Citation
The archaeology of Afro-American slavery in coastal Georgia : a regional perception of slave household and community patterns

Material Information

Title:
The archaeology of Afro-American slavery in coastal Georgia : a regional perception of slave household and community patterns
Creator:
Singleton, Theresa A
Publication Date:
Language:
English
Physical Description:
xiii, 248 leaves : ill., maps ; 28 cm.

Subjects

Subjects / Keywords:
Anthropology thesis Ph. D ( lcsh )
Antiquities -- Georgia ( lcsh )
Dissertations, Academic -- Anthropology -- UF ( lcsh )
Slavery -- Georgia ( lcsh )
Butler Island ( local )
Slavery ( jstor )
Rice ( jstor )
Plantations ( jstor )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1980.
Bibliography:
Bibliography: leaves 229-247.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Theresa A. Singleton.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
023443427 ( ALEPH )
07333676 ( OCLC )

Downloads

This item has the following downloads:

ERTWT0LXS_WYG5HC.xml

AA00024758_00001.pdf

AA00024758_00001_0165.txt

archaeologyofafr00sing_0094.txt

AA00024758_00001_0188.txt

archaeologyofafr00sing_0140.txt

AA00024758_00001_0193.txt

AA00024758_00001_0034.txt

archaeologyofafr00sing_0079.txt

archaeologyofafr00sing_0173.txt

archaeologyofafr00sing_0081.txt

AA00024758_00001_0071.txt

archaeologyofafr00sing_0230.txt

archaeologyofafr00sing_0219.txt

archaeologyofafr00sing_0043.txt

archaeologyofafr00sing_0068.txt

archaeologyofafr00sing_0051.txt

archaeologyofafr00sing_0204.txt

archaeologyofafr00sing_0178.txt

archaeologyofafr00sing_0018.txt

archaeologyofafr00sing_0227.txt

archaeologyofafr00sing_0259.txt

archaeologyofafr00sing_0089.txt

archaeologyofafr00sing_0253.txt

AA00024758_00001_0263.txt

archaeologyofafr00sing_0057.txt

archaeologyofafr00sing_0241.txt

archaeologyofafr00sing_0104.txt

AA00024758_00001_0198.txt

archaeologyofafr00sing_0198.txt

archaeologyofafr00sing_0093.txt

archaeologyofafr00sing_0005.txt

archaeologyofafr00sing_0158.txt

AA00024758_00001_0175.txt

archaeologyofafr00sing_0047.txt

AA00024758_00001_0159.txt

AA00024758_00001_0026.txt

archaeologyofafr00sing_0146.txt

AA00024758_00001_0120.txt

archaeologyofafr00sing_0086.txt

AA00024758_00001_0096.txt

AA00024758_00001_0019.txt

AA00024758_00001_0111.txt

archaeologyofafr00sing_0067.txt

archaeologyofafr00sing_0020.txt

AA00024758_00001_0218.txt

AA00024758_00001_0113.txt

archaeologyofafr00sing_0002.txt

archaeologyofafr00sing_0119.txt

AA00024758_00001_0210.txt

archaeologyofafr00sing_0222.txt

archaeologyofafr00sing_0058.txt

archaeologyofafr00sing_0142.txt

archaeologyofafr00sing_0162.txt

AA00024758_00001_0138.txt

archaeologyofafr00sing_0097.txt

AA00024758_00001_0242.txt

AA00024758_00001_0166.txt

AA00024758_00001_0023.txt

AA00024758_00001_0014.txt

archaeologyofafr00sing_0108.txt

AA00024758_00001_0184.txt

archaeologyofafr00sing_0147.txt

AA00024758_00001_0029.txt

archaeologyofafr00sing_0175.txt

archaeologyofafr00sing_0193.txt

AA00024758_00001_0043.txt

archaeologyofafr00sing_0023.txt

archaeologyofafr00sing_0124.txt

archaeologyofafr00sing_0202.txt

archaeologyofafr00sing_0075.txt

archaeologyofafr00sing_0184.txt

AA00024758_00001_0035.txt

AA00024758_00001_0148.txt

archaeologyofafr00sing_0252.txt

AA00024758_00001_0072.txt

AA00024758_00001_0089.txt

AA00024758_00001_0260.txt

archaeologyofafr00sing_0042.txt

archaeologyofafr00sing_0118.txt

AA00024758_00001_0053.txt

AA00024758_00001_0182.txt

AA00024758_00001_0006.txt

AA00024758_00001_0152.txt

AA00024758_00001_0061.txt

AA00024758_00001_0108.txt

AA00024758_00001_0052.txt

AA00024758_00001_0157.txt

AA00024758_00001_0017.txt

AA00024758_00001_0187.txt

AA00024758_00001_0156.txt

AA00024758_00001_0151.txt

archaeologyofafr00sing_0059.txt

AA00024758_00001_0173.txt

archaeologyofafr00sing_0026.txt

AA00024758_00001_0202.txt

AA00024758_00001_0194.txt

archaeologyofafr00sing_0004.txt

archaeologyofafr00sing_0028.txt

AA00024758_00001_0082.txt

archaeologyofafr00sing_0213.txt

AA00024758_00001_0048.txt

archaeologyofafr00sing_0163.txt

AA00024758_00001_0001.txt

AA00024758_00001_0016.txt

archaeologyofafr00sing_0069.txt

AA00024758_00001_0103.txt

AA00024758_00001_pdf.txt

AA00024758_00001_0204.txt

archaeologyofafr00sing_0171.txt

archaeologyofafr00sing_0153.txt

AA00024758_00001_0196.txt

archaeologyofafr00sing_0101.txt

archaeologyofafr00sing_0196.txt

AA00024758_00001_0212.txt

archaeologyofafr00sing_0111.txt

AA00024758_00001_0244.txt

archaeologyofafr00sing_0040.txt

AA00024758_00001_0010.txt

archaeologyofafr00sing_0181.txt

AA00024758_00001_0174.txt

AA00024758_00001_0127.txt

AA00024758_00001_0251.txt

archaeologyofafr00sing_0256.txt

AA00024758_00001_0032.txt

archaeologyofafr00sing_0065.txt

AA00024758_00001_0246.txt

archaeologyofafr00sing_0009.txt

AA00024758_00001_0215.txt

archaeologyofafr00sing_0229.txt

archaeologyofafr00sing_0167.txt

AA00024758_00001_0257.txt

archaeologyofafr00sing_0096.txt

archaeologyofafr00sing_0224.txt

archaeologyofafr00sing_0199.txt

AA00024758_00001_0238.txt

archaeologyofafr00sing_0131.txt

AA00024758_00001_0247.txt

archaeologyofafr00sing_0182.txt

archaeologyofafr00sing_0258.txt

archaeologyofafr00sing_0120.txt

AA00024758_00001_0097.txt

archaeologyofafr00sing_0170.txt

AA00024758_00001_0003.txt

AA00024758_00001_0070.txt

archaeologyofafr00sing_0165.txt

archaeologyofafr00sing_0207.txt

archaeologyofafr00sing_0041.txt

archaeologyofafr00sing_0180.txt

AA00024758_00001_0239.txt

archaeologyofafr00sing_0022.txt

AA00024758_00001_0249.txt

AA00024758_00001_0020.txt

archaeologyofafr00sing_0077.txt

archaeologyofafr00sing_0031.txt

archaeologyofafr00sing_0200.txt

AA00024758_00001_0229.txt

archaeologyofafr00sing_0141.txt

AA00024758_00001_0191.txt

archaeologyofafr00sing_0201.txt

archaeologyofafr00sing_0145.txt

archaeologyofafr00sing_0098.txt

archaeologyofafr00sing_0255.txt

archaeologyofafr00sing_0261.txt

archaeologyofafr00sing_0032.txt

archaeologyofafr00sing_0080.txt

archaeologyofafr00sing_0016.txt

AA00024758_00001_0056.txt

archaeologyofafr00sing_0166.txt

AA00024758_00001_0172.txt

archaeologyofafr00sing_0064.txt

AA00024758_00001_0133.txt

AA00024758_00001_0101.txt

AA00024758_00001_0041.txt

archaeologyofafr00sing_0007.txt

archaeologyofafr00sing_0117.txt

AA00024758_00001_0141.txt

AA00024758_00001_0255.txt

AA00024758_00001_0200.txt

AA00024758_00001_0100.txt

archaeologyofafr00sing_0076.txt

AA00024758_00001_0128.txt

archaeologyofafr00sing_0055.txt

archaeologyofafr00sing_0130.txt

AA00024758_00001_0064.txt

AA00024758_00001_0024.txt

archaeologyofafr00sing_0121.txt

AA00024758_00001_0231.txt

archaeologyofafr00sing_0050.txt

archaeologyofafr00sing_0105.txt

archaeologyofafr00sing_0107.txt

AA00024758_00001_0054.txt

AA00024758_00001_0155.txt

archaeologyofafr00sing_0192.txt

AA00024758_00001_0250.txt

AA00024758_00001_0025.txt

archaeologyofafr00sing_0052.txt

AA00024758_00001_0226.txt

archaeologyofafr00sing_0083.txt

AA00024758_00001_0073.txt

archaeologyofafr00sing_0074.txt

AA00024758_00001_0028.txt

archaeologyofafr00sing_0010.txt

archaeologyofafr00sing_0027.txt

AA00024758_00001_0046.txt

AA00024758_00001_0062.txt

archaeologyofafr00sing_0048.txt

archaeologyofafr00sing_0220.txt

archaeologyofafr00sing_0129.txt

archaeologyofafr00sing_0033.txt

archaeologyofafr00sing_0062.txt

AA00024758_00001_0099.txt

archaeologyofafr00sing_0024.txt

AA00024758_00001_0186.txt

AA00024758_00001_0195.txt

archaeologyofafr00sing_0072.txt

archaeologyofafr00sing_0245.txt

archaeologyofafr00sing_0087.txt

archaeologyofafr00sing_0061.txt

AA00024758_00001_0259.txt

archaeologyofafr00sing_0157.txt

archaeologyofafr00sing_0125.txt

archaeologyofafr00sing_0046.txt

AA00024758_00001_0119.txt

archaeologyofafr00sing_0176.txt

AA00024758_00001_0121.txt

archaeologyofafr00sing_0211.txt

AA00024758_00001_0102.txt

archaeologyofafr00sing_0236.txt

AA00024758_00001_0227.txt

archaeologyofafr00sing_0063.txt

archaeologyofafr00sing_0126.txt

archaeologyofafr00sing_0115.txt

archaeologyofafr00sing_0001.txt

AA00024758_00001_0256.txt

AA00024758_00001_0232.txt

AA00024758_00001_0063.txt

AA00024758_00001_0012.txt

AA00024758_00001_0058.txt

AA00024758_00001_0067.txt

archaeologyofafr00sing_0100.txt

archaeologyofafr00sing_0206.txt

archaeologyofafr00sing_0138.txt

AA00024758_00001_0199.txt

AA00024758_00001_0078.txt

archaeologyofafr00sing_0034.txt

archaeologyofafr00sing_0095.txt

AA00024758_00001_0253.txt

archaeologyofafr00sing_0186.txt

archaeologyofafr00sing_0168.txt

archaeologyofafr00sing_0149.txt

archaeologyofafr00sing_0179.txt

AA00024758_00001_0177.txt

AA00024758_00001_0201.txt

AA00024758_00001_0124.txt

AA00024758_00001_0105.txt

archaeologyofafr00sing_0228.txt

AA00024758_00001_0084.txt

archaeologyofafr00sing_0185.txt

AA00024758_00001_0047.txt

AA00024758_00001_0153.txt

AA00024758_00001_0081.txt

AA00024758_00001_0180.txt

AA00024758_00001_0163.txt

AA00024758_00001_0258.txt

AA00024758_00001_0123.txt

AA00024758_00001_0130.txt

AA00024758_00001_0126.txt

archaeologyofafr00sing_0066.txt

archaeologyofafr00sing_0231.txt

AA00024758_00001_0107.txt

AA00024758_00001_0225.txt

AA00024758_00001_0079.txt

AA00024758_00001_0171.txt

archaeologyofafr00sing_0160.txt

AA00024758_00001_0066.txt

archaeologyofafr00sing_0234.txt

AA00024758_00001_0140.txt

archaeologyofafr00sing_0188.txt

archaeologyofafr00sing_0128.txt

archaeologyofafr00sing_0177.txt

AA00024758_00001_0080.txt

archaeologyofafr00sing_0099.txt

AA00024758_00001_0208.txt

AA00024758_00001_0197.txt

archaeologyofafr00sing_0217.txt

archaeologyofafr00sing_0251.txt

archaeologyofafr00sing_0114.txt

archaeologyofafr00sing_0189.txt

AA00024758_00001_0092.txt

AA00024758_00001_0216.txt

archaeologyofafr00sing_0249.txt

AA00024758_00001_0090.txt

AA00024758_00001_0149.txt

AA00024758_00001_0069.txt

AA00024758_00001_0005.txt

AA00024758_00001_0146.txt

archaeologyofafr00sing_0215.txt

AA00024758_00001_0036.txt

archaeologyofafr00sing_0109.txt

archaeologyofafr00sing_0174.txt

archaeologyofafr00sing_0148.txt

AA00024758_00001_0134.txt

archaeologyofafr00sing_0152.txt

AA00024758_00001_0018.txt

AA00024758_00001_0147.txt

archaeologyofafr00sing_0082.txt

archaeologyofafr00sing_0021.txt

archaeologyofafr00sing_0194.txt

archaeologyofafr00sing_0037.txt

archaeologyofafr00sing_0091.txt

archaeologyofafr00sing_0155.txt

archaeologyofafr00sing_0012.txt

AA00024758_00001_0008.txt

archaeologyofafr00sing_0139.txt

archaeologyofafr00sing_0197.txt

AA00024758_00001_0233.txt

AA00024758_00001_0077.txt

archaeologyofafr00sing_0218.txt

AA00024758_00001_0050.txt

AA00024758_00001_0045.txt

AA00024758_00001_0083.txt

archaeologyofafr00sing_0029.txt

archaeologyofafr00sing_0150.txt

archaeologyofafr00sing_0190.txt

AA00024758_00001_0007.txt

archaeologyofafr00sing_0242.txt

AA00024758_00001_0115.txt

archaeologyofafr00sing_0172.txt

archaeologyofafr00sing_0238.txt

AA00024758_00001_0190.txt

archaeologyofafr00sing_0183.txt

AA00024758_00001_0106.txt

archaeologyofafr00sing_0053.txt

archaeologyofafr00sing_0106.txt

archaeologyofafr00sing_0246.txt

archaeologyofafr00sing_0237.txt

AA00024758_00001_0164.txt

archaeologyofafr00sing_0025.txt

AA00024758_00001_0221.txt

AA00024758_00001_0144.txt

AA00024758_00001_0169.txt

archaeologyofafr00sing_0216.txt

AA00024758_00001_0135.txt

AA00024758_00001_0160.txt

AA00024758_00001_0118.txt

AA00024758_00001_0040.txt

AA00024758_00001_0223.txt

archaeologyofafr00sing_0208.txt

AA00024758_00001_0132.txt

archaeologyofafr00sing_0092.txt

AA00024758_00001_0109.txt

archaeologyofafr00sing_0060.txt

archaeologyofafr00sing_0134.txt

archaeologyofafr00sing_0212.txt

AA00024758_00001_0189.txt

archaeologyofafr00sing_0102.txt

AA00024758_00001_0068.txt

archaeologyofafr00sing_0144.txt

archaeologyofafr00sing_0203.txt

AA00024758_00001_0131.txt

archaeologyofafr00sing_0085.txt

archaeologyofafr00sing_0011.txt

archaeologyofafr00sing_0019.txt

archaeologyofafr00sing_0132.txt

archaeologyofafr00sing_0133.txt

AA00024758_00001_0049.txt

AA00024758_00001_0219.txt

archaeologyofafr00sing_0169.txt

AA00024758_00001_0224.txt

archaeologyofafr00sing_0122.txt

AA00024758_00001_0091.txt

AA00024758_00001_0214.txt

AA00024758_00001_0183.txt

AA00024758_00001_0230.txt

archaeologyofafr00sing_0136.txt

AA00024758_00001_0211.txt

archaeologyofafr00sing_0154.txt

AA00024758_00001_0114.txt

archaeologyofafr00sing_0014.txt

archaeologyofafr00sing_0156.txt

archaeologyofafr00sing_0233.txt

AA00024758_00001_0110.txt

archaeologyofafr00sing_0159.txt

AA00024758_00001_0179.txt

AA00024758_00001_0178.txt

archaeologyofafr00sing_0015.txt

AA00024758_00001_0203.txt

AA00024758_00001_0261.txt

archaeologyofafr00sing_0044.txt

AA00024758_00001_0170.txt

archaeologyofafr00sing_0191.txt

archaeologyofafr00sing_0035.txt

AA00024758_00001_0262.txt

AA00024758_00001_0222.txt

AA00024758_00001_0015.txt

archaeologyofafr00sing_0038.txt

archaeologyofafr00sing_0071.txt

archaeologyofafr00sing_0045.txt

AA00024758_00001_0136.txt

archaeologyofafr00sing_0240.txt

archaeologyofafr00sing_0088.txt

AA00024758_00001_0021.txt

AA00024758_00001_0042.txt

archaeologyofafr00sing_0250.txt

archaeologyofafr00sing_0209.txt

AA00024758_00001_0185.txt

archaeologyofafr00sing_0151.txt

AA00024758_00001_0240.txt

archaeologyofafr00sing_0262.txt

AA00024758_00001_0150.txt

archaeologyofafr00sing_0078.txt

AA00024758_00001_0125.txt

AA00024758_00001_0095.txt

archaeologyofafr00sing_0112.txt

archaeologyofafr00sing_0113.txt

AA00024758_00001_0248.txt

AA00024758_00001_0055.txt

archaeologyofafr00sing_0137.txt

archaeologyofafr00sing_0039.txt

AA00024758_00001_0022.txt

AA00024758_00001_0087.txt

AA00024758_00001_0145.txt

AA00024758_00001_0241.txt

AA00024758_00001_0057.txt

AA00024758_00001_0004.txt

archaeologyofafr00sing_0017.txt

AA00024758_00001_0143.txt

archaeologyofafr00sing_0210.txt

archaeologyofafr00sing_0003.txt

AA00024758_00001_0002.txt

archaeologyofafr00sing_0232.txt

AA00024758_00001_0033.txt

archaeologyofafr00sing_0036.txt

AA00024758_00001_0051.txt

AA00024758_00001_0217.txt

AA00024758_00001_0161.txt

AA00024758_00001_0085.txt

AA00024758_00001_0013.txt

archaeologyofafr00sing_0244.txt

AA00024758_00001_0094.txt

archaeologyofafr00sing_0257.txt

archaeologyofafr00sing_0248.txt

archaeologyofafr00sing_0205.txt

AA00024758_00001_0213.txt

AA00024758_00001_0030.txt

AA00024758_00001_0254.txt

archaeologyofafr00sing_0161.txt

AA00024758_00001_0139.txt

AA00024758_00001_0117.txt

archaeologyofafr00sing_0070.txt

archaeologyofafr00sing_0223.txt

AA00024758_00001_0098.txt

AA00024758_00001_0234.txt

archaeologyofafr00sing_0054.txt

archaeologyofafr00sing_0110.txt

AA00024758_00001_0167.txt

archaeologyofafr00sing_0056.txt

AA00024758_00001_0158.txt

AA00024758_00001_0192.txt

AA00024758_00001_0060.txt

archaeologyofafr00sing_0254.txt

AA00024758_00001_0252.txt

archaeologyofafr00sing_0164.txt

archaeologyofafr00sing_0127.txt

AA00024758_00001_0220.txt

archaeologyofafr00sing_0195.txt

AA00024758_00001_0065.txt

AA00024758_00001_0038.txt

AA00024758_00001_0112.txt

AA00024758_00001_0243.txt

AA00024758_00001_0236.txt

archaeologyofafr00sing_0030.txt

AA00024758_00001_0207.txt

archaeologyofafr00sing_0008.txt

archaeologyofafr00sing_0006.txt

AA00024758_00001_0122.txt

archaeologyofafr00sing_0239.txt

AA00024758_00001_0009.txt

archaeologyofafr00sing_0247.txt

archaeologyofafr00sing_0187.txt

AA00024758_00001_0075.txt

archaeologyofafr00sing_0123.txt

AA00024758_00001_0044.txt

archaeologyofafr00sing_0235.txt

AA00024758_00001_0206.txt

AA00024758_00001_0011.txt

AA00024758_00001_0116.txt

archaeologyofafr00sing_0084.txt

AA00024758_00001_0235.txt

AA00024758_00001_0209.txt

AA00024758_00001_0168.txt

AA00024758_00001_0037.txt

AA00024758_00001_0088.txt

AA00024758_00001_0093.txt

archaeologyofafr00sing_0214.txt

archaeologyofafr00sing_0225.txt

AA00024758_00001_0129.txt

archaeologyofafr00sing_0221.txt

AA00024758_00001_0205.txt

AA00024758_00001_0181.txt

AA00024758_00001_0228.txt

AA00024758_00001_0142.txt

AA00024758_00001_0104.txt

AA00024758_00001_0162.txt

archaeologyofafr00sing_0013.txt

AA00024758_00001_0059.txt

AA00024758_00001_0039.txt

AA00024758_00001_0176.txt

AA00024758_00001_0027.txt

AA00024758_00001_0237.txt

archaeologyofafr00sing_0073.txt

AA00024758_00001_0076.txt

archaeologyofafr00sing_0049.txt

archaeologyofafr00sing_0116.txt

archaeologyofafr00sing_0103.txt

AA00024758_00001_0086.txt

archaeologyofafr00sing_0260.txt

AA00024758_00001_0154.txt

archaeologyofafr00sing_0143.txt

AA00024758_00001_0074.txt

AA00024758_00001_0137.txt

archaeologyofafr00sing_0000.txt

archaeologyofafr00sing_0090.txt

AA00024758_00001_0031.txt

archaeologyofafr00sing_0226.txt

archaeologyofafr00sing_0243.txt

AA00024758_00001_0245.txt

archaeologyofafr00sing_0135.txt


Full Text













THE :; .-.! iGY OF AFRO-ARICAN SLAVERY IN COASTAL GEORGIA:
A REGIC:"K PEkCEPTION OF SLAVE HOUSEHOLD AND CC:,' TY PATERNS












BY

iHER:'\ A. S":71 7TO1


A DISSERTATION P-7-'NTED TO THE GRAD;' TE CO 'IL
F THE ..VRSITY OF FL.ORIDA Tl
PARTIAL FULFILLMENT OF THE RElU H
FOR THE GREE OF DOCTOR OF 'HILOSOHiY


UNIVERSITY OF ,i'RTDA





























SCopyright 1980O

by

Theresa A. Singleton





























DedicaIted to the fiemory of father, illiam E Singleton r.













AC I ;. I is


Many persons made this effort possible. Several faculty members

at the University of Florida have been central in helping me develop ny

research and writing skills, particularly Drs. Fairbanks, Jerald T.

Milanich, Maxine -'-golis, Prudence Rice, Samuel Proctor, and Ronald

Foreman. Dr. Fairbanks has been a mentor and friend throughout my gradu-

ate career and I have enjoyed studying with him. I also would like to

thank Dr. Vorn -*ing for reading the final draft for my defense.

Dr. Kathleen A. Deagan, "' first instructor in field methods of histori.-

cal archaeology, has always been supportive and interested in my career

goals.

Very special thanks go to Dr. Lewis Larson, State Archaeologist

for Georgia. I ' pleased that he asked me to undertake the Putler

Island Project. Also, his staff deserve special mention, particularly

Linda Timmings, for undertaking all of the bookkeeping responsibilities

for the project.

I em very grateful to the field crews of summer 19 78, spring 1979,

arnd summ!'er 1979 for working along with mue in .the Butler Island -arsh.

I think we are the only ones who can truly appreciate what slave life

rM'st have been like at Butter Island. I a;n particularly grateful to

Ernest Despai : for the detaileAd drawings r the chimney Biill Fisher

"O r hiding out tiih ie 'laboratoy man',sis and for sskin' S his fOther,

Dr. Richard Finher, to undertake the soil analyses; and Lorraine kcCosker

for his ass stance with th photographs'.










Appreciation is extended to Gene Love, David Edwards, and the re-

mainder of the staff of the Altamaha Waterfowl' Man.agement Area for their

advice, equipment, and time.

I would like to thank the coimmunity of Darien, Georgia, for the

hospitality they extended to me and the crew during our stay there. Spe-

cial thanks go to Mis. Bessie Lewis, Jim Cook, Nurse Campbell, Mrs. Bates,

Nan Earl Wylly, and Rudolph Capers for their valuable in format ion relating

to the findings presented in this study.

The archival research was an extremely important aspect of this study.

I would like to thank the staffs of the Pennsylvania Historical Society,

the Georgia Historical Society, and the P. K. Yonge Library at the Univer-

sity of Florida for their assistance. Also acknowledged are the staff of

the Inter-Library Loan Department at the University of Florida Library for

securing many of the materials I needed.

A warm appreciation is extended to my colleagues for their comments

and suggestions concerning my research objectives: Kenneth Lewis, Patrick

Garrow, Thomas W'heaton, Leland Fergueson, Jennifer Hamilton, Chad Bradly,

Sue Mulli.ns-Moore, John Morgan, Jr., Morgan R. Crook, Elizabeth Wing,

Rochelle Marrinan, Larry Banks, Nicholas Honerkemp, Robin Smith, and

Leslie S. Libberman. Also, I would like to thank John A. Scott for his

- ..c:,.tions; Richard Fisher for undertaking the soil analyses; Kathy

Bordua, my lab assistant and confident during the analysis and writing of

the dissertation; Ashley Wood, the draftsnan and illustrator; rand Sue

Kirkpatrick for typing the fi-ial manuscript.

Finally, I would like to thank .- family and friends for their

encouragement through the years.















TABLE OF C"-ITENTS


A.' 7 zE I I .7


LIST I F . . . . .

LIST OF FIGURES . . . . . . . . . . .


ABS-T RACT . . . . .. . . .

CHAPTER

1 IN'1FCD UCTON . . . . . . . .


Region) 2a0ys.is in Archeoio.y ... .. ....
Regional 1 ,- ..ac-s- to the St. y ou Slavey' . .. .
,The Arc aeo uy of Afro-A',Tric ns .. . . ..
ProbIei''s ad . . . .
3J PO'C W : i k S . .

2. PLI'ANTATION '-ICULJTURE IN TH!" E O' Til .O. T r

Ernviroamsaentu Sup:: vry and Intrroducticn .... .
t, ys r Ge r a.
The Ee -ice of the, PlanLation rScte. in rgia .
Te Rice Industry . . . . . . . . .
Ri ce 'a.nd Slave rry . . . . .
M,1hods of T.. rri atior n n a 'd P t Rice . .i
-3 .g-
H arves.ig, 1 v li and ...in.. .......
o-Stape Colton Cu"'l Cure......
MIel tods of P1 artiing and Cultivati.:o .
H arv esg- s.. t. -. C -. . . .. . . ..
S u 'ia y . . . . . . . . . . .
F...,. a, hu -

I I-STOPRICPL A"JD ARCA EOL', -O-ICAL .VE STIGATI Al BUT1 0LR
1 i1. -.7 :. . . .. . . . .


Ownership and Occup_-,ation o
PDI an-ttion 1c22se eit at Butler
Records of S Life .. ..

A rchaIo 1' '- o-.'c: e .
S'\.ve Vfi11?;;jp Sitcs -...


S r. .. .. .. ..... PATTER i COASPTAL SiA' C'
-.~ ~~~ AL3- 221


is:d 0 802-- 18G0


ITIES . . .


uS o m 0' ..........t t ef 1, tL2
"Te o .t-in ,nts of PO anI a-tioI e. t'21 .' Patt. ,rns2
. u.ve Co:i'r-'.inityv orgnizat-;Ions a' Pui er Island . . .


.. . . iV


]ii


'1


. 20

20
S. 23
. 26
'o
31
6 ;
39

4'
, 4 ^




. 58
o B9
15





0 43









Page


CHAPTER

The Community Plan at Slave Setlement 4 .. . 1
Slave Dwellings . . ... .............. . .. 126
Chimney construction . . . . . . . 130
Building hardware . . . . . . 137
Sumnmary . . . . ........................ . 138

5, THE ,-,,IAL CULTURE OF SLAVERY . . . . ... . 140

Kitchen Artifacts .. . . . . . . .. 141
Ceramics ........... ......................42
Ceramic styles and vessel forms .. ......... .. i43
Dating the ceramics from Butler Island ..... 155
,.,raminc Artifacts . . . . . . . . .57
Glass bottles and tablewares . . . ... 157
Cooking and processing equipment . . . ...9..
E.atin utencil.s 1....................
Ec~t r~g Lit nci s . . . . . . . . 1 0
Kitchen Artifacts: Summary . . . . .i..5
Food Procurement . . . . .................... 166
Food Resources . . . . . . . . 169
Domestic Food Resources . ........... . ... 13
.....'.. ,-.:tic Food Resources . . . . . . .
Fishs "-u tutls....................... 179
Fishes and turtles . . . . . . . . 179
Birds and Mammals . . . . . . . .. 183
Sumoairy of Food Resources . . . . ...... 184
Clothing . . . . . . . . . . . . 1 6
Buttons . . . . . . . . . . . 6
Footwear... .... ..................... *
Accessories and Ornaments . . . . . .. .. 0..'
Beads *........... . ... . . .. ... 190
Miscellaneous Items . . . . . . .. 193
Sum ry . . . . . . . * * * _'3
Household Ites . . . . . . . ... 193
Tobacco Equipment . . .
Farming Tools and Specialized Crafts . . . . . 99
Summary . . . . . . . . . .......... 109

6. PAk -7[a REW'ITION IN SLAVE '1 TRIAL CULTURE . . . 211

7. S, : .SIONS . . . . . . . . . 219

APPnD-.IX

1 K. SW Y Or SOIL A 5";YS.s FROM SE01EKET 4, BUDILR ISLAND 12,11

2. SLAVE SUBSISTENCE AR1IFACTS FR SETTLF711. "I . . . 225

ABBREVIATI ', USED N REFER NCES. . .. . ...... . . . . 228









Page

REFERE CES. . . . . . . . . . 2
. . . . . . . . . . . 229


BIOGRPHICI\L S' :'.i9i...
. . . . . . . . . . . 248'-


v1 ii












LIST OF TABLES


Table Pa,,e

1. Ownership of Butler IslanI-d, 1790? to the Present 50

2. Rice Production at Buter Tsland, 1T02-1833 4

3. Annual Births a'.d Deth of Butler Slaves 71

4. An Apcroxi Mmate List of S Iaves by Age and Task Rate at Butler
is Ia nd, I803-1845: Field Hands
5. An Approximate List of Slaves, by Age and Occupation, at
Butler Island, 1803-1845: Plantation Srecialists 76

6. 1830 Census of Slaves at. Butler Island 77

7. Mortality Rates of Infants and Ch-ildren of Sieves on the
Butter Estate, 18138-384 81

8. Vessel Forms, of Ceramic St.yles 14

9. Freuences of Decoraited Taberes 5

10. App icati on of the eaI Ceramic, Date Formula to Tab ewd re s 156

i. Styles of Glass Bott.es Pepresented at Butler Island,
SIttlemert ,'4 158

12. identifedFa rom Slave Site Q4, Butier Island 171

13. Relative Weihts of Domestic and Nonduesti c Animal Food
fro-m ";,",!le, s a'
fro Buter t Islaend 1/2

1 4. T e e ,l iht,- of Doniestic Sp-,c-ies t But.ler Islad 7;4

15. Re '- t- v Wei qit of Td t rfiFed domestic Food by Anal,
Class 1 79

16 ATfac Profilrs fc o'; f. r Slavery ites in 'eorgiaFlri.a 2,
17. j t-': -A if ac. P o' .o. i g o.'s


18. Tie Slave Arti. fact tttern in Coastal Georni /.ori.da 216















LIST OF FJGUfI.S


Figure Page

1. Tnvestigted Slavery S ites in Coastal Geor' gia. 15

2. The Physiographic Regiuns of Georgia. 21

3. ExIenL,--' of Rice Cultivation in Geor(ia ia in 1860 27

4. Butler Island and Environs. 47

5. The Butler Georgian Estate. 52

6. The Subdivision of Butlter and Genera l's islands inIto
Tenant Plots (circa 1883). 57

7. Historic Sites at Butler Island. 5

8. Present-Day View of Butler Island. .

9. Extent of Excavations at Settlement 5

10. Stratigraphic Profile of Settleiient #l. 17

t. Eig chteenth-Century S1 i-, re F-ra A.ts from Settlement #l. 100

1?. Spanish real, 'i:..ico Mint frci , Settlejei.nt 4. 02

13. Contour Map of Settlement 10i

14. Excavaltion of Set.tlement 14. 10

is. tr.a1ti g'rap ic Pr 10 Se17tlemen1t : S ....

16. 0 erseer' s Ho u. 5I

17. Tidal Mi 11 ad .o.t Landing. 15

13. Sa ve CbI. 6

19. Steoam Mill Ruins. .i

20. Linear Arr. -ir.nt of Slave Dvwelings. -

21. Extent of Excavation at Structure Four. 122


. '/^^^yAq^s;^'^^"'-^'^









Figure Page

22. Stratigraphic Profile of Drainage Ditch. 124

23. The Extent of Excavation at Structure One, Settlement fX. 123

24. Hearth of Chimney. 132

25. Chimney Brickwork. 133

26. Arch-Shaped Hearth Support. 135

27. Red Coarse Earthenwar'es. 147

28. Transfer Printed Perlw,'are Fragments. 150

29. Annular Ceramics. 153

30. ;. .l.'ar Ceramics. 153

31. Eating Utencils. 154

32. Carved Bone Fragments. 164

33. Food Procurement Artifacts. 168

34. Clothing and Personal Adornments. 189

35. Shoe Soles. 192

36. Parasol Attachment. 195

37. Eye Lens. 197

38. Porcelain Pipe Bowl. 201

39. Woodvorking Toos. 203

40. Unidentified Tools. 206

41. Sia' -- de Ceramics, 208














Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy

THEC ARCHAEOLOGY OF AFRO-AMERICAN SLA'.' IN COASTAL ...' 4:
A REGIC'"A' PERCEPTI : OF SLAVE HOUSEHOLD f' i& ...ITY PAT TERNS

By

Theresa A. Singleton

December Ic .

Chairman: Charles H. Fairbanks
Major Department: Anthropology

A variety of methodological and theoretical orientations has been

utilized in the study of Afro-American slavery. R--.onal approaches are

not new to sla' ..', but few studies have examined the impact of local

ecological factors upon slave lifeways within a narrowly defined area.

In this study, coastal Georgia (the barrier islands and the. adjacent

deltas) provides the region for the c oration of an ecological approach

to slave life and material conditions. Specifically, slavery at a rice

coast plantation is compared with slavery at long-staple cotton planta-

tions. Archaeological data provide the primary basis for interpretation,

and historical resources are used to supplement the archaeology.

Primary .settlement patterns and subsistence data recovered from the

Butler island rice plantation in 1978 and 1979 are used to investigate

save life and material co editions on the rice coast. The But'er Islari

data are co pc..rec with t arch-"aeolog ical data recovered frome sV:.ra1

previously inves tiogated slave sites of long-staple cotton plantations.

The comparison reveals si nilarities as we'll as differences in the










slave lifeways oF the two cash crop regions. Archaeologica.ly, the

differences are primarily discernible in slave comui-nity organizations,

the natural resources exploited, and slave crafts. Similarities are

evident in most household artifact pLatterns: plantation food rations,

food preparation equipment, personal possessions, and leisure-tie activi-

ties. it is s- res ted that the differences in slave life reflect the

dissimilarity between th-e habitats where rice and long-staple cotton w ere

produced. Oni the other hand, the similarities may reflect general labor

mnane gem.ent practices adapted for the production of coastal staples.


xA i i













CHAPTER 1
1 -i ,.,UCTION


Tho anthropological study of Afro-American slavery has increas-

ingiy used archaeological data as a primary source for interpreta-

tion. Because traditional historic sources are subject to problems of

falsification and bias, archaeological data are frequently used to

supplement historic records. It has been suggested that archaeology be

the key to writing the cultural history of the inarticulatet" (Ascher

1974:10-12). Slavery sites have been investigated specifical iy for this

reason. As a consequence of the widespread illiteracy i.:.- slaves,

few documents written by slaves exist. "".-: of our understanding of

slavery has come from accounts written by whites. inferences derived

from the archaeological record of slavery can provide insights into

slave lifeways and slave worldview. Although planter records: travelers'

accounts, and oral interviews are used extensively here to supplement the

archaeological record, this study assumes that archaeological data are

one of th:e best indices of slave behavior.

The purpose of this study is to offer a model capable of dis-

cerning slave behavioral patterns in coastal Georgia. Slavery on the

Georgia coast was characterized by distinctive demographic, environmen-

tal econoic, as wel! as historical factors 'which were absent in the

interior.. -doultedly, these conditions affected the quality of slave'

life, save activities, and ultimately the developmeInt of save cUN Urc

patterns.









AL present, slave sites from coastal Georgia are perhaps better

known than those from any other regions of the Old South. In recent

years, a number of slave sites have been identified, located, and exca-

vO, taed n this area. Un til n,, no attemtL has beE made to syrthJesize

these archaeological finngs thin a fra6meloark larger than he singlIle

pl antation.

Archaeological investigations undertake, at the Butler Is and
pl nt t on in "T.' Is County, .. .. -" '.. .
p1Cant-ton in *Ttosh CountyS Georgia, provide the first rcheo'ogical

evidence of slave life at a Georgian rice plantation. AlI reported

excavated slave sites in Georgia are former long-staple cotton pl-a.-

tirons on the barrier islands. Thus, the significance of the Butler Ildr '

research is t.o.fod: first, iL sr.-lies archaeological evidence for the

specific adapt&Lions of slavevv to rice ci'ture; second, comparisons

o e uler I -s ld at wi dat, fro nong-s at e coton, s ve si tes

ndi ca e several ident ie cal patterns. This evoideIce suggests 1ht h Ie

cultural system of slavery h-i ch deveel "1- in costa I Georgia xe(:.eded

adaptations to specific cash crop requirements.

U! timately, the goal cf this dissertation is to dent-fy End ofrer

possile explanati ons for slave coccrnity and househo patterns- This

objective is accompl ished through the analysis s of primary settle -' L
ad ssis ter ce d gta recovered from Butler Island save sites. Comn-

par ,e data from o'etlr roast al slave sites provide additional evi-

d for regional interpretations.. 1Th theor-tica orent nation of
dlelce s fr regiuadl i dLa E- .... n s i

c'!i "pp oac i n: t"e " .- er p Cre ation r.. o c t l rft,- n n.

Lccc ygi Cal Fcpfr'oF,,ii, "h Iai~rrt~in -L e r -, ptr










P,-^- .. l *^ :.:- :- il ,'' --' ,/

Systematic approaches toward the organization of archaeological

data within temporal and spatial frameworks have been undertaken in

archaeoi:,.' for most of this century. Although early regional studies

lacked the sophisticated methodology and theoretical orientations of

recent years, large amounts of archaeological data were collected for

many states and regions in the United States (Hole .and Heizer 1977:

13-14). These data have provided the basis for regional chronologies,

cultures, and the frameworks for the "more-analytic causal ly-orierted

approaches" of today (Adams 1968:1188).

The concepts end methods for the regional analysis of the "new

archaeolcc have been derived from geography. Stated simply, the

region is a "part of the earth's surface which is distinguished in

some defined way from the surrounding area." This disti nctiveness may

be based upon a single criterion or upon a number of criteria (Grigg

1967:464). The objective of regional analysis is to understand

spatial associations within the confines of a specified geographic area

and eventually within a hierarchy of larger areas (Berry '1964, 1968;

Isard 1956; Haggett 1965). its methodology utilizes probability sam-

pling andmultiv riate mathematical techniques for the identification,

interpretation, and explanation of spatial organization.

Recent applications of regional analysis to archaeology have

resulted in a trend toward replacing the single site as the .nit of

archaeological analysis with d larger entity, e reion, t resu!-y,

the importance of a single site cannot be sufficiently understood

taken out of its regional context (Piog 1978). iewis Binford was the









first to define this usage of regional archaeology:

the detailed and systematic st-'y of regions that can be
expected to have supported cultural systems. The extent of
such regions will vary because it is recognized that cul-
tural systems differ greatly in the limits of their adap-
tive range and milieu. As cultural systems become more
comp ex, they : rally spai greater eco ,lo gical ranges and
enter more complex, widespread, extra-societal interaction.
The isolation and definition of content, the structure, and
the range of a cultural system together with its ecological
relationships may be viewed as a research objective.
(1964:426)

Central to the above definition is the view that culture is an

adaptive system. As an anthropological concept, adaptation refers to

the ability of a population to adjust to the environment by develop-

ing effective social and technological structures. Adaptation is an

ecological process (Steward 1955:30). The interaction between human

behavior and the environment in which culture is the rmediatirg

variable is fundamental to ecological approaches in anthropology

(Harris 1968:659). Cultural ecology as a research strategy. involves

three procedures: first, it analyzes the interrelationships between

technology and the environment; second, it analyzes behavioral patterns

involving the exploitation of a particular area (the region) by a

particular mode of technology; third, it ascertains the extent to

which the behavior patterns involved in exploiting the environment

affect other aspects of culture (Steward 1955:40-1). Th utility of

regional analysis in cultura-l ecol-:' is very clear. As the locus

occupied by a cultural system, the region provides the unit of analysis

for ascertaining huwan ecological relationships.

Because the region is the locus of human activities, regional

a.rchaeology requires the investigation cof diverse activity areas within








the geographic area occupied by a cultural system. It assumes that 'the

region, not the single site, embraces the total range of behavioral

variability for a population. The single site, on the other hand,

reflects only those activities that had taken place at that

specific locality (Streuver 1971:11). Thus the goal of regional

archaeology is to determine patterned interrelationships among sites.

Regional archaeology attempts to understand the cultural and

environmental processes richh have affected behavioral patterns. Its

application to historic site archaeology is slowly emerging. Specifi-

cally at historic sites, artifact pattern recognition (South 1977,

1978) has been suggested as a technique of discerning behavioral

variability within a region or time period. Ultimately, pattern recogni-

tion seeks to identify by means of archaeological data the functional

and behavioral processes which have taken place at sites (South 1378:

223; e.g. Lewis 1977).

Besides a unit of analysis, regional archaeology also s.pp.lis a

method for the study of site location. Locational analysis examines

spatial relationships about ecological or cultural phenomiena

(Ha0 gett 1965) A developing focus of locational analysis is its

application to historic sites (Swendlund 1975; Langhorne 1976; House

1977). Such studi es attempt to determine the importance of c ;tural

or natural resources within the functioning of a historic cultural

system (House 1977:243-4).

A third concept of regional archaeoliogy establishes a fame-

wourk for the coi election of cultural data for a defined area. nhe

regional plan or overview is the first phase in multi-phase








archaeological research. Frequently, regional plans are used as

planning devices in cultural resource management (King et al. 1977:

145-173; McGimsey and Davis 1977,:47; Schiffer and Gummerrnman 1977;

121-131). As planning tools, regional plans involve the review of all

known records of the project area, including historical records, sur-

vey and excavation reports, and other existing field data as well as

informant consultation. The overview should summarize the present

knowledge, evaluate the available evidence, estimate the resource base,
and attempt to forecast long-range regional development and ongoing

destructive processes (Schiffer and Gummerman 1977: 12-13). Although

regional plans are only as good as our present-day knowledge permits

(Glassow' 1977), their utility as rudimentary planning tools is essen-

tial in historic preservation.

From the above discussion four concepts of regional archaeology

are identified. The first is a unit of analysis which represents

the broad cultural milieu of human activities for a cultural system.

The second is a method of discerning spatial organizations within a

cultural system. The third is a "--.-.. .rk for tihe collection and

-.* -.i s of cultural and ecological data within a defined geographic
area. Finally, the fourth is the explanation for the patterns thus
-:T- ned.

In this study, the first, third, and fourth concepts provide the

orientation for analysis and interpretation. .Coastal Georgia, speci-

cally the barrier islands and the adjacent river deltas, forms the

geographic p.raweter. The temporal parameter is the nineteenth cen-

tury (1800 to 186 1 ) he unit of archaeological analysis for the

investigation of the slave system is the slave community site.








TIe slave vi llag or community site was the locus of most slave

activities. Admittedly, slaves spent considerable time in the fields

and at other locations of the plantation complex. But it is doubtful

that these sites will provide archaeological data relating to slave

behavior. Another site perhaps central to the slave belief system is

the slave graveyard. Because of the sensitivity involved in excavat-

ing burials, this archaeological resource is not treated here. Slave

graveyards, however, were at least located at Butler Island. Because

of the restricted movement enforced upon slaves, the slave village site

should embrace a major portion of slave behavioral variability. It

also preserves evid-nce for a broader range of cultural activities than

does a simple activity area such as a rice field or threshing floor.

A methodological departure of this study from regional research

strategies advocated by Einford and others is the degree of use of

probability s',.,ling at all levels of analysis (Binford 1964). Proba-

bility sampling is used in this study but minimally. Additionally, the

mulitivariate techniques of geographers are not used at all. Admittedly,

this methodological departure may present some serious disadvantages

in offering regional interpretations. The purpose of this study,

however, is to synthesize the presently available data recovered from

slave sites in coastal Georgia. The irn terpretations offered here

are tentative and, hopefully, wil e tested in future regional studies

utilizing regional methods.




Defined cultural, spatial, and temporal variable s form t.

parameters for most discussions of Afro-American slavery.










Cross-cultural comparisons of slavery are often concerned with the

issue of slave treatment within the confines of political entities

(e.g. Tannebaum 1949; Davis 1966; Degler 1971). Similarly, in the

United States discussions of slavery are Confined to geogrc.,ic areas

such as states (e.g. Flanders 1933) or time periods (e.g. Stampp 1956).

Both comparative studies and those within the United States are fre-

quently concerned with slave manaq.,-_...,, or the legal aspects of

slavery. Less often has the issue of slave behavior been addressed.

When it has, the entire slaveholding South has formed the unit of

analysis (e.g. Blassingame 1972; Genovese 1974). It is questionable

whether these discussions are applicable to slavery everywhere in tIhe

Old South.

Although the slave cultural system which developed in the South

was characterized by general behavioral patterns (Gutman 1977), local

variations of this tradition were inevitable. Evidence of this has

been indicated in r.. -y localized Afro-American traditions, some of

which have survived until the present day. The coastal areas of

Georgia and South Carolina have been focal areas of African retention

(herskovits 1958:120), and in the development of distinctive Afro-

American traditions, particularly in the decorative arts, language,

cuisine, and music.

Certainly the quality of slave life must have varied regionally.

Although this acspmption remains untested, indications from arcoeu .ogi-

calI resources are that the mati.erial standard of slaves in coastal

Georgia was better than that of slaves documented elsewhere in the

Old South (MacFarlane 1975; Otto 1975). Further, a recent study of









slave nutritional adequacy indicates that slaves in the coastal areas

of South Carolina, Georgia, and northeast Florida were adequately

nourished (Gibbs ot al. I'1-). These studies demonstrate that regional

approaches in the study of sia'.,,' are needed.

It is possible that certain economic, environmental, and demo-

graphic conditions in coastal Georgia favored a distinctive develop-

menit of slave behavior. In Georgia, the largest percentages of blacks

to whites were found on the coast. In addition, the most valuable farms,

the largest slaveholding, and the most livestock were located there.

Another factor was the value of lands and buildings per farm. This

was highest on the coast than in E part of Georgia (Flanders 1933:81).

Ecologically, the region provided abundant plant and animal resources

which were exploited for both domestic and industrial purposes. Employ-

ment of the task slave labor system on most coastal plantations allowed

slaves more time for recreation and to improve their material lots

than the gang labor system employed elsewhere in the Old South (see

Otto 1975). For the most part, these conditions were not present in the

interior. It is likely that variations in these conditions produced

variations in slave behavior.

Historic factors were also i :.i..nt. Unlike their counterparts

of the short-staple cotton belt, most planters that came to coastal

Georgia were experienced slaveholders from the West Indies or South

Ca'rolina (Vanstory 1970:74-79; Davis 1976:132). They were presumably

acquainted ,witi the problem1s of sl.av'e m'ingemntl. Conceivably, they

attempted to establish the best possible conditions for a "stable

regime within which their slave could live" (Genovese 1974:6). Perhaps,








this was why the plantations of the Georgia coast have been described as

"patriarchal" (Wylly 1910:12; Lewis and Huie 1974). Another contribut-

ing factor was the infrequency in which planters of the coast migrated

to the new lands of the West. With a few exceptions, slaveholders on

the coast of Georgia established their roots in the late eighteenth

century and remained there until the Civil War. It was "only in the tide-

water dreas of Georgia [that] wealth [had] matured for several genera-

tions" (Boney in Coleman 19:7174)1 This may have been the result of

special local ecological factors.

This study assumes that the previously described variables sub-

stantially influenced a development of nslave behavior patterns pecu-

liar to the Georgia coast. Some of these patterns are believed to be

archaeologically discernible. These archaeological indices a.re dis-

cussed in detail in subsequent chapters.

1. fl'1 TP -n i n

Archaeologists havw become increasingly involved in te investiga...

tion of sites formerly occupied by black Americans. Because few his-

toric sources have been written by blacks, this research interest

developed out of a need to expand upon traditional interpretation of

past black American 1ifeways. Contemporary eighteenth and ninetpenth

-c... accounts relating to black Americans were written by whive-s.

As a result, these are often very prejudiced views of Afro-Ar'erican

cultural life.- The major objective of Afro-American archaeology hias

been to investigate aspcts of black A..ricpn life not valuable in


documents.









With a few exceptions, archaeological studies of black "ricans

are of two types: slavery sites and northern free black sites.

Because of the preliminary quality of this research area, most of these

studies have been more descriptive than interpretative. Within this

year, however, an attempt has been made to synthesize the findings from

several Afrc-L- -."ican sites (see Sc.' .ler 1980). From these studies

and other published works, two themes dominate this research interest:

first, the search for African retentions in black American material

culture (Schuyler 1980:2), and second, the recognition of Afro-American

subsistence patterns.

It was the search for material correlates of African survivals as

well as subsistence information which initiated preliminary testing at

the Kingsley slave site by Fairbanks in 1968 (Fairbanks 1974). Un-

fortunately, "no surely African elements" were identified (Fairbanks

1974:90). Since that time studies of slavery sites in the Old South

have been directed toward the definition of subsistence patterns

associated with the socioeconomic status of slavery (Ascher and

Fairbanks 1971; .: -arlane 1975; Otto 1975; Drucker and Anthony 1979).

In.estigatons of slave burials in Barbadoes have provided very

convincing evidence of Africanisms (Handler and Lange 1978, i979).

At the Newton planation, the orientations of save bodies and the

associated burial goods are indicative of West African burial patterns.

Burial practices peculiar to Afro-Americans in the United St&ates hav,,

been reported but only at sites occupied by freed blacks (Cate and

Wrightman 1955;:207-215; Combes 1974; Crosby and 'Emerson 1979).








Perhaps, the most striking and yet controversial evidence (see

Schuyler 1980:2) of Afro-Americans attempting to recreate their African

past has been uncovered at the Parting O'.ys site (Leetz 1977:135-154).

This site near Plymouth, Massachusetts, was occupied by four Revolution-

ary s'a-r veterans from the 1ate eighteenth to the mnid-nineteenth centuries.

Deetz suggests that the settlement pattern, huuse floor plans, a utii-

tarian earthenware, and possibly the culnnery practices are derived from

an African past.

The recent suggestion that black ; -,icons influenced and miranu-

factured some Colono-Indian ceramics (Ferguson 1980) is another example

of this search for African survivals. These ceramiriics occur in highl

frequencies at seventeenth-and eighteenth-century southern sites, Speci-
fically in the coastal areass of the Carolinas ( Fer. ..- 1980). Previous

to this SuC.-:,tion these cera.mics were though to have -been made

exclusively by native Americans ( '..-1-Hume 1962). It is well documented

that some of these ceramics 'were made by native Americans and some are-

still made by them today. Indications that blacks may have had a hand

in the mranufacture of these ceramics include the possible similarity of

this ceramic style with West African ceram ic traditions, the inre-

quency of Colorno-Indian ceramics at historic Indian sites, and its

high frequency at black occupied sites (Ferguson 1980). But the

fact that Colono-Indian ceramics have been uncovered, in most cases,

where native American populations were nearby* o.kes the suggestion

that blacks manufactured these ceramics uncErtin.



*Cha rules I. Fairbanks, personal communication 1979, Gainpsville,
Florida.








The archaeological investigation of African survivals at Afro-

American sites has resulted in very little definitive evidence to date.

This may be the result of a number of factors. Although his discussion

is specifically directed to slavery sites, Otto details three processes

that limited the re-creation of African materials. These include

(1) selection arid simplification, (2) availability and substi tu ion,

and (3) differential acculturation (Otto 1975:375). It is very likely

that northern free blacks also substituted traditional items vUith

available ones. The archaeological record is another factor. African-

styled artifacts such as wooden objects or basketry are not likely to

be preserved (Otto 15:382). Regardless of the reason for the scarcity

of African-styled artifacts at black-occupied sites, research designed

solely to uncover Africanisms ignores the fact that Afro-me.rican

culture, like all cultures, is an adaptive system (Schuyler 198.:?).

A data base of Afro-American subsistence patterns is slto,-,y emerg-

ing from diverse archaeological resources. These data are crucial
to gener+atng future hypotheses regarding Afro-American cultural

patterns. A major problem has developed in the idlen.ti.fication of

Afro-American economic cs. To what extent are the subsistence patterns

at black sites indicative of Afr,- .',ican ethnicity or a culture of

poverty (Kely and Kelly I-)? Only when archaeoogical data

are available from poor white subsistence farmers can the archaeo-

logical visibity of low-status Afro-Americans be ascertained

( Baker 1978) Presen ty, these nonplc ntc s ares archa nc& i

uLnknown.









The archaeological evidence of slavery, however, may not be as

difficult to reco;iize as are other Afro-American sites. Recently,

an excellent attempt was made to identify an undocumented site as a

slave site at Spiers Landing in Berkley, South Ca rolina (Drucker and

Anthony 1979). Although Hanidler and Lange (19783:228) conclude that

archaeological data do not identify the slave status or slavery,

the Spiers Landing exai;ple refutes this. To suggest that slavery sites

can be identified entirely from, archaeological remains- is very pre-

mature at this stage of research. This objective, however, way be

realized in the future through the use of regional approaches to Afro-

American sites, particularly the utilization cf pattern recognition

techniques.

Finally, in their discussion of ethnic idenrtiiication in his-

torical archaeol ogy, the Kellys recommend the use of regi onal research

designs (1980:135-5). They suggest that research designs geared toward

the recognition of ethnic cultural patterns within a defined region may

aid in the delineation of geographic variations of a broad cultural

tradition (1980:130). The delineation of a geographic and tIemporal

variation of a broad Afro-American cultural tradition is precisely the

objective of this study.




From the available reports, it appears that slave sites are known

better archaeologic ally in coastal Georgin than in ,' other region

of the Old South (see Figurie 1, for the location of these sitss. De-

spite this knowledge, these studies have suffered from three "mjor



























K..'


->2


~

C) c) 2D


'1 ,, "' ,





._. ,_...4s."



-~ ..*






i :
'/1 7


^--.^"- "'* '- -. ,f ^


,-.--* I



~

7.
V '7~4-~ ,-







K J~V~-K 1~iJ~'' ' ''/ L~1E~' 1(1 L4t7,1 ~ jji.


..i ,- ,









limitations: first, with the exception of one study (MacFarlane 1975)

excavations of slave settlements have sampled only a small portion of

the site, usually one or two structures (Ascher and Fairban ks 1971;

Fairbanks 1974; Otto 1975). Second, as a result of this sampling,

archaeological evidence of the internal social organizationwithin the

s lave community has not been uncovered. Third, slave sites investi-

gated have been exclusively from long-staple cotton plantations. Con-

parative data from the other dominant cash crop of the coast rice, have

been lacking.

Investigations at the south end slave cabins at the Cannon's

Point plantation on St. Simons Island, Georgia ([i1acFalralne, l975 were

the first attempt to excavate an entire slave community site.. This

investigation provided details of slave housing, household itens, cloth-

ing, personal items, and food resources. Little or no data, however,

were reported reg riding g the community organization, special i7ed crafts,

or social hierarchy.

Documentary resources of the Butler Island rice plantation indi-

cate that each slave settlement had plantation drivers, craftsmen, cooks,

as well as field laborers. it was hypothesized that testing an entire

slave cormmuni-ty should uncover archaeological evidence of these special-

ized roles and status differences. Initially, the objective of arcraeo-

logical investigations at B:utler Island was twofold: firs-:, to uncover

a.rchaeological evidence of status difference and craft specializations

withinn the slave _o .i uiy; seco d, to compare; the a.rch.-eo-ogical re-

sources of slavery at a ri e) p1 antatio. wtith o v hose ni o.g-s ap C c t n

plantations, in an attempt to ascertain similarities and differences









in slave life from two cash crop sites within a narrowly defined region.

It became apparent in the field that the time needed to investi-

gate the first objective was insufficient for the time and funds

available for fieldwork. As a result, attention was then directed

toward uncovering archaeological evidence associated with the slave

community plan. Presumably, settlement and certain subsistence data

would provide archaeological evidence for the special adaptations of

slavery to rice cultivation. At the same time, comparison of the

Butler Island data with long-staple cotton plantations would supply

additional evidence of slave life which exceeded adaptations to cash

crops. In other words, the comparison would provide data specifically

related to the economic function of the plantation on one hand and

data which were not so related on the other hand. Further, it is

hypothesized by the author that similar artifact patterns occurring inI

both crop loci are indicative of the material conditions of slave life

and slave behavior patterns in a larger coastal Ge Q-. a area.

The factors which are believed to have influenced a distinctive

development of slavery in coastal Georgia have been briefly discussed.

It is important, however, to define more accurately their role in the

development of coastal slavery. ThIe demographic, economic, end

historic factors (see pages 9 and 10) are all related to prA:ntation

management practices. These management practices developed out of an

adaptation to the coastal environnt of Georgia. Although the habitats

for the cultivation of rice and lon.g-stipIe cotton are very dissimilar,

both habitats are part of a larger ecosystem (Johnson et al. 1974).

Thus, this is a study of adaptation on two lvels: adaptation to










specific crop requirements and adaptation to a more generalized coastal

environment.

The purpose of this dissertation is to test the following

hypo theses and their ipli cat ions:

!,thesis A: Slave co.muniLty and household patterns in coastal

Georgia reflect adaptations to the specialized habitats where tidewater

staples were produced. Archaeologically this will be identified in the

fo lowing:

1. Patterned placement of slave village sites relative to

habitats.

2. Community plan of spatial arr" -; its

a. Specialize'd bildings and materials culture

b. '- tructural features

3. Slave dwellings

a. Size and available space

b. Construction details and materials

4. Farming implements and specialized crafts

5. V. i: tic plant and animal resources exploited

Hypothesis B: Slave material conditions and behavior patterns

reflect managemenL practices adopted for coastal Georgia. Archaeologi-

cal y these will be identified iin the followi ng househod artifact

patterns:

t. Food-r .lated activities

a. Fool preparation equipment and techniques

b. Food ser'v.'ing equip .: t

c. Oomestic plant and animal food resources










d. Evidence of food elementsment s in plantation rations

1. Equipment for producing food

2. Equipment for procuring wild food resources

3. Food processing equipment

4. Remains of food supplements

2. Personal possessions

a. Clothing

b. ,::.--.sories and ornaments

c. Household items

3. Leisure-time activities

a. Smoking equipment

b. Games and t:-. s

c. Miscellaneous items

These hypotheses and their underlying assumptions are discussed

in greater detail in later chapters. Chapters 2 and 3 provide back-

ground information on the coastal environment of Georgia, the produc-

tion of rice and long-staple cotton, and the archaeological and his-

toricaI investigations at Butler Island. Slave community organizations

at Butler Island are discussed in Chapter 4. Chapter 5 describes the

slave material culture recovered from Butler Island. Slave artifact

patterns derived from statistical pattern recognition techniques are

presented in i,.,.ter 6. i-,.pter 7 discusses the results from testing

the hypotheses and the conclusions for the study.














C.H/', !.:, 2
PL'ATION AGRICULTURE IN THIE rF7.GIA TIDEWATER


Envi ronment.al 7".-.--, y -, ,' T, -...'. .;- ,,
Tidewater refers to the geo '.hic area of the coastal plain,

which is affected by the 'influence of the tides. ii Georgia, this

region includes the barrier islands and extends approximately 75 kilo-

meters inland to the point where extensive areas of pine forests begin

to occur (Crook 78-28-30) (see fi re 2). Three jor subdivisions

characterize the Georgia tidewater: the strand section, the salL Iarsh

and lagc-i section, nd the delta section. These are discussed in de

tail elsewhere (Larson 1970; Johnson et a!. 1974), and are briefly sum-

marized here.

The strand borders on r.- ocean ea d is composed of beach and

dunes. it is the most rapidly changing of the three sections. Vegeta-

tion is restricted and m;3st animal resources which occur there are found

aore frequently in other habitats along the coast.

Of the three s'bdi visions, the salt marsh and lagoon section is

the most ecologically diverse. It is composed of four habitats: marsh,

tidal creeks, hamrocks or highlands, and freshwater swamps. As a

result of its diversity, this section is rich in both plant and animal

resources. The food chain begins in these salt marshes. Several

species of freish, s-e i-k, tIr lcs, ned wtcrfowl thrive in the norm,

and tidal creeks. The hacimock forest community includes several



















































Figure 2. The Physiographic Regions of Georgia.
(Adapted from Coleman 1977.)









species of oak, p-gnuK hickory, red cedar, southern magnolia, cabbage

palm, wax *tle, saw palmetto, and many herbs and vines. Faunal

resources in these highlands are also abundant. .Kitetail deer,

oppossum, raccoon, and cottontail rabbit are among the dominant species.

The freshwater swamcp is an important breeding ground for re pti. les and

amrphibiarns, and the dominant vegetati on is cypress.

Fewer data are available for delta ecology h than e other sections.

The delta embraces those areas where freshwater rivers enter saltwater.

Deltas are primarily freshwater but are affected by the tides. The

area is flat and is frequently inundated. :.,t vegetation is water-

tolerant. Near the ocean, the area is a grass-covered marsh; where

there is little or no intrusion of saltwater, the delta is co'mposed of

c :;;s swamp (Larson 1970:35). Much of this swamp was cleared for rice

cultivation. Deer, otter, and raccoon occur in portions of the delta,

but the most distinctive faunal resources are the anadromous fish such

as glut herring, striped bass, sturgeon, and the ..-ican shad.

The narrow belt of the tidewater, which includes the barrier

islands and the adjacent river deltas, is part of the oldest agricultural

region of Georgia. The sandy loams of the sea islands were the -t

productive lands" of the coast. During the colonial period, these

lands yielded crops of corn, inrdi go, potatoes, and vegetables. After

the Revolutionary War, these lands were devoted al!;ost exclusively to

the production of long-staple cotton (Bonnrer 1964:-).

By the -late eighteenth ri, rc in he li

deltaic soils for rice cultivation hid begun. Tideater rice culture

was limited to ,the areas coated far enough upstream to avoid contact










with saltwater but close enough to the ocean to be affected by the tide

(House 1954b:23). The restricted spatial limits of tidewater rice cul-

ture made the alluvial soils of the delta the "most valuable" of this

old agricultural region (Bonner 1964:2).

In general, most plantations on the barrier islands were devoted

to long-staple cotton and those of the delta to rice. Some rice, however,

was grown on the barrier islands, presumably in the freshwater swamps,

and some cotton on the delta. Sugar, a third important cash crop (see

Sitterson 1937, 1953:31-35), was frequently produced at both, but in

smaller quantities. For example, at Butler Island, considerable quanti-

ties of cotton and su;:r were cultivated., although rice was the island's

.1j'or cash crop.

The purpose of this chapter is to review the historical develop-

ment of plantation agriculture in the Georgia tidewater, specifically as

it relates to the production of rice and long-staple cotton, the two

dominant crops of the post-Colonial period. Also, planting and process-

ing methods and marketing for each crop are described. Rice culture is

considered in greater detail for two reasons: It was a more involved

process than that ocf cotton culture and a detailed description of this

culture is central to understanding the archaeological resources of

Butler Island (long-staple cottonis treated extensively in another

archaeological report-see Otto 1975:51-61).


The EEnmeroence of the Pl ttion Systes iS G

On June 9, 1732, the developers of Georgia, known as the trustees,

were granted a charter for the lands between the Savannah and the








Altamaha Rivers. The trustees desired to establish a colony with small

settlements of independent farmers (Bonner 1964:2). To obtain this

objective, the trustees prohibited large land grants and Negro slavery.

The plantation system was outlawed for several reasons. First, the

colony was purportedly founded for debtors, although debtors made up

fewer than a dozen of the settlers during the entire trusteeship period

(Spalding in Coleman 1977:18). A large portion of the early settlers to

Georgia, however, were small businessmen, farmers, tradesmen, and

unemployed laborers (Coleman 1976:36--54), who could not be expected to

c.:-,-;la: with a planter class. Second, Georgia was selected by the

trustees to produce commodities needed by England, such as silk, which

was not adapted to plantation agriculture. Third, the colony was to serve

as a military buffu;. to protect Carolina from the Spaniards in Florida.

Plantations would spread the people over the land instead of no :t-.-at-

ing them into settlements. Moreover; a large slave population could sub-

ject the colony to slave insurrection, making it vulnerable to Spanish

attack (Potter 1932:116-117; Calloway 1948:30-31).

From the '.-,;y beginning some colonists objected to the restrictions

placed upon Georgia. These "malcontents" were particularly displeased

with the prohibition of slavery. They were convinced that the use of

a cheap labor force was the only way to develop the colony. As dis-

satisfaction grew, many colonists abandoned Georgia. On the brink of

economic distress, it became apparent to the trustees that if the

colony was to succeed it must be ma de attractive to potential colonists

(Davis 1976:126). Finally, in 1750, slavery was made legal. But,

discouraged by their failures, the trustees surrendered their charter

to the British Crown in 1752 (Spaiding in Coleman 1977:44).








Under royal gc vernment, the colony experienced an accelerated

growth A sig nificant source of immiigration to the coastal area was

the South Carolina plant er. South Carolinians were aware of the rice

growing potential of the Georgia tidewater o reas, nd they mi orated to

Georgia wi th their famii lies and s av es (Wilms 19/2); rice soec n he a

Georgia's most i",portc;nt crop. Later, the introduction of indigo gave

the plantation system an added boost (Spalding in Coleman 1977:52).

Planters were attracted to the coastal lands, but good lrnds 'were

not abundant. -.iver, the develop -;t of these lands was very costly.

An eighteenth-cen' ., observer in coastal Georgia estimated that Vie

initial expense of settling a rice estate was about $10,000, and, for

cotton or indigo $9,OCO (Thayer 1957:81). Consequently, only large,

wealthy pa nters could afford to acquire and cultivate these lands. The

small planter unable to compete settled further in the interior. Thus,

coastall Georgia became a region of a few, wealthy planters shortly

after the e nrgence of plantations to the area (Calloway 1948:47-49).

Te Rev olutionary War b. Lht disastrous setbacks to the coastal

plantation eco .... The bounty on indigo formerly received from the

B.ritish cvernent ceased and -the e::xporta.tion of rice declined. Pros-

perity was restored, however, with the successful introduction of Iong-

staple cotton in 1786. Plnters eof rice plantations acquired land and

equipmEnt for cotton production. Thus, rice w'as te:mporacrily dis-

placed as a -jor staple of the tidrewater (Fl anders 1 933: 55-57).

E, the mid-ni.net' nth century, the rel.,nce upon sea-is!F"d
,- .. .. : ,, K 1. !,... *. 5 --, -U.' .'0 e*'l
cotton i rdual.y OfEM to., ricu Appa irenlry, this change iou(C'{y

coincid,- with periods nf depressed long-st-p-o cotton prices (Gray









194'1:38). During this time of renewed rice interests, Carolina planters
intensifiedl in re se "- ..1 '
intensified their activities in Georgia. They not only increased their

acreage in the old rice producing areas along the Savannah (see Clifton

1978) and Al tan'ah ,,Rivers, but also expanded further southva rd to 11clurIe

the Sati11a and St. Mary's Rivers (see Figure 3). In the 1850s, cotton

prices were on the upswing (Gray 1941 :739) and, rice1 cultivatLion w;as at

its peak (Hilliard 1975:65). Planters producing both crops were enjoy-

ing sizeable profits on the eve of the Civil War. Devastation from, the

Civil Wfar prevented the successful revival of either culture in the

post-war years. The reign of rice and long-staple cotton as the dominant

cash crops of the coast had come to an end.


The Rice Industry

Early experiments with rice cultivation were attempted in

clonial Virginia, but rice was first established as a stale in the
South Carolina low country (Gray 1941:277). Although the details of its

introduction remain unclear, apparently early promoters of the colony

encouraged rice cultivation shortly after the founding of Carolina in

1670. By 1695, South Carolina's rice industry had begun (Saiiey 1919).


Rice and Slavery

Pt.nhods of ric c ul-tivation during most of -the plantation caera

tir ~. rpendent upn, hand labor. As a resu,,,,t, large-scale rice pro-

dc tion required a enormous labor force. y,, the mid-nine.eenth cent ry

ep-,o ';a-teley i"5 la -orers rwee needed to clt. i vat "" acres of ri c.e

land or 6 to 7 acres per haEnd (House 1954a:151). Presumably, colonial

practices demanded greater 'labor requirements. For example, the average.



















__, - __
^ } i.
i ,f o u *;., .

O^.- h r-.; A r '-
'/^
Ri' : 72


2.


-.


~1
a 0


K..


Fl.',, -) F 2


Fi Cure 3. enutrc of Rice Cui vi- t-ionn in Georgia hy 1860.
(Adapted from Hitliard 1975.)


...... R ;w Is










rice plantation in colonial Georgia had an average of "3 working hands,

but the average acreage was not recorded (Flanders 1933:43).

Given this labor requirement, a demand for African slave labor

developed. Prior to 1695, most slaves in colonial South Carolina had

come there with their masters from the West Indies. These Afro-Americans

made notable contributions to settling the frontier of South Carolina

(Wood 1974:27-34). It was the rice industry, however, which estab-

lished a need for black slave labor. As the new staple became profitable,
considerable quantities of Africans were imported (Wood 1974:57).

Africans were thought to be better equipped to labor in rice fields than

either indentured whites or Indian slaves for several reasons (AWood

1974:37-42). Perhaps, two of the most significant were the African

adaptability to the swampy lowlands and the African knowledge of rice

growing.

Malaria thrived in the stagnant, swampy lowlands where rice was

cultivated (Childs 1940:32). Settlers to the Carolina-Georgia coast

frequently complained of the prevalence of "fevers." Malaria or

'"intermittent fe .," was considered to be the worst among these, par-

ticularly during its pernicious season in the Spring and Summer (Childs

1940:33).

Only in recent years has medical research shown that populations

exposed to malarial climates display a genetic trait which produces a

part al but heritable immunLity to the disease. This trait, referred to

as "sickle-ce1 trait," is not harmful in itself, bu.t if present in both

parents can produce offspring who suffer from sickle-cell anemia and

are likely to die in adolescence (Wood 1974:80; Savitt 1978:27-?8).






29


P-,,le who are heterzygous for the sickle-cell trait are evidently

.;,or able to survive and reproduce in a malarial region (Weiss and

'-.,i 1975:356). In the United States, studies have shown that a high

incidence of the sickle-cell trait presently exists SImc. the "CGullah"

Negroes of the South Carolina and C- .:'ia coasts. This fi ending sug-

gests that Afro-Americans in these former rice-growing areas have per-

petuated this genetic characteristic (Gun n 1975-5).

Given the medical knowledge of the day, it is doubtful that rice

planters understood the sickle-cell trait. It appears, however, that

some planters observed fairly early that blacks were less likely to

suffer from malaria and other diseases prevalent in swampy lowlands

than whites (Postell 1951:74-75; Savitt 1978:22-24), which established

a preference for black labor in rice fields. Thus, it is clear that the

SNegro's adaptability to the coastal environment partially contributed to

the tradition which defended Afro-American sla,... on the basis that

blacks were more suitable to perform work whites were unwilling to do

(Gunn 1975:6).

Another possible explanation for the preference of African labor

in the cultivation of rice ;as the African knowledge of rice growing.

Historic ns have ov erlooked this possibility until recently. Primarily,

through the pioneering efforts of Peter Wood, a preliminary investiga-

tion of this suggestion has been undertaken (see Wood 1974:55-62).

Rice planting and processing invol. ved obhstcles to Eiropeanls,

since they were generally unfamil iar with rice culture. Unlike

Europeans, Africars from several regions in Nest Africa were successful

rice producers (Wood 1974:58-59). In fact, some West Africans were









cultivating rice as early as the fifteenth century (Lruer 1969:17).

A preference for slaves from rice growing regions of West Africa was

established early ',' white Carolinians. These areas included the

Congo-Angola area (iHe -:d 1932:172-175), Senegambia area (Clark

et el. i9G2:225-239; Lauer 1969:47-48), and the "'indwrd" or "fold"

coast of present-day Sierra Leone and -n',', (Corry 1960). Slaves from

the rice coast of Africa were advertised by many firms. Between 1732

and 1765, C' rleston alone handled "53 ship loads from Gambia rnd 29 from

the Gold coast" (Lauer 1969:48).

Additional evidence for the presence of slaves from West African

rice growing areas is provided by the prevalence of Afric-n personal

[naris. The names ".anyun,' "Bram," and "Bu'l,1unda" among slaves and

their later descenden;ts on the Carolina-Georgia coasts strongly su-<,..

that the peon'e ', the rice growing areas of Sene-.aia were repre-

sented among the slave imports (Lauer 1969:48).

Besides the actual physical labor, Afro- -ricans made signifi-

cant contributions in the planting and processing of rice. The moti ons

used in planting rice seeds such as the pressing of a hole with the

heel and covering the seed with a foot were also practiced in West

African rice planting. The flat, wide, coiled winnowing baskets were

nade by slaves and were typical of West African ones both in stlve rnd

techno ques of m nua.ctu-re. ..hods of pounding and polishing rice by

-.nd were possibly told to white planters by their African slaves

(Wood 1974:61-62).

Other factors, ho wtver, ay have establish od a need for African

slave labor. Briefly, these include the abundance of black abo.r,







the difficulties posed by the enslavement of Indians, and the insuf-

ficient supply of indentured whites (Wood 1974:37-48). Despite the

fundamental role these factors contributed to the existence of African

enslavement in the New World, it is suggested that the practical advan-

tages offered by African physical adaptability and African knowledge of

rice growing fostered the Carolina-Georgia rice industry.


,jhds of Irrigjat-ion and Plantint Rice

Two different species of rice were cultivated in the Southern

United States: lowland rice, Oryza sativa, and upland rice, Oryza mutica.

Lowland rice, a marsh plant, required an abundant supply of water for

ripening (Austin 1893:9). Of its many varieties, "coldseed" rice and
"white" rice were cultivated in the lowland areas of the Atlantic Sea-

board (lQ':pp 1900:5). Upland rice was grown by dry culture in various

hilly areas of the cotton belt. It was not a commercial crop, but

grown essentially for home use (Gray 1941i:723).

Lowl.and rice was probably first grown in moist soils without

irrigation (Gray IS417229). ; .....j 1724, irrigation was employed in

what were called "inland s.saps." ;: .e were small fields adjacent

to freshwater stre.ams. The rice fields were irrigated with water

stored in reservoirs frrmed by dams. After irrigation, the water was

drained through ditches into streams. Because the water -. -ly in the

reservoirs was dependent upon rainfall, the crop was subject to either

frequent floods or drought, often making yields small and uncertain

(He rdJ 1937: i?...] ;-..;, s "9,72:50-51).
By the iate eighteenth century or perhaps earlier (see Wi ins

1972), rice planters began to recognize the advantages of cultivat ing







32



on the tidal swamps to permit the inflow of freshwt:.r backed up at

high tide and the outflow at low tide. The fields could be flooded and

drained as desired (Cole 1927:599). But before the tidal marsh could

be cultivated, it had to be cleared, diked, and drained. Reclaiming the

tidal swamp was a tre-me'.dou. s undertaking which required considerable

expertise and its completion often took several rs (Heyward 1937:

18-20; Wiilmrs 1972:53; Hilliard 1975:60).

The innovation of transforming a marshland into a piece of culti-

vable land, a polder, cannot be credited to the tidewater planter.

Construction of polderlands has been undertaken since ancient times,

and today they are located all over the world (Wagret 1968). Perhaps,

the best known are those of Holland. in fact, indications are that

several Carolina-Georgia tidewater planters employed Dutch expctise in

designing their hydraulic systems (Doar 1970:13; Scott 1973:73-74;

Johnson 1930:61).

Articles describing the construction of the irrigation systems

used by tidewater rice culture are found in several contemporary

journals (e.g. Aliston 1846, 1854; OeBovws Review i847; Spa-lding 1845).

Several secondary sources have provided additional details for this

procedure (e.g. Heyward 1937; House 1939, 1954b; Cray 1941; Doar 1970;

Hilliiard 1975). In very general terms the procedure for constructing

a tidewater rice irrigation system involved six major steps. first

t"I length end width of the required acreage was measured off aim t&

trees were cut down. and noved out of the way. Second, a temporary ditch

and embankment were built around the entire area to, keep water out, while

tree stumps and debris were removed; third, the ditch was filled and







elevated to form the perma ent L embankment. Fourth, small channels were

bridged and the trunks were installed. (Trunks were thie devices which

controlled the water flow.) Fifth, individual fields 12 to 22 acres were

laid out. (Each of these fields could be filled or drained individually.)

Sixth, these larger fields were subdivided into plots of 1/4 or 1/8

acre to aid the water movement (Gray 1941:726-727; Doar 1970:9-13;

Hilliard 1975:59-60). These smaller subdivisions, sometimes referred to

as "tasks" (Gray i941:734), provided not only drainage but also the

standard measurement for certain labor requi regents (House 1,: 153).

Thus, the basis for the task labor manar.:,-,,2. systEm, which was used

:extensively in the cultivation of tidewater staples, was derived from the

small segments formed by the irrigation system used in rice culture

(Stampp 1956:55).

Regular maintenance of the irrigation system was essential in

order to keep the fields in proper working order. The ditches and

drains were thoroughly cleared of mud and trash annually (Austin 1893:

18). Breaks in the embankments caused by storms, high tides, and

"freshets" or floods were frequent, despite precautions taken to pro-

tect the outer embankr"ents (SFpalding 1830; Souen the Agriculturalist

1834). in addition, small animals including rats, snakes, and crayfish

often made holes on the outside of the bank which very often became

leaks and then breaks. if left unattended, these small breaks would

interfere with the drainac.e system (Doar 1970:30). Such breaks and

eaks had to be mended, and "ditchers" or "bank menders" (Heyw.'rd 1937:

236) wee engaged r-eularly in repairin g dy'ks


too "HoWnn









Usually the -'1.-,l cycle began in the fall and early winter with

the "breaking" of the soil. This was done by hand with heavy hoes.

Plows were used by some planters, but the small subdi vided rice fields

and the numerous ditches made plowing difficult. .,reover, many fields

were not dry enough and draft animals were expensi ve to ma inLtain (Gray

1941:729; House 1954b:27-28).

Planting occurred beginning in the middle of .y, depending upon

the weather (Austin 1893:19). Techniques varied among planters, but

either of two methods was generally employed. In the traditional method,
"covered rice" seeds were carefully sowed in trenches measuring 3 to 4

inches wide, 2 inches deep, and 13 to 14 inches apart. After sowing,

the trenches were covered with 2 inches of soil by hoes, rakes, or

covering boards (Gray 1941:727-728). The other method, open trench,,"

required that the seeds be coated in clay before sowing. Upon comple-

tion of sowing, water was let in the open trenches without covering them

with soil (House 194?2a:188). Open trench planting was not widely used

in Georgia. It was apparently developed by South Carolina planters in

an attempt to increase yields from exhausted soils (House 1942a).

Once the seed rice was planted, inundation of the fields was needed
at various stages of growth. irrigation practices, like planting methods,

varied according to the planter. All irrigation, however, was utilized

for three purposes: the initial flow or sprout flow was applied to aid

the seed in sprouting; subsequent flows known as stretch flows were

applied to keep down the growth of grass, weeds, and other pests; and

the final or harvest flow strengthened and supported the stalks bearing

ripened grain before the rice was harvested (House 1954b:3).









A number of pests pnr,- I upon the growing rice (see Austin 1893:

31-37 for a detailed discussion). Two of the most frequently mentioned

were volunteer rice and rice birds. Volunteer rice or "red rice" was the

product of seed from the previous year which remained in the soil and

sprouted with the new crop (Austin 1893:23). A few of these red grains

among the white rice were believed to seriously lower the quality of the

crop. Consequently, planters carefully selected the seed rice and had

slaves to remove these self-planted sprouts whenever they were detected

(Gray 1941:729). Rice birds presented another serious problem for they

often consumed la. portions of the crop. These birds were unavoidable

since they appeared in May for approximately two weeks and then re-

appear.ed in September for the same length of time. Some planters planted

in between May and September, but yields were small (0' -.rd 1937-31-.33).

:st planters, however, employed "bird minders" to frighten the birds

Eway with fire, noise, or decoys (Heyward 1937:32; Gray 1941:729).


Harvewtin__ fillilf qI. and iarketin1

Rice was harvested by hand sickles in late ,,.,,ust or early

ptembe-r. The cut rice was allowed to dry for a day or so and then it

was tied together in sheaves. These sheaves were carried to the mill

o.r barnyard by flats which floated on the main canals or overland by

ox-carts (Spalding 1835). The rice was stacked in the barn until

threshing season (Pry 1941:729).

P'i- prin th, rice for the markt involved several processes:

''reshin"ci," the "f reoval of he rice from the plant; cleaning or
"winnow'ng," ',: separation of the grain from the chaff; "pounding,'









or "grinding," the removal of the outer hull and blowing it caway from

thoe sheled rice; finally, "polishing," which resulted in the clean wh te

rice grains plus powdered fragments, known as rice flour (House I954b:

59-60).

Threshing was done by hno.,d using flails for nost of the planta-

tion period. Sometimes a flor was constructed fr threshing; a.n

efficient" floor w,'as 110 feet long and 50 feet wide end composed of

three r .--- of boards (souse 1954b:60). Rice was cleaned usin-g w.ind

or a fan. More often a winnowing house was constructed for this pur-

pose. ,jinnowing houses were small, usually square-shaped buildings

raised upon stilts. The rice was dropped through a hole in the floor

of the house, and the wind blew away Lhe chaff. The grain landed below.

the hole on a clay or earthen floor. It was swept up and placed in

conOtainers (Smith 1936:20, and illustrated). Not until the mid-

nireteenth century was a reliable tlhreshing mil developed which

threshed and cleaned the rice (Gray 1941:729-730).

DTi colonial days, the outer husks of the rice grain were removed

by hand-po,,ered mil s arnd hand fans blew away the chaff. These "u, skiing"

machin-es consisted of siTabs of wood about 2 feet in diameter revolving

one on -.e other. The inner shell was removed by hand by use of a

rmortar and pestle (Smith 1936:21; Gray 1941:-287). Later in the nine-

leen-th cent-ury, mills were developed which combined aP I of the processes

of milling rice-grinidig, cleaning, pounding, screening, and polish-

ing (!-ay S,41:,730; Lw -_on 1975:11, illust'atin of the devices used in

a rice pounding in r; ). Power frr these mii s vwas supplied by t i des

steami engines, e animal power, or wind. -he s te. &!-powered mills were se e









most efficient (House 1954b:62-65). Plantations with threshing machines

often housed machinery for threshing, pounding, and polishing within the

same structure (Smith 1936:22; House 1954b:62). Only .. large planta-

tions, however, could afford to be fully ekipped with milling facili-

ties. The planter with potundin .g and polishing mil" could ship pounded,

clean grain to the Parket, whereas the planter without had to ship

rough rice, or "paddy." Also, the planter without a mill had to

rely upon toUl mills to complete the processing of the rice for a per-

centage of the grain. Several European buyers, however, preferred to

urch-ase rough rice and mill it in their own countries. Because of

this there was always a substantial market for rough rice. By 1850,

most foreign shipment was pad,:', and pounding mi1Is were utilized pri-

marily for the domestic market (Gray i941I:730; House 1954b:68).

World prices for rice were based upon polished rice and weight

(House 1954b:67). The commercial standard weight of rough rice was 44

to 45 pounds a bushel (Austin 1893:24; Doar 1970:18), and the average

yield of rice lands varied from 253 to 60 bushels of rough rice per

acre (Gray 1941:730). Polished rice vwas packed in "tierces," each

weighi j 600 pounds. The quality or grade of the rice was judged accord

ing to the number of bushels it took to make one tierce. Generally,

20 bushels or less of rough rice to make one tierce of polished rice

vas considered a good crop, wore than 20 an inferior one (Doar 1970:

18). Prices varied consider rably during the antebelumn period. The

best prices for Georgian rice, however, v.erc obtained between 1840 rd

186;0, or the "golden age" of Georgian rice production. At that time










rice sold for $.80 to $2.23 per bushel or 2 to 4.75 cents per pound

(House 171:78).

Rice, like all other plantation crops, was marketed through fac-

tors in major port cities. The factor served as the middlemann" be-

tween the planters and the rice merchants. He was the banker and stock-

broker for the planter and provided him with credit in meeting opera-

tional costs. Since the planter depended upon the re.:: from his

crop for cash, he was often in debt to the factor when rice sales were

low (Easterby 1941). Prior to 1840, most Georgian rice was sold thrc.,'

factors based in Charleston. After that date, Georgian planters in-

creasingly sold their rice from Savannah. By 1859, Savannah had 82

factor establishments. Savannah experienced a rapid growth as a rice

port. Its population increased; a thriving trade developed with ports

in the West Indies, Mobile, and "onj Orleans; and it became an i portant

social and cultural center (House 1954b:70-79).

The rice industry on the Carolina-Georgia costs was seriously

beginning to decline on the eve of the Civil Par, although large-scale

rice production did not vanish completely from the area until the end

of the nineteenth century. Competition from rice-growing states in the

southwest and the rising cost and scarcity of slaves (Cole 1927:600-

601) were the major causes for this new crisis facing the rice planter.

Nonetheless, the industry managed to survive because of large crop

yields. After the Civil i.ar, however, attempts to restore the former

pros-prity to the industry were futile. The new labor system arnd

increased flooding from freshets caused by accelerated erosion in the

interior (TFrimble 1969) made rice cultivation unprofitable. Gradually,








the interest in rice vwas abandoned. Finally, a major storm destr oyed

the last remaining vestiges of rice culture at the turn of the c- :;,..,y

(Vanstory 1970:83). k following passage taken from an article in the

Darien Tlimber Gzette summarizes the sentiment toward rice growing amon,

post-bellun plantLers in Georgia after a devastating flood:

the loss will be a heavy blow to many of the planters
especially where money was borrowed to plant with. Several
have already declared they will not plant any more as the
crop is so uncertain. Every year the acreage is decreasing
and the -.'iver planters are nearly discouraged.
(27 August 1887, Darien, Georgia)


Long-Stajple Cotton Culture

Like rice, the production of long-staple cotton was restricted to

the coastal fringes for most of the antebellum period. It became known

as "sea-island cotton" because the staple was believed to degenerate

when removed fronmi the influence of saltwater (All ston 1854:593). Its

introduction to the coastal states, unlike that of rice, occurred in

Georgia not South Carolina. :itiallygrown in the Caribbean, long-staple

cotton made its way to coastal Georgia through expatriated Georgians who

had made their home in the Bahamas after the Revolutionary Ear. The

success of the cotton in the Bahamas led the former Georgians to send a

few bags of the Cseed to their friends and relatives in Georgia (Coulter

S1941:65-68; Hammnd 1897:1- 17). cGradually, the fruit yielded a finrr

and bettor cotton tha that grown in the ah Wmas (johnson ,930:2-2).

By 1789, substantia quantities were produced in -Georgia and its cultiva-

tion spread to nnary Scuth Carolina.

For most of the antebt 1W-H ; period, spa-islad cotton. was grwul

exclusively along the narrow strip from Charleston, South Carolina, to








the St. Johns River in Florida (Hammond 1897:19). In the early forties,

it was discovered that the crop was adaptable to sea marshes if these were

reclaimed and to interior lands as much as 100 miles from the Atlantic

Ocean. This discovery led to the exp.,-. -ion of ithe industry into the

interior lands of Georgia and Florida. By 1858, Florida had become a

leader in the production of sea-island cotton, and the volume of exports

for the staple increased by 50 percent as a result of the expansion in

cultivation (Gray 1941:733-734).

Lcng-staple cotton was strikingly distinctive from the short--

st!aple which was cultivated elsewhere in the Old South. As the name

implies, the fiber was longer than the short-staple. ihe fiber length

of sea-island cotton reanged from 1 1/2 to 2 inches as contrasted with

5/3 to I inch for ithe shcrt-staple (Gray 1941:731). The quality of the

two was also dissimilar. The fine, silky long-staple was used in the

manufact ure of lace, thread, and cloth of silky luster. On the other

hand, short-staple cotton was used primarily for making coarse cotton

cloth (Gray 1941:731; Phihllips 1929:91). Moreover, the !(.. -stapie

required a lc,... &growing season, more hands per acroe, and more tedions

methods for cuitivation, harvesting, and processing (see Otto 1975:53-61

for a dectailed comparison of the two varieties of cottonn.

:dhods. of Plantuin aad Cultivation

The soil wns prepared for planting in February or early March.

Soils varied frro iight sandy loams to heavy clays, but the sandy ones.

were preferred (Fh.7lips 1969:271). On lowlands "ditching" similar to

that employed in rice culture was often necessary to drain an area before









planting (Gray 1941:731). High rid:., 4 1/2 to 5 feet apart were con-

structed before the seed was planted. These were usually made with

hoes, particularly in the old "sea-island region" (Gray 1941:735).

Plows came into wider use during the late anteblclum period, when more

time was needed to manure exhausted soils (Gray 1941:734). Between

late .rch and the middle of April, the seed was sown. Once the cotton

plant sprouted, it had to be "thinned" with hoes several times. Gener-

ally cultivation consisted of 4 to 8 hoe-ings, and grass was pulled by

hand.

With a few notable exceptions, most sea-island cotton planters

did little to enhance the productivity of their crop prior to 1830. At

that time, confronted with fluctuating prices and soil exhaustion,

experiments with manuring and crop rotLation were undertaken (Johnson

1930:51-54). Manur"ing soon came into wide use,and various types of

manures were utilized such as marsh mud, crushed shell, and guano. The

practice of crop rotation, on the other hand, was limited to a few enter-

prising planters (johnson 1930:60-64). Careful selection of seeds was

perhaps the universal practice adopted by most planters to improve the

quality of the cotton. Several varieties were 'known and some reputed

to yield cotton that sold between $1.50 to $2.00 a pound. Unfortunately,

the production of these seeds was kept secret and this knowledge is now

lost to agriculture (DeBows Review 1,67:87).


Harvostine, OMAN-occs nd V~rKetmoq

A great deal of care had to be taken while harvesting thI cotton.

The fiber was gathered as soon as the pod opened in order to protect it

from injury. Also, trash, leaves, and dirt had to be separated frcm the









fiber carefully. Harvesting required 10 to 12 pickings, whereas the

short-staple only required three (Gray 1941:735). ;i'.i it was picked,

the cotton was placed on a platform to dry in the sun and later stored

in the barn to await ginning.

The famous "Whitney saw gin," invented for the removal of seed

from short-staple cotton, was not used for ginning sea-island cotton;

instead, "roller gins" were used. In the early days these wErc hand-

powered mills derived from the Indian churka. Later improvements to

this apparatus led to the invention of gins powered by animals or

steam (1, ....- 1897:72). After the cotton was ginned, it was Jited":

that is the residue from the broken seeds was removed. The clean

cotton was then hand-packed in bales, since hand-packed bales vwere less

injurious to cotton and consequently brought higher prices than those

packed by a press (Gray 1941:736).

Both acreage -per hand and yields per acre were severely limited

because of the laborious cultivation and processing methods. :" re cul-

tivation relied exclusively upon hand labor, the average number of acres

cultivated per laborer was 3 to 4. Where the plow was substituted for the

hoe, the average number of acres per laborer was 6 to 7. .'.: average

yield per acre ranged from 125 to 400 pounds, althc -5i yields recorded as

high as 2,000 pounds per acre were known (Gray 1941:737; Phillips 1969:

275).

Only superior long-staple cotton commanded very high prices, but

its production was limited to a small group of planters (All stou I'":

596). For the rest of staple, price fluctuations were considerable.
S-: average price range during periods of good to high prices was $.40

to $.80 a pound, and at periods of &'-,,; sed prices $.14 to $.25 a









pound. -he decade before the Civil War was a period of rising prices

after a long period of depressed prices. At that time, the average

price of cotton was between $.37 and $.47. However, after the Civil

War, the quality of the crop deteriorated and its production became less

valuable. Machinery was soon developed which produced thread from the

short-staple as delic-ate as that made from the long-staple (DeBows

Review 1867:,..). F-inally, devastation from the bol1 weevil made the

production of the crop totally unprofitable (Brown and Ware 1958:69).


S UTnlain

Rice and long-staple cotton are two very dissimilar culitigens.

Each crop requires distinctive modes of cultivation and processing. For

the meost part, each is cultivated in a very specialized habitat and is

produced for its own peculiar market. Yet analogies between the two crops

exist in the hiscIory of their production in the Georgia tidewater.

Both crops were restricted to a synonymous geographic range and

required a considerable labor force. These similarities were particularly

significant to the demography and economy of coastal Georgia. The

Georgia tidewater, like that of South Carolirna, became a region with

a heavy co ncentradtion of slaves and a few wealthy slaveholders where

absenteeism fre,.- .,y prevailed.

Another ao cr I was evident in general cultivation practices.

Often, sea-island cotton growers were also rice growers or former rice

growers. Appar.ently, certain practices established for rice, the

old r staple, were Nr.ieueintly adopted fur cotton. Sone of these, though

harmless to the rice, proved to be detrimental to the cotton industry.









Clearly, the conservatism toward the use of hand labor, manures, and

crop rotation had its origins in rice culture. ":- .ring was not essen-

tial to rice growing since silt loam from irrigation aided soil fer-

tility (Cole 1927:601). Similarly, hand labor was vital to rice cul-

ture because of the difficulties of plowing. Perhaps an early implemen-

tation of plowing and man uri-.g in the cultivation of 1o:, -sta.ple cotton

would have resulted in lower production costs and higher yields than

those actually e. _'rienced by most coastal planters. At least. the

success of the sea-island cotton industry in Florida was partial ly re-

lated to the greater use of the plow (Gray 1941:737). Unfortunately,

the failure to employ crop rotation methods by both cultures resulted in

early so'I exhaustion, which had disastrous effects upon post-bellum
agriculture in the area.

Likewise, labor practices were very similar. Here again, the

task system which developed from rice culture was later used for cotton

culture. Under the task system, field hands were classified as full,

three-quarter, half, and one-quarter hands (Flanders 1933:143-144).

Generally, each full laborer was expected to work a task of 1/4 to 1/2

acre everyday, whether in rice fields or cotton fields. This system

was particularly advantageous if the worker completed the task by mid-

afternoon. Upon completion of the task, he or she had the remainder of

the day to tend to personal needs. Conversely, under the gang system,

the other major labor system for plantation slavery, workers labored in

the field from sunrise to sunset (Phi lips 19299:279-280; S.amp 1 56:

53-56).








It has bee suggested that the task system provided slaves at

sea-island cotton plantations in coastal Georgia with sufficient leisure

time to improve their material lots (Otto 1975; "-cFarlane 1975). Und;r

the task system, the quality of slave life could be greatly improved

through hunting and fishing, selling handicrafts or plantation produce,

or by growing foods to supplement plantation rations Presumably, the

use of the task system at a rice plantation should have resul ted in the

saiC conditions. But how similar were the conditions of slavery in two:)

very different crop loci? The adaptations of slavery to long-staple

cotton culture in coastal Georgia has been investigated utilizing both

documentary cand archaeological resources (Otto 1975; MacFarlane 1975).

A similar investigation of slave adaptations to rice culture along the

coast of Georgia is presented in the fol lowing three chapters.















CHAPTER 3
HISTORICAL r'iIL ARCHAEOLOGICAL INVESTIGATIONS
AT BUTLER ISLAND

Butler Island is one of several "brackish marsh islands" (Long

1958:11) within the Altamaha drainage in fMlcIntosh County, Georgia (see

Figure 4). These islands are termed brackish because they become

covered with saltwater when tides are pushed by high wind and are in-

undated with freshwater when rivers overflow as in the case of freshets

(U.S. Dept. of Agriculture 1961: 21). The land consists of sediments

washed down f'.: streams flowing out of the piedmont and the coastal

plain. Soils vary but are primarily clayey. Generally, these marsh

soils are poorly drained and can be extremely acidic, particularly on

unreclaimed marsh (Long 1958:13).

During the nineteenth century, most of these islands were con-

verted for rice culture. Later in the twentieth century, truck farmi-ng

replaced rice culture. Most recently, these marshlands have become

managed breeding grounds and refuge for waterfowl. Since 1954, several

of these islands have been acquired by the Game and Fish Commission of

the Georgia Department on Natural Resources for the development of a

waterfowl management area. Butler Island and adjacent Charipney Island

were the first to be acquired for the program (Anonymous nd). Today,

Butler island has become the headquarters for it he Altamaha Waterfowl

Management Area.






47






^-.... .. .,- i"-i- i-- .... ,-:; .... . --^ 3-. ^, ), --

1. i Ll 4. N, .
\ - -"- "-
^ - \g -^ y x. . ^ _.,- ^ .. ,
i .- BUTLER iSL- - - -" -- 1



\ - 4, _- ... ..-~ -
,.,.,Jz S, L At N O \ ;

/^ ". l ^ ^ < ----- -+- .. . -- \ \,I


y : i' ~ ~ __- k -" '- -'' -' (. .f ., --,
IjJ' >7; -(1/^'\--: ---v-*'/




7 7
.......... y ..................... . -t .













F r Bul.er is lan md rc ivi ois.
\- ,
\~ -. /*- _.^- - g


-. 1 ~ 4

\ [ '~~~LL-.L]6'^ I"^^







Figure 4. ButIer Island rtnci Env'ons.











Archaeological investigations were undertaken at Butler Island in

response to a need to locate and identify archaeologically sensitive areas

within the confines of waterfowl mana .-._, activity. Certain manage-

ment practices such as seasonal flooding, plowing, and dr..' ng activities

would adversely affect or destroy archaeological sites. In past years

on several occasions, archaeological materials have been unearthed as a

consequence of these practices. To prevent these practices, funds 'were

obtained for survey, inventory, and testing at Butler Island from the

grant-in-aide program of Heritage Conservation and Recreation Service

(HCRS), U.S. D-<:' riment of the Interior, and administered by the Historic

Preservation Section, Georgia Department of Natural Resources. Nine

sites were located and identified. All were found to be historic sites,

but three have prehistoric components.

in this study the slave village sites are the subject of discussion,

although the other plantation period sites are briefly described. A de-

tailed description of all the sites will be part of another archaeological

report (Singleton rid).

Another major aspect of the Butler Island project has been the

archival research. Documents relating to the Butler family ownership

of But ler Island are voluminous. These records have not only provided

the background data presented in this chapter, but have also been util-

ized in the identification, dating, and interpretation of the archaeologi-

cal resources.


OWnership and Occuoation

Unfortunately, the first colonial grantees of Butler Island

could not be ascertained. it is possible that the Lochian aMclntosh









family acquired Butler Islian duin
family acquired Butler Island during the colonial period. Between the

years 1758 and 1775, Lochlan Mc,'ITntosh i,"..;. -,ed several large tracts of

marshland in the Al tamnaha delta (Jackson 1973:17-18). The precise iden-

tification for some of these islands is unclear.: Butler Island

may have been one of them.

Although the first owners of Butler Island are unknown, Mlajor Pierce

Butler apparently made the first attempts toward reclamation ad cultiva-

tion of the island 1:".e the family's ownership was complex,*' an

attempt has been made to summarize it (see Table I). The islailnd was in

the possession of the Butlers for close to 120 years, and several family

members have made this island a very significant historic site.

Born in Ireland, Major Butler came to America in 1766 as an

officer in the British Army. He l iked America and decided to settle

here. Later in 1771, he married Polly Middleton, the daughter of a

wealthy South Carolina planter. He and Polly had one son and four

daughters. During the American Revol Iution '1_-or Butler sided with the

revolutionaries and became an importaant political figure in South

Carolina. In 1787, he was a delegate to the Constitutional Convention

in Phinadelphia, and he was elected twice as a senator from South

Carolina to the U.S. Senate (Scott 1973:53, 1978:6-7).

At the Constitutional Convention, Major But er was one of the

principal prporne nts arguing that state repressentation in the Hous of

Representatives should be proportional to the total population, both

slave end ,free (Alden 1971:8 6--87; Scott 1973:6). Butle-r and his



*See Scott (1961 :xx-xxi) for a ,-'.- .1ogy of the Butler fmily.










Table 1
of Butler Islanrj 1790? the resent
t,,ur I s ri, 1790? to the rsn


1790?-1822

18 22 i0 6


36-18 /47


i847-186i


In ;cr ~1Crcu f~etir2~r


John B ier arnd Pierce Butler, Jr.
(forinerly John Mease and Butler
.-e), but estate as dni"nis-
tered by the Mlaj or's dauhtr
Frar c eS and his so -l Thoias.

John and Pierce, Jr., a, sued ful
resi oni bili ties of owe r: hi p until I
JoIns death in 1847.

Properties jointly owned by Pierce,
Jr. and John, .utler heirs, but
slaves had been divided equil iv
be-ween the two.

,doned.
d'8 o iC e-, d.,.

Frances BIe. r Leigh (Pierce
Butler, Jr., ,nd F .... Ke.e's
daughter).

Lelhs St:il own l but have leased
Sian re p enters.

Po.s es t4uIo1Vugh sevc'ral hn idC!s, pri-
marity rice 9rew.rs.


i861-1865 (Civil War)

18665-1878



1978-1910


1910-1920


l20- 1949

4 9 Tioc1


Colorel us C-on.

R. J. R,-yv,:,o ds


Grn and Fiish Cm-i Sion, Georia
pzrtr ient of .atural FResotrces.


Ov.)nersh i p


Own aer s


1954"'.-pr"esent









supporters believed that the slave states needed this type of represen-

tation in order to prevent the possibility that a majority of representa-

tives from the free states would d attempt to abolish slavery. ihe "-,jor's

view epitomized that of the small but powerful elite who desired to

perpetuate and -....! the institution of sla'.-.,- (Scott 1978:7).

In 1790, Polly died and the Major took his family and slaves to

Georgia. Like many South Carolinians:, the :7 ir moved southward in

search of new lands. The developing market for sea-island cotton and

the revival of the rice industry made the fresh, fertile lands of

Georgia attractive to the immigrants.

By 1799, Major Butler had acquired several thousand acres in

Mcintosh and Glynn counties. Included in these were Butler Island,

Butler Point on St. Simons Island,* L.ittle St. Simmrons Island, and

pine land on the mainland known as ',ooodville (Butler EstatQ Papers,

1799)(see Figure 5 for locations). In addition to land, the major

also acquired such a vast number of slaves that by the time of his

death he owned close to 700 (Butler Ustate Papers, 1821). Ev-,ntually,

the jor became one of the wealthiest planters in Georgia.

The Major huitt a fine mansion at Hampton plantation on St. Simons

island, where he spent most w,,inters. It was here that Aaron Purr sought

refuge aft.r the du..el in which he killed Alexander Hamilton. Burr

visited the plantation for a month and he described it in many of

his letters to his daughter, Theodosia (Van Doren 1929). The Major


.'Conc'rfentr with the archaeological research at rutler Island,
similar research hs been undertaken at several sites of the Hampton
plantation (see Muins, 1978; Fairbanks and Mullins- i,-re, 1980; and
i. ins-Moo, re, nd).




52










^^^~ '-:-"
{1 ,_ ",- J .. / ,,,-v 1L,-..: ." """" ^!, "

fT- 7r



/y~ ,I/n
... -"-- '( !i ,,f ~\V\ \}. J! /
:../ '.." -y :. ... "*y ^ u
. ./ L .. / .,*'
.. .: s ^ ^' ,... ,,- :., ... t .
/, -",,'-'.-.- " --" -. t
"\,.' ('!." '" .. . .-
"\ "" \ .} ".M -,--C
V(



"- \ "/- -"






M'" Wl 5lei'







Fgr 5. TG"e a F s a 1
-.. '- ; r < *




Figure 5. -The But!ler Georgian' -,sta!e. ...(Adapted -From Scott 1961.~l )










and his daughters, however, made Philadelphia their permanent home.

Thus, the responsibility of running the estate during the "-or's

absence was taken care of by an overseer.

,--. the Major died in 1822, his properties were inherited by his

grandsons, John and Butler -..,e. The Major had wanted his only son,

Thnomas, to take on the responsibility of administering the eorgian

estate after his death, but Thomas, who hated the Major, refused and

went to Europe to live (Scott 1973:54). Thus, the ]or disinherited his

son and made his grandsons heirs. After the :jor's death, however,

Thomnas helped his sister Frances to administer the estate until the

Mease boys came of age and were able to do so. At the Major's request,

his grandsons had to change their surnames to Butler. Consequently,

John Mease became John Butler and Butler '. -.*e became Pierce Butler, Jr.

In 1834, Pierce Butler, Jr., married Fanny ;- Alie, the famous British

actress. During her first and only trip to her husband's Gecrgian

estate in 1839, she wrote the controversial account of slave life,

A Journal of A Res idence on A Georg ianPlantatio n To ]--1]939, which was

published more than 20 years later.

Fanny and Pierce came to Georgia with their two young daughters

for approsinimately four months. Two months were spent at Butler Island

and another two at Hampton Plantation. Fanny was oitra,"d at the con-

ditions under which the slaves lived, particularly the suffering enrduredc

hy women. Fanny's reaction resulted in man'y heated arguments with

Pi erce, which intensified the couple's preexistin rit problems.
,,=i..ch. winter_ /id co euple s "s'' in rimr

After their return to Philadelphia, the marriage grew increasingly worse.

in 1849, Fanny 'returned to Pritain, the couple were divorced, and

Pierce gained custody of their dau-ters (Scott 1973:11).









F :.,." always wanted to publish the journal she had kept while in

Georgia. In 1859, she began giving this serious thought for two rea-

sons: first, she was no longer married to Butler; second, the estate

she describes in the journal no longer existed (Scott 1973:119).

Pierce Butler had fallen into serious financial debt as a result of the

crash in 1857, and more than 400 slaves were sold to pay his debts

('x,.:-i 1863). In light of these new circumstances, F.-,, reasoned

that her journal would be thought of less in a personal way than as a

social statement of slavery in the United States (Scott 1973:120).

The Emancipation Proclamation and the events of the Civil War gave er

the final impetus toward publication. Fanny felt that the journal would

help "explain Lincoln's war aims and show the nature of the system that

the North was fighting to abolish" (Scott 1973:122). The journal was

published May 1863 (Kemble 1882:205).

It is difficult to estimate the reaction to the journal imme-

diately after its publication. A few Southerners have suggested that

the journal single-handedly swayed British opinion against the South,

which made aid to the Confederate War effort impossible (e.g. Perkerson

1952:133-134). Even among Fanny's admirers this opinion has prevailed

(Armstrong 1938:432). It has also been argued, however, that the

journal had little influence ..:., the contemporary pulbic ( Lombard

1930). Recent evidence has indicated that by the time the journal '.'as

published, British opinion had already turned against the South

(Scott 1973:124). More significant than the criticism related to the

influence of the journal upon the contemporary public is the cri-

ticism related to its accuracy and authencity. Among historians









viewpoints vary. Apologists for slavery tend to be very critical (e.g.

Phillips 1929:259-265; Eaton 949) Other historians have found a greet

deal of accuracy in Fanny's descriptions of slave life, although some

suggest that the journal is o'-,: .iized (e.g. S.tampp 195C; Genovese

1974; Oweins 1975). Of all the critics, Georgians have Men the most

hostile (Holmes 1978; Lovell 1932:196, 213; Cate 1960). Lven today,

this sentiment persists (Farncher 1970:102; e.g. Lewis and Hi-e 19/8:120).

Butler- Island was dese rted during the Civil War, with the e. :-

tion of a few slaves who remained. In 1865, Pierce Butler, Jr., and

his daughter Frances, returned to Butler Island to restore rice plant.-

ing (Vanstory 1970:87). Frances kept a journal, which was later pub--

lished (Leigh 1883). Her journal became one of the few doceuments th at

describe labor conditions in the South during the postwar years (Hous.

1942b:157). Pierce Butler died in 1857. Frances, accompanied by Ler

Gerran maid, m-ana- the plantations alone for four years. -Frae.s

miarried Reverend James Leigh in 1871 while she was in England and they

remained there untit 1873, at 'ich time they returned to Georgia.

T Reverend Leigh; was very much interested in the South and helped to

establish St. Cyprian Church for the blacks of D-rien. Leigh liwrote of

his experiences in coastl Georgia in Other Days (1921).

iMY in Englanrd, thahe Leighs contracted some English workers to

supplem ent the black 1or at Butler island. These British e brners

raidedd at But e Island and occupied the former slave hospital. As

.hnre's; however ''' wer.a a disoppointr t (Leigh 153:M -21 3).

Discourh.ed, in 1876, the .ihs decided to return to England and

placed the mrnur,-. A of the islands in the hands of a manager. Yn

1878 they g3ve up their planting interest entirely.








Information of ownership and land use at Butler Island in the

years between the Leighs' departure and 1920 has been very sketchy. It

has been assumed that the island was abandoned and left unattended

(Vanstcry 1970:89). Yet deed records, local informants, newspapers,

and a valuable map (see Figure 6) have suggested a different interpreta-

tion.

Evidently, the Leighs leased and sold portions of their proper-

ties in the Altamaha delta between 1878 and 1910. By 1910, all of their

properties were sold (Mcintosh County Records, Deed Books H-52, 1-120,

1-160). These properties, which embraced all of Butler Island and

General's island, were sold and leased to rice planters, who in turn con-

tracted with tenants to farm small plots for a portion of the crop (as

shown in Figure 6). Local informants have indicated that by the 1880s,

tenants no longer lived at Butler Island. Instead, they lived at several

settlements on the mainland and commuted daily by small boats to Butler

and General's Islands. Two of these informants, both descendants of

slaves in the area, remember vividly planting rice at Butler Island in

the early 1900s.*

Long after the Leighs left Georgia they continued to correspond

with the local blacks (Letters from Descendants of Butler Slaves, I7.3-

1903). Particularly, their only child, Alice Leigh, visited Georgia

often and on these visits she would give parties for the people (Cate

and Wrightman 1955:203). As children, present-day black residents of

Darien performed for "Lady Alice" on these occasions;.** The impact of

Elizabeth Bates and Jim Cook, Personal communication, 1978,
Darien Georgia.
--;'.,--. ,r1 Wylly and Mrs. Gamble, personal communication, 1978,
Darien, Georgia.




57







'; L--.__ S..A.S. -.- ,-<,,. '- y
-\, -<
.*:TAM.\.\ v V.A- <' -1



4 '" ", '* \" "-- -' ' ^' ^ "s: "
t '- *,] -" X- .A ^ .... ,^ ^



7 i i \ ,,. \ \. :\
St" 5.-' -.--.''- .A

"' -. v V.\<^. ^ j < ." : % .
-.-*[ T.-< /irs / 7"<* 1. '* ** ..*. -" ) <> \ .-.'*. > f S ~..
-:. .,,, -<, :
---^ \jy ^ .>yj,-^,
^ \ \.y \jr^


Cr": \\, -,,:'.7 <'j-

^ ~ ~ ~ f-- / : '-- '*- .- ~^-'> *' .i ii >'i ,o i
I '' t- -, ,;\.y' -- ." c s* '- "'" '. ? *
--'-- / 3 i' ? \ .\

-^.' "* <-'v--.-7-- ''- ...- A" -'"- .-;< 'J : "< -- y'- < ;.s-' t
;. -... ,A ? / s /"

,\ ,<,' "- i' "' a 12 / -
) J. 5, /. I'^ y ^ ; -- ;\
i- '-> .... / ''"-N '.-" -"
| -"5 _S '- --- ."7 ---" "k. ', .- -<- .
. . .. ,, . . S

I,,-f- S *'^?< ^ '"

F -tij ,.re 6." lhe .Subri wv s :rciui ... ,.E';t er il, G ,_ne.,o Isl nd ...- '-
7 -e nta n ti P lots (c ir"ca. 1883.,').









the Butler family upon the Afro-Americans in the area had lasted well

over a century.

In 1920, Colonel Tilli.i-,h:t L'Hommedieu Huston, part owner of the

New York Yankees, purchased Butler Island. He built a two-story house,

which is still standing today on the presumed site of the overseer's

cottage. Huston established a dairy farm and planted fruit trees and

vegetables. His special interest, however, was the cultivation of ice-

berg lettuce. He revamped the old rice irrigation system and used it

for his fields. Huston remained at Butler Island until his death. In

1949, Butler Island was sold to Richard J. Reynolds, the tobacco million-

aire, who made additional improvements to lettuce cultivation (Vanstory

1970:90). In 1954, Reynolds donated Butler island to the state of

Georgia for the development of a wildlife preserve.

Plantation ":.,.gement at Butler island,
1802-1860

The overseer on the Carolina-Georgia coast has teen described as

the most important "single element in the mana,- -'. hierarchy of the

rice belt"' (Scarborough 1964:183). At least, this appears to have been

the case at Butler Island, where the Roswell King family successfully

managed the estate fur over thirty years during the Butler family's

periodic absences.

Absenteeism as a practice among rice planters was established

fairly early by colonial predecessors. This tradition developed, in

part, out of a need to escape the hot, malarial season as it was

.- .rally believed that "...asmas" caused the disease. The ownership of

multiple plantations also necessitated the periodic absence from one or









more of these productive units (Bridenbaugh i952:69). In the early days,

planters traveled to the North or to Europe, while others migrated to

their town houses in Charleston and other coastal cities. Seashore

settleM.ents and pinelanLd, piedmont, and mountain villages began to sup-

plement the coastal towns as sun-er resorLs later in the nineteenth

century (Brewster 1970:109). The degree of absenteeism, however,

apparently varied along the rice coast. For example, in the Georgetown

district in South Carolina, one planter noted that of Ill rice planta-

tions in the area only 51 of the owners actually lived on their planta-

tions, and of that number all were absent during the crop season from

" y to November (Allston quoted in Cathey 1956:142). Some planters, who

could not afford a "dual existence" or who were too far from Charleston

or Savannah, remained on their estates throughout the lp r or most of

the year (Bridenba 1h 1952:70). In the Altahama Basin, rice planters

spent a major portion of the -r on their rice estates and moved to

nea,! pirneland retreats or barrier island plantations for the summer

months. Often, they conmuted each day to keep a watchful eye on the

progress of the crop and other :. ,-,- t problems (House 195401Q). The

Butlers were the only exception to this rule (see '."1iy 1910.48-53).

They were never full-tim.e residents of Georgia. Because they were

generally absernt from their Georgian estate, the overseers regularly

corresponded with them. Fortunately, a large portion of this corre-

spondence has survived to the present day.*


*The. Roswe'l Kings' Correspondence is found under two manuscript
collections of the Pennsylvania Historical Society, in Philadelphia:
The Butler Family Papers (primarily Boxes 9, 10, 11, 12), and the
Wister Family Papers (The Butler Family Section, Boxes 31 33, 34).










This correspondence provides a wealth of information related to
the day-to-day activities at Butler Island and the other Georgian proper-

ties. In many ways these letters resemble similar correspondence

written by overseers at other plantations with absentee landlords (e.g.

Easterlby, 11.,; Bassaett 1968; Clifton 1978). All have in common a con-

cern with planti-., harvesting, and marketing activities, labor manage-

ment problems; and the health and welfare of the slaves. Occasionally,

the Roswell Kings' letters also supply some indices of slave behavior.

Most often, however, only deviant behavior was reported, and therefore

these references tend to be biased and one-sided. Yet, despite this

limitation, the Kings' observations provide valuable insights into

slave lifeways at Butler island.

As overseers, Roswell King, Sr., and later his son, also named

Rosv;eil, belonged to the "overseer elite" (Scarborough 1966!:58-177).

This minority among overseers was frequently employed on the Carolina--

Georgia rice coast but was found in other localities as we ll (e.g.

Sitterson 1943). For the most part, members of the overseer elite were

conscientious and efficient men. They made significant contributions

to agriculture, and some became planters themselves. Often, they were

the sons of planters. In contrast with their colleagues, they were

usually well paid. Both the Kings earned as much as $2j_ a year.

Roswell1 King, Sr., had come to the Georgia coast from Sharon,

Connecticut, in fhe 178is (Lewis 1975:39). In iW02, he became the

manager of the Butler estate, a positicio he reta.:ined until 1819. Some-

time after that date, he became the manager of the Darien Saw Mill

(Darien Gazette, 6 January 182'1). King's son, Roswell. Jr., uas









selected to replace him upon recommendations from Roswetl, Sr., and

John Couper of Cannon's Point on St. Simons island (Couper to 7i.jor

Butler, 15 February 1818). Roswel, Jr., presided as overseer until

1839. It was young King's resi.,.tio n which resulted in the trip Pierce

Butler, Jr., and his wife, Fanny 'e..,e made to Georgia. Butler had

come to Georgia to hire a replacement for Roswel I, Jr. Thomas Olden,

formerly the manager at James HamilIton Couper's Hopeton plantation

(Anc,., .,s 1965?), was hired to managed the estate. Olden died in 1841,

and Roswell, Jr., once again became involved with the miana-.-,t of the

Butler estate, but this time in the capacity of a steward. As a

steward, he supervised two overseers and managed the account books.

Roswel, Jr., held this post until the mid-SlSOs (Scarborough 1966:168).

Unfortunately, most of the plantation records for the Butler

estate, particularly the overseers' records, are between the years 1803

and 1835, prior to Fanny -w bt's visit. Within these 32 years, however,

it is possible to obtain a very clear picture of planting and labor

manav- ..t activities. After 1835, it becomes difficult to document

management practi-ces, but it may be ,.-e to assume that earlier,

established practices prevailed in later years.

Under the management of the Ki ., the Butler estate was operated

efficiently. Both Kings had taken an active interest in planting. With

regard to labor management, runaways were few, and slaves received treat-

mert comparable to other slaves in the area. To a great extent, their

success as m,.na.. s may hMave been related to the dutiful drivers of the

estate. '...t of the drivers were loyal and reliable, and the Kings felt

a high regard for them. For example, in the selection of a new driver









for Butler Island, King sought the counsel of two drivers at Hmpton

Plantation (Roswell King, Sr., 15 :.. Ler 1818). Several years later,

Roswell, Jr., praised the work of a driver by sayin-g: "The hands a.-.-c

to be fit enough to attack another crop, which can only be attributed

to Sambo's mana At,--t in varying the work as rimuch ,s possible" (Roswell

King, Jr., 24 September 1825). The death of a good driver was con-

sidered a great loss to the estate, as well as to the entire planter

community: "Driver Brain . was worth a great deal to the estate

his death is regretted by every white person who know him" (Roslwell

King, Jr., 1 February 1825). But problems with drivers were .ot un-

common. One at Butler Island was involved in steal iig la ..- ar.o. its of

corn from the warehouses and selling the corn to merchants in Darien

(PRoswell King, Sr.., 22 November 1806). Another driver, a1to at Butler

Island, apparently was an alcoholic (Roswe11 Kinig, ,Jr, 29 August 180).

Occasionally, a driver's cruelty to the field hands required thaFt he be

removed:

I have been obliged to displace Morris at St. Anne.
He was always tyranical and brutal, 'which I c. -
ally kept -under, but lately he had become intemperate,
in fits of that kind, and has used great severity to
some of the negroes. (Ros1elMl King, Jr. 16 July
2826)

Despite occasional problems, most drivers were *,-.-'d at their jLobs and

to the field hands (see Van Deburg 1979, for a reassessment t of the sieve

driver). Generally, drivic-rs "strove to mediate between th. Big Hease

and the quart.s &.' (Genoveso i974:378). At Bu-ier Island, r' ive'-s

were particulalry vital to the managei-m ent of a planitioln whichh saw

few whi tes









Of all the properties in the estate, Butler Island was the largest

in both acreage and in production. More slaves were utilized in the

plantation operation than at any of the other properties. As many as

300 to 400 slaves lived and worked at this rice island and they occupied

four slave settlements or villages. Rice production, however, was not

the major interest of the estate until the 1820s. Prior to that time,

long-staple cotton dominated the planting interest (Scott 1961:xxxviii-

xxxix). Most of the cotton was planted at St. Simons and Little

St. Simons, but some was also cultivated at Butler Island, apparently

for crop rotation purposes (see Coulter 1941:104). Despite the interest

in cotton, considerable quantities of rice were produced before 1820.

As indicated in Table 2, the average annual production of rice was 712

tierces between the years 1802 and 1820. From 1821 to 1833, the

average annual production (1827 and 1829 are not included) was 1,246

tierces. Thus, in later years, rice production increased by approxi-

mately 43 percent. Years of low production coincided with losscs due

to storms, as well as to .-.rs when the acreage planted in cotton may

have exceeded that of rice.

The rice cultivation methods utilized at Butler Island were
i.i-rally typical of the region. As a rule, the mnost sophisticated

rnh&ods and equi. .. ,t were employed. As early as 1802, rice fields were

marurr-ed, possibly with lime and marl (Barker, 1841-1844) to enhance soil

productivity (Roswell King, Sr., 17 September 1803). Fields were

pi'r, ted with rice oe year and then rotated vith cotton to keep dnwn

the growth of grasses (Coulter 1941:104). Occasionally, the fields

were prepared with plows, but hand labor predominated. In the processing









Table 2
Rice Production at Butler Island, 1802-1833


Year Acreage .er of Tierces a


1802 -- 128
1803 337 1/2
1804 -- 44
1805 -- 1326
1806 -- 468 1/2
1 or7] 8
i07 --1186
1808 --- 565 1/2
1809 -- 578 1/2
1810 587 1/2
1811 -- 857
1812 1739
1813 253 1/2b
1814 -- 829
1815 591
1816 -- 513
1817 -- 1058
1818 -- 650
1819 407 1173
1820 227 548 1/2
1821 234 832
1822 272 1080
1823 358 1240
1824 342 84.b
1825 318 1024
1826 399 1455
1827
13828 395 1331
1829
1830 505 1747
1831 552 1557
1832 271 1695
1833 591 1660


Source: Compiled from Annual Crop and Livestock Reports. 1803-1833.

Note: Average annual tierces., 1802-1830, is 712; avedrege anmu-i
tierces, 1821-1833 is 1 246; increaC equals 43%.

,Tierce = 600 lbs.
bLow production due to hurricanes.









of rice, experiments were attempted with every conceivable threshing and

pounding device. By 1835, the machinery for processing rice included

an animal-powered mill, tw:o tidal mills, and a steam mill. Tihe steam

mill was built in 1833; it housed both threshing and pounding machinery

and its cost was over '.<000 (Roswell King, Jr., 25 March 1932, 22

April 1832, 22 February 1833). Evidently, the threshinrg mills were not

sufficient for the amount of rice harvested, anid a portion of it was

threshed by hand on "tarred" threshing floors (Roswel1 King, Sr.,

9 August 1806; ;- ble 1961 (1863):109-110). Rough rice was also mar-

keted, but large quantities were not sold until the 1850s (Pierce

Butler, Jr., Ledger, 1851-1855).

Besides rice arid cotton, sugar was planted at Butler Island be-

ginning in 1812. In fact, :?-Jor Butler had considered "turning the

island into a sugar plantation" (Roswe1l King, Sr., 1 er 1812).

Between the years 1812 and 1815, complete sugar processing equipment was

added to the plantation complex (Butler Estate Papers, 1815a).

Roswell, Sr., had become very interested in sugar culture and he sought

the advice of West Indian pla nters (Roswel1 King, Sr. 12 August 1815).

This interest in sugar was later pursued by Roswel1. Jr. (King 1828).

Regardless of their enthusiasm for sugar culture, the acreage allotted d to

it was always considerably smaller than that of rice or cott-n.

Like most large plantai ions of the Old Sout.h (Galliman 19700:23),
some provision foods -were produced at Butler Island. Livestock were

raised and some fruit trees and vegetables were cultivated. ihe list

of stock included oxen, barrows, mules, mares, C ttle, pigs, sheep,

turkeys, and other fowl (Butler Estate Papers, 1803-1830) The draft









animals were used primarily for operating machinery and pulling carts

and wagons. Wool was sheared from the sheep to make cloth (Roswell1

King, Sr., 18 August 1813). On special occasions, cattle, pigs, or

sheep were butchered to feed the slaves (PRoswell King, Sr., 16 May 1813;

Roswell1 King, Jr., 13 November 1823). But these occasions were appar-

ently rare, and slaves were most often fed "barrelled pork and salted

fish" (Roswel1 King, Sr., 7 April 1816; Rosweli King, Jr., 18 March

1821). Citrus fruit trees were planted along the outer bank of the

island, an area approximately five miles in length. The w' nose was to

protect the dykes from storms or freshets (Spa ding 1830). Several

visitors to the area marveled at the beauty of the orange and lemon

groves (Hall 1931 (1828):234; Coulter 1941:96). Most of the fruit was

marketed and the remainder was sent to the Butlers or distributed v-.rj

the plantation occupants (Roswell King, Jr., 25 December 1812, 8 "7v-'.-

ber 1818). Although experiments with planting corn, barley, and pota-

toes ,were attempted (Roswell King, Sr., 31 May 1813, 1 June 1818),

Butler Island lacked sufficient "high lands" or hammocks (see House 1939;

213, 1954a:153) for the production of provision crops. Consequently,

most provision crops were either produced at the other plantations of

the estate (Roswell King, Sr., 9 June 1814; Roswell King, Jr., 25 Jan.uary

1819) or were purchased from other planters.

Plantation structures were numerous at Butler Island. in addi-

tion to the outbuildings needed to process and store rice, cotton, and

s *. r, a sawmill, cornmi11, warehouses, aud a slave hospital were iamo

the structures (Butler Estate Pc-pers,1815a). The maintenance

of these buildings was an ongoing process, and both slave mechanics and








hired white craF.. were regularly engaged in making improvements on

the structures.

Efforts were made toward making Butler Island self-sufficient In

the production of plantation supplies. Construction materials were

manufactured, including clay and tabby bricks, shingles, nails, and iron

tools. Slave clothing and shoes were also frequently made. if trained

slaves were unavailable, local whites were hired to undertake these

tasks (Roswell King, Sr., 9 July 1805, 28 May 1815, 4 October 1915).

Sometimes whites were employed to train the slaves in specific skills

(Rosweli King, Sr., 26 April 1816). The extent to which ce-tain pro-

ducts were made depended on the needs of the plantation and the cost of

manufacture at the time. This was clearly the case with regard to shoe-

making: "As for tanning and shoemaking, I recommend you give it up.

Hides have gotten extravagant. Our shoes cost $4.00 a pair, you can get

better shoes out of the state prison for $1.00" (Roswell King, Sr., 15

November 1818). Thus "self-suffici- I" at the Butler estate varied

through time and with the needs and interests of the owners.

Taken as a whole, the Butler estate was at most times very profit-

able. The estimated value of the slaves, structures, and lands was

$288,027 in 1818 (Rosweil Ling, Sr., 19 July 1819). By the 1850s, the

slaves alone were valued at $300 .' (Pierce Butler, Jr., Ledger,

1851-1855). ; .ua returns on the c ps ranged from $20,000 to $50,000

except in years of losses (Butler Estate Papers, 1814; Butler Estate

Paers, 1824-1835). Th-estntp, however, was not entirely a success

st. ry. Storms dmged Lrmendou.r s Emounis of (crops in the years

1804, 1813, and 1824. The hurricane of 1804 was particularly disastrous


............. 01110111L. I -."









to coastal Georgia (Luodlum 1953:116), and the Butler estate lost not

only crops but 19 slaves as well (Van Doren 1929:186). During the

British raids along the coast in the War of 1i12, 139 slaves were taken

from the estate (Roswell King, Jr., 28 April 1815). In later years,

freshets and epidemics among slaves had far-reaching effects upon prof-

its. Finally, the crash of 1857 resulted in the sale oF 500 slaves.

It is possible that Major Butler found the losses of his day over-

whelming. Historic records indicate that he put the entire estate up for

sale in 1818, a total of 15,000 acres and 535 Negroes (Darien Gazette,

81l8-1819). The factors which motivated his actions are not clear.

Perhaps he was of the opinion that the estate was becoming unprofitable.

it is evident that Roswell, Sr., had begun urging the ..jor to sell por-

tions of the estate in 1816: "I think you have too many eggs in one

basket to keep the land is worn out, the droughts and gales make

our cotton the smallest I ever knew . . Your interest is too large.

't. enough laborers for the amount of land" (Roswell King, Sr.,

21 January 1816). In 1818, King's urging had become more persistent:

I have .- ,-. you to sell your estate for a long time cut
of the purest of motives, thinking Negroes was above their
value and that was a time to sell, but I find I have
ered in judgement, Negroes are rising in value like the
tide, every day. . I recommend you to sell all your
Negroes and land you hold in this state, say in a lump
for $555,000 and in my opinion in fifteen 'mi nutes after
you read this and if you will not sell I E!, :.ly recom-
mend to you without loss of your time make arrangements to
lay cut one hundred thousand in Negroes. Let them cost
what they may-as I have often observed your estate must
dwindle without more force. (Roswell King, Sr., 28 June
1 81 8)

...--ther it was King's persuasion or other factors which resulted in the

decision to sell cannot be ascertained. The financial crashes between









1817 and 1819 may have ':.. factors. During that time the entire south-

ern economy suffered great financial losses (Eaton 1961:274-273). P.r-

sonal family relations are another possibility. Unable to convince

Thomas to take over the administration of his estate, the Major may have

decided to sell before considering the alternative of leaving his prc, .r-

ties to his grandsons. ..tever his reasons for selling, he never found

a buyer: "I have done all i can to sell your estate to promote your

interests and to close your accounts in Georgia. I fear I shall not be

able to do much for you (Roswell King, Sr., 165 January 1819) Thus the

estate was retained and portions of it remained in the Butler family for

another 90 years.


Records of Slave Life

One of the most astonishing aspects of the Butler estate was the

size of the labor force. By 1859, a total of 919 slaves were included

in the estate (Butler Estate Papers, 1859). Even in Major Butler s

day, the total number of Butler slaves ranged from 500 to 700. Slave-

holding of this magnitude was clearly in a class by itself. Slaveholders

who o.wned from 500 to 1,000 slaves represented less than 2 percent of

all slaveholding in the entire South. In Georgia, this "unusually large

slaveholding was confined to less than I percent cf all planters in the

state" (Gray 1941:L530-538). Thus, the size of the Butler estate made

it atypical among southern plantations

ow the MajIor m %a(jd to acquire such a vast number of sleeves is

an interesting subject in i tself. Most of the ior's initial transc-

tions had taken place between the :'i-rs 1786 and 1804 (George Hooper

1785-1302; Butler Estate Papers, IdO01-1804). Although some of the







70


Butler slaves were direct imports from Africa (Roswell King, Sr., 13

May 1803), most were acquired through the domestic slave trade based

in the older slaveholding regions, particularly the Virginia Tidewater.

Theenterprising :-'jor, of course, found ways to add to his slave"hod-

ings illegally. Before the Major left for Georgia, he took a consider-

able number of slaves belonging to his wife's family. Although the

total ,i7.'--'er of slaves acquired illegally for the -..,or's in-laws is

not certain, in 1817, 127 slaves were found to be the property of the

Major's sister-in-law (Roswell King, Sr., 29 January 1817). Apparently,

these slaves were never returned to their former owner. As planting

interests expanded, the estate was supplemented with additional pur-

chases of slaves. Natural increase also added to the slave population.

Rarely did the death rate exceed the birth rate (see Table 3). The

fact that few slaves were sold* provided an additional factor which

helped to maintain the size of the slave population.

Several slave lists exist for the Butler estate, but few specifi-

cally list the sleeves at Butler Island. Tables 4 and 5 combi ne a

number of lists taken from various time periods (Butler Estate Papers,

1803, 1821; Roswell King, Jr., Daybook, 1844). indications are

from the plantation records that this combined list reflects the

average number of slaves regularly engaged in the labor force at

Butler island. he list, however, does not inc lude children

under age. 10, which usually veraed 100 in number (Posve.'ll King, Sr.



Qc lose examination of the slave records reveals that the occa-
sions on which slaves were sold from the estate wre very rare. iThis
is in 'eepi ng with the evidence concerning other rice p1 anta ti ons in
the area (see House 1939:215).










Annual Births and


Years)


1802-1812

1819

1820

1821

1822

1823

1824

1825

1826

1827

1828

1829

1830

1831

1832

1833


T o ta.
Total


CPote: Natnural increase v


Table 3
Deaths of Butler Slaves


Births Deathss


257 191

29 19

25 165

19 27

19 22

32 14

30 18

20 28

35 23

41 39

46 25

31 16

41 25

45 20

31 29

43 34

53 ..34

756 530


176.






72


Tab] e 4
An Approximate List of Slaves by Age and Task Rate at Butler Island
1803-1845: Field Hands


,Naame Age Task Ratea


Samibo (driver) 33
Pender 31 1
Robin 25 1
Cloe 26 1
Eve 23 1
John 20 1
L e,,i s 18 1/4
Robin 25 1
Cornelia 30 1
Joe 43 1
Hannah 44 1
Limas 23 1
18 1
Charlotte 15 1
Mol ly 15 1/4
July 46
i: cy 26
Renter 56
Flora 53 1
Mia ry 32
Listern 29 1
tol ly 25 1
Bil y 49 1
Quacer 35 1
Mary 35 1
Toby 34 1
Adam 35 1
inii under 35 1
Total -- 25 1/ 2

Jack (driver) 57 --
London 28 1
Dido 21 1
John 29 1
Beck 16 i/4
Harry 29 1
Na, 18 cy/2
Ji 35 1
Toby 37 1
Nancy 52
Constant 16 1/4
Tin.h 14 1/4









Table 4--(contlinued.)


: Age Task Ratea


Q.-:,h 36 ]
,, Ii I /92
Flora 32 1/2
Auber 3
Jeffrey 37 3/4
Dally 45
Dago 26
Jean 346
Hagar 34
Sarah 4 8
Lisbon 18 1/2
Ssor24
F]u25 1
Tamfsor 25
Mari a 42 1

Total -22

Ley (driver) 30
Peggy 3 31
Sarah 19 1
Randail 465 3/4
Hector 24 /4
Peggy 17 /4
Sam 33 1/4
Libby 27 1
Provamy 23 l
Toby 19 /2
Peter 16 1/4
Hagar 52 1/2
Leira 52 1/2
Leipia 26 112
Cos !-,a 42
Quami n 24 ,
Haar 17 2/4
Robin 16 1/4
Frank 29 I 4
May 22
Abrav 42
oo1ba 43
IJ i l I y ';9 i f '
~1 9 /
L ,,,! 201 ''/2
L /- cy / /i
Lucy 29 1 4
J-in,., 47
Rose 20 ../,A
1 /,









Table 4-(contiinued)


Name Age Task Ratea


Clarinda
Violet 15 1/4
Fanny 20 1/4
Molly 17 1/2
Harry 37 /2
Lilly 44
Charlotte 28 1
Tpie 23 1

Total 25 1/2

Captain (driver) 52
Dorcas 57 /2
Lagette 25 1
Jenny 21
Cate 19 3/4
Dick 29 14
Chari ty 35 1
Morris 37 1
Melinda 43 1
Peter 40
London 1 I
York 16 /2

Affa 29
Lilla 42 1
Mira 42 1
Violet 17 ]/,4
Pompe 12 1/4
Cncy 37 1/2
Joe 33 1 l
3u ba 30 1
Betty 22 1
George 19 3/4
Lender 20 0
Pena 25
Philo 25
Elce 24
Silva 1 /1
.... 1/2
To ta1 2 3/4











Talb e 4---(cont i nued)


-.,. Age Task Ra,:ea


Frank (driver) 42
Elce 23
Albert 2
Lender 44
Breste r 33
Lookey 35 1
[o lv 13 /4
16 1/4
Jer-ry 16 1
3ru lhs 36 1
,,. ..34 1
g" :"e 39 1
Jack. 21
Tom 37 1
l9y 9
iacar 22 1
Tina 25
117 !/4
i' go 17 1/4
lus ice 32
John 33 1
Phoebe 32 1
Petty 3-5
y 41 3/4
Ga rrett 25
Ley 25 ]
Rose 42 1 2
Patty 18 /4
-Ls Li'er 15 14
rate 29 l
- m "' Q" I
Cooper c ro 9 19
rVary 29 l

-26 3/4

Grand Total 122 1/2

,dote. rotal] n i. .r 156.


oil ~ 1/?









Table 5
An Approximate List of Slaves, by Age and Occupation,
at Butler Island, 1803-]845: Plantation Specialists


Occupation Namile


Blacksmiths


Sawyers


Carpenters


Jack
Sammy
John
Andrew
Sambo
Pompe
Gorham
Jacob
Sunday
John
Bram
Harry
Ned
Pri mis
Frank
Sambo
George
Abraham
George
Ra ndy
Alex
Issac
Robin
Bram
Abraham
Abraham Carter
Justice
Justice (yo ., .r)
Sam
Joe
Robin
To by
Quash
Sancho
Peter
Tuco


Shoemakers

Ditchers


Crafts nri,
Ditchers


Age










13 Hay 1827), or retired slaves involved in menial tasks. A census

taken of the Butler slaves in 1830 indicates that 365 slaves lived at

Butler Island at that time. The census provides the female-to-male ratio

and age distributions, as shown in Table 6.


Table 6
1830 Census of Slaves at Butler island


Age Male Female Total


Under 10 44 49 93

10-24 57 52 109

24-36 36 33 69

36-55 39 29 68

55-100 12 14 26

Total 1388 177 365





Details obtained from plantation records regarding the material

conditions of the Butler slaves are similar to standard slave care prac-

tices throughout the Old South. At Butler Island, these practices

mirror those established for other rice plantations (e.g. Allston 1858;

House 1954a; Phillips "969;i14-130). Generally, these practices include

the following: the provision. of two sets of clothing annually, one for

wiKntor -;d the other for su'im'r ; p antution rations, which consisted

primrily of corn, pork, salted fish, and molasses; a sick house and

medical attention furnished by the overseer or slave doctor; and hous-

ing. Superficially, the records tend to suggest that the Butler slaves










were well provided for,as some recent historians have suggested to have

been the general nature of slavery in the United States (see Foge0 and

Engerman 1974); yet it is apparent from close inspection of these

records, particularly the overseers' letters, that inadequacies were

prevalent. Because the overseers' greatest concern was slave diseases,

most of the information relating to slave care is concern with health

care, and to a much lesser extent diet. References to clothing or

improvements to slave dwellings are few. Thus, it is difficult to assess

the material conditions of slavery at Butler Island on the basis of

plantation records alone. It is intended that the archaeological data

presented in Chapters 4 and 5 will amplify the descriptions of slave

material conditions found in the records. On the other hand, some as-

pects of slave life mentioned in the records have no archaeological

correlates, including slave diseases, mortality rates, disciplinary

problems, and punishment. In the remainder of this chapter these topics

are briefly examined.

The Kings' greatest concern was with the slaves' health care.

Efforts therefore were made to place at Butler Island slaves who were

adapted to the ......y lowlands. Just how these adaptive capacities

were determined is not clear. Through mere observation such an assess-

ment could have possibly been made. Roswell King, Sr., stated:

I landed the Negroes all safe and in good health, . .
I carried them to Tide Island [Butler Island] and you
would have been much gratified to see their rejoicing,
when Qrv found V/ coulcd drink the water out of the
river, which is proof that they have been acquainted
with the same soil & hopefully they wiil be healthy.
. They landed very cheerful & happy, you have no
people that can talk with them. (13 May 1803)









The above staA i--i supports the suggestion that planters deliber-

ately selected slaves adapted to a malarial environment for labor in rice

fields. But it appears that even among members of the old, established

slave population, their abilities to withstand the rice swamp was an

ongoing concern:

It appears Re have few ;-groes here I[ampton] that are
suitable for the rice swamp. We have at least 20 at the
rice island which I would like to move if we had
suitable highland for them. "-., you purchase more of
course some of them will not be so profitable in a rice
swamp and it is necessary we should arrange for land more
suitable. (Roswell King, Sr., 22 August 1813)

Yet sickness was a prevailing problem at Butler Island despite

efforts to place slaves there that were adapted to a malarial environ-

ment. An array of diseases plagued the island, including fevers of all

kinds, measles, intestinal disorders, cholera, influenza, whooping

cough, and others. Of these,cholera, influenza, and intestinal dis-

orders were the most dreaded, and reports of these diseases frequently

followed the occurrence of freshets, particularly in the 1830s,when

freshets became annual events. At that time, slaves were often re-

located to keep down debilitating diseases (Roswell King, Jr.,

15 February 1835, 8 "rch 1835). Evidently, freshets contaminated the

food and water supplies, which when consumed, caused illness and pos-

sibly death. The drinking of2 contaminated water, however, was seem-

ingly a health problem at other times as well: "1his is the sickly

season for Negroes at the rice island. There is 14 to 15 in the hos-

pitni. MW'oe dangerously ill. I believe it is owning to their drinkiii.

river after" (Roswell King, Sr., 30 April 1815) In general, in the

South contaminated water was fre ,;tly the source for epidemics of










of cholera, d ..,try, diarrhea, typhoid, hepatitis (Savitt 1978:59).

By far the greatest sufferers among the slave population at hButler

Island were infants and children. The infant mortality rate was higher

there than at any of the other plantations in the Butler estate:

You wi11 perceive that the mortality among infants is
very high particularly at 16 this place [Butler Island]
the state of the atmosphere is certainly injurious to
infants. The proportion of young people at this place is
far behind St. Siimons. (Roswell King, Jr., 5 January 1829)

Deaths among infants at the Butler estate often comprised 50 percent or

more of the annual deaths as indicated in Table 7. Infant mortality

rates such as these were typical of the late antebellum period (Steckel

1979:95-96). Several conditions generated high levels of slave infant

mortality, but recent studies have suggested that the predominant causes

of slave infant deaths were Sudden Infant Death Syndrome (SIDS), or crib

death (Sutch 1976:283-292; Savitt 1978:122-127), tetani-related

diseases (Savitt 1978:121-122), and occasionally infanticide (Savitt

1978:127-128). Although no definaite cases of infanticide wre reported

at Butler Island, carelessness or neglect on the part of mothers was

given as the cause of death for some infants (Roswell King, Sr.,

16 Jcae 1816; PRosweiI King, Jr., 17 September 1826). iMost deaths among

children between the ages of 2 and 10 years were believed to be duo to

worm infestation (Roswell King, Jr., 10 September 1835). Worms were a

general health prohl-iem among black children in the Oldi South (Savitt

1978:128).

The treatment of diseases was usually handI-'d by the Kings. Yn

cases of major emergencies, a doctor was called upon to tend to the

slaves. The f i 'ntly u-ed medications listed en annual expense lists













C Q0 C(D C) (o r- C CO C) CD CD CD -- Q0 CC'
r- LI) <-- u0 cCO k- Lo r-. co )o Co k-o C) Q0o r-- Vo




L-) r- C U (D .--- Lo cr LU) Co nc Co C)
I- -- (- r --- C(' (J -- -- r- i- "I C-I


k") Co C '- C) C.D r) O U <- c- j CO L- C,- C: o C





CO C) c Cc Co) CD o- -,i c\j r- (N (NJ o i


CO



CO


Cc
n3

4-)
(A

LU



<-7


4-,
OJ
4.




C:
0

(J
C)
C)


0 4-
r~ 0
2

-0


S-c
0

-..)


cc
Cc'




1-I-



(4-
0
(A



ci
CO




4-




(U


r--
cc









CD
(SI






cr<
021'a



2Z
(L!


m C) C-- (N -' O CC-c M2 C) C I c c) c:C- C) :- "
(C\ (N0 C"] o0j (*-- co co en C,4 c0 1 -o (") Cc |




c-o
0
I---

C) cO C-- c, Cc cdi tCj) 'C- C-.- O C) C) (NJ CC ) T

CO 0) 02) CO CO CO C C O, CC" CO CO CO4 CO CO C CO CO)
t-- r- r-- r-- C-.r I- r-,-' r- C- --- r r -- r r- r -'-


LI)

I C ) , CD o --- CO c'' (i C. (Ni 0 C -.- j )
i-. Cc r- r-- r- r-- r-- r- r- r r---





C") N (C t C" C) ('I ULO) -- LO CO C os OQ) L)) LO .) C)














- r U- C)- "C) C C(J L.O cJ 01- o C) I') ,- Ni (cD'
<-j- 't -- r -- C, L C) o -O ) O CO L-O ( 0- 2 () LC)
Ii




-- Cc( C-c i c 0 L- c C r-) c--. i r--
CJ CM i- i- i-* cJ iI


(CD
ci-
C.)









10
0n







iO.


,QJ tU

0.)
L-

F-'




S-
CC3









included rhubarb, Spirits of Turpentine, red lave der, aloes, altum,

magnesia, camphor, peppermint, castor oil, opium, and others. These

medications were typical of the antebellum apothecary (Poste'Il 1951:93-

99). Black folk medicine was discouraged, although some slaves sought

help from "negro doctors" despite objections (Roswell King, Sr.,

10 Au .,.i 1815).

improvements in slave diet were apparently the primary actions

taken to prevent illness. This was especially true for children. Meals

were sometimes prepared separately for them to insure that they were

properly fed:

There is no estate with so fine of crop of young Negroes
as this Butler Island. There are over 100 hominy eaters,
during the summer once or twice a week they are collected
and have an extra meal of fish or pork soup which make
each little -..-- grow in the river swamp, I think.
(Roswell King, Jr., 13 May 1827)

From time to time improvements were also made on the diets of adult

slaves, such as increases in the quality or quantity of plantation

rations. Generally, the amount of plantation ration varied according

to the slave's occupation. Pitchers received the largest amounts of

meat and corn; next in line were the drivers and mechanics; then came

the field workers; and last were invalids and retired slaves whlo re-

ceived the least amount of food (Butler Estate Papers, 1815b). Slaves

kept vegetable gardens and raised chickens at Butler Island, but these

were periodically destroyed by freshets (Roswel l King, Jr. ?83 July

1829). Hunting (.Roswel King, Sr., 28 Januay 1S?9) and fishing also

supple ented the diet, which is indicated by arciheoogical evidence as

wel l.









Despite all indications of an adequate diet, slaves frequently

stole plantation rations stored in warehouses. Although most of these

foods were sold for cash, some of them were probably used to supplement

deficient diets. Stealing corn was particularly a problem: "" jor

Hopkins was telling me one day :.,'" .Pgroes had more to sell than any

Negroes he ever known (I told him they were industrious) and he said

Q y supported Darien in corn & cheap to" (Roswe' King, Sr. 22 Novem-

ber 1806).

Stealing plantation produce, in fact, had become so widespread in

the Altamaha estuary that planters attempted to prosecute local mer-

chants in Darien for buying from slaves:

Tomorrow I am to go to a meeting in Darien to form a
proclamation against the traders there for dealing with
-groes. The planters on St. Simons and in the vicinity
of Darien have formed a coalition to prosecute all dealers
with Negroes. (Roswell King, Sr., 23 September ,?1C)

Because incidences of stealing were frequently reported, stealing was

evidently the number one disciplinary problem among the Butler slaves.

Second to stealing., slave runaways appear to have been a problem.

It is curious that most runaways were intended to be short-term absences

from the plantation. Oftentimes, one or two slaves, usually male, would

go off to another plantation or sneak into Darien without permission

(Rosweli King, Sr., 31 December 1808, 10 May 1818; Roswell King, Jr.,

2 January 1821). Few incidences of permanent runaways are indicated,

and it Eppears that these were always unsuccessful attempts (Roswell

King. Sr.., 12 1;y 1804).

A variety of methods was used to punish slaves for stealing,

running away, and other deviant behavior. The lash was used, but










banishment to hard labor at "Experiment" (Roswell King, Sr., 6 March

1813) or "Five Pound" (Kemble 1961 (1861):270) was prescribed for chronic

offenders.* Punis hment also entailed the denial of fish, meat, molasses,

clothes, rice, and forced labor on Sundays (Rosweil King, Jr., 29

Feburary 1829). Children were submerged in cold water as a means of

punishment. On one occasion a young boy died shortly after he was dumped

in water (Roswell King, Jr., 3 February 1829).

For the most part, overseers' letters provide very little informa-

tion on the family or social life among the slaves. This is unfortunate,

since this is an area in which archaeological data supply little, if

information. A few scattered references do exist regarding fa-Olily

domestic quarrels, conflicts between drivers and workers at the slave

villages, and the religious fervor .,.r. slaves. Perhaps oral history

is the best resource of information for investigating these aspects of

slave life (e.g. Blassingame 1972, 1976; Killion and Waller 1973;

Escott 1979). It has been pointed out that overseers tended to v.,rite

about issues they thought the planter should know about (Bassett tl9c.:

261). Consequently, overseers' letters .-:i-rally reflect concerns re-

lating to slave il niess, death, and disciple inary problems. Given these

concerns, it should be obvious that gaps would exist in interpretations

of slavery derived solely from this resource. Although archaeology may

not be able to provide data relating to the social or religious life of

slaves. it can shed light on the specifics of slave material conditions.


-'In the plantation records there is no reference to the planta-
tion"Five Pound but there are a number of references to "Experirment."
Because they were both apparently located on Little St. Simons, the
author is of the opinion that these names refer to the same plantation.









Virtually little or no information is contained in plantation records
regarding the building materials used in slave housing, the size of

these dwellings, or the wild food resources utilized by slaves, anrd many

other details. These items of slave life are discussed in the remainder

cf this study.


Archaeolopical Resour'ces
Arch e i 1, 1 .. .
Archaeological investigations were undertaken at Butler Iml-nd

in the summer of "1973 under the direction of Dr. Lewis H. Larson, of

West Georgia College. and in the spring and summ ier of 1979 by the author.

Scyclical inundation practices of the Waaterfowl .:r:.. t Progrc..

necessitated that all fieldwcrk be conducted in the spring and summer

months. This schedule, which involves the flooding of most of Butler
Island in the fall and. g'adually removing the water from lfate winter to

early spring, prohibited fieldwork before April. Even in Apri, l, many

areas were still very damp, and some were not completely dry for field-

work activities until June or July. Thus, the time spent at any one

site depended, to a great extent, upon the degree of moisture present at

the site during the time allocated for fieldwork. Fieldwork a-ctiv, es

included surface reconnaissance by foot and by boat, mapping of sites,

and limited excavations at slave settlement #1, the southernmost tidal

milIl, and slave settlement :4 (see Figure 7).


Site As s .. ts
1I Pl S.~A1
T, "es ...... Figure- 7 are pres r rpsu :ly nte eriod

s te. T.hes have bee.n 'I cated and identified by use of a comii-ii nation

of resources incl, uding histri c mraps, pihotographcs, records, aeri al

photographs, inf4r 'nts, an. d archaeological assessments.P Recent '. land

























K
















S-- ^ -.


"0 -





\


C-


iJ~C
I---'


(2)





4-)


(Cl
w
4-)

(I.)

U


C)
C-)
C')

--C--


C










modifications at Butler Island are shown in Figure 8. The most obvious

modifications are U.S. Highway 17, which roughly coincides with the

canal east of Settlement :1 and Interstate 95, which crosses the island

in the vicinity of settlement #2. Also, twentieth-century buildings,

from the Huston occupation, are present at settlement #1. These are

presently used for waterfowl management activities.

It was possible to predict potential areas for site location by the

use of aerial photographs (of course with photographs having better

resolution than that of Figure 8). Because Butler Island is generally

a low-lying, grass-covered marshland, sites tended to be located on

infrequent high spots, which are presently covered with trees and shrubs,

notably, hackberry, persimmon, and fig. These areas of tree groves

were easily identified in the aerial photographs. From surface recon-

naissance, however, not all of these tree clumps were found to be sites.

But all sites located were associated with this tree growth. Addi-

tionally, some sites, such as the slave settlements, were also indicated

in aerial photographs by the detection of antebellum drainage ditches,

which surround and separate these from former rice fields.

The identification for several sites is well documented. The

slave settlements correspond exactly with the area designations I, II,

III, and IV en the 1877 USGS Map (or Figure 6). Historic photographs

provide suggestions for the placement of structures designated in slave

settlement -i. The steam-powered mill is precisely identified by ruins

of the. t'wo chi 's which are prE gently standing. A sketch drawn by

- well King pinpoints the location of the sawmill and both tidal mills

(Roswell King, Sr. 13 January 1813).




Full Text
Figures 29 (upper and 30 (lower). Annular Ceramics.
Key to Artifacts
A. Whiteware, black stenciled design
with red bands on white body
B. Rouletted with swirled bands
C. Mocha design on yellow body
D. Blue stripes on yellow body
E. Circular brown and tan bands on
pearl ware body


175
made into pork for the slaves (Hilliard 1972:57). This meant that
slaves viere likely to have been fed the entire pig. Also, the planta
tion records for Butler Island indicate that this was often the case:
"We have killed 5 bbl of pork this year [which] with the two you sent
out will serve the ditchers" (Roswell King, Jr., 18 March 1821).
Hilliard has made the point that antebellum connotations for the terms
"bacon" or "pork" referred to much more than sides of a hog, often
meaning lean meat free of bone (1972:57-59). If the curing of pork at
Butler Island involved the removal of some or all bones, there would be
no archaeological evidence of hams, shoulders, or ribs. Additionally,
salt pork ("fat back") would not yield bone evidence at the point of
consumption and was likely to have been the most common form preserved
in barrels. The jaw and feet parts recovered may simply represent por
tions of the pig that were not cured but were given to the slaves just
the same.
On the other hand, cattle remains also included head parts as
well as humeri and scapula. These "shoulder roast" remains may indi
cate that the entire animal was slaughtered for the slaves and supplied
to them as fresh meat. Rarely was beef cured in the antebellum South;
it was most often eaten fresh (Hilliard 1972:44). References to the
slaughtering of cattle at the Butler estate suggest that all of the
beef was rationed to the slaves as fresh meat.
The point to be made here is not to underestimate the importance
of beef in the slave diet but to suggest that the higher bone weight
for beef than for pork may reflect differential usages of beef and pork.
Pork, the major domestic animal food resource (as indicated by the


180
waters. Gar is frequently seen near the surface of the water to expel
gases and gulp air (McClane 1978:179-183).
Freshwater catfishes are known for their diverse eating habits.
They are bottom feeding scavangers and most species are fond of turbid
slow moving waters, the exceptions being the blue catfish (Ictalurus
furcatus) and channel catfish (Ictalurus punctatus) which inhabit lakes
and larger rivers that have clean sand or gravel bottoms.
The only two fishes typically found in saltwater were identified:
sheepshead (Archosargus probateocephalus) and drum (Sciam'dae). Sheeps-
head are found in diverse habitats, typically inshore but frequently
around off-shore reefs (Dalhberg 1975:73-74). Generally, they travel in
schools and feed upon rnollusks. They will enter brackish waters and
on. occasion even freshwater (Larson 1970:66).
Drums (Scianidae) are among the most common fish found along the
coast (Dahlberg 1975:69). Although several species of freshwater drum
are known, they are not found along the south Atlantic Seaboard (McClane
1978:194). It is likely that the drum represented in the collection
are saltwater species which also occur in freshwater, such as the red
drum (Sciaenops ocellata) or spotted seatrout (Cynoscion nebulosus).
Both are gregarious bottom feeders found in diverse inshore habitats.
Although the two identified saltwater fishes could have entered
the freshwater environment in the vicinity of Butler Island, their
presence may suggest that the slaves were exploiting the more brackish
habitats of the delta. Evidently, fresh fish from brackish waters were
occasionally provided for the slaves.
This has been one of the hardest freshets in years. . .
The freshet destroyed much of their comforts. I have
-


I
I
Figure 2. The Physiographic Regions of Georgia.
(Adapted from Coleman 1977.)


the Butler family upon the Afro-Americans in the area had lasted well
over a century.
In 1920, Colonel Tillinghast L'Hommedieu Huston, part owner of the
New York Yankees, purchased Butler Island. He built a two-story house,
which is still standing today on the presumed site of the overseer's
cottage. Huston established a dairy farm and planted fruit trees and
vegetables. His special interest, however, was the cultivation of ice
berg lettuce. He revamped the old rice irrigation system and used it
for his fields. Huston remained at Butler Island until his death. In
1949, Butler Island was sold to Richard J. Reynolds, the tobacco million
aire, who made additional improvements to lettuce cultivation (Vanstory
1970:90). In 1954, Reynolds donated Butler Island to the state of
Georgia for the development of a wildlife preserve.
Plantation Management at Butler Island,
1802-i860
The overseer on the Carolina-Georgia coast has been described as
the most important "single element in the management hierarchy of the
rice belt" (Scarborough 1964:18). At least, this appears to have been
the case at Butler Island, where the Roswell King family successfully
managed the estate for over thirty years during the Butler family's
periodic absences.
Absenteeism as-a practice among rice planters was established
fairly early by colonial predecessors. This tradition developed, in
part, out of a need to escape the hot, malarial season as it was
generally believed that "miasmas" caused the disease. The ownership of
multiple plantations also necessitated the periodic absence from one or


I certify that I
conforms to acceptable
adequate, in scope and
Doctor of Philosophy.
have read this study and that in my opinion it
standards of scholarly presentation and is fully
quality, as a dissertation for the degree of
Ronald C. Foreman
Associate Professor of English
This dissertation was submitted to the Graduate Faculty of the Department
of Anthropology in the College of Liberal Arts and Sciences and to the
Graduate Council, and was accepted as partial fulfillment of the require
ments for the degree of Doctor of Philosophy.
December 1980
/
Dean, Graduate School


165
All of the nondomestic food resources identified are found in the
delta section of the tidewater in either brackish or freshwater habitats,
although, of course, many food resources not identified were possibly
also consumed. Therefore, it is difficult to ascertain which animal
species were utilized most frequently. At the same time, fishes and
turtles are among the dominant animal species found in the delta, and
the faunal collection suggests that these were the predominant non
domestic species consumed. Birds, which are greatly underrepresented
in the collection are also plentiful, but the slaves' ability to procure
them was dependent upon their access to firearms. Also, birds, par
ticularly ducks and other migratory fowl, are generally more seasonal
resources than fishes or turtles. Consequently, birds may not have been
consumed as often as the fishes and turtles. For the most part, mammalian
food resources are more restricted in the delta than in other habitats
of the coast. This may explain the infrequent presence of nondomestic
mammals within the collection.
In conclusion, the archaeological evidence of slave diet at
Butler Island and other coastal plantations indicates that domestic
food resources played a greater role in slave diet than nondomestic food
resources at both rice and long-staple cotton plantations. Yet the
presence of nondomestic food remains implies the need to supplement
monotonous and perhaps diffident plantation rations. Furthermore,
slaves supplemented their rations with food resources available to them
in nearby habitats.




Figure 27. Red Coarse Earthenwares.
Key to Artifacts
1. Brown-glazed
2. Black-glazed
3 Brown-glazed
4. Black-glazed
5. Unglazed lip
large bowl
chamber pot
shoulder fragment
basal fragment
of storage jar


The archaeological investigation of African survivals at Afro-
American sites has resulted in very little definitive evidence to date.
This may be the result of a number of factors. Although his discussion
is specifically directed to slavery sites, Otto details three processes
that limited the re-creation of African materials. These include
(1) selection and simplification, (2) availability and substitution,
and (3) differential acculturation (Otto 1975:375). It is very likely
that northern free blacks also substituted traditional items with
available ones. The archaeological record is another factor. African-
styled artifacts such as wooden objects or basketry are not likely to
be preserved (Otto 1975:382). Regardless of the reason for the scarcity
of African-styled artifacts at black-occupied sites, research designed
solely to uncover Africanisms ignores the fact that Afro-American
culture, like all cultures, is an adaptive system (Schuyler 1980:2).
A data base of Afro-American subsistence patterns is slowly emerg
ing from diverse archaeological resources. These data are crucial
to generating future hypotheses regarding Afro-American cultural
patterns. A major problem has developed in the identification of
Afro-American economics. To what extent are the subsistence patterns
at black sites indicative of Afro-American ethnicity or a culture of
poverty (Kelly and Kelly 1980)? Only when archaeological data
are available from poor white subsistence farmers can the archaeo
logical visibity of low-status Afro-Americans be ascertained
(Baker 1978). Presently, these nonplanter whites are archaeologically
unknown.
... ' . - " : '


N
r-
c.
Zone A
Zone B
Zone C
n
EY
SILTY LOAM
p p V I
'Ll ii % L I O La.
AYEY LOAM
l METER
CLAY WITH
:llow flecks
Figure 15. Stratigraphic Profile, Settlement #4.
O


83
Despite all indications of an adequate diet, slaves frequently
stole plantation rations stored in warehouses. Although most of these
foods were sold for cash, some of them were probably used to supplement
deficient diets. Stealing corn was particularly a problem: "Major
Hopkins was telling me one day your Negroes had more to sell than any
Negroes he ever known (I told him they were industrious) and he said
they supported Darien in corn & cheap to" (Roswell King, Sr., 22 Novem
ber 1806).
Stealing plantation produce, in fact, had. become so widespread in
the Altamaha estuary that planters attempted to prosecute local mer
chants in Darien for buying from slaves:
Tomorrow I am to go to a meeting in Darien to form a
proclamation against the traders there for dealing with
Negroes. The planters on St. Simons and in the vicinity
of Darien have formed a coalition to prosecute all dealers
with Negroes. (Roswell King, Sr., 23 September 1808)
Because incidences of stealing were frequently reported, stealing was
evidently the number one disciplinary problem among the Butler slaves.
Second to stealing, slave runaways appear to have been a problem.
It is curious that most runaways were intended to be short-term absences
from the plantation. Oftentimes, one or two slaves, usually male, would
go off to another plantation or sneak into Darien without permission
(Roswell King, Sr., 31 December 1808, 10 May 1818; Roswell King, Jr.,
28 January 1821). Few incidences of permanent runaways are indicated,
and it appears that these were always unsuccessful attempts (Roswell
King, Sr., 12 May 1804).
A variety of methods was used to punish slaves for stealing,
running away, and other deviant behavior. The lash was used, but


6
archaeological research. Frequently, regional plans are used as
planning devices in cultural resource management (King et al. 1977:
145-173; McGimsey and Davis 1977:47; Schiffer and Gurnmerman 1977;
121-131). As planning tools, regional plans involve the review of all
known records of the project area, including historical records, sur
vey and excavation reports, and other existing field data as well as
informant consultation. The overview should summarize the present
knowledge, evaluate the available evidence, estimate the resource base,
and attempt to forecast long-range regional development and ongoing
destructive processes (Schiffer and Gurnmerman 1977: 12-13). Although
regional plans are only as good as our present-day knowledge permits
(Glassow 1977), their utility as rudimentary planning tools is essen
tial in historic preservation.
From the above discussion four concepts of regional archaeology
are identified. The first is a unit of analysis which represents
the broad cultural milieu of human activities for a cultural system.
The second is a method of discerning spatial organizations within a
cultural system. The third is a framework for the collection and
synthesis of cultural and ecological data within a defined geographic
area. Finally, the fourth is the explanation for the patterns thus
defined.
In this study, the first, third, and fourth concepts provide the
orientation for analysis and interpretation. Coastal Georgia, speci-
cally the barrier islands and the adjacent river deltas, forms the
geographic parameter. The temporal parameter is the nineteenth cen
tury (1800 to 1861). The unit of archaeological analysis for the
investigation of the slave system is the slave community site.


98
On the basis of the artifacts recovered from other slave sites
in coastal Georgia, the antebellum materials at settlement #1 are sug
gestive of the slavery status. The small sample of artifacts recovered
and their disturbed nature, however, prohibits the use of these
materials in most of the interpretations offered in this study. Per
haps the greatest contribution of the archaeological resources un
earthed at settlement #1 is to provide evidence of the earliest his
toric occupation at Butler Island. A few eighteenth-century artifacts
were recovered, including fragments of slipware (Noel-Hume 1976:98-108)
(see Figure 11) and an eighteenth-century Spanish reale (see Figure 12).
The precise date of the coin is unclear, but the first two digits, 1 and
7, suggest that the coin was manufactured in the 1700s. These arti
facts provide the only vestiges of a pre-nineteenth century occupation
at Butler Island. Further excavations at settlement #1 are recom
mended to uncover additional evidence of this occupation.
Slave sites #2 and #3 were not tested. From field inspection
and aerial photographs, it was decided that slave site #2 was com
pletely destroyed by the construction of Interstate 95. In the vicin
ity where 1-95 crosses Butler Island, a tremendous quantity of sand was
deposited after dredging activities associated with the highway's con
struction. On the surface of this sand, antebellum materials were
found, including pipe bowls, shell-edged pearlware, and Savannah grey
bricks. The presence of these artifacts to this specific area supplied
the only archaeological evidence of slave settlement #2. Slave settle
ment #3 was surveyed very briefly. Two chimney falls (identical to the
ones discussed at slave settlement #4) were located. Unfortunately,


Figure 40. Unidentified Tools.
Designations
1. U-shaped object
2. Mold for lead shot


100


197
" . >
'MMTfc
r
tkeas
j


125
possible that the pigs and chickens kept by the slaves consumed a great
deal of the organic refuse. At any rate, no intentional patterns of
trash disposal were identified.
The large areas of the site in which no cultural materials were
uncovered were possibly used for gardening. The stratigraphy of the
posthole tests taken from the area above the structures (see Figure 14)
suggests that the ground has been turned by a pi gw more intensively than
in areas near the dwellings. This overturned soil may be the artifact
of postoccupational plowing which is evident from visible plow furrows
all over the site. The plowing activity appears to have been minimal as
indicated by little scattering of the artifacts.
The spatial arrangements at slave settlement #4 provide some indi
cations of an accommodation to rice culture. The overall layout of the
site is typical of tidewater patterns in general, and these patterns
have been observed at rice plantations (Olmsted 1968 (1856):416-417, 421-
422). On the other hand, the absence of wells, the specialized struc
tures associated with rice production, and the use of drainage ditches
for refuse disposal and possibly for human elimination appear to have
been practices developed specifically for slave life in a deltaic, marsh
land habitat. Limited testing, however, cannot be ruled out as a factor
which may have skewed the findings used for the interpretations offered
here. Additional excavations are needed to demonstrate that these
archaeological resources of community organizations reflect slave life
t
on the Georgia rice coast.


REFERENCES
Adams, Robert McC.
1963 Archaeological research strategies: Past and present. Science
160:1187-1192.
Alderi, John R.
1971 The First South. Louisiana State University Press, Batton Rouge.
Allston, R. F. W.
1846 Rice. DeBows Review 1:320-357.
1854 The essay on sea coast crops. DeBows Review 16:589-615.
1858 Notes on the management of a southern rice estate. DeBows
Review 24:324-326.
Anderson, Adrienne
1968 The archaeology of mass-produced footwear. Historical
Archaeology 2:56-65.
Anonymous
1S65? History of Altama Plantation, 1763-1965. (Pamphlet, no
copyright).
nd Untitled manuscript. On file, Altamaha Waterfowl Management
Area Office, Butler Island, Georgia.
Anthony, Carl
1976 The big house and slave quarters, part I: Prelude to
New World architecture. Landscape 21(1):8-l9.
Armstrong, Margaret
1S33 Fanny Kemble: A Passionate Victorian. Macmillan, New York.
Ascher, Robert
1574 Tin can archaeology. Historical Archaeology 8:7-17.
Ascher, Robert and Charles H, Fairbanks
1971 Excavation of a slave cabin: Georgia, U.S.A. Historical
Archaeology 5:3-17.
Austin, Amory
1893 Rice: Its cultivation, production and distribution in the
United States and foreign countries. U.S. Department of
Agriculture Bulletin #6. Government Printing Office, Washington,
d.c:
229




10
this was why the plantations of the Georgia coast have been described as
"patriarchal" (Wylly 1910:12; Lewis and Huie 1974). Another contribut
ing factor was the infrequency in which planters of the coast migrated
to the new lands of the West. With a few exceptions, slaveholders on
the coast of Georgia established their roots in the late eighteenth
century and remained there until the Civil War. It was "only in the tide
water areas of Georgia [that] wealth [had] matured for several genera
tions" (Boney in Coleman 1977:174). This may have been the result of
special local ecological factors. ^
This study assumes that the previously described variables sub
stantially influenced a development of slave behavior patterns pecu
liar to the Georgia coast. Some of these.patterns are believed to be
archaeologically discernible. These archaeological indices are dis
cussed in detail in subsequent chapters.
The Archaeoiogy of Afro-Americans
Archaeologists have become increasingly involved in the investiga
tion of sites formerly occupied by black Americans. Because few his
toric sources have been written by blacks, this research interest
developed out of a need to expand upon traditional interpretation of
past black American lifeways. Contemporary eighteenth and nineteenth
century accounts relating to black Americans were written by whites.
As a result, these are often very prejudiced views of Afro-American
cultural life. The major objective of Afro-American archaeology has
been to investigate aspects of black American life not available in
documents.


CHAPTER 1
INTRODUCTION
The anthropological study of Afro-American slavery has increas
ingly used archaeological data as a primary source for interpreta
tion. Because traditional historic sources are subject to problems of
falsification and bias, archaeological data are frequently used to
supplement historic records. It has been suggested that archaeology be
the key to writing the cultural history of the "inarticulate" (Ascher
1974:10-12). Slavery sites have been investigated specifically for this
reason. As a consequence of the widespread illiteracy among slaves,
few documents written by slaves exist. Most of our understanding of
slavery has come from accounts written by whites. Inferences derived
from the archaeological record of slavery can provide insights into
slave lifeways and slave worldview. Although planter records," travelers'
accounts, and oral interviews are used extensively here to supplement the
archaeological record, this study assumes that archaeological data are
one of the best indices of slave behavior.
The purpose of this study is to offer a model capable of dis
cerning slave behavioral patterns in coastal Georgia. Slavery on the
Georgia coast was characterized by distinctive demographic, environmen
tal, economic, as well as historical factors which were absent in the
interior. Undoubtedly, these conditions affected the quality of slave
life, slave activities, and ultimately the development of slave cultural
patterns.
1


171
Identified
Table 12
Fauna from Slave Site #4,
Butler Island
Species
Common Name
Bone Weight
in Grams
Percent
Amia calva
Bowfin
21.1
.96
Lepiososteus sp.
Gar
9.5
.43
Si 1uriformes
Catfishes
13.5
.61
Icaturidae
Freshwater catfish
23.0
1.04
Archosargus
probatocephal us
Sheepshead
3.0
.14
Scianeidae
Drum
4.2
,20
Unidentified fish
62.8
r\)
o
Kinosternon sp.
Mud turtles
6.0
.23
Chrysemys sp.
Coolers and sliders
13.0
.59
Trionyx ferox
Softshell turtles
14.1
.64
Unidentified turtle
252.5
11.50
Squamata
Snakes
1.2
.05
Aythya sp.
Diving ducks
2.0
.09
Gall us gal I us
Domestic chicken
5.0
.23
Unidentified birds
25.5
1.16
Rodentia
Unidentified
rodent
2.0
.09
Cri citidae
New World rats
and mice
6.0
.28
Procyon lotor
Raccoon
2.5
.11
Artiodactyl
Even toe
Ungulates
200.0
9.11
Sus scrofa
Domestic and
feral pig
205.8
9.38


limitations: first, with the exception of one study (MacFarlane 1975)
excavations of slave settlements have sampled only a small portion of
the site, usually one or two structures (Ascher and Fairbanks 1971;
Fairbanks 1974; Otto 1975). Second, as a result of this sampling,
archaeological evidence of the internal social organization within the
slave community has not been uncovered. Third, slave sites investi
gated have been exclusively from long-staple cotton plantations. Com
parative data from the other dominant cash crop of the coast, rice, have
been lacking.
Investigations at the south end slave cabins at the Cannon's
Point plantation on St. Simons Island, Georgia (MacFarlane 1975) were
the first attempt to excavate an entire slave community site. This
investigation provided details of slave housing, household items, cloth
ing, personal items, and food resources. Little or no data, however,
were reported regarding the community organization, specialized crafts,
or social hierarchy.
Documentary resources of the Butler Island rice plantation indi
cate that each slave settlement had plantation drivers, craftsmen, cooks
as well as field laborers. It was hypothesized that testing an entire
slave community should uncover archaeological evidence of these special
ized roles and status differences. Initially, the objective of archaeo
logical investigations at Butler Island was twofold: first, to uncover
archaeological evidence of status difference and craft specializations
within the slave community; second, to compare the archaeological re
sources of slavery at a rice plantation with those of long-staple cotton
plantations, in an attempt to ascertain similarities and differences


cultivating rice as early as the fifteenth century (Latter 1969:17).
A preference for slaves from rice growing regions of West Africa was
established early by white Carolinians. These areas included the
Congo-Angola area (Heyward 1932:172-175), Senegambia area (Clark
et al. 1962:226-239; Lauer 1969:47-48), and the "Windward" or "fold"
coast of present-day Sierra Leone and Ghana (Corry i960). Slaves from
the rice coast of Africa were advertised by many firms. Between 1732
and 1765, Charleston alone handled "53 ship loads from Gambia and 29 fr
the Gold coast" (Lauer 1969:48).
Additional evidence for the presence of slaves from West African
rice growing areas is provided by the prevalence of African personal
names. The names "Banyun," "Bram," and "Bu'lunda" among slaves and
their later descendants on the Carolina-Georgia coasts strongly suggest
that the people from the rice growing areas of Senegambia were repre
sented among the slave imports (Lauer 1969:48).
Besides the actual physical labor, Afro-Americans made signifi
cant contributions in the planting end processing of rice. The motions
used in planting rice seeds such as the pressing of a hole with the
heel and covering the seed with a foot were also practiced in West
African rice planting. The flat, wide, coiled winnowing baskets were
trade by slaves and were typical of West African ones both in style and
techniques of manufacture. Methods of pounding and polishing rice by
hand were possibly told to white planters by their African slaves
(Wood 1974:61-62).
Other factors, however, may have established a need for African
slave labor. Briefly, these include the abundance of black labor,


22
species of oak, pignut hickory, red cedar, southern magnolia, cabbage
palm, wax myrtle, saw palmetto, and many herbs and vines. Faunal
resources in these highlands arc also abundant. Whitetail deer,
oppossum, raccoon, and cottontail rabbit are among the dominant species.
The freshwater swamp is an important breeding ground for reptiles and
amphibians, and the dominant vegetation is cypress.
Fewer data are available for delta ecology than the other sections.
The delta embraces those areas where freshwater rivers enter saltwater.
Deltas are primarily freshwater but are affected by the tides. The
area is flat and is frequently inundated. Most vegetation is water-
tolerant. Near the ocean, the area is a grass-covered marsh; where
there is little or no intrusion of saltwater, the delta is composed of
cypress swamp (Larson 1970:35). Much of this swamp was cleared for rice
cultivation. Deer, otter, and raccoon occur in portions of the delta,
but the most distinctive faunal resources are the anadromous fish such
as glut herring, striped bass, sturgeon, and the American shad.
The narrow belt of
the tidewater, which includes the barrier
islands and the adjacent river deltas,is part of the oldest
region of Georgia. The sandy loams of the sea islands were
agricultural
the "most
productive lands" of the coast. During the colonial period, these
lands yielded crops of corn, indigo, potatoes, and vegetables. After
the Revolutionary War, these lands were devoted almost exclusively to
the production of long-staple cotton (Bonner 1964:1).
By the late eighteenth century, reclamation of the wet,
deltaic soils for rice cultivation had begun. Tidewater rice
alluvial
culture
was limited to the areas located far enough upstream to avoid contact




59
more of these productive units (Bridenbaugh 1952:69). In the early days,
planters traveled to the North or to Europe, while others migrated to
their town houses in Charleston and other coastal cities. Seashore
settlements and pine!and, piedmont, and mountain villages began to sup
plement the coastal towns as summer resorts later in the nineteenth
century (Brewster 1970:109). The degree of absenteeism, however,
apparently varied along the rice coast. For example, in the Georgetown
district in South Carolina, one planter rioted that of 111 rice planta
tions in the area only 51 of the owners actually lived on their planta
tions, and of that number all were absent during the crop season from
May to November (/"Allston quoted in Cathey 1956:142). Some planters, who
could not afford a "dual existence" or who were too far from Charleston
or Savannah, remained on their estates throughout the year or most of
the year (Bridenbaugh 1952:70). In the Altahama Basin, rice planters
spent a major portion of the year on their rice estates and moved to
nearby pi riel and retreats or barrier island plantations for the summer
months. Often, they commuted each day to keep a watchful eye on the
progress of the crop and other management problems (House 1954b;10). The
Butlers were the only exception to this rule (see Wylly 1910:48-53).
They were never full-time residents of Georgia. Because they were
generally absent from their Georgian estate, the overseers regularly
corresponded with them. Fortunately, a large portion of this corre
spondence has survived to the present day.*
*The Roswell Kings Correspondence is found under two manuscript
collections of the Pennsylvania Historical Society, in Philadelphia:
The Butler Family Papers (primarily Boxes 9, 10, 11, 12), and the
Ulster Family Papers (The Butler Family Section, Boxes 31, 33, 34).


76
Table 5
An
Approximate List of Slaves, by Age and Occupation,
at
Butler Island, 1803-1845: Plantation Specialists
Occupation
Name
Age
Blacksmiths
Jack
70
Sammy
50
John
30
Sawyers
Andrew
30
Sambo
55
Masons
Pompe
55
Gorham
78
Jacob
---
Carpenters
Sunday
56
John
50
Bram
5-
Harry
3-
Ned
3-
Primis

Frank
--
Sambo

George
--
Abraham
37
George
47
Randy
30
A1 ex
28
Issac
18
Shoemakers
Robi n
49
Bram
48
Ditchers
Abraham
30
Abraham Carter
30
Justice
26
Justice (younger)
21
Sam
22
Joe
25
Robin
40
Toby
22
Quash
25
Sancho
25
Peter
29
Tuco
20
Craftsmen
Ditchers
24
12


247
Van Rensselaer, Susan
1978 Banded creamware. In English Pottery and Porcelain: A His
torical Survey, edited by Paul Atterburn, pp. 240-244.
Universe Books, New York.
Vanstory, Burnette
1970 Georgia's Land of the Golden Isles. The University of Georgia
Press, Athens.
Vlach, John Michael
1978 The Afro-American Tradition in Decorative Arts. The Cleveland
Museum of Art, Cleveland, Ohio.
Wagret, Paul
1968 Polder!ends. Barnes and Noble, New York.
Watkins, Malcolm
1970 Artifact From the Sites of Three 19th Century Houses and
Ditches at Darien Bluff, Georgia. University of Georgia
Laboratory of Archaeology Series Report #9, Athens.
Weiss, Mark L. and Alan E. Mann
1975 Human Biology and Behavior: An Anthropological Perceptive.
Little, Brown, and Company, Boston.
Wells, Tom Henderson
1967 The Slave Ship Wanderer. University of Georgia Press, Athens.
Wilms, D. C.
1972 The development of rice culture in 18th century Georgia.
Southeastern Geographer 12:45-57.
Wing, Elizabeth and Antoinette B. Brown
1979 Paleo-Nutrition: Method and Theory in Prehistoric Foodways.
Academic Press, New York.
Wolf, Eric
1959 Aspects of plantation systems in the New World: Community
sub-cultures and social classes. In Seminar on Plantation
Systems of the New World. Social Science Monographs 7:136-
147. Research Institute for the Study of Man and the Pan
American Union, Washington, D.C.
Wood, Peter H.
1974 Negroes In Colonial South Carolina From 1670 Through the
Stono Rebel lion. W. W. Norton, New York.
Lylly, Charles Spalding
1910 The Seed That Was Sown in the Colony of Georgla: The Harvest
and the Aftermath, 1740-1370. Neale Publishing, New York.


40
the St. Johns River in Florida (Hammond 1897:19). In the early forties,
it was discovered that the crop 'was adaptable to sea marshes if these were
reclaimed and to interior lands as much as 100 miles from the Atlantic
Ocean. This discovery led to the expansion of the industry into the
interior lands of Georgia and Florida. By 1858, Florida had become a
leader in the production of sea-island cotton, and the volume of exports
for the staple increased by 50 percent as a result of the expansion in
cultivation (Gray 1941:733-734).
Long-staple cotton was strikingly distinctive from the short-
staple which was cultivated elsewhere in the Old South. As the name
implies, the fiber was longer than the short-staple. The fiber length
of sea-island cotton ranged from 1 1/2 to 2 inches as contrasted with
5/8 to 1 inch for the short-staple (Gray 1941:731). The quality of the
two was also dissimilar. The fine, silky long-staple was used in the
manufacture of lace, thread, and cloth of silky luster. On the other
hand, short-staple cotton was used primarily for making coarse cotton
cloth (Gray 1941:731; Phillips 1929:91). Moreover, the long-staple
required a longer growing season, more hands per acre, and more tedious
methods for cultivation, harvesting, and processing (see Otto 1975:53-61
for a detailed comparison of the two varieties of cotton).
Methods of Planting and Cultivation
The soil was prepared for planting in Februa
Soils varied from light sandy loams to heavy clays
were preferred (Phillips 1969:271). On lowlands "
that employed in rice culture was often necessary
y or early March.
, but the sandy ones
ditching" similar to
to drain an area before


169
that this flint was used to make the gun flints. Flint is not native to
the Georgia coast. Moreover, black,or Dover, flint is found only in the
Old World. Therefore, its presence at Butler Island indicates that it
was brought there. Possibly it was obtained from nearby ballast dumps.
Ballast materials frequently occur at coastal sites (e.g. Jones 1976).
How this material turned up at the slave sites is unclear. A sugges
tion is provided by Roswell King, Sr., who indicated that the ballast
was used in repairing the sugar works (Roswell King, Sr., 17 March 1816).
If ballast was periodically brought to Butler Island, the presence of
the Old World flint at the slave settlements would be explained.
Food Resources
Unfortunately, the food bone remains at Butler Island were found
to be very poorly preserved. Although acidic, waterlogged conditions
are favorable for the preservation of hide, leather, hair, and wool,
bone is preserved best in alkaline situations (Cornwall 1958:69). The
fluctuating water conditions presently found at Butler Island provide an
additional unfavorable condition for bone preservation. Such situa
tions are usually characterized by increased organic decay.*
Besides unfavorable bone preservation conditions, a large portion
of the collection had been severely charred, apparently by exposure to
very hot, low-oxygenating fires, as indicated by the whitish blue color
of many fragments.** (Cremated bone often exhibits similar physical
*Larry Banks, Chief, Army Corps of Engineers, Southwestern
Division, Dallas, Texas, personal communication, I960.
**Elizabet.h Wing, Florida State Museum, Gainesville, Florida,
personal communication, 1980.


TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ..... ...... iv
LIST OF TABLES ix
LIST OF FIGURES
ABSTRACT
XI1
CHAPTER
1. INTRODUCTION .......... 1
Regional Analysis in Archaeology ... ... 3
Regional Approaches to the Study cf Slavery 7
The Archaeology of Afro-Americans ..... 10
Problems and Hypotheses ..... ..... ¡4
2. PLANTATION AGRICULTURE IN THE GEORGIA TIDEWATER ...... 20
Environmental Summary and Introduction .... 20
The Emergence of the Plantation System in Georgia .... 23
The Rice Industry 26
Rice and Slavery 26
Methods of Irrigation and Planting Rice ........ 31
Harvesting, Milling, and Marketing .... 35
Long-Staple Cotton Culture 39
Methods of Planting arid Cultivation 40
Harvesting, Processing, and Marketing ........ 41
Summary 43
3.HISTORICAL AND ARCHAEOLOGICAL INVESTIGATIONS AT BUTLER
ISLAND 46
Ownership and Occupation 43
Plantation Management at Butler Island, 1802-1860 .... 58
Records of Slave Life 69
Archaeological Resources .... ..... 85
Site Assessments 85
Slave Village Sites 93
4.SETTLEMENT PATTERNS OF COASTAL SLAVE COMMUNITIES 109
The Determinants of Plantation Settlement Patterns ... 109
Slave Community Organizations at Butler Island ..... ¡14
vi


139
Cypress house constructions may have been found more often on the rice
coast, but cypress was likely to have been used at the barrier island
plantations as wel1.


mmmmm mm aajggjg
43
Archaeological investigations were undertaken at Butler Island in
response to a need to locate and identify archaeologically sensitive areas
within the confines of waterfowl management activity. Certain manage
ment practices such as seasonal flooding, plowing, and dredging activities
would adversely affect or destroy archaeological sites. In past years
on several occasions, archaeological materials have been unearthed as a
consequence of these practices. To prevent these practices, funds were
obtained for survey, inventory, and testing at Butler Island from the
grant-in-aide program of Heritage Conservation and Recreation Service
(HCRS),U.S. Department of the Interior, and administered by the Historic
Preservation Section, Georgia Department of Natural Resources. Nine
sites were located and identified. All were found to be historic sites,
but three have prehistoric components.
In this study the slave village sites are the subject of discussion,
although the other plantation period sites are briefly described. A de
tailed description of all the sites will be part of another archaeological
report (Singleton nd).
Another major aspect of the Butler Island project has been the
archival research. Documents relating to the Butler family ownership
of Butler Island are voluminous. These records have not only provided
the background data presented in this chapter, but have also been util
ized in the identification, dating, and interpretation of the archaeologi
cal resources.
Ownership and Qccupation
Unfortunately, the first colonial grantees of Butler Island
could not be ascertained. It is possible that the Lochlan McIntosh


le
st
o
Toe* !
i ca i
-s* North
3IS
zizi
20
meter.
nqure 9,
L X l
tenc ot
:avc
ion:
iti]ement
#1
cn


39
the interest in rice was abandoned. Finally, a major storm destroyed
the last remaining vestiges of rice culture at the turn of the century
(Vanstory 1970:83). The following passage taken from an article in the
Darien Timber Gazette summarizes the sentiment toward rice growing among
post-bellum planters in Georgia after a devastating flood:
the loss will be a heavy blow to many of the planters
especially where money was borrowed to plant with. Several
have already declared they will not plant any more as the
crop is so uncertain. Every year the acreage is decreasing
end the upriver planters are nearly discouraged.
(27 August 1887, Darien, Georgia)
Long-Staple. Cotton Culture
Like rice, the production of long-staple cotton was restricted to
the coastal fringes for most of the antebellum period. It became known
as "sea-island cotton" because the staple was believed to degenerate
when removed from the influence of saltwater (Alisten 1854:593). Its
introduction to the coastal states, unlike that of rice, occurred in
Georgia not South Carolina. Initially grown in the Caribbean, long-staple
cotton made its way to coastal Georgia through expatriated Georgians who
had made their home in the Bahamas after the Revolutionary War. The
success of the cotton in the Bahamas led the former Georgians to send a
few bags of the seed to their friends and relatives in Georgia (Coulter
1941:65-68: Hammond 1897:16-17). Gradually, the fruit yielded a finer
and better cotton titan that grown in the Bahamas (Johnson 1930:24-25).
By 1789, substantial quantities were produced in Georgia and its cultiva
tion spread to nearby South Carolina.
For most of the antebellum period, sea-island cotton was grown
exclusively along the narrow strip from Charleston, South Carolina, to
W vi S : W '.**' --5 : y '.,y SSSjaBJ*? J a ;- peegS-


144
Table 3
Vessel Forms of Ceramic Styles
Ceramic Style
Vessel Form
Number
Coarse red earthenwares
Ungazed
Storage jar
1
1
Black lead glazed
Indeterminate
1
Chamber pot
1
Brown lead glazed
Large bowl
1
'ined earthenwares
Undecorated whiteware
Chamber pots
3
Serving bowls
and
cups
4
Plates/platters
4
Mug
1
Pitcher
1
Transfer printed pearl ware
Serving bowls
or
cups
2
Platter
1
Edged Ware
Shell edged pearl ware
Plat.es/plattei
rs
6
Molded edged whiteware
Plate/platter
1
Handpainted
Pearl ware underglazed blue
on white
Serving bowls
2
Pearl ware polychrome '
Whiteware polychrome
Shallow bowls
or
cups
2
Bowls or cups
2
Annular
Circular bands
Pearl ware
Serving bowls
and
cups
3
Whiteware
Serving bowls
and
cups
2
Yellow paste
Serving bowl
1
Rouletted
Pearlware
Serving bowl
1
Whiteware
serving bowl
1
Finger painted
Serving bowl
1
Mug
i


4
first to define this usage of regional archaeology:
the detailed and systematic study of regions that can be
expected to have supported cultural systems. The extent of
such regions will vary because it is recognized that cul
tural systems differ greatly in the limits of their adap
tive range and milieu. As cultural systems become more
complex, they generally span greater ecological ranges and
enter more complex, widespread, extra-societal interaction.
The isolation and definition of content, the structure, and
the range of a cultural system together with its ecological
relationships may be viewed as a research objective.
(1964:426)
Central to the above definition is the view that culture is an
adaptive system. As an anthropological concept, adaptation refers to
the ability of a population to adjust to the environment by develop
ing effective social and technological structures. Adaptation is an
ecological process (Steward 1955:30). The interaction between human
behavior and the environment in which culture is the mediating
variable is fundamental to ecological approaches in anthropology
(Harris 1963:659). Cultural ecology as a research strategy involves
three procedures: first, it analyzes the interrelationships between
technology and the environment; second, it analyzes behavioral patterns
involving the exploitation of a particular area (the region) by a
particular mode of technology; third, it ascertains the extent to
which the behavior patterns involved in exploiting the environment
affect other aspects of culture (Steward 1955:40-1). The utility of
regional analysis in cultural ecology is very clear. As the locus
occupied by a cultural system, the region provides the unit of analysis
for ascertaining human ecological relationships.
Because the region is the locus of human activities, regional
archaeology requires the investigation of diverse activity areas within


Table 14
The Relative Weight of Domestic Species at Butler Island
Species
Weight
Percent
Gallus gallus
Domestic chicken
5.0
.8
Sus serofa
Pig
205.8
31.5
Bos taurus
Domestic cattle
442.5
67.7
Total
653.3
100.0
coast, and these beef rations tended to be larger than rations of pork
(Hilliard 1972:59, 130-131). Also, at Butler Island, more cattle was
available than pig from the plantation livestock. Although some of the
cattle included oxen, which were used as draft animals, they were peri
odically slaughtered to feed the slaves: "Buy 20 more steers to add to
the oxen for Christmas beef" (Roswell King, Sr., 15 May 1813). The
reference to "Christmas beef" may indicate that beef was only provided
on very special occasions.
Perhaps the confusion between the archaeological evidence and the
Butler estate records lies in differential butchering and curing prac
tices of cattle and pigs. The pig remains identified within the sample
included primarily mandible, teeth, and metapodial fragments, suggest
ing that the remainder of the pig was utilized elsewhere. While it is
possible that the Kings kept the remainder of it "on plantations that
were overseer operated, the great majority.of the hogs were killed and


143
3. As for decorated tablewares, annular decorated ceramics pre
dominate, whereas the transfer printed styles are minority
items.
Because these patterns have been observed consistently at coastal
cotton plantations, it is hypothesized that similar ceramic patterns
would be observed at rice plantations. It is assumed that slaves at
both rice and cotton plantations had a similar access to ceramics and
other material possessions. This slave ceramic pattern would suggest
either that special purchases were supplied to slaves or that perhaps
slaves purchased ceramics made available to them from local merchants.
Ceramic styles and vessel forms
For the most part, the ceramics uncovered at Butler Island were
utilized for the serving of food or the storage of it. The exceptions
are those vessels used for hygienic purposes. The range of vessel forms
is given in Table 8. All the ceramics have been described in detail
elsewhere and are very briefly summarized here.
Red coarse earthenwares. Red paste ceramics have been associated
with dairy and general purpose kitchen activities. Their manu
facture is well documented for New England and adjacent areas ("Stradling
and Garrison 1977; see related to redwares); Less information,
however, is available for the South. In Georgia, these ceramics were
manufactured in Savannah during the mid-eighteenth century, but
the location of later kilns is not clear (Ketchem 1971:31). At Butler
Island, three styles were identified, including unglazed, brown glazed,
and black glazed varieties (see Figure 27). All vessels appear to have
r storage or some other utilitarian purposes.
been used for eithe


138
At least one padlock was found in association with each structure
excavated. Structure one alone yielded three of these. Two of the pad
locks had brass keyhole covers, which establishes a date of 1840 for
their manufacture (Noel-Hume 1976:251). The frequency of these padlocks,
particularly at structure one, supports the observations made by antebellum
travelers to the rice coast, that slaves had to lock up their few, mea
ger possessions because of the rampant stealing which existed among them
(Olmsted 1968 (1856):432; Lane 1973:197). It is also possible that
padlocks were used to secure slave houses at night to ensure that the
slaves obtained sufficient sleep and did not spend time in nocturnal re
creation (Genovese 1974:535).
Summary
Although a detailed investigation of plantation spatial organiza
tions has not been undertaken, the settlement patterns of most large
plantations were characterized by broad similarities. Apparently, this
was also true for the plans of slave villages. Generally, both the inter
nal arrangements of the slave villages as well as the location of these
within the plantation complex were clearly functional. These functional
attributes, however, appear to have exceeded specific cash crop require
ments; therefore, it is difficult to identify plantation settlement
pattern data which are exclusively associated with rice culture. At
Butler Island tentative evidence of an accommodation to rice culture has
been indicated from the slave community plan, but additional archaeo
logical testing is needed to support these suggestions.
As for slave housing, the evidence from Butler Island conforms to
that of most tidewater plantations at both cotton and rice plantations.


160
indications that the Butler Island slaves prepared their meals in the
privacy of their own homes.
Several polished stone fragments were found within the slave dwell
ings (see Figure 23SG). Some of these, flat slabs of lateritic sand
stone containing mica, appear to have been used for grinding,* possibly
to grind hominy into grits (Hilliard 1972:49) or for some wild vegetable
items. Quartzite pounders were also found and were evidently used to
crack items such as nuts. Similar quartzite pounders were found at
Cannon's Point plantation on St. Simons, and it was suggested that these
were perhaps taken from prehistoric middens (MacFarlane 1975:107).
Eating utencils
Fanny Kemble observed that the Butler Island slaves ate with
broken iron spoons, pieces of wood, and with their fingers (Kemble 1S61
(1863):100). Yet evidence of more substantial eating implements was
provided by the archaeological data, including four spoons., two cutlery
knife blades, and several fragments of bone handles. Presumably, the
bone handles were attached to iron cutlery. The St. Simons slave sites
have yielded very similar evidence of eating utencils. Four spoons
were identified, two of iron, one of whitemetal, and one of pewter
(see Figure 31). Possibly, the pewter spoon was a luxury item and a
very special personal possession. The knife blades are made of iron and
were possibly attached to bone or wooden handles.
The bone handles are perhaps the most curious of all the eating
utencils (Figure 32). Although these resemble manufactured bone cutlery
*Charles H. Fairbanks, personal communication, 1980.


189
100 110 120 130 140 150 160


181
been fortunate enough this week to give them a fine portion
of fresh fish. 4 hands took over 300 bass and drum at
least 4000 lbs., which when divided among them are smoked
and cured and will last for weeks. (Roswell King, Jr.,
14 July 1833}
The reference to both bass and drum in large quantities suggests that
fish were taken from brackish waters, if not marine waters.
It is unfortunate that no surely identifiable anadromous fish were
recognized. Anadromous fish, which enter the rivers during the spring
and summer months and move upstream to spawn, include herring and shad
(Clupeidae), stripped bass (Morona saxtil is), and the strugeons
(Acipenseridae) (Dahlberg 1975).
Assuming that the identified fish species can be generalized as
representative samples of fish taken and consumed by the Butler Island
slaves, a preference for fish found in nearby stagnant waters is sug
gested. It may be that convenience dictated this pattern rather than
an actual preference, since these fish were taken probably from the rice
canals, where they are frequently found today. At Cannon's Point, slaves
evidently exploited nearby species rather than those more distant (Otto
1975:345). The drum and sheepshead may simply represent fish occasion
ally taken from brackish waters or they could have made their way into
a-freshwater environment. Given the limitations of the collection, these
interpretations must remain tentative, but the sample does provide indi
cations that fish typically found in the delta were regularly exploited.
Turtles represent the greatest nondomestic food resource exploited.
The three turtle species identified are all semi aquatic species and are
all found in deltaic habitats.
The mud turtle (Kinosternon sp.), only 7.5-12.0 centimeters in
size, frequently inhabits quiet, slow-moving, shallow waters along


69
1817 and 1819 may have been factors. During that time the entire south
ern economy suffered great financial losses (Eaton 1961:274-873). Per
sonal family relations are another possibility. Unable to convince
Thomas to take over the administration of his estate, the Major may have
decided to sell before considering the alternative of leaving his proper
ties to his grandsons. Whatever his reasons for selling, he never found
a buyer: "I have done all I can to sell your estate to promote your
interests and to close your accounts in Georgia. I fear I shall not be
able to do much for you" (Roswell King, Sr., 15 January 1819). Thus the
estate was retained and portions of it remained in the Butler family for
another 90 years.
Records of Slave Life
One of the most astonishing aspects of the Butler estate was the
size of the labor force. By 1859, a total of 919 slaves were included
in the estate (Butler Estate Papers, 1859). Even in Major Butler's
day, the total number of Butler slaves ranged from 500 to 700. Slave
holding of this magnitude was clearly in a class by itself. Slaveholders
who owned from 500 to 1,000 slaves represented less than 2 percent of
all slaveholding in the entire South. In Georgia, this "unusually large
slaveholding was confined to less than 1 percent of all planters in the
state" (Gray 1941:530-538). Thus, the size of the Butler estate made
it atypical among southern plantations.
How the Major managed to acquire such a vast number of slaves is
an interesting subject in itself. Most of the Major's initial transac
tions had taken place between the years 1786 and 1804 (George Hooper
1785-1802; Butler Estate Papers, 1801-1804). Although some of the


35
A number of pests preyed upon the growing rice (see Austin 1893:
31-37 for a detailed discussion). Two of the most frequently mentioned
were volunteer rice and rice birds. Volunteer rice or "red rice" was the
product of seed from the previous year which remained in the soil and
sprouted with the new crop (Austin 1893:23). A few of these red grains
among the white rice were believed to seriously lower the quality of the
crop. Consequently5 planters carefully selected the seed rice and had
slaves to remove these self-planted sprouts whenever they were detected
(Gray 1941:729). Rice birds presented another serious problem for they
often consumed large portions of the crop. These birds were unavoidable
since they appeared in May for approximately two weeks and then re
appeared in September for the same length of time. Some planters planted
in between May and September, but yields were small (Heyward 1937-31-33).
Most planters, however, employed "bird minders" to frighten the birds
away with gunfire, noise, or decoys (Heyward 1937:32; Gray 1941:729).
Harvesting, Milling, and Marketing
Rice was harvested by hand sickles in late August or early
September. The cut rice was allowed to dry for a day or so and then it
was tied together in sheaves. These sheaves were carried to the mill
or barnyard by flats which floated on the main canals or overland by
o-carts (Spalding 1835). The rice was stacked in the barn ur.ti 1
thresh!ng season (Gray 1S41:729).
Preparing the rice for the market involved several processes:
"threshing," the removal of the rice from the plant; cleaning or
"winnowing," the separation of the grain from the chaff; "pounding,"


- - -

-;
it.
THE ARCHAEOLOGY OF AFRO-AMERICAN SLAVERY IN COASTAL GEORGIA:
A REGIONAL PERCEPTION OF SLAVE HOUSEHOLD AND COMMUNITY PATTERNS
BY
THERESA A. SINGLETON
ft DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1980


3
Regional Analysis in Archaeology
Systematic approaches toward the organization of archaeological
data within temporal and spatial frameworks have been undertaken in
archaeology for most of this century. Although early regional studies
lacked the sophisticated methodology and theoretical orientations of
recent years, large amounts of archaeological data were collected for
many states and regions in the United States (Hole and Heizer 1977:
13-14). These data have provided the basis for regional chronologies,
cultures, and the frameworks for the "more-analytic causally-oriented
approaches" of today (Adams 1968:1188).
The concepts and methods for the regional analysis of the "new
archaeology" have been derived from geography. Stated simply, the
region is a "part of the earth's surface which is distinguished in
some defined way from the surrounding area." This distinctiveness may
be based upon a single criterion or upon a number of criteria (Grigg
1967:464). The objective of regional analysis is to understand
spatial associations within the confines of a specified geographic area
and eventually within a hierarchy of larger areas (Berry 1864, 1968;
Isard 1956; Haggett 1965). Its methodology utilizes probability sam
pling and multi variate mathematical techniques for the identification,
interpretation, and explanation of spatial organization.
Recent applications of regional analysis to archaeology have
resulted in a trend toward replacing the single site as the unit of
archaeological analysis with a larger entity, the region. Presumably,
the importance of a single site cannot be sufficiently understood
taken out of its regional context (Flog 1978). Lewis Binford was the


of the tide (Doer 1970:23). These canals and the impounding pond were
detected in aerial photographs. Subsurface testing at the site un
covered the structural underpinnings of the mill, the bed of the water-
wheel, and a millstone.
Previous archaeological assessments made in 1956 were used to
locate the slave cemetery (Caldwell nd; Sanders nd). Again, through
waterfowl management dredging activities, a number of coffins were
unearthed in the area indicated as the slave cemetery (see Figure 7).
At that time, a dragline cut through the south edge of the cemetery
smashing several wooden coffins and exposing others in the profile north
of the dragline. It was noted that the coffins were made of cypress and
that the skeletal material was poorly preserved. Most of the coffins
were evidently those of infants, a finding which supports the historical
evidence of a high infant mortality rate at Butler Island. On the basis
of information obtained from local informants, Sanders (nd) identified
the cemetery as a slave burial ground. These informants claimed to have
had ancestors buried there. The location of a cemetery could not be
found in the plantation records, but both Fanny Kemble (1961 (1863):141)
and Francis Leigh (1883:72) mention a slave burial ground. Unfor-
tunately, neither identified its location. Given the number of infant
burials, and the fact that a slave cemetery or two was located at
Butler Island, it was decided that this site once served as a slave
graveyard.
The sites
of "unknown function" could not be
identified on the
basis of visible archaeological remains occurring on the surface,
gestions of possible function, however, can be offered on the bas
Sug-
s of


- - -

-;
it.
THE ARCHAEOLOGY OF AFRO-AMERICAN SLAVERY IN COASTAL GEORGIA:
A REGIONAL PERCEPTION OF SLAVE HOUSEHOLD AND COMMUNITY PATTERNS
BY
THERESA A. SINGLETON
ft DISSERTATION PRESENTED TO THE GRADUATE COUNCIL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1980

SM 83- 82- 2685 Si
(S) Copyright 1980
by
Theresa A. Singleton
'

V .. - "" ?V.- ....
' .
Dedicated to the memory of my father William
Singleton. Sr

ACKNOW!. EDGMENTS
Many persons made this effort possible. Several faculty members
at the University of Florida have been central in fiel ping me develop my
research and writing skills, particularly Drs. Fairbanks, Jerald T.
Milanich, Maxine Margolis, Prudence Rice, Samuel Proctor, and Ronald
Foreman. Dr. Fairbanks has been a mentor and friend throughout my gradu
ate career and I have enjoyed studying with him. I also would like to
thank Dr. Von Mering for reading the final draft for my defense.
Dr. Kathleen A. Deagan, my first instructor in field methods of histori
cal archaeology, has always been supportive and interested in my career
goals.
Very special thanks go to Dr. Lewis Larson, State Archaeologist
for Georgia. I am very pleased that he asked me to undertake the Butler
Island Project. Also, his staff deserve special mention, particularly
Linda Tiironings, for undertaking all of the bookkeeping responsibilities
for the project.
I am very grateful to the field crews of summer 1978, spring 1979,
and summer 1979 for working along with me in the Butler Island marsh.
I think we are the only ones who can truly appreciate what slave life
must have been like at Butler Island. I am particularly grateful to
Ernest Despainge for the detailed drawings of the chimney; Bill Fisher
for helping out with the laboratory analysis and for asking his father,
Dr. Richard Fisher, to undertake the soil analyses; and Lorraine McCoske
for his assistance with the photographs.

Appreciation is extended to Gene Love, David Edwards, and the re
mainder of the staff of the Al tamaha Waterfowl Management Area for their
advice, equipment, and time.
I would like to thank the community of Darien, Georgia, for the
hospitality they extended to me and the crew during our stay there. Spe
cial thanks go to Ms. Bessie Lewis, Jim Cook, Nurse Campbell, Mrs. Bates,
Nan Earl Wylly, and Rudolph Capers for their valuable information relating
to the findings presented in this study.
The archival research was an extremely important aspect of this study.
I would like to thank the staffs of the Pennsylvania Historical Society,
the Georgia Historical Society, and the P. K. Yonge Library at the Univer
sity of Florida for their assistance. Also acknowledged are the staff of
the Inter-Library Loan Department at the University of Florida Library for
securing many of the materials I needed.
A warm appreciation is extended to my colleagues for their comments
and suggestions concerning my research objectives: Kenneth Lewis, Patrick
Garrow, Thomas Wheaton, Lei and Fergueson, Jennifer Hamilton, Chad Bradly,
Sue Mullins-Moore, John Morgan, Jr., Morgan R. Crook, Elizabeth Wing,
Rochelle Marrinan, Larry Banks, Nicholas Honerkemp, Robin Smith, and
Leslie S. Lieberman. Also, I would like to thank John A. Scott for his
suggestions; Richard Fisher for undertaking the soil analyses; Kathy
Bordua, my lab assistant and confident during the analysis and writing of
the dissertation; Ashley Wood, the draftsman and illustrator; and Sue
Kirkpatrick for typing the final manuscript.
Finally, I would like to thank my family and friends for their
encouragement through the years.
v

TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ..... ...... iv
LIST OF TABLES ix
LIST OF FIGURES
ABSTRACT
XI1
CHAPTER
1. INTRODUCTION .......... 1
Regional Analysis in Archaeology ... ... 3
Regional Approaches to the Study cf Slavery 7
The Archaeology of Afro-Americans ..... 10
Problems and Hypotheses ..... ..... ¡4
2. PLANTATION AGRICULTURE IN THE GEORGIA TIDEWATER ...... 20
Environmental Summary and Introduction .... 20
The Emergence of the Plantation System in Georgia .... 23
The Rice Industry 26
Rice and Slavery 26
Methods of Irrigation and Planting Rice ........ 31
Harvesting, Milling, and Marketing .... 35
Long-Staple Cotton Culture 39
Methods of Planting arid Cultivation 40
Harvesting, Processing, and Marketing ........ 41
Summary 43
3.HISTORICAL AND ARCHAEOLOGICAL INVESTIGATIONS AT BUTLER
ISLAND 46
Ownership and Occupation 43
Plantation Management at Butler Island, 1802-1860 .... 58
Records of Slave Life 69
Archaeological Resources .... ..... 85
Site Assessments 85
Slave Village Sites 93
4.SETTLEMENT PATTERNS OF COASTAL SLAVE COMMUNITIES 109
The Determinants of Plantation Settlement Patterns ... 109
Slave Community Organizations at Butler Island ..... ¡14
vi

Page
CHAPTER
The Community Plan at Slave Settlement #4
Slave Dwellings
Chimney construction
Building hardware
Summary .............
5. THE MATERIAL CULTURE OF SLAVERY
Kitchen Artifacts . .
Ceramics
Ceramic styles and vessel forms . .
Dating the ceramics from Butler Island
Nonceramic Artifacts
Glass bottles and tablewares . . .
Cooking and processing equipment . .
Eating utencils .
Kitchen Artifacts: Summary .......
Food Procurement . . .
Food Resources . .
Domestic Food Resources
Nondomestic Food Resources .
Fishes and turtles ..........
Birds and Mammals
Summary of Food Resources
Clothing
Buttons
Footwear
Accessories and Ornaments .........
Beads .
Miscellaneous Items ... .
Summary .
Household Items ..... .
Tobacco Equipment . .
Farming Tools and Specialized Crafts . .
Summary . .
6. PATTERN RECOGNITION IN SLAVE MATERIAL CULTURE .......
7. SUMMARY AND CONCLUSIONS . .
APPENDIX
1. SUMMARY OF SOIL ANALYSIS FROM SETTLEMENT if4, BUTLER ISLAND
2, SLAVE SUBSISTENCE ARTIFACTS FROM SETTLEMENT i/4 ......
ABBREVIATIONS USED IN REFERENCES . ........
i l D
126
130
137
138
140
141
142
143
155
157
157
159
160
i 6 b
166
169
173
1 78
179
183
184
186
186
187
190
190
193
193
1 S3
198
199
109
211
219
225
VI 1

REFERENCES
Page
229
BIOGRAPHICAL SKETCH
248
vm
- m n

LIST OF TABLES
¡ble
Page
1. Ownership of Butler Island, 1790? to the Present 50
2. Rice Production at Butler Island, 1802-1833 64
3. Annual Births and Deaths of Butler Slaves 71
4. An Approximate List of Slaves by Age and Task Rate at Butler
Island, 1803-1845: Field Hands 72
5. An Approximate List of Slaves, by Age and Occupation, at
Butler Island, 1803-1845: Plantation Specialists 76
6. 1830 Census of Slaves at Butler Island 77
7. Mortality Rates of Infants and Children of Sieves on the
Butler Estate, 1818-1834 81
8. Vessel Forms of Ceramic Styles 144
9. Frequencies of Decorated Tablewares 155
10. Application of the Mean Ceramic Date Formula to Tablewares 156
11. Styles of Glass Bottles Represented at Butler Island,
Settlemen
t h 4
158
12.
Identified Fauna from
Slav
e Site #4, Butler Island
171
13.
Relative Weights of Domest
from Butler Island
ic and Nondomestic Animal Food
1 72
14.
The ReTat
ive Weight of
Domestic Species at Butler Island
174
15.
Relative
Class
Weight of Ide
nti f
led Nondomestic Food by Animal
179
15,
Artifact
Profiles for
F o u r
Slavery Sites in Georgia/Florida
214
17.
Adjusted
Artifact Prof
lie
for Kingsley P1 antation
215
18.
The Slave
1 Artifact Pat
te rn
in Coastal Georgia/FIorida
216

LIST OF FIGURES
Figure Page
1. Investigated Slavery Sites in Coastal Georgia. 15
2. The Physiographic Regions of Georgia. 21
3. Extent of Rice Cultivation in Georgia in I860 27
4. Butler Island and Environs. 47
5. The Butler Georgian Estate. 52
6. The Subdivision of Butler and General's Islands into
Tenant Plots (circa 1883). 57
7. Historic Sites at Butler Island. £6
8. Present-Day View of Butler Island. 88
9. Extent of Excavations at Settlement #1. 95
10. Stratigraphic Profile of Settlement #1. 97
11. Eighteenth-Century Slipware Fragments from Settlement #1. 100
12. Spanish reale, "Mexico Mint," from Settlement '4. 102
13. Contour Map of Settlement #4. 104
14. Excavation of Settlement #4. 105
15. Stratigraphic Profile, Settlement #4. 107
16. Overseer's House. 115
17. Tidal Mill and Boat Landing. 115
13. Slave Cabin. 116
19. Steam Mi 11 Rui ns. 116
20. Linear Arrangement of Slave Dwellings. 116
21. Extent of Excavation at Structure Four. 122

gure
Page
22.
Stratigraphic Profile of Drainage Ditch.
124
23.
The Extent of Excavation at Structure One, Settlement #4.
123
24.
Hearth of Chimney.
132
25.
Chimney Brickwork.
133
26.
Arch-Shaped Hearth Support.
135
27.
Red Coarse Earthenwares.
147
23.
Transfer Printed Pearlware Fragments.
150
29.
Annular Ceramics.
153
30.
Annular Ceramics.
153
31.
Eating Utenc.ils.
154
32.
Carved Bone Fragments.
164
33.
Food Procurement Artifacts.
168
34.
Clothing and Personal Adornments.
189
LO
CO
Shoe Soles.
192
36.
Parasol Attachment.
195
37.
Eye Lens.
197
33.
Porcelain Pipe Bowl.
201
39.
Woodwork!ng Tools.
203
40.
Unidentified Tools.
206
41.
Slave-Made Ceramics.
208
xi
s-'r:::-: -:.Vx

Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
THE ARCHAEOLOGY OF AFRO-AMERICAN SLAVERY IN COASTAL GEORGIA:
A REGIONAL PERCEPTION OF SLAVE HOUSEHOLD AND COMMUNITY PATTERNS
Gy
Theresa A. Singleton
December 1980
Chairman: Charles H. Fairbanks
Major Department: Anthropology
A variety of methodological and theoretical orientations has been
utilized in the study of Afro-American slavery. Regional approaches are
not new to slavery, but few studies have examined the impact of local
ecological factors upon slave 1ifeways within a narrowly defined area.
In this study, coastal Georgia (the barrier islands and the adjacent
deltas) provides the region for the exploration of an ecological approach
to slave life and material conditions. Specifically, slavery at a rice
coast plantation is compared with slavery at long-staple cotton planta
tions. Archaeological data provide the primary basis for interpretation,
and historical resources are used to supplement the archaeology.
Primary settlement patterns and subsistence data recovered from the
Butler Island rice plantation in 1978 and 1979 are used to investigate
slave life and material conditions on the rice coast. The Butler Island
data are compared with the archaeological data recovered from several
previously mves
mi gated slave sites of long-staple cotton plantations.
The comparison reveals similarities as well as differences in the
xn

slave 1 i fe ways of the two cash crop regions. Archaeological!./, the
differences are primarily discernible in slave community organizations,
the natural resources exploited, and slave crafts. Similarities are
evident in most household artifact patterns: plantation food rations,
food preparation equipment, personal possessions, and leisure-time activi
ties. It is suggested that the differences in slave life reflect the
dissimilarity between the habitats where rice and long-staple cotton were
produced. On the other hand, the similarities may reflect general labor
management practices adapted for the production of coastal staples.

CHAPTER 1
INTRODUCTION
The anthropological study of Afro-American slavery has increas
ingly used archaeological data as a primary source for interpreta
tion. Because traditional historic sources are subject to problems of
falsification and bias, archaeological data are frequently used to
supplement historic records. It has been suggested that archaeology be
the key to writing the cultural history of the "inarticulate" (Ascher
1974:10-12). Slavery sites have been investigated specifically for this
reason. As a consequence of the widespread illiteracy among slaves,
few documents written by slaves exist. Most of our understanding of
slavery has come from accounts written by whites. Inferences derived
from the archaeological record of slavery can provide insights into
slave lifeways and slave worldview. Although planter records," travelers'
accounts, and oral interviews are used extensively here to supplement the
archaeological record, this study assumes that archaeological data are
one of the best indices of slave behavior.
The purpose of this study is to offer a model capable of dis
cerning slave behavioral patterns in coastal Georgia. Slavery on the
Georgia coast was characterized by distinctive demographic, environmen
tal, economic, as well as historical factors which were absent in the
interior. Undoubtedly, these conditions affected the quality of slave
life, slave activities, and ultimately the development of slave cultural
patterns.
1

2
At present, slave sites from coastal Georgia are perhaps better
known than those from any other regions of the Old South. In recent
years, a number of slave sites have been identified, located, and exca
vated in this area. Until now, no attempt has been made to synthesize
these archaeological findings within a framework larger than the single
piantation.
Archaeological investigations undertaken at the Butler Island
plantation in McIntosh County, Georgia, provide the first archeological
evidence of slave life at a Georgian rice plantation. A1! reported
excavated slave sites in Georgia are former long-staple cotton planta
tions on the barrier islands. Thus, the significance of the Butler Island
research is twofold: first, it supplies archaeological evidence for the
specific adaptations of slavery to rice culture; second, comparisons
of the Butler Island data with data from long-staple cotton slave sites
indicate several identical patterns. This evidence suggests that the
cultural system of slavery which developed in coastal Georgia exceeded
adaptations to specific cash crop requirements.
Ultimately, the goal of this dissertation is to identify and offer
possible explanations for slave community and household patterns. This
objective is accomplished through the analysis of primary settlement
and subsistence data recovered from Butler Island slave sites. Com
parative d
dence for
this study
ecologicai
ata from other coastal slave sites provide additional evi-
regional interpretations. The theoretical orientation of
is derived from regional archaeology, which utilized an
approach in the interpretation of cultural phenomena.

3
Regional Analysis in Archaeology
Systematic approaches toward the organization of archaeological
data within temporal and spatial frameworks have been undertaken in
archaeology for most of this century. Although early regional studies
lacked the sophisticated methodology and theoretical orientations of
recent years, large amounts of archaeological data were collected for
many states and regions in the United States (Hole and Heizer 1977:
13-14). These data have provided the basis for regional chronologies,
cultures, and the frameworks for the "more-analytic causally-oriented
approaches" of today (Adams 1968:1188).
The concepts and methods for the regional analysis of the "new
archaeology" have been derived from geography. Stated simply, the
region is a "part of the earth's surface which is distinguished in
some defined way from the surrounding area." This distinctiveness may
be based upon a single criterion or upon a number of criteria (Grigg
1967:464). The objective of regional analysis is to understand
spatial associations within the confines of a specified geographic area
and eventually within a hierarchy of larger areas (Berry 1864, 1968;
Isard 1956; Haggett 1965). Its methodology utilizes probability sam
pling and multi variate mathematical techniques for the identification,
interpretation, and explanation of spatial organization.
Recent applications of regional analysis to archaeology have
resulted in a trend toward replacing the single site as the unit of
archaeological analysis with a larger entity, the region. Presumably,
the importance of a single site cannot be sufficiently understood
taken out of its regional context (Flog 1978). Lewis Binford was the

4
first to define this usage of regional archaeology:
the detailed and systematic study of regions that can be
expected to have supported cultural systems. The extent of
such regions will vary because it is recognized that cul
tural systems differ greatly in the limits of their adap
tive range and milieu. As cultural systems become more
complex, they generally span greater ecological ranges and
enter more complex, widespread, extra-societal interaction.
The isolation and definition of content, the structure, and
the range of a cultural system together with its ecological
relationships may be viewed as a research objective.
(1964:426)
Central to the above definition is the view that culture is an
adaptive system. As an anthropological concept, adaptation refers to
the ability of a population to adjust to the environment by develop
ing effective social and technological structures. Adaptation is an
ecological process (Steward 1955:30). The interaction between human
behavior and the environment in which culture is the mediating
variable is fundamental to ecological approaches in anthropology
(Harris 1963:659). Cultural ecology as a research strategy involves
three procedures: first, it analyzes the interrelationships between
technology and the environment; second, it analyzes behavioral patterns
involving the exploitation of a particular area (the region) by a
particular mode of technology; third, it ascertains the extent to
which the behavior patterns involved in exploiting the environment
affect other aspects of culture (Steward 1955:40-1). The utility of
regional analysis in cultural ecology is very clear. As the locus
occupied by a cultural system, the region provides the unit of analysis
for ascertaining human ecological relationships.
Because the region is the locus of human activities, regional
archaeology requires the investigation of diverse activity areas within

5
the geographic area occupied by a cultural system. It assumes that the
region, not the single site, embraces the total range of behavioral
variability for a population. The single site, on the other hand,
reflects only those activities that had taken place at that
specific locality (Streuver 1971:11). Thus the goal of regional
archaeology is to determine patterned interrelationships among sites.
Regional archaeology attempts to understand the cultural and
environmental processes which have affected behavioral patterns. Its
application to historic site archaeology is slowly emerging. Specifi
cally at historic sites, artifact pattern recognition (South 1977,
1978) has been suggested as a technique of discerning behavioral
variability within a region or time period. Ultimately, pattern recogni
tion seeks to identify by means of archaeological data the functional
and behavioral processes which have taken place at sites (South 1378:
223; e.g. Lewis 1977).
Besides a unit of analysis, regional archaeology also supplies a
method for the study of site location. Locational analysis examines
spatial relationships about ecological or cultural phenomena
(Haggett1965). A developing focus of locational analysis is its
application to historic sites (Swendlund 1975; Langhorne 1976; House
1977). Such studies attempt to determine the importance of cultural
or natural resources within the functioning of a historic cultural
system (House 1977:243-4).
A third concept of regional archaeology establishes a frame
work for the collection of cultural data for a defined area. The
regional plan or overview is the first phase in multi-phase

6
archaeological research. Frequently, regional plans are used as
planning devices in cultural resource management (King et al. 1977:
145-173; McGimsey and Davis 1977:47; Schiffer and Gurnmerman 1977;
121-131). As planning tools, regional plans involve the review of all
known records of the project area, including historical records, sur
vey and excavation reports, and other existing field data as well as
informant consultation. The overview should summarize the present
knowledge, evaluate the available evidence, estimate the resource base,
and attempt to forecast long-range regional development and ongoing
destructive processes (Schiffer and Gurnmerman 1977: 12-13). Although
regional plans are only as good as our present-day knowledge permits
(Glassow 1977), their utility as rudimentary planning tools is essen
tial in historic preservation.
From the above discussion four concepts of regional archaeology
are identified. The first is a unit of analysis which represents
the broad cultural milieu of human activities for a cultural system.
The second is a method of discerning spatial organizations within a
cultural system. The third is a framework for the collection and
synthesis of cultural and ecological data within a defined geographic
area. Finally, the fourth is the explanation for the patterns thus
defined.
In this study, the first, third, and fourth concepts provide the
orientation for analysis and interpretation. Coastal Georgia, speci-
cally the barrier islands and the adjacent river deltas, forms the
geographic parameter. The temporal parameter is the nineteenth cen
tury (1800 to 1861). The unit of archaeological analysis for the
investigation of the slave system is the slave community site.

7
The slave village or community site was the locus of most slave
activities. Admittedly, slaves spent considerable time in the fields
and at other locations of the plantation complex. But it is doubtful
that these sites will provide archaeological data relating to slave
behavior. Another site perhaps central to the slave belief system is
the slave graveyard. Because of the sensitivity involved in excavat
ing burials, this archaeological resource is not treated here. Slave
graveyards, however, were at least located at Butler Island. Because
of the restricted movement enforced upon slaves, the slave village site
should embrace a major portion of slave behavioral variability. It
also preserves evidence for a broader range of cultural activities than
does a simple activity area such as a rice field or threshing floor.
A methodological departure of this study from regional research
strategies advocated by Binford and others is the degree of use of
probability sampling at all levels of analysis (Binford 1964). Proba
bility sampling is used in this study but minimally. Additionally, the
multivariate techniques of geographers are not used at all. Admittedly,
this methodological departure may present some serious disadvantages
in offering regional interpretations. The purpose of this study,
however, is to synthesize the presently available data recovered from
slave sites in coastal Georgia. The interpretations offered here
are tentative and, hopefully, will be tested in future regional studies
utilizing regional methods.
Reg!oral Approaches to_the Study_of Slavery
Defined cultural, spatial, and temporal variables form the
parameters for most discussions of Afro-American slavery.

8
Cross-cultural comparisons of slavery are often concerned with the
issue of slave treatment within the confines of political entities
(e.g. Tannebaum 1949; Davis 1966; Degler 1971). Similarly, in the
United States discussions of slavery are confined to geographic areas
such as states (e.g. Flanders 1933) or time periods (e.g. Stampp 1956).
Both comparative studies and those within the United States are fre
quently concerned with slave management or the legal aspects of
slavery. Less often has the issue of slave behavior been addressed.
When it has, the entire slaveholding South has formed the unit of
analysis (e.g. Blassngame 1972; Genovese 1974). It is questionable
whether these discussions are applicable to slavery everywhere in the
Old South.
Although the slave cultural system which developed in the South
was characterized by general behavioral patterns (Gutman 1977), local
variations of this tradition were inevitable. Evidence of this has
been indicated in many localized Afro-American traditions, some of
which have survived until the present day. The coastal areas of
Georgia and South Carolina have been focal areas of African retentions
(Kerskovits 1958:120), and in the development of distinctive Afro-
American traditions, particularly in the decorative arts, language,
cuisine, and music.
Certainly the quality of slave life must have varied regionally.
Although this assumption remains untested, indications from archaeologi
cal resources are that the material standard of slaves in coastal
Georgia was better than that of slaves documented elsewhere in the
Old South (MacFarlane 1975; Otto 1975).
Further, a recent study of

9
slave nutritional adequacy indicates that slaves in the coastal areas
of South Carolina, Georgia, and northeast Florida were adequately
nourished (Gibbs et al. 1980). These studies demonstrate that regional
approaches in the study of slavery are needed.
It is possible that certain economic, environmental, and demo
graphic conditions in coastal Georgia favored a distinctive develop
ment of slave behavior. In Georgia, the largest percentages of blacks
to whites were found on the coast. In addition, the most valuable farms,
the largest slaveholding, and the most livestock were located there.
Another factor was the value of lands and buildings per farm. This
was highest on the coast than in any part of Georgia (Flanders 1933:81).
Ecologically, the region provided abundant plant and animal resources
which were exploited for both domestic and industrial purposes. Employ
ment of the task slave labor system on most coastal plantations allowed
slaves more time for recreation and to improve their material lots
than the gang labor system employed elsewhere in the Old South (see
Otto 1975). For the most part, these conditions were not present in the
interior. It is likely that variations in these conditions produced
variations in slave behavior.
Historic factors were also important. Unlike their counterparts
of the short-staple cotton belt, most planters that came to coastal
Georgia were experienced slaveholders from the West Indies or South
Carolina (Vanstory 1970:74-79; Davis 1976:132). They were presumably
acquainted with the problems of slave management. Conceivably, they
attempted to establish the best possible conditions for a "stable
regime within which their slave could live" (Genovese 1974:6). Perhaps.

10
this was why the plantations of the Georgia coast have been described as
"patriarchal" (Wylly 1910:12; Lewis and Huie 1974). Another contribut
ing factor was the infrequency in which planters of the coast migrated
to the new lands of the West. With a few exceptions, slaveholders on
the coast of Georgia established their roots in the late eighteenth
century and remained there until the Civil War. It was "only in the tide
water areas of Georgia [that] wealth [had] matured for several genera
tions" (Boney in Coleman 1977:174). This may have been the result of
special local ecological factors. ^
This study assumes that the previously described variables sub
stantially influenced a development of slave behavior patterns pecu
liar to the Georgia coast. Some of these.patterns are believed to be
archaeologically discernible. These archaeological indices are dis
cussed in detail in subsequent chapters.
The Archaeoiogy of Afro-Americans
Archaeologists have become increasingly involved in the investiga
tion of sites formerly occupied by black Americans. Because few his
toric sources have been written by blacks, this research interest
developed out of a need to expand upon traditional interpretation of
past black American lifeways. Contemporary eighteenth and nineteenth
century accounts relating to black Americans were written by whites.
As a result, these are often very prejudiced views of Afro-American
cultural life. The major objective of Afro-American archaeology has
been to investigate aspects of black American life not available in
documents.

11
With a few exceptions, archaeological studies of black Americans
are of two types: slavery sites and northern free black sites.
Because of the preliminary quality of this research area, most of these
studies have been more descriptive than interpretative. Within this
year, however, an attempt has been made to synthesize the findings from
several Afro-American sites (see Schuyler 1980). From these studies
and other published works, two themes dominate this research interest:
first, the search for African retentions in black American material
culture (Schuyler 1980:2), and second, the recognition of Afro-American
subsistence patterns.
It was the search for material correlates of African survivals as
well as subsistence information which initiated preliminary testing at
the Kingsley slave site by Fairbanks in 1968 (Fairbanks 1974). Un
fortunately, "no surely African elements" were identified (Fairbanks
1974:90). Since that time studies of slavery sites in the Old South
have been directed toward the definition of subsistence patterns
associated with the socioeconomic status of slavery (Ascher and
Fairbanks 1971; MacFarlane 1975; Otto 1975; Drucker and Anthony 1979).
Investigations of slave burials in Barbadoes have provided very
convincing evidence of Africanisms (Handler and Lange 1978, 1979).
At the Newton planation, the orientations of slave bodies and the
associated burial goods are indicative of West African burial patterns.
Burial practices peculiar to Afro-Americans in the United States have
been reported but only at sites occupied by freed blacks (Cate and
Wrightman 1955:207-215; Combes 1974; Crosby and Emerson 1979).

12
Perhaps, the most striking and yet controversial evidence (see
Schuyler 1980:2) of Afro-Americans attempting to recreate their African
past has been uncovered at the Parting Ways site (Deetz 1977:135-154).
This site near Plymouth, Massachusetts, was occupied by four Revolution
ary War veterans from the late eighteenth to the mid-nineteenth centuries.
Deetz suggests that the settlement pattern, house floor plans, a utili
tarian earthenware, and possibly the culinary practices are derived from
an African past.
The recent suggestion that black Americans influenced and manu
factured some Colono-Indian ceramics (Ferguson 1980) is another example
of this search for African survivals. These ceramics occur in high
frequencies at seventeenth-and eighteenth-century southern sites, speci
fically in the coastal areas of the Carolinas (Ferguson 1980). Previous
to this suggestion, these ceramics were thought to have been made
exclusively by native Americans (Noel-Hume 1962). It is well documented
that some of these ceramics were made by native Americans and some are
still made by them today. Indications that blacks may have had a hand
in the manufacture of these ceramics include the possible similarity of
this ceramic style with West African ceramic traditions, the infre
quency cf Colono-Indian ceramics at historic Indian sites, and its
high frequency at black occupied sites (Ferguson 1980). But the
fact that Colono-Indian ceramics have been uncovered, in most cases,
where native American populations were nearby* makes the suggestion
that blacks manufactured these ceramics uncertain.
^Charles H. Fairbanks, personal communication 19/9, Gainesville,
Florida.

The archaeological investigation of African survivals at Afro-
American sites has resulted in very little definitive evidence to date.
This may be the result of a number of factors. Although his discussion
is specifically directed to slavery sites, Otto details three processes
that limited the re-creation of African materials. These include
(1) selection and simplification, (2) availability and substitution,
and (3) differential acculturation (Otto 1975:375). It is very likely
that northern free blacks also substituted traditional items with
available ones. The archaeological record is another factor. African-
styled artifacts such as wooden objects or basketry are not likely to
be preserved (Otto 1975:382). Regardless of the reason for the scarcity
of African-styled artifacts at black-occupied sites, research designed
solely to uncover Africanisms ignores the fact that Afro-American
culture, like all cultures, is an adaptive system (Schuyler 1980:2).
A data base of Afro-American subsistence patterns is slowly emerg
ing from diverse archaeological resources. These data are crucial
to generating future hypotheses regarding Afro-American cultural
patterns. A major problem has developed in the identification of
Afro-American economics. To what extent are the subsistence patterns
at black sites indicative of Afro-American ethnicity or a culture of
poverty (Kelly and Kelly 1980)? Only when archaeological data
are available from poor white subsistence farmers can the archaeo
logical visibity of low-status Afro-Americans be ascertained
(Baker 1978). Presently, these nonplanter whites are archaeologically
unknown.
... ' . - " : '

The archaeological evidence of slavery, however* may not be as
difficult to recognize as are other Afro-American sites. Recently,
an excellent attempt was made to identify an undocumented site as a
slave site at Spiers Landing in Berkley, South Carolina (Drucker and
Anthony 1979). Although Handler and Lange (1973:228) conclude that
archaeological data do not identify the slave status or slavery,
the Spiers Landing example refutes this. To suggest that slavery sites
can be identified entirely from archaeological remains is very pre
mature at this stage of research. This objective, however, may be
realized in the future through the use of regional approaches to Afro-
American sites, particularly the utilization cf pattern recognition
techniques.
Finally, in their discussion of ethnic identification in his
torical archaeology, the Kellys recommend the use of regional research
designs (1980:135-6). They suggest that research designs geared toward
the recognition of ethnic cultural patterns within a defined region may
aid in the delineation of geographic variations of a broad cultural
tradition (1980:130). The delineation of a geographic and temporal
variation of a broad Afro-American cultural tradition is precisely the
objective of this study.
Problems and Hypotheses
From the available reports, it appears that slave sites are known
better archaeological!:/ in coastal Georgia than in any other region
of the Old South (see Figure 1, for,the location of these sites). De
spite this knowledge, these studies have suffered from three major

15
Figure 1. Investigated Slavery Sites in Coastal Georgia.

limitations: first, with the exception of one study (MacFarlane 1975)
excavations of slave settlements have sampled only a small portion of
the site, usually one or two structures (Ascher and Fairbanks 1971;
Fairbanks 1974; Otto 1975). Second, as a result of this sampling,
archaeological evidence of the internal social organization within the
slave community has not been uncovered. Third, slave sites investi
gated have been exclusively from long-staple cotton plantations. Com
parative data from the other dominant cash crop of the coast, rice, have
been lacking.
Investigations at the south end slave cabins at the Cannon's
Point plantation on St. Simons Island, Georgia (MacFarlane 1975) were
the first attempt to excavate an entire slave community site. This
investigation provided details of slave housing, household items, cloth
ing, personal items, and food resources. Little or no data, however,
were reported regarding the community organization, specialized crafts,
or social hierarchy.
Documentary resources of the Butler Island rice plantation indi
cate that each slave settlement had plantation drivers, craftsmen, cooks
as well as field laborers. It was hypothesized that testing an entire
slave community should uncover archaeological evidence of these special
ized roles and status differences. Initially, the objective of archaeo
logical investigations at Butler Island was twofold: first, to uncover
archaeological evidence of status difference and craft specializations
within the slave community; second, to compare the archaeological re
sources of slavery at a rice plantation with those of long-staple cotton
plantations, in an attempt to ascertain similarities and differences

17
in slave life from two cash crop sites within a narrowly defined region.
It became apparent in the field that the time needed to investi
gate the first objective was insufficient for the time and funds
available for fieldwork. As a result, attention was then directed
toward uncovering archaeological evidence associated with the slave
community plan. Presumably, settlement and certain subsistence data
would provide archaeological evidence for the special adaptations of
slavery to rice cultivation. At the same time, comparison of the
Butler Island data with long-staple cotton plantations would supply
additional evidence of slave life which exceeded adaptations to cash
crops. In other words, the comparison would provide data specifically
related to the economic function of the plantation, on one hand and
data which were not so related on the other hand. Further, it is
hypothesized by the author that similar artifact patterns occurring in
both crop loci are indicative of the material conditions of slave life
and slave behavior patterns in a larger coastal Georgia area.
The factors which are believed to
have influenced a
riistinctive
development of slavery in coastal Georgia have been briefly discussed.
It is important, however, to define more accurately their role in the
development of coastal slavery. The demographic, economic, and
historic factors (see pages 9 and 10} are all related to plantation
management practices. These management practices developed out of an
adaptation to the coastal environment of Georgia. Although the habitats
for the cultivation of rice and long-staple cotton are very dissimilar,
both habitats are part of a larger ecosystem (Johnson et al. 1974).
Thus, this is a study of adaptation on two levels: adaptation to

18
specific crop requirements and adaptation to a more generalized coastal
environment.
The purpose of this dissertation is to test the following
hypotheses and their implications:
Hypothesis A: Slave community and household patterns in coastal
Georgia reflect adaptations to the specialized habitats where tidewater
staples were produced. Archaeologically this will be identified in the
following:
1. Patterned placement of slave village sites relative to
habitats.
2. Community plan of spatial arrangements
a. Specialized buildings and material culture
b. Nonstructural features
3. Slave dwellings
a. Size and available space
b. Construction details and materials
4. Farming implements and specialized crafts
5. Nondomestic plant and animal resources exploited
Hypothesis B: Slave material conditions and behavior patterns
ref1ect management
cally these will be
practices adapted for coastal Georgia. Archaeologi-
identified in the following household artifact
patterns:
1. Food-reated sctivi ties
a. Food preparation equipment and techniques
b. Food serving equipment
c. Domestic plant and animal food resources

d. Evidence of food supplements in plantation rations
1. Equipment for producing food
2. Equipment for procuring wild food resources
3. Food processing equipment
4. Remains of food supplements
2. Personal possessions
a. Clothing
b. Accessories and ornaments
c. Household items
3, Leisure-time activities
a. Smoking equipment
b. Games and toys
c. Miscellaneous items
These hypotheses and their underlying assumptions are discussed
in greater detail in later chapters. Chapters 2 and 3 provide back
ground information on the coastal environment of Georgia, the produc
tion of rice and long-staple cotton, and the archaeological and his
torical investigations at Butler Island. Slave community organizations
at Butler Island are discussed in Chapter 4. Chapter 5 describes the
slave material culture recovered from Butler Island. Slave artifact
patterns derived from statistical pattern recognition techniques are
presented in Chapter 6. Chapter 7 discusses the results from testing
the hypotheses and the conclusions for the study.

CHAPTER 2
PLANTATION AGRICULTURE IN THE GEORGIA TIDEWATER
Environmental Summary and Introduct i_o n
Tidewater refers to the geographic area of the coastal plain,
which is effected by the influence of the tides. In Georgia, this
region includes the barrier islands and extends approximately 75 kilo
meters inland to the point where extensive areas of pine forests begin
to occur (Crook 1978:28-30) (see figure 2). Three major subdivisions
characterize the Georgia tidewater: the strand section, the salt marsh
.and lagoon .section, and the delta section. These are discussed in de
tail elsewhere (Larson 1970; Johnson et al. 1374), and are briefly sum
marized here.
The strand borders on open ocean and is composed of beach and
dunes. It is the most rapidly changing of the three sections. Vegeta
tion is restricted and most animal resources which occur there are found
more frequently in other habitats along the coast.
Of the three subdivisions, the salt marsh and lagoon section is
the most ecologically diverse. It is composed of four habitats: marsh,
tidal creeks, hammocks or highlands, and freshwater swamps. As a
result of its diversity, this section is rich in both plant and animal
resources. The food chain begins in these salt marshes. Several
species of fish, shellfish, turtles, and waterfowl thrive in the marsh
The hammock forest community includes several
20
and tidal creeks.

I
I
Figure 2. The Physiographic Regions of Georgia.
(Adapted from Coleman 1977.)

22
species of oak, pignut hickory, red cedar, southern magnolia, cabbage
palm, wax myrtle, saw palmetto, and many herbs and vines. Faunal
resources in these highlands arc also abundant. Whitetail deer,
oppossum, raccoon, and cottontail rabbit are among the dominant species.
The freshwater swamp is an important breeding ground for reptiles and
amphibians, and the dominant vegetation is cypress.
Fewer data are available for delta ecology than the other sections.
The delta embraces those areas where freshwater rivers enter saltwater.
Deltas are primarily freshwater but are affected by the tides. The
area is flat and is frequently inundated. Most vegetation is water-
tolerant. Near the ocean, the area is a grass-covered marsh; where
there is little or no intrusion of saltwater, the delta is composed of
cypress swamp (Larson 1970:35). Much of this swamp was cleared for rice
cultivation. Deer, otter, and raccoon occur in portions of the delta,
but the most distinctive faunal resources are the anadromous fish such
as glut herring, striped bass, sturgeon, and the American shad.
The narrow belt of
the tidewater, which includes the barrier
islands and the adjacent river deltas,is part of the oldest
region of Georgia. The sandy loams of the sea islands were
agricultural
the "most
productive lands" of the coast. During the colonial period, these
lands yielded crops of corn, indigo, potatoes, and vegetables. After
the Revolutionary War, these lands were devoted almost exclusively to
the production of long-staple cotton (Bonner 1964:1).
By the late eighteenth century, reclamation of the wet,
deltaic soils for rice cultivation had begun. Tidewater rice
alluvial
culture
was limited to the areas located far enough upstream to avoid contact

23
With saltwater but close enough to the ocean to be affected by the tide
(House 1954b:23). The restricted spatial limits of tidewater rice cul
ture made the alluvial soils of the delta the "most valuable" of this
old agricultural region (Bonner 1964:2).
In general, most plantations on the barrier islands were devoted
to long-staple cotton and those of the delta to rice. Some rice, however,
was grown on the barrier islands, presumably in the freshwater swamps,
and some cotton on the delta. Sugar, a third important cash crop (see
Sitterson 1937, 1953:31-35), was frequently produced at both, but in
smaller quantities. For example, at Butler Island, considerable quanti
ties of cotton and sugar were cultivated, although rice was the island's
major cash crop.
The purpose of this chapter is to review the historical develop- .
ment of plantation agriculture in the Georgia tidewater, specifically as
it relates to the production of rice and long-staple cotton, the two
dominant crops of the post-Colonial period. Also, planting and process
ing methods and marketing for each crop are described. Rice culture is
considered in greater detail for two reasons: It was a more involved
process than that of cotton culture and a''detailed description of this
culture is central to understanding the archaeological resources of
Butler Island (long-staple cotton is treated extensively in another
archaeological reportsee Otto 1975:51-61 ).
The Emergence of the Plantation System in Georgia
On June 9, 1732, the developers of Georgia, known as the trustees,
were granted a charter for the lands between the Savannah and the

24
Al tamaha Rivers. The trustees desired to establish a colony with small
settlements of independent farmers (Bonner 1964:2). To obtain this
objective, the trustees prohibited large land grants and Negro slavery.
The plantation system was outlawed for several reasons. First, the
colony was purportedly founded for debtors, although debtors made up
fewer than a dozen of the settlers during the entire trusteeship period
(Spalding in Coleman 1977:18). A large portion of tfie early settlers to
Georgia, however, were small businessmen, farmers, tradesmen, and
unemployed laborers (Coleman 1976:36-54), who could not be expected to
compete with a planter class. Second, Georgia was selected by the
trustees to produce commodities needed by England, such as silk, which
was not adapted to plantation agriculture. Third, the colony was to serve
as a military buffer to protect Carolina from the Spaniards in Florida.
Plantations would spread the people over the land instead of concentrat
ing them into settlements. Moreover, a large slave population could sub
ject the colony to slave insurrection, making it vulnerable to Spanish
attack (Potter 1932:116-117; Calloway 1948:30-31).
From the very beginning some colonists objected to the restrictions
placed upon Georgia. These "malcontents" were particularly displeased
with the prohibition of slavery. They were convinced that the use of
a cheap labor force was the only way to develop the colony. As dis
satisfaction grew, many colonists abandoned Georgia. On the brink of
economic distress, it became apparent to the trustees that if the
colony was to succeed it must be made attractive to potential colonists
(Davis 1976:126). Finally, in 1750, slavery was made legal. But,
discouraged by their failures, the trustees surrendered their charter
to the British Crown in 1752 (Spalding in Coleman 1977:44).

25
Under royal government, the colony experienced an accelerated
growth. A significant source of immigration to the coastal area was
the South Carolina planter. South Carolinians were aware of the rice
growing potential of the Georgia tidewater areas, end they migrated to
Georgia with their families and slaves (Wilms 1972); rice soon became
Georgia's most important crop. Later, the introduction of indigo gave
the plantation system an added boost (Spalding in Coleman 1977:52).
Planters were attracted to the coastal lands, but good lands were
not abundant. Moreover, the development of these lands was very costly.
An eighteenth-century observer in coastal Georgia estimated that the
initial expense of settling a rice estate was about $10,000, and for
cotton or indigo $9,000 (Thayer 1957:81). Consequently, only large,
wealthy planters could afford to acquire and cultivate these lands. The
small planter unable to compete settled further in the interior. Thus,
coastal Georgia became a region of a few, wealthy planters shortly
after the emergence of plantations to the area (Calloway 1948:47-49).
The Revolutionary War brought disastrous setbacks to the coastal
plantation economy. The bounty on indigo formerly received from the
British government ceased and the'exportation of rice declined. Pros
perity was restored, however, with the successful introduction of long-
staple cotton in 1786. Planters of rice plantations acquired land and
equipment for cotton production. Thus, rice was temporarily dis
placed as a major staple of the tidewater (Flanders 1933:55-57),
By the mid-nineteenth century, the reliance upon sea-island
cotton gradually shifted to rice. Apparently, this change roughly
coincided with periods of depressed long-staple cotton prices (Gray

1941:38). During this time of renewed rice interests, Carolina planters
intensified their activities in Georgia. They not only increased their
acreage in the old rice producing areas along the Savannah (see Clifton
1978) and Altamha Rivers, but also expanded further southward to include
the Satilla and St. Mary's Rivers (see Figure 3). In the 1850s, cotton
prices were on the upswing (Gray 1941:739) and rice cultivation was at
its peak (Hilliard 1975:65). Planters producing both crops were enjoy
ing sizeable profits on the eve of the Civil War
Devastation from the
Civil War prevented the successful revival of either culture in the
post-war years. The reign of rice and long-staple cotton as the dominant
cash crops of the coast had come to an end.
The Rice Industry
Early experiments with rice cultivation were attempted in
colonial Virginia, but rice was first established as a staple in the
South Carolina low country (Gray 1941:277). Although the details of its
introduction remain unclear, apparently early promoters of the colony
encouraged rice cultivation shortly after the founding of Carolina in
1670. By 1695, South Carolina's rice industry had begun (Salley 1919).
Rice and Slavery
si
Methods of rice cultivation during most of the plantation era were
entirely dependent upon hand labor. As a result, large-scale rice pro
duction required an enormous labor force. By the mid-nineteenth century,
approximately 15 laborers were needed to cultivate 100 acres of rice
land or 6 to 7 acres per hand (House 1954a:151). Presumably, colonial
practices demanded greater labor requirements. For example, the average

27
Figure 3.
Extent of Rice Cultivation in Georgia by 1860.
(Adapted from Hilliard 1975.)

rice plantation in colonial Georgia had an average of 43 working hands,
but the average acreage was not recorded (Flanders 1933:43).
Given this labor requirement, a demand for African slave labor
developed. Prior to 1635, most slaves in colonial South Carolina had
come there with their masters from the West Indies. These Afro-Americans
made notable contributions to settling the frontier of South Carolina
(Wood 1974:27-34). It was the rice industry, however, which estab
lished a need for black slave labor. As the new staple became profitable,
considerable quantities of Africans were imported (Wood 1974:57).
Africans were thought to be better equipped to labor in rice fields than
either indentured whites or Indian slaves for several reasons (Wood
1974:37-42). Perhaps, two of the most significant were the African
adaptability to the swampy lowlands and the African knowledge of rice
growing.
Malaria thrived in the stagnant, swampy lowlands where rice was
cultivated (Childs 1940:32). Settlers to the Carolina-Georgia coast
frequently complained of the prevalence of "fevers." Malaria or
intermittent fever" was considered to be the worst among these, par
ticularly during its pernicious season in the Spring and Summer (Childs
1940:33).
Only in recent years has medical research shown that populations
exposed to malarial climates display a genetic trait which produces a
partial but heritable immunity to the disease. This trait, referred .to
as sickle-cell trait," is not harmful in itself, but if present in both
parents can produce offspring who suffer from sickle-cell anemia and
are likely to die in adolescence (Wood 1974:80; Savitt 1978:27-28).

29
People who are heterzygous for the sickle-cell trait are evidently
better able to survive and reproduce in a malarial region (Weiss and
Mann 1975:356). In the United States, studies have shown that a high
incidence of the sickle-cell trait presently exists among the "Gullah"
Negroes of the South Carolina and Georgia coasts. This finding sug
gests that Afro-Americans in these former rice-growing areas have per
petuated this genetic characteristic (Gunn 1975:5).
Given the medical knowledge of the day, it is doubtful that rice
planters understood the sickle-cell trait. It appears, however, that
some planters observed fairly early that blacks were less likely to
suffer from malaria and other diseases prevalent in swampy lowlands
than whites (Poste!1 1951:74-75; Savitt 1978:22-24), which established
a preference for black labor in rice fields. Thus, it is clear that the
Negro's adaptability to the coastal environment partially contributed to
the tradition which defended Afro-American slavery on the basis that
blacks were more suitable to perform work whites were unwilling to do
(Gunn 1975:6).
Another possible explanation for the preference of African labor
in the cultivation of rice was the African knowledge of rice growing.
Historians have overlooked this possibility until recently. Primarily,
through the pioneering efforts of Peter Hood, a preliminary investiga
tion of this suggestion has been undertaken (see Wood 1974:55-62).
Rice planting and processing involved obstacles to Europeans,
since they were generally unfamiliar with rice culture. Unlike
Europeans, Africans from several regions in West Africa were successful
rice producers (Wood 1974:58-59). In fact, some West Africans were

cultivating rice as early as the fifteenth century (Latter 1969:17).
A preference for slaves from rice growing regions of West Africa was
established early by white Carolinians. These areas included the
Congo-Angola area (Heyward 1932:172-175), Senegambia area (Clark
et al. 1962:226-239; Lauer 1969:47-48), and the "Windward" or "fold"
coast of present-day Sierra Leone and Ghana (Corry i960). Slaves from
the rice coast of Africa were advertised by many firms. Between 1732
and 1765, Charleston alone handled "53 ship loads from Gambia and 29 fr
the Gold coast" (Lauer 1969:48).
Additional evidence for the presence of slaves from West African
rice growing areas is provided by the prevalence of African personal
names. The names "Banyun," "Bram," and "Bu'lunda" among slaves and
their later descendants on the Carolina-Georgia coasts strongly suggest
that the people from the rice growing areas of Senegambia were repre
sented among the slave imports (Lauer 1969:48).
Besides the actual physical labor, Afro-Americans made signifi
cant contributions in the planting end processing of rice. The motions
used in planting rice seeds such as the pressing of a hole with the
heel and covering the seed with a foot were also practiced in West
African rice planting. The flat, wide, coiled winnowing baskets were
trade by slaves and were typical of West African ones both in style and
techniques of manufacture. Methods of pounding and polishing rice by
hand were possibly told to white planters by their African slaves
(Wood 1974:61-62).
Other factors, however, may have established a need for African
slave labor. Briefly, these include the abundance of black labor,

31
the difficulties posed by the enslavement of Indians, and the insuf
ficient supply of indentured whites (Wood 1974:37-48). Despite the
fundamental role these factors contributed to the existence of African
enslavement in the New World, it is suggested that the practical advan
tages offered by African physical adaptability and African knowledge of
rice growing fostered the Carolina-Georgia rice industry.
Methods of Irrigation and PI anting Rice
Two different species of rice were cultivated in the Southern
United States: lowland rice, Oryza sativa, and upland rice, Oryza mutica.
Lowland rice, a marsh plant, required an abundant supply of water for
ripening (Austin 1893:9). Of its many varieties, "goldseed" rice and
"white" rice were cultivated in the lowland areas of the Atlantic Sea
board (Knapp 1900:5). Upland rice was grown by dry culture in various
hilly areas of the cotton belt. It was not a commercial crop, but
grown essentially for home use (Gray 1941:723).
Lowland rice was probably first grown in moist soils without
irrigation (Gray 1941:229). Around 1724, irrigation was employed in
what were called "inland swamps." These were small fields adjacent
to freshwater streams. The rice fields were irrigated with water-
stored in reservoirs formed by dams. After irrigation, the water was
drained through ditches into streams. Because the water supply in the
reservoirs was dependent upon rainfall, the crop was subject to either
frequent floods or drought, often making yields small and uncertain
(Heyward 1937:12-16; Wi1 ms 1972:50-51 ).
By the late eighteenth century or perhaps earlier (see Wilms
1972), rice planters began to recognize the advantages of cultivating
-
- ,

32
on the tidal swamps to permit the inflow of freshwater backed up at
high tide and the outflow at low tide. The fields could be flooded and
drained as desired (Cole 1927:599). But before the tidal marsh could
be cultivated, it had to be cleared, diked, and drained. Reclaiming the
tidal swamp was a tremendous undertaking which required considerable
expertise and its completion often took several years (Heyward 1937:
13-20; Wilms 1972:53; Hilliard 1975:60).
The innovation of transforming a marshland into a piece of culti
vable land, a polder, cannot be credited to the tidewater planter.
Construction of polderlands has been undertaken since ancient times,
and today they are located all over the world (Wagret 1968). Perhaps,
the best known are those of Holland. In fact, indications are that
several Carolina-Georgia tidewater planters employed Dutch expertise in
designing their hydraulic systems (Doar 1970:13; Scott 1973:73-74;
Johnson 1930:61).
Articles describing the construction of the irrigation systems
used by tidewater rice culture are found in several contemporary
journals (e.g. Alisten 1846, 1854; DeBows Review 1847; Spalding 1845).
Several secondary sources have provided additional details for this
procedure (e.g. Heyward 1937; House 1939, 1954b; Gray 1941; Doar 1970;
Hilliard 1975). In very general terms, the procedure for constructing
a tidewater rice irrigation system involved six major steps, 'first,
the'length end width of the required acreage was measured off and the
trees were cut down- and moved out of the way. Second, a temporary ditch
and embankment were built around the entire area to keep water out, while
tree stumps and debris were removed; third, the ditch was filled and

33
elevated to form the permanent embankment. Fourth, small channels were
bridged and the trunks were installed. (Trunks were the devices which
controlled the water flow.) Fifth, individual fields 12 to 22 acres were
laid out. (Each of these fields could be filled or drained individually.)
Sixth, these larger fields were subdivided into plots of 1/4 or 1/3
acre to aid the water movement (Gray 1941:726-727; Doar 1970:9-13;
Hilliard 1975:59-60). These smaller subdivisions, sometimes referred to
as "tasks" (Gray 1941:734), provided not only drainage but also the
standard measurement for certain labor requirements (House 1954a:153).
Thus, the basis for the task labor management system, which was used
extensively in the cultivation of tidewater staples, was derived from the
small segments formed by the irrigation system used in rice culture
(Stampp 1956:55).
Regular maintenance of the irrigation system was essential in
order to keep the fields in proper working order. The ditches and
drains were thoroughly cleared of mud and trash annually (Austin 1393:
18). Breaks in the embankments caused by storms, high tides, and
"freshets" or floods were frequent, despite precautions taken to pro
tect the outer embankments (Spalding 1830; Southern Agriculturalist
1834). In addition, small animals including rats, snakes, and crayfish
often made holes on the outside of the bank which very often became
leaks and then breaks. If left unattended, these small breaks would
interfere with the drainage system (Doar 1970:30). Such breaks and
leaks had to be mended, and "ditchers" or "bank menders" (Heyward 1937:
236) were engaged regularly in repairing dykes.

34
Usually the annual cycle began in the fall and early winter with
the "breaking" of the soil. This was done by hand with heavy hoes.
Plows were used by some planters, but the small subdivided rice fields
and the numerous ditches made plowing difficult. Moreover, many fields
were not dry enough and draft animals were expensive to maintain (Gray
1941:729; House 1954b:27-28).
Planting occurred beginning in the middle of May, depending upon
the weather (Austin 1893:19). Techniques varied among planters, but
either of two methods was generally employed. In the traditional method,
"covered rice" seeds were carefully sowed in trenches measuring 3 to 4
inches wide, 2 inches deep, and 13 to 14 inches apart. After sowing,
the trenches were covered with 2
inches of soil
by hoes, rakes, or
covering boards (Gray 1941:727-728). The other method, "open trench,"
required that the seeds be coated in clay before sowing. Upon comple
tion of sowing, water was let in the open trenches without covering them
with soil (House 1942a:188). Open trench planting was not widely used
in Georgia. It was apparently developed by South Carolina planters in
an attempt to increase yields from exhausted soils (House 1942a).
Once the seed rice was planted, inundation of the fields was needed
at various stages of growth. Irrigation practices, like planting methods,
varied according to the planter. All irrigation, however, was utilized
for three purposes: the initial flow or sprout flow was applied to aid
the seed in sprouting; subsequent flows known as stretch flows were
applied to keep down the growth of grass, weeds, and other pests; and
the final or harvest flow strengthened and supported the stalks bearing
ripened grain before the rice was harvested (House 1954b:3).

35
A number of pests preyed upon the growing rice (see Austin 1893:
31-37 for a detailed discussion). Two of the most frequently mentioned
were volunteer rice and rice birds. Volunteer rice or "red rice" was the
product of seed from the previous year which remained in the soil and
sprouted with the new crop (Austin 1893:23). A few of these red grains
among the white rice were believed to seriously lower the quality of the
crop. Consequently5 planters carefully selected the seed rice and had
slaves to remove these self-planted sprouts whenever they were detected
(Gray 1941:729). Rice birds presented another serious problem for they
often consumed large portions of the crop. These birds were unavoidable
since they appeared in May for approximately two weeks and then re
appeared in September for the same length of time. Some planters planted
in between May and September, but yields were small (Heyward 1937-31-33).
Most planters, however, employed "bird minders" to frighten the birds
away with gunfire, noise, or decoys (Heyward 1937:32; Gray 1941:729).
Harvesting, Milling, and Marketing
Rice was harvested by hand sickles in late August or early
September. The cut rice was allowed to dry for a day or so and then it
was tied together in sheaves. These sheaves were carried to the mill
or barnyard by flats which floated on the main canals or overland by
o-carts (Spalding 1835). The rice was stacked in the barn ur.ti 1
thresh!ng season (Gray 1S41:729).
Preparing the rice for the market involved several processes:
"threshing," the removal of the rice from the plant; cleaning or
"winnowing," the separation of the grain from the chaff; "pounding,"

36
or "grinding," the removal
of the outer hull and blowing it away from
the shelled rice; finally
rice grains plus powdered
"polishing," which resulted in the clean wh
fragments, known as rice flour (House 1954b:
te
59-60).
Threshing was done by hand using flails for most of the planta
tion period. Sometimes a floor was constructed for threshing; an
"efficient" floor was 110 feet long and 50 feet wide and composed of
three ranges of boards" (House 1954b:60). Rice was cleaned using wind
or a fan. More often a winnowing house was constructed for this pur
pose. Winnowing houses were small, usually square-shaped buildings
raised upon stilts. The rice was dropped through a hole in the floor
of the house, and the wind blew away the chaff. The grain landed below
the hole on a clay or earthen floor. It was swept up and placed in
containers (Smith 1936:20, and illustrated). Not until the mid-
nineteenth century was a reliable threshing mill developed which
threshed and cleaned the rice (Gray 1541:729-730).
In colonial days, the outer husks of the rice grain were removed
by hand-powered mills and hand fans blew away the chaff. These "husking"
machines consisted of slabs of wood about 2 feet in diameter revolving
one on the other. The inner shell was removed by hand by use of a
mortar and pestle (Smith 1936:21; Gray 1941:287). Later in the nine
teenth century, mills were developed which combined ail of the processes
of milling ricegrinding, cleaning, pounding, screening, and polish
ing (Gray 1941:730; Lawson V975:ll, illustration of the devices used in
a rice pounding mill). Power for these mills was supplied by tides,
steam engines, animal power, or wind. The steam-powered mills were the

37
most efficient (House 1954b:62-65). Plantations with threshing machines
often housed machinery for threshing, pounding, and polishing within the
same structure (Smith 1936:22; House 1954b:62). Only very large planta
tions, however, could afford to be fully equipped with milling facili
ties. The planter with pounding and polishing mill could ship pounded,
clean grain to the market, whereas the planter without had to ship
rough rice, or "paddy." Also, the planter without a mill had to
rely upon toll mills to complete the processing of the rice for a per
centage of the grain. Several European buyers, however, preferred to
purchase rough rice and mill it in their own countries. Because of
this there was always a substantial market for rough rice. By 1850,
most foreign shipment was paddy, and pounding mills were utilized pri
marily for the domestic market (Gray 1941:730; House 19545:68).
World prices for rice were based upon polished rice and weight
(House 1954b:67). The commercial standard weight of rough rice was 44
to 45 pounds a bushel (Austin 1893:24; Doar 1970:18), and the average
yield of rice lands varied from 25 to 60 bushels of rough rice per
acre (Gray 1941:730). Polished rice was packed in "tierces," each
weighing 600 pounds. The quality or grade of the rice was judged accord
ing to the number of bushels it took to make one tierce. Generally.
20 bushels or less of rough rice to make one tierce of polished rice
was considered a good crop, more than 20 an inferior one (Doar 1970:
18), Prices varied considerably during the antebellum period. The
best prices for Georgian rice, however, were obtained between 1840 and
1860, or the "golden age" of Georgian rice production. At that time

38
rice sold for $.80 to $2.23 per bushel or 2 to 4.75 cents per pound
(House 1954b:78).
Rice, like all other plantation crops, was marketed through fac
tors in major port cities. The factor served as the "middleman" be
tween the planters and the rice merchants. He was tine banker and stock
broker for the planter and provided him with credit in meeting opera
tional costs. Since the planter depended upon the returns from his
crop for cash, he was often in debt to the factor when rice sales were
low (Easterby 1941). Prior to 1840, most Georgian rice was sold through
factors based in Charleston. After that date, Georgian planters in
creasingly sold their rice from Savannah. By 1859, Savannah had 82
factor establishments. Savannah experienced a rapid growth as a rice
port. Its population increased; a thriving trade developed with ports
in the West Indies, Mobile, and New Orleans; and it became an important
social and cultural center (House 1954b:70-79).
The rice industry on the Carolina-Georgia costs was seriously
beginning to decline on the eve of the Civil War, although large-scale
rice production did not vanish completely from the area until the end
of the nineteenth century. Competition from rice-growing states in the
southwest and the rising cost and scarcity of slaves (Cole 1927:600-
601) were the major causes for this new crisis facing the rice planter.
Nonetheless, the industry managed to survive because of large crop
yields. After the Civil War, however,
prosperity to the industry were futile,
increased flooding from freshets caused
attempts to restore the former
The new labor system and
by accelerated erosion in the
interior (Trimble 1969) made rice cultivation unprofitable. Gradually,

39
the interest in rice was abandoned. Finally, a major storm destroyed
the last remaining vestiges of rice culture at the turn of the century
(Vanstory 1970:83). The following passage taken from an article in the
Darien Timber Gazette summarizes the sentiment toward rice growing among
post-bellum planters in Georgia after a devastating flood:
the loss will be a heavy blow to many of the planters
especially where money was borrowed to plant with. Several
have already declared they will not plant any more as the
crop is so uncertain. Every year the acreage is decreasing
end the upriver planters are nearly discouraged.
(27 August 1887, Darien, Georgia)
Long-Staple. Cotton Culture
Like rice, the production of long-staple cotton was restricted to
the coastal fringes for most of the antebellum period. It became known
as "sea-island cotton" because the staple was believed to degenerate
when removed from the influence of saltwater (Alisten 1854:593). Its
introduction to the coastal states, unlike that of rice, occurred in
Georgia not South Carolina. Initially grown in the Caribbean, long-staple
cotton made its way to coastal Georgia through expatriated Georgians who
had made their home in the Bahamas after the Revolutionary War. The
success of the cotton in the Bahamas led the former Georgians to send a
few bags of the seed to their friends and relatives in Georgia (Coulter
1941:65-68: Hammond 1897:16-17). Gradually, the fruit yielded a finer
and better cotton titan that grown in the Bahamas (Johnson 1930:24-25).
By 1789, substantial quantities were produced in Georgia and its cultiva
tion spread to nearby South Carolina.
For most of the antebellum period, sea-island cotton was grown
exclusively along the narrow strip from Charleston, South Carolina, to
W vi S : W '.**' --5 : y '.,y SSSjaBJ*? J a ;- peegS-

40
the St. Johns River in Florida (Hammond 1897:19). In the early forties,
it was discovered that the crop 'was adaptable to sea marshes if these were
reclaimed and to interior lands as much as 100 miles from the Atlantic
Ocean. This discovery led to the expansion of the industry into the
interior lands of Georgia and Florida. By 1858, Florida had become a
leader in the production of sea-island cotton, and the volume of exports
for the staple increased by 50 percent as a result of the expansion in
cultivation (Gray 1941:733-734).
Long-staple cotton was strikingly distinctive from the short-
staple which was cultivated elsewhere in the Old South. As the name
implies, the fiber was longer than the short-staple. The fiber length
of sea-island cotton ranged from 1 1/2 to 2 inches as contrasted with
5/8 to 1 inch for the short-staple (Gray 1941:731). The quality of the
two was also dissimilar. The fine, silky long-staple was used in the
manufacture of lace, thread, and cloth of silky luster. On the other
hand, short-staple cotton was used primarily for making coarse cotton
cloth (Gray 1941:731; Phillips 1929:91). Moreover, the long-staple
required a longer growing season, more hands per acre, and more tedious
methods for cultivation, harvesting, and processing (see Otto 1975:53-61
for a detailed comparison of the two varieties of cotton).
Methods of Planting and Cultivation
The soil was prepared for planting in Februa
Soils varied from light sandy loams to heavy clays
were preferred (Phillips 1969:271). On lowlands "
that employed in rice culture was often necessary
y or early March.
, but the sandy ones
ditching" similar to
to drain an area before

41.
planting (Gray 1941:731). High ridges 4 1/2 to 5 feet apart were con
structed before the seed was planted. These were usually made with
hoes, particularly in the old "sea-island region" (Gray 1941:735).
Plows came into wider use during the late antebellum period, when more
time was needed to manure exhausted soils (Gray 1941:734). Between
late March and the middle of April, the seed was sown. Once the cotton
plant sprouted, it had to be "thinned" with hoes several times. Gener
ally cultivation consisted of 4 to 8 hoeings, and grass was pulled by
hand.
With a few notable exceptions, most sea-island cotton planters
did little to enhance the productivity of their crop prior to 1830. At
that time, confronted with fluctuating prices and soil exhaustion,
experiments with manuring and crop rotation were undertaken (Johnson
1930:51-54). Manuring soon came into wide use,and various types of
manures were utilized such as marsh mud, crushed shell, and guano. The
practice of crop rotation, on the other hand, was limited to a few enter
prising planters (Johnson 1930:60-64). Careful selection of seeds was
perhaps the universal practice adopted by most planters to improve the
quality of the cotton. Several varieties were known and some reputed
to yield cotton that sold between $1.50 to $2.00 a pound. Unfortunately
the production of these seeds was kept secret and this knowledge is now
lost to agriculture (DeBows Review 1867:87).
Harvesting, Processing, and Marketing
A great deal of care had to be taken while harvesting the cotton.
The fiber was gathered as soon as the pod opened in order to protect it
from injury. Also, trash, leaves, and dirt had to be separated from the

42
fiber carefully. Harvesting required 10 to 12 pickings, whereas the
short-staple only required three (Gray 1941:735). After it was picked,
the cotton was placed on a platform to dry in the sun and later stored
in the barn to await ginning.
The famous "Whitney saw gin," invented for the removal of seed
from short-staple cotton, was not used for ginning sea-island cotton;
instead, "roller gins" were used. In the early days these were hand-
powered mills derived from the Indian churka, Later improvements to
this apparatus led to the invention of gins powered by animals or
steam (Hammond 1897:72). After the cotton was ginned, it was "mated"
that is the residue from the broken seeds was removed. The clean
cotton was then hand-packed in bales, since hand-packed bales were less
injurious to cotton and consequently brought higher prices than those
packed by a press (Gray 1941:736).
Both acreage-per hand and yields per acre were severely limited
because of the laborious cultivation and processing methods. Where cul
tivation relied exclusively upon hand labor, the average number of acres
cultivated per laborer was 3 to 4. Where the plow was substituted for the
hoe, the average number of acres per laborer was 6 to 7. The average
yield per acre ranged from 125 to 400 pounds, although yields recorded as
high as 2,000 pounds per acre were known (Gray 1941:737; Phillips 1969:
275).
Only superior long-staple cotton commanded very high prices, but
its production was limited to a small group of planters (Allston 1854:
5S6). For the rest of staple, price fluctuations were considerable.
The average price range during periods of good to high prices was $.40
to $.80 a pound, and at periods of depressed prices $.14 to $.25 a

pound. The decade before the Civil War was a period of rising prices
after a long period of depressed prices. At that time, the average
price of cotton was between $.37 and $.47. However, after the Civil
War, the quality of the crop deteriorated and its production became less
valuable. Machinery was soon developed which produced thread from the
short-staple as delicate as that made from the long-staple (DeBows
Review 1867:83). Finally, devastation from the boll weevil made the
production of the crop totally unprofitable (Brown and Ware 1958:69).
Summary
Rice and long-staple cotton are two very dissimilar cultigens.
Each crop requires distinctive modes of cultivation and processing. For
the most part, each is cultivated in a very specialized habitat and is
produced for its own peculiar market. Yet analogies between the two crops
exist in the history of their production in the Georgia tidewater.
Both crops were restricted to a synonymous geographic range and
required a considerable labor force. These similarities were particularly
significant to the demography and economy of coastal Georgia. The
Georgia tidewater, like that of South Carolina, became a region with
a heavy concentration of slaves and a few wealthy slaveholders where
absenteeism frequently prevailed.
Another analogy was evident in general cultivation practices.
Often, sea-island cotton growers were also rice growers or former rice
growers. Apparently, certain practices established for rice, the
older staple, were frequently adopted for cotton. Some of these, though
harmless to the rice, proved to be detrimental to the cotton industry.

44
%
Clearly, the conservatism toward the use of hand labor, manures, and
crop rotation had its origins in rice culture. Manuring was not essen
tial to rice growing since silt loam from irrigation aided soil fer
tility (Cole 1927:601). Similarly, hand labor was vital to rice cul
ture because of the difficulties of plowing. Perhaps an early implemen
tation of plowing and manuring in the cultivation of long-staple cotton
would have resulted in lower production costs and higher yields than
those actually experienced by most coastal planters. At least, the
success of the sea-island cotton industry in Florida was partially re
lated to the greater use of the plow (Gray 1941:737). Unfortunately,
the failure to employ crop rotation methods by both cultures resulted in
early soil exhaustion, which had disastrous effects upon post-bellum
agriculture in the area.
Likewise, labor practices were very similar. Here again, the
task system which developed from rice culture was later used for cotton
culture. Under the task system, field hands were classified as full,
three-quarter, half, and one-quarter hands (Flanders 1933:143-144).
Generally, each full laborer was expected to work a task of 1/4 to 1/2
acre everyday, whether in rice fields or cotton fields. This system
was particularly advantageous if the worker completed the task by mid
afternoon. Upon completion of the task, he or she had the remainder of
the day to tend to personal needs. Conversely, under the gang system,
the other major labor system for plantation slavery, workers labored in
the fields from sunrise to sunset (Phillips 1929:279-280; Stampp 1956:
53-56).

45
f-'-'jf
It has been suggested that the task system provided slaves at
sea-island cotton plantations in coastal Georgia with sufficient leisure
time to improve their material lots (Otto 1975; MacFarlane 1975). Under
the task system, the quality of slave life could be greatly improved
through hunting and fishing, selling handicrafts or plantation produce,
or by growing foods to supplement plantation rations. Presumably, the
use of the task system at a rice plantation should have resulted in the
same conditions. But how similar were the conditions of slavery in two
very different crop loci? The adaptations of slavery to long-staple
cotton culture in coastal Georgia has been investigated utilizing both
documentary and archaeological resources (Otto 1975; MacFarlane 1975).
A similar investigation of slave adaptations to rice culture along the
coast of Georgia is presented in the following three chapters.

-s;vsjssrw^ %¡ i
CHAPTER 3
HISTORICAL AND ARCHAEOLOGICAL INVESTIGATIONS
AT BUTLER ISLAND
Butler Island is one of several "brackish marsh islands" (Long
1953:11} within the Altamaha drainage in McIntosh County, Georgia (see
Figure 4). These islands are termed brackish because they become
covered with saltwater when tides are pushed by high wind and are in
undated with freshwater when rivers overflow as in the case of freshets
(U.S. Dept, of Agriculture 1961: 21), The land consists of sediments
washed down from streams flowing out of the piedmont and the coastal
plain. Soils vary but are primarily clayey. Generally, these marsh
soils are poorly drained and can be extremely acidic, particularly on
unreclaimed marsh (Long 1958:13).
During the nineteenth century, most of these islands were con
verted for rice culture. Later in the twentieth century, truck farming
replaced rice culture. Most recently, these marshlands have become
managed breeding grounds and refuge for waterfowl. Since 1954, several
of these islands have been acquired by the Game and Fish Commission of
the Georgia Department on Natural Resources for the development of a
waterfowl management area. Butler Island and adjacent Champney Island
were the first to be acquired for the program (Anonymous nd). Today,
Butler Island has become the headquarters for the Altamaha Waterfowl
Management Area.
46

47
Butler Island and Environs.
Figure 4

mmmmm mm aajggjg
43
Archaeological investigations were undertaken at Butler Island in
response to a need to locate and identify archaeologically sensitive areas
within the confines of waterfowl management activity. Certain manage
ment practices such as seasonal flooding, plowing, and dredging activities
would adversely affect or destroy archaeological sites. In past years
on several occasions, archaeological materials have been unearthed as a
consequence of these practices. To prevent these practices, funds were
obtained for survey, inventory, and testing at Butler Island from the
grant-in-aide program of Heritage Conservation and Recreation Service
(HCRS),U.S. Department of the Interior, and administered by the Historic
Preservation Section, Georgia Department of Natural Resources. Nine
sites were located and identified. All were found to be historic sites,
but three have prehistoric components.
In this study the slave village sites are the subject of discussion,
although the other plantation period sites are briefly described. A de
tailed description of all the sites will be part of another archaeological
report (Singleton nd).
Another major aspect of the Butler Island project has been the
archival research. Documents relating to the Butler family ownership
of Butler Island are voluminous. These records have not only provided
the background data presented in this chapter, but have also been util
ized in the identification, dating, and interpretation of the archaeologi
cal resources.
Ownership and Qccupation
Unfortunately, the first colonial grantees of Butler Island
could not be ascertained. It is possible that the Lochlan McIntosh

49
family acquired Butler Island during the colonial period. Between the
years 1758 and 1775, Lochlan McIntosh purchased several large tracts of
marshland in the Altamaha delta (Jackson 1973:17-18). The precise iden
tification for some of these islands is unclear.;. Butler Island
may have been one of them.
Although the first owners of Butler Island are unknown, Major Pierce
Butler apparently made the first attempts toward reclamation and cultiva
tion of the island. While the family's ownership was complex,* an
attempt has been made to summarize it (see Table 1). The island was in
the possession of the Butlers for close to 120 years, and several family
members have made this island a very significant historic site.
Born in Ireland, Major Butler came to America in 1766 as an
officer in the British Army/, h'e liked America and decided to settle
here. Later in 1771, he married Polly Middleton, the daughter of a
wealthy South Carolina planter. He and Polly had one son and four
daughters. During the American Revolution, Major Butler sided with the
revolutionaries and became an important political figure in South
Carolina. In 1787, he was a delegate to the Constitutional Convention
in Philadelphia, and he was elected twice as a senator from South
Carolina to the U.S. Senate (Scott 1973:53, 1978:6-7).
At the Constitutional Convention, Major Butler was one of the
principal proponents arguing that state representation in the House of
Representatives should be proportional to the total population, both
slave and free (Alden 1971:86-87; Scott 1973:6). Butler and his
*See Scott (1961:xx-xxi) for a genealogy of the Butler family.

Table 1
Ownership of Butler Island, 1790? to the Present
Time Owners
1790?-1822
Major Pierce Butler
1822-1036
John Butler and Pierce Butler, Jr.
(formerly John Mease and Butler
Me ase), bu t e s tate was adminis-
tered by the Major's daughter
Frances and his son Thomas.
1336-1847
John and Pierce, Jr., assumed full
responsibilities of ownership until
John's death in 1847.
1847-186!
Properties jointly owned by Pierce,
Jr., and John Butler heirs, but
slaves had been divided equally
between the two.
1861-1865 (Civil War)
Abandoned.
1866-1878
Frances Butler Leigh (Pierce
Butler, Jr., and Fanny Kemble's
daughter).
1978-1910
Leighs still own but have leased
the island to rice planters.
1910-1920
Passes through several hands, pri
marily rice growers.
1920-1949
Colonel Huston.
1949-1954
R. J. Reynolds.
1954-present
Game and Fish Commission, Georgia
Department of Natural Resources.

51
supporters believed that the slave states needed this type of represen
tation in order to prevent the possibility that a majority of representa
tives from the free states would attempt to abolish slavery. The Major's
view epitomized that of the small but powerful elite who desired to
perpetuate and expand the institution of slavery (Scott 1978:7).
In 1790, Polly died and the Major took his family and slaves to
Georgia. Like many South Carolinians, the Major moved southward in
search of new lands. The developing market for sea-island cotton and
the revival of the rice industry made the fresh, fertile lands of
Georgia attractive to the immigrants.
By 1799, Major Butler had acquired several thousand acres in
McIntosh and Glynn counties. Included in these were Butler Island,
Butler Point on St. Simons Island,* Little St. Simmons Island, and
pine land on the mainland known as Wocdville (Butler Estate Papers,
1799)(see Figure 5 for locations). In addition to land, the major
also acquired such a vast number of slaves that by the time of his
death he owned close to 700 (Butler Estate Papers, 1821). Eventually,
the Major became one of the wealthiest planters in Georgia.
The Major built a fine mansion at Hampton plantation on St. Simons
Island, where he spent most winters. It was here that Aaron Burr sought
refuge after the duel in which he killed Alexander Hamilton. Burr
visited the plantation for a month and he described it in many of
his letters to his daughter, Theodosia (Van Doren 1929). The Major
Concurrent with the
archaeoloqical
research at
4-,
Butler Island,
similar research has been undertaken at several sites of the Hampton
plantation (see Mullins, 1978; Fairbanks and Mull ins-Moore, 1980; anc
Mullins-Moore, nd).

Figure 5. The Butler Georgian
Estate.
(Adapted from Scott 1561.)

53
and iris daughters, however, made Philadelphia their permanent home.
Thus, the responsibility of running the estate during the Major's
absence was taken care of by an overseer.
When the Major died in 1322, his properties were inherited by his
grandsons, John and Butler Mease. The Major had wanted his only son,
Thomas, to take on the responsibility of administering the Georgian
estate after his death, but Thomas, who hated the Major, refused and
went to Europe to live (Scott 1973:54). Thus, the Major disinherited his
son and made his grandsons heirs. After the Major's death, however,
Thomas helped his sister Frances to administer the estate until the
Mease boys came of age and were able to do so. At the Major's request,
his grandsons had to change their surnames to Butler. Consequently,
John Mease became John Butler and Butler Mease became Pierce Butler, Jr.
In 1834, Pierce Butler, Jr., married Fanny Kemble, the famous British
actress. During her first and only trip to her husband's Georgian
estate in 1839, she wrote the controversial account of slave life,
A Journal of A Residence on A Georgian Plantation In 1838-1939, which was
published more than 20 years later.
Fanny and Pierce came to Georgia with their two young daughters
for approximately four months. Two months were spent at Butler Island
and another two at Hampton Plantation. Fanny was outraged at the con
ditions under which the slaves lived, particularly the suffering endured
by women. Fanny's reaction resulted in many heated arguments with
Pierce, which intensified the couple's preexisting marital problems.
After their return to Philadelphia, the marriage grew increasingly worse.
In 1849, Fanny returned to Britain, the couple were divorced, and
Pierce gained custody of their daughters (Scott 1973:11).

54
Fanny always wanted to publish the journal she had kept while in
Georgia. In 18593 she began giving this serious thought for two rea
sons: first, she was no longer married to Butler; second, the estate
she describes in the journal no longer existed (Scott 1973:119).
Pierce Butler had fallen into serious financial debt as a result of the
crash in 1857, and more than 400 slaves were sold to pay his debts
(Thomsen 1863). In light of these new circumstances, Fanny reasoned
that her journal would be thought of less in a personal way than as a
social statement of slavery in the United States (Scott 1973:120).
The Emancipation Proclamation and the events of the Civil War gave her
the final impetus toward publication. Fanny felt that the journal would
help "explain Lincoln's war aims and show the nature of the system that
the North was fighting to abolish" (Scott 1973:122). The journal was
published May 1863 (Kemble 1882:205).
It is difficult to estimate the reaction tc the journal imme
diately after its publication. A few Southerners have suggested that
the journal single-handedly swayed British opinion against the South,
which made aid to the Confederate War effort impossible (e.g. Perkerson
1952:133-134), Even among Fanny's admirers this opinion has prevailed
(Armstrong 1938:432). It has also been argued, however, that the
journal had little influence upon the contemporary public (Lombard
1930). Recent evidence has indicated that by the time the journal was
published, British opinion had already turned against the South
(Scott 1973:124). More significant than the criticism related to the
influence of the journal upon the contemporary public is the cri
ticism related to its accuracy and authencity. Among historians

viewpoints vary. Apologists for slavery tend to be very critical (e.g.
Phillips 1929:259-265; Eaton 1949). Other historians have found a great
deal of accuracy in Fanny's descriptions of slave life, although some
suggest that the journal is overromaticized (e.g. Stampp 1955; Genovese
1974; Owens 1975). Of all the critics, Georgians have been the most
hostile (Holmes 1978; Lovell 1932:196, 213; Cate I960). Even today,
this sentiment persists (Fancher 1971:102; e.g. Lewis and Hule 1978:120).
Butler Island was deserted during the Civil War, with the excep
tion of a few slaves who remained. In 1866, Pierce Butler, Jr., and
his daughter, Frances, returned to Butler Island to restore rice plant
ing (Vanstory 1970:87). Frances kept a journal, which was later pub
lished (Leigh 1883). Her journal became one of the few documents that
describe labor conditions in the South during the postwar years (House
1942b:157). Fierce Butler died in 1867. Frances, accompanied by her
German maid, managed the plantations alone for four years. Frances
married Reverend James Leigh in 1871 while she was in England and they
remained there until 1873, at which time they returned to Georgia.
The Reverend Leigh was very much interested in the South and helped to
establish St. Cyprian Church for the blacks of Darien. Leigh wrote of
his experiences in coastal Georgia in Other Days (1921).
While in England, the Leighs contracted some English workers to
supplement the black labor at Butler Island. These British laborers
resided at Butler Island and occupied the former slave hospital. As
laborers, however, they were a disappointment (Leigh 1883:202-213).
Discouraged, in 1876, the Leighs decided to return to England and
placed the management of the islands in the hands of a manager.
1878 they gave up their planting interest entirely.
In

56
Information of ownership and land use at Butler Island in the
years between the Leighs' departure and 1920 has been very sketchy. It
has been assumed that the island was abandoned and left unattended
(Vanstory 1970:89). Yet deed records, local informants, newspapers,
and a valuable map (see Figure 6) have suggested a different interpreta
tion.
Evidently, the Leighs leased and sold portions of their proper
ties in the Altamaha delta between 1878 and 1910. By 1910, all of their
properties were sold (McIntosh County Records, Deed Books H--52, I-I20,
1-160). These properties, which embraced all of Butler Island and
Generals Island, were sold and leased to rice planters, who in turn con
tracted with tenants to farm small plots for a portion of the crop (as
shown in Figure 6). Local informants have indicated that by the 1880s,
tenants no longer lived at Butler Island. Instead, they lived at several
settlements on the mainland and commuted daily by small boats to Butler
and General's Islands. Two of these informants, both descendants of
slaves in the area, remember vividly planting rice at Butler Island in
the early 1900s.*
Long after the Leighs left Georgia they continued to correspond
with the local blacks (Letters from Descendants of Butler Slaves, 1890-
1903). Particularly, their only child, Alice Leigh, visited Georgia
often and on these visits she would give parties for the people (Cate
end Wrightman 1955:203). As children, present-day black residents of
Darien performed for "Lady Alice" on these occasions.** The impact of
Elizabeth Bates and Jim Cook, Personal communication, 1978,
Darien Georgia.
**Nan-Earl Wylly and Mrs. Gamble, personal communication, 1978,
Darien, Georgia.

57
Figure 6.
I he Subdivision of Butler and General's Islands
Tenant Plots (circa 1883).
i nto

the Butler family upon the Afro-Americans in the area had lasted well
over a century.
In 1920, Colonel Tillinghast L'Hommedieu Huston, part owner of the
New York Yankees, purchased Butler Island. He built a two-story house,
which is still standing today on the presumed site of the overseer's
cottage. Huston established a dairy farm and planted fruit trees and
vegetables. His special interest, however, was the cultivation of ice
berg lettuce. He revamped the old rice irrigation system and used it
for his fields. Huston remained at Butler Island until his death. In
1949, Butler Island was sold to Richard J. Reynolds, the tobacco million
aire, who made additional improvements to lettuce cultivation (Vanstory
1970:90). In 1954, Reynolds donated Butler Island to the state of
Georgia for the development of a wildlife preserve.
Plantation Management at Butler Island,
1802-i860
The overseer on the Carolina-Georgia coast has been described as
the most important "single element in the management hierarchy of the
rice belt" (Scarborough 1964:18). At least, this appears to have been
the case at Butler Island, where the Roswell King family successfully
managed the estate for over thirty years during the Butler family's
periodic absences.
Absenteeism as-a practice among rice planters was established
fairly early by colonial predecessors. This tradition developed, in
part, out of a need to escape the hot, malarial season as it was
generally believed that "miasmas" caused the disease. The ownership of
multiple plantations also necessitated the periodic absence from one or

59
more of these productive units (Bridenbaugh 1952:69). In the early days,
planters traveled to the North or to Europe, while others migrated to
their town houses in Charleston and other coastal cities. Seashore
settlements and pine!and, piedmont, and mountain villages began to sup
plement the coastal towns as summer resorts later in the nineteenth
century (Brewster 1970:109). The degree of absenteeism, however,
apparently varied along the rice coast. For example, in the Georgetown
district in South Carolina, one planter rioted that of 111 rice planta
tions in the area only 51 of the owners actually lived on their planta
tions, and of that number all were absent during the crop season from
May to November (/"Allston quoted in Cathey 1956:142). Some planters, who
could not afford a "dual existence" or who were too far from Charleston
or Savannah, remained on their estates throughout the year or most of
the year (Bridenbaugh 1952:70). In the Altahama Basin, rice planters
spent a major portion of the year on their rice estates and moved to
nearby pi riel and retreats or barrier island plantations for the summer
months. Often, they commuted each day to keep a watchful eye on the
progress of the crop and other management problems (House 1954b;10). The
Butlers were the only exception to this rule (see Wylly 1910:48-53).
They were never full-time residents of Georgia. Because they were
generally absent from their Georgian estate, the overseers regularly
corresponded with them. Fortunately, a large portion of this corre
spondence has survived to the present day.*
*The Roswell Kings Correspondence is found under two manuscript
collections of the Pennsylvania Historical Society, in Philadelphia:
The Butler Family Papers (primarily Boxes 9, 10, 11, 12), and the
Ulster Family Papers (The Butler Family Section, Boxes 31, 33, 34).

60
This correspondence provides a wealth of information related to
the day-to-day activities at Butler Island and the other Georgian proper
ties. In many ways these letters resemble similar correspondence
written by overseers at other plantations with absentee landlords (e.y.
Easterby, 1945; Bassett 1368; Clifton 1978). All have in common a con
cern with planting, harvesting, and marketing activities; labor manage
ment problems; and the health and welfare of the slaves. Occasionally,
the Roswell Kings' letters also supply some indices of slave behavior.
Most often, however, only deviant behavior was reported, and therefore
these references tend to be biased and one-sided. Yet, despite this
limitation, the Kings' observations provide valuable insights into
slave lifeways at Butler Island.
As overseers, Roswell King, Sr., and later iris son, also named
Roswell, belonged to the "overseer elite" (Scarborough 1966:158-177).
This minority among overseers was frequently employed on the Carolina-
Georgia rice coast but. was found in other localities as well (e.g.
Sitterson 1943). For the most part, members of the overseer elite were
conscientious and efficient men. They made significant contributions
to agriculture, and some became planters themselves. Often, they were
the sons of planters. In contrast with their colleagues, they were
usually well paid. Both the Kings earned as much as $2,000 a year.
Roswell King, Sr., had come to the Georgia coast from Sharon,
Connecticut, in the 1780s (Lewis 1975:39). In 1802, he became the
manager of the Butler estate, a position he retained until 1819. Some
time after that date, he became the manager of the Darien Saw Mill
(Darien Gazette, 6 January 1821). Kings son, Roswell, Jr., was

61
selected to replace him upon recommendations from Roswell, Sr., and
John Couper of Cannon's Point on St. Simons Island (Couper to Major
Butler, 15 February 1818). Roswell, Jr., presided as overseer until
1839. It was young King's resignation which resulted in the trip Pierce
Butler, Jr., and iris wife, Fanny Kemble, made to Georgia. Butler had
come to Georgia to hire a replacement for Roswell, Jr. Thomas Olden,
formerly the manager at James Hamilton Couper1s Hopeton plantation
(Anonymous 1965?), was hired to managed the estate. Olden died in 1841,
and Roswell, Jr., once again became involved with the management of the
Butler estate, but this time in the capacity of a steward. As a
steward, he supervised two overseers and managed the account hooks.
Roswell. Jr., held this post until the nrid-1850s (Scarborough 1966:168).
Unfortunately, most of the plantation records for the Butler
estate, particularly the overseers' records, are between the years 1803
and 1835, prior to Fanny Kemble's visit. Within these 32 years, however,
it is possible to obtain a very clear picture of planting and labor
management activities. After 1835, it becomes difficult to document
management practices, but it may be reasonable to assume that earlier,
established practices prevailed in later years.
Under the management of the Kings, the Butler estate was operated
efficiently. Both Kings had taken an active interest in planting. With
regard to labor management, runaways were few, and slaves received treat
ment comparable.to other slaves in the area. To a great extent, their
success as managers may have been related to the dutiful drivers of the
estate. Most of the drivers were loyal and reliable, and the Kings felt
a high regard for them. For example, in the selection of a new driver

62
for Butler Island, King sought the counsel of two drivers at Hampton
Plantation (Roswell King, Sr., 15 November 1818). Several years later,
Roswell, Jr., praised the work of a driver by saying: "The hands appear
to be fit enough to attack another crop, which can only be attributed
to Sambo's management in varying the work as much as possible" (Roswell
King, Jr., 24 September 1826). The death of a good driver was con
sidered great loss to the estate, as well as to the entire planter
community: "Driver Bram . was worth a great deal to the estate .
his death is regretted by every white person who know him" (Roswell
King, Jr., 1 February 1825). But problems with drivers were not un
common. One at Butler Island was involved in stealing large amounts of
corn from the warehouses and selling the corn to merchants in Darien
(Roswell King, Sr., 22 November 1806). Another driver, also at Butler
Island, apparently was an alcoholic (Roswell King, Jr., 29 August 1806).
Occasionally, a driver's cruelty to the field hands required that he be
removed:
I have been obliged to displace Morris at St. Anne.
He was always tyranical and brutal, which I gener
ally kept under, but lately he had become intemperate,
in fits of that kind, and has used great severity to
some of the negroes. (Roswell King, Jr., 16 July
1826)
Despite occasional problems, most drivers were good at their jobs and
to the field hands (see Van Deburg 1979,for a reassessment of the slave
driver). Generally, drivers "strove to mediate between the Big House
and the quarters" (Genovese 1974:378). At Butler Island, drivers
were particulalry vital to the management of a plantation which saw
few whites.

63
Of all the properties in the estate, Butler Island was the largest
in both acreage and in production. More slaves were utilized in the
plantation operation than at any of the other properties. As many as
300 to 400 slaves 1 ived and worked at this rice island and they occupied
four slave settlements or villages. Rice production, however, was not
the major interest of the estate until the 1820s. Prior to that time,
long-staple cotton dominated the planting interest (Scott 1961:xxxviii-
xxxix). Most of the cotton was planted at St. Simons and Little
St. Simons, but some was also cultivated at Butler Island, apparently
for crop rotation purposes (see Coulter 1941:104). Despite the interest
in cotton, considerable quantities of rice were produced before 1820.
As indicated in Table ?., the average annual production of rice was 712
tierces between the years 1802 and 1820. From 1821 to 1833, the
average annual production (1827 and 1829 are not included) was 1,246
tierces. Thus, in later years, rice production increased by approxi
mately 43 percent. Years of low production coincided with losses due
to storms, as well as to years when the acreage planted in cotton may
have exceeded that of rice.
The rice cultivation methods utilized at Butler Island were
generally typical of the region. As a rule, the most sophisticated
methods and equipment were employed. As early as 1802, rice fields were
manured, possibly with lime and marl (Barker, 1841-1844) to enhance soil
productivity (Roswell King, Sr., 17 September 1803). Fields were
planted with rice one year and then rotated with cotton to keep down
the growth of grasses (Coulter 1941:104). Occasionally, the fields
were prepared with plows, but hand labor predominated. In the processing

64
Table 2
Rice Production at Butler Island, 1802-1833
Year
Acreage
Number of Tierces3
1802
128
1803
337 1/2
1804
44
1805
1326
1806
--
468 1/2
1807
--
1186
1808
--
565 1/2
1809
--
678 1/2
1810
--
587 1/2
1811
__
. 857
1812
1739
1813

253 1/2
1814
--
829
1815
591
1816
--
513
1817
--
1058
1818
--
650
1819
407
1173
1820
227
543 1/2
1821
234
832
1822
272
1080
1823
358
1240
1824
342
84b
1825
318
1024
1826
399
1455
1827
--
1828
395
1331
1829
1830
505
1747
1831
552
¡557
1832
271
1695
1833
591
1660
Source:
Compiled from Annual Crop and Lives
lock Reports, 1803-1833
Note:
Average annual tierces, 1802-1830, is
712; average annual
fierce
- 3 O 1
-1833 is 1,246; increase equals 43%.
3
Tierce
= 600 lbs.
k|_ow production due to hurricanes.

of rice, experiments were attempted with every conceivable threshing and
pounding device. By 1835, the machinery for processing rice included
an animal-powered mill, two tidal mills, and a steam mill. The steam
mill was built in 1833; it housed both threshing and pounding machinery
and its cost was over $20,000 (Roswell King, Jr., 25 March 1932, 22
April 1832, 22 February 1833). Evidently, the threshing mills were not
sufficient for the amount of rice harvested, and a portion of it was
threshed by hand on "tarred" threshing floors (Roswell King, Sr.,
9 August 1806; Kemble 1961 (1863):109-110). Rough rice was also mar
keted, but large quantities were not sold until the 1850s (Pierce
Butler, Jr., Ledger, 1851-1855).
Besides rice and cotton, sugar was planted at Butler Island be
ginning in 1812. In fact, Major Butler had considered "turning the
island into a sugar plantation" (Roswell King, Sr., 1 November 1812).
Between the years 1812 and 1815, complete sugar processing equipment was
added to the plantation complex (Butler Estate Papers, 1815a).
Roswell, Sr., had become very interested in sugar culture and he sought
the advice of West Indian planters (Roswell King, Sr., 12 August 1815).
This interest in sugar was later pursued by Roswell, Jr. (King 1823).
Regardless of their enthusiasm for sugar culture, the acreage allotted t
it was always considerably smaller than that of rice or cotton.
Like most large plantations of the Old South (6a11 man 1970:23),
some provision foods were produced at Butler Island. Livestock were
raised and some fruit trees and vegetables were cultivated. The list
of stock included oxen, barrows, mules, mares, cattle, pigs, sheep,
turkeys, and other fowl(Butler Estate Papers, 1803-1830). The draft

66
animals were used primarily for operating machinery and pulling carts
and wagons. Wool was sheared from the sheep to make cloth (Roswell
King, Sr., 18 August 1813). On special occasions, cattle, pigs, or
sheep were butchered to feed the slaves (Roswell King, Sr., 16 May 1813;
Roswell King, Jr., 13 November 1823). But these occasions were appar
ently rare, and slaves were most often fed "barrelled pork and salted
fish" (Roswell King, Sr., 7 April 1816; Roswell King, Jr., 18 March
1821). Citrus fruit trees were planted along the outer bank of the
island, an area approximately five miles in length. The purpose was to
protect the dykes from storms or freshets (Spalding 1830). Several
visitors to the area marveled at the beauty of the orange and lemon
groves (Hall 1931 (1828}:234; Coulter 1941:96). Most of the fruit was
marketed and the remainder was sent to the Butlers or distributed among
the plantation occupants (Roswell King, Jr., 25 December 1812, 8 Novem
ber 1818). Although experiments with planting corn, barley, and pota
toes were attempted (Roswell King, Sr., 31 May 1813, 1 June 1818),
Butler Island lacked sufficient "high lands" or hammocks (see House 193S;
213, 1954a:153) for the production of provision crops. Consequently,
most provision crops were either produced at the other plantations of
the estate (Roswell King, Sr., 9 June 1814; Roswell King, Jr., 25 January
1819) or were purchased from other planters.
Plantation structures were numerous at Butler Island. In addi
tion to the outbuildings needed to process and store rice, cotton, and
sugar, a sawmill, cornmill, warehouses, and a slave hospital were among
the structures (Butler Estate Papers,1815a). The maintenance
of these buildings was an ongoing process, and both slave mechanics and

67
hired white craftsmen were regularly engaged in making improvements on
the structures.
Efforts were made toward making Butler Island self-sufficient in
the production of plantation supplies. Construction materials were
manufactured, including clay arid tabby bricks, shingles, nails, and iron
tools. Slave clothing and shoes were also frequently made. If trained
slaves were unavailable, local whites were hired to undertake these
tasks (Roswell King, Sr., 9 July 1805, 28 May 1815, 4 October 1915).
Sometimes whites were employed to train the slaves in specific skills
(Roswell King, Sr., 26 April 1816). The extent to which certain pro
ducts were made depended on the needs of the plantation and the cost of
manufacture at the time. This was clearly the case with regard to shoe-
making: "As for tanning and shoemaking, I recommend you give it up.
Hides have gotten extravagant. Our shoes cost $4.00 a pair, you can get
better shoes out of the state prison for $1.00" (Roswell King, Sr., 15
November 1818). Thus "self-sufficiency" at the Butler estate varied
through time and with the needs and interests of the owners.
Taken as a whole, the Butler estate was at most times very profit
able. The estimated value of the slaves, structures, and lands was
$288,027 in 1818 (Roswell King, Sr., 19 July 1819). By the 1850s, the
slaves alone were valued at $300,000 (Pierce Butler, Jr., Ledger,
1851-1855). Annual returns on the crops ranged from $20,000 to $50,000
except in years of losses (Butler Estate Papers, 1814: Butler Estate
Papers, 1824-1835). The estate, however, was not entirely a success
story. Storms damaged tremendous amounts of crops in the years
1804, 1813, and 1824. The hurricane of 1804 was particularly disastrous

6.8
to coastal Georgia (Ludlum 1963:116), and the Butler estate lost not
only crops but 19 slaves as well (Van Doren 1929:186). During the
British raids along the coast in the War of 1812, 139 slaves were taken
from the estate (Roswell King, Jr., 28 April 1315). In later years,
freshets and epidemics among slaves had far-reaching effects upon prof
its. Finally, the crash of 1857 resulted in the sale of 500 slaves.
It is possible that Major Butler found the losses of his day over
whelming. Historic records indicate that he put the entire estate up for
sale in 1818, a total of 15,000 acres and 535 Negroes (Darien Gazette,
1818-1819). The factors which motivated his actions are not clear.
Perhaps he was of the opinion that the estate was becoming unprofitable.
It is evident that Roswell, Sr., had begun urging the Major to sell por
tions of the estate in 1816: "I think you have too many eggs in one
basket to keep . the land is worn out, the droughts and gales make
our cotton the smallest I ever knew .... Your interest is too large.
. . Not enough laborers for the amount of land" (Roswell King, Sr.,
21 January 1816). In 1318, King's urging had become more persistent:
I have urged you to sell your estate for a long time out
of the purest of motives, thinking Negroes was above their
value and that was a time to sell, but I find I have
ered in judgement, Negroes are rising in value like the
tide, every day. ... I recommend you to sell all your
Negroes and land you hold in this state, say in a lump
for $555,000 and in my opinion in fifteen minutes after
you read this and if you will not sell I strongly recom
mend to you without loss of your time make arrangements to
lay out one hundred thousand in Negroes. Let them cost
what they mayas I have often observed your estate must
dwindle without more force. (Roswell King, Sr., 28 June
1 O 1 o \
i O i O j
Whether it was King's persuasion or other factors which resulted in the
decision to sell cannot be ascertained. The financial crashes between

69
1817 and 1819 may have been factors. During that time the entire south
ern economy suffered great financial losses (Eaton 1961:274-873). Per
sonal family relations are another possibility. Unable to convince
Thomas to take over the administration of his estate, the Major may have
decided to sell before considering the alternative of leaving his proper
ties to his grandsons. Whatever his reasons for selling, he never found
a buyer: "I have done all I can to sell your estate to promote your
interests and to close your accounts in Georgia. I fear I shall not be
able to do much for you" (Roswell King, Sr., 15 January 1819). Thus the
estate was retained and portions of it remained in the Butler family for
another 90 years.
Records of Slave Life
One of the most astonishing aspects of the Butler estate was the
size of the labor force. By 1859, a total of 919 slaves were included
in the estate (Butler Estate Papers, 1859). Even in Major Butler's
day, the total number of Butler slaves ranged from 500 to 700. Slave
holding of this magnitude was clearly in a class by itself. Slaveholders
who owned from 500 to 1,000 slaves represented less than 2 percent of
all slaveholding in the entire South. In Georgia, this "unusually large
slaveholding was confined to less than 1 percent of all planters in the
state" (Gray 1941:530-538). Thus, the size of the Butler estate made
it atypical among southern plantations.
How the Major managed to acquire such a vast number of slaves is
an interesting subject in itself. Most of the Major's initial transac
tions had taken place between the years 1786 and 1804 (George Hooper
1785-1802; Butler Estate Papers, 1801-1804). Although some of the

> 'i r r-- --^r ,tfj3 V.T'Saj'' i-'\ * W* J ~"r-' '^'SiJ £>-i \ --, w*,. ;> ''J&&--'' ',-=iV 2 '^J-V
70
Butler slaves were direct imports from Africa (Roswell King, Sr., 13
May 1803), most were acquired through the domestic slave trade based
in the older slaveholding regions, particularly the Virginia Tidewater.
The enterprising Major, of course, found ways to add to his slavehold-
ings illegally. Before the Major left for Georgia, he took a consider
able number of slaves belonging to his wife's family. Although the
total number of slaves acquired illegally for the Major's in-laws is
not certain, in 1817, 127 slaves were found to be the property of the
Major's sister-in-law (Roswell King, Sr., 29 January 1817). Apparently,
these slaves were never returned to their former owner. As planting
interests expanded, the estate was supplemented with additional pur
chases of slaves. Natural increase also added to the slave population.
Rarely did the death rate exceed the birth rate (see Table 3). The
fact that few slaves were sold* provided an additional factor which
helped to maintain the size of the slave population.
Several slave lists exist for the Butler estate, but few specifi
cally list the slaves at Butler Island. Tables 4 and 5 combine a
number of lists taken from various time periods (Butler Estate Papers,
1803, 1821; Roswell King, Jr., Daybook, 1844). Indications are
from the plantation records that this combined list reflects the
average number of slaves regularly engaged in the labor force at
Butler Island. The list, however, does not include children
under age 10, which usually averaged 100 in number (Roswell King, Sr.,
*A close examination of the slave records reveals that the occa
sions on which slaves were sold from the estate were very rare. This
is in keeping with the evidence concerning other rice plantations in
the area (see House 1939:215).

71
Table 3
Annual Births and Deaths of Butler Slaves
Year(s)
Births
Deaths
1.302-1812
257
191
1819
29
19
1820
25
16
1821
19
27
1822
19
22
1823
32
14
1824
30
18
1825
20
28
1826
35
23
1827
41
39
1828
46
25
1829
31
16
1830
41
25
1831
45
20
1832
31
29
1833
43
34
1834
53
34
Total
756
580
Note; i
Natural Increase 176.

72
An Approximate
Table 4
List of Slaves by Age and Task Rate at Butler
1803-1845: Field Hands
Island,
Name
Age
Task Rate3
Sambo (driver)
33
Pender
31
1
Robi n
25
1
Cloa
26
1
Eve
23
1
John
20
1
Lewi s
18
1/4
Robin
25
1
Cornelia
30
1
Joe
43
1
Hannah
44
1
Limas
23
1
Amy
18
1
Charlotte
15
1
Molly
15
1/4
July
46
1
i
Nancy
26
1
Renter
56
i
FI ora
53
1
Mary
32
1
Li stern
29
1
Molly
25
1
Billy
49
1
Quacer
35
1
Mary
35
1
Toby
34
1
Adam
35
1
Minder
35
1
Total
25 1/2
Jack (driver)
57
__ _
London
28
I
Dido
21
I
John
29
1
Beck
16
1/4
Harry
29
1
Nancy
18
1/2
J1 mmy
35
i
l
Toby
37
1
Nancy
52
1
Constant
16
1/4
Tinsh
14
1/4

73
Table 4~(continued)
Name
Age
Task Rate
a
Quash
36
Will
13
FI ora
32
Auber
18
Jeffrey
37
Dally
45
Dago
26
Jean
34
Hagar
48
Sarah
18
Lisbon
24
Flu
2.5
Tamfsor
22
Mari a
42
1
1/2
i
i
3/4
1
1
i
i
1
1
1
Total
Ley (driver)
30
Peggy
31
Sarah
19
Randal 1
45
Hector
24
Peggy
17
Sam
33
Libby
27
Provamy
23
Toby
19
Peter
16
Hagar
52
Le i ra
52
Lei pi a
26
Ccsmba
42
Quanri n
24
Hagar
17
Robi n
16
Frank
29
May
22
Abram
42
Cootnba
43
Billy
19
Lew
20
Lucy
29
Jimmy
47
Rose
20
1
1
1
1
i
1
1
1
3/4
1/4
1/2
1/4
1/2
1/2
3/4
1/4
1/4
1
1
1
1/2
3/4
3/4

74
jble 4(continued)
Name
Agt
Task Rate0
Clarinda
Violet
Fanny
Molly
Harry
Lilly
Charlotte
Temple
Total
Captain (driver)
Dorcas
Lagette
Jenny
Cate
Dick
Charity
Morris
Melinda
Peter
London
York
Judy
Affa
Lilia
Mi ra
Violet
Pompe
Oncy
Joe
Juba
Betty
George
Lender
Pena
Philo
Elce
Silva
15
20
17
37
44
28
23
52
57
25
21
19
29
35
37
43
40
1
16
25
29
42
42
17
12
37
33
30
22
19
20
25
25
24
1/4
1/4
1/2
1/2
1
]
25 1/2
1/2
3/4
1/2
1/4
1/4
1/2
3/4
1 o ta i
1/2
22 3/4

Table 4-(continued)
Name
Age
Task Ratea
Frank (driver)
42
Elce
23
1
A1bert
52
1
Lender
44
1
Brewster
33
1
Lookey
35
1
Molly
13
1/4
Jerry
16
1/4
Bruths
36
1
Jenny
34
1
K.egew
39
1
Jack
21
1
Tom
37
1
Amy
29
1
Hagar
22
i
Ti na
25
1
Mingo
17
1/4
Justice
32
3/4
John
33
1
Phoebe
32
1
Betty
35
i
Mary
41
3/4
Garrett
25
1
Ley
25
7
\
Rose
42
1
Patty
18
3/4
Esther
15
1/4
Cate
29
1
Cooper Nero
29
1
Mary
29
1
i
Total
--
26 3/4
Grand Total
122 1/2
Note. Total number = 156.
U1 = full time, 1/2 half time, etc.

76
Table 5
An
Approximate List of Slaves, by Age and Occupation,
at
Butler Island, 1803-1845: Plantation Specialists
Occupation
Name
Age
Blacksmiths
Jack
70
Sammy
50
John
30
Sawyers
Andrew
30
Sambo
55
Masons
Pompe
55
Gorham
78
Jacob
---
Carpenters
Sunday
56
John
50
Bram
5-
Harry
3-
Ned
3-
Primis

Frank
--
Sambo

George
--
Abraham
37
George
47
Randy
30
A1 ex
28
Issac
18
Shoemakers
Robi n
49
Bram
48
Ditchers
Abraham
30
Abraham Carter
30
Justice
26
Justice (younger)
21
Sam
22
Joe
25
Robin
40
Toby
22
Quash
25
Sancho
25
Peter
29
Tuco
20
Craftsmen
Ditchers
24
12

77
13 May 1827), or retired slaves involved in menial tasks, A census
taken of the Butler slaves in 1830 indicates that 365 slaves lived at
Butler Island at that time. The census provides the female-to-male ratio
and age distributions, as shown in Table 6.
Table 6
1830 Census of Slaves at Butler Island
Age
Male
Female
Tota'
Under 10
44
49
93
10-24
57
52
109
24-36
36
33
69
36-55
39
29
68
55-100
12
14
26
Total
183
177
365
Details obtained from plantation records regarding the material
conditions of the Butler slaves are similar to standard slave care prac
tices throughout the Old South. At Butler Island, these practices
mirror those established for other rice plantations (e.g. Allston 1858;
House 1954a; Phillips 1969;!14-130}. Generally, these practices include
the following; the provision of two sets of clothing annually, one for
winter and the other for summer; plantation rations, which consisted
primarily of corn, pork, salted fish, and molasses; a sick house and
medical attention furnished by the overseer or slave doctor; and hous
ing. Superficially, the records tend to suggest that the Butler slaves

78
were well provided for,as some recent historians have suggested to have
been the general nature of slavery in the United States (see Fogel and
Engeriran 1974); yet it is apparent from close inspection of these
records, particularly the overseers' letters, that inadequacies were
prevalent. Because the overseers' greatest concern was slave diseases,
most of the information relating to slave care is concern with health
care, and to a much lesser extent diet. References to clothing or
improvements to slave dwellings are few. Thus, it is difficult to assess
the material conditions of slavery at Butler Island on the basis of
plantation records alone. It is intended that the archaeological data
presented in Chapters 4 and 5 will amplify the descriptions of slave
material conditions found in the records. On the other hand, some as
pects of slave life mentioned in the records have no archaeological
correlates, including slave diseases, mortality rates, disciplinary
problems, and punishment. In the remainder of this chapter these topics
are briefly examined.
The Kings' greatest concern was with the slaves' health care.
Efforts therefore were made to place at Butler Island slaves who were
adapted to the swampy lowlands. Just how these adaptive capacities
were determined is not clear. Through mere observation such an assess
ment could have possibly been made. Roswell King. Sr., stated:
I landed the Negroes all safe and in good health. . .
I carried them to Tide Island [Butler Island] and you
would have been much gratified to see their rejoicing,
when they found they could drink the water out of the
river, which is proof that they have been acquainted
with the same soil & hopefully they will be healthy.
. . They landed very cheerful & happy, you have no
people that can talk with them. (13 May 1803)

7 9
The above statement supports the suggestion that planters deliber
ately selected slaves adapted to a malarial environment for labor in rice
fields. But it appears that even among members of the old, established
slave population,
their abilities to withstand the
rice swamp was an
ongoing concern:
It appears we have few Negroes here [Hampton] that are
suitable for the rice swamp. We have at least 20 at the
rice island which I would like to move if we had
suitable highland for them. When you purchase more of
course some of them will not be so profitable in a rice
swamp and it is necessary we should arrange for land more
suitable. (Roswell King, Sr., 22 August 1813)
Yet sickness was a prevailing problem at Butler Island despite
efforts to place slaves there that were adapted to a malarial environ
ment. An array of diseases plagued the island, including fevers of all
kinds, measles, intestinal disorders, cholera, influenza, whooping
cough, and others. Of these,cholera, influenza, and intestinal dis
orders were the most dreaded, and reports of these diseases frequently
followed the occurrence of freshets, particularly in the 1830s,when
freshets became annual events. At that time, slaves were often re
located to keep down debilitating diseases (Roswell King, Jr.,
15 February 1835, 8 March 1835). Evidently, freshets contaminated the
food and water supplies, which when consumed, caused illness and pos
sibly death. The drinking of contaminated water, however, was seem
ingly a health problem at other times as well: "This is the sickly
season for Negroes at the rice island. There is 14 to 15 in the hos
pital. None dangerously ill. I believe it is owning to their drinking
river water" (Roswell King, Sr., 30 April 1815). In general, in the
South
contaminated water was frequently the source for epidemics of

of cholera, dysentry, diarrhea, typhoid, hepatitis (Savitt 1978:59).
By far the greatest sufferers among the slave population at Butler
Island were infants and chi 1 dren. The infant mortality rate was higher
there than at any of the other plantations in the Butler estate:
You will perceive that the mortality among infants is
very high particularly at #6 this place [Butler Island]
the state of the atmosphere is certainly injurious to
infants. The proportion of young people at this place is
far behind St. Simons. (Roswell King, Jr., 5 January 1829)
Deaths among infants at the Butler estate often comprised 50 percent or
more of the annual deaths as indicated in Table 7. Infant mortality
rates such as these were typical of the late antebellum period (Steckel
1979:95-96). Several conditions generated high levels of slave infant
mortality, but recent studies have suggested that the predominant causes
of slave infant deaths were Sudden Infant Death Syndrome (SIDS), or crib
death (Sutch 1976:283-292; Savitt 1978:122-127), tetani-related
diseases (Savitt 1978:121-122), and occasionally infanticide (Savitt
1978:127-128). Although no definite cases of infanticide were reported
at Butler Island, carelessness or neglect on the part of mothers was
given as the cause of death for some infants (Roswell King, Sr.,
16 June 1816; Roswell King, Jr., 17 September 1826). Most deaths among
children between the ages of 2 and 10 years were believed to be due to
worm infestation (Roswell King, Or., 10 September 1835). Worms were a
general health problem among black children in the Old South (Savitt
1978:128).
The treatment of diseases was usually handled by the Kings. In
cases of major emergencies, a doctor was called upon to tend to the
The frequently used medications listed on annual expense lists
slaves.

¡able 7
Mortality Rates
of Infants and
Children of
Slaves on
the Butler Esta
te, 1819-1834
Number
Ini
rants
Chi1dv
~en (1-4)
Children
(5-14)
Total
Year
(Total deaths)
#
%
It
0/
io
#
%
Jf
ii
0/
10
1819
19
9
47
16
3
15
15
79
1820
16
7
43
2
13
0
0
9
56
1821
27
3
11
5
18.5
3
11
11
40.5
1822
22
3
14
7
32
3
14
13
60
1823
14
12
86
0
0
0
0
12
86
1 £24
18
9
50
2
11
0
0
11
61
1825
28
9
32
C
j
18
i
3
15
53
1826
23
15
65
i
4
2
9
13
78
1827
39
24
62
5
13
2
5
31
80
1328
25
2]
84
3
12
1
4
25
100
1829
16
8
50
3
19
2
12
13
81.5
1830
25
10
40
3
12
2
8
! 5
60
1831
n ^
au
15
75
r'
£.
10
1
5
18
90
1832
29
12
41
5
17
1
3
18
61
1833
34
21
62
r*
0
14
0 .
0
26
76
1834
34
17
50
5
15
1
3
23
68
Total 389
200
51
56
15
25
5
281
72
CO

82
included rhubarb, Spirits of Turpentine, red lavender, aloes, alum,
magnesia, camphor, peppermint, castor oil, opium, and others. These
medications were typical of the antebellum apothecary (Postell 1951:98-
99). Black folk medicine 'was discouraged, although some slaves sought
help from "negro doctors" despite objections (Roswell King, Sr.,
10 August 1815).
Improvements in slave diet were apparently the primary actions
taken to prevent illness. This was especially true for children. Meals
were sometimes prepared separately for them to insure that they were
properly fed:
There is no estate with so fine of crop of young Negroes
as this Butler Island. There are over 100 hominy eaters,
during the summer once or twice a week they are collected
and have an extra meal of fish or pork soup which make
each little negroe grow in the river swamp, I think.
(Roswel1 King, Jr., 13 May 1827)
From time to time improvements were also made on the diets of adult
slaves, such as increases in the quality or quantity of plantation
rations. Generally, the amount of plantation ration varied according
to the slave's occupation. Ditchers received the largest amounts of
meat and corn; next in line were the drivers and mechanics; then came
the field workers; and last were invalids and retired slaves who re
ceived the least amount of food (Butler Estate Papers, 1815b). Slaves
kept vegetable gardens and raised chickens at Butler Island, but these
were periodically destroyed by freshets (Roswell King, Jr., 23 July
1829). Hunting (Roswell King, Sr., 28 January 1829) and fishing also
supplemented the diet, which is indicated by archeological evidence as
wel 1.

83
Despite all indications of an adequate diet, slaves frequently
stole plantation rations stored in warehouses. Although most of these
foods were sold for cash, some of them were probably used to supplement
deficient diets. Stealing corn was particularly a problem: "Major
Hopkins was telling me one day your Negroes had more to sell than any
Negroes he ever known (I told him they were industrious) and he said
they supported Darien in corn & cheap to" (Roswell King, Sr., 22 Novem
ber 1806).
Stealing plantation produce, in fact, had. become so widespread in
the Altamaha estuary that planters attempted to prosecute local mer
chants in Darien for buying from slaves:
Tomorrow I am to go to a meeting in Darien to form a
proclamation against the traders there for dealing with
Negroes. The planters on St. Simons and in the vicinity
of Darien have formed a coalition to prosecute all dealers
with Negroes. (Roswell King, Sr., 23 September 1808)
Because incidences of stealing were frequently reported, stealing was
evidently the number one disciplinary problem among the Butler slaves.
Second to stealing, slave runaways appear to have been a problem.
It is curious that most runaways were intended to be short-term absences
from the plantation. Oftentimes, one or two slaves, usually male, would
go off to another plantation or sneak into Darien without permission
(Roswell King, Sr., 31 December 1808, 10 May 1818; Roswell King, Jr.,
28 January 1821). Few incidences of permanent runaways are indicated,
and it appears that these were always unsuccessful attempts (Roswell
King, Sr., 12 May 1804).
A variety of methods was used to punish slaves for stealing,
running away, and other deviant behavior. The lash was used, but

84
banishment to hard labor at "Experiment" (Roswell King, Sr., 6 March
1813) or "Five Pound" (Kemble 1961 (1861):270) was prescribed for chronic
offenders.* Punishment also entailed the denial of fish, meat, molasses,
clothes, rice, and forced labor on Sundays (Roswell King, Jr., 29
Fehurary 1829). Children were submerged in cold water as a means of
punishment. On one occasion a young boy died shortly after he was dumped
in water (Roswell King, Jr., 3 February 1829).
For the most part, overseers' letters provide very little informa
tion on the family or social life among the slaves. This is unfortunate,
since this is an area in which archaeological data supply little, if any,
information. A few scattered references do exist regarding family
domestic quarrels, conflicts between drivers and workers at the slave
villages, and the religious fervor among slaves. Perhaps oral history
is the best resource of information for investigating these aspects of
slave life (e.g. Blassingame 1972, 1976; Kill!on and Waller 1973:
Escott 1979). It has been pointed out that overseers tended to write
about issues they thought the planter should know about (Bassett 1S6S:
261). Consequently, overseers' letters generally reflect concerns re
lating to slave illness, death, and disciplinary problems. Given these
concerns, it should be obvious that gaps would exist in interpretations
of slavery derived solely from this resource. Although archaeology may
not be able to provide data relating to the social or religious life of
slaves, it can shed light on the specifics of slave material conditions.
*In the plantation records there is no reference to the planta-
tion"Five Pound" but there are a number of references to "Experiment."
Because they were both apparently located on Little St. Simons, the
author is of the opinion that these names refer to the same plantation.

Virtually little or no information is contained in plantation records
regarding the building materials used in slave housing, the size of
these dwellings, or the wild food resources utilized by slaves, and many
other details. These items of slave life are discussed in the remainder
of this Study.
Archaeological Resources
Archaeological investigations were undertaken at Butler Island
in the summer of 1978 under the direction of Dr. Lewis H. Larson, of
West Georgia College, and in the spring and summer of 1979 by the author.
The cyclical inundation practices of the Waterfowl Management Program
necessitated that all fieldwork be conducted in the spring and summer
months. This schedule, which involves the flooding of most of Butler
Island in the fall and gradually removing the water from late winter to
early spring, prohibited fieldwork before April. Even in April, many
areas were still very damp, and some were not completely dry for field-
work activities until June or July. Thus, the time spent at any one
site depended, to a great extent, upon the degree of moisture present at
the site during the time allocated for fieldwork. Fieldwork activities
included surface reconnaissance by foot and by boat, mapping of sites,
and limited excavations at slave settlement #1, the southernmost tidal
mill, and slave settlement #4 (see Figure 7).
Site Assessmen
t,T>
The sites designated in Figure 7 are presumably antebellum period
sites. These have been located and identified by use of a combination
of resources including historic maps, photographs, records, aerial
photographs, informants, and archaeological assessments. Recent land


87
modifications at Butler Island are shown in Figure 3. The most obvious
modifications are U.S. Highway 17, which roughly coincides with the
canal east of Settlement #1 and Interstate 95, which crosses the island
in the vicinity of settlement #2. Also, twentieth-century buildings,
from the Huston occupation, are present at settlement #1. These are
presently used, for waterfowl management activities.
It was possible to predict potential areas for site location by the
use of aerial photographs (of course with photographs having better
resolution than that: of Figure 8). Because Butler Island is generally
a low-lying, grass-covered marshland, sites tended to be located on
infrequent high spots, which are presently covered with trees and shrubs,
notably, hackberry, persimmon, and fig. These areas of tree groves
were easily identified in the aerial photographs. From surface recon
naissance, however, not all of these tree clumps were found to be sites.
But ail sites located were associated with this tree growth. Addi
tionally, some sites, such as the slave settlements, were also indicated
in aerial photographs by the detection of antebellum drainage ditches,
which surround and separate these from former rice fields.
The identification for several sites is well documented. The
slave settlements correspond exactly with the area designations I, II,
III, and IY on the 1877 USGS Map (or Figure 6). Historic photographs
provide suggestions for the placement of structures designated in slave
settlement #1. The steam-powered mill is precisely identified by ruins
of the two chimneys which are presently standing. A sketch drawn by
Roswell King pinpoints the location of the sawmill and both tidal mills
(Roswell King, Sr., 13 January 1813).

88
Figure 8.
Present-Day View of Butler
construction. 1-95 crosses
ment #2 and US-17 coincides
Island. (Note: Modern
approximate location of
with historic canal.)
hi ohway
settle-

The tidal mills were further identified by archaeological resources.
In 1977, wooden timbers, presumably parts of a waterwheel,were unearthed
following waterfowl management dredging activities. These timber frag
ments were found directly to the east (that is east of IKS. 17) of the
tidal mill designated in settlement #1. Evidently, these timber frag
ments were the only remains suggestive of a mill or other structure. In
other words, no other structural indications were evident. It was sug
gested by John Morgan, the archaeologist making the assessment at the
time, that these remains did not identify a structural site at that
location. He suggested, instead, that, the timber fragments may have been
deposited in that location by some other means, such as tidal action,
which could have caused loose mill fragments to be moved to the area;
activities associated with the construction of U.S. 17; or by an un
known agent.* The present author agrees with this interpretation. In
light of the historic evidence which indicates that a tidal mill was
located at settlement #1, Morgan's interpretation has been substantiated.
Evidently, these wooden timbers were part of that tidal mill structure.
The other tidal mill located on the southern edge of the island
was partially excavated. Evidently, the mill structure which housed
the milling machinery was built upon a rich deposit of oyster shell
associated with aboriginal ceramics and food bone remains. Tidal action
for operating the mill was supplied by a series of canals that were
controlled by floodgates and trunks. The water was stored in an impound
ing pond, and an undershot waterwheel was turned by the rise and fall
*John Morgan, Jr., Staff Archaeologist, Historic Preservation
Section, Georgia Department of National Resources, Atlanta, Georgia,
personal communication, 1978.

of the tide (Doer 1970:23). These canals and the impounding pond were
detected in aerial photographs. Subsurface testing at the site un
covered the structural underpinnings of the mill, the bed of the water-
wheel, and a millstone.
Previous archaeological assessments made in 1956 were used to
locate the slave cemetery (Caldwell nd; Sanders nd). Again, through
waterfowl management dredging activities, a number of coffins were
unearthed in the area indicated as the slave cemetery (see Figure 7).
At that time, a dragline cut through the south edge of the cemetery
smashing several wooden coffins and exposing others in the profile north
of the dragline. It was noted that the coffins were made of cypress and
that the skeletal material was poorly preserved. Most of the coffins
were evidently those of infants, a finding which supports the historical
evidence of a high infant mortality rate at Butler Island. On the basis
of information obtained from local informants, Sanders (nd) identified
the cemetery as a slave burial ground. These informants claimed to have
had ancestors buried there. The location of a cemetery could not be
found in the plantation records, but both Fanny Kemble (1961 (1863):141)
and Francis Leigh (1883:72) mention a slave burial ground. Unfor-
tunately, neither identified its location. Given the number of infant
burials, and the fact that a slave cemetery or two was located at
Butler Island, it was decided that this site once served as a slave
graveyard.
The sites
of "unknown function" could not be
identified on the
basis of visible archaeological remains occurring on the surface,
gestions of possible function, however, can be offered on the bas
Sug-
s of

91
historic resources. The site east of settlement #1 is purported to be
a slave cemetery. But surface evidence of a burial ground in the form
of headboards, tombstones, or distinctive Afro-American burial patterns
(Combes 1974) is lacking. Because of the dense tree cover, it is pos
sible that subtle evidence of a graveyard was overlooked. Attempts to
locate coffins by probing with a metal rod proved to be unsuccessful.
Although no superficial evidence indicated that this area was a grave
yard or, for that matter, a site, its size and location strongly suggest
that it was a site of some type. Because of the lack of time and
resources, its function could not be ascertained. Additional historical
research and fieldwork are needed to determine its function.
The site located between slave settlements #3 and #4 appears to
have been a mill or other industrial site. No structural features are
evident, but a nearby impounding pond and the site's layout appear to
be similar to that of the excavated tidal mill. More specifically,
historic evidence suggests that this was the site of a sugar mill.
Roswell King placed the "sugar works" at Butler Island halfway between
slave settlements #3 and #4 (Roswell King, Sr., 20 April 1816). This is
the only location in that area indicative of a site. Only through
future excavations can this suggestion be confirmed.
Like the previously described site, it is believed that the site
in the northwest corner of Butler Island is also a mill or an industrial
site. The evidence pointing to this assessment is more conclusive than
for the other sites of unknown function. Again, an impounding pond
similar to that of the tidal mill is present at this site. The recovery
of wooden fragments identical to the excavated waterwheel further

supports this interpretation. Additionally, the base of a brick wall
along the drainage ditch indicated that a rather large structure was
once present on the premises, and a scatter of bricks, presumably from
a chimney fall was also found. A padlock lying on the surface has been
found to be identical to ones uncovered at the slave site. It is be
lieved that the padlock and brick type (Savannah greys) suggest that
this component of the site is temporally antebellum. Unfortunately, its
antebellum function could not he determined. On the other hand, the
sites's postbellum function is well known by local informants. In the
twentieth century, the site was utilized for the manufacture of illegal
moonshine whiskey until it was discovered by the authorities in the
early 1950s. Archaeological evidence of this moonshine operation is
still apparent in the form surface debris, including still parts, corn
liquor crockery fragments, and Mason jars. This site also has a pre
history c component.
Prehistoric components were found to be associated with the
southern tidal mill, settlement #3, and the presumed mill site in the
northwest corner of the island. These prehistoric occupations are
identified by oyster shell and aboriginal ceramics ranging in time from
fiber tempered to complicated stamped. From the very limited samples
collected, most of these ceramics appear to be cord marked and checked
stamped. It is, however, premature to suggest a cultural association
from such slim evidence. A problem is presented by these prehistoric
deposits. It is not at. all certain whether the oyster shell was brought
in at a later time to make "high spots" in the swamp or was placed in
its present locations by prehistoric peoples. At the tidal ¡¡¡ill site,

93
the former interpretation does seem plausible, but historic evidence
suggests otherwise. Prior to the construction of this tidal mill, King
describes the "oyster bank" and suggests placing a mill at that loca
tion (Roswell King, Sr.s 13 January 1813). In a later letter, King
describes the depth of the oyster shell with astonishment: "The oyster
shell are 6 feet underground, and the water runs in like a spring"
(Roswell King, Sr., 6 March 1813). The two letters seem to suggest that
the oyster bank was there long before King or Butler thought of building
a mill at that location. At least, it does appear that King was not
responsible for bringing the oyster shell to Butler Island. It is
possible that King's predecessor was responsible for this action.
These aboriginal shell dumps have been greatly disturbed by post-
depositional construction and the robbing of shell, presumably for the
making of tabby. Therefore, it is doubtful that the present size and
shape of these shell deposits can help to determine whether or not they
were made by prehistoric occupants of Butler Island. Unfortunately, the
prehistory of the Alternaba delta is virtually unknown. Consequently,
these shell deposits may be prehistoric sites and not aboriginal shell
brought in at a later time.
Slave Village Sites
The investigators were most interested in the slave settlements,
and an effort was made to determine the archaeological integrity of these.
Investigations began at settlement #1 for several reasons. First, the
site's location, just off U.S. 17, makes it more accessible to the
vehicle than the other slave sites. Second, unlike the other slave
sites, it is not subjected to management flooding in the fall and winter

94
(it only becomes inundated during naturally occurring floods), thereby
making the preservation of organic materials better than that of the
other sites.* Third, the approximate locations of certain structures
at settlement #1 are 'well documented in historic photographs. However,
with the exception of the steam mill ruins, no surface indications of
these structures presently exist.
The areas in and around settlement #1 have been subject to con
siderable plowing and dredging. Consequently, a great deal of cultural
debris was found to be lying on the ground surface. The first objective
for fieldwork activities at settlement #1 was to walk the area, collect
the surface materials, and note the location of these scatters. It was
hoped that these collections would help to identify structural features
or other specialized activity areas. These collected materials ranged
from the mid-nineteenth century to the present. Unfortunately, they did
not provide indications for the location of features.
The next objective was to locate the two rows of slave dwellings
known to have existed at the site. It was decided to make a contour map
to help identify these slave dwellings. Presumably, a contour map would
indicate by subtle rises the location of structures. A very rudimentary
map was prepared, and it provided some indications for the approximate
location of the two rows. Two test trenches (1.5 x 4 meters), de
signated as Test 1 and Test 2, were laid out to test one of these subtle
rises (see Figure 9). This particular area was chosen because a con
siderable amount of cultural debris was present on the ground surface,
notably construction materials.
*See discussion regarding the preservation of archaeological
materials at Butler Island in Chapter 5.

le
st
o
Toe* !
i ca i
-s* North
3IS
zizi
20
meter.
nqure 9,
L X l
tenc ot
:avc
ion:
iti]ement
#1
cn

96
The units were excavated in arbitrary levels of 15 centimeters.
Excavated materials were dry-screened by use of a 1/4 inch mesh hardware
cloth. All beginning levels were screened because it was not certain
where the antebellum component began. It was during this first attempt
at excavation that the archaeologists became painfully aware of the
difficulty of removing, by shovel, and screening this wet, clayey soil.
Screening was found to be a very slow arid time-consuming process, but
alternatives to screening were unacceptable. The Noel-Hume method of
excavation, in which every artifact is recovered by trowel (1974:104),
would have been just as time consuming, and later troweling activities
attested to this presumption. To have abandoned screening altogether
would have resulted in the recovery of few small items characteristic of
slave material culture such as fragments of pipe bowls and stems, beads,
and buttons. In light of the alternatives, screening was not cur
tailed. But the commitment made to screening sacrificed the areal
extent of the subsurface testing. This was found to be the case at
other sites as well.
The excavated materials indicated that the site is very much
disturbed. No j_n situ structural features were uncovered although most
of the materials were in fact clay bricks. Antebellum artifacts
dominated the recovered materials but were mixed with modern debris
throughout the first two strata. Below these layers, few cultural
materials were recovered. At the base of the excavation, evidence of
the old cypress root system was unearthed and the watertable was en
countered (see Figure 10).

fVP
i'C jw,
TAKE
Tfl?rErcvSdw{thShs and Brick/^ortgLiMfel^-
STAXE
Figure 10. Stratigraphic Profile of Settlement #1.

98
On the basis of the artifacts recovered from other slave sites
in coastal Georgia, the antebellum materials at settlement #1 are sug
gestive of the slavery status. The small sample of artifacts recovered
and their disturbed nature, however, prohibits the use of these
materials in most of the interpretations offered in this study. Per
haps the greatest contribution of the archaeological resources un
earthed at settlement #1 is to provide evidence of the earliest his
toric occupation at Butler Island. A few eighteenth-century artifacts
were recovered, including fragments of slipware (Noel-Hume 1976:98-108)
(see Figure 11) and an eighteenth-century Spanish reale (see Figure 12).
The precise date of the coin is unclear, but the first two digits, 1 and
7, suggest that the coin was manufactured in the 1700s. These arti
facts provide the only vestiges of a pre-nineteenth century occupation
at Butler Island. Further excavations at settlement #1 are recom
mended to uncover additional evidence of this occupation.
Slave sites #2 and #3 were not tested. From field inspection
and aerial photographs, it was decided that slave site #2 was com
pletely destroyed by the construction of Interstate 95. In the vicin
ity where 1-95 crosses Butler Island, a tremendous quantity of sand was
deposited after dredging activities associated with the highway's con
struction. On the surface of this sand, antebellum materials were
found, including pipe bowls, shell-edged pearlware, and Savannah grey
bricks. The presence of these artifacts to this specific area supplied
the only archaeological evidence of slave settlement #2. Slave settle
ment #3 was surveyed very briefly. Two chimney falls (identical to the
ones discussed at slave settlement #4) were located. Unfortunately,

Figure 11,
Eighteenth-Century Slipware Fragments
from Settlement #1.

100

Figure 12. Spanish reaie, "Mexico Mint,
from Settlement #4.


the funds available were insufficient for the time needed to map and
test these sites.
Ultimately, the investigations of slave villages at Butler Island
centered upon slave settlement #4. This site was evidently the best
preserved of the four settlements. Visually apparent above the ground
were five chimney falls. The clearing of the chimneys revealed that
these were H-shaped hearths, presumably of duplex slave dwellings with
central fireplaces. A contour map of this site was made in the summer
of 1S78 (see Figure 13), and it was decided at that time to return the
following spring and summer to investigate the site further.
The field strategy devised for settlement #4 during the 1979 field
season was twofold: first, two or more of the slave dwellings would be
tested for structural information and, second, a subsurface testing pro
cedure would be developed to locate other activity areas hidden beneath
the ground surface. It was hoped that outbuildings, concentrated refuse
dumps, wells, and privies could be identified from subsurface testing.
The testing scheme developed utilized a systematic random sampling pro
cedure. At 10 meter intervals, a row of posthole tests was made ev
ery 5 meters. In the beginning, a hand posthole digger was used, but it
was soon replaced with a pointed shovel, the latter being better suited
for the clayey soil. Although the shovel did not yield tests of a con
sistent size, most tests ranged from 25 to 33 centimeters. A total of
222 tests were made. These tests, however, did not provide useful in
formation for delineating areas of refuse dumps, wells, or privies.
Artifacts recovered from the tests were concentrated in and around slave
dwellings. The exception to this rule was the uncovering of structures

104
mentis
Figure 13. Contour Map of Settlement #4. (Note five chimney falls
in lower left.)

105
six and seven to the right of the slave dwellings (see Figure 14). Evi
dence of these structures was first identified from posthole tests.
Later, through probing and small test pits, the approximate dimensions of
the structures were determined. Structure 6 was found to be a brick floor
and structure seven, the foundation of a large structure, possibly a rice
barn.
Excavation of the slave dwellings proceeded with the establishment
of excavation units from an arbitrary point designated N10QE100. Ini
tially, the soil was removed in arbitrary levels of 15 centimeters, but
later this was changed to the removal of natural strata. For the most
part, three strata were recognized at slave settlement #4. These include
grey silty loam, grey clayey loam, and grey clay with red and yellow
flecks (see Figure 15). These soil classifications were determined by
soil analyses undertaken after fieldwork activities (see Appendix 1
for soil descriptions). All cultural materials were deposited within
the first two strata, but the clayey loam yielded a higher concentra
tion of artifacts than the silty loam. Evidently, the clayey loam
represents the old humus zone of the antebellum period. With the
exception of intrusive features such as postholes, no cultural materials
were found in the clay. All excavated materials were water-screened
through 1/4 inch mesh hardware cloth. From the experience at
settlement #1, it was felt that water-screening would maximize recovery
as well as aid in the removal of the sticky clay and silt from the
artifacts.
The vast majority of
antebellum date. In fact,
the artifacts from settlement #4 were of
only one artifact was found to be postbellum.


N
r-
c.
Zone A
Zone B
Zone C
n
EY
SILTY LOAM
p p V I
'Ll ii % L I O La.
AYEY LOAM
l METER
CLAY WITH
:llow flecks
Figure 15. Stratigraphic Profile, Settlement #4.
O

'¡08
This was a modern-day shotgun shell, presumably the artifact of recent
duck hunting in the area. Unlike in settlement #1, no eighteenth-
century materials were recovered. This may be due to the fact that
settlement #4 was not constructed until 1803 (Roswell King, Sr., 22
October 1803). Thus, the date 1803 established a terminus post quern
for the site. The site was evidently not intensively occupied after the
Civil War, since postbellum artifacts were virtually absent. All indica
tions have suggested that the site is temporally antebellum, thereby
making it a slavery site. Three of the five slave dwellings were tested.
These structures and the excavated materials are discussed in Chapters
4 and 5.

CHAPTER 4
SETTLEMENT PATTERNS OF COASTAL SLAVE COMMUNITIES
The Determinants of Plantation Sett!ement Patterns
The determinants of settlement patterns have been defined as those
"classes of factors that interact with each other to produce the spatial
organizations of a social group" (Trigger 1968:53). These determinants
can be conceived of in terms of three levels of settlement: first, the
individual structure; second, the arrangements of structures within a.
community; and third, the manner in which communities are distributed
over the landscape (Trigger 1968:55).
At slave villages, what were the factors that determined spatial
arrangements? In order to answer this question, it may be necessary to
examine the plantation settlement as a whole. Certainly, topographic,
economic, and social factors influenced, if they did not determine, the
spatial arrangements of most plantations. Precisely what these factors
were at coastal plantations and the ways these were manifested at Butler
Island are briefly explored in this chapter. To date, very little has
been written on the subject of plantation settlement pattern. Archaeo
logical studies are especially few in number (see Otto 1975:18-28; Lewis
1979). The attempt is made in this discussion to synthesize from
a v a Hable settle me n t
lished works in anth
pattern data obtained from plantation plats; pub-
ropology, history, and geography; and the archaeo
logical resources uncovered
Sutler Island.
109

no
Generally, the spatial aspects of plantation communities were
similar throughout the South. At the 11 administrative nucleus" were
located the planter's or manager's house, service buildings, and the
dwellings of house or skilled servants. Processing equipment and
storage houses formed the "technical nucleus" (Wolf 1959:137). Ap
parently, convenience provided the key for the location of technical
equipment (Smith 1936:22) since it was placed at the administrative
center, or located separately, or associated with slave quarters
(Prunty 1955; Otto 1975:18-19). On small plantations, slave dwellings
were located near the plantation house, but large plantations tended
to have one or more slave villages located a considerable distance
away (Stampp 1956:272).
Reasons for the division of slaves into separate villages are not
clear. Polderland construction at rice plantations may have necessi
tated this pattern. The placement of structures or the centers of plan
tation activities was often dependent upon the availability of high
lands at rice plantations. Structures were located either on hammocks
adjacent to rice fields (Smith 1936:22), or high areas had to be
created. But the advantages of a close proximity to the fields and the
danger associated with a large assemblage of black people (Scott 1973:
75) better explain the prevalence of the segmented slave village pattern
in other cash crop regions. The Butler family plantation records pro
vide further suggestions for the establishment of this pattern. Slave
villages were efficient' means to keep account of a large number of
ss aves.
Slaves were listed and the amounts of crops planted or har
vested were often recorded by slave villages. It is also possibl<

Ill
that provisions were accounted for and distributed by these settlements.
Despite the advantages of segregating a large number of slaves into
smaller villages, some planters of large productive units evidently
preferred to keep all their slaves in the same location. This was par-
ticularly true of interior plantations (e.g. Smith 1973:144, plat of
Glendownwer estate). Exceptions such as these were also to be found in
the Georgia tidewater. For example, a plat of the Hopeton rice planta
tion (see Johnson 1930:63; Prunty 1955:464) reveals that all "500 slaves"
(Lane 1973:196) were evidently located in quarters near the administra
tive nucleus. Here again, the land use practices associated with
Hopeton's unique polder!and construction may account for its arrangement
rather than a preference to have slaves nearby.
Within the slaves' quarters, dwellings were typically arranged in
rows along roads or streets forming a square or rectangle of buildings
(Prunty 1955:465). This pattern was established for tidewater planta
tions and later replicated in other areas of the South (Stampp 1956:
292). The basis for this arrangement appears to have been clearly func
tional as it permitted easy cleaning and inspection of slave dwellings
(Flanders 1533:153; Otto 1975:24) and perhaps easy access to them by
wagons or carts.
But plantation layouts were not just functional. These spatial
arrangements expressed the social structure of plantation society,
which reflects a chain of command of owners, managers, overseers,
permanent laborers, and seasonal workers (Wolf 1959:137). The large
plantation plan (which has been under discussion here) with the big
house, numerous outbuildings, offices, and slave villages was a standard

112
set by the superordinate planter class (Corbett 1941:14). Although
regional differences were evident in architectural styles and construc
tion materials, plantation arrangements were characterized by broad
similarities (Otto 1975:24). It has been suggested that the presence of
these similarities expressed the social ideals of the planter elite
(Corbett 1941:13-14).
Yet, to a lesser extent, plantation settlement patterns may have
expressed ideals other than those of a generalized planter elite. In
some cases, planter ethnicity or for that matter slaves' ethnicities may
have influenced settlement patterns. It is a well acknowledged fact
that southern antebellum architecture represents a blend of ethnic in
fluences (Kniffen 1965; Glassie 1968). Recently, the Afro-American
contribution to this architectural style has been investigated (Fine
1973:11-26; Anthony 1976; Viach 1978:122-138). Ethnic influences are
also evident in plantation layouts. Many of the early colonial planta
tions were intended to resemble medieval manors (e.g. Phillips 1929:49,
photograph of Mulberry Castle). Some postcolonial plantations also
exhibit the owner's ethnicity. For example, the spatial arrangement of
the big house and outbuildings at the Choclat plantation at Sapelo Island,
Georgia, is similar to a French Chatelet.* Choclat, however, was
atypical among Georgia tidewater plantations. It was not engaged in the
^Morgan R. Crook, West Georgia College, Carrollton, Georgia,
personal coiniunication, 1980. The Choclat ruins were mapped in 1974 by
Crook, who later discovered that the arrangement of these buildings was
similar to a French Chatelet. The present author pursued this sugges
tion further and found that Choclat1s layout was not only characteristic
of a French Chatelet, but that it was similar to contemporary French
settlement plans, circa 1786-1800 in general (see Mauban 1945:plate 18).

113
full-time production of cash crops for a world economy. Instead, it was
a retirement home and subsistence-based farm for its owners, which
utilized numerous slaves in the work force (Lovell 1932:110-111). Per
haps, this difference in economic function may account for Choclat's
distinctive spatial arrangement.
The Afro-American influence in plantation settlement patterns is
more difficult to document than that of the Euro-American planter. Al
though the evidence of African-styled architecture has been observed in
a variety of plantation contexts, notably in South Carolina, Virginia,
and Louisiana (Anthony 1976), the evidence of a similar influence in
plantation layouts is slim and remains untested. The possibility of
this influence, however, is suggested in the community layouts of slave
quarters at two antebellum plantations on the Atlantic Seaboard. The
arc-shaped arrangement of slave dwellings at the Kingsley plantation
(Fairbanks 1974) and a horseshoe-shaped arrangement recently uncovered
in coastal South Carolina* are suggestive of traditional African village
layouts found in several areas of Africa (Denyer 1978:20). At this
stage of research, however, such a suggestion is speculative and awaits
further investigation.
Because plantation settlement patterns have not been systemati
cally investigated, it is difficult to
determined these spatial organizations,
were designed to be functional to meet
enumerate the factors which
Undoubtedly, most arrangements
the environmental challenge and
economic needs.
It Is likely that cultural ideals,
ethnicity, and
*Patrick Garrow, Chief Archaeologist, Soil Systems Incorporated,
Marietta, Georgia, personal communication, 1379.

114
other unexplored factors also had an impact on these patterns. The
extent that these additional factors influenced plantation arrangements
remains unanswered.
Slave Community Organizations at Butler Island
The hypotheses outlined in Chapter 1 assume that plantation
spatial organizations were primarily functional and that these reflect
adaptations to specific crop requirements. Butler Island's settlement
patterns conform to those of most tidewater rice plantations, specifi
cally those lacking sufficient highlands. Plantation structures were
evidently built upon high areas that were either naturally occurring or
artifacts of polderland construction. In the placement of slave vil
lages, the latter appears to have been the case. The slave villages were
strategically located at the southern end of the island, where they were
accessible overland by the main dikes and by water along the main canals
and the river (see Figures 7 and 8).
Settlement #1 was the administrative nucleus for the plantation
and it included the overseer's house, blacksmithing and shoemaking
sheds, warehouses, and a meat-curing house. Also,the largest technical
nucleus for the island, a tidal mill and steam mill, and the greatest
number of slave dwellings were located there. Photographs of structures
once located at settlement #1 (Cate 1925; Figures 16, 17, 18, 19, and 20}
provide some indications of the former spatial arrangements at this site.
The technical equipment, was east of the overseer's house, the slave quar
ters to the south, and the other administrative buildings are believed to
have been to the west (see Scott 1961:57). For the Butler Island slaves,
settlement 41 must also have served as headquarters.
I he slave

115
Figure 17.
Tidal Mill and Boat Landing

116
ITW ft
?* > ; V-- 1 " ,; ?
- : mk
. -
Figure 19. Steam Mill Ruins.
Figure 20. Linear Arrangement of Slave Dwellings.

117
hospital, which occasionally served as a chapel and recreational center,
was located there (Kemble 1961 (1863):131).
Each of the remaining slave villages contained technical equipment
for the processing and storage of plantation staples, in addition to
slave dwellings. As indicated from the archaeological evidence from
settlement #4, these structures were located opposite the slave dwellings
forming a small technical nucleus. Settlement #2 contained long-staple
cotton processing equipment, and the slaves living there were possibly
engaged only in the production of cotton. Both settlements #3 and #4
contained rice storage facilities and were equidistant from sugar pro
cessing machinery (Butler Estate Papers, 1815a.; Roswell King, Sr.,
20 April 1816), thereby making these units equipped for the produc
tion of rice and sugar. Additionally, each settlement had its own plan
tation specialists who produced items to be used for planting activities
as well as for household needs.
To a certain extent, the slave villages at Butler Island repre
sent smaller self-contained productive units of a larger plantation com
plex. Frequently large areas of ricelands of 1,000 to 1,200 acres were
divided up into smaller units of 300 to 400 acres. These smaller units
operated as separate plantations (House 1954a:150). Apparently, planters
found these smaller units to be more efficient than larger ones. Thus,
the subdivision of slaves into villages on ricelands may have developed
from several factors, including polderland construction, economic effi
ciency, and social control. The impact these self-contained villages
had upon slave lifeways needs to be investigated. At Butler Island
these villages tended to isolate the slaves from the other centers of the

118
plantation complex. The only time a slave needed to go to the planta
tion headquarters at settlement #1 was to see the overseer, head driver,
to go to the hospital, or for an occasional church or other meeting.
Although slaves at one village probably intermingled with slaves at the
other villages, most of their time was spent in the rice fields adjacent
to their settlement or at their own quarters. From the planter's per
spective, this isolation maintained economic efficiency and social con
trol. But for the slaves, it may have substantially reinforced the
very limited contact prevailing between blacks and whites as well as
between acculturated blacks and those that were
not so
acculturated.
Within these
isolated villages slaves had the opportunity to retain
African folkways. Perhaps this explains why the rice coast has pre
served more evidence of African retentions than any other former ante
bellum crop region within the United States (Herskovits (1958:120). The
addition of African imports to the rice coast long after the ban on the
Atlantic Slave Trade (see Wells 1967) may have further encouraged this
cultural development.
The Community Plan at Slave Settlement #4
Although very limited testing has been undertaken at settlement
#4, some general statements can be made regarding the site's internal
spatial arrangements. While it may be possible that the periodic in
undation of the site has resulted in the lateral displacement of archaeo
logical materials, evidence of this disruptive action is lacking. If
the materials were displaced, the horizontal distribution of the
archaeological materials would appear to be random, that is, having no

patterned relationship (Garrison in Schiffer and Gummerinan 1977:154).
Fortunately, the evidence from both the posthole testing and the mapped
artifacts in and around the slave dwellings strongly indicated a non-
random, patterned distribution. Thus, it is doubtful that inundation has
adversely affected the distributional patterns of cultural features at the
present time.
According to the 1815 inventory, eight dwellings were located at
this site. Assuming that no additional dwellings were added after that
date, slave structures were possibly arranged in two rows, three in each,
and with two structures at the end of the two rows (like structure three;
see Figure 14). It is suggested that the dwellings were arranged in
this fashion because of the considerable construction debris, presumably
from slave dwellings,found in the canal directly below the site. The
canal is an artifact of Waterfowl Management activity, and it appears
that when it was constructed the archaeological remains for several struc
tures at the slave site were destroyed.
On a main dike, adjacent to the canal and formed by the dredging
for it, artifacts characteristic of slave material culture were found
to be lying on the surface. Both materials recovered from the canal and
the dike suggest that the additional slave dwellings once located at
this slave site were in the location of the present canal, in line with
the mapped structures located at the site.
The two very large structures, six and seven, were apparently part
of the technical nucleus. All indications suggest that structure seven
was a rice barn. On the 1815 inventory, a rice barn with the dimen
sions 60 by 44 feet is listed. The dimensions of structure seven were
found to be 18.4 by 13.4 meters, which closely matches these dimensions.

The only remains uncovered of the rice barn were its cypress foundations.
No artifacts were found in association with it.
The function of structure six, the brick floor, could not be
determined. Unfortunately, the 1315 inventory does not list any other
buildings at settlement #4 that would have been used for rice production.
Therefore, it must have been added after 1815. It is possible that it
may have been used as a threshing floor, since evidence of a super
structure was not identified with the floor. But no indications of brick
threshing floors have been found in the plantation records. All refer
ences indicate the use of tar threshing floors. Moreover, a soil analy
sis taken from below the floor suggests that the brick floor was super
imposed upon an earthen floor. From small test pits made to deter
mine the dimensions of the floor, numerous soil concretions resembling
hardpan in texture were recovered. The analysis suggested that these
concretions may have been part of an old earthen floor, and what
appeared to have been straw remains were found in these concretions..
This interpretation is plausible, however, only if the distribution of
these concretions was restricted to a limited area (see Appendix 1).
Because this soil type was limited to the area around structure six,
it is highly probably that these materials represent the former presence
of an earthen floor. In rice processing, earthen floors were often
associated with winnowing houses, but no archaeological evidence of
winnowing posts was identified. Perhaps, the floor was part of a shed
or warehouse. The possibilities regarding the function of the floor-
are numerous. But the size of the structure and its relation to the
rice barn strongly suggest that it was a specialized structure,

121
presumably used in the production of rice, rather than a domestic dwell
ing.
No surely identifiable nonstructural features, such as wells or
privies, were uncovered. As for wells, the historical evidence indi
cates that the Butler Island slaves drank water from the river (Roswell
King, Sr., 30 April 1815; Kemble 1961 (1853):189). Looking at the
Altamaha River today, it is hard to imagine that the river was ever
clean enough for drinking or bathing purposes. Nonetheless, if the
river regularly supplied water for drinking, then wells would not be
needed, and this archaeological feature would not be present. The ab
sence of privies is more easily explained than that of wells. On many
plantations slaves were not provided with privies. In fact, privies were
not used extensively in many areas of the South until the twentieth
century (Savitt 1978:60). Evidence of a privy is suggested from the
drainage ditch next to structure four (see Figure 14). From a trench
designed to test the drainage ditch adjacent to structure four (see
Figure 21), several vertical posts were uncovered. These posts appeared
to have been part of a small structure, possibly a privy or washstand.
In addition, soil samples taken from the drainage ditch had an extremely
high phosphorus content when compared with samples from other areas
of the site. Although the high phosphoric level suggests that this
area was a privy, the test for human bacteria was negative (see
Appendix 1). Therefore, positive evidence of a well or privy at
settlement #4 is lacking.
Whatever the function of the small structure, wooden fragments
recovered from the ditch and the soil analysis indicate that the

A22

123
structure was painted with red oil-base paint. Red ochre was frequently
listed on the annual expense and supply lists. Evidently, the ochre was
used to paint some of the plantation structures. The reasons for paint
ing the small structure red, or for that matter painting it at all, is
unknown.
If the ditch did not contain human waste, another explanation for
the high phosphoric content of the soil is suggested by the rich deposit
of organic refuse. Apparently, the ditch was used intensively for the
deposit of refuse, particularly organic refuse. Peach pits, shoe
leather, and charred bone, as well as inorganic materials, were recovered
below the mottled silt (evidently fill) and continued below the water
table (see Figure 22). Unfortunately, excavations of the ditch had to
be terminated since the archaeologists lacked an appropriate water pump
ing device to continue excavations below the water table. Clearly, this
portion of the drainage ditch was used for refuse disposal.
No other concentrated refuse dumps were uncovered. Small post-
hole tests placed along other locations of the drainage ditch did not
yield any artifacts. This suggests that the refuse from the ditch next
to structure four is most likely from that household. Another possible
location for concentrated refuse disposal was the river. The site's
close proximity to the river suggests that trash was likely to have been
regularly discarded in it. All other refuse recovered at the site was
concentrated in and around the slave dwellings. No evidence of trash
pits or patterns of refuse disposal could be delineated (see South 1977:
47-80). Both inorganic and organic refuse were found together in
similar proportions both inside and outside the house. Also, it is

Zone
Zone
Zone
Zone
N 120
WATERTABLE
(=:=! MOTTLED SREY/BROWN/YELLOW
SILT
T| GREY SILTY LOAM
I METER
c MM GREY CLAYEY LOAM
0U77\ GREY CLAY WITH RED FLECKS
Figure 22. Stratigraphic Profile of Drainage Ditch.

125
possible that the pigs and chickens kept by the slaves consumed a great
deal of the organic refuse. At any rate, no intentional patterns of
trash disposal were identified.
The large areas of the site in which no cultural materials were
uncovered were possibly used for gardening. The stratigraphy of the
posthole tests taken from the area above the structures (see Figure 14)
suggests that the ground has been turned by a pi gw more intensively than
in areas near the dwellings. This overturned soil may be the artifact
of postoccupational plowing which is evident from visible plow furrows
all over the site. The plowing activity appears to have been minimal as
indicated by little scattering of the artifacts.
The spatial arrangements at slave settlement #4 provide some indi
cations of an accommodation to rice culture. The overall layout of the
site is typical of tidewater patterns in general, and these patterns
have been observed at rice plantations (Olmsted 1968 (1856):416-417, 421-
422). On the other hand, the absence of wells, the specialized struc
tures associated with rice production, and the use of drainage ditches
for refuse disposal and possibly for human elimination appear to have
been practices developed specifically for slave life in a deltaic, marsh
land habitat. Limited testing, however, cannot be ruled out as a factor
which may have skewed the findings used for the interpretations offered
here. Additional excavations are needed to demonstrate that these
archaeological resources of community organizations reflect slave life
t
on the Georgia rice coast.

126
Slave Dwel 1 J ngs
Unfortunately none of the slave dwellings at settlement #4 were
completely excavated. Testing at structure one and the historic photo
graphs of slave houses at settlement #1 (Figures 18 and 20) provide
evidence of architectural style and some specific construction details.
All of the slave structures at Butler Island were evidently duplex
dwellings with central chimneys. This "two pen, saddle-bag" house type
(Kniffen 1965:556) was commonly used for slave housing throughout the
South. Single-pen dwellings, however, were equally as common. Regional
differences to this basic architectural style were most evident in con
struction materials. In the piedmount and interior coastal plain, slave
houses were frequently built with timber logs. Frame, brick, and tabby
constructions were typical of the tidewater (Otto 1975:104-105).
All indications are that the slave dwellings at Butler Island were
of frame construction. Contemporary accounts of nineteenth-century
visitors to Butler Island (Kemble 1961 (1863):67; Lane 1973:194), his
toric photographs, and archaeological resources support this finding.
These structures were evidently made of cypress. The walls were pre
sumably vertical board-and-batten as indicated in Figure 18. This is
not surprising because cypress, a very durable building material
(Panshin 1964:489), was abundant in the rice swamps. In fact, all
wooden remains uncovered at Butler Island have been identified as
cypress. At settlement #4, structure one provided evidence of cypress
construction. In Figure 23 the dotted line represents the outline of a
large fragment of hand-hewn cypress, possibly the remains of the floor.
Additional fragments were found scattered around the chimney.

Figure 23. The Extent of Excavations at Structure One, Settlement #4.
Key to Mapped Artifacts
A. Hinge
B. Adze
C. Scissors
D. Rounded chisel or gouge
E. Large fragment of black flint
F. Pintle
G. Grinding or polishing stone?
H. Axe head
I. Padlock
J. Hinge
K. Wood fragment
L. Wood fragment
M. Hoe
N. Padlock
O. Rice sickle
Dotted line: Large hand-hewn
cypress fragment

SOI N
123

129
The hearth level of the chimney for structure one indicates that
it was elevated off the ground approximately 69 centimeters or 2 1/4
feet. No evidence of brick footing piers was uncovered, but the round
posthole approximately 3.72 meters from the center of the chimney (see
Figure 22) may have functioned as a timber building pier. If this
posthole represents the outer wall of the structure, then the width of
this dwelling was approximately 7.44 meters,or 24 feet. The two small
square posts to the left of structure four's chimney (see Figure 21)
indicate that it was approximately the same width. Given a width of 24
feet, the length must be close to 43 feet. These dimensions, 24 by 48
feet are slightly larger than the dimensions given for most duplex slave
houses. Dimensions of slave houses at the other Butler plantations
appear to have been 20 by 40 feet (Roswell King, Jr., 6 Feburary 1835,
1 February 1835). Moreover, Olmsted noted that the duplex family tene
ments on the Georgia rice coast were 21 by 42 feet (1968 (1856):466).
Both Olmsted's and King's dimensions suggest that the slaves at Butler
Island had a little more space available to them than the average slave
family on the Georgia coast.
In 1811, 86 slaves were listed on the provision list for
settlement #4. Assuming that eight duplexes were at that time on the
premises, the average number of slaves to one duplex was 10.7 or So
slayes per family unit. These figures correspond remarkably with the
5.2 figure recently suggested as an estimate for the number of slaves
living in a single family structure (see Fogel and Engerman (1974:115).
These figures also adhere to those given by Fanny Kemble for Butler
Island (1961 (1863):67) and those of Olmsted for other rice plantations
(1968 (1856):422).

130
The function of the irregular pattern of postholes located out
side of structure one could not be determined (see Figure 23). Although
only one side of this feature vas uncovered, it is suggestive of a pen
or coop. Historic sources indicate that the Butler slaves raised both
pigs and chickens. Possibly, these postholes represent a portion of an
enclosure to keep pigs or chickens fenced in. Postwar photographs of
rice coast slaves or freedmen houses frequently show fenced yards.
One of the posts of the presumed enclosure intruded into another
feature, which was possibly a fire pit. Charred bone and wood frag
ments were recovered from this feature. Such a fire pit could have
been used for a number of purposes, including to ward off bugs, burn
trash, wash clothes, or melt lead for the making of fishweights or buck
shots.
Chimney constructions
The most detailed information regarding the construction of the
cabins was derived from the chimneys. With the exception of structure
five, all of the chimneys appear to have been constructed of tabby
bricks. Even the clay brick chimney was found to have tabby bricks in
its lowest course.
Tabby, a construction material used extensively along the Atlantic
Seaboard in the eighteenth and nineteenth centuries, has been defined
as a concrete construction-like material made from equal or nearly equal
portions of lime, shells, water, and sand (Gritzner 1978:9). Like
adobe and modern concrete, tabby bricks are not fired in the manufac
turing process. Instead, the semi-fluid mixture was poured into brick
molds and left to harden for a period of 6 to 18 months. Close
.

131
adherence of this procedure often resulted in a very durable building
material (Gritzner 1978:36). The tabby bricks made at Butler Island
evidently included shell robbed from prehistoric shell middens. This
was indicated by the presence of burnt aboriginal ceramics found within
the bricks.
Although tabby brick may have been a durable construction material,
it was not at all fire resistant. The fire box of structure one was
lined with clay bricks (see Figure 24). This has been found to have been
the case with most tabby chimneys. In addition, the heat from the fire
had apparently caused the hearth of structure one to buckle inward and
to separate from the outer walls of the chimney. Thus the durability of
tabby as a construction material for chimneys is questionable. Perhaps,
structure five was made primarily with clay bricks to guard against the
lack of fire resistance inherent with tabby bricks.
Two bond styles were employed in the brickwork, of the outer chim
ney walls. The "English bond" (Noel-Hume 1974:122), which alternates
courses of stretchers (the side of the bricks) and headers (the end of
the brick) was used at the base of the chimney. The above grade brick
work was laid in "common bond" (Noel-Hume 1974:122), that is, stretchers
in every course for five or more courses (see Figure 25).
Perhaps the most curious aspect of the chimney is the arch-shaped
support feature found below the hearth (see Figure 26). This architec
tural feature has been observed frequently at coastal plantations in
Georgia on planter, overseer, and slave dwellings. Its function is
unclear. It may simply reflect a shortage of bricks needed to complete
the chimney or it may have served a functional as well as an economic
-1 mam

132
Figure 24. Hearth of Chimney.
(Clay bricks: straight edges;
tabby bricks: rounded edges)
- '


Figure 26. Arch-Shaped Hearth Support. (Note how the hearth buckles inward.)
%
I
I
I
I
f
I
m
w
$
%£
f A'
m
*$-
m-

'r
f:
ft:
I
hr


136
purpose.* Possibly, it provided additional support to the hearth.
A similar arch-shaped feature was identified with structure four, but
none could be detected with structure five, the clay brick chimney.
Limited testing at the clay brick chimney may account for the failure to
detect this feature.
Part of the arch was removed in order to uncover artifacts below
it that would provide a construction date for the chimney. The only
dateable item recovered was a sherd of banded pearlware ceramic, which
establishes that the chimney had to be constructed after 1790. Because
the diking of settlement #4 did not begin until 1803, the chimney had to
have been built after that date. Unfortunately, the pearlware sherd does
not aid in the refinement of a construction date for structure one.
The bricks utilized in the clay brick chimney are known as
Savannah greys. They were given this name because they were made with
clay found along the Savannah River drainage. Several plantations in
the Savannah area had been engaged in brickmaking as early as 1750
(Granger 1947:9, 435). At Butler Island bricks were produced from the
"swamp clay" (Roswell King, Sr., 20 August 1815), but bricks were also
frequently purchased from Charleston and Savannah. The source of the
bricks used in structure five could not be determined. It may be pos
sible to tentatively identify bricks that were made at Butler Island.
Many of the clay bricks used to line the fire boxes of the chimneys and
used in the construction of the brick floor were very crudely made, low-
fired, and with large grey clay inclusions. It is suggested that these
bricks were probably made at Butler Island because brickmaking at the
^Charles H. Fairbanks, personal communication, 1980.

137
plantation was primarily experimental and never really became sophisti
cated. Therefore, it is likely that the very well made clay bricks
used in the construction of structure five were bricks obtained from the
Savannah River region.
Building haroware
Although more nails were recovered than any other artifact class
(Appendix 2). few other construction artifacts could be positively iden
tified. Most iron materials were severely corroded. These materials
were weighed but provided few clues as to their identity. Two hinges
and a small pintle were the only hardware recognized from structure one.
Fewer construction materials were found in association with structures
four and five. It is possible that some building hardware was salvaged
and reused on other buildings once these structures fell into disuse.
Additional testing, however, is needed to support this suggestion.
Of the 4,367 nails and nail fragments, only 4 could be identified.
All 4 nails were classified as hand-wrought nails, including 2 rose heads,
1 L-headed, and 1 headless. The rose head was a general purpose nail
while the L-headed and the headless were most often used for flooring
or trimming (Noel-Hume 1976:253). The identification of hand-wrought
nails cannot be used to assume that most nails used in the construction
of slave dwellings were of this type. Although nails were made by black
smiths at Butler Island, .machine cut nails were frequently purchased,
as indicated by the annual expense lists. Thus, chances are that a
large portion of the nails recovered were in fact machine cut.
Virtually no window glass was present. This finding supports the
general consensus that slave houses had few, if any, glazed windows.
Windows were usually closed by shutters.

138
At least one padlock was found in association with each structure
excavated. Structure one alone yielded three of these. Two of the pad
locks had brass keyhole covers, which establishes a date of 1840 for
their manufacture (Noel-Hume 1976:251). The frequency of these padlocks,
particularly at structure one, supports the observations made by antebellum
travelers to the rice coast, that slaves had to lock up their few, mea
ger possessions because of the rampant stealing which existed among them
(Olmsted 1968 (1856):432; Lane 1973:197). It is also possible that
padlocks were used to secure slave houses at night to ensure that the
slaves obtained sufficient sleep and did not spend time in nocturnal re
creation (Genovese 1974:535).
Summary
Although a detailed investigation of plantation spatial organiza
tions has not been undertaken, the settlement patterns of most large
plantations were characterized by broad similarities. Apparently, this
was also true for the plans of slave villages. Generally, both the inter
nal arrangements of the slave villages as well as the location of these
within the plantation complex were clearly functional. These functional
attributes, however, appear to have exceeded specific cash crop require
ments; therefore, it is difficult to identify plantation settlement
pattern data which are exclusively associated with rice culture. At
Butler Island tentative evidence of an accommodation to rice culture has
been indicated from the slave community plan, but additional archaeo
logical testing is needed to support these suggestions.
As for slave housing, the evidence from Butler Island conforms to
that of most tidewater plantations at both cotton and rice plantations.

139
Cypress house constructions may have been found more often on the rice
coast, but cypress was likely to have been used at the barrier island
plantations as wel1.

CHAPTER 5
THE MATERIAL CULTURE OF SLAVERY
The hypotheses outlined in Chapter 1 are based on the following
three assumptions concerning slave material conditions in coastal
Georgia:
1. Artifact classes identified as food-related activities, cloth
ing, personal possessions, and leisure, reflect general condi
tions in slave lifeways which may have been related to labor
management practices established for the production of tide
water staples. Thus, these conditions should be evident at
both rice and long-staple cotton plantations.
2. The habitats where certain cash crops were produced will be
evident in slave lifeways primarily in community organizations
(Chapter 4) and in exploitable food and nonfood resources.
3. Tools, farming implements, and specialized crafts should be
directly related to cash crop production.
In order to test these assumptions, it is necessary to refine the arti
fact groups indicated on pages 18 and 19. Functional artifact groups
are used here which consider material attributes when appropriate.
The following artifacts are included and are listed in the order in which
they are discussed:
1. Food-related activities
a. Kitchen artifacts (food preparation, processing, serving
storage, and containers)
140

141
b. Food procurement activities (hunting and fishing)
c. Food resources
1. Domestic
2. Nondomestic
2. Personal posessions
a. Clothing
b. Accessories and ornaments
c. Household items
3. Lei sure-time activities
a. Tobacco equipment
b. Games and toys
c. Miscellaneous objects
4. Farming and specialized crafts
Obviously many of these groups crosscut categories, which is
generally a problem with the establishment of functional categories.
But to discuss these artifacts on the basis of material characteristics
alone would be meaningless to this study. These artifact groups very
closely resemble those established by South (1977:95-102) for the defi
nition of artifact patterns, and his classes are used in Chapter 6 to
derive a slave artifact pattern. Counts of artifacts recovered from
slave settlement #4 are listed in Appendix 2.
Kitchen Artifacts
According to historic sources, food
kitchens in each of the slave villages at
was prepared at central
Butler Island (Kemble 1961
(1863):55).
'et it is curious that kitchen artifacts form the second
highest artifact class next to building hardware. Although most of

these artifacts were found to be ceramics and glass pertaining to food
serving and storage, food preparation and processing artifacts were
also found, which has been the case at other slave sites in Georgia as
well. Presently, the archaeological evidence from Butler Island and
elsewhere in Georgia strongly suggests that most culinary activities took
place in the slaves' cabins and not at historically documented central
kitchens. Thus far, no central kitchens have been identified archaeologi
cally; therefore, it may be questionable whether they existed. Perhaps
one cabin served as a kitchen for feeding slave children, who were often
fed separately from adults (Genovese 1974:507-508), but archaeological
evidence of this very specialized activity is lacking. At present, the
archaeological evidence from coastal Georgia points to the slave cabin as
having been the locus for most cooking and eating activities.
Ceramics
More attention has probably been directed toward the study of
ceramics then any other artifact class in archaeological contexts. Con
sequently, ceramics have become vital indices for the documentation of
human activities (Fontana 1973). At slavery sites the following tenta
tive patterns have been observed:
1. A varied range of types and styles suggests that slaves were
not furnished with a special class of wares but obtained
either planter castoffs of unmatched assortments of ceramics
marketed for slave use.
2. Cups and bowls and to a lesser extent plates and platters
are the primary vessel forms of slave ceramic assemblages.

143
3. As for decorated tablewares, annular decorated ceramics pre
dominate, whereas the transfer printed styles are minority
items.
Because these patterns have been observed consistently at coastal
cotton plantations, it is hypothesized that similar ceramic patterns
would be observed at rice plantations. It is assumed that slaves at
both rice and cotton plantations had a similar access to ceramics and
other material possessions. This slave ceramic pattern would suggest
either that special purchases were supplied to slaves or that perhaps
slaves purchased ceramics made available to them from local merchants.
Ceramic styles and vessel forms
For the most part, the ceramics uncovered at Butler Island were
utilized for the serving of food or the storage of it. The exceptions
are those vessels used for hygienic purposes. The range of vessel forms
is given in Table 8. All the ceramics have been described in detail
elsewhere and are very briefly summarized here.
Red coarse earthenwares. Red paste ceramics have been associated
with dairy and general purpose kitchen activities. Their manu
facture is well documented for New England and adjacent areas ("Stradling
and Garrison 1977; see related to redwares); Less information,
however, is available for the South. In Georgia, these ceramics were
manufactured in Savannah during the mid-eighteenth century, but
the location of later kilns is not clear (Ketchem 1971:31). At Butler
Island, three styles were identified, including unglazed, brown glazed,
and black glazed varieties (see Figure 27). All vessels appear to have
r storage or some other utilitarian purposes.
been used for eithe

144
Table 3
Vessel Forms of Ceramic Styles
Ceramic Style
Vessel Form
Number
Coarse red earthenwares
Ungazed
Storage jar
1
1
Black lead glazed
Indeterminate
1
Chamber pot
1
Brown lead glazed
Large bowl
1
'ined earthenwares
Undecorated whiteware
Chamber pots
3
Serving bowls
and
cups
4
Plates/platters
4
Mug
1
Pitcher
1
Transfer printed pearl ware
Serving bowls
or
cups
2
Platter
1
Edged Ware
Shell edged pearl ware
Plat.es/plattei
rs
6
Molded edged whiteware
Plate/platter
1
Handpainted
Pearl ware underglazed blue
on white
Serving bowls
2
Pearl ware polychrome '
Whiteware polychrome
Shallow bowls
or
cups
2
Bowls or cups
2
Annular
Circular bands
Pearl ware
Serving bowls
and
cups
3
Whiteware
Serving bowls
and
cups
2
Yellow paste
Serving bowl
1
Rouletted
Pearlware
Serving bowl
1
Whiteware
serving bowl
1
Finger painted
Serving bowl
1
Mug
i

145
Table 8-
-(continued)
Ceramic Style
Vessel Form
Number
Mocha
Whiteware
Serving bowl
1
Yellow paste
Serving bowl
1
Martelized
Bowl and cups
2
Sponge decorated
Serving bowl?
1
Stonewares
Alkaline glazed
Storage jars
2
Brown salt glazed
Storage jars
2
Low fired unidenti
tied
Storage jar
1
Porcelain
Indeterminate
1
Total
59
Grand Total
Bowls or cups
32
Plates/platters
11
Storage jars
6
Mugs
2
Chamber pots
4
Other
4

Figure 27. Red Coarse Earthenwares.
Key to Artifacts
1. Brown-glazed
2. Black-glazed
3 Brown-glazed
4. Black-glazed
5. Unglazed lip
large bowl
chamber pot
shoulder fragment
basal fragment
of storage jar

'
' 0- -i:
, .
' - '
ttVi : Ife :
\ k m /.>'
. ' 3 : ..
w I
.->v-y
-
:
wi&w ;
gsgSJji;
¡itis
3 R;
0-%&
.
-P

148
Undecorated white ceramics. The most frequently occurring ceramics
were undecorated white fragments, possibly including creamware, pearlware,
and whiteware, with most appearing to be whiteware. Those classed as
whiteware were defined as such on the basis of a white overall glaze tint
and by the absence of the blue puddling characteristic of pearlware
(Price 1979:13-15) or by the creamy white color and greenish-yellow
puddling found on the creamwares (Noel-Hume 1973:217-254). The undeco
rated whitewares are found in a variety of vessel forms. The Butler
Island collection includes chamber pots, bowls, mugs, platters, and
plates.
Transfer printed pearlware and whiteware. The surface decorations
on transfer printed ceramics were mass produced by the use of copper plate
engravings. These ceramics are one of the most frequently occurring
at late eighteenth- and nineteenth-century sites (Noel-Hume 1976:129-
130). At slave sites, however, they appear to be minority wares, and are
frequently associated with high status households. Only 17 sherds were
recovered, but three vessel forms are represented: a platter, a plate,
and a bowl or cup (see Figure 28).
Handpalnted pearlware and whiteware. Blue-on-white and later
polychrome floral designs were painted on plain white bodies. These
ceramics seem to have been particularly popular among poorer classes.
Beginning in 1835, bright floral designs were stenciled on these ceramics
(Noel-Hume 1976:129).
Edged wares. White bodied vessels with embossed feathers, shells,
or fish scales placed on the edge were usually painted in blue or green.
Typical vessel forms are plates and platters, but other vessels are
known (see Watkins 1970, Figure 3a),

:> isscbftaw* -'

Figure 28
Transfer Printed Pearlware Fragments.

y
I
1
O
LO
U5UI
LUIA.'

av s
'i A ,
- r > *v,-. s\ / .. >-
Sv:
:
-
-'K-'W?SW$4W;-
a
:vg£#|iS£v^SfeJ
i: " ."
\
' 'V; /./' ' ; '; ''
'- '- 1 ' V '' J, = \
.
¡¡pip
mm'Axm
Baiisaaaassr
3lpi#
!,V. LAB : i:
Cn?
rtv^'V
,. ,p.
mimmmm
; ; '-a
fet§iSi£a -- 5 /. :,:,,' a /a .'vvf
¡i*. -, a -a-va
a:. v; 4
m m¡
a:;;vvvv:a.V v/ ; %
I
- * -v:jM
i0lme!
V.V-. V:V
fSp'SfWfti
|:Vvvva'#';:jv'Sv
fW''a'sr -i;
feiWftiail
358$ r? f-V ^:j---Va.r-;
'.s'&v
i . ,
';% ;Al .
;
ftngra
\ &^KsSdSS,'iiuSj"%3t^i^¡:¡
¡

Annular wares. Several decorative motifs are characteristically
found cn annular ceramics, often more than one on the same vessel (see
van Rensselaer 1978:240-24). The motifs include slip banded circles,
mocha, swirl, circle and cube, and engine turned or rouletted (Price
19/9:18). Vessel forms include mugs, bowls, jugs, teapots, and tureens.
At Butler Island, only cups, bowls, and mugs were recognized (Figures
29 and 30).
Spong decorated. Ceramics known as sponge decorated, or spatter-
ware, consist of the application of colors to a white bodied vessel by
means of an inked sponge. Plates, cups, and saucers are the most
common vessels (Price 1979:19). Only one vessel was represented at
Butler Island, and it appears to be a small cup or bowl.
Alkaline-glazed stonewares. A glaze made from wood ash and lye
water was frequently applied to stoneware storage vessels in the South
(Greer 1971). These were probably produced cheaply by small potters on
the Georgia or South Carolina piedmont.
Salt-glazed stoneware. The technique of manufacturing stoneware
in which salt was thrown in the kiln while the vessels were being fired
resulted in salt-glazed stoneware. The effect is an orange-peel tex
ture on the surface of the vessels. Most nineteenth-century salt-glazed
stoneware was made into storage forms, including jugs, jars, pitchers,
and bottles (Sarnmis in Stradlirig end Garrison 1977:112).
Porcelain. Generally found in all tableware vessel forms, porce
lain is associated with high status households. Cne undecorated white
fragment was uncovered. It appears to have been part of an ornamental
vessel form, most probably salvaged from the overseer's house.

Figures 29 (upper and 30 (lower). Annular Ceramics.
Key to Artifacts
A. Whiteware, black stenciled design
with red bands on white body
B. Rouletted with swirled bands
C. Mocha design on yellow body
D. Blue stripes on yellow body
E. Circular brown and tan bands on
pearl ware body


As at other coastal slave sites, a variety of ceramic styles were
found to be present at Butler Island. These assemblages seem to indi
cate that slaves received an assortment of unmatched ceramics. In the
Butler Island collection some vessels appear to have been "wasters,"
those which have no market value (Noel-Hume 1974:170). Generally the
wasters were found to be poorly fired, warped, or with a sloppily applied
glaze. Apparently, ceramics were not regularly supplied to the Butler
Island slaves, as these items do not appear on the annual expense lists.
However, they may have been provided occasionally, and there are two ref
erences to the "purchase of 50 crates of crockery" (Butler Estate Papers,
1811; Roswell King, Jr., 18 March 1929). The size of this purchase sug
gests that these ceramics were for the slaves. How often ceramics were
supplied to the slaves could not be ascertained, but it appears to have
been infrequent. Most likely, slaves purchased their own crockery with
money they earned from selling produce. Merchants in Darien frequently
advertised the sale of ceramics, and indications are that these were
perhaps special wares obtained exclusively for slave use (Otto 1975:
160).
While a variety of vessel forms was recognized, bowls and cups
predominated (see Table 8). The frequency of bowls and cups strongly
suggests that most slaves' meals were one-dish soup, stew, or porridges.
A high frequency of bowls has also been observed at other Afro-American
sites (e.g. Baker 1978). Baker suggested that the high frequency of
bowls may be not only an artifact of poverty, but also an indicator of
Afro-American cuisine (1978:81).
Of the decorated ceramics, the annular wares occurred in very
high frequencies. At Butler Island, these ceramics comprised over 52
percent of all the decorated ceramics (see Table 9).

155
Table 9
Frequencies of Decorated Tablewares
Ceramic Style
Fragments
Percent
Edged ware
53
20.78
Transfer printed
17
6.66
Handpainted
48
18.82
Sponge decorated
2
.80
Annular
135
52.94
Total
255
100.00
The high frequencies of banded ceramics at Georgian slave sites suggest
that these were inexpensive ceramics obtained for slave use.
Dating the ceramics from But!er Island
One method to determine slave usage of outmoded ceramics would be
to apply the Mean Ceramic Data Formula (South 1977:201-230) to those
uncovered at settlement #4. Although the application of the formula to
nineteenth-century ceramics has not been refined, it has been used to
demonstrate the presence of out-of-date ceramics in another slavery con
text (see Fairbanks 1974).
The formula was applied only to ceramic styles of the Butler
Island collection for which South had determined median dates of manu
facture (South 1977:210-212). A median date of 1835.01 was derived
(see Table 10). This date closely approximates the actual median date
of antebellum occupation, assuming that the site was occupied between
1804 and 1861.

156
Appplication of the M
Table 10
ean Ceramic Date Formula
to Tablewares
Ceramic Style
# of Fragments Mean Date
Product
Whiteware, undecorated
217
1860
403620
Green and blue
shell edged pearl ware
48
1805
86640
Transfer printed
pearlware
17
1818
30906
Underglazed blue and
white pearlware
11
85
19855
Underglazed polychrome
26
1805
46930
Finger painted
1
1805
1805
Mocha
10
1843
18430
Annular pearlware
91
1805
164355
Total
421

772541
Note: Mean ceramic date = 1835.01.
The application of the Mean Ceramic Date Formula to the Butler
Island collection, unlike its application to other slavery sites in
coastal Georgia, indicates that the Butler Island slaves were not using
outmoded ceramics. Two factors may account for the difference between the
Butler Island ceramic assemblage and the assemblages of the other slavery
sites: imprecise dating of the whitewares and the absence of eighteenth-
century ceramics. The precise manufacturing dates for the whiteware are
unknown (Noel-Hume 1976:131). Their assigned median date of 1860 is at
best an approximation, and the actual median date is likely to have been
much later. Because dates of manufacture for whiteware are unclear, many
authors do not apply the formula to these ceramics. The inclusion of
these ceramics to the formula here, however, suggests that a satisfac
tory median date can be derived.

Eighteenth-century ceramics form significant proportions of slave
ceramic assemblages at other sites, where the presence of outmoded
ceramics was indicated by the Mean Ceramic Formula. The absence of
eighteenth-century ceramics in the Butler Island collection could account
for the derivation of a fairly accurate median date of antebellum occupa
tion. Thus, the Butler Island slaves were receiving more stylish ceramics
than other slaves in the area but were perhaps using some slightly out
moded ceramics such as the annular pearlwares.
Nonceramlc Artlfacts
Nonceramic kitchen artifacts included glassware, cooking and
food processing equipment, and eating utencils.
Glass bottles and tablewares
The most abundant nonceramic artifact was glass, a total of 448
fragments having been recovered. Only 5 of these fragments were iden
tified as tablewares, and two vessel forms were represented: a possible
tumbler and a decanter or vase. The remaining fragments were found to
be bottle glass, but no whole or nearly whole bottles were recovered.
Table 11 provides the approximate style and number of bottles repre
sented.
As can be seen in Table 11, dark green bottles predominated. This
has been found to be the case at other slave sites. Dark green bottles
are particularly significant because they were generally used as con
tainers for brewed alcoholic beverages (Switzer 1974). Evidence for
alcohol consumption among the slaves at other sites in Georgia is
another household pattern observed in both cash crop contexts. It has

Table 11
Styles of Glass Bottles Represented at Butler Island, Settlement #4
Glass Color # of Fragments Bottle Form # of Bottles
Dark green
317
3-hinge mold
3
free blown
1
dip mold
2
Light green
35
could not be .
determined
1
Clear
51
could not be
determined
1
Aqua
40
medicine vial
3
Total
443
11
been suggested that alcohol was an item obtained by slaves, not one that
was issued to them (MacFarlane 1975:117). The historic evidence for
Butler Island, however, indicates that alcoholic beverages were occa
sional items issued to the slaves, presumably for medicinal purposes:
He drinks no rum and he can be left to give rum to the Negroes
every wet morning. I think it will be profitable even if
they drank 100 gallons [of rum] a year. The people have a
number of discomforts that they would not get elsewhere.
Scant of water, wood, little or no vegetable, we must make
some way to make them comfortable & content. A number of
them are given to swelling at times. I think rum is a good
preventative. (Roswell King, Sr., 22 October 1803)
Alcoholic beverages were undoubtedly infrequently given to the
Butler slaves. Nonetheless, alcohol was issued to them at times. In
fact,
some rum was manufactured on the
plantation (Roswell King, Sr.,
19 January 181/), and it is possible that the recovered case bottle

159
fragments contained rum made at Butler Island. Yet it is doubtful that
all alcohol consumed by the slaves was issued to them. The overseers'
concern about slave overindulgence in alcoholic consumption attests to
this. Some alcohol was probably purchased by the slaves. Perhaps the
light green "champagne" bottle (see Switzer 1974:23-24) was such a pur
chase.
Unfortunately, the former contents of recovered medicine vials
could not be ascertained. A variety of medicines was regularly ordered
for the treatment of slave diseases (see page 82) and presumably these
were kept at the plantation dispensary. The remains of medicine vials
uncovered from the slave quarters may have been plantation issues or
slave purchases.
Whether slaves were issued liquor or medicines or purchased these
items for themselves, the archaeological evidence indicates that they
were regularly consumed by them. Although alcohol was possibly con
sumed for its medicinal effect, historical evidence indicates that it
was used as a lei sure-time activity, particularly during the holidays.
Cooking and processing equipment
Several curve-shaped iron fragments suggestive of pots or kettles
were recovered. Two kettle legs, however, were the only diagnostic
artifacts of cooking equipment. These were evidently legs to trpoda!
caldrons, which often weighed up to 60 pounds when empty (Booth 1971:
38). Iron cooking pots were often supplied to the slaves for cooking
purposes, and remains of these pots have been uncovered at other slave
sites as well (Ascher and Fairbanks 1971; MacFarlane 1975:106). One
leg was found in the hearth of structure four, which further supports
u.
1
-
& -
:T' y-' >-,!< f "r

160
indications that the Butler Island slaves prepared their meals in the
privacy of their own homes.
Several polished stone fragments were found within the slave dwell
ings (see Figure 23SG). Some of these, flat slabs of lateritic sand
stone containing mica, appear to have been used for grinding,* possibly
to grind hominy into grits (Hilliard 1972:49) or for some wild vegetable
items. Quartzite pounders were also found and were evidently used to
crack items such as nuts. Similar quartzite pounders were found at
Cannon's Point plantation on St. Simons, and it was suggested that these
were perhaps taken from prehistoric middens (MacFarlane 1975:107).
Eating utencils
Fanny Kemble observed that the Butler Island slaves ate with
broken iron spoons, pieces of wood, and with their fingers (Kemble 1S61
(1863):100). Yet evidence of more substantial eating implements was
provided by the archaeological data, including four spoons., two cutlery
knife blades, and several fragments of bone handles. Presumably, the
bone handles were attached to iron cutlery. The St. Simons slave sites
have yielded very similar evidence of eating utencils. Four spoons
were identified, two of iron, one of whitemetal, and one of pewter
(see Figure 31). Possibly, the pewter spoon was a luxury item and a
very special personal possession. The knife blades are made of iron and
were possibly attached to bone or wooden handles.
The bone handles are perhaps the most curious of all the eating
utencils (Figure 32). Although these resemble manufactured bone cutlery
*Charles H. Fairbanks, personal communication, 1980.

Figure 31
Eating Utencils. (A is a white metal spoon;
B is a pewter spoon.)


Figure 32. Carved Bone Fragments. (Note scoring in upper fragment.)

164
gas*

165
handles of the eighteenth and nineteenth centuries (Noel-Hume 1976:182
(8 and 9); South 1977:98, below), they appear to have been homemade
items. Several uncarved bone fragments crudely sawed to form the same
shape were later recognized within the faunal collection. Also, the
fact that several woodworking tools were identified from structure one
(the provenance of the bone fragments) provides additional evidence to
suggest that the bone handles were homemade. At least it seems possible
that the slaves were able to carve these bone fragments. The largest
fragment has several scoring marks which evidently postdate the carved
motifs. The significance of the scoring is unknown but may indicate
later whittling activity after the handle was discarded (see Figure 32).
Kitchen Artifacts: Summary
The similarities in the artifact styles and the frequencies of the
kitchen artifacts between Butler Island and other slave sites, notably
Cannon's Point, are astonishing. These similarities provide a pre
liminary basis for the establishment of a slave kitchen artifact pat
tern for coastal Georgia. Subtle differences are evident. Perhaps the
most notable is the absence of identifiable eighteenth-century ceramics.
It cannot be ruled out that future excavations may uncover these.
Another possibility may be that Butler Island slaves obtained more
stylish ceramics than their barrier island counterparts. Their close
proximity to the town of Darien may have been a factor which enabled
them to purchase ceramics more frequently. Also, the barrier island
slaves had resident planters and were perhaps getting their castoffs.
But, in spite of these two differences, the kitchen artifact group tends
to reflect a regional pattern.

Food Procurement
Two categories of food procurement artifacts were uncovered:
fishing equipment and firearms (see Figure 33). The only evidence of
fishing activity was provided by lead net sinkers. Numerous fragments
of lead slag suggest that these were melted down for making fish net
sinkers, weights, and buckshots.
The archaeological evidence of slaves' access to firearms is per
haps the most significant conflict with contemporary accounts of slavery.
At Butler Island, the evidence is more convincing than that previously
recovered from other slavery sites. Besides buckshots, musket balls,
and gunflints, a brass trigger guard was uncovered. To the author's
knowledge, a trigger guard has not been previously found in other slave
sites in Georgia. Fortunately, at Butler Island, the plantation records
also indicate that slaves were occasionally given the privilege of using
guns for hunting.
At Butler Island, a fellow asks me permission to get a gun
for ducks. Plenty of which are at the landing and in the
fields. I have sometime since taken all firearms from them
' as I think they have forfeited their charter from the swamps
and their conduct. I am often glad my phi 1 entropy in'allow
ing them to have guns did not extend further than this.
(Roswell King, Jr., 28 June 1829)
The archaeological and historical evidence is conclusive that slaves
had access to firearms for hunting at Butler Island. It is likely that
this was an established practice for other plantations in the area.
Gun flints were made from both black, or Dover flint,and French,or
honey-colored,flint. Curiously, several large flint nodules of both
black and honey-colored flint were found in and around the slave houses.
Although little evidence of flint debitage was uncovered, it is possible

Figure 33.
Food Procurement Artifac
Key to Artifacts
1. Lead fish net sinkers
2. Lead slag fragment
3. Buckshot
4. Gun flints
5. Brass trigger guard
I.'
t
£

168
'S,'39r-

169
that this flint was used to make the gun flints. Flint is not native to
the Georgia coast. Moreover, black,or Dover, flint is found only in the
Old World. Therefore, its presence at Butler Island indicates that it
was brought there. Possibly it was obtained from nearby ballast dumps.
Ballast materials frequently occur at coastal sites (e.g. Jones 1976).
How this material turned up at the slave sites is unclear. A sugges
tion is provided by Roswell King, Sr., who indicated that the ballast
was used in repairing the sugar works (Roswell King, Sr., 17 March 1816).
If ballast was periodically brought to Butler Island, the presence of
the Old World flint at the slave settlements would be explained.
Food Resources
Unfortunately, the food bone remains at Butler Island were found
to be very poorly preserved. Although acidic, waterlogged conditions
are favorable for the preservation of hide, leather, hair, and wool,
bone is preserved best in alkaline situations (Cornwall 1958:69). The
fluctuating water conditions presently found at Butler Island provide an
additional unfavorable condition for bone preservation. Such situa
tions are usually characterized by increased organic decay.*
Besides unfavorable bone preservation conditions, a large portion
of the collection had been severely charred, apparently by exposure to
very hot, low-oxygenating fires, as indicated by the whitish blue color
of many fragments.** (Cremated bone often exhibits similar physical
*Larry Banks, Chief, Army Corps of Engineers, Southwestern
Division, Dallas, Texas, personal communication, I960.
**Elizabet.h Wing, Florida State Museum, Gainesville, Florida,
personal communication, 1980.

170
characteristics; see Gejvall 1963.) Consequently, faunal materials
collected are very brittle and fragmentary. However, the presence of
charred food remains (not as severely charred as the Butler Island
collection) in and around slave dwellings has been reported (Fairbanks
1974:87: MacFarlane 1975:167). Perhaps this reflects a refuse dis
posal practice in which food remains were burned in the fireplace and
may have been subjected to repeated burnings. Periodically, the food
remains must have been swept out and onto the floor.*
Faunal analyses were conducted at the Zooarchaeology Laboratory of
the Florida State Museum. As a consequence of the previously described
conditions of the bone remains, most faunal materials could not be
identified beyond animal class. Yet some classifications to lower taxa
were possible (see Table 12). All bone fragments were weighed but not
counted. Although charring and other modifications can reduce bone
weights by as much as 50 percent, bone weight appears to be a more
accurate basis for quantification than mere fragment counts (Wing and
Brown 1979:126). Therefore, bone weight was preferred to bone count
in the Butler Island collection where so much of the bone was frag
mented. Here fragment counts would tend to misrepresent the relative
frequencies of the animal species identified within the sample.
Yet, despite the limitations of the faunal collection, the rela
tive proportions of domestic to nondomestic food animals, as indicated
in Table 13, correspond remarkably well with the collection from the
Cannon's Point north slave cabin (Otto 1975:310). Additionally, the
identified nondomestic species are typically found in the deltaic
habitats of the Georgia tidewater. Given the limitations of the faunal
*Char!es H. Fairbanks, Personal communication, 1980.

171
Identified
Table 12
Fauna from Slave Site #4,
Butler Island
Species
Common Name
Bone Weight
in Grams
Percent
Amia calva
Bowfin
21.1
.96
Lepiososteus sp.
Gar
9.5
.43
Si 1uriformes
Catfishes
13.5
.61
Icaturidae
Freshwater catfish
23.0
1.04
Archosargus
probatocephal us
Sheepshead
3.0
.14
Scianeidae
Drum
4.2
,20
Unidentified fish
62.8
r\)
o
Kinosternon sp.
Mud turtles
6.0
.23
Chrysemys sp.
Coolers and sliders
13.0
.59
Trionyx ferox
Softshell turtles
14.1
.64
Unidentified turtle
252.5
11.50
Squamata
Snakes
1.2
.05
Aythya sp.
Diving ducks
2.0
.09
Gall us gal I us
Domestic chicken
5.0
.23
Unidentified birds
25.5
1.16
Rodentia
Unidentified
rodent
2.0
.09
Cri citidae
New World rats
and mice
6.0
.28
Procyon lotor
Raccoon
2.5
.11
Artiodactyl
Even toe
Ungulates
200.0
9.11
Sus scrofa
Domestic and
feral pig
205.8
9.38

Table 12 (continued)
Species
Common Name
Bone Weight
in Grams
Percent
Odocoileus
virginianus
Whitetailed
deer
12.0
.54
Bos taurus
Domestic cow
442.5
20.16
Unidentified
mammals
867.00
39.51
Total
2194.20
100.00
Table 13
Relative Weights of Domestic and Nondomestic Animal Food
from Butler Island
Food Animals
Weight
Percent
Domestic
653.3
59.85
Nondomestic
438.2
40.15
Total
1091.5
100.00
aRodentia, Cricitidae, and Squamata not included.

173
collection, it is still possible for the purposes of this study to test
the postulates regarding domestic and nondomestic food resources at
slavery sites.
Domestic Food Resources
In Chapter 3, the diet of the Butler Island slaves is briefly
described (see Pages 67, 77, 82). The Butler family estate records
indicate that this diet consisted of corn, purchased pork and salted
fish, and occasionally rice and molasses. Chronic shortages of meat
rations, however, were known, and meat may not have been supplied to
the majority of slaves on a regular basis. In fact, the ditchers were
apparently the only slaves to whom a ration of pork was regularly supplied.
Most slaves seem to have been fed a great deal of salted fish. Occa
sionally cattle, sheep, and pigs were slaughtered to supplement the meager
rations.
Taken together the dietary information obtained from the plan
tation records and the archaeological record present a confused pi cutre
of slave foodways at Butler Island,, Although the plantation records
indicate that the slaves had very little meat, when it was made
available, pork was the primary domestic meat consumed by them. The
archaeological evidence, on the other hand, indicates that beef may have
been the primary meat in slave dietary remains; the relative weight of
the recovered cattle (Bos taurus) remains exceeds that of pig (Sus scrofa)
remains (see Table 14). The implication that beef contributed more to
slave diet than pork is in keeping with the faunal evidence from Cannon's
Point (Otto 1975:327-328, 332-334). Furthermore, beef often supplanted
pork rations on many coastal plantations, particularly along the rice

Table 14
The Relative Weight of Domestic Species at Butler Island
Species
Weight
Percent
Gallus gallus
Domestic chicken
5.0
.8
Sus serofa
Pig
205.8
31.5
Bos taurus
Domestic cattle
442.5
67.7
Total
653.3
100.0
coast, and these beef rations tended to be larger than rations of pork
(Hilliard 1972:59, 130-131). Also, at Butler Island, more cattle was
available than pig from the plantation livestock. Although some of the
cattle included oxen, which were used as draft animals, they were peri
odically slaughtered to feed the slaves: "Buy 20 more steers to add to
the oxen for Christmas beef" (Roswell King, Sr., 15 May 1813). The
reference to "Christmas beef" may indicate that beef was only provided
on very special occasions.
Perhaps the confusion between the archaeological evidence and the
Butler estate records lies in differential butchering and curing prac
tices of cattle and pigs. The pig remains identified within the sample
included primarily mandible, teeth, and metapodial fragments, suggest
ing that the remainder of the pig was utilized elsewhere. While it is
possible that the Kings kept the remainder of it "on plantations that
were overseer operated, the great majority.of the hogs were killed and

175
made into pork for the slaves (Hilliard 1972:57). This meant that
slaves viere likely to have been fed the entire pig. Also, the planta
tion records for Butler Island indicate that this was often the case:
"We have killed 5 bbl of pork this year [which] with the two you sent
out will serve the ditchers" (Roswell King, Jr., 18 March 1821).
Hilliard has made the point that antebellum connotations for the terms
"bacon" or "pork" referred to much more than sides of a hog, often
meaning lean meat free of bone (1972:57-59). If the curing of pork at
Butler Island involved the removal of some or all bones, there would be
no archaeological evidence of hams, shoulders, or ribs. Additionally,
salt pork ("fat back") would not yield bone evidence at the point of
consumption and was likely to have been the most common form preserved
in barrels. The jaw and feet parts recovered may simply represent por
tions of the pig that were not cured but were given to the slaves just
the same.
On the other hand, cattle remains also included head parts as
well as humeri and scapula. These "shoulder roast" remains may indi
cate that the entire animal was slaughtered for the slaves and supplied
to them as fresh meat. Rarely was beef cured in the antebellum South;
it was most often eaten fresh (Hilliard 1972:44). References to the
slaughtering of cattle at the Butler estate suggest that all of the
beef was rationed to the slaves as fresh meat.
The point to be made here is not to underestimate the importance
of beef in the slave diet but to suggest that the higher bone weight
for beef than for pork may reflect differential usages of beef and pork.
Pork, the major domestic animal food resource (as indicated by the

176
plantation records) was possibly cured more often than beef. In the
curing process, some or all of the bone could have been removed, which
would leave little or no evidence at the point of consumption. Beef may
have been used more often to supply fresh meat, and the archaeological
record reflects this. Moreover, it appears that the Butler slaves could
have consumed all edible parts of both cattle and hogs, simply because
a full-time planter was not present to receive choice cuts. The size of
the slave population also suggests that this was the.pattern. To
supply 300 to 400 slaves with only the forequarters (heads, necks, tails,
and viscera) would require that more animals be slaughtered than if a
few whole animals were butchered exclusively for them.
Assuming that purchased pork supplied the bulk of the meat ration
to the Butler slaves, it is questionable What portion of the diet the
preserved food remains represent. Although purchased pork contained
some joints, most of it was bacon sides (Hilliard 1972:58). Purchased
pork, like plantation cured pork, may not be preserved in the archaeo
logical record. Thus, it is possible that the domestic animal remains
reflect only occasional dietary items. At least this suggestion sup
ports the plantation records as well as the need to supplement planta
tion rations with nondomestic animals.
No goat (Capra hlrcus) or sheep (Ovis aries) remains were identi
fied. Possibly they have been placed with the family of Artiodacyl
(see Table 12). The plantation records indicate that sheep were rarely
butchered for the slaves. Although Fanny Kemble apparently had mutton
nearly everyday at Butler Island (1961 (1863):184), the slaves were
evidently not as fortunate. In addition, few sheep and goats were

177
present at Butler Island before 1835, which predates Fanny's visit,
and it is possible that the sheep and goat population increased after
that date. But the archaeological record does not suggest that mutton
was consumed by the Butler Island slaves.
The sample indicates that chicken (Gal 1 us gal 1 us) was a very
occasional dietary item. A similar pattern has been observed at the
Cannon's Point slave sites (MacFarlane 1975:165: Otto 1975:311).
Chickens were apparently kept for the provision of eggs that were most
often sold or traded to local markets. In fact, most slave-owned live
stock was sold rather than consumed by the slaves (Hilliard 1972:60).
Peach pits were the only plant food remains, domestic or otherwise,
which provided archaeological evidence at Butler Island. A total of 82
whole pits and 62 fragments were recovered. Peach trees were present
on Butler Island (Kemble 1961 (1863):167)3 but no references to supply
ing the slaves with peaches could be found. In this case, the archaeo
logical evidence provided information not available in the documents.
The evidence of slave peach consumption supports the claim that
planters supplied slaves with seasonal food items such as fruits and
vegetables, which are generally not mentioned in the planter records
(Fogel and Engerrnan 1974:111), The records, however, do indicate that
the slaves were given some of the oranges and lemons from the Butler-
Island groves. Present-day fig trees at Butler Island, presumably
dating from the antebellum period, may also have been used to supply
fruit to the slaves. Both archaeological and historical evidence indi
cate that the Butler Island slaves were able to add fresh fruit to their
diets from fruit trees available to them.
ms

Nondomestic Food Resources
Traditional historic studies of slave diet have tended to over
look the potential contribution of nondomestic food resources. A recent
study of slave diet demonstrated that slaves residing within the coastal
eco-zone were able to make their diet adequately nutritious. Primarily
through the slaves' own efforts to supplement plantation rations with the
abundant food resources available within the coastal environment,their
diet was made nutritious (Gibbs et al. 1980). The study relies heavily
upon archaeological and zooarchaeological data uncovered from slave sites
in Georgia and Florida for evidence of food supplements. All of these
sites were formerly barrier island, long-staple cotton plantations which
lie within the marsh and lagoon section of the tidewater (see discussion
on pages 20-22). Slaves living on these plantations were able to ex
ploit a variety of habitats. Among them, the salt marshes form the
richest portion of the coast, and the archaeological evidence from long-
staple cotton plantations indicates that this habitat was most fre
quently exploited by the slaves (Fairbanks 1974:87; Otto 1975:345).
On the other hand, slaves on deltaic rice plantations were perhaps
limited to the variety of species that could be exploited. The delta,
being freshwater and to a lesser extent brackish, appears to have fewer
species of wild resources than the salt marshes. It is hypothesized
that the nondomestic species utilized by rice coast slaves would be
food items typically found in deltaic habitats.
Table 15 provides the relative weight of nondomestic resources
by animal class recovered from Butler Island. It indicates that fishes
and turtles were predominant while birds and mammals were less frequent


resources. Even taking into consideration the grave limitations of the
Butler Island faunal collection, freshwater fishes and turtles were
very abundant resources in the delta and possibly were exploited most
often by the slaves.
Table 15
Relative Weight of Identified Nondomestic Food by Animal Class
Class
Weight
Percent
Fishes
137.1
31.1
Turtles
285.6
64.7
Bi rbs
2.0
.5
Mammals
14.0
3.7
Total
438.2
100.00
Fishes and turtles
The three predominant fishes identified include bowfin (Amia calva),
gar (Lepiososteus sp.), and freshwater catfish (Icaturidae), all found
most frequently in turbid, stagnant waters.'
The bowfin is a primitive fish of a large family, most of which
now exist only as fossils. The bowfin has a unique air bladder which
enables it to live in water unsuitable for other fish. It is a bottom
feeder and usually inhabits weedy lakes and sluggish streams. Seldom is
it found in fast currents (McClane 1978:178-179).
Gar is another primitive fish. It is primarily found in fresh
water but has a high tolerance for saltwater. Like the bowfin, gar has
a modified air bladder which it can use for breathing in very stagnant

180
waters. Gar is frequently seen near the surface of the water to expel
gases and gulp air (McClane 1978:179-183).
Freshwater catfishes are known for their diverse eating habits.
They are bottom feeding scavangers and most species are fond of turbid
slow moving waters, the exceptions being the blue catfish (Ictalurus
furcatus) and channel catfish (Ictalurus punctatus) which inhabit lakes
and larger rivers that have clean sand or gravel bottoms.
The only two fishes typically found in saltwater were identified:
sheepshead (Archosargus probateocephalus) and drum (Sciam'dae). Sheeps-
head are found in diverse habitats, typically inshore but frequently
around off-shore reefs (Dalhberg 1975:73-74). Generally, they travel in
schools and feed upon rnollusks. They will enter brackish waters and
on. occasion even freshwater (Larson 1970:66).
Drums (Scianidae) are among the most common fish found along the
coast (Dahlberg 1975:69). Although several species of freshwater drum
are known, they are not found along the south Atlantic Seaboard (McClane
1978:194). It is likely that the drum represented in the collection
are saltwater species which also occur in freshwater, such as the red
drum (Sciaenops ocellata) or spotted seatrout (Cynoscion nebulosus).
Both are gregarious bottom feeders found in diverse inshore habitats.
Although the two identified saltwater fishes could have entered
the freshwater environment in the vicinity of Butler Island, their
presence may suggest that the slaves were exploiting the more brackish
habitats of the delta. Evidently, fresh fish from brackish waters were
occasionally provided for the slaves.
This has been one of the hardest freshets in years. . .
The freshet destroyed much of their comforts. I have
-

181
been fortunate enough this week to give them a fine portion
of fresh fish. 4 hands took over 300 bass and drum at
least 4000 lbs., which when divided among them are smoked
and cured and will last for weeks. (Roswell King, Jr.,
14 July 1833}
The reference to both bass and drum in large quantities suggests that
fish were taken from brackish waters, if not marine waters.
It is unfortunate that no surely identifiable anadromous fish were
recognized. Anadromous fish, which enter the rivers during the spring
and summer months and move upstream to spawn, include herring and shad
(Clupeidae), stripped bass (Morona saxtil is), and the strugeons
(Acipenseridae) (Dahlberg 1975).
Assuming that the identified fish species can be generalized as
representative samples of fish taken and consumed by the Butler Island
slaves, a preference for fish found in nearby stagnant waters is sug
gested. It may be that convenience dictated this pattern rather than
an actual preference, since these fish were taken probably from the rice
canals, where they are frequently found today. At Cannon's Point, slaves
evidently exploited nearby species rather than those more distant (Otto
1975:345). The drum and sheepshead may simply represent fish occasion
ally taken from brackish waters or they could have made their way into
a-freshwater environment. Given the limitations of the collection, these
interpretations must remain tentative, but the sample does provide indi
cations that fish typically found in the delta were regularly exploited.
Turtles represent the greatest nondomestic food resource exploited.
The three turtle species identified are all semi aquatic species and are
all found in deltaic habitats.
The mud turtle (Kinosternon sp.), only 7.5-12.0 centimeters in
size, frequently inhabits quiet, slow-moving, shallow waters along

182
the south Atlantic coast. Two species are typically found in coastal
Georgia, the common mud turtle (K. subrubrutn) and the striped mud
turtle (K. bauri). Both are bottom dwellers, rarely seen at the water
line, and are fond of brackish 'waters (Ernst and Barbour 1972:50-59).
The genera of turtles which indues the cooters and sliders
(Chrysemys sp.) are found in freshwater lakes, ponds, and swamps. They
are gregarious and are often seen basking above the waterline on logs.
Their flesh is very palatable and is often eaten locally in the South
today (Ernst and Barbour 1972:157).
As a result of its distinctive marginal ridge on its carapace,
the Florida softshell turtle (Trionyx ferox) is easily recognized. It
occurs in all freshwater habitats but prefers deep waters with mud or
sand bottoms, and is occasionally found in brackish waters near the
mouths of streams (Ernst and Barbour 1972:266-272).
It is not surprising that turtles are well represented within
the faunal collection. All identified species could be easily procured
in the delta. Mud turtles will take a baited hook or they could have
been obtained with nets (Pope 1939:55), whereas the cooter and soft-
shell might have been collected on land as they were migrating or nest
ing (Larson 1970:26). A local informant who was around during the last
days of Georgia rice cultivation claims that "the softshell turtle and
terrapin were frequently caught by rice ditchers."* His observations
suggest that the prevalence of semi aquatic turtles in canals and ditches
would have made them easily obtainable food items for rice coast slaves.
^Rudolph Capers, Hofwyl Plantation Historic Site, Glynn County,
Georgia, personal communication, 1980.

133
Birds and mammals
A diving duck (Aythya sp.) was the only nondomestic bird identi
fied. Ducks were apparently taken by the Butler slaves, as the planta
tion records indicate, but the slaves' accessabi1ity to guns may have
regulated duck hunting activity. Fields of planted rice attracted
numerous predatory fowl (Austin 1893:34), and slaves were possibly en
couraged to capture them at times. The rice bird, or the bobolink,
(Dolichonyx oryzivorus) was paramount among these predatory birds. Per
haps its size (6 1/2 inches,or 16 centimeters; see Peterson 1947:209)
and the fact that the entire bird, both flesh and bone, was generally
consumed* have resulted in no archaeological evidence of this food item.
Only two nondomestic mammalian species were identified, raccoon
(Procyon lotor) and deer (Odocoileus virginianus). Both occur more fre
quently in the hammocks and salt marshes along,the coast than in the
delta but appear to adapt well to deltaic marshland. The raccoon with
its varied diet could thrive on frogs, insects, berries, seeds, and
eggs of birds and turtles found in the delta (Johnson et al. 1974:42,
85). The deer were possibly attracted to garden crops (Reitz 1979:78).
Traps could be set to capture the nocturnal raccoon while, presumably,
the deer were taken with guns.
The unidentified rodent (Rodentia) and rats and mice (Cricitidae)
as well as the snake (Squamata) were probably not consumed and for that
reason have not been included in either Table 13 or Table 16. In the
South Carolina rice coasts, however, slaves have been reported to have
consumed the rice rat (Oryzomys sp.), which they considered a delicacy
*Rudolph Capers, personal communication, 1980.

184
(Doar 1970:29). Whether or not the Butler Island slaves ate rats and
mice could not be determined from the food remains.
Summary of Food Resources
Despite the limitations of the faunal collection, it is possible
to offer some interpretations regarding slave diet at Butler Island,
particularly as these relate to the test postulates for this study.
The relative proportions of the domestic food remains were found
to be very similar to those recovered from Cannon's Point plantation.
This finding lends support to the hypotheses that similar patterns
existed in slave consumption of domestic food items throughout coastal
Georgia, at both cotton and rice plantations. Beef was found to be the
predominant food item at both plantations. Although beef was evidently
an important food resource at coastal plantations, the records for the
Butler estate indicate that pork was the major domestic animal food
supplied to the slaves. Assuming that the records are accurate, it may
be that the archaeological record preserves more evidence of beef con
sumption than of pork, particularly if pork was cured and the bony
parts removed and beef was served as freshmeat.
Plantation managment practices may have influenced domestic food
consumption. Because of the absence of a planter family at Butler
Island, it is possible that slaves received all parts of butchered
animals, whereas, in the case of a plantation with a resident planter,
slaves were perhaps more likely to receive the portions of an animal
that the planter did not want. The archaeological evidence is very
tentative, but the presence of cattle humeri ai: Butler Island and the
absence of such fragments at Cannon's Point do suggest that this may
have been the case.
... .

165
All of the nondomestic food resources identified are found in the
delta section of the tidewater in either brackish or freshwater habitats,
although, of course, many food resources not identified were possibly
also consumed. Therefore, it is difficult to ascertain which animal
species were utilized most frequently. At the same time, fishes and
turtles are among the dominant animal species found in the delta, and
the faunal collection suggests that these were the predominant non
domestic species consumed. Birds, which are greatly underrepresented
in the collection are also plentiful, but the slaves' ability to procure
them was dependent upon their access to firearms. Also, birds, par
ticularly ducks and other migratory fowl, are generally more seasonal
resources than fishes or turtles. Consequently, birds may not have been
consumed as often as the fishes and turtles. For the most part, mammalian
food resources are more restricted in the delta than in other habitats
of the coast. This may explain the infrequent presence of nondomestic
mammals within the collection.
In conclusion, the archaeological evidence of slave diet at
Butler Island and other coastal plantations indicates that domestic
food resources played a greater role in slave diet than nondomestic food
resources at both rice and long-staple cotton plantations. Yet the
presence of nondomestic food remains implies the need to supplement
monotonous and perhaps diffident plantation rations. Furthermore,
slaves supplemented their rations with food resources available to them
in nearby habitats.

Clothing
Few references to slave clothing viere found in the Butler planta
tion records. Cloth was generally purchased but sometimes was made on
the estate. 11 We are making our own cloth, the cotton has been spun and
Molly's son, James, put the loom up and we will weave it out" (Roswell
King, Sr., 15 November 1812). Woolen cloth and slave shoes were also
made from time to time. The only archaeological evidence for clothing
manufacture, however, was a pair of scissors recovered from structure
one.
Buttons
Metal buttons were the predominant clothing artifact, with brass
and whitemetal flat disc buttons being the-most frequently occurring,
particularly Type 7 (South 1964), or Type D (Olsen 1963), and Type 31
(South 1964). Of the total of 72 buttons recovered, 52 were of these
two types. The sizes of these buttons ranged from 10 to 24 millimeters,
which suggest that they were used for a variety of garments. Addi
tionally, the frequency of the flat disc buttons may indicate that
they were purchased specifically for slave clothing. In fact, refer
ences to "metal buttons" appear on the annual expense lists. The
occurrence of the remaining metal buttons was sporadic, possibly because
they came from clothing the slaves purchased, for themselves. These
metal buttons" include two Type 32 (South 1964) and one Type 9 (South
1964) with a hand-stamped design of a person on a horse.
Twelve porcelain buttons Type 23 (South 1964) were found. Al
though it has been suggested that porcelain buttons were used for
undergarments (MacFarlane 1975:135), it is more likely that they were

187
used for shirts and blouses. Nine of the porcelain buttons are plain
white. Two of the remaining three are known as "small china buttons,"
one with a. blue transfer print design known as "calico" and the other
with a painted blue-on-white design known as "bullseye) (Luscomb 1974:
23-24)(see Figure 34). These may have been used specifically on
female apparel. A dark green porcelain button identical, to the plain
white ones was also found.
Bone buttons form the third class. These appear to have been
made by machine and not on the plantation. Three 5-hole Type 19 (South
1964),or Type J (Olsen 1963),and one 4-hole Type 20 (South 1S64) were
recovered. Bone buttons were strictly utilitarian end were most fre
quently found on underwear and trousers (Luscomb 1975:26).
Identical button styles have been found in all the previously men
tioned slave contexts but in different frequenciespossibly due to varia
tions in sample sizes. But the predominance of metal flat disc buttons
at Butler Island seems to suggest that they were purchased specifically
for slave clothing, possibly for both males and females. On smaller
plantations, it would seem that slaves were more apt to receive hand-me-
down clothing from the planter and his family. Yet the si mi 1iarities in
button styles indicate that slave clothing was virtually the same in
diverse plantation contexts.
Footwear
Fortunately, the preservation conditions at Butler Island have
resulted in archaeological evidence of slave footwear. Slaves, both
men and women, were provided with work shoes known as "brogans." (Stampp
1956:291). Three complete or nearly complete shoe soles and numerous

Figure 34. Clothing and Personal Adornments.
Designations
A.
Flat disc metal
button
B.
Bone buttons
C.
Porcelain button
is (bul
transfer print,
green)
D.
Glass beads
E.
Hematite bead

189
100 110 120 130 140 150 160

190
smaller sole fragments viere uncovered. For this reason, it has been
estimated that four shoes were represented within the collection. The
soles all appear to have been stitched rather than nailed (see Anderson
1968:56-65). Thread fragments are still apparent in some of the stitch
holes (Figure 35). Because they seem to be machine stitched, these are
most likely the remains of purchased shoes and not homemade items.
Accessories and Ornaments
Beads
Of ail the artifacts within this functional class, glass beads
were the most numerous. Blue, faceted, hexagonal beads found in several
slave contexts in coastal Georgia and recently in Tennessee (Smith 1977:
157) predominated. In addition to 16 blue beads, 15 beads of the same
style but in various colors viere found, including green, purple, orange,
red, and brown. Besides being found at slave sites, these cane beads
"occur in great quantities" at eighteenth- and nineteenth-century
Seminole Indian grave sites (Fairbanks 1974:90).
Three round, wire wound beads (Kidd 1970; type Wlbl7, Wlbl5,
Wlh!2)(see Figure 34) were also collected. These beads, though of a
style not reported at any other slave site, are probably contemporaneous
with the faceted beads. Both bead styles were possibly slave pur
chases.
The most curious bead, origin unknown, is one made of hematite
(Figure 34). While it may possibly have been brought from Africa or
made by the slaves, it may simply be an item taken from a prehistoric
site.*
*Jerald T. Milanich, Florida State Museum, Gainesville, Florida,
personal communication, 1980.

Figure 35. Shoe Soles.

1 9?.

193
Mls cel 1aneous I terns
Included within miscellaneous items are a brass object, possibly
a parasol attachment (Figure 36); a glass lens (Figure 37); a hard
rubber tooth, presumably of a comb; and a bone handled pocket knife.
The brass object appears to have been part of the apparatus used
to open and close an umbrella belonging to a female. This attachment
and the little china buttons were uncovered from structure one, which
provided convincing evidence of a female's presence at that household.
The glass lens is likely to have been part of a pair of glass
spectacles, but its magnification is presently very slight. If not an
eye piece, it could have been a lens for an unknown object.
The size or type of comb represented by the uncovered hard rubber
tooth could not be ascertained, but it provides evidence that the
Butler Island slaves had a few amenities.
The pocket knive is the only personal possession that has also
been uncovered at other slave sites. Its occurrence at Butler Island
and other slave sites may indicate that pocket knives were typically
owned by slaves.
Summary
The previously described personal items and those mentioned else
where in this study are very significant. Whether these items were
obtained through purchase, stealing, or gift, they indicate that slaves
made efforts to improve upon their material conditions. Assuming that
the items were obtained through purchase, the task labor system is
likely to have been a crucial factor for making the purchases possible.

Figure 36. Parasol Attachmen


Figure.37. Eye Lens.

197
" . >
'MMTfc
r
tkeas
j

198
Household Items
Items such as chamber pots, a pewter spoon, and furnishings are
designated household items. Only one artifact suggestive of furniture
was recovered, a brass drawer pull. Generally, the remains of furni
ture. at slave sites have been very scant, implying that furniture was
a rare item in slave dwellings.
Tobacco Equipment
Tobacco pipes were the only leisure-activity artifact found, there
being no toys or personal items related to leisure recovered. Numeri
cally, pipes constituted the third highest artifact group in the Butler
Island collection. A total of 642 pipe stems and bowls were collected,
a number considerably higher than that found at other slavery sites (see
Chapter 6). This may indicate that tobacco and pipes were periodically
supplied to the Butler Island slaves and that these commodities were not
supplied to slaves at the other sites. Both tobacco and "Negro pipes"
are listed on some of the annual expense lists. Like ceramics and
alcoholic beverages, tobacco and pipes may have been occasionally pro
vided to the slaves, with the slaves making additional purchases to sup
plement the provisions. Here again, the close proximity of Butler
Island to Darien may have made tobacco and pipes more easily obtainable
than to slaves on the barrier islands.
Most pipes appear to be of British or Dutch manufacture. No
American made pipes 'were identified. In fact, most pipes appear to be
Dutch. The recovery of a Peter Dori pipe indicates that the site was
occupied after 1850 (Humphrey 1969:15). Unfortunately, few of the
other pipes could be as precisely dated. Fluted bowls were well

199
represented and several varieties were noted. A fragment of a "TD"
pipe (Humphrey 1969:14; Hanson 1971:92) was identified.
All pipes were white clay with the exception of one which was
brown and white porcelain. The latter was the only pipe with a de
tachable stem pipe bowl. It was presumably of German manufacture and
would have been a fairly expensive item for a slave to have owned (see
Figure 38).
Farming Tools and Specialized Crafts
All of the tools recovered were those to be expected at a rice
plantation. Fragments of six eye hoes and one rice hook were the only
farming tools, but a considerable number of woodworking tools were
present, particularly at structure one. Such woodworking tools are
also characteristic of rice production. The wooden tierces used for
shipping rice were generally made at the plantation.
Included among the woodworking tooks are an adze, an axe, a
rounded chisel or gouge (see Figure 39), a number of smaller chisels,
and a saw knife blade. Adzes and axes are general woodworking tools,
whereas the chisels and saw knife are more specialized tools. The
rounded chisel was used to make holes and the saw knife for shaping
rough wood (Sloans 1964:54, 84). Perhaps the abundance of these wood
working tools at structure one indicates that the structure was oc
cupied by a cooper and his family. This may explain the presence of
some of the "luxury" items. Coopers, like other plantation specialists,
viere often able to acquire extra cash through hiring themselves out or
by building items which they sold for cash (Kemble 1961 (1863);65).
With the extra cash they were able to acquire a standard of living
slightly above the field hand.

Figure 38.
Porcelain Pipe Bowl.


201

Figure 39. Woodworking Tools.
Key to Artifacts
1. Adze
2. Axe
3. Rounded chisel or gouge

203

204
Two tools, the function of which could not be determined, are
shown in Figure 40. The first item is a thick U-shaped object. The
other appears to have been used as a mold for making small rounded ob
jects, possibly buckshot or fish net sinkers, but the function of the
screw holes on each side is unknown.
It is interesting that most of the tools and implements viere
uncovered within the house. At structure one the distribution of some
of the tools has been plotted (see Figure 22). The presence of tools
within the slave cabin structure strongly suggests that slave dwell
ings doubled as draft workshops and tool sheds. Apparently, tool sheds
were not available at the site.
Although woodworking tools were identified, no carved wooden ob
jects were found. Thus far, very little evidence of slave crafts has
been revealed archaeologically in coastal Georgia, yet the area is well
known for its postbellum Afro-American crafts (Georgia Writers Project
1940; Vlach 1978). There are, however, some items at Butler Island
which may be suggestive of slave craft activity, for example, the
carved bone fragments. Also, a hugh hand-hewn cypress log found close
to settlement #4 was possibly smoothed down for the construction of a
canoe or other object. The log was approximately 25 feet long (7.47
meters) and was found in the vicinity of the old antebellum dike. It
would be difficult to demonstrate that this log was intended to be
a canoe, but it appears to have been hand-hewn, and someone had begun
hollowing out the inside. Assuming that this is an antebellum artifact,
it is likely that slaves were responsible for it.
More convincing evidence of slave crafts is provided by the
occurrence of three fragments of a crudely made ceramic (Figure 41).
£
t.

Figure 40. Unidentified Tools.
Designations
1. U-shaped object
2. Mold for lead shot

X ~
>
ii;
WM0%

Figure 41
Slave-Made Cerami

jrfTTr^ft
" 'Tf
TO 100 110
I20 100 140 150 1G0
ilU
r
¡ n c n c :
no
o
CD
....-ys'Y.fl'ii;-.'.; .tv-.- V

209
The colono-ceramics which have been recently suggested to have been
made by Afro-Americans (see page 12) are frequently found at black
occupied sites in coastal South Carolina but have been nonexistent in
Georgia. The ceramics recovered from Butler Island, however, do not
resemble the sand and grit tempered colono-wares at all. Instead, they
are very reminiscent of the St. Johns series (Goggin 1952) and other
chalky wares in texture. It is possible that these are later manifesta
tions of St. Johns, but the sherd thickness, as well as the vessel form,
is strikingly different from the known St. Johns series. The vessel is
suggestive of a shallow bowl and the makers seem to have been trying to
imitate that European ceramic form. It is very likely that the slaves
at Butler Island experimented with the manufacture of ceramics. Both
the abundance of clay and the fact that clay bricks were regularly made
there are conditions which would have prompted ceramic making among the
slaves, assuming that a potter was among them. The manufacture of these
chalky textured ceramics may have been another way in which the Butler
Island slaves utilized the resources of their deltaic environment. It
is, however, difficult to draw conclusive evidence from what appears
to be one vessel. Thus, the suggestion that these ceramics were manu
factured by slaves remains an untested assumption until further evi
dence is presented.
Summary
Most of the artifacts of slavery recovered from Butler Island
are very similar to those found at long-staple cotton slave sites.
Differences exist, particularly with regards to items directly related
to rice production or the habitat of rice production (the delta).

210
Additionally, subtle differences such as higher frequencies of buttons,
beads, and pipes may reflect differences in management practices or the
fact that the Butler Island slaves were closer to local merchants than
were slaves on the barrier islands. On the whole, the similarities in
the artifact patterns are overwhelming, and it is suggested that these
are regional characteristics in slave material culture.

CHAPTER 6
PATTERN RECOGNITION IN SLAVE MATERIAL CULTURE
To demonstrate further that slavery assemblages from sites in
coastal Georgia display regional characteristics, the artifact groups
of four slavery sites were quantified and compared. Specifically, the
artifact frequencies for each slave assemblage were calculated and the
percentage range compared. Initially, it was intended that the examina
tion of artifact assemblages would include all of the sites indicated
in Figure 1. Unfortunately, the artifact data for some of these sites
are not available.* The sites used in this comparison include the pre
viously mentioned sites: Cannon's Point south cabins (MacFarlane 1975),
Cannon's Point north slave cabin (Otto 1975, 1977), Kingsley Plantation
(Fairbanks 1974), and Butler Island. Although Kingsley Plantation is
technically in Florida, it is located on a coastal barrier island and
it produced long-staple cotton.
Another reason for the selection of these three plantation con
texts is their differences in economic functions and plantation manage
ment. It h.as been pointed out that Butler Island was a rice plantation
and the other two long-staple cotton plantations. Each plantation,
however, was apparently managed very differently: Cannon's Point had a
*The analyses of Butler Point and Sinclair are presently in
process (Mullins-Mopre nd). Artifact counts were not published in the
report of the Rayfield plantation (Ascher and Fairbanks 1971). The
LeConte-Woodmanston plantation did not provide a slavery occupation
(Hamilton 1980).
211

212
resident planter, and white overseers managed the estate; Butler Island
had an absentee landowner; and Kingsley had a resident planter, and
black drivers presumably managed the estate. Also, Kingsley was known
to have been a slave importing station (Fairbanks 1974:63),which adds
another dimension to its economic function. All three plantations had
in common slave labor forces which exceeded 100 individuals and which
were located in a coastal environment.
Possibly, the diversity of these plantation contexts has resulted
in variations within specific artifact groups. Some of these were
noted in Chapter 5, but the quantitative differences in artifact frequen
cies were not examined. Artifact pattern recognition attempts to
examine frequency variations in artifacts from within intrasite or inter
site contexts (South 1977:99-102). Through such examinations patterned
similarities and differences may be delineated (South 1378:223). These
patterns presumably reflect the functional and behavioral processes
which Have taken place at sites. Thus artifact similarities occurring
within slavery contexts are suggested here to reflect general patterns
in coastal slave lifeways. Differences, on the other hand, may be
related to functional characteristics of specific plantations.
South defined eight artifact groups for the definition of arti
fact patterns (1977:92-102). These include architecture, kitchen,
arms, clothing, personal, furniture, tobacco, and activities. These
groups are essentially the same as those used in Chapters 4 and 5, but
there are differences which include, first, furniture and arms have
been assigned to separate groups; second, glass beads are included with
clothing; and, third, farming equipment, specialized crafts, and
l
r;-'

213
fishing equipment have been assigned to activities. South's categories
have been used because they may be more insightful for assessing the
importance of personal artifacts, furniture, and arms. Also, the slave
artifact pattern could later be compared with other ethnic artifact
patterns.
Each site, however, was excavated by different investigators, and
collection techniques varied. Also, artifact analyses were conducted
by a variety of individuals. Unfortunately, very little could be done
to correct for these difficulties, and it must be assumed that these
differences have not severely altered the artifact frequencies. Most
artifact counts were obtained from published reports. Occasionally,
however, artifacts needed to complete the artifact groups were not quan
tified. In these cases, the author examined the original classification
cards, all of which are located at the Florida State Museum.
Table 16 provided the artifact profile for each site. It reveals
that the percentage range for Butler Island and the two Cannon's Point
slave sites are very similar for most artifact groups. The Kingsley
profile is very dissimilar with regard to architecture and kitchen
groups. The raw artifact count for Kingsley's architecture group is
considerably lower than that of the other three sites. This is pos
sibly related to the fact that nails constitute a large portion of the
artifact group at Butler Island and Cannon's Point, where slave dwell
ings were frame constructions. The nail frequencies at both Butler
Island and Cannon's Point are considerably higher than at Kingsley.
The dwellings at Kingsley were made from poured tabby and this explains
why few nails were'found.

Table 16
Artifact Profiles for Four Slavery Sites in Georgia/Florida
Butler
Island
Cannon'
s Point3 S
Cannon'
s Point, N
Ki ngsley
Artifact Group
#
Of
10
it
%
#
7o
#
%
Architecture
4494
67.90
3824
71.38
3789
70.60
754
34.34
Kitohen
¡325
20.01
1388
25.91
1383
25.77
1384
62.97
Furniture
1
.01
5
.09

0.00

0.00
Guns
15
.23
2
.04
6
.11
10
.45
Clothing
111
1.68
45
.84
67
1.25
18
.82
Personal
5
.08
9
;i7
3
.05
5
.23
Tobacco
642
9.70
71
1.33
107
2.00
15
.68
Activities
26
.39
13
.24
12
.22
12
. 55
Total
6619
100.00
5357
100.00
5397
100.00
2198
100.00
214

215
To correct for the difference in the architectural frequency of
Kingsley, a correction factor was added to this artifact group. The
corrected artifact count and frequency for the architectural group in
dicated in Table 17 were derived by adding approximately the number of
nails recovered, if the Kingsley structure had been frame, to the
actual architectural count indicated in Table 16. The approximation for
the number of nails recovered from a frame dwelling was derived from
taking the average number of nails recovered from the three slave sites
with frame structures. The adjusted artifact profile for the Kingsley
plantation is given below. Admittedly, the adjustment of the Kingsley
Table 17
Adjusted Artifact Profile for Kingsley Plantation
Artifact Group
Artifact Count
Frequency
Architecture
3950
73.23a
Ki tchen
1385
25.66
Furniture

0.00
Clothing
18
.34
Guns
10
.18
Personal
5
.09
Tobacco
15
.28
Activities
12
. 22
ocal
5394
ICO.00
Corrected
artifact count and frequency.

216
artifact profile by the addition of the correction factor may not be
the best procedure to correct for the disproportionate nail frequencies.
South makes similar adjustments to his Carolina pattern (1977:104-106).
And herein lies the major defect with the artifact pattern recognition:
It offers no real solution to the problem of extraordinary dispropor
tionate artifact counts (Chance 1977:127).
Derived from the adjusted Kingsley profile and the artifact pro
files from the other sites in Table 17, a slave artifact pattern is sug
gested, as indicated in Table 18.
Table 18
The Slave Artifact Pattern in Coastal Georgia/Florida
Artifact Group
Mean Percent
Percentage/Range
Architecture
70.78
67.90-73.23
Kitchen
24.34
20.01-25.77
Furniture
.02
.00-00.09
Guns
.14
.04.00.23
Clothing
1.03
.34-01.68
Personal
.09
.05-00.17
Tobacco
3.32
.28-09.70
Activities
.28
.22-00.39
Total
100.00
The slave artifact f
pattern reveals that the arch'
itecture group is
the overwhelming group of the artifact pattern. This is evidently

directly related to the vast amount of nails uncovered from the frame
slave dwellings. Also, it indicates that slave cabins have been the
focus of previous excavations. Perhaps, if more trash or midden deposits
were excavated, this artifact group would not dominate the profile. It
also strongly suggests that houses were the material aspect of slave life,
at least as seen archaeologically.
High proportions of kitchen artifacts have again been indicated.
The proportion of this artifact group within the total slave assem
blages further supports that the cabins were central to slave cooking and
eating activities.
Furniture and personal possessions are very scarce. Although
slaves may have been able to improve their material lots as a result of
the task system, luxury items such as furniture and personal possessions
were still hard to come by.
Firearms were represented at all the sites, but occupy a small
proportion of the total assemblages. Perhaps this indicates that slaves
had very limited access to guns.
Tobacco has the widest percentage range of all the artifact groups.
As previously stated, this possibly reflects whether or not slaves re
ceived provisions of tobacco. At Butler Island, tobacco and pipes were
provided occasionally.
Clothing is somewhat variable and could possibly be related to
status of the slave occupants at the various sites. Presumably, a high
status slave would have more clothing than a common field laborer. It
is difficult to make such an
assessment based upon such scanty evidence.
On the other hand, the activities group indicated very little variation.

218
among the sites. Its percentage range suggests that few artifacts
related to slave craft activity or specialized farming equipment have
been uncovered.
The slave artifact pattern suggested here is very tentative as it
only relies upon data taken from four sites. Differences in collection
techniques as well as artifact analyses may have skewed artifact fre
quencies. In spite of these difficulties and problems inherent in the
pattern recognition method, it seems that artifact pattern recognition
can be a useful tool in making interpretations regarding Afro-American
slavery subsistence patterns. Hopefully, future pattern recognition
studies in coastal Georgia will refine this suggested regional artifact
pattern.

CHAPTER 7
SUMMARY AND CONCLUSIONS
The Butler Island data have suggested that the adaptations of
slavery to rice culture are archaeologically discernible in slave com
munity organizations, in the natural resources exploited, and to a
lesser extent in farming implements, tools, and slave crafts. These
patterns are presumably adaptations to the deltaic habitat in which rice
is produced.
On the other hand, the archaeological correlates for slave cloth
ing, plantation food rations, personal possessions, and leisure activi
ties were found to be similar to those found at long-staple cotton plan
tations in coastal Georgia. Subtle qualitative and quantitative varia
tions exist between the Butler Island data and those from other sites, but
the variations may be related to differences in plantation operation.
While it is possible that these slave material conditions existed
in other cash crop regions of the Old South, it is important to remember
that several factors in coastal Georgia favored a distinctive develop
ment in slave lifeways. Paramount among these factors were local ecologi
cal factors. The abundant natural resources of the coast, both in the
marsh and lagoon section and on the delta, were exploited for numerous
subsistence needs. Additionally, cultivation techniques and labor manage
rnent systems were adaptations to the coastal environment. Specifically,
the task labor system developed from polderland construction of the
219

220
tidal marsh. Small fields known as "tasks" became the unit cf measure
for certain labor requirements in the cultivation of tidewater staples.
This labor system, which provided slaves with the incentive to improve
their material standard, was later employed in other cash crop regions
throughout the South. Its origin, however, is clearly an adaptation to
coastal habitats. Lush ecological conditions like those of the coast
were not present in interior regions of the Old South, and the absence of
these conditions in the interior was likely to have been reflected in
slave material conditions. It is, however, possible that slave life-
ways were similar in all coastal areas of the South, but this assumption
will have tc be demonstrated archaeologically.
Slave demography along the coast was another factor. The high
concentration of slaves in coastal areas was again an adaptation to
the production of coastal cultigens. The total reliance upon hand labor
required vast amounts of slave labor. It may have been that in "black
belt" areas slaves were given a few privileges, such as the use of
firearms, that slaves in other areas were forbidden. This assumption
remains untested, but it does seem that coastal slaves, particularly
those on rice plantations, had at least the opportunities to manipulate
the system during the periodic absences of whites.
Historic and economic factors were also found to have been very
important. Georgia was settled relatively late by planters when com
pared with other tidewater areas. When the plantation system was just
beginning in Georgia, planters from the older tidewater areas were
suffering from rapidly depleting soils. Many of these planters came to
the virgin lands of Georgia as a result. Major Pierce Butler was one

of these immigrants, and he became one of the wealthiest slaveholders
in the South. In fact, the Georgia coast became known as a region of
wealthy planters. Possibly, this wealth affected the quality of slave
life. At least this seems to have been the case with the Butler estate.
The Butler slaves are documented to have received items (alcohol, pipes,
tobacco, and ceramics) which were apparently not supplied to slaves at
many other plantations.
Regional interpretations of slave lifeways have been offered for
the entire coastal Georgia area. Yet variations are likely to have
existed. Variations resulting from habitat specific differences have
been discussed. The proximity of plantations to towns or cities may
also have been an influential condition. It has been suggested that
the higher frequences at Butler Island of beads, pipes, and stylish
ceramics may have been related to the fact that the Butler Island slaves
were able to trade more often with nearby merchants than were slaves on th
barrier islands. Slave life styles in towns and cities have been docu
mented to be very dissimilar to those on rural plantations (see Miller
and Genovese 1974:337-452; articles on slavery in town and city). This
finding, however, awaits archaeological investigation. In coastal
Georgia, no urban or town slave sites have been archaeologically inves
tigated, and none of the urban-orientated Savannah River plantations
has been excavated. Therefore, it is not known whether the regional
interpretations offered here extend to those situations. At the
southern extreme of the coast, variations to the ceramic patterns dis
cussed in Chapter 5 have already been noted in the St. Mary's River

2.22
district.* Perhaps these variations are related to the fact that planta
tions in Camden County or along the St. Mary's River were exposed to
more of a frontier environment than those of the Altamaha estuary. The
St. Mary's district was settled considerably later and it is further
away from the port cities of Savannah and Charleston than is the Altamaha
district. It is possible that gradations in slave material conditions
existed from the more urban-centered plantations along the coast to
more rural ones. Thus, even along the narrow belt of the Georgia coast
variations in slave lifeways existed.
Recently, it has been suggested that the unit of analysis for the
study of slavery be limited to the county (Otto 1979). The county may in
fact be the maximal areal extent for any regional interpretation of
slavery. Still it has been possible here to tentatively suggest that
regularities existed in slave life which exceeded both habitat specific
influences or county lines. To suggest that these patterns extend beyond
the limits of the Georgia tidewater needs to be investigated. Hope
fully, future regional studies of slavery in Georgia and elsewhere will
help to refine the suggestions offered here.
Perhaps the greatest contribution of the Butler Island study is
that it has further demonstrated that archaeological data can provide
information not available in historic documents (see Fairbanks 1977).
Butler Island is a very wel1-documented historic site. In many ways
the documentary resources outshine the very limited archaeological data
recovered. Yet, in spite of the limitations of the archaeology, a
*Chad Bradley, Assistant Archaeologist, Kings Bay Project of
Anthropology, University of Florida, Gainesville, personal communica
tion, 1980.

223
number of findings were uncovered that were not only absent in the docu
ments, but also, in some cases, were contrary to them. Clearly, archaeo
logical data need to be incorporated with historical sources to investi
gate the slave subsistence patterns. It may be that slave artifact
pattern recognition should begin with the establishment of artifact pro
files from well-documented sites. Possibly, the artifact profile as well
as the community organizations at Butler Island can serve as models for
future investigations of rice coast slave sites.

APPENDIX 1
SUMMARY OF SOIL ANALYSIS FROM SETTLEMENT #4, BUTLER ISLAND*
Provenance
Zone A (Figure 15)
Zone B (Figure 15)
Zone C (Figure 15)
Structure six
Drainage ditch
(adjacent to structure four)
10YR 5/1** (grey) with/7.5YR 5/6
(strong brown) flecks; silt loam; crumb
structure; slightly hard dry, friable
moist, slightly sticky wet (soil acidity)
4.9; P (phorphorus) content 21 ppm
(parts per mi Ilion).
1OYR 5/1 (grey) with/7.5YR 6/8 (redish
yellow) flecks and charcoal; clay loam;
blocky structure; hard dry, firm moist,
slightly sticky wet; pH 5.1; P content
38 ppm.
10YR 6/1 (grey) with/7.5YR 5/8 (strong
brown) mottles; silty clay; massive
structure; very dry, very firm moist,
sticky wet; pH 5.3; P content 27 ppm;
contains, old cypress roots.
1OYR 3/2 (very dark grey brown); loamy
coarse sand;'massive, cemented with
iron oxide and organic matter (organic
matter 12%); hard dry, friable moist,
non-sticky; pH 5.9; P content 14 ppm;
spodic horizon (iron hard pan) with
old roots that have largely decayed or
old earthen floor with decayed straw.
7.5YR 3/2 (dark brown) with/7.5R 3/8
(dark red) pigment chunks (possibly old
paint pigment); silty clay; massive
structure; slightly hard dry, friable
moist, slightly sticky wet; pH 5.5;
P content 135 ppm; organic matter 20%;
high P and organic matter suggest old
privy but test for human colioform
bacteria was negative.
*Kobert F. Fisher, School of Forest Resources and Conservation,
University of Florida, Gainesville, written communication.
**Moist Munsell colors.
224

APPENDIX 2
SLAVE SUBSISTENCE ARTIFACTS FROM SETTLEMENT 4
Artifact Class
it of Items
Building hardward
Nails and spikes
Hinges
Pintle
Padlocks
Window glass
Total
4468
3
1
c
17
4494
Farming tools and equipment
Hoes
Chisels
Gouge or rounded chisel
Saw knife blade
Axe head
Adze
Fil e
Sickle
Unidentified objects (Figure 40)
Total
Food procurement equipment
Gun flints
Buckshots and musket balls
Trigger guard
Fish net sinkers
Total
6
5
1
1
1
1
1
1
2
19
6
8
1
4
Food preparation and serving equipment
Kettle legs 2
Grinding stone 5
Spoons
Pewter 1
Iron 2
Whitemetal handle
225

226
Artifact Class
# of Items
Cutlery
Bone handles
Knife blades
Ceramics
Coarse red earthenwares
Unglazed
Black lead glazed
Brown lead glazed
Colono-ceramics (slave made)
Refined earthenwares
Unclassified white
Whiteware, undecorated
Transfer print, pearl ware
Edged ware
Shell edged pearl ware
Molded pearl ware
Handpainted
Pearl ware undergalzed blue on white
Pearlware polychrome
Whitewa re polychrome
Annular
Circular bands
Pearlware
Whiteware
Yellow paste
Rouletted
Pearlware
Whiteware
Finger painted
Pearlware
Whiteware
4
2
2
13
13
298
217
17
48
11
26
11
87
28
2
2
2
1
1
Mocha
Whiteware
Yellow paste
Marbelized
Sponge decorated
Stonewares
Alkaline glazed
Brown salt glazed
Unidentitied
9
1
2
2
22
15
10
Porcelain
1
Unidentifi ed
l
Total ceramics
863
Glassware
Dark green
317
Light green
35

Artifact Class
# of Items
Clear
Bottle
51
Tableware
5
Aqua
40
Total glassware
448
Clothing, adornments, and personal possessions
Buttons
Porcelain
12
Bone
4
Brass
56
Beads, glass
Hexagonal, faceted
32
Roung, wire-wound
3
Beads, hematite
1
Total
35
Miscellaneous items
Shoe soles (reconstructured)
4
Parasol attachment
1
Comb tooth
1
Pocket knife
E
Scissors
1
Tobacco equipment
White clay pipes
Stems
236
Bowl s
Plai n
154
Decorated
251
Porcelain
1
Total tobacco equipment
642
Whole
82
Fragments
62
Furniture
Brass drawer pull
i
',* rw.iv,.'t-.'o#**

ABBREVIATIONS USED IN REFERENCES
BFPC
Butler Family Papers Collection #1447
GHS
Georgia Historical Society, Savannah, Georgia
MDCC
Margaret Davis Cate Collection
PHS
Pennsylvania Historical Society, Philadelphia, Pennsylvania
WFPC
Wister Family Papers Collection (Butler Section)
228

REFERENCES
Adams, Robert McC.
1963 Archaeological research strategies: Past and present. Science
160:1187-1192.
Alderi, John R.
1971 The First South. Louisiana State University Press, Batton Rouge.
Allston, R. F. W.
1846 Rice. DeBows Review 1:320-357.
1854 The essay on sea coast crops. DeBows Review 16:589-615.
1858 Notes on the management of a southern rice estate. DeBows
Review 24:324-326.
Anderson, Adrienne
1968 The archaeology of mass-produced footwear. Historical
Archaeology 2:56-65.
Anonymous
1S65? History of Altama Plantation, 1763-1965. (Pamphlet, no
copyright).
nd Untitled manuscript. On file, Altamaha Waterfowl Management
Area Office, Butler Island, Georgia.
Anthony, Carl
1976 The big house and slave quarters, part I: Prelude to
New World architecture. Landscape 21(1):8-l9.
Armstrong, Margaret
1S33 Fanny Kemble: A Passionate Victorian. Macmillan, New York.
Ascher, Robert
1574 Tin can archaeology. Historical Archaeology 8:7-17.
Ascher, Robert and Charles H, Fairbanks
1971 Excavation of a slave cabin: Georgia, U.S.A. Historical
Archaeology 5:3-17.
Austin, Amory
1893 Rice: Its cultivation, production and distribution in the
United States and foreign countries. U.S. Department of
Agriculture Bulletin #6. Government Printing Office, Washington,
d.c:
229

230
Baker, Vernon G.
1978 Historical archaeology at Black Lucys garden, Andover,
Massachusetts: Ceramics from the site of a 19th century
Afro-American. Papers of the Robert S. Peabody Foundation for
Archaeology #8. Phillips Academy, Andover, Massachusetts.
1580 Archaeological visibility of Afro-American culture: An example
from Black Lucy's garden, Andover, Massachusetts. In Archaeo
logical Perceptives on Ethnicity in America, edited by Robert
Schuyler, pp. 29-37. Baywood Publishing, Farmingdale, New York.
Barker, Sandford W.
1841- Ledger. Barker Plantation Records. (Experiments on Rice
1844 Culture). South Carolina Historical Society Manuscript
Collection, Charleston, South Carolina.
Bassett, John S.
1968 Southern Plantation Overseer as Revealed in His Letters. Negro
Universities Press, New York.
Berry, Brian
1964 Approaches to regional analysis: A synthesis. Annals of the
Association of American Geographers 54:2-11.
1968 A synthesis of formal and functional regions using a general
field theory of spatial behavior. In Spatial Analysis: A
Reader in Statistical Geography, edited by Brian J. L. Berry
and Duane F. Marble, pp. 419-430. Prentice-Hall, Englewood
Cliffs, New Jersey.
Binford, Lewis R.
1964 A consideration of archaeological research design. American
Antiquity 29:425-441.
Blassingame, John W.
1972 The Slave Community: Plantation Life in the_Antebel1um South.
Oxford University Press, New York.
1976 Status and social structure in the slave community: Evidence
from new sources. In Perceptives and Irony in American
SI avery, edited by Harry P. Owens, pp. 137-152. University
of Mississippi Press, Jackson.
Boney, F. N.
1977 Part three: 1820-1865. In A History of Georgia, edited by
Kenneth Coleman, pp. 129-187. University of Georgia, Athens.
Bonner,
1964
James C.
A History of Georgia Agriculture. 1732-1860.
Georgia Press, Athens.
University of

231
Booth, Sally S.
1971 Hung, Strung, & Potted: A His to ry of Eating in Co1onial
America. Crown Publishers, New York.
Brewster, Lawrence F.
1970 Summer Migrations and Resorts of the South Carolina Low
Country Planters. AMS Press, New York.
Bridenbaugh, Carl
1952 Myths and Real i11es: Socleties of the Colonial South.
Louisiana State University Press, Baton Rouge.
Brown, Harry B. and Jacob 0. Ware
1953 Cotton. McGraw-Hill, New York.
Butler Estate Papers
1799 Copy of taxable property (McIntosh and Glynn Counties, Georgia).
BFPC, box 3S folder 2, PHS.
1301- Slave transactions. BFPC, box 3, folder 5, PHS.
1803
1803 List of ditchers at Tide Island, BFPC, box 3, folder 3, PHS.
1803- Crop and livestock reports, BFPC, box 7, folder 4, PHS.
1830
1811 Annual expense list, BFPC, box 9, folder 7, PHS.
1314 Crop proceeds since 1801, BFPC, box 9, folder 10, PHS.
1815a Inventory of the Butler Estate, WFPC, box 37, folder 8, PHS.
1815b Provisions for Negroes, BFPC, box 3, folder 3, PHS.
1821 List of field hands, BFPC, box 3, folder 3, PHS.
1824- Statement of crops. BFPC, box 11, folder 5, PHS.
1835
1859 A list of slaves belonging to John Butler, WFPC, box 41,
folder 5, PHS.
Butler, Pierce, Jr.
1851- Ledge>'\ WFPC, loose, PHS.
1855
Caldwell, Sheila
nd Report on Butler Island burial site. Report on file, Historic
Preservation Section, Georgia Department of Natural Resources,
Al tonta, Georgia.
Calloway, James E.
1943 Early Settlement of Georgia.. University of Georgia Press, Athen
; i ' ; -"v - : m*. mmmmmmvm

232
Cate, Margaret D.
1925 Photographs. (Butler Island slave cabins, rice mill chimney,
tidal mill and boat landing, overseer's house). MDCC,
folder 206, GHS.
1960 Mistakes in Fanny Kemble's Georgia Journal. Georgia Histori
cal Quarterly 44(1):1-17.
Cate, Margaret D. and 0. Wrightman
1955 Early Days of Coastal Georgia. Gallery Press, New York.
Cathey, Cornelius
1956 Agricultural Developments in North Carolina, 1783-1820.
University of North Carolina Press, Chapel Hill.
Chance, David H.
1977 South: Method and theory in historical archaeology (a review).
Historical Archaeology 11:126-129.
Childs, St. Julien Pxavenel
1940 Malaria and Colonization in the Carolina Low Country, 1526-1696.
John Hopkins Press, Baltimore, Maryland.
Clark, J. D., H. G. Baker, and W. B, Morgan
1562 The history of sub-Saharan food crops. Journal of African
History 3:211-239.
Clifton, James
1978 Life and Labor on Argyle Island: Letters and Documents of
a Savannah River Rice Plantation-, 1833-1867. Beehive Press,
Savannah, Georgia.
Cole, Arthur H.
1927
The American
advantages.
rice growing industry: A study of comparative
Quarterly Journal of Economics 41:495-643.
Coleman,
1976
t
Kenneth
Colonial Georgia: A History.
Charles Scribner's Sons, New
York.
1977
(Editor) A \
Athens.
listory of Georgia.
University of Georgia Press,
ombes, John D.
1974 Ethnography, archaeology, and burial practices among coastal
South Carolina blacks. Conference on Historic Site Archaeciocjy
Papers 7:52-61.
Michael F.
A preliminary study of the planter aristocracy as a folk level
of study in the Old South. Unpublished H.A. Thesis. Univer
sity of North Carolina, Chapel Hill.
1941

233
Cornwall, I. W.
1958 Soils for the Archaeologists. Phoenix House, London.
Corry, Joseph
1960 Observations Upon the Windward Coast of Africa Made in the
Years 1305 and 1806. Franklin Cass, London.
Coulter, Merton E.
1941 Thomas Spalding of Sapelo. University of Louisiana Press,
Baton Rouge.
Couper, John
1818 Manuscript. (Letter to Major Butler). WFPC, box 31, folder 7,
PHS.
Crook, Morgan R., Jr.
1978 Mississippian Period Community Organizations on the Georgia
Coast. Ph.D. Dissertation. University of Florida, Gainesville.
Crosby, Constance and Mathew C. Emerson
1979 Identifying Afro-American mortuary custom: An example from
the Parting Ways Settlement, Plymouth, Massachusetts. Paper
delivered at the 12th annual meeting of the Society for
Historical Archaeology, Nashville, Tennessee.
Dahl berg, Michael D.
1975 Guide to Coastal Fishes of Georgia and Nearby States.
University of Georgia Press, Athens.
Darien Gazette
1818- Newspaper on Micro-film. University of Georgia at Athens
1828 Library.
Darien Timber Gazette
1887 Newspaper. (Rice Crop on the Savannah River, 27 August).
Originals on file, McIntosh County Court House, Darien, Georgia.
Davis, David B.
1966 The Problem of Slavery in Western Culture. Cornell University
Press, Ithaca, New York.
i- r-
t/G v o
1976
Herold
The Fledging Province: Social and Cultural Life in Colonial
Georgia 1733-1776. University of North Carolina Press,
Chapel Hill.
DeBows Review
1847
K1C
4:506-511
DeBows Review (After the War Series)
1867 The Sea Island cotton of the South: Its history, characteris
tics, cultivation. 3(1):84-83.
S t _-J y* fe -*AV' f iVY I* H 1 'I I I A

234
Deetz, James
1977 In Small Things Forgotten. Doubleday, New York.
Degler, Carl N.
1971 Neither Black nor White: Slavery and Race Relations in
Brazil and the United States. MacmiHan, New York.
Denyer, Susan
1978 African traditional architecture: An Historical and Geographi
cal Perspective. Africana Publishing, New York.
Descendants of the Butler Slaves
1898- Manuscript. (Letters to the Leighs and Wisters). WFPC, box. 41,
1903 folder 8, PHS.
Doar, David
1970 Rice and Rice Planting in South Carolina. The Charleston
Museum, Charleston, South Carolina.
Drucker, Lesley M. and Ronald W. Anthony
1979 The Spiers Landing Site: Archaeological Investigations in
Berkeley County, South Carolina. Carolina Archaeological
Services, Columbia, South Carolina.
Easterby, James H.
1941 The South Carolina rice factor as revealed in the papers of
Robert F. W. Alistan. Journal of Southern History 7:160-172.
1945 The South Carolina Rice Plantation as Revealed in the Papers
of Robert F. W. Allston. University of Chicago Press, Illionis.
Eaton, Clement
1949 A History of the Old South. Macmillan, New York.
1961 The Growth of Southern Civilization, 1790-1860. Harper & Row,
New York.
Ernst, Carl H. and Roger W. Barbour
1972 Turtles of the United States. University Press of Lexington,
Kentucky.
Escott, Paul D.
1979 Slavery Remembered: 20th Century Slave Narratives. Uni versity
of North Carolina Press, Chapel~H'i'Tl.
Fairbanks, Charles H.
197.4 The Kingsley slave cabins in Duval County, Florida, 1968.
Conference on Historic Sites Archaeology Papers 7:62-93.
1977 Backyard archaeology as research strategy. Conference on
Historic Site Archaeology 11:133-139.

235
Fairbanks, Charles H. and Sue A. Mullins-Moore
1980 The archaeology of slavery. Early Man, summer (in press).
Fancher, Betsy
1971 The Lost Legacy of Georgia's Golden Isles. Doubleday, Garden
City, New York.
Ferguson, Lei and
1980 Looking for the "Afro" in Colono-Indi an pottery. In
Archaeological Perceptivas on Ethnicity in America.
edited by Robert L. Schuyler, pp. 14-28, Baywood Publishing,
Farmingdale, New York.
Fine, Elsa Honig
1973 The Afro-American Artist. Holt, Rinehart, and Winston,
New York.
Flanders, Ralph B.
1933 Plantation Slavery in Georgia. University of North Carolina
Press, Chapel Hill.
Fogel, Robert W. and Stanley L. Engeriran
1974 Time on the Cross: The Economics of American Negro Slavery.
Little, Brown, and Company, Boston.
Fontana, Bernard L,
1973 The cultural dimensions of pottery: Ceramics as social docu
ments. In Ceramics in America, edited by George Quirnby, pp. 1-
18. University of Virginia Press, Charlottesville.
Gall man, Robert E.
1970 Self-sufficiency in the cotton economy of the antebellum South.
Agrieu1 turn! History 44:5-24.
Garrison, Ervan
1977 Modeling inundation effects for planning and prediction. In
Ccnservatlon Archaeology: A Guide for Cultural Resource
Management. Studies, edited by Michael B. Schiffer and George
Gutnmernia.n, pp. 151-157. Academic Press, New York.
Gejvall, Nils-Gustaf
1S63 Cremations. In Science in Archaeology: A Comprehensive Survey
of Progress and Research, edited by Don Brothwell and Eric Higgs,
pp. 468-479. Basic Books, New York.
Genovese, Eugene D.
1974 Roll Jordon Roll: The World the Slaves Made. Pantheon Books,
New York.
Georgia Writers Project
1940 Drums and Shadows: Survival Studies Among Georgia Coastal
Negroes. University of Georgia Press, Athens.
" ' -

236
Gibbs, Tyson, Kathleen Cargill, Leslie S. Lieberman, and Elizabeth Reitz
1980 An anthropological examination of slave nutritional adequacy.
Medical Anthropology: Cross-cultural Studies in Health and
Illness 4(3) (In press).
Glass!e, Henry
1968 Pattern in the Material Culture of Folk Culture of the Eastern
United States. University of Pennsylvania Press, Philadelphia.
Glassow, Michael
1977 Issues in evaluating the significance of archaeological re
sources. American Antiquity 44(3):413-420.
Goggin, John M.
1952 Space and Time Perceptive in Northern St. Johns Archaeology,
Florida. Yale University Press, New Haven.
Granger, Mary
1947 (Editor) Savannah River Plantations. Savannah Writers Project,
Gray, Lewis C.
IS41 History of Agriculture in the Southern United States to 1860.
Peter Smith, New York.
Greer, Georgeanna
1971 Preliminary Information on the use of the alkaline glaze on
stoneware in the South, 1800-1970. Conference on Historic
Site Archaeology Papers 4:155-170.
Grigg, David
1967 Regions, models, and classes. In Models in Georgraphy, edited
by Richard Chorely and Peter Haggett, pp. 461-501. Methuen,
London.
Gritzner,
¡978
Janet B.
[abby in the
Coastal Southeas
c:
The Culture History of an
American Building Material. Ph.D. Dissertation. Louisiana
State University, Baton Rouge.
Gunn, Victoria R.
1975 Hofwyl Plantation. Report on file, Historic Preservation
Section, Georgia Department of Natural Resources, Atlanta,
Georgia.
Gutman, Herbert G.
1977 The Black Famlly in Slavery and Freedom 1750-1925. Vintage
Books, New York. *
Haggett, P.
1965 Locational Analysis in Human Geography.
E. Arnold, London.

237
Hall, Margaret
1931 The Aristocratic Journey (1828), edited by Una Pope-Kennessey.
G. P. Putnam & Sons, London.
Hamilton, Jennifer
1980 Early history and excavation of Woodmanston plantation.
Unpublished M.A. Thesis. University of Florida, Gainesville.
Hammond, M. B.
1897 The cotton industry: An essay in American economic history,
part I. Publications of the American Economic Association
#1, New York.
Handler, Jerome and Frederick W. Lange
1978 Plantation Slavery in Barbados: An Archaeological and
Historical Investigation. Harvard University Press, Cambridge,
Massachusetts.
1979 Plantation slavery in Barbados, West Indies. Archaeology
32(4):45-52.
Hanson, Lee H.
1971 Pipes from Rome, New York. Historical Archaeology 5:92-99.
Harris, Marvin
1968 The Rise of Anthropological Theory: A_History of Theories_of
Culture. Thomas Crowell, New York.
Herskovits, Mervilie, J.
1958 The Myth of the Negro Past. Beacon Press, Boston.
Heyward, Dubose
1932 The Negro in the low country. In The Carolina Low Country,
edited by Augustine T. Smythe, pp. 171-191. William Morrow,
New York.
Heyward, Duncan C.
1937 Seed from Madagascar. University of North Carolina Press,
Chapel Hill.
Hilliard, Sam
log Meat and Hoe Cake: Food Supply 1n the Old South 1840-18¡
lincis University Press, Carbondale.
ar rice plantation: An ingenious adaptation to
Geoscience and Man: Coastal Resources 12:57-66.
1372
Hog Meat
Southern
1975
The tidev
nature.
Ho
i
Frank and Robert F. Keizer
Prehistoric Archaeoloqy: A Brief Iotrodectiojn.
hart, and Winston, New York.
iolt, Rine-
Holmes,
1878
D.
Comments on Fanny Kemble
13 September. Newspaper
at Athens.
s J o u r nal. D a ren Ti mb e r_ Gazette,
on Micro-film. University of Georgia

Hooper, George
1786- Manuscript. (Letters to Major Pierce Butler). BFPC, box 7,
1802 folder 2, PHS.
House, Albert J.
1939 The management of a rice plantation in Georgia 1834-1861, as
revealed in the journal of Hugh Fraser Grant. Agricuitura!
Hi story 13:208-217.
1942a Charles Manigault's essay on the planting of rice. Agricultura!
Hi story 16:184-193.
1942b A reconstruction share-cropper contract on a Georgia rice
piantation. Georgia Historical Quarterly 26:156-165.
1954a Labor.management problems on a Georgia rice plantation 1840-
1860. Agricultural History 28:149-153.
1954b Planter Management and Capitalism in Antebellum Georgia: The
Journal of Hugh Fraser Grant, Rice Grower. Columbia University
Press, New York.
House, John H.
1977 Survey data and regional models 'in archaeology. In Research
Strategies in Historical Archaeology, edited by Stanley South,
pp. 243-257. Academic Press, New York.
Humphrey, Richard V.
1969 Clay pipes from Old Sacramento. Historical Archaeology
3:12-23.
Isard, Walter
1956 Regional science: The concept of the region and regional struc
ture. Papers and Proceedings of the Regional Science Association
2:13-26.
Jackson, Harvey H.
1S73 General Lochlan McIntosh 1727-1806: A Biography. Ph.D.
Dissertation. University of Georgia, Athens.
Johnson, A. S., Hilbium 0., Hillestad, Sheryl Shanbotzer, and
1974 G. Frederick
An ecological survey of the coastal region of Georgia. National
Park ServiceScientific Monograph Series #3. Government
Printing Office, Washington, D.C.
Johnson, Gurion G.
1930 A Social History of the Sea Islands. University of North
Carolina Press, Chapel Hill.
Jones, William M.
1976 The source of ballast at a Florida site. Historical
Archaeology 10:42-45.

Kelly, Marsha C. S. and Roger Kelly
1980 Approaches to ethnic identifications in historical archeology.
In Archaeological Perceptivas on Ethnicity in America, edited
by Robert L. Schuyler' pp. 133-143. Baywood Publishing,
New York.
Kemble, Frances A.
1882 Record of a Later Life. Henry Holt, New York.
1961 Journa1 of A Residence on A Georgian Plantation In 1833-1839
(18637 edited by John A. Scott. Alfred A. Knopf, New York.
Ketchum, William C., Jr.
1971 The Pottery and Porcelain Collectors Handbook: A Guide to
Early American Ceramics From Main to California.'" Funk' &
Wagnalls, New York
Kidd, Kenneth and M. A. Kidd
1970 A classification system for glass beads for the use of field
archaeologists. In Occasional Papers in Archaeology and
History #1. Ontario, Canada.
Kill on, Ronald and Charles Waiter
1973 (Edi to rs) Slavery Time When I Was Chi Hun Down on Master's
Plantations. Beehive Press, Savannah, Georgia.
King, Roswell, Jr.
1819- Manuscript. (Letters to Frances Butler, Major Butler, and
1835 Thomas Butler). BFPC, boxes 10.11, 12 and WFPC, boxes 33,
40, PHS.
1828 On the management of the Butler estate, and the cultivation of
sugar cane. Southern Agriculturalist 1:523-529.
1844 Daybook. (Estate of John and Pierce Butler in Georgia).
MDCC, folder 205, GHS.
King, Roswell, Sr.
1803- Manuscript. (Letters to Major Pierce Butler). BFPC, boxes
1819 9, 10, PHS.
King, Thomas, Patricia Parker Hickman, and Gary Berg
Anthropology in Historic Preservation:___ Caring for Culture's
Clutter. Academic Press, New York.
Knapp, S. A.
1900 Rice culture in the United States. U.S. Department of
Agriculture Farmers Bulletin jfT!0_. Government Printing Office,
Washington, D.C.
Kniffen, Fred
1965 Folk housing: Key to diffusion. Annals of the Association
of American Geographers 55(4):549-577.
''y --
, : i t \ ....
:V'ns-f -w,..

240
Lane, Mil lis
1973 (Editor) Charles Lye11. In The Rambler in Georgia, pp. 189-
201. Beehive Press, Savannah, Georgia.
Langhorne, William I., Jr.
1976 Mill based settlement patterns in Schoharie County, New York:
A regional study. Historical Archaeology 10:73-92.
Larson, Lewis, H., Jr.
1S70 Aboriginal Subsistence Technology in the Southeastern Coastal
Plain During the Late" Prehistoric Period. Ph.D. Dissertation.
University Microfilms, Ann Arbor, Michigan.
Lauer,
1969
Joseph Jerome
Rice in the history of the lower Gambia-Geba area.
M.A, Thesis. University of Wisconsin, Madison.
Unpublished
Lawson, Dennis T.
1975 No Heir To Take Its Place: The Study of Rice in Georgetown
County, South Carolina. The Rice Museum, Georgetown, South
Carolina.
Leigh, Frances Butler
1883 Ten Years on a Georgian Plantation, Since the War. Bentley
and Sons, London.
Leigh, John
1921 Other Days. Bentley and Sons, London.
Lewis, Bessie
1975 They Called Their Town Darien. Darien News, Darien, Georgia.
Lewis, Bessie and Mildred Huie
1974 Patriarchi a 1_P1antatlons o f_StSimons. L ewis and Huie,
St. Simons, Georgia.
1978 Hampton Plantation. Plantations of Coastal Georgia #1.
Argyle Publications, St. "Simons Island, Georgia.
Lewis, Kenneth
1377 Sampling the archaeological frontier: Regional models and
component analysis. In Research Strategies in Historical
Archaeology, edited by Stanley South, pp. 151-196. Academic
Press, New York.
1979 Settlement and activity patterning of two rice plantations
in the South Carolina low country. Paper delivered at the
20th Annual Conference on Historic Site Archaeology,
St. August-!ng, Florida.
Lombard, Mildred E.
1930 Contemporary opinion of Mrs. Kemble's Journal of a Residence
on a Georgian Plantation. Georgia Historical Quarterly 14(4):
335-343.

Long, F. L.
1958 Chemical properties and the fertility requirements of some
brackish marsh soils of Georgia. Unpublished M.S. Thesis.
University of Georgia, Athens.
Lovell, Carole Couper
1932 Golden Isles of Georgia. Little, Brown and Company, Boston.
Ludlum, David M.
1963 Early American Hurricanes, 1492-1878. American Meteorological
Society, Boston.
Luscomb, Sally C.
1975 The Co11ector1 $ Encyclopedia of Buttons. Crown Pub1ishers,
New York.
MacFarlane, Suzanne
1975 The ethnoarcheology of a slave cabin community: Couper
Plantation site. Unpublished M.A. Thesis. University of
Florida, Gainesville.
Mauban, A.
1S45 L1Architecture Francaise.
De Jean Maviette, Vanoest, Paris.
McClane
1978
A. J.
Field Guide to_ Freshwater Fishes of North America.
Rinehart and Winston, New York.
Hoi t,
McGimsey Charles R. III and Hester A. Davis
1977 (Editors) The Management of Archaeological Resources: Airlie
House Report. Special Publication of the Society for American
Archaeology, Washington, D.C.
McIntosh County Records
MS Deed Books. H-52, 1-20, 1-160.
Courthouse, Darien, Georgia.
On file in McIntosh County
Miller, Elinor and Euguene D. Genovese
1974 Piantation, Town,_and County: Essays in the Local History
of American Slave Society. University of Illinois Press,
Urbana.
Sue A.
An archaeological survey of Butler Point, St. Simons
Island, Georgia. Unpublished Report on File. Department
of Anthropology, University of Florida.
M u 11 i n s Mo o re, S u. e A.
nd The Antebellum Barrier Island Plantation: A Search for an
Archaeological_ Pattern. Ph.D. Dissertation. University of
Florida, Gainesville (In preparation).
Tul Iins,
1978

242
Noel-Hume, Ivor
1962 An Indian Ware of the colonial period. Archaeological Society
of VIrglnla Quarterly BulletIn 17(1):1 -68.
1973 From crearnware to pearlware: A Williamsburg perspective.
In Ceramics in America, edited by George Quimbys pp. 217-
254. University of Virginia Press, Charlottesville.
1974 Historical Archaeology. Alfred A. Knopf, New York.
1976 Artifacts of Colonial America. Alfred A. Knopf, New York.
Olmsted, Frederick L.
1968 A Journal in the Seaboard Slave States With Remarks on Their
Economy "(18567. Negro Universities Press, Detroit.
Olsen, Stanley
1963 Dating early plain buttons by their forms. American Antiquity
28(4):551 -554.
Otto, John Solomon
1S75 Status Differences and the Archaeological RecordA Comparison
of Planter, Overseer, and Slave Sites From Cannon's Point
Plantation (1794-186f), St. Simons Island, Georgia. Ph.D.
Dissertation. University of Florida, Gainesville.
1977 Artifacts and status differencesa comparison of ceramics
from planter, overseer, and slave sites on an antebellum
plantation. In Pvesearch Strategies in Historical Archaeology,
edited by Stanley South,"pp. 91-116. Academic Press, New York.
1979 Slavery in a Coastal Community Glynn County (1790-1860).
Georgia Historical Quarterly 64(2):161-168.
Owens, Leslie H.
1975 This Species of Property: Slave Life and Culture in the Old
South. Oxford University Press, New York.
Panshin, A. J.
1964Textbook of Wood Technology. McGraw-Hill, New York.
Perkerson, Medora
1952 White Columns in Georgia. Macmillan, New York.
Peterson, Roger T.
1947 A Field Guide to Birds of the United States. Houghton
Mifflin, Boston.
Phillips, Ulrich
1929 Life and Labor in the Old South. Little, Brown, and Company,
Boston.
1969 (Editor) Plantation and Frontier. Burt Franklin, New York.

243
Plog, Fred
1978 Cultural resource management and the "new archaeology."
In Social Archaeology: Beyond Subsistence and Dating, edited by
Charles L. Redman, Mary A. Berman, Edward Curtin, William
Langhorne, Jr., Nina M. Veroggi, Jeffery Winser, pp. 402 422.
Academic Press, New York.
Pope, Clifford H.
1939 Turtles of the United States, Canada. Alfred A. Knopf, New
York.
Postell, William D.
1951 The Health of SI aves on S o uthern Plant a ti o ns_. Louisiana
State University, Baton Rouge.
Potter, David M.
1932 The rise of the plantation system in Georgia. Georgia
Historical Quarterly 16(1):114-135.
Price, Cynthia R.
1979 19th Century Ceramics in the Eastern Ozark Border Region.
Center for Archaeological Research, Southwest Missouri"State
Uni versity, Springfield.
Prunty, Merle
1S55 The renaissance of the southern plantation. Geographical
Review 45(4):459-491.
Reitz, Elizabeth J.
1979 Spanish and British Subsistence Strategies at St. Augustine,
Rond a_ and Frederica. Georgia Between 1565 and 1783. Ph.D.
Dissertation. University of Florida, Gainesville.
Salley, A. S., Jr.
1919 The introduction of rice into South Carolina. Bulletins of
the Historical Commission of South Carolina, #6. The State
Company, Columbia, South Carolina.
Sammis, I. S.
1977 Pottery at Huntington, Long Island. In The Art of the Potter,
edited by Diana Stradling and J. Garrison, pp. 103-112.
Antiques Magazine Library, Universe Books, New York.
Sanders, William T.
nd Report on Excavations on Butler Island, July 1956. Report on
file, Archaeological Laboratory, University of Georgia, Athens.
Savitt, Todd L.
1973 Medicine and Slavery: The Diseases and Health Care of Blacks
in Ante be11 urn Virginia. University of Illinois Press, Urbana.

244
Scarborough, William M.
1964 The plantation overseer reevaluated. Agricultural History
38:13-20.
1966 The Overseer: Plantation Management in_the Old South.
Louisiana State University, Baton Rouge.
Schiffer, Michael B. and George J. Gummernan
1977 (Editors) Conservation Archaeology: A Gulde for Cultural
Resource Management Studies. Academic Press, New York.
Schuyler, Robert
1980 (Editor) Archaeological Perspectives on Ethnicity in America:
Afro-American and Asian American Culture History. Baywood
Publishing, Farmingdale, New York.
Scott, John Anthony
1961 (Editor) A Journal of A Residence On a Georgian Plantation
in 1838-1839, by Fanny Kemble (1863). Alfred A. Knopf, New
York.
1973 Fanny Kemble's America. Dell Publishing Company, New York.
1978 Hard Trials on My Way: Slavery and the Struggle Against It,
1800-1860. A Mentor Book, New York.
Singleton, Theresa
nd Historical and Archaeological Investigations at Butler Island,
Report to the Office of the State Archaeologist, Georgia
Department of Natural Resources, Carrollton, Georgia (In prepara
tion) .
Sitterson, J. Carlyle
1937 Antebellum sugar cane in the South Atlantic States, Journal
of Southern History 9:59-74.
1943 The William J. Minor plantations: A study in antebellum
ownership. Journal of Southern History 9:59-74.
1953 Sugar Country: The Cane Sugar Industry in the SouthJ753-
1950. Unversity of Kentucky Press, Lexington'.
Slosne, Eric
1564 A Museum of Early American Tools. Ball atine Books, Toronto,
Canada.
Smith, Alice R. Huger
1936 A Carolina Rice Plantation of the Fifties. Hi 11iam Morrow,
New York.
Smith, Julia F.
1973 Slavery and Plantation Growth in Antebellum Florida, 1821-1860.
University of Florida Press, Gainesville,

245
Smith, Samuel D.
1977 Plantation archaeology at the Hermitage: Some suggested
patterns. Tennessee Anthropology 2:152-163.
South, Stanley
1964 Analysis of buttons from Brunswick Town and Fort Fisher.
Florida Anthropologist 17(2):113-133.
1977 Method and Theory in Historical Archaeology. Academic Press,
New York.
1978 Pattern recognition in historical archaeology. American
Antiquity 43(2):223-230.
Southern Agricultural 1st
1834 On the protection and repairs of rice embankments. 7:517-519.
Spalding, Phinizy
1977 Part One: Colonial Period. In The History of Georgia, edited
Kenneth Coleman, pp. 9-70. Univeristy of Georgia Press, Athens.
Spalding, Thomas
1830 Observations on the bordering of river banks with orange trees
and protecting the embankments with Bermuda grass. Southern
Agriculturist 3:73-96.
1835 On the culture, harvesting, and threshing of rice and in the
rust in cotton. Southern Agricultural 1st 8:169-174.
1845 Culture of rice. Amen can Agriculturai1st 4:53-55.
Stampp, Kenneth M.
1955 The Perculiar Institution: Slavery in the Antebellum South.
Alfred A. Knopf, New York.
Steckel, Richard H.
1379 Slave mortality: Analysis of evidence from plantation records.
Social Science History 3 and 4:86-114.
Steward, Julian H.
1955 The Theory of Culture Change:_ The Methodology of Mill til inear
Evolution. University of Illinois Press, Chicago.
Stradling, Diana and J. Garrison
1977 The Art_of the Potter: _Redware and Stoneware. Antiques
Magazine Library, Universe Books, New York.
Sireover, Stuart
1971 Comments on archaeological data requirements and research
strategy. American Antiquity 36:9-19.

246
Sutch, Richard
1976 The care and feeding of slaves. In Reckoning With Slavery:
A Critical Study in the Qu anti tat ive History of American Negro
Slavery, edited Paul A. David, pp. 231-301. Oxford University
Press, New York.
Swendlund, Alan C.
1975 Population growth and settlement pattern in Franklin and
Hampshire Counties Massachusetts, 1650-1850. Memoirs of the
Society for American Archaeology #30, Washington, D.C.
Switzer, Ronald
1974 The Betrand Bottles: A Study of 19th Century Glass and Ceramlc
Contalners. National Park Service, U.S. Department of the
Interior, Government Printing Office, Washington, D.C.
Tannenbaum, Frank
1949 Slave and Citizen: The Negro in the Americas. Alfred A.
Knopf, New York.
Thayer, Theodore
1957 Nathaniel Pendletons short account of the sea coast of Georgia
in respect to agriculture, shipbuilding, navigation and timber
trade. Georgia Histrica1 Quarterly 41:70-81.
Thomsen, Mortimer
1863 What became of the slaves on a Georgia Plantation: Great
auction of slaves at Savannah, Georgia, March 2nd and 3rd,
1859, A sequel to Mrs. Kemble's Journal. American Anti-
slavery Society, Boston.
Trigger, Bruce G.
1968 The determinants of settlement patterns. In Settlement
Archaeology, edited by K. C. Chang, pp. 53-78. National Press
Books, Palto Alto, California.
Trimble, S. W.
1969 Culturally accelerated sedimentation on the middle Georgia
Piedmont. Unpublished M.S. Thesis. University of Georgia,
Athens.
U.S. Department of Agriculture
1961 Soil Survey of McIntosh County Georgia. Government Printing
Office, Washington, D.C.
n DeBurg, William
i 070
\ y / y
The Slave Drivers: Black Agricultural Labor Supervisors in
the Antebellum South. Greenwood Press, London
Van Doren, Mark
1929 (Editor) Correspondence of Aaron Burr and His Daughter,
Theodosia. Stratford Press, New York.

247
Van Rensselaer, Susan
1978 Banded creamware. In English Pottery and Porcelain: A His
torical Survey, edited by Paul Atterburn, pp. 240-244.
Universe Books, New York.
Vanstory, Burnette
1970 Georgia's Land of the Golden Isles. The University of Georgia
Press, Athens.
Vlach, John Michael
1978 The Afro-American Tradition in Decorative Arts. The Cleveland
Museum of Art, Cleveland, Ohio.
Wagret, Paul
1968 Polder!ends. Barnes and Noble, New York.
Watkins, Malcolm
1970 Artifact From the Sites of Three 19th Century Houses and
Ditches at Darien Bluff, Georgia. University of Georgia
Laboratory of Archaeology Series Report #9, Athens.
Weiss, Mark L. and Alan E. Mann
1975 Human Biology and Behavior: An Anthropological Perceptive.
Little, Brown, and Company, Boston.
Wells, Tom Henderson
1967 The Slave Ship Wanderer. University of Georgia Press, Athens.
Wilms, D. C.
1972 The development of rice culture in 18th century Georgia.
Southeastern Geographer 12:45-57.
Wing, Elizabeth and Antoinette B. Brown
1979 Paleo-Nutrition: Method and Theory in Prehistoric Foodways.
Academic Press, New York.
Wolf, Eric
1959 Aspects of plantation systems in the New World: Community
sub-cultures and social classes. In Seminar on Plantation
Systems of the New World. Social Science Monographs 7:136-
147. Research Institute for the Study of Man and the Pan
American Union, Washington, D.C.
Wood, Peter H.
1974 Negroes In Colonial South Carolina From 1670 Through the
Stono Rebel lion. W. W. Norton, New York.
Lylly, Charles Spalding
1910 The Seed That Was Sown in the Colony of Georgla: The Harvest
and the Aftermath, 1740-1370. Neale Publishing, New York.

BIOGRAPHICAL SKETCH
Theresa A. Singleton was born in Charleston, South Carolina,
15 April 1952. She grew up in Charleston and graduated from Bishop
England High School in 1970. She then moved to Washington, D.C., to
enter Trinity College, where in 1974 she received the degree of Bachelor
of Arts with a major in anthropology-sociology. In 1975, she entered
the Graduate School at the University of Florida, where in 1977 she
received the degree Master of Arts with a major in anthropology and in
1980, the degree Doctor of Philosophy with a major in anthropology.
248
1 J 'fc'SU'-y /tv*' y r- \S-

I certify that I have read this study and that in rny opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Charles H. Fairbanks, Chairman
Distinguished Service Professor
of Anthropology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
/
/ Jerald T. Mi tarn ch
/ Associate Professor of
l Anthropology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Otto Von Men ng""
Professor of Anthropology
I certify that I
conforms to acceptable
have read this study and that in my opinion it
standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Samuel Proctor
Distinguished Service Professor
of History
- ' *

I certify that I
conforms to acceptable
adequate, in scope and
Doctor of Philosophy.
have read this study and that in my opinion it
standards of scholarly presentation and is fully
quality, as a dissertation for the degree of
Ronald C. Foreman
Associate Professor of English
This dissertation was submitted to the Graduate Faculty of the Department
of Anthropology in the College of Liberal Arts and Sciences and to the
Graduate Council, and was accepted as partial fulfillment of the require
ments for the degree of Doctor of Philosophy.
December 1980
/
Dean, Graduate School



Figure 35. Shoe Soles.


133
Birds and mammals
A diving duck (Aythya sp.) was the only nondomestic bird identi
fied. Ducks were apparently taken by the Butler slaves, as the planta
tion records indicate, but the slaves' accessabi1ity to guns may have
regulated duck hunting activity. Fields of planted rice attracted
numerous predatory fowl (Austin 1893:34), and slaves were possibly en
couraged to capture them at times. The rice bird, or the bobolink,
(Dolichonyx oryzivorus) was paramount among these predatory birds. Per
haps its size (6 1/2 inches,or 16 centimeters; see Peterson 1947:209)
and the fact that the entire bird, both flesh and bone, was generally
consumed* have resulted in no archaeological evidence of this food item.
Only two nondomestic mammalian species were identified, raccoon
(Procyon lotor) and deer (Odocoileus virginianus). Both occur more fre
quently in the hammocks and salt marshes along,the coast than in the
delta but appear to adapt well to deltaic marshland. The raccoon with
its varied diet could thrive on frogs, insects, berries, seeds, and
eggs of birds and turtles found in the delta (Johnson et al. 1974:42,
85). The deer were possibly attracted to garden crops (Reitz 1979:78).
Traps could be set to capture the nocturnal raccoon while, presumably,
the deer were taken with guns.
The unidentified rodent (Rodentia) and rats and mice (Cricitidae)
as well as the snake (Squamata) were probably not consumed and for that
reason have not been included in either Table 13 or Table 16. In the
South Carolina rice coasts, however, slaves have been reported to have
consumed the rice rat (Oryzomys sp.), which they considered a delicacy
*Rudolph Capers, personal communication, 1980.


199
represented and several varieties were noted. A fragment of a "TD"
pipe (Humphrey 1969:14; Hanson 1971:92) was identified.
All pipes were white clay with the exception of one which was
brown and white porcelain. The latter was the only pipe with a de
tachable stem pipe bowl. It was presumably of German manufacture and
would have been a fairly expensive item for a slave to have owned (see
Figure 38).
Farming Tools and Specialized Crafts
All of the tools recovered were those to be expected at a rice
plantation. Fragments of six eye hoes and one rice hook were the only
farming tools, but a considerable number of woodworking tools were
present, particularly at structure one. Such woodworking tools are
also characteristic of rice production. The wooden tierces used for
shipping rice were generally made at the plantation.
Included among the woodworking tooks are an adze, an axe, a
rounded chisel or gouge (see Figure 39), a number of smaller chisels,
and a saw knife blade. Adzes and axes are general woodworking tools,
whereas the chisels and saw knife are more specialized tools. The
rounded chisel was used to make holes and the saw knife for shaping
rough wood (Sloans 1964:54, 84). Perhaps the abundance of these wood
working tools at structure one indicates that the structure was oc
cupied by a cooper and his family. This may explain the presence of
some of the "luxury" items. Coopers, like other plantation specialists,
viere often able to acquire extra cash through hiring themselves out or
by building items which they sold for cash (Kemble 1961 (1863);65).
With the extra cash they were able to acquire a standard of living
slightly above the field hand.


231
Booth, Sally S.
1971 Hung, Strung, & Potted: A His to ry of Eating in Co1onial
America. Crown Publishers, New York.
Brewster, Lawrence F.
1970 Summer Migrations and Resorts of the South Carolina Low
Country Planters. AMS Press, New York.
Bridenbaugh, Carl
1952 Myths and Real i11es: Socleties of the Colonial South.
Louisiana State University Press, Baton Rouge.
Brown, Harry B. and Jacob 0. Ware
1953 Cotton. McGraw-Hill, New York.
Butler Estate Papers
1799 Copy of taxable property (McIntosh and Glynn Counties, Georgia).
BFPC, box 3S folder 2, PHS.
1301- Slave transactions. BFPC, box 3, folder 5, PHS.
1803
1803 List of ditchers at Tide Island, BFPC, box 3, folder 3, PHS.
1803- Crop and livestock reports, BFPC, box 7, folder 4, PHS.
1830
1811 Annual expense list, BFPC, box 9, folder 7, PHS.
1314 Crop proceeds since 1801, BFPC, box 9, folder 10, PHS.
1815a Inventory of the Butler Estate, WFPC, box 37, folder 8, PHS.
1815b Provisions for Negroes, BFPC, box 3, folder 3, PHS.
1821 List of field hands, BFPC, box 3, folder 3, PHS.
1824- Statement of crops. BFPC, box 11, folder 5, PHS.
1835
1859 A list of slaves belonging to John Butler, WFPC, box 41,
folder 5, PHS.
Butler, Pierce, Jr.
1851- Ledge>'\ WFPC, loose, PHS.
1855
Caldwell, Sheila
nd Report on Butler Island burial site. Report on file, Historic
Preservation Section, Georgia Department of Natural Resources,
Al tonta, Georgia.
Calloway, James E.
1943 Early Settlement of Georgia.. University of Georgia Press, Athen
; i ' ; -"v - : m*. mmmmmmvm


LIST OF FIGURES
Figure Page
1. Investigated Slavery Sites in Coastal Georgia. 15
2. The Physiographic Regions of Georgia. 21
3. Extent of Rice Cultivation in Georgia in I860 27
4. Butler Island and Environs. 47
5. The Butler Georgian Estate. 52
6. The Subdivision of Butler and General's Islands into
Tenant Plots (circa 1883). 57
7. Historic Sites at Butler Island. £6
8. Present-Day View of Butler Island. 88
9. Extent of Excavations at Settlement #1. 95
10. Stratigraphic Profile of Settlement #1. 97
11. Eighteenth-Century Slipware Fragments from Settlement #1. 100
12. Spanish reale, "Mexico Mint," from Settlement '4. 102
13. Contour Map of Settlement #4. 104
14. Excavation of Settlement #4. 105
15. Stratigraphic Profile, Settlement #4. 107
16. Overseer's House. 115
17. Tidal Mill and Boat Landing. 115
13. Slave Cabin. 116
19. Steam Mi 11 Rui ns. 116
20. Linear Arrangement of Slave Dwellings. 116
21. Extent of Excavation at Structure Four. 122


Figure 31
Eating Utencils. (A is a white metal spoon;
B is a pewter spoon.)


114
other unexplored factors also had an impact on these patterns. The
extent that these additional factors influenced plantation arrangements
remains unanswered.
Slave Community Organizations at Butler Island
The hypotheses outlined in Chapter 1 assume that plantation
spatial organizations were primarily functional and that these reflect
adaptations to specific crop requirements. Butler Island's settlement
patterns conform to those of most tidewater rice plantations, specifi
cally those lacking sufficient highlands. Plantation structures were
evidently built upon high areas that were either naturally occurring or
artifacts of polderland construction. In the placement of slave vil
lages, the latter appears to have been the case. The slave villages were
strategically located at the southern end of the island, where they were
accessible overland by the main dikes and by water along the main canals
and the river (see Figures 7 and 8).
Settlement #1 was the administrative nucleus for the plantation
and it included the overseer's house, blacksmithing and shoemaking
sheds, warehouses, and a meat-curing house. Also,the largest technical
nucleus for the island, a tidal mill and steam mill, and the greatest
number of slave dwellings were located there. Photographs of structures
once located at settlement #1 (Cate 1925; Figures 16, 17, 18, 19, and 20}
provide some indications of the former spatial arrangements at this site.
The technical equipment, was east of the overseer's house, the slave quar
ters to the south, and the other administrative buildings are believed to
have been to the west (see Scott 1961:57). For the Butler Island slaves,
settlement 41 must also have served as headquarters.
I he slave


Hooper, George
1786- Manuscript. (Letters to Major Pierce Butler). BFPC, box 7,
1802 folder 2, PHS.
House, Albert J.
1939 The management of a rice plantation in Georgia 1834-1861, as
revealed in the journal of Hugh Fraser Grant. Agricuitura!
Hi story 13:208-217.
1942a Charles Manigault's essay on the planting of rice. Agricultura!
Hi story 16:184-193.
1942b A reconstruction share-cropper contract on a Georgia rice
piantation. Georgia Historical Quarterly 26:156-165.
1954a Labor.management problems on a Georgia rice plantation 1840-
1860. Agricultural History 28:149-153.
1954b Planter Management and Capitalism in Antebellum Georgia: The
Journal of Hugh Fraser Grant, Rice Grower. Columbia University
Press, New York.
House, John H.
1977 Survey data and regional models 'in archaeology. In Research
Strategies in Historical Archaeology, edited by Stanley South,
pp. 243-257. Academic Press, New York.
Humphrey, Richard V.
1969 Clay pipes from Old Sacramento. Historical Archaeology
3:12-23.
Isard, Walter
1956 Regional science: The concept of the region and regional struc
ture. Papers and Proceedings of the Regional Science Association
2:13-26.
Jackson, Harvey H.
1S73 General Lochlan McIntosh 1727-1806: A Biography. Ph.D.
Dissertation. University of Georgia, Athens.
Johnson, A. S., Hilbium 0., Hillestad, Sheryl Shanbotzer, and
1974 G. Frederick
An ecological survey of the coastal region of Georgia. National
Park ServiceScientific Monograph Series #3. Government
Printing Office, Washington, D.C.
Johnson, Gurion G.
1930 A Social History of the Sea Islands. University of North
Carolina Press, Chapel Hill.
Jones, William M.
1976 The source of ballast at a Florida site. Historical
Archaeology 10:42-45.


d. Evidence of food supplements in plantation rations
1. Equipment for producing food
2. Equipment for procuring wild food resources
3. Food processing equipment
4. Remains of food supplements
2. Personal possessions
a. Clothing
b. Accessories and ornaments
c. Household items
3, Leisure-time activities
a. Smoking equipment
b. Games and toys
c. Miscellaneous items
These hypotheses and their underlying assumptions are discussed
in greater detail in later chapters. Chapters 2 and 3 provide back
ground information on the coastal environment of Georgia, the produc
tion of rice and long-staple cotton, and the archaeological and his
torical investigations at Butler Island. Slave community organizations
at Butler Island are discussed in Chapter 4. Chapter 5 describes the
slave material culture recovered from Butler Island. Slave artifact
patterns derived from statistical pattern recognition techniques are
presented in Chapter 6. Chapter 7 discusses the results from testing
the hypotheses and the conclusions for the study.


233
Cornwall, I. W.
1958 Soils for the Archaeologists. Phoenix House, London.
Corry, Joseph
1960 Observations Upon the Windward Coast of Africa Made in the
Years 1305 and 1806. Franklin Cass, London.
Coulter, Merton E.
1941 Thomas Spalding of Sapelo. University of Louisiana Press,
Baton Rouge.
Couper, John
1818 Manuscript. (Letter to Major Butler). WFPC, box 31, folder 7,
PHS.
Crook, Morgan R., Jr.
1978 Mississippian Period Community Organizations on the Georgia
Coast. Ph.D. Dissertation. University of Florida, Gainesville.
Crosby, Constance and Mathew C. Emerson
1979 Identifying Afro-American mortuary custom: An example from
the Parting Ways Settlement, Plymouth, Massachusetts. Paper
delivered at the 12th annual meeting of the Society for
Historical Archaeology, Nashville, Tennessee.
Dahl berg, Michael D.
1975 Guide to Coastal Fishes of Georgia and Nearby States.
University of Georgia Press, Athens.
Darien Gazette
1818- Newspaper on Micro-film. University of Georgia at Athens
1828 Library.
Darien Timber Gazette
1887 Newspaper. (Rice Crop on the Savannah River, 27 August).
Originals on file, McIntosh County Court House, Darien, Georgia.
Davis, David B.
1966 The Problem of Slavery in Western Culture. Cornell University
Press, Ithaca, New York.
i- r-
t/G v o
1976
Herold
The Fledging Province: Social and Cultural Life in Colonial
Georgia 1733-1776. University of North Carolina Press,
Chapel Hill.
DeBows Review
1847
K1C
4:506-511
DeBows Review (After the War Series)
1867 The Sea Island cotton of the South: Its history, characteris
tics, cultivation. 3(1):84-83.
S t _-J y* fe -*AV' f iVY I* H 1 'I I I A


Figure 39. Woodworking Tools.
Key to Artifacts
1. Adze
2. Axe
3. Rounded chisel or gouge


244
Scarborough, William M.
1964 The plantation overseer reevaluated. Agricultural History
38:13-20.
1966 The Overseer: Plantation Management in_the Old South.
Louisiana State University, Baton Rouge.
Schiffer, Michael B. and George J. Gummernan
1977 (Editors) Conservation Archaeology: A Gulde for Cultural
Resource Management Studies. Academic Press, New York.
Schuyler, Robert
1980 (Editor) Archaeological Perspectives on Ethnicity in America:
Afro-American and Asian American Culture History. Baywood
Publishing, Farmingdale, New York.
Scott, John Anthony
1961 (Editor) A Journal of A Residence On a Georgian Plantation
in 1838-1839, by Fanny Kemble (1863). Alfred A. Knopf, New
York.
1973 Fanny Kemble's America. Dell Publishing Company, New York.
1978 Hard Trials on My Way: Slavery and the Struggle Against It,
1800-1860. A Mentor Book, New York.
Singleton, Theresa
nd Historical and Archaeological Investigations at Butler Island,
Report to the Office of the State Archaeologist, Georgia
Department of Natural Resources, Carrollton, Georgia (In prepara
tion) .
Sitterson, J. Carlyle
1937 Antebellum sugar cane in the South Atlantic States, Journal
of Southern History 9:59-74.
1943 The William J. Minor plantations: A study in antebellum
ownership. Journal of Southern History 9:59-74.
1953 Sugar Country: The Cane Sugar Industry in the SouthJ753-
1950. Unversity of Kentucky Press, Lexington'.
Slosne, Eric
1564 A Museum of Early American Tools. Ball atine Books, Toronto,
Canada.
Smith, Alice R. Huger
1936 A Carolina Rice Plantation of the Fifties. Hi 11iam Morrow,
New York.
Smith, Julia F.
1973 Slavery and Plantation Growth in Antebellum Florida, 1821-1860.
University of Florida Press, Gainesville,


APPENDIX 2
SLAVE SUBSISTENCE ARTIFACTS FROM SETTLEMENT 4
Artifact Class
it of Items
Building hardward
Nails and spikes
Hinges
Pintle
Padlocks
Window glass
Total
4468
3
1
c
17
4494
Farming tools and equipment
Hoes
Chisels
Gouge or rounded chisel
Saw knife blade
Axe head
Adze
Fil e
Sickle
Unidentified objects (Figure 40)
Total
Food procurement equipment
Gun flints
Buckshots and musket balls
Trigger guard
Fish net sinkers
Total
6
5
1
1
1
1
1
1
2
19
6
8
1
4
Food preparation and serving equipment
Kettle legs 2
Grinding stone 5
Spoons
Pewter 1
Iron 2
Whitemetal handle
225


234
Deetz, James
1977 In Small Things Forgotten. Doubleday, New York.
Degler, Carl N.
1971 Neither Black nor White: Slavery and Race Relations in
Brazil and the United States. MacmiHan, New York.
Denyer, Susan
1978 African traditional architecture: An Historical and Geographi
cal Perspective. Africana Publishing, New York.
Descendants of the Butler Slaves
1898- Manuscript. (Letters to the Leighs and Wisters). WFPC, box. 41,
1903 folder 8, PHS.
Doar, David
1970 Rice and Rice Planting in South Carolina. The Charleston
Museum, Charleston, South Carolina.
Drucker, Lesley M. and Ronald W. Anthony
1979 The Spiers Landing Site: Archaeological Investigations in
Berkeley County, South Carolina. Carolina Archaeological
Services, Columbia, South Carolina.
Easterby, James H.
1941 The South Carolina rice factor as revealed in the papers of
Robert F. W. Alistan. Journal of Southern History 7:160-172.
1945 The South Carolina Rice Plantation as Revealed in the Papers
of Robert F. W. Allston. University of Chicago Press, Illionis.
Eaton, Clement
1949 A History of the Old South. Macmillan, New York.
1961 The Growth of Southern Civilization, 1790-1860. Harper & Row,
New York.
Ernst, Carl H. and Roger W. Barbour
1972 Turtles of the United States. University Press of Lexington,
Kentucky.
Escott, Paul D.
1979 Slavery Remembered: 20th Century Slave Narratives. Uni versity
of North Carolina Press, Chapel~H'i'Tl.
Fairbanks, Charles H.
197.4 The Kingsley slave cabins in Duval County, Florida, 1968.
Conference on Historic Sites Archaeology Papers 7:62-93.
1977 Backyard archaeology as research strategy. Conference on
Historic Site Archaeology 11:133-139.


84
banishment to hard labor at "Experiment" (Roswell King, Sr., 6 March
1813) or "Five Pound" (Kemble 1961 (1861):270) was prescribed for chronic
offenders.* Punishment also entailed the denial of fish, meat, molasses,
clothes, rice, and forced labor on Sundays (Roswell King, Jr., 29
Fehurary 1829). Children were submerged in cold water as a means of
punishment. On one occasion a young boy died shortly after he was dumped
in water (Roswell King, Jr., 3 February 1829).
For the most part, overseers' letters provide very little informa
tion on the family or social life among the slaves. This is unfortunate,
since this is an area in which archaeological data supply little, if any,
information. A few scattered references do exist regarding family
domestic quarrels, conflicts between drivers and workers at the slave
villages, and the religious fervor among slaves. Perhaps oral history
is the best resource of information for investigating these aspects of
slave life (e.g. Blassingame 1972, 1976; Kill!on and Waller 1973:
Escott 1979). It has been pointed out that overseers tended to write
about issues they thought the planter should know about (Bassett 1S6S:
261). Consequently, overseers' letters generally reflect concerns re
lating to slave illness, death, and disciplinary problems. Given these
concerns, it should be obvious that gaps would exist in interpretations
of slavery derived solely from this resource. Although archaeology may
not be able to provide data relating to the social or religious life of
slaves, it can shed light on the specifics of slave material conditions.
*In the plantation records there is no reference to the planta-
tion"Five Pound" but there are a number of references to "Experiment."
Because they were both apparently located on Little St. Simons, the
author is of the opinion that these names refer to the same plantation.


V .. - "" ?V.- ....
' .
Dedicated to the memory of my father William
Singleton. Sr


57
Figure 6.
I he Subdivision of Butler and General's Islands
Tenant Plots (circa 1883).
i nto


226
Artifact Class
# of Items
Cutlery
Bone handles
Knife blades
Ceramics
Coarse red earthenwares
Unglazed
Black lead glazed
Brown lead glazed
Colono-ceramics (slave made)
Refined earthenwares
Unclassified white
Whiteware, undecorated
Transfer print, pearl ware
Edged ware
Shell edged pearl ware
Molded pearl ware
Handpainted
Pearl ware undergalzed blue on white
Pearlware polychrome
Whitewa re polychrome
Annular
Circular bands
Pearlware
Whiteware
Yellow paste
Rouletted
Pearlware
Whiteware
Finger painted
Pearlware
Whiteware
4
2
2
13
13
298
217
17
48
11
26
11
87
28
2
2
2
1
1
Mocha
Whiteware
Yellow paste
Marbelized
Sponge decorated
Stonewares
Alkaline glazed
Brown salt glazed
Unidentitied
9
1
2
2
22
15
10
Porcelain
1
Unidentifi ed
l
Total ceramics
863
Glassware
Dark green
317
Light green
35


236
Gibbs, Tyson, Kathleen Cargill, Leslie S. Lieberman, and Elizabeth Reitz
1980 An anthropological examination of slave nutritional adequacy.
Medical Anthropology: Cross-cultural Studies in Health and
Illness 4(3) (In press).
Glass!e, Henry
1968 Pattern in the Material Culture of Folk Culture of the Eastern
United States. University of Pennsylvania Press, Philadelphia.
Glassow, Michael
1977 Issues in evaluating the significance of archaeological re
sources. American Antiquity 44(3):413-420.
Goggin, John M.
1952 Space and Time Perceptive in Northern St. Johns Archaeology,
Florida. Yale University Press, New Haven.
Granger, Mary
1947 (Editor) Savannah River Plantations. Savannah Writers Project,
Gray, Lewis C.
IS41 History of Agriculture in the Southern United States to 1860.
Peter Smith, New York.
Greer, Georgeanna
1971 Preliminary Information on the use of the alkaline glaze on
stoneware in the South, 1800-1970. Conference on Historic
Site Archaeology Papers 4:155-170.
Grigg, David
1967 Regions, models, and classes. In Models in Georgraphy, edited
by Richard Chorely and Peter Haggett, pp. 461-501. Methuen,
London.
Gritzner,
¡978
Janet B.
[abby in the
Coastal Southeas
c:
The Culture History of an
American Building Material. Ph.D. Dissertation. Louisiana
State University, Baton Rouge.
Gunn, Victoria R.
1975 Hofwyl Plantation. Report on file, Historic Preservation
Section, Georgia Department of Natural Resources, Atlanta,
Georgia.
Gutman, Herbert G.
1977 The Black Famlly in Slavery and Freedom 1750-1925. Vintage
Books, New York. *
Haggett, P.
1965 Locational Analysis in Human Geography.
E. Arnold, London.


7
The slave village or community site was the locus of most slave
activities. Admittedly, slaves spent considerable time in the fields
and at other locations of the plantation complex. But it is doubtful
that these sites will provide archaeological data relating to slave
behavior. Another site perhaps central to the slave belief system is
the slave graveyard. Because of the sensitivity involved in excavat
ing burials, this archaeological resource is not treated here. Slave
graveyards, however, were at least located at Butler Island. Because
of the restricted movement enforced upon slaves, the slave village site
should embrace a major portion of slave behavioral variability. It
also preserves evidence for a broader range of cultural activities than
does a simple activity area such as a rice field or threshing floor.
A methodological departure of this study from regional research
strategies advocated by Binford and others is the degree of use of
probability sampling at all levels of analysis (Binford 1964). Proba
bility sampling is used in this study but minimally. Additionally, the
multivariate techniques of geographers are not used at all. Admittedly,
this methodological departure may present some serious disadvantages
in offering regional interpretations. The purpose of this study,
however, is to synthesize the presently available data recovered from
slave sites in coastal Georgia. The interpretations offered here
are tentative and, hopefully, will be tested in future regional studies
utilizing regional methods.
Reg!oral Approaches to_the Study_of Slavery
Defined cultural, spatial, and temporal variables form the
parameters for most discussions of Afro-American slavery.


216
artifact profile by the addition of the correction factor may not be
the best procedure to correct for the disproportionate nail frequencies.
South makes similar adjustments to his Carolina pattern (1977:104-106).
And herein lies the major defect with the artifact pattern recognition:
It offers no real solution to the problem of extraordinary dispropor
tionate artifact counts (Chance 1977:127).
Derived from the adjusted Kingsley profile and the artifact pro
files from the other sites in Table 17, a slave artifact pattern is sug
gested, as indicated in Table 18.
Table 18
The Slave Artifact Pattern in Coastal Georgia/Florida
Artifact Group
Mean Percent
Percentage/Range
Architecture
70.78
67.90-73.23
Kitchen
24.34
20.01-25.77
Furniture
.02
.00-00.09
Guns
.14
.04.00.23
Clothing
1.03
.34-01.68
Personal
.09
.05-00.17
Tobacco
3.32
.28-09.70
Activities
.28
.22-00.39
Total
100.00
The slave artifact f
pattern reveals that the arch'
itecture group is
the overwhelming group of the artifact pattern. This is evidently


Nondomestic Food Resources
Traditional historic studies of slave diet have tended to over
look the potential contribution of nondomestic food resources. A recent
study of slave diet demonstrated that slaves residing within the coastal
eco-zone were able to make their diet adequately nutritious. Primarily
through the slaves' own efforts to supplement plantation rations with the
abundant food resources available within the coastal environment,their
diet was made nutritious (Gibbs et al. 1980). The study relies heavily
upon archaeological and zooarchaeological data uncovered from slave sites
in Georgia and Florida for evidence of food supplements. All of these
sites were formerly barrier island, long-staple cotton plantations which
lie within the marsh and lagoon section of the tidewater (see discussion
on pages 20-22). Slaves living on these plantations were able to ex
ploit a variety of habitats. Among them, the salt marshes form the
richest portion of the coast, and the archaeological evidence from long-
staple cotton plantations indicates that this habitat was most fre
quently exploited by the slaves (Fairbanks 1974:87; Otto 1975:345).
On the other hand, slaves on deltaic rice plantations were perhaps
limited to the variety of species that could be exploited. The delta,
being freshwater and to a lesser extent brackish, appears to have fewer
species of wild resources than the salt marshes. It is hypothesized
that the nondomestic species utilized by rice coast slaves would be
food items typically found in deltaic habitats.
Table 15 provides the relative weight of nondomestic resources
by animal class recovered from Butler Island. It indicates that fishes
and turtles were predominant while birds and mammals were less frequent



pound. The decade before the Civil War was a period of rising prices
after a long period of depressed prices. At that time, the average
price of cotton was between $.37 and $.47. However, after the Civil
War, the quality of the crop deteriorated and its production became less
valuable. Machinery was soon developed which produced thread from the
short-staple as delicate as that made from the long-staple (DeBows
Review 1867:83). Finally, devastation from the boll weevil made the
production of the crop totally unprofitable (Brown and Ware 1958:69).
Summary
Rice and long-staple cotton are two very dissimilar cultigens.
Each crop requires distinctive modes of cultivation and processing. For
the most part, each is cultivated in a very specialized habitat and is
produced for its own peculiar market. Yet analogies between the two crops
exist in the history of their production in the Georgia tidewater.
Both crops were restricted to a synonymous geographic range and
required a considerable labor force. These similarities were particularly
significant to the demography and economy of coastal Georgia. The
Georgia tidewater, like that of South Carolina, became a region with
a heavy concentration of slaves and a few wealthy slaveholders where
absenteeism frequently prevailed.
Another analogy was evident in general cultivation practices.
Often, sea-island cotton growers were also rice growers or former rice
growers. Apparently, certain practices established for rice, the
older staple, were frequently adopted for cotton. Some of these, though
harmless to the rice, proved to be detrimental to the cotton industry.


159
fragments contained rum made at Butler Island. Yet it is doubtful that
all alcohol consumed by the slaves was issued to them. The overseers'
concern about slave overindulgence in alcoholic consumption attests to
this. Some alcohol was probably purchased by the slaves. Perhaps the
light green "champagne" bottle (see Switzer 1974:23-24) was such a pur
chase.
Unfortunately, the former contents of recovered medicine vials
could not be ascertained. A variety of medicines was regularly ordered
for the treatment of slave diseases (see page 82) and presumably these
were kept at the plantation dispensary. The remains of medicine vials
uncovered from the slave quarters may have been plantation issues or
slave purchases.
Whether slaves were issued liquor or medicines or purchased these
items for themselves, the archaeological evidence indicates that they
were regularly consumed by them. Although alcohol was possibly con
sumed for its medicinal effect, historical evidence indicates that it
was used as a lei sure-time activity, particularly during the holidays.
Cooking and processing equipment
Several curve-shaped iron fragments suggestive of pots or kettles
were recovered. Two kettle legs, however, were the only diagnostic
artifacts of cooking equipment. These were evidently legs to trpoda!
caldrons, which often weighed up to 60 pounds when empty (Booth 1971:
38). Iron cooking pots were often supplied to the slaves for cooking
purposes, and remains of these pots have been uncovered at other slave
sites as well (Ascher and Fairbanks 1971; MacFarlane 1975:106). One
leg was found in the hearth of structure four, which further supports
u.
1
-
& -
:T' y-' >-,!< f "r


The only remains uncovered of the rice barn were its cypress foundations.
No artifacts were found in association with it.
The function of structure six, the brick floor, could not be
determined. Unfortunately, the 1315 inventory does not list any other
buildings at settlement #4 that would have been used for rice production.
Therefore, it must have been added after 1815. It is possible that it
may have been used as a threshing floor, since evidence of a super
structure was not identified with the floor. But no indications of brick
threshing floors have been found in the plantation records. All refer
ences indicate the use of tar threshing floors. Moreover, a soil analy
sis taken from below the floor suggests that the brick floor was super
imposed upon an earthen floor. From small test pits made to deter
mine the dimensions of the floor, numerous soil concretions resembling
hardpan in texture were recovered. The analysis suggested that these
concretions may have been part of an old earthen floor, and what
appeared to have been straw remains were found in these concretions..
This interpretation is plausible, however, only if the distribution of
these concretions was restricted to a limited area (see Appendix 1).
Because this soil type was limited to the area around structure six,
it is highly probably that these materials represent the former presence
of an earthen floor. In rice processing, earthen floors were often
associated with winnowing houses, but no archaeological evidence of
winnowing posts was identified. Perhaps, the floor was part of a shed
or warehouse. The possibilities regarding the function of the floor-
are numerous. But the size of the structure and its relation to the
rice barn strongly suggest that it was a specialized structure,


88
Figure 8.
Present-Day View of Butler
construction. 1-95 crosses
ment #2 and US-17 coincides
Island. (Note: Modern
approximate location of
with historic canal.)
hi ohway
settle-


rice plantation in colonial Georgia had an average of 43 working hands,
but the average acreage was not recorded (Flanders 1933:43).
Given this labor requirement, a demand for African slave labor
developed. Prior to 1635, most slaves in colonial South Carolina had
come there with their masters from the West Indies. These Afro-Americans
made notable contributions to settling the frontier of South Carolina
(Wood 1974:27-34). It was the rice industry, however, which estab
lished a need for black slave labor. As the new staple became profitable,
considerable quantities of Africans were imported (Wood 1974:57).
Africans were thought to be better equipped to labor in rice fields than
either indentured whites or Indian slaves for several reasons (Wood
1974:37-42). Perhaps, two of the most significant were the African
adaptability to the swampy lowlands and the African knowledge of rice
growing.
Malaria thrived in the stagnant, swampy lowlands where rice was
cultivated (Childs 1940:32). Settlers to the Carolina-Georgia coast
frequently complained of the prevalence of "fevers." Malaria or
intermittent fever" was considered to be the worst among these, par
ticularly during its pernicious season in the Spring and Summer (Childs
1940:33).
Only in recent years has medical research shown that populations
exposed to malarial climates display a genetic trait which produces a
partial but heritable immunity to the disease. This trait, referred .to
as sickle-cell trait," is not harmful in itself, but if present in both
parents can produce offspring who suffer from sickle-cell anemia and
are likely to die in adolescence (Wood 1974:80; Savitt 1978:27-28).


SOI N
123


242
Noel-Hume, Ivor
1962 An Indian Ware of the colonial period. Archaeological Society
of VIrglnla Quarterly BulletIn 17(1):1 -68.
1973 From crearnware to pearlware: A Williamsburg perspective.
In Ceramics in America, edited by George Quimbys pp. 217-
254. University of Virginia Press, Charlottesville.
1974 Historical Archaeology. Alfred A. Knopf, New York.
1976 Artifacts of Colonial America. Alfred A. Knopf, New York.
Olmsted, Frederick L.
1968 A Journal in the Seaboard Slave States With Remarks on Their
Economy "(18567. Negro Universities Press, Detroit.
Olsen, Stanley
1963 Dating early plain buttons by their forms. American Antiquity
28(4):551 -554.
Otto, John Solomon
1S75 Status Differences and the Archaeological RecordA Comparison
of Planter, Overseer, and Slave Sites From Cannon's Point
Plantation (1794-186f), St. Simons Island, Georgia. Ph.D.
Dissertation. University of Florida, Gainesville.
1977 Artifacts and status differencesa comparison of ceramics
from planter, overseer, and slave sites on an antebellum
plantation. In Pvesearch Strategies in Historical Archaeology,
edited by Stanley South,"pp. 91-116. Academic Press, New York.
1979 Slavery in a Coastal Community Glynn County (1790-1860).
Georgia Historical Quarterly 64(2):161-168.
Owens, Leslie H.
1975 This Species of Property: Slave Life and Culture in the Old
South. Oxford University Press, New York.
Panshin, A. J.
1964Textbook of Wood Technology. McGraw-Hill, New York.
Perkerson, Medora
1952 White Columns in Georgia. Macmillan, New York.
Peterson, Roger T.
1947 A Field Guide to Birds of the United States. Houghton
Mifflin, Boston.
Phillips, Ulrich
1929 Life and Labor in the Old South. Little, Brown, and Company,
Boston.
1969 (Editor) Plantation and Frontier. Burt Franklin, New York.


187
used for shirts and blouses. Nine of the porcelain buttons are plain
white. Two of the remaining three are known as "small china buttons,"
one with a. blue transfer print design known as "calico" and the other
with a painted blue-on-white design known as "bullseye) (Luscomb 1974:
23-24)(see Figure 34). These may have been used specifically on
female apparel. A dark green porcelain button identical, to the plain
white ones was also found.
Bone buttons form the third class. These appear to have been
made by machine and not on the plantation. Three 5-hole Type 19 (South
1964),or Type J (Olsen 1963),and one 4-hole Type 20 (South 1S64) were
recovered. Bone buttons were strictly utilitarian end were most fre
quently found on underwear and trousers (Luscomb 1975:26).
Identical button styles have been found in all the previously men
tioned slave contexts but in different frequenciespossibly due to varia
tions in sample sizes. But the predominance of metal flat disc buttons
at Butler Island seems to suggest that they were purchased specifically
for slave clothing, possibly for both males and females. On smaller
plantations, it would seem that slaves were more apt to receive hand-me-
down clothing from the planter and his family. Yet the si mi 1iarities in
button styles indicate that slave clothing was virtually the same in
diverse plantation contexts.
Footwear
Fortunately, the preservation conditions at Butler Island have
resulted in archaeological evidence of slave footwear. Slaves, both
men and women, were provided with work shoes known as "brogans." (Stampp
1956:291). Three complete or nearly complete shoe soles and numerous


87
modifications at Butler Island are shown in Figure 3. The most obvious
modifications are U.S. Highway 17, which roughly coincides with the
canal east of Settlement #1 and Interstate 95, which crosses the island
in the vicinity of settlement #2. Also, twentieth-century buildings,
from the Huston occupation, are present at settlement #1. These are
presently used, for waterfowl management activities.
It was possible to predict potential areas for site location by the
use of aerial photographs (of course with photographs having better
resolution than that: of Figure 8). Because Butler Island is generally
a low-lying, grass-covered marshland, sites tended to be located on
infrequent high spots, which are presently covered with trees and shrubs,
notably, hackberry, persimmon, and fig. These areas of tree groves
were easily identified in the aerial photographs. From surface recon
naissance, however, not all of these tree clumps were found to be sites.
But ail sites located were associated with this tree growth. Addi
tionally, some sites, such as the slave settlements, were also indicated
in aerial photographs by the detection of antebellum drainage ditches,
which surround and separate these from former rice fields.
The identification for several sites is well documented. The
slave settlements correspond exactly with the area designations I, II,
III, and IY on the 1877 USGS Map (or Figure 6). Historic photographs
provide suggestions for the placement of structures designated in slave
settlement #1. The steam-powered mill is precisely identified by ruins
of the two chimneys which are presently standing. A sketch drawn by
Roswell King pinpoints the location of the sawmill and both tidal mills
(Roswell King, Sr., 13 January 1813).


115
Figure 17.
Tidal Mill and Boat Landing


51
supporters believed that the slave states needed this type of represen
tation in order to prevent the possibility that a majority of representa
tives from the free states would attempt to abolish slavery. The Major's
view epitomized that of the small but powerful elite who desired to
perpetuate and expand the institution of slavery (Scott 1978:7).
In 1790, Polly died and the Major took his family and slaves to
Georgia. Like many South Carolinians, the Major moved southward in
search of new lands. The developing market for sea-island cotton and
the revival of the rice industry made the fresh, fertile lands of
Georgia attractive to the immigrants.
By 1799, Major Butler had acquired several thousand acres in
McIntosh and Glynn counties. Included in these were Butler Island,
Butler Point on St. Simons Island,* Little St. Simmons Island, and
pine land on the mainland known as Wocdville (Butler Estate Papers,
1799)(see Figure 5 for locations). In addition to land, the major
also acquired such a vast number of slaves that by the time of his
death he owned close to 700 (Butler Estate Papers, 1821). Eventually,
the Major became one of the wealthiest planters in Georgia.
The Major built a fine mansion at Hampton plantation on St. Simons
Island, where he spent most winters. It was here that Aaron Burr sought
refuge after the duel in which he killed Alexander Hamilton. Burr
visited the plantation for a month and he described it in many of
his letters to his daughter, Theodosia (Van Doren 1929). The Major
Concurrent with the
archaeoloqical
research at
4-,
Butler Island,
similar research has been undertaken at several sites of the Hampton
plantation (see Mullins, 1978; Fairbanks and Mull ins-Moore, 1980; anc
Mullins-Moore, nd).


11
With a few exceptions, archaeological studies of black Americans
are of two types: slavery sites and northern free black sites.
Because of the preliminary quality of this research area, most of these
studies have been more descriptive than interpretative. Within this
year, however, an attempt has been made to synthesize the findings from
several Afro-American sites (see Schuyler 1980). From these studies
and other published works, two themes dominate this research interest:
first, the search for African retentions in black American material
culture (Schuyler 1980:2), and second, the recognition of Afro-American
subsistence patterns.
It was the search for material correlates of African survivals as
well as subsistence information which initiated preliminary testing at
the Kingsley slave site by Fairbanks in 1968 (Fairbanks 1974). Un
fortunately, "no surely African elements" were identified (Fairbanks
1974:90). Since that time studies of slavery sites in the Old South
have been directed toward the definition of subsistence patterns
associated with the socioeconomic status of slavery (Ascher and
Fairbanks 1971; MacFarlane 1975; Otto 1975; Drucker and Anthony 1979).
Investigations of slave burials in Barbadoes have provided very
convincing evidence of Africanisms (Handler and Lange 1978, 1979).
At the Newton planation, the orientations of slave bodies and the
associated burial goods are indicative of West African burial patterns.
Burial practices peculiar to Afro-Americans in the United States have
been reported but only at sites occupied by freed blacks (Cate and
Wrightman 1955:207-215; Combes 1974; Crosby and Emerson 1979).


218
among the sites. Its percentage range suggests that few artifacts
related to slave craft activity or specialized farming equipment have
been uncovered.
The slave artifact pattern suggested here is very tentative as it
only relies upon data taken from four sites. Differences in collection
techniques as well as artifact analyses may have skewed artifact fre
quencies. In spite of these difficulties and problems inherent in the
pattern recognition method, it seems that artifact pattern recognition
can be a useful tool in making interpretations regarding Afro-American
slavery subsistence patterns. Hopefully, future pattern recognition
studies in coastal Georgia will refine this suggested regional artifact
pattern.


41.
planting (Gray 1941:731). High ridges 4 1/2 to 5 feet apart were con
structed before the seed was planted. These were usually made with
hoes, particularly in the old "sea-island region" (Gray 1941:735).
Plows came into wider use during the late antebellum period, when more
time was needed to manure exhausted soils (Gray 1941:734). Between
late March and the middle of April, the seed was sown. Once the cotton
plant sprouted, it had to be "thinned" with hoes several times. Gener
ally cultivation consisted of 4 to 8 hoeings, and grass was pulled by
hand.
With a few notable exceptions, most sea-island cotton planters
did little to enhance the productivity of their crop prior to 1830. At
that time, confronted with fluctuating prices and soil exhaustion,
experiments with manuring and crop rotation were undertaken (Johnson
1930:51-54). Manuring soon came into wide use,and various types of
manures were utilized such as marsh mud, crushed shell, and guano. The
practice of crop rotation, on the other hand, was limited to a few enter
prising planters (Johnson 1930:60-64). Careful selection of seeds was
perhaps the universal practice adopted by most planters to improve the
quality of the cotton. Several varieties were known and some reputed
to yield cotton that sold between $1.50 to $2.00 a pound. Unfortunately
the production of these seeds was kept secret and this knowledge is now
lost to agriculture (DeBows Review 1867:87).
Harvesting, Processing, and Marketing
A great deal of care had to be taken while harvesting the cotton.
The fiber was gathered as soon as the pod opened in order to protect it
from injury. Also, trash, leaves, and dirt had to be separated from the


these artifacts were found to be ceramics and glass pertaining to food
serving and storage, food preparation and processing artifacts were
also found, which has been the case at other slave sites in Georgia as
well. Presently, the archaeological evidence from Butler Island and
elsewhere in Georgia strongly suggests that most culinary activities took
place in the slaves' cabins and not at historically documented central
kitchens. Thus far, no central kitchens have been identified archaeologi
cally; therefore, it may be questionable whether they existed. Perhaps
one cabin served as a kitchen for feeding slave children, who were often
fed separately from adults (Genovese 1974:507-508), but archaeological
evidence of this very specialized activity is lacking. At present, the
archaeological evidence from coastal Georgia points to the slave cabin as
having been the locus for most cooking and eating activities.
Ceramics
More attention has probably been directed toward the study of
ceramics then any other artifact class in archaeological contexts. Con
sequently, ceramics have become vital indices for the documentation of
human activities (Fontana 1973). At slavery sites the following tenta
tive patterns have been observed:
1. A varied range of types and styles suggests that slaves were
not furnished with a special class of wares but obtained
either planter castoffs of unmatched assortments of ceramics
marketed for slave use.
2. Cups and bowls and to a lesser extent plates and platters
are the primary vessel forms of slave ceramic assemblages.


237
Hall, Margaret
1931 The Aristocratic Journey (1828), edited by Una Pope-Kennessey.
G. P. Putnam & Sons, London.
Hamilton, Jennifer
1980 Early history and excavation of Woodmanston plantation.
Unpublished M.A. Thesis. University of Florida, Gainesville.
Hammond, M. B.
1897 The cotton industry: An essay in American economic history,
part I. Publications of the American Economic Association
#1, New York.
Handler, Jerome and Frederick W. Lange
1978 Plantation Slavery in Barbados: An Archaeological and
Historical Investigation. Harvard University Press, Cambridge,
Massachusetts.
1979 Plantation slavery in Barbados, West Indies. Archaeology
32(4):45-52.
Hanson, Lee H.
1971 Pipes from Rome, New York. Historical Archaeology 5:92-99.
Harris, Marvin
1968 The Rise of Anthropological Theory: A_History of Theories_of
Culture. Thomas Crowell, New York.
Herskovits, Mervilie, J.
1958 The Myth of the Negro Past. Beacon Press, Boston.
Heyward, Dubose
1932 The Negro in the low country. In The Carolina Low Country,
edited by Augustine T. Smythe, pp. 171-191. William Morrow,
New York.
Heyward, Duncan C.
1937 Seed from Madagascar. University of North Carolina Press,
Chapel Hill.
Hilliard, Sam
log Meat and Hoe Cake: Food Supply 1n the Old South 1840-18¡
lincis University Press, Carbondale.
ar rice plantation: An ingenious adaptation to
Geoscience and Man: Coastal Resources 12:57-66.
1372
Hog Meat
Southern
1975
The tidev
nature.
Ho
i
Frank and Robert F. Keizer
Prehistoric Archaeoloqy: A Brief Iotrodectiojn.
hart, and Winston, New York.
iolt, Rine-
Holmes,
1878
D.
Comments on Fanny Kemble
13 September. Newspaper
at Athens.
s J o u r nal. D a ren Ti mb e r_ Gazette,
on Micro-film. University of Georgia


12
Perhaps, the most striking and yet controversial evidence (see
Schuyler 1980:2) of Afro-Americans attempting to recreate their African
past has been uncovered at the Parting Ways site (Deetz 1977:135-154).
This site near Plymouth, Massachusetts, was occupied by four Revolution
ary War veterans from the late eighteenth to the mid-nineteenth centuries.
Deetz suggests that the settlement pattern, house floor plans, a utili
tarian earthenware, and possibly the culinary practices are derived from
an African past.
The recent suggestion that black Americans influenced and manu
factured some Colono-Indian ceramics (Ferguson 1980) is another example
of this search for African survivals. These ceramics occur in high
frequencies at seventeenth-and eighteenth-century southern sites, speci
fically in the coastal areas of the Carolinas (Ferguson 1980). Previous
to this suggestion, these ceramics were thought to have been made
exclusively by native Americans (Noel-Hume 1962). It is well documented
that some of these ceramics were made by native Americans and some are
still made by them today. Indications that blacks may have had a hand
in the manufacture of these ceramics include the possible similarity of
this ceramic style with West African ceramic traditions, the infre
quency cf Colono-Indian ceramics at historic Indian sites, and its
high frequency at black occupied sites (Ferguson 1980). But the
fact that Colono-Indian ceramics have been uncovered, in most cases,
where native American populations were nearby* makes the suggestion
that blacks manufactured these ceramics uncertain.
^Charles H. Fairbanks, personal communication 19/9, Gainesville,
Florida.


213
fishing equipment have been assigned to activities. South's categories
have been used because they may be more insightful for assessing the
importance of personal artifacts, furniture, and arms. Also, the slave
artifact pattern could later be compared with other ethnic artifact
patterns.
Each site, however, was excavated by different investigators, and
collection techniques varied. Also, artifact analyses were conducted
by a variety of individuals. Unfortunately, very little could be done
to correct for these difficulties, and it must be assumed that these
differences have not severely altered the artifact frequencies. Most
artifact counts were obtained from published reports. Occasionally,
however, artifacts needed to complete the artifact groups were not quan
tified. In these cases, the author examined the original classification
cards, all of which are located at the Florida State Museum.
Table 16 provided the artifact profile for each site. It reveals
that the percentage range for Butler Island and the two Cannon's Point
slave sites are very similar for most artifact groups. The Kingsley
profile is very dissimilar with regard to architecture and kitchen
groups. The raw artifact count for Kingsley's architecture group is
considerably lower than that of the other three sites. This is pos
sibly related to the fact that nails constitute a large portion of the
artifact group at Butler Island and Cannon's Point, where slave dwell
ings were frame constructions. The nail frequencies at both Butler
Island and Cannon's Point are considerably higher than at Kingsley.
The dwellings at Kingsley were made from poured tabby and this explains
why few nails were'found.


60
This correspondence provides a wealth of information related to
the day-to-day activities at Butler Island and the other Georgian proper
ties. In many ways these letters resemble similar correspondence
written by overseers at other plantations with absentee landlords (e.y.
Easterby, 1945; Bassett 1368; Clifton 1978). All have in common a con
cern with planting, harvesting, and marketing activities; labor manage
ment problems; and the health and welfare of the slaves. Occasionally,
the Roswell Kings' letters also supply some indices of slave behavior.
Most often, however, only deviant behavior was reported, and therefore
these references tend to be biased and one-sided. Yet, despite this
limitation, the Kings' observations provide valuable insights into
slave lifeways at Butler Island.
As overseers, Roswell King, Sr., and later iris son, also named
Roswell, belonged to the "overseer elite" (Scarborough 1966:158-177).
This minority among overseers was frequently employed on the Carolina-
Georgia rice coast but. was found in other localities as well (e.g.
Sitterson 1943). For the most part, members of the overseer elite were
conscientious and efficient men. They made significant contributions
to agriculture, and some became planters themselves. Often, they were
the sons of planters. In contrast with their colleagues, they were
usually well paid. Both the Kings earned as much as $2,000 a year.
Roswell King, Sr., had come to the Georgia coast from Sharon,
Connecticut, in the 1780s (Lewis 1975:39). In 1802, he became the
manager of the Butler estate, a position he retained until 1819. Some
time after that date, he became the manager of the Darien Saw Mill
(Darien Gazette, 6 January 1821). Kings son, Roswell, Jr., was


15
Figure 1. Investigated Slavery Sites in Coastal Georgia.


33
elevated to form the permanent embankment. Fourth, small channels were
bridged and the trunks were installed. (Trunks were the devices which
controlled the water flow.) Fifth, individual fields 12 to 22 acres were
laid out. (Each of these fields could be filled or drained individually.)
Sixth, these larger fields were subdivided into plots of 1/4 or 1/3
acre to aid the water movement (Gray 1941:726-727; Doar 1970:9-13;
Hilliard 1975:59-60). These smaller subdivisions, sometimes referred to
as "tasks" (Gray 1941:734), provided not only drainage but also the
standard measurement for certain labor requirements (House 1954a:153).
Thus, the basis for the task labor management system, which was used
extensively in the cultivation of tidewater staples, was derived from the
small segments formed by the irrigation system used in rice culture
(Stampp 1956:55).
Regular maintenance of the irrigation system was essential in
order to keep the fields in proper working order. The ditches and
drains were thoroughly cleared of mud and trash annually (Austin 1393:
18). Breaks in the embankments caused by storms, high tides, and
"freshets" or floods were frequent, despite precautions taken to pro
tect the outer embankments (Spalding 1830; Southern Agriculturalist
1834). In addition, small animals including rats, snakes, and crayfish
often made holes on the outside of the bank which very often became
leaks and then breaks. If left unattended, these small breaks would
interfere with the drainage system (Doar 1970:30). Such breaks and
leaks had to be mended, and "ditchers" or "bank menders" (Heyward 1937:
236) were engaged regularly in repairing dykes.


49
family acquired Butler Island during the colonial period. Between the
years 1758 and 1775, Lochlan McIntosh purchased several large tracts of
marshland in the Altamaha delta (Jackson 1973:17-18). The precise iden
tification for some of these islands is unclear.;. Butler Island
may have been one of them.
Although the first owners of Butler Island are unknown, Major Pierce
Butler apparently made the first attempts toward reclamation and cultiva
tion of the island. While the family's ownership was complex,* an
attempt has been made to summarize it (see Table 1). The island was in
the possession of the Butlers for close to 120 years, and several family
members have made this island a very significant historic site.
Born in Ireland, Major Butler came to America in 1766 as an
officer in the British Army/, h'e liked America and decided to settle
here. Later in 1771, he married Polly Middleton, the daughter of a
wealthy South Carolina planter. He and Polly had one son and four
daughters. During the American Revolution, Major Butler sided with the
revolutionaries and became an important political figure in South
Carolina. In 1787, he was a delegate to the Constitutional Convention
in Philadelphia, and he was elected twice as a senator from South
Carolina to the U.S. Senate (Scott 1973:53, 1978:6-7).
At the Constitutional Convention, Major Butler was one of the
principal proponents arguing that state representation in the House of
Representatives should be proportional to the total population, both
slave and free (Alden 1971:86-87; Scott 1973:6). Butler and his
*See Scott (1961:xx-xxi) for a genealogy of the Butler family.




173
collection, it is still possible for the purposes of this study to test
the postulates regarding domestic and nondomestic food resources at
slavery sites.
Domestic Food Resources
In Chapter 3, the diet of the Butler Island slaves is briefly
described (see Pages 67, 77, 82). The Butler family estate records
indicate that this diet consisted of corn, purchased pork and salted
fish, and occasionally rice and molasses. Chronic shortages of meat
rations, however, were known, and meat may not have been supplied to
the majority of slaves on a regular basis. In fact, the ditchers were
apparently the only slaves to whom a ration of pork was regularly supplied.
Most slaves seem to have been fed a great deal of salted fish. Occa
sionally cattle, sheep, and pigs were slaughtered to supplement the meager
rations.
Taken together the dietary information obtained from the plan
tation records and the archaeological record present a confused pi cutre
of slave foodways at Butler Island,, Although the plantation records
indicate that the slaves had very little meat, when it was made
available, pork was the primary domestic meat consumed by them. The
archaeological evidence, on the other hand, indicates that beef may have
been the primary meat in slave dietary remains; the relative weight of
the recovered cattle (Bos taurus) remains exceeds that of pig (Sus scrofa)
remains (see Table 14). The implication that beef contributed more to
slave diet than pork is in keeping with the faunal evidence from Cannon's
Point (Otto 1975:327-328, 332-334). Furthermore, beef often supplanted
pork rations on many coastal plantations, particularly along the rice


182
the south Atlantic coast. Two species are typically found in coastal
Georgia, the common mud turtle (K. subrubrutn) and the striped mud
turtle (K. bauri). Both are bottom dwellers, rarely seen at the water
line, and are fond of brackish 'waters (Ernst and Barbour 1972:50-59).
The genera of turtles which indues the cooters and sliders
(Chrysemys sp.) are found in freshwater lakes, ponds, and swamps. They
are gregarious and are often seen basking above the waterline on logs.
Their flesh is very palatable and is often eaten locally in the South
today (Ernst and Barbour 1972:157).
As a result of its distinctive marginal ridge on its carapace,
the Florida softshell turtle (Trionyx ferox) is easily recognized. It
occurs in all freshwater habitats but prefers deep waters with mud or
sand bottoms, and is occasionally found in brackish waters near the
mouths of streams (Ernst and Barbour 1972:266-272).
It is not surprising that turtles are well represented within
the faunal collection. All identified species could be easily procured
in the delta. Mud turtles will take a baited hook or they could have
been obtained with nets (Pope 1939:55), whereas the cooter and soft-
shell might have been collected on land as they were migrating or nest
ing (Larson 1970:26). A local informant who was around during the last
days of Georgia rice cultivation claims that "the softshell turtle and
terrapin were frequently caught by rice ditchers."* His observations
suggest that the prevalence of semi aquatic turtles in canals and ditches
would have made them easily obtainable food items for rice coast slaves.
^Rudolph Capers, Hofwyl Plantation Historic Site, Glynn County,
Georgia, personal communication, 1980.


246
Sutch, Richard
1976 The care and feeding of slaves. In Reckoning With Slavery:
A Critical Study in the Qu anti tat ive History of American Negro
Slavery, edited Paul A. David, pp. 231-301. Oxford University
Press, New York.
Swendlund, Alan C.
1975 Population growth and settlement pattern in Franklin and
Hampshire Counties Massachusetts, 1650-1850. Memoirs of the
Society for American Archaeology #30, Washington, D.C.
Switzer, Ronald
1974 The Betrand Bottles: A Study of 19th Century Glass and Ceramlc
Contalners. National Park Service, U.S. Department of the
Interior, Government Printing Office, Washington, D.C.
Tannenbaum, Frank
1949 Slave and Citizen: The Negro in the Americas. Alfred A.
Knopf, New York.
Thayer, Theodore
1957 Nathaniel Pendletons short account of the sea coast of Georgia
in respect to agriculture, shipbuilding, navigation and timber
trade. Georgia Histrica1 Quarterly 41:70-81.
Thomsen, Mortimer
1863 What became of the slaves on a Georgia Plantation: Great
auction of slaves at Savannah, Georgia, March 2nd and 3rd,
1859, A sequel to Mrs. Kemble's Journal. American Anti-
slavery Society, Boston.
Trigger, Bruce G.
1968 The determinants of settlement patterns. In Settlement
Archaeology, edited by K. C. Chang, pp. 53-78. National Press
Books, Palto Alto, California.
Trimble, S. W.
1969 Culturally accelerated sedimentation on the middle Georgia
Piedmont. Unpublished M.S. Thesis. University of Georgia,
Athens.
U.S. Department of Agriculture
1961 Soil Survey of McIntosh County Georgia. Government Printing
Office, Washington, D.C.
n DeBurg, William
i 070
\ y / y
The Slave Drivers: Black Agricultural Labor Supervisors in
the Antebellum South. Greenwood Press, London
Van Doren, Mark
1929 (Editor) Correspondence of Aaron Burr and His Daughter,
Theodosia. Stratford Press, New York.




'¡08
This was a modern-day shotgun shell, presumably the artifact of recent
duck hunting in the area. Unlike in settlement #1, no eighteenth-
century materials were recovered. This may be due to the fact that
settlement #4 was not constructed until 1803 (Roswell King, Sr., 22
October 1803). Thus, the date 1803 established a terminus post quern
for the site. The site was evidently not intensively occupied after the
Civil War, since postbellum artifacts were virtually absent. All indica
tions have suggested that the site is temporally antebellum, thereby
making it a slavery site. Three of the five slave dwellings were tested.
These structures and the excavated materials are discussed in Chapters
4 and 5.


The tidal mills were further identified by archaeological resources.
In 1977, wooden timbers, presumably parts of a waterwheel,were unearthed
following waterfowl management dredging activities. These timber frag
ments were found directly to the east (that is east of IKS. 17) of the
tidal mill designated in settlement #1. Evidently, these timber frag
ments were the only remains suggestive of a mill or other structure. In
other words, no other structural indications were evident. It was sug
gested by John Morgan, the archaeologist making the assessment at the
time, that these remains did not identify a structural site at that
location. He suggested, instead, that, the timber fragments may have been
deposited in that location by some other means, such as tidal action,
which could have caused loose mill fragments to be moved to the area;
activities associated with the construction of U.S. 17; or by an un
known agent.* The present author agrees with this interpretation. In
light of the historic evidence which indicates that a tidal mill was
located at settlement #1, Morgan's interpretation has been substantiated.
Evidently, these wooden timbers were part of that tidal mill structure.
The other tidal mill located on the southern edge of the island
was partially excavated. Evidently, the mill structure which housed
the milling machinery was built upon a rich deposit of oyster shell
associated with aboriginal ceramics and food bone remains. Tidal action
for operating the mill was supplied by a series of canals that were
controlled by floodgates and trunks. The water was stored in an impound
ing pond, and an undershot waterwheel was turned by the rise and fall
*John Morgan, Jr., Staff Archaeologist, Historic Preservation
Section, Georgia Department of National Resources, Atlanta, Georgia,
personal communication, 1978.


Table 16
Artifact Profiles for Four Slavery Sites in Georgia/Florida
Butler
Island
Cannon'
s Point3 S
Cannon'
s Point, N
Ki ngsley
Artifact Group
#
Of
10
it
%
#
7o
#
%
Architecture
4494
67.90
3824
71.38
3789
70.60
754
34.34
Kitohen
¡325
20.01
1388
25.91
1383
25.77
1384
62.97
Furniture
1
.01
5
.09

0.00

0.00
Guns
15
.23
2
.04
6
.11
10
.45
Clothing
111
1.68
45
.84
67
1.25
18
.82
Personal
5
.08
9
;i7
3
.05
5
.23
Tobacco
642
9.70
71
1.33
107
2.00
15
.68
Activities
26
.39
13
.24
12
.22
12
. 55
Total
6619
100.00
5357
100.00
5397
100.00
2198
100.00
214


243
Plog, Fred
1978 Cultural resource management and the "new archaeology."
In Social Archaeology: Beyond Subsistence and Dating, edited by
Charles L. Redman, Mary A. Berman, Edward Curtin, William
Langhorne, Jr., Nina M. Veroggi, Jeffery Winser, pp. 402 422.
Academic Press, New York.
Pope, Clifford H.
1939 Turtles of the United States, Canada. Alfred A. Knopf, New
York.
Postell, William D.
1951 The Health of SI aves on S o uthern Plant a ti o ns_. Louisiana
State University, Baton Rouge.
Potter, David M.
1932 The rise of the plantation system in Georgia. Georgia
Historical Quarterly 16(1):114-135.
Price, Cynthia R.
1979 19th Century Ceramics in the Eastern Ozark Border Region.
Center for Archaeological Research, Southwest Missouri"State
Uni versity, Springfield.
Prunty, Merle
1S55 The renaissance of the southern plantation. Geographical
Review 45(4):459-491.
Reitz, Elizabeth J.
1979 Spanish and British Subsistence Strategies at St. Augustine,
Rond a_ and Frederica. Georgia Between 1565 and 1783. Ph.D.
Dissertation. University of Florida, Gainesville.
Salley, A. S., Jr.
1919 The introduction of rice into South Carolina. Bulletins of
the Historical Commission of South Carolina, #6. The State
Company, Columbia, South Carolina.
Sammis, I. S.
1977 Pottery at Huntington, Long Island. In The Art of the Potter,
edited by Diana Stradling and J. Garrison, pp. 103-112.
Antiques Magazine Library, Universe Books, New York.
Sanders, William T.
nd Report on Excavations on Butler Island, July 1956. Report on
file, Archaeological Laboratory, University of Georgia, Athens.
Savitt, Todd L.
1973 Medicine and Slavery: The Diseases and Health Care of Blacks
in Ante be11 urn Virginia. University of Illinois Press, Urbana.


CHAPTER 7
SUMMARY AND CONCLUSIONS
The Butler Island data have suggested that the adaptations of
slavery to rice culture are archaeologically discernible in slave com
munity organizations, in the natural resources exploited, and to a
lesser extent in farming implements, tools, and slave crafts. These
patterns are presumably adaptations to the deltaic habitat in which rice
is produced.
On the other hand, the archaeological correlates for slave cloth
ing, plantation food rations, personal possessions, and leisure activi
ties were found to be similar to those found at long-staple cotton plan
tations in coastal Georgia. Subtle qualitative and quantitative varia
tions exist between the Butler Island data and those from other sites, but
the variations may be related to differences in plantation operation.
While it is possible that these slave material conditions existed
in other cash crop regions of the Old South, it is important to remember
that several factors in coastal Georgia favored a distinctive develop
ment in slave lifeways. Paramount among these factors were local ecologi
cal factors. The abundant natural resources of the coast, both in the
marsh and lagoon section and on the delta, were exploited for numerous
subsistence needs. Additionally, cultivation techniques and labor manage
rnent systems were adaptations to the coastal environment. Specifically,
the task labor system developed from polderland construction of the
219


Table 1
Ownership of Butler Island, 1790? to the Present
Time Owners
1790?-1822
Major Pierce Butler
1822-1036
John Butler and Pierce Butler, Jr.
(formerly John Mease and Butler
Me ase), bu t e s tate was adminis-
tered by the Major's daughter
Frances and his son Thomas.
1336-1847
John and Pierce, Jr., assumed full
responsibilities of ownership until
John's death in 1847.
1847-186!
Properties jointly owned by Pierce,
Jr., and John Butler heirs, but
slaves had been divided equally
between the two.
1861-1865 (Civil War)
Abandoned.
1866-1878
Frances Butler Leigh (Pierce
Butler, Jr., and Fanny Kemble's
daughter).
1978-1910
Leighs still own but have leased
the island to rice planters.
1910-1920
Passes through several hands, pri
marily rice growers.
1920-1949
Colonel Huston.
1949-1954
R. J. Reynolds.
1954-present
Game and Fish Commission, Georgia
Department of Natural Resources.


Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements
for the Degree of Doctor of Philosophy
THE ARCHAEOLOGY OF AFRO-AMERICAN SLAVERY IN COASTAL GEORGIA:
A REGIONAL PERCEPTION OF SLAVE HOUSEHOLD AND COMMUNITY PATTERNS
Gy
Theresa A. Singleton
December 1980
Chairman: Charles H. Fairbanks
Major Department: Anthropology
A variety of methodological and theoretical orientations has been
utilized in the study of Afro-American slavery. Regional approaches are
not new to slavery, but few studies have examined the impact of local
ecological factors upon slave 1ifeways within a narrowly defined area.
In this study, coastal Georgia (the barrier islands and the adjacent
deltas) provides the region for the exploration of an ecological approach
to slave life and material conditions. Specifically, slavery at a rice
coast plantation is compared with slavery at long-staple cotton planta
tions. Archaeological data provide the primary basis for interpretation,
and historical resources are used to supplement the archaeology.
Primary settlement patterns and subsistence data recovered from the
Butler Island rice plantation in 1978 and 1979 are used to investigate
slave life and material conditions on the rice coast. The Butler Island
data are compared with the archaeological data recovered from several
previously mves
mi gated slave sites of long-staple cotton plantations.
The comparison reveals similarities as well as differences in the
xn


45
f-'-'jf
It has been suggested that the task system provided slaves at
sea-island cotton plantations in coastal Georgia with sufficient leisure
time to improve their material lots (Otto 1975; MacFarlane 1975). Under
the task system, the quality of slave life could be greatly improved
through hunting and fishing, selling handicrafts or plantation produce,
or by growing foods to supplement plantation rations. Presumably, the
use of the task system at a rice plantation should have resulted in the
same conditions. But how similar were the conditions of slavery in two
very different crop loci? The adaptations of slavery to long-staple
cotton culture in coastal Georgia has been investigated utilizing both
documentary and archaeological resources (Otto 1975; MacFarlane 1975).
A similar investigation of slave adaptations to rice culture along the
coast of Georgia is presented in the following three chapters.


54
Fanny always wanted to publish the journal she had kept while in
Georgia. In 18593 she began giving this serious thought for two rea
sons: first, she was no longer married to Butler; second, the estate
she describes in the journal no longer existed (Scott 1973:119).
Pierce Butler had fallen into serious financial debt as a result of the
crash in 1857, and more than 400 slaves were sold to pay his debts
(Thomsen 1863). In light of these new circumstances, Fanny reasoned
that her journal would be thought of less in a personal way than as a
social statement of slavery in the United States (Scott 1973:120).
The Emancipation Proclamation and the events of the Civil War gave her
the final impetus toward publication. Fanny felt that the journal would
help "explain Lincoln's war aims and show the nature of the system that
the North was fighting to abolish" (Scott 1973:122). The journal was
published May 1863 (Kemble 1882:205).
It is difficult to estimate the reaction tc the journal imme
diately after its publication. A few Southerners have suggested that
the journal single-handedly swayed British opinion against the South,
which made aid to the Confederate War effort impossible (e.g. Perkerson
1952:133-134), Even among Fanny's admirers this opinion has prevailed
(Armstrong 1938:432). It has also been argued, however, that the
journal had little influence upon the contemporary public (Lombard
1930). Recent evidence has indicated that by the time the journal was
published, British opinion had already turned against the South
(Scott 1973:124). More significant than the criticism related to the
influence of the journal upon the contemporary public is the cri
ticism related to its accuracy and authencity. Among historians


Clothing
Few references to slave clothing viere found in the Butler planta
tion records. Cloth was generally purchased but sometimes was made on
the estate. 11 We are making our own cloth, the cotton has been spun and
Molly's son, James, put the loom up and we will weave it out" (Roswell
King, Sr., 15 November 1812). Woolen cloth and slave shoes were also
made from time to time. The only archaeological evidence for clothing
manufacture, however, was a pair of scissors recovered from structure
one.
Buttons
Metal buttons were the predominant clothing artifact, with brass
and whitemetal flat disc buttons being the-most frequently occurring,
particularly Type 7 (South 1964), or Type D (Olsen 1963), and Type 31
(South 1964). Of the total of 72 buttons recovered, 52 were of these
two types. The sizes of these buttons ranged from 10 to 24 millimeters,
which suggest that they were used for a variety of garments. Addi
tionally, the frequency of the flat disc buttons may indicate that
they were purchased specifically for slave clothing. In fact, refer
ences to "metal buttons" appear on the annual expense lists. The
occurrence of the remaining metal buttons was sporadic, possibly because
they came from clothing the slaves purchased, for themselves. These
metal buttons" include two Type 32 (South 1964) and one Type 9 (South
1964) with a hand-stamped design of a person on a horse.
Twelve porcelain buttons Type 23 (South 1964) were found. Al
though it has been suggested that porcelain buttons were used for
undergarments (MacFarlane 1975:135), it is more likely that they were


of rice, experiments were attempted with every conceivable threshing and
pounding device. By 1835, the machinery for processing rice included
an animal-powered mill, two tidal mills, and a steam mill. The steam
mill was built in 1833; it housed both threshing and pounding machinery
and its cost was over $20,000 (Roswell King, Jr., 25 March 1932, 22
April 1832, 22 February 1833). Evidently, the threshing mills were not
sufficient for the amount of rice harvested, and a portion of it was
threshed by hand on "tarred" threshing floors (Roswell King, Sr.,
9 August 1806; Kemble 1961 (1863):109-110). Rough rice was also mar
keted, but large quantities were not sold until the 1850s (Pierce
Butler, Jr., Ledger, 1851-1855).
Besides rice and cotton, sugar was planted at Butler Island be
ginning in 1812. In fact, Major Butler had considered "turning the
island into a sugar plantation" (Roswell King, Sr., 1 November 1812).
Between the years 1812 and 1815, complete sugar processing equipment was
added to the plantation complex (Butler Estate Papers, 1815a).
Roswell, Sr., had become very interested in sugar culture and he sought
the advice of West Indian planters (Roswell King, Sr., 12 August 1815).
This interest in sugar was later pursued by Roswell, Jr. (King 1823).
Regardless of their enthusiasm for sugar culture, the acreage allotted t
it was always considerably smaller than that of rice or cotton.
Like most large plantations of the Old South (6a11 man 1970:23),
some provision foods were produced at Butler Island. Livestock were
raised and some fruit trees and vegetables were cultivated. The list
of stock included oxen, barrows, mules, mares, cattle, pigs, sheep,
turkeys, and other fowl(Butler Estate Papers, 1803-1830). The draft


240
Lane, Mil lis
1973 (Editor) Charles Lye11. In The Rambler in Georgia, pp. 189-
201. Beehive Press, Savannah, Georgia.
Langhorne, William I., Jr.
1976 Mill based settlement patterns in Schoharie County, New York:
A regional study. Historical Archaeology 10:73-92.
Larson, Lewis, H., Jr.
1S70 Aboriginal Subsistence Technology in the Southeastern Coastal
Plain During the Late" Prehistoric Period. Ph.D. Dissertation.
University Microfilms, Ann Arbor, Michigan.
Lauer,
1969
Joseph Jerome
Rice in the history of the lower Gambia-Geba area.
M.A, Thesis. University of Wisconsin, Madison.
Unpublished
Lawson, Dennis T.
1975 No Heir To Take Its Place: The Study of Rice in Georgetown
County, South Carolina. The Rice Museum, Georgetown, South
Carolina.
Leigh, Frances Butler
1883 Ten Years on a Georgian Plantation, Since the War. Bentley
and Sons, London.
Leigh, John
1921 Other Days. Bentley and Sons, London.
Lewis, Bessie
1975 They Called Their Town Darien. Darien News, Darien, Georgia.
Lewis, Bessie and Mildred Huie
1974 Patriarchi a 1_P1antatlons o f_StSimons. L ewis and Huie,
St. Simons, Georgia.
1978 Hampton Plantation. Plantations of Coastal Georgia #1.
Argyle Publications, St. "Simons Island, Georgia.
Lewis, Kenneth
1377 Sampling the archaeological frontier: Regional models and
component analysis. In Research Strategies in Historical
Archaeology, edited by Stanley South, pp. 151-196. Academic
Press, New York.
1979 Settlement and activity patterning of two rice plantations
in the South Carolina low country. Paper delivered at the
20th Annual Conference on Historic Site Archaeology,
St. August-!ng, Florida.
Lombard, Mildred E.
1930 Contemporary opinion of Mrs. Kemble's Journal of a Residence
on a Georgian Plantation. Georgia Historical Quarterly 14(4):
335-343.


164
gas*


Figure 34. Clothing and Personal Adornments.
Designations
A.
Flat disc metal
button
B.
Bone buttons
C.
Porcelain button
is (bul
transfer print,
green)
D.
Glass beads
E.
Hematite bead


Ill
that provisions were accounted for and distributed by these settlements.
Despite the advantages of segregating a large number of slaves into
smaller villages, some planters of large productive units evidently
preferred to keep all their slaves in the same location. This was par-
ticularly true of interior plantations (e.g. Smith 1973:144, plat of
Glendownwer estate). Exceptions such as these were also to be found in
the Georgia tidewater. For example, a plat of the Hopeton rice planta
tion (see Johnson 1930:63; Prunty 1955:464) reveals that all "500 slaves"
(Lane 1973:196) were evidently located in quarters near the administra
tive nucleus. Here again, the land use practices associated with
Hopeton's unique polder!and construction may account for its arrangement
rather than a preference to have slaves nearby.
Within the slaves' quarters, dwellings were typically arranged in
rows along roads or streets forming a square or rectangle of buildings
(Prunty 1955:465). This pattern was established for tidewater planta
tions and later replicated in other areas of the South (Stampp 1956:
292). The basis for this arrangement appears to have been clearly func
tional as it permitted easy cleaning and inspection of slave dwellings
(Flanders 1533:153; Otto 1975:24) and perhaps easy access to them by
wagons or carts.
But plantation layouts were not just functional. These spatial
arrangements expressed the social structure of plantation society,
which reflects a chain of command of owners, managers, overseers,
permanent laborers, and seasonal workers (Wolf 1959:137). The large
plantation plan (which has been under discussion here) with the big
house, numerous outbuildings, offices, and slave villages was a standard


supports this interpretation. Additionally, the base of a brick wall
along the drainage ditch indicated that a rather large structure was
once present on the premises, and a scatter of bricks, presumably from
a chimney fall was also found. A padlock lying on the surface has been
found to be identical to ones uncovered at the slave site. It is be
lieved that the padlock and brick type (Savannah greys) suggest that
this component of the site is temporally antebellum. Unfortunately, its
antebellum function could not he determined. On the other hand, the
sites's postbellum function is well known by local informants. In the
twentieth century, the site was utilized for the manufacture of illegal
moonshine whiskey until it was discovered by the authorities in the
early 1950s. Archaeological evidence of this moonshine operation is
still apparent in the form surface debris, including still parts, corn
liquor crockery fragments, and Mason jars. This site also has a pre
history c component.
Prehistoric components were found to be associated with the
southern tidal mill, settlement #3, and the presumed mill site in the
northwest corner of the island. These prehistoric occupations are
identified by oyster shell and aboriginal ceramics ranging in time from
fiber tempered to complicated stamped. From the very limited samples
collected, most of these ceramics appear to be cord marked and checked
stamped. It is, however, premature to suggest a cultural association
from such slim evidence. A problem is presented by these prehistoric
deposits. It is not at. all certain whether the oyster shell was brought
in at a later time to make "high spots" in the swamp or was placed in
its present locations by prehistoric peoples. At the tidal ¡¡¡ill site,


71
Table 3
Annual Births and Deaths of Butler Slaves
Year(s)
Births
Deaths
1.302-1812
257
191
1819
29
19
1820
25
16
1821
19
27
1822
19
22
1823
32
14
1824
30
18
1825
20
28
1826
35
23
1827
41
39
1828
46
25
1829
31
16
1830
41
25
1831
45
20
1832
31
29
1833
43
34
1834
53
34
Total
756
580
Note; i
Natural Increase 176.


34
Usually the annual cycle began in the fall and early winter with
the "breaking" of the soil. This was done by hand with heavy hoes.
Plows were used by some planters, but the small subdivided rice fields
and the numerous ditches made plowing difficult. Moreover, many fields
were not dry enough and draft animals were expensive to maintain (Gray
1941:729; House 1954b:27-28).
Planting occurred beginning in the middle of May, depending upon
the weather (Austin 1893:19). Techniques varied among planters, but
either of two methods was generally employed. In the traditional method,
"covered rice" seeds were carefully sowed in trenches measuring 3 to 4
inches wide, 2 inches deep, and 13 to 14 inches apart. After sowing,
the trenches were covered with 2
inches of soil
by hoes, rakes, or
covering boards (Gray 1941:727-728). The other method, "open trench,"
required that the seeds be coated in clay before sowing. Upon comple
tion of sowing, water was let in the open trenches without covering them
with soil (House 1942a:188). Open trench planting was not widely used
in Georgia. It was apparently developed by South Carolina planters in
an attempt to increase yields from exhausted soils (House 1942a).
Once the seed rice was planted, inundation of the fields was needed
at various stages of growth. Irrigation practices, like planting methods,
varied according to the planter. All irrigation, however, was utilized
for three purposes: the initial flow or sprout flow was applied to aid
the seed in sprouting; subsequent flows known as stretch flows were
applied to keep down the growth of grass, weeds, and other pests; and
the final or harvest flow strengthened and supported the stalks bearing
ripened grain before the rice was harvested (House 1954b:3).


CHAPTER 5
THE MATERIAL CULTURE OF SLAVERY
The hypotheses outlined in Chapter 1 are based on the following
three assumptions concerning slave material conditions in coastal
Georgia:
1. Artifact classes identified as food-related activities, cloth
ing, personal possessions, and leisure, reflect general condi
tions in slave lifeways which may have been related to labor
management practices established for the production of tide
water staples. Thus, these conditions should be evident at
both rice and long-staple cotton plantations.
2. The habitats where certain cash crops were produced will be
evident in slave lifeways primarily in community organizations
(Chapter 4) and in exploitable food and nonfood resources.
3. Tools, farming implements, and specialized crafts should be
directly related to cash crop production.
In order to test these assumptions, it is necessary to refine the arti
fact groups indicated on pages 18 and 19. Functional artifact groups
are used here which consider material attributes when appropriate.
The following artifacts are included and are listed in the order in which
they are discussed:
1. Food-related activities
a. Kitchen artifacts (food preparation, processing, serving
storage, and containers)
140


6.8
to coastal Georgia (Ludlum 1963:116), and the Butler estate lost not
only crops but 19 slaves as well (Van Doren 1929:186). During the
British raids along the coast in the War of 1812, 139 slaves were taken
from the estate (Roswell King, Jr., 28 April 1315). In later years,
freshets and epidemics among slaves had far-reaching effects upon prof
its. Finally, the crash of 1857 resulted in the sale of 500 slaves.
It is possible that Major Butler found the losses of his day over
whelming. Historic records indicate that he put the entire estate up for
sale in 1818, a total of 15,000 acres and 535 Negroes (Darien Gazette,
1818-1819). The factors which motivated his actions are not clear.
Perhaps he was of the opinion that the estate was becoming unprofitable.
It is evident that Roswell, Sr., had begun urging the Major to sell por
tions of the estate in 1816: "I think you have too many eggs in one
basket to keep . the land is worn out, the droughts and gales make
our cotton the smallest I ever knew .... Your interest is too large.
. . Not enough laborers for the amount of land" (Roswell King, Sr.,
21 January 1816). In 1318, King's urging had become more persistent:
I have urged you to sell your estate for a long time out
of the purest of motives, thinking Negroes was above their
value and that was a time to sell, but I find I have
ered in judgement, Negroes are rising in value like the
tide, every day. ... I recommend you to sell all your
Negroes and land you hold in this state, say in a lump
for $555,000 and in my opinion in fifteen minutes after
you read this and if you will not sell I strongly recom
mend to you without loss of your time make arrangements to
lay out one hundred thousand in Negroes. Let them cost
what they mayas I have often observed your estate must
dwindle without more force. (Roswell King, Sr., 28 June
1 O 1 o \
i O i O j
Whether it was King's persuasion or other factors which resulted in the
decision to sell cannot be ascertained. The financial crashes between


Figure 33.
Food Procurement Artifac
Key to Artifacts
1. Lead fish net sinkers
2. Lead slag fragment
3. Buckshot
4. Gun flints
5. Brass trigger guard
I.'
t
£


y
I
1
O
LO
U5UI
LUIA.'

av s
'i A ,
- r > *v,-. s\ / .. >-
Sv:
:
-
-'K-'W?SW$4W;-
a
:vg£#|iS£v^SfeJ
i: " ."
\
' 'V; /./' ' ; '; ''
'- '- 1 ' V '' J, = \
.
¡¡pip
mm'Axm
Baiisaaaassr
3lpi#
!,V. LAB : i:
Cn?
rtv^'V
,. ,p.
mimmmm
; ; '-a
fet§iSi£a -- 5 /. :,:,,' a /a .'vvf
¡i*. -, a -a-va
a:. v; 4
m m¡
a:;;vvvv:a.V v/ ; %
I
- * -v:jM
i0lme!
V.V-. V:V
fSp'SfWfti
|:Vvvva'#';:jv'Sv
fW''a'sr -i;
feiWftiail
358$ r? f-V ^:j---Va.r-;
'.s'&v
i . ,
';% ;Al .
;
ftngra
\ &^KsSdSS,'iiuSj"%3t^i^¡:¡
¡


245
Smith, Samuel D.
1977 Plantation archaeology at the Hermitage: Some suggested
patterns. Tennessee Anthropology 2:152-163.
South, Stanley
1964 Analysis of buttons from Brunswick Town and Fort Fisher.
Florida Anthropologist 17(2):113-133.
1977 Method and Theory in Historical Archaeology. Academic Press,
New York.
1978 Pattern recognition in historical archaeology. American
Antiquity 43(2):223-230.
Southern Agricultural 1st
1834 On the protection and repairs of rice embankments. 7:517-519.
Spalding, Phinizy
1977 Part One: Colonial Period. In The History of Georgia, edited
Kenneth Coleman, pp. 9-70. Univeristy of Georgia Press, Athens.
Spalding, Thomas
1830 Observations on the bordering of river banks with orange trees
and protecting the embankments with Bermuda grass. Southern
Agriculturist 3:73-96.
1835 On the culture, harvesting, and threshing of rice and in the
rust in cotton. Southern Agricultural 1st 8:169-174.
1845 Culture of rice. Amen can Agriculturai1st 4:53-55.
Stampp, Kenneth M.
1955 The Perculiar Institution: Slavery in the Antebellum South.
Alfred A. Knopf, New York.
Steckel, Richard H.
1379 Slave mortality: Analysis of evidence from plantation records.
Social Science History 3 and 4:86-114.
Steward, Julian H.
1955 The Theory of Culture Change:_ The Methodology of Mill til inear
Evolution. University of Illinois Press, Chicago.
Stradling, Diana and J. Garrison
1977 The Art_of the Potter: _Redware and Stoneware. Antiques
Magazine Library, Universe Books, New York.
Sireover, Stuart
1971 Comments on archaeological data requirements and research
strategy. American Antiquity 36:9-19.


no
Generally, the spatial aspects of plantation communities were
similar throughout the South. At the 11 administrative nucleus" were
located the planter's or manager's house, service buildings, and the
dwellings of house or skilled servants. Processing equipment and
storage houses formed the "technical nucleus" (Wolf 1959:137). Ap
parently, convenience provided the key for the location of technical
equipment (Smith 1936:22) since it was placed at the administrative
center, or located separately, or associated with slave quarters
(Prunty 1955; Otto 1975:18-19). On small plantations, slave dwellings
were located near the plantation house, but large plantations tended
to have one or more slave villages located a considerable distance
away (Stampp 1956:272).
Reasons for the division of slaves into separate villages are not
clear. Polderland construction at rice plantations may have necessi
tated this pattern. The placement of structures or the centers of plan
tation activities was often dependent upon the availability of high
lands at rice plantations. Structures were located either on hammocks
adjacent to rice fields (Smith 1936:22), or high areas had to be
created. But the advantages of a close proximity to the fields and the
danger associated with a large assemblage of black people (Scott 1973:
75) better explain the prevalence of the segmented slave village pattern
in other cash crop regions. The Butler family plantation records pro
vide further suggestions for the establishment of this pattern. Slave
villages were efficient' means to keep account of a large number of
ss aves.
Slaves were listed and the amounts of crops planted or har
vested were often recorded by slave villages. It is also possibl<


117
hospital, which occasionally served as a chapel and recreational center,
was located there (Kemble 1961 (1863):131).
Each of the remaining slave villages contained technical equipment
for the processing and storage of plantation staples, in addition to
slave dwellings. As indicated from the archaeological evidence from
settlement #4, these structures were located opposite the slave dwellings
forming a small technical nucleus. Settlement #2 contained long-staple
cotton processing equipment, and the slaves living there were possibly
engaged only in the production of cotton. Both settlements #3 and #4
contained rice storage facilities and were equidistant from sugar pro
cessing machinery (Butler Estate Papers, 1815a.; Roswell King, Sr.,
20 April 1816), thereby making these units equipped for the produc
tion of rice and sugar. Additionally, each settlement had its own plan
tation specialists who produced items to be used for planting activities
as well as for household needs.
To a certain extent, the slave villages at Butler Island repre
sent smaller self-contained productive units of a larger plantation com
plex. Frequently large areas of ricelands of 1,000 to 1,200 acres were
divided up into smaller units of 300 to 400 acres. These smaller units
operated as separate plantations (House 1954a:150). Apparently, planters
found these smaller units to be more efficient than larger ones. Thus,
the subdivision of slaves into villages on ricelands may have developed
from several factors, including polderland construction, economic effi
ciency, and social control. The impact these self-contained villages
had upon slave lifeways needs to be investigated. At Butler Island
these villages tended to isolate the slaves from the other centers of the


113
full-time production of cash crops for a world economy. Instead, it was
a retirement home and subsistence-based farm for its owners, which
utilized numerous slaves in the work force (Lovell 1932:110-111). Per
haps, this difference in economic function may account for Choclat's
distinctive spatial arrangement.
The Afro-American influence in plantation settlement patterns is
more difficult to document than that of the Euro-American planter. Al
though the evidence of African-styled architecture has been observed in
a variety of plantation contexts, notably in South Carolina, Virginia,
and Louisiana (Anthony 1976), the evidence of a similar influence in
plantation layouts is slim and remains untested. The possibility of
this influence, however, is suggested in the community layouts of slave
quarters at two antebellum plantations on the Atlantic Seaboard. The
arc-shaped arrangement of slave dwellings at the Kingsley plantation
(Fairbanks 1974) and a horseshoe-shaped arrangement recently uncovered
in coastal South Carolina* are suggestive of traditional African village
layouts found in several areas of Africa (Denyer 1978:20). At this
stage of research, however, such a suggestion is speculative and awaits
further investigation.
Because plantation settlement patterns have not been systemati
cally investigated, it is difficult to
determined these spatial organizations,
were designed to be functional to meet
enumerate the factors which
Undoubtedly, most arrangements
the environmental challenge and
economic needs.
It Is likely that cultural ideals,
ethnicity, and
*Patrick Garrow, Chief Archaeologist, Soil Systems Incorporated,
Marietta, Georgia, personal communication, 1379.


94
(it only becomes inundated during naturally occurring floods), thereby
making the preservation of organic materials better than that of the
other sites.* Third, the approximate locations of certain structures
at settlement #1 are 'well documented in historic photographs. However,
with the exception of the steam mill ruins, no surface indications of
these structures presently exist.
The areas in and around settlement #1 have been subject to con
siderable plowing and dredging. Consequently, a great deal of cultural
debris was found to be lying on the ground surface. The first objective
for fieldwork activities at settlement #1 was to walk the area, collect
the surface materials, and note the location of these scatters. It was
hoped that these collections would help to identify structural features
or other specialized activity areas. These collected materials ranged
from the mid-nineteenth century to the present. Unfortunately, they did
not provide indications for the location of features.
The next objective was to locate the two rows of slave dwellings
known to have existed at the site. It was decided to make a contour map
to help identify these slave dwellings. Presumably, a contour map would
indicate by subtle rises the location of structures. A very rudimentary
map was prepared, and it provided some indications for the approximate
location of the two rows. Two test trenches (1.5 x 4 meters), de
signated as Test 1 and Test 2, were laid out to test one of these subtle
rises (see Figure 9). This particular area was chosen because a con
siderable amount of cultural debris was present on the ground surface,
notably construction materials.
*See discussion regarding the preservation of archaeological
materials at Butler Island in Chapter 5.


212
resident planter, and white overseers managed the estate; Butler Island
had an absentee landowner; and Kingsley had a resident planter, and
black drivers presumably managed the estate. Also, Kingsley was known
to have been a slave importing station (Fairbanks 1974:63),which adds
another dimension to its economic function. All three plantations had
in common slave labor forces which exceeded 100 individuals and which
were located in a coastal environment.
Possibly, the diversity of these plantation contexts has resulted
in variations within specific artifact groups. Some of these were
noted in Chapter 5, but the quantitative differences in artifact frequen
cies were not examined. Artifact pattern recognition attempts to
examine frequency variations in artifacts from within intrasite or inter
site contexts (South 1977:99-102). Through such examinations patterned
similarities and differences may be delineated (South 1378:223). These
patterns presumably reflect the functional and behavioral processes
which Have taken place at sites. Thus artifact similarities occurring
within slavery contexts are suggested here to reflect general patterns
in coastal slave lifeways. Differences, on the other hand, may be
related to functional characteristics of specific plantations.
South defined eight artifact groups for the definition of arti
fact patterns (1977:92-102). These include architecture, kitchen,
arms, clothing, personal, furniture, tobacco, and activities. These
groups are essentially the same as those used in Chapters 4 and 5, but
there are differences which include, first, furniture and arms have
been assigned to separate groups; second, glass beads are included with
clothing; and, third, farming equipment, specialized crafts, and
l
r;-'


66
animals were used primarily for operating machinery and pulling carts
and wagons. Wool was sheared from the sheep to make cloth (Roswell
King, Sr., 18 August 1813). On special occasions, cattle, pigs, or
sheep were butchered to feed the slaves (Roswell King, Sr., 16 May 1813;
Roswell King, Jr., 13 November 1823). But these occasions were appar
ently rare, and slaves were most often fed "barrelled pork and salted
fish" (Roswell King, Sr., 7 April 1816; Roswell King, Jr., 18 March
1821). Citrus fruit trees were planted along the outer bank of the
island, an area approximately five miles in length. The purpose was to
protect the dykes from storms or freshets (Spalding 1830). Several
visitors to the area marveled at the beauty of the orange and lemon
groves (Hall 1931 (1828}:234; Coulter 1941:96). Most of the fruit was
marketed and the remainder was sent to the Butlers or distributed among
the plantation occupants (Roswell King, Jr., 25 December 1812, 8 Novem
ber 1818). Although experiments with planting corn, barley, and pota
toes were attempted (Roswell King, Sr., 31 May 1813, 1 June 1818),
Butler Island lacked sufficient "high lands" or hammocks (see House 193S;
213, 1954a:153) for the production of provision crops. Consequently,
most provision crops were either produced at the other plantations of
the estate (Roswell King, Sr., 9 June 1814; Roswell King, Jr., 25 January
1819) or were purchased from other planters.
Plantation structures were numerous at Butler Island. In addi
tion to the outbuildings needed to process and store rice, cotton, and
sugar, a sawmill, cornmill, warehouses, and a slave hospital were among
the structures (Butler Estate Papers,1815a). The maintenance
of these buildings was an ongoing process, and both slave mechanics and


Table 11
Styles of Glass Bottles Represented at Butler Island, Settlement #4
Glass Color # of Fragments Bottle Form # of Bottles
Dark green
317
3-hinge mold
3
free blown
1
dip mold
2
Light green
35
could not be .
determined
1
Clear
51
could not be
determined
1
Aqua
40
medicine vial
3
Total
443
11
been suggested that alcohol was an item obtained by slaves, not one that
was issued to them (MacFarlane 1975:117). The historic evidence for
Butler Island, however, indicates that alcoholic beverages were occa
sional items issued to the slaves, presumably for medicinal purposes:
He drinks no rum and he can be left to give rum to the Negroes
every wet morning. I think it will be profitable even if
they drank 100 gallons [of rum] a year. The people have a
number of discomforts that they would not get elsewhere.
Scant of water, wood, little or no vegetable, we must make
some way to make them comfortable & content. A number of
them are given to swelling at times. I think rum is a good
preventative. (Roswell King, Sr., 22 October 1803)
Alcoholic beverages were undoubtedly infrequently given to the
Butler slaves. Nonetheless, alcohol was issued to them at times. In
fact,
some rum was manufactured on the
plantation (Roswell King, Sr.,
19 January 181/), and it is possible that the recovered case bottle


53
and iris daughters, however, made Philadelphia their permanent home.
Thus, the responsibility of running the estate during the Major's
absence was taken care of by an overseer.
When the Major died in 1322, his properties were inherited by his
grandsons, John and Butler Mease. The Major had wanted his only son,
Thomas, to take on the responsibility of administering the Georgian
estate after his death, but Thomas, who hated the Major, refused and
went to Europe to live (Scott 1973:54). Thus, the Major disinherited his
son and made his grandsons heirs. After the Major's death, however,
Thomas helped his sister Frances to administer the estate until the
Mease boys came of age and were able to do so. At the Major's request,
his grandsons had to change their surnames to Butler. Consequently,
John Mease became John Butler and Butler Mease became Pierce Butler, Jr.
In 1834, Pierce Butler, Jr., married Fanny Kemble, the famous British
actress. During her first and only trip to her husband's Georgian
estate in 1839, she wrote the controversial account of slave life,
A Journal of A Residence on A Georgian Plantation In 1838-1939, which was
published more than 20 years later.
Fanny and Pierce came to Georgia with their two young daughters
for approximately four months. Two months were spent at Butler Island
and another two at Hampton Plantation. Fanny was outraged at the con
ditions under which the slaves lived, particularly the suffering endured
by women. Fanny's reaction resulted in many heated arguments with
Pierce, which intensified the couple's preexisting marital problems.
After their return to Philadelphia, the marriage grew increasingly worse.
In 1849, Fanny returned to Britain, the couple were divorced, and
Pierce gained custody of their daughters (Scott 1973:11).


Figure 23. The Extent of Excavations at Structure One, Settlement #4.
Key to Mapped Artifacts
A. Hinge
B. Adze
C. Scissors
D. Rounded chisel or gouge
E. Large fragment of black flint
F. Pintle
G. Grinding or polishing stone?
H. Axe head
I. Padlock
J. Hinge
K. Wood fragment
L. Wood fragment
M. Hoe
N. Padlock
O. Rice sickle
Dotted line: Large hand-hewn
cypress fragment


67
hired white craftsmen were regularly engaged in making improvements on
the structures.
Efforts were made toward making Butler Island self-sufficient in
the production of plantation supplies. Construction materials were
manufactured, including clay arid tabby bricks, shingles, nails, and iron
tools. Slave clothing and shoes were also frequently made. If trained
slaves were unavailable, local whites were hired to undertake these
tasks (Roswell King, Sr., 9 July 1805, 28 May 1815, 4 October 1915).
Sometimes whites were employed to train the slaves in specific skills
(Roswell King, Sr., 26 April 1816). The extent to which certain pro
ducts were made depended on the needs of the plantation and the cost of
manufacture at the time. This was clearly the case with regard to shoe-
making: "As for tanning and shoemaking, I recommend you give it up.
Hides have gotten extravagant. Our shoes cost $4.00 a pair, you can get
better shoes out of the state prison for $1.00" (Roswell King, Sr., 15
November 1818). Thus "self-sufficiency" at the Butler estate varied
through time and with the needs and interests of the owners.
Taken as a whole, the Butler estate was at most times very profit
able. The estimated value of the slaves, structures, and lands was
$288,027 in 1818 (Roswell King, Sr., 19 July 1819). By the 1850s, the
slaves alone were valued at $300,000 (Pierce Butler, Jr., Ledger,
1851-1855). Annual returns on the crops ranged from $20,000 to $50,000
except in years of losses (Butler Estate Papers, 1814: Butler Estate
Papers, 1824-1835). The estate, however, was not entirely a success
story. Storms damaged tremendous amounts of crops in the years
1804, 1813, and 1824. The hurricane of 1804 was particularly disastrous




184
(Doar 1970:29). Whether or not the Butler Island slaves ate rats and
mice could not be determined from the food remains.
Summary of Food Resources
Despite the limitations of the faunal collection, it is possible
to offer some interpretations regarding slave diet at Butler Island,
particularly as these relate to the test postulates for this study.
The relative proportions of the domestic food remains were found
to be very similar to those recovered from Cannon's Point plantation.
This finding lends support to the hypotheses that similar patterns
existed in slave consumption of domestic food items throughout coastal
Georgia, at both cotton and rice plantations. Beef was found to be the
predominant food item at both plantations. Although beef was evidently
an important food resource at coastal plantations, the records for the
Butler estate indicate that pork was the major domestic animal food
supplied to the slaves. Assuming that the records are accurate, it may
be that the archaeological record preserves more evidence of beef con
sumption than of pork, particularly if pork was cured and the bony
parts removed and beef was served as freshmeat.
Plantation managment practices may have influenced domestic food
consumption. Because of the absence of a planter family at Butler
Island, it is possible that slaves received all parts of butchered
animals, whereas, in the case of a plantation with a resident planter,
slaves were perhaps more likely to receive the portions of an animal
that the planter did not want. The archaeological evidence is very
tentative, but the presence of cattle humeri ai: Butler Island and the
absence of such fragments at Cannon's Point do suggest that this may
have been the case.
... .


7 9
The above statement supports the suggestion that planters deliber
ately selected slaves adapted to a malarial environment for labor in rice
fields. But it appears that even among members of the old, established
slave population,
their abilities to withstand the
rice swamp was an
ongoing concern:
It appears we have few Negroes here [Hampton] that are
suitable for the rice swamp. We have at least 20 at the
rice island which I would like to move if we had
suitable highland for them. When you purchase more of
course some of them will not be so profitable in a rice
swamp and it is necessary we should arrange for land more
suitable. (Roswell King, Sr., 22 August 1813)
Yet sickness was a prevailing problem at Butler Island despite
efforts to place slaves there that were adapted to a malarial environ
ment. An array of diseases plagued the island, including fevers of all
kinds, measles, intestinal disorders, cholera, influenza, whooping
cough, and others. Of these,cholera, influenza, and intestinal dis
orders were the most dreaded, and reports of these diseases frequently
followed the occurrence of freshets, particularly in the 1830s,when
freshets became annual events. At that time, slaves were often re
located to keep down debilitating diseases (Roswell King, Jr.,
15 February 1835, 8 March 1835). Evidently, freshets contaminated the
food and water supplies, which when consumed, caused illness and pos
sibly death. The drinking of contaminated water, however, was seem
ingly a health problem at other times as well: "This is the sickly
season for Negroes at the rice island. There is 14 to 15 in the hos
pital. None dangerously ill. I believe it is owning to their drinking
river water" (Roswell King, Sr., 30 April 1815). In general, in the
South
contaminated water was frequently the source for epidemics of


203


77
13 May 1827), or retired slaves involved in menial tasks, A census
taken of the Butler slaves in 1830 indicates that 365 slaves lived at
Butler Island at that time. The census provides the female-to-male ratio
and age distributions, as shown in Table 6.
Table 6
1830 Census of Slaves at Butler Island
Age
Male
Female
Tota'
Under 10
44
49
93
10-24
57
52
109
24-36
36
33
69
36-55
39
29
68
55-100
12
14
26
Total
183
177
365
Details obtained from plantation records regarding the material
conditions of the Butler slaves are similar to standard slave care prac
tices throughout the Old South. At Butler Island, these practices
mirror those established for other rice plantations (e.g. Allston 1858;
House 1954a; Phillips 1969;!14-130}. Generally, these practices include
the following; the provision of two sets of clothing annually, one for
winter and the other for summer; plantation rations, which consisted
primarily of corn, pork, salted fish, and molasses; a sick house and
medical attention furnished by the overseer or slave doctor; and hous
ing. Superficially, the records tend to suggest that the Butler slaves


136
purpose.* Possibly, it provided additional support to the hearth.
A similar arch-shaped feature was identified with structure four, but
none could be detected with structure five, the clay brick chimney.
Limited testing at the clay brick chimney may account for the failure to
detect this feature.
Part of the arch was removed in order to uncover artifacts below
it that would provide a construction date for the chimney. The only
dateable item recovered was a sherd of banded pearlware ceramic, which
establishes that the chimney had to be constructed after 1790. Because
the diking of settlement #4 did not begin until 1803, the chimney had to
have been built after that date. Unfortunately, the pearlware sherd does
not aid in the refinement of a construction date for structure one.
The bricks utilized in the clay brick chimney are known as
Savannah greys. They were given this name because they were made with
clay found along the Savannah River drainage. Several plantations in
the Savannah area had been engaged in brickmaking as early as 1750
(Granger 1947:9, 435). At Butler Island bricks were produced from the
"swamp clay" (Roswell King, Sr., 20 August 1815), but bricks were also
frequently purchased from Charleston and Savannah. The source of the
bricks used in structure five could not be determined. It may be pos
sible to tentatively identify bricks that were made at Butler Island.
Many of the clay bricks used to line the fire boxes of the chimneys and
used in the construction of the brick floor were very crudely made, low-
fired, and with large grey clay inclusions. It is suggested that these
bricks were probably made at Butler Island because brickmaking at the
^Charles H. Fairbanks, personal communication, 1980.


56
Information of ownership and land use at Butler Island in the
years between the Leighs' departure and 1920 has been very sketchy. It
has been assumed that the island was abandoned and left unattended
(Vanstory 1970:89). Yet deed records, local informants, newspapers,
and a valuable map (see Figure 6) have suggested a different interpreta
tion.
Evidently, the Leighs leased and sold portions of their proper
ties in the Altamaha delta between 1878 and 1910. By 1910, all of their
properties were sold (McIntosh County Records, Deed Books H--52, I-I20,
1-160). These properties, which embraced all of Butler Island and
Generals Island, were sold and leased to rice planters, who in turn con
tracted with tenants to farm small plots for a portion of the crop (as
shown in Figure 6). Local informants have indicated that by the 1880s,
tenants no longer lived at Butler Island. Instead, they lived at several
settlements on the mainland and commuted daily by small boats to Butler
and General's Islands. Two of these informants, both descendants of
slaves in the area, remember vividly planting rice at Butler Island in
the early 1900s.*
Long after the Leighs left Georgia they continued to correspond
with the local blacks (Letters from Descendants of Butler Slaves, 1890-
1903). Particularly, their only child, Alice Leigh, visited Georgia
often and on these visits she would give parties for the people (Cate
end Wrightman 1955:203). As children, present-day black residents of
Darien performed for "Lady Alice" on these occasions.** The impact of
Elizabeth Bates and Jim Cook, Personal communication, 1978,
Darien Georgia.
**Nan-Earl Wylly and Mrs. Gamble, personal communication, 1978,
Darien, Georgia.


Appreciation is extended to Gene Love, David Edwards, and the re
mainder of the staff of the Al tamaha Waterfowl Management Area for their
advice, equipment, and time.
I would like to thank the community of Darien, Georgia, for the
hospitality they extended to me and the crew during our stay there. Spe
cial thanks go to Ms. Bessie Lewis, Jim Cook, Nurse Campbell, Mrs. Bates,
Nan Earl Wylly, and Rudolph Capers for their valuable information relating
to the findings presented in this study.
The archival research was an extremely important aspect of this study.
I would like to thank the staffs of the Pennsylvania Historical Society,
the Georgia Historical Society, and the P. K. Yonge Library at the Univer
sity of Florida for their assistance. Also acknowledged are the staff of
the Inter-Library Loan Department at the University of Florida Library for
securing many of the materials I needed.
A warm appreciation is extended to my colleagues for their comments
and suggestions concerning my research objectives: Kenneth Lewis, Patrick
Garrow, Thomas Wheaton, Lei and Fergueson, Jennifer Hamilton, Chad Bradly,
Sue Mullins-Moore, John Morgan, Jr., Morgan R. Crook, Elizabeth Wing,
Rochelle Marrinan, Larry Banks, Nicholas Honerkemp, Robin Smith, and
Leslie S. Lieberman. Also, I would like to thank John A. Scott for his
suggestions; Richard Fisher for undertaking the soil analyses; Kathy
Bordua, my lab assistant and confident during the analysis and writing of
the dissertation; Ashley Wood, the draftsman and illustrator; and Sue
Kirkpatrick for typing the final manuscript.
Finally, I would like to thank my family and friends for their
encouragement through the years.
v




23
With saltwater but close enough to the ocean to be affected by the tide
(House 1954b:23). The restricted spatial limits of tidewater rice cul
ture made the alluvial soils of the delta the "most valuable" of this
old agricultural region (Bonner 1964:2).
In general, most plantations on the barrier islands were devoted
to long-staple cotton and those of the delta to rice. Some rice, however,
was grown on the barrier islands, presumably in the freshwater swamps,
and some cotton on the delta. Sugar, a third important cash crop (see
Sitterson 1937, 1953:31-35), was frequently produced at both, but in
smaller quantities. For example, at Butler Island, considerable quanti
ties of cotton and sugar were cultivated, although rice was the island's
major cash crop.
The purpose of this chapter is to review the historical develop- .
ment of plantation agriculture in the Georgia tidewater, specifically as
it relates to the production of rice and long-staple cotton, the two
dominant crops of the post-Colonial period. Also, planting and process
ing methods and marketing for each crop are described. Rice culture is
considered in greater detail for two reasons: It was a more involved
process than that of cotton culture and a''detailed description of this
culture is central to understanding the archaeological resources of
Butler Island (long-staple cotton is treated extensively in another
archaeological reportsee Otto 1975:51-61 ).
The Emergence of the Plantation System in Georgia
On June 9, 1732, the developers of Georgia, known as the trustees,
were granted a charter for the lands between the Savannah and the


121
presumably used in the production of rice, rather than a domestic dwell
ing.
No surely identifiable nonstructural features, such as wells or
privies, were uncovered. As for wells, the historical evidence indi
cates that the Butler Island slaves drank water from the river (Roswell
King, Sr., 30 April 1815; Kemble 1961 (1853):189). Looking at the
Altamaha River today, it is hard to imagine that the river was ever
clean enough for drinking or bathing purposes. Nonetheless, if the
river regularly supplied water for drinking, then wells would not be
needed, and this archaeological feature would not be present. The ab
sence of privies is more easily explained than that of wells. On many
plantations slaves were not provided with privies. In fact, privies were
not used extensively in many areas of the South until the twentieth
century (Savitt 1978:60). Evidence of a privy is suggested from the
drainage ditch next to structure four (see Figure 14). From a trench
designed to test the drainage ditch adjacent to structure four (see
Figure 21), several vertical posts were uncovered. These posts appeared
to have been part of a small structure, possibly a privy or washstand.
In addition, soil samples taken from the drainage ditch had an extremely
high phosphorus content when compared with samples from other areas
of the site. Although the high phosphoric level suggests that this
area was a privy, the test for human bacteria was negative (see
Appendix 1). Therefore, positive evidence of a well or privy at
settlement #4 is lacking.
Whatever the function of the small structure, wooden fragments
recovered from the ditch and the soil analysis indicate that the


5
the geographic area occupied by a cultural system. It assumes that the
region, not the single site, embraces the total range of behavioral
variability for a population. The single site, on the other hand,
reflects only those activities that had taken place at that
specific locality (Streuver 1971:11). Thus the goal of regional
archaeology is to determine patterned interrelationships among sites.
Regional archaeology attempts to understand the cultural and
environmental processes which have affected behavioral patterns. Its
application to historic site archaeology is slowly emerging. Specifi
cally at historic sites, artifact pattern recognition (South 1977,
1978) has been suggested as a technique of discerning behavioral
variability within a region or time period. Ultimately, pattern recogni
tion seeks to identify by means of archaeological data the functional
and behavioral processes which have taken place at sites (South 1378:
223; e.g. Lewis 1977).
Besides a unit of analysis, regional archaeology also supplies a
method for the study of site location. Locational analysis examines
spatial relationships about ecological or cultural phenomena
(Haggett1965). A developing focus of locational analysis is its
application to historic sites (Swendlund 1975; Langhorne 1976; House
1977). Such studies attempt to determine the importance of cultural
or natural resources within the functioning of a historic cultural
system (House 1977:243-4).
A third concept of regional archaeology establishes a frame
work for the collection of cultural data for a defined area. The
regional plan or overview is the first phase in multi-phase


Figure 26. Arch-Shaped Hearth Support. (Note how the hearth buckles inward.)
%
I
I
I
I
f
I
m
w
$
%£
f A'
m
*$-
m-

'r
f:
ft:
I
hr


REFERENCES
Page
229
BIOGRAPHICAL SKETCH
248
vm
- m n


220
tidal marsh. Small fields known as "tasks" became the unit cf measure
for certain labor requirements in the cultivation of tidewater staples.
This labor system, which provided slaves with the incentive to improve
their material standard, was later employed in other cash crop regions
throughout the South. Its origin, however, is clearly an adaptation to
coastal habitats. Lush ecological conditions like those of the coast
were not present in interior regions of the Old South, and the absence of
these conditions in the interior was likely to have been reflected in
slave material conditions. It is, however, possible that slave life-
ways were similar in all coastal areas of the South, but this assumption
will have tc be demonstrated archaeologically.
Slave demography along the coast was another factor. The high
concentration of slaves in coastal areas was again an adaptation to
the production of coastal cultigens. The total reliance upon hand labor
required vast amounts of slave labor. It may have been that in "black
belt" areas slaves were given a few privileges, such as the use of
firearms, that slaves in other areas were forbidden. This assumption
remains untested, but it does seem that coastal slaves, particularly
those on rice plantations, had at least the opportunities to manipulate
the system during the periodic absences of whites.
Historic and economic factors were also found to have been very
important. Georgia was settled relatively late by planters when com
pared with other tidewater areas. When the plantation system was just
beginning in Georgia, planters from the older tidewater areas were
suffering from rapidly depleting soils. Many of these planters came to
the virgin lands of Georgia as a result. Major Pierce Butler was one


64
Table 2
Rice Production at Butler Island, 1802-1833
Year
Acreage
Number of Tierces3
1802
128
1803
337 1/2
1804
44
1805
1326
1806
--
468 1/2
1807
--
1186
1808
--
565 1/2
1809
--
678 1/2
1810
--
587 1/2
1811
__
. 857
1812
1739
1813

253 1/2
1814
--
829
1815
591
1816
--
513
1817
--
1058
1818
--
650
1819
407
1173
1820
227
543 1/2
1821
234
832
1822
272
1080
1823
358
1240
1824
342
84b
1825
318
1024
1826
399
1455
1827
--
1828
395
1331
1829
1830
505
1747
1831
552
¡557
1832
271
1695
1833
591
1660
Source:
Compiled from Annual Crop and Lives
lock Reports, 1803-1833
Note:
Average annual tierces, 1802-1830, is
712; average annual
fierce
- 3 O 1
-1833 is 1,246; increase equals 43%.
3
Tierce
= 600 lbs.
k|_ow production due to hurricanes.


37
most efficient (House 1954b:62-65). Plantations with threshing machines
often housed machinery for threshing, pounding, and polishing within the
same structure (Smith 1936:22; House 1954b:62). Only very large planta
tions, however, could afford to be fully equipped with milling facili
ties. The planter with pounding and polishing mill could ship pounded,
clean grain to the market, whereas the planter without had to ship
rough rice, or "paddy." Also, the planter without a mill had to
rely upon toll mills to complete the processing of the rice for a per
centage of the grain. Several European buyers, however, preferred to
purchase rough rice and mill it in their own countries. Because of
this there was always a substantial market for rough rice. By 1850,
most foreign shipment was paddy, and pounding mills were utilized pri
marily for the domestic market (Gray 1941:730; House 19545:68).
World prices for rice were based upon polished rice and weight
(House 1954b:67). The commercial standard weight of rough rice was 44
to 45 pounds a bushel (Austin 1893:24; Doar 1970:18), and the average
yield of rice lands varied from 25 to 60 bushels of rough rice per
acre (Gray 1941:730). Polished rice was packed in "tierces," each
weighing 600 pounds. The quality or grade of the rice was judged accord
ing to the number of bushels it took to make one tierce. Generally.
20 bushels or less of rough rice to make one tierce of polished rice
was considered a good crop, more than 20 an inferior one (Doar 1970:
18), Prices varied considerably during the antebellum period. The
best prices for Georgian rice, however, were obtained between 1840 and
1860, or the "golden age" of Georgian rice production. At that time


32
on the tidal swamps to permit the inflow of freshwater backed up at
high tide and the outflow at low tide. The fields could be flooded and
drained as desired (Cole 1927:599). But before the tidal marsh could
be cultivated, it had to be cleared, diked, and drained. Reclaiming the
tidal swamp was a tremendous undertaking which required considerable
expertise and its completion often took several years (Heyward 1937:
13-20; Wilms 1972:53; Hilliard 1975:60).
The innovation of transforming a marshland into a piece of culti
vable land, a polder, cannot be credited to the tidewater planter.
Construction of polderlands has been undertaken since ancient times,
and today they are located all over the world (Wagret 1968). Perhaps,
the best known are those of Holland. In fact, indications are that
several Carolina-Georgia tidewater planters employed Dutch expertise in
designing their hydraulic systems (Doar 1970:13; Scott 1973:73-74;
Johnson 1930:61).
Articles describing the construction of the irrigation systems
used by tidewater rice culture are found in several contemporary
journals (e.g. Alisten 1846, 1854; DeBows Review 1847; Spalding 1845).
Several secondary sources have provided additional details for this
procedure (e.g. Heyward 1937; House 1939, 1954b; Gray 1941; Doar 1970;
Hilliard 1975). In very general terms, the procedure for constructing
a tidewater rice irrigation system involved six major steps, 'first,
the'length end width of the required acreage was measured off and the
trees were cut down- and moved out of the way. Second, a temporary ditch
and embankment were built around the entire area to keep water out, while
tree stumps and debris were removed; third, the ditch was filled and


> 'i r r-- --^r ,tfj3 V.T'Saj'' i-'\ * W* J ~"r-' '^'SiJ £>-i \ --, w*,. ;> ''J&&--'' ',-=iV 2 '^J-V
70
Butler slaves were direct imports from Africa (Roswell King, Sr., 13
May 1803), most were acquired through the domestic slave trade based
in the older slaveholding regions, particularly the Virginia Tidewater.
The enterprising Major, of course, found ways to add to his slavehold-
ings illegally. Before the Major left for Georgia, he took a consider
able number of slaves belonging to his wife's family. Although the
total number of slaves acquired illegally for the Major's in-laws is
not certain, in 1817, 127 slaves were found to be the property of the
Major's sister-in-law (Roswell King, Sr., 29 January 1817). Apparently,
these slaves were never returned to their former owner. As planting
interests expanded, the estate was supplemented with additional pur
chases of slaves. Natural increase also added to the slave population.
Rarely did the death rate exceed the birth rate (see Table 3). The
fact that few slaves were sold* provided an additional factor which
helped to maintain the size of the slave population.
Several slave lists exist for the Butler estate, but few specifi
cally list the slaves at Butler Island. Tables 4 and 5 combine a
number of lists taken from various time periods (Butler Estate Papers,
1803, 1821; Roswell King, Jr., Daybook, 1844). Indications are
from the plantation records that this combined list reflects the
average number of slaves regularly engaged in the labor force at
Butler Island. The list, however, does not include children
under age 10, which usually averaged 100 in number (Roswell King, Sr.,
*A close examination of the slave records reveals that the occa
sions on which slaves were sold from the estate were very rare. This
is in keeping with the evidence concerning other rice plantations in
the area (see House 1939:215).


Page
CHAPTER
The Community Plan at Slave Settlement #4
Slave Dwellings
Chimney construction
Building hardware
Summary .............
5. THE MATERIAL CULTURE OF SLAVERY
Kitchen Artifacts . .
Ceramics
Ceramic styles and vessel forms . .
Dating the ceramics from Butler Island
Nonceramic Artifacts
Glass bottles and tablewares . . .
Cooking and processing equipment . .
Eating utencils .
Kitchen Artifacts: Summary .......
Food Procurement . . .
Food Resources . .
Domestic Food Resources
Nondomestic Food Resources .
Fishes and turtles ..........
Birds and Mammals
Summary of Food Resources
Clothing
Buttons
Footwear
Accessories and Ornaments .........
Beads .
Miscellaneous Items ... .
Summary .
Household Items ..... .
Tobacco Equipment . .
Farming Tools and Specialized Crafts . .
Summary . .
6. PATTERN RECOGNITION IN SLAVE MATERIAL CULTURE .......
7. SUMMARY AND CONCLUSIONS . .
APPENDIX
1. SUMMARY OF SOIL ANALYSIS FROM SETTLEMENT if4, BUTLER ISLAND
2, SLAVE SUBSISTENCE ARTIFACTS FROM SETTLEMENT i/4 ......
ABBREVIATIONS USED IN REFERENCES . ........
i l D
126
130
137
138
140
141
142
143
155
157
157
159
160
i 6 b
166
169
173
1 78
179
183
184
186
186
187
190
190
193
193
1 S3
198
199
109
211
219
225
VI 1




177
present at Butler Island before 1835, which predates Fanny's visit,
and it is possible that the sheep and goat population increased after
that date. But the archaeological record does not suggest that mutton
was consumed by the Butler Island slaves.
The sample indicates that chicken (Gal 1 us gal 1 us) was a very
occasional dietary item. A similar pattern has been observed at the
Cannon's Point slave sites (MacFarlane 1975:165: Otto 1975:311).
Chickens were apparently kept for the provision of eggs that were most
often sold or traded to local markets. In fact, most slave-owned live
stock was sold rather than consumed by the slaves (Hilliard 1972:60).
Peach pits were the only plant food remains, domestic or otherwise,
which provided archaeological evidence at Butler Island. A total of 82
whole pits and 62 fragments were recovered. Peach trees were present
on Butler Island (Kemble 1961 (1863):167)3 but no references to supply
ing the slaves with peaches could be found. In this case, the archaeo
logical evidence provided information not available in the documents.
The evidence of slave peach consumption supports the claim that
planters supplied slaves with seasonal food items such as fruits and
vegetables, which are generally not mentioned in the planter records
(Fogel and Engerrnan 1974:111), The records, however, do indicate that
the slaves were given some of the oranges and lemons from the Butler-
Island groves. Present-day fig trees at Butler Island, presumably
dating from the antebellum period, may also have been used to supply
fruit to the slaves. Both archaeological and historical evidence indi
cate that the Butler Island slaves were able to add fresh fruit to their
diets from fruit trees available to them.
ms


93
the former interpretation does seem plausible, but historic evidence
suggests otherwise. Prior to the construction of this tidal mill, King
describes the "oyster bank" and suggests placing a mill at that loca
tion (Roswell King, Sr.s 13 January 1813). In a later letter, King
describes the depth of the oyster shell with astonishment: "The oyster
shell are 6 feet underground, and the water runs in like a spring"
(Roswell King, Sr., 6 March 1813). The two letters seem to suggest that
the oyster bank was there long before King or Butler thought of building
a mill at that location. At least, it does appear that King was not
responsible for bringing the oyster shell to Butler Island. It is
possible that King's predecessor was responsible for this action.
These aboriginal shell dumps have been greatly disturbed by post-
depositional construction and the robbing of shell, presumably for the
making of tabby. Therefore, it is doubtful that the present size and
shape of these shell deposits can help to determine whether or not they
were made by prehistoric occupants of Butler Island. Unfortunately, the
prehistory of the Alternaba delta is virtually unknown. Consequently,
these shell deposits may be prehistoric sites and not aboriginal shell
brought in at a later time.
Slave Village Sites
The investigators were most interested in the slave settlements,
and an effort was made to determine the archaeological integrity of these.
Investigations began at settlement #1 for several reasons. First, the
site's location, just off U.S. 17, makes it more accessible to the
vehicle than the other slave sites. Second, unlike the other slave
sites, it is not subjected to management flooding in the fall and winter


Annular wares. Several decorative motifs are characteristically
found cn annular ceramics, often more than one on the same vessel (see
van Rensselaer 1978:240-24). The motifs include slip banded circles,
mocha, swirl, circle and cube, and engine turned or rouletted (Price
19/9:18). Vessel forms include mugs, bowls, jugs, teapots, and tureens.
At Butler Island, only cups, bowls, and mugs were recognized (Figures
29 and 30).
Spong decorated. Ceramics known as sponge decorated, or spatter-
ware, consist of the application of colors to a white bodied vessel by
means of an inked sponge. Plates, cups, and saucers are the most
common vessels (Price 1979:19). Only one vessel was represented at
Butler Island, and it appears to be a small cup or bowl.
Alkaline-glazed stonewares. A glaze made from wood ash and lye
water was frequently applied to stoneware storage vessels in the South
(Greer 1971). These were probably produced cheaply by small potters on
the Georgia or South Carolina piedmont.
Salt-glazed stoneware. The technique of manufacturing stoneware
in which salt was thrown in the kiln while the vessels were being fired
resulted in salt-glazed stoneware. The effect is an orange-peel tex
ture on the surface of the vessels. Most nineteenth-century salt-glazed
stoneware was made into storage forms, including jugs, jars, pitchers,
and bottles (Sarnmis in Stradlirig end Garrison 1977:112).
Porcelain. Generally found in all tableware vessel forms, porce
lain is associated with high status households. Cne undecorated white
fragment was uncovered. It appears to have been part of an ornamental
vessel form, most probably salvaged from the overseer's house.


jrfTTr^ft
" 'Tf
TO 100 110
I20 100 140 150 1G0
ilU
r
¡ n c n c :
no
o
CD
....-ys'Y.fl'ii;-.'.; .tv-.- V


131
adherence of this procedure often resulted in a very durable building
material (Gritzner 1978:36). The tabby bricks made at Butler Island
evidently included shell robbed from prehistoric shell middens. This
was indicated by the presence of burnt aboriginal ceramics found within
the bricks.
Although tabby brick may have been a durable construction material,
it was not at all fire resistant. The fire box of structure one was
lined with clay bricks (see Figure 24). This has been found to have been
the case with most tabby chimneys. In addition, the heat from the fire
had apparently caused the hearth of structure one to buckle inward and
to separate from the outer walls of the chimney. Thus the durability of
tabby as a construction material for chimneys is questionable. Perhaps,
structure five was made primarily with clay bricks to guard against the
lack of fire resistance inherent with tabby bricks.
Two bond styles were employed in the brickwork, of the outer chim
ney walls. The "English bond" (Noel-Hume 1974:122), which alternates
courses of stretchers (the side of the bricks) and headers (the end of
the brick) was used at the base of the chimney. The above grade brick
work was laid in "common bond" (Noel-Hume 1974:122), that is, stretchers
in every course for five or more courses (see Figure 25).
Perhaps the most curious aspect of the chimney is the arch-shaped
support feature found below the hearth (see Figure 26). This architec
tural feature has been observed frequently at coastal plantations in
Georgia on planter, overseer, and slave dwellings. Its function is
unclear. It may simply reflect a shortage of bricks needed to complete
the chimney or it may have served a functional as well as an economic
-1 mam


156
Appplication of the M
Table 10
ean Ceramic Date Formula
to Tablewares
Ceramic Style
# of Fragments Mean Date
Product
Whiteware, undecorated
217
1860
403620
Green and blue
shell edged pearl ware
48
1805
86640
Transfer printed
pearlware
17
1818
30906
Underglazed blue and
white pearlware
11
85
19855
Underglazed polychrome
26
1805
46930
Finger painted
1
1805
1805
Mocha
10
1843
18430
Annular pearlware
91
1805
164355
Total
421

772541
Note: Mean ceramic date = 1835.01.
The application of the Mean Ceramic Date Formula to the Butler
Island collection, unlike its application to other slavery sites in
coastal Georgia, indicates that the Butler Island slaves were not using
outmoded ceramics. Two factors may account for the difference between the
Butler Island ceramic assemblage and the assemblages of the other slavery
sites: imprecise dating of the whitewares and the absence of eighteenth-
century ceramics. The precise manufacturing dates for the whiteware are
unknown (Noel-Hume 1976:131). Their assigned median date of 1860 is at
best an approximation, and the actual median date is likely to have been
much later. Because dates of manufacture for whiteware are unclear, many
authors do not apply the formula to these ceramics. The inclusion of
these ceramics to the formula here, however, suggests that a satisfac
tory median date can be derived.


A22


'
' 0- -i:
, .
' - '
ttVi : Ife :
\ k m /.>'
. ' 3 : ..
w I
.->v-y
-
:
wi&w ;
gsgSJji;
¡itis
3 R;
0-%&
.
-P


105
six and seven to the right of the slave dwellings (see Figure 14). Evi
dence of these structures was first identified from posthole tests.
Later, through probing and small test pits, the approximate dimensions of
the structures were determined. Structure 6 was found to be a brick floor
and structure seven, the foundation of a large structure, possibly a rice
barn.
Excavation of the slave dwellings proceeded with the establishment
of excavation units from an arbitrary point designated N10QE100. Ini
tially, the soil was removed in arbitrary levels of 15 centimeters, but
later this was changed to the removal of natural strata. For the most
part, three strata were recognized at slave settlement #4. These include
grey silty loam, grey clayey loam, and grey clay with red and yellow
flecks (see Figure 15). These soil classifications were determined by
soil analyses undertaken after fieldwork activities (see Appendix 1
for soil descriptions). All cultural materials were deposited within
the first two strata, but the clayey loam yielded a higher concentra
tion of artifacts than the silty loam. Evidently, the clayey loam
represents the old humus zone of the antebellum period. With the
exception of intrusive features such as postholes, no cultural materials
were found in the clay. All excavated materials were water-screened
through 1/4 inch mesh hardware cloth. From the experience at
settlement #1, it was felt that water-screening would maximize recovery
as well as aid in the removal of the sticky clay and silt from the
artifacts.
The vast majority of
antebellum date. In fact,
the artifacts from settlement #4 were of
only one artifact was found to be postbellum.


27
Figure 3.
Extent of Rice Cultivation in Georgia by 1860.
(Adapted from Hilliard 1975.)


210
Additionally, subtle differences such as higher frequencies of buttons,
beads, and pipes may reflect differences in management practices or the
fact that the Butler Island slaves were closer to local merchants than
were slaves on the barrier islands. On the whole, the similarities in
the artifact patterns are overwhelming, and it is suggested that these
are regional characteristics in slave material culture.


patterned relationship (Garrison in Schiffer and Gummerinan 1977:154).
Fortunately, the evidence from both the posthole testing and the mapped
artifacts in and around the slave dwellings strongly indicated a non-
random, patterned distribution. Thus, it is doubtful that inundation has
adversely affected the distributional patterns of cultural features at the
present time.
According to the 1815 inventory, eight dwellings were located at
this site. Assuming that no additional dwellings were added after that
date, slave structures were possibly arranged in two rows, three in each,
and with two structures at the end of the two rows (like structure three;
see Figure 14). It is suggested that the dwellings were arranged in
this fashion because of the considerable construction debris, presumably
from slave dwellings,found in the canal directly below the site. The
canal is an artifact of Waterfowl Management activity, and it appears
that when it was constructed the archaeological remains for several struc
tures at the slave site were destroyed.
On a main dike, adjacent to the canal and formed by the dredging
for it, artifacts characteristic of slave material culture were found
to be lying on the surface. Both materials recovered from the canal and
the dike suggest that the additional slave dwellings once located at
this slave site were in the location of the present canal, in line with
the mapped structures located at the site.
The two very large structures, six and seven, were apparently part
of the technical nucleus. All indications suggest that structure seven
was a rice barn. On the 1815 inventory, a rice barn with the dimen
sions 60 by 44 feet is listed. The dimensions of structure seven were
found to be 18.4 by 13.4 meters, which closely matches these dimensions.


96
The units were excavated in arbitrary levels of 15 centimeters.
Excavated materials were dry-screened by use of a 1/4 inch mesh hardware
cloth. All beginning levels were screened because it was not certain
where the antebellum component began. It was during this first attempt
at excavation that the archaeologists became painfully aware of the
difficulty of removing, by shovel, and screening this wet, clayey soil.
Screening was found to be a very slow arid time-consuming process, but
alternatives to screening were unacceptable. The Noel-Hume method of
excavation, in which every artifact is recovered by trowel (1974:104),
would have been just as time consuming, and later troweling activities
attested to this presumption. To have abandoned screening altogether
would have resulted in the recovery of few small items characteristic of
slave material culture such as fragments of pipe bowls and stems, beads,
and buttons. In light of the alternatives, screening was not cur
tailed. But the commitment made to screening sacrificed the areal
extent of the subsurface testing. This was found to be the case at
other sites as well.
The excavated materials indicated that the site is very much
disturbed. No j_n situ structural features were uncovered although most
of the materials were in fact clay bricks. Antebellum artifacts
dominated the recovered materials but were mixed with modern debris
throughout the first two strata. Below these layers, few cultural
materials were recovered. At the base of the excavation, evidence of
the old cypress root system was unearthed and the watertable was en
countered (see Figure 10).


176
plantation records) was possibly cured more often than beef. In the
curing process, some or all of the bone could have been removed, which
would leave little or no evidence at the point of consumption. Beef may
have been used more often to supply fresh meat, and the archaeological
record reflects this. Moreover, it appears that the Butler slaves could
have consumed all edible parts of both cattle and hogs, simply because
a full-time planter was not present to receive choice cuts. The size of
the slave population also suggests that this was the.pattern. To
supply 300 to 400 slaves with only the forequarters (heads, necks, tails,
and viscera) would require that more animals be slaughtered than if a
few whole animals were butchered exclusively for them.
Assuming that purchased pork supplied the bulk of the meat ration
to the Butler slaves, it is questionable What portion of the diet the
preserved food remains represent. Although purchased pork contained
some joints, most of it was bacon sides (Hilliard 1972:58). Purchased
pork, like plantation cured pork, may not be preserved in the archaeo
logical record. Thus, it is possible that the domestic animal remains
reflect only occasional dietary items. At least this suggestion sup
ports the plantation records as well as the need to supplement planta
tion rations with nondomestic animals.
No goat (Capra hlrcus) or sheep (Ovis aries) remains were identi
fied. Possibly they have been placed with the family of Artiodacyl
(see Table 12). The plantation records indicate that sheep were rarely
butchered for the slaves. Although Fanny Kemble apparently had mutton
nearly everyday at Butler Island (1961 (1863):184), the slaves were
evidently not as fortunate. In addition, few sheep and goats were


viewpoints vary. Apologists for slavery tend to be very critical (e.g.
Phillips 1929:259-265; Eaton 1949). Other historians have found a great
deal of accuracy in Fanny's descriptions of slave life, although some
suggest that the journal is overromaticized (e.g. Stampp 1955; Genovese
1974; Owens 1975). Of all the critics, Georgians have been the most
hostile (Holmes 1978; Lovell 1932:196, 213; Cate I960). Even today,
this sentiment persists (Fancher 1971:102; e.g. Lewis and Hule 1978:120).
Butler Island was deserted during the Civil War, with the excep
tion of a few slaves who remained. In 1866, Pierce Butler, Jr., and
his daughter, Frances, returned to Butler Island to restore rice plant
ing (Vanstory 1970:87). Frances kept a journal, which was later pub
lished (Leigh 1883). Her journal became one of the few documents that
describe labor conditions in the South during the postwar years (House
1942b:157). Fierce Butler died in 1867. Frances, accompanied by her
German maid, managed the plantations alone for four years. Frances
married Reverend James Leigh in 1871 while she was in England and they
remained there until 1873, at which time they returned to Georgia.
The Reverend Leigh was very much interested in the South and helped to
establish St. Cyprian Church for the blacks of Darien. Leigh wrote of
his experiences in coastal Georgia in Other Days (1921).
While in England, the Leighs contracted some English workers to
supplement the black labor at Butler Island. These British laborers
resided at Butler Island and occupied the former slave hospital. As
laborers, however, they were a disappointment (Leigh 1883:202-213).
Discouraged, in 1876, the Leighs decided to return to England and
placed the management of the islands in the hands of a manager.
1878 they gave up their planting interest entirely.
In


118
plantation complex. The only time a slave needed to go to the planta
tion headquarters at settlement #1 was to see the overseer, head driver,
to go to the hospital, or for an occasional church or other meeting.
Although slaves at one village probably intermingled with slaves at the
other villages, most of their time was spent in the rice fields adjacent
to their settlement or at their own quarters. From the planter's per
spective, this isolation maintained economic efficiency and social con
trol. But for the slaves, it may have substantially reinforced the
very limited contact prevailing between blacks and whites as well as
between acculturated blacks and those that were
not so
acculturated.
Within these
isolated villages slaves had the opportunity to retain
African folkways. Perhaps this explains why the rice coast has pre
served more evidence of African retentions than any other former ante
bellum crop region within the United States (Herskovits (1958:120). The
addition of African imports to the rice coast long after the ban on the
Atlantic Slave Trade (see Wells 1967) may have further encouraged this
cultural development.
The Community Plan at Slave Settlement #4
Although very limited testing has been undertaken at settlement
#4, some general statements can be made regarding the site's internal
spatial arrangements. While it may be possible that the periodic in
undation of the site has resulted in the lateral displacement of archaeo
logical materials, evidence of this disruptive action is lacking. If
the materials were displaced, the horizontal distribution of the
archaeological materials would appear to be random, that is, having no


36
or "grinding," the removal
of the outer hull and blowing it away from
the shelled rice; finally
rice grains plus powdered
"polishing," which resulted in the clean wh
fragments, known as rice flour (House 1954b:
te
59-60).
Threshing was done by hand using flails for most of the planta
tion period. Sometimes a floor was constructed for threshing; an
"efficient" floor was 110 feet long and 50 feet wide and composed of
three ranges of boards" (House 1954b:60). Rice was cleaned using wind
or a fan. More often a winnowing house was constructed for this pur
pose. Winnowing houses were small, usually square-shaped buildings
raised upon stilts. The rice was dropped through a hole in the floor
of the house, and the wind blew away the chaff. The grain landed below
the hole on a clay or earthen floor. It was swept up and placed in
containers (Smith 1936:20, and illustrated). Not until the mid-
nineteenth century was a reliable threshing mill developed which
threshed and cleaned the rice (Gray 1541:729-730).
In colonial days, the outer husks of the rice grain were removed
by hand-powered mills and hand fans blew away the chaff. These "husking"
machines consisted of slabs of wood about 2 feet in diameter revolving
one on the other. The inner shell was removed by hand by use of a
mortar and pestle (Smith 1936:21; Gray 1941:287). Later in the nine
teenth century, mills were developed which combined ail of the processes
of milling ricegrinding, cleaning, pounding, screening, and polish
ing (Gray 1941:730; Lawson V975:ll, illustration of the devices used in
a rice pounding mill). Power for these mills was supplied by tides,
steam engines, animal power, or wind. The steam-powered mills were the


X ~
>
ii;
WM0%


CHAPTER 6
PATTERN RECOGNITION IN SLAVE MATERIAL CULTURE
To demonstrate further that slavery assemblages from sites in
coastal Georgia display regional characteristics, the artifact groups
of four slavery sites were quantified and compared. Specifically, the
artifact frequencies for each slave assemblage were calculated and the
percentage range compared. Initially, it was intended that the examina
tion of artifact assemblages would include all of the sites indicated
in Figure 1. Unfortunately, the artifact data for some of these sites
are not available.* The sites used in this comparison include the pre
viously mentioned sites: Cannon's Point south cabins (MacFarlane 1975),
Cannon's Point north slave cabin (Otto 1975, 1977), Kingsley Plantation
(Fairbanks 1974), and Butler Island. Although Kingsley Plantation is
technically in Florida, it is located on a coastal barrier island and
it produced long-staple cotton.
Another reason for the selection of these three plantation con
texts is their differences in economic functions and plantation manage
ment. It h.as been pointed out that Butler Island was a rice plantation
and the other two long-staple cotton plantations. Each plantation,
however, was apparently managed very differently: Cannon's Point had a
*The analyses of Butler Point and Sinclair are presently in
process (Mullins-Mopre nd). Artifact counts were not published in the
report of the Rayfield plantation (Ascher and Fairbanks 1971). The
LeConte-Woodmanston plantation did not provide a slavery occupation
(Hamilton 1980).
211


78
were well provided for,as some recent historians have suggested to have
been the general nature of slavery in the United States (see Fogel and
Engeriran 1974); yet it is apparent from close inspection of these
records, particularly the overseers' letters, that inadequacies were
prevalent. Because the overseers' greatest concern was slave diseases,
most of the information relating to slave care is concern with health
care, and to a much lesser extent diet. References to clothing or
improvements to slave dwellings are few. Thus, it is difficult to assess
the material conditions of slavery at Butler Island on the basis of
plantation records alone. It is intended that the archaeological data
presented in Chapters 4 and 5 will amplify the descriptions of slave
material conditions found in the records. On the other hand, some as
pects of slave life mentioned in the records have no archaeological
correlates, including slave diseases, mortality rates, disciplinary
problems, and punishment. In the remainder of this chapter these topics
are briefly examined.
The Kings' greatest concern was with the slaves' health care.
Efforts therefore were made to place at Butler Island slaves who were
adapted to the swampy lowlands. Just how these adaptive capacities
were determined is not clear. Through mere observation such an assess
ment could have possibly been made. Roswell King. Sr., stated:
I landed the Negroes all safe and in good health. . .
I carried them to Tide Island [Butler Island] and you
would have been much gratified to see their rejoicing,
when they found they could drink the water out of the
river, which is proof that they have been acquainted
with the same soil & hopefully they will be healthy.
. . They landed very cheerful & happy, you have no
people that can talk with them. (13 May 1803)



201


170
characteristics; see Gejvall 1963.) Consequently, faunal materials
collected are very brittle and fragmentary. However, the presence of
charred food remains (not as severely charred as the Butler Island
collection) in and around slave dwellings has been reported (Fairbanks
1974:87: MacFarlane 1975:167). Perhaps this reflects a refuse dis
posal practice in which food remains were burned in the fireplace and
may have been subjected to repeated burnings. Periodically, the food
remains must have been swept out and onto the floor.*
Faunal analyses were conducted at the Zooarchaeology Laboratory of
the Florida State Museum. As a consequence of the previously described
conditions of the bone remains, most faunal materials could not be
identified beyond animal class. Yet some classifications to lower taxa
were possible (see Table 12). All bone fragments were weighed but not
counted. Although charring and other modifications can reduce bone
weights by as much as 50 percent, bone weight appears to be a more
accurate basis for quantification than mere fragment counts (Wing and
Brown 1979:126). Therefore, bone weight was preferred to bone count
in the Butler Island collection where so much of the bone was frag
mented. Here fragment counts would tend to misrepresent the relative
frequencies of the animal species identified within the sample.
Yet, despite the limitations of the faunal collection, the rela
tive proportions of domestic to nondomestic food animals, as indicated
in Table 13, correspond remarkably well with the collection from the
Cannon's Point north slave cabin (Otto 1975:310). Additionally, the
identified nondomestic species are typically found in the deltaic
habitats of the Georgia tidewater. Given the limitations of the faunal
*Char!es H. Fairbanks, Personal communication, 1980.


directly related to the vast amount of nails uncovered from the frame
slave dwellings. Also, it indicates that slave cabins have been the
focus of previous excavations. Perhaps, if more trash or midden deposits
were excavated, this artifact group would not dominate the profile. It
also strongly suggests that houses were the material aspect of slave life,
at least as seen archaeologically.
High proportions of kitchen artifacts have again been indicated.
The proportion of this artifact group within the total slave assem
blages further supports that the cabins were central to slave cooking and
eating activities.
Furniture and personal possessions are very scarce. Although
slaves may have been able to improve their material lots as a result of
the task system, luxury items such as furniture and personal possessions
were still hard to come by.
Firearms were represented at all the sites, but occupy a small
proportion of the total assemblages. Perhaps this indicates that slaves
had very limited access to guns.
Tobacco has the widest percentage range of all the artifact groups.
As previously stated, this possibly reflects whether or not slaves re
ceived provisions of tobacco. At Butler Island, tobacco and pipes were
provided occasionally.
Clothing is somewhat variable and could possibly be related to
status of the slave occupants at the various sites. Presumably, a high
status slave would have more clothing than a common field laborer. It
is difficult to make such an
assessment based upon such scanty evidence.
On the other hand, the activities group indicated very little variation.


198
Household Items
Items such as chamber pots, a pewter spoon, and furnishings are
designated household items. Only one artifact suggestive of furniture
was recovered, a brass drawer pull. Generally, the remains of furni
ture. at slave sites have been very scant, implying that furniture was
a rare item in slave dwellings.
Tobacco Equipment
Tobacco pipes were the only leisure-activity artifact found, there
being no toys or personal items related to leisure recovered. Numeri
cally, pipes constituted the third highest artifact group in the Butler
Island collection. A total of 642 pipe stems and bowls were collected,
a number considerably higher than that found at other slavery sites (see
Chapter 6). This may indicate that tobacco and pipes were periodically
supplied to the Butler Island slaves and that these commodities were not
supplied to slaves at the other sites. Both tobacco and "Negro pipes"
are listed on some of the annual expense lists. Like ceramics and
alcoholic beverages, tobacco and pipes may have been occasionally pro
vided to the slaves, with the slaves making additional purchases to sup
plement the provisions. Here again, the close proximity of Butler
Island to Darien may have made tobacco and pipes more easily obtainable
than to slaves on the barrier islands.
Most pipes appear to be of British or Dutch manufacture. No
American made pipes 'were identified. In fact, most pipes appear to be
Dutch. The recovery of a Peter Dori pipe indicates that the site was
occupied after 1850 (Humphrey 1969:15). Unfortunately, few of the
other pipes could be as precisely dated. Fluted bowls were well


Figure 12. Spanish reaie, "Mexico Mint,
from Settlement #4.


fVP
i'C jw,
TAKE
Tfl?rErcvSdw{thShs and Brick/^ortgLiMfel^-
STAXE
Figure 10. Stratigraphic Profile of Settlement #1.


Food Procurement
Two categories of food procurement artifacts were uncovered:
fishing equipment and firearms (see Figure 33). The only evidence of
fishing activity was provided by lead net sinkers. Numerous fragments
of lead slag suggest that these were melted down for making fish net
sinkers, weights, and buckshots.
The archaeological evidence of slaves' access to firearms is per
haps the most significant conflict with contemporary accounts of slavery.
At Butler Island, the evidence is more convincing than that previously
recovered from other slavery sites. Besides buckshots, musket balls,
and gunflints, a brass trigger guard was uncovered. To the author's
knowledge, a trigger guard has not been previously found in other slave
sites in Georgia. Fortunately, at Butler Island, the plantation records
also indicate that slaves were occasionally given the privilege of using
guns for hunting.
At Butler Island, a fellow asks me permission to get a gun
for ducks. Plenty of which are at the landing and in the
fields. I have sometime since taken all firearms from them
' as I think they have forfeited their charter from the swamps
and their conduct. I am often glad my phi 1 entropy in'allow
ing them to have guns did not extend further than this.
(Roswell King, Jr., 28 June 1829)
The archaeological and historical evidence is conclusive that slaves
had access to firearms for hunting at Butler Island. It is likely that
this was an established practice for other plantations in the area.
Gun flints were made from both black, or Dover flint,and French,or
honey-colored,flint. Curiously, several large flint nodules of both
black and honey-colored flint were found in and around the slave houses.
Although little evidence of flint debitage was uncovered, it is possible


165
handles of the eighteenth and nineteenth centuries (Noel-Hume 1976:182
(8 and 9); South 1977:98, below), they appear to have been homemade
items. Several uncarved bone fragments crudely sawed to form the same
shape were later recognized within the faunal collection. Also, the
fact that several woodworking tools were identified from structure one
(the provenance of the bone fragments) provides additional evidence to
suggest that the bone handles were homemade. At least it seems possible
that the slaves were able to carve these bone fragments. The largest
fragment has several scoring marks which evidently postdate the carved
motifs. The significance of the scoring is unknown but may indicate
later whittling activity after the handle was discarded (see Figure 32).
Kitchen Artifacts: Summary
The similarities in the artifact styles and the frequencies of the
kitchen artifacts between Butler Island and other slave sites, notably
Cannon's Point, are astonishing. These similarities provide a pre
liminary basis for the establishment of a slave kitchen artifact pat
tern for coastal Georgia. Subtle differences are evident. Perhaps the
most notable is the absence of identifiable eighteenth-century ceramics.
It cannot be ruled out that future excavations may uncover these.
Another possibility may be that Butler Island slaves obtained more
stylish ceramics than their barrier island counterparts. Their close
proximity to the town of Darien may have been a factor which enabled
them to purchase ceramics more frequently. Also, the barrier island
slaves had resident planters and were perhaps getting their castoffs.
But, in spite of these two differences, the kitchen artifact group tends
to reflect a regional pattern.


190
smaller sole fragments viere uncovered. For this reason, it has been
estimated that four shoes were represented within the collection. The
soles all appear to have been stitched rather than nailed (see Anderson
1968:56-65). Thread fragments are still apparent in some of the stitch
holes (Figure 35). Because they seem to be machine stitched, these are
most likely the remains of purchased shoes and not homemade items.
Accessories and Ornaments
Beads
Of ail the artifacts within this functional class, glass beads
were the most numerous. Blue, faceted, hexagonal beads found in several
slave contexts in coastal Georgia and recently in Tennessee (Smith 1977:
157) predominated. In addition to 16 blue beads, 15 beads of the same
style but in various colors viere found, including green, purple, orange,
red, and brown. Besides being found at slave sites, these cane beads
"occur in great quantities" at eighteenth- and nineteenth-century
Seminole Indian grave sites (Fairbanks 1974:90).
Three round, wire wound beads (Kidd 1970; type Wlbl7, Wlbl5,
Wlh!2)(see Figure 34) were also collected. These beads, though of a
style not reported at any other slave site, are probably contemporaneous
with the faceted beads. Both bead styles were possibly slave pur
chases.
The most curious bead, origin unknown, is one made of hematite
(Figure 34). While it may possibly have been brought from Africa or
made by the slaves, it may simply be an item taken from a prehistoric
site.*
*Jerald T. Milanich, Florida State Museum, Gainesville, Florida,
personal communication, 1980.


BIOGRAPHICAL SKETCH
Theresa A. Singleton was born in Charleston, South Carolina,
15 April 1952. She grew up in Charleston and graduated from Bishop
England High School in 1970. She then moved to Washington, D.C., to
enter Trinity College, where in 1974 she received the degree of Bachelor
of Arts with a major in anthropology-sociology. In 1975, she entered
the Graduate School at the University of Florida, where in 1977 she
received the degree Master of Arts with a major in anthropology and in
1980, the degree Doctor of Philosophy with a major in anthropology.
248
1 J 'fc'SU'-y /tv*' y r- \S-


Eighteenth-century ceramics form significant proportions of slave
ceramic assemblages at other sites, where the presence of outmoded
ceramics was indicated by the Mean Ceramic Formula. The absence of
eighteenth-century ceramics in the Butler Island collection could account
for the derivation of a fairly accurate median date of antebellum occupa
tion. Thus, the Butler Island slaves were receiving more stylish ceramics
than other slaves in the area but were perhaps using some slightly out
moded ceramics such as the annular pearlwares.
Nonceramlc Artlfacts
Nonceramic kitchen artifacts included glassware, cooking and
food processing equipment, and eating utencils.
Glass bottles and tablewares
The most abundant nonceramic artifact was glass, a total of 448
fragments having been recovered. Only 5 of these fragments were iden
tified as tablewares, and two vessel forms were represented: a possible
tumbler and a decanter or vase. The remaining fragments were found to
be bottle glass, but no whole or nearly whole bottles were recovered.
Table 11 provides the approximate style and number of bottles repre
sented.
As can be seen in Table 11, dark green bottles predominated. This
has been found to be the case at other slave sites. Dark green bottles
are particularly significant because they were generally used as con
tainers for brewed alcoholic beverages (Switzer 1974). Evidence for
alcohol consumption among the slaves at other sites in Georgia is
another household pattern observed in both cash crop contexts. It has


I certify that I have read this study and that in rny opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Charles H. Fairbanks, Chairman
Distinguished Service Professor
of Anthropology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
/
/ Jerald T. Mi tarn ch
/ Associate Professor of
l Anthropology
I certify that I have read this study and that in my opinion it
conforms to acceptable standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Otto Von Men ng""
Professor of Anthropology
I certify that I
conforms to acceptable
have read this study and that in my opinion it
standards of scholarly presentation and is fully
adequate, in scope and quality, as a dissertation for the degree of
Doctor of Philosophy.
Samuel Proctor
Distinguished Service Professor
of History
- ' *


209
The colono-ceramics which have been recently suggested to have been
made by Afro-Americans (see page 12) are frequently found at black
occupied sites in coastal South Carolina but have been nonexistent in
Georgia. The ceramics recovered from Butler Island, however, do not
resemble the sand and grit tempered colono-wares at all. Instead, they
are very reminiscent of the St. Johns series (Goggin 1952) and other
chalky wares in texture. It is possible that these are later manifesta
tions of St. Johns, but the sherd thickness, as well as the vessel form,
is strikingly different from the known St. Johns series. The vessel is
suggestive of a shallow bowl and the makers seem to have been trying to
imitate that European ceramic form. It is very likely that the slaves
at Butler Island experimented with the manufacture of ceramics. Both
the abundance of clay and the fact that clay bricks were regularly made
there are conditions which would have prompted ceramic making among the
slaves, assuming that a potter was among them. The manufacture of these
chalky textured ceramics may have been another way in which the Butler
Island slaves utilized the resources of their deltaic environment. It
is, however, difficult to draw conclusive evidence from what appears
to be one vessel. Thus, the suggestion that these ceramics were manu
factured by slaves remains an untested assumption until further evi
dence is presented.
Summary
Most of the artifacts of slavery recovered from Butler Island
are very similar to those found at long-staple cotton slave sites.
Differences exist, particularly with regards to items directly related
to rice production or the habitat of rice production (the delta).


2
At present, slave sites from coastal Georgia are perhaps better
known than those from any other regions of the Old South. In recent
years, a number of slave sites have been identified, located, and exca
vated in this area. Until now, no attempt has been made to synthesize
these archaeological findings within a framework larger than the single
piantation.
Archaeological investigations undertaken at the Butler Island
plantation in McIntosh County, Georgia, provide the first archeological
evidence of slave life at a Georgian rice plantation. A1! reported
excavated slave sites in Georgia are former long-staple cotton planta
tions on the barrier islands. Thus, the significance of the Butler Island
research is twofold: first, it supplies archaeological evidence for the
specific adaptations of slavery to rice culture; second, comparisons
of the Butler Island data with data from long-staple cotton slave sites
indicate several identical patterns. This evidence suggests that the
cultural system of slavery which developed in coastal Georgia exceeded
adaptations to specific cash crop requirements.
Ultimately, the goal of this dissertation is to identify and offer
possible explanations for slave community and household patterns. This
objective is accomplished through the analysis of primary settlement
and subsistence data recovered from Butler Island slave sites. Com
parative d
dence for
this study
ecologicai
ata from other coastal slave sites provide additional evi-
regional interpretations. The theoretical orientation of
is derived from regional archaeology, which utilized an
approach in the interpretation of cultural phenomena.


123
structure was painted with red oil-base paint. Red ochre was frequently
listed on the annual expense and supply lists. Evidently, the ochre was
used to paint some of the plantation structures. The reasons for paint
ing the small structure red, or for that matter painting it at all, is
unknown.
If the ditch did not contain human waste, another explanation for
the high phosphoric content of the soil is suggested by the rich deposit
of organic refuse. Apparently, the ditch was used intensively for the
deposit of refuse, particularly organic refuse. Peach pits, shoe
leather, and charred bone, as well as inorganic materials, were recovered
below the mottled silt (evidently fill) and continued below the water
table (see Figure 22). Unfortunately, excavations of the ditch had to
be terminated since the archaeologists lacked an appropriate water pump
ing device to continue excavations below the water table. Clearly, this
portion of the drainage ditch was used for refuse disposal.
No other concentrated refuse dumps were uncovered. Small post-
hole tests placed along other locations of the drainage ditch did not
yield any artifacts. This suggests that the refuse from the ditch next
to structure four is most likely from that household. Another possible
location for concentrated refuse disposal was the river. The site's
close proximity to the river suggests that trash was likely to have been
regularly discarded in it. All other refuse recovered at the site was
concentrated in and around the slave dwellings. No evidence of trash
pits or patterns of refuse disposal could be delineated (see South 1977:
47-80). Both inorganic and organic refuse were found together in
similar proportions both inside and outside the house. Also, it is


8
Cross-cultural comparisons of slavery are often concerned with the
issue of slave treatment within the confines of political entities
(e.g. Tannebaum 1949; Davis 1966; Degler 1971). Similarly, in the
United States discussions of slavery are confined to geographic areas
such as states (e.g. Flanders 1933) or time periods (e.g. Stampp 1956).
Both comparative studies and those within the United States are fre
quently concerned with slave management or the legal aspects of
slavery. Less often has the issue of slave behavior been addressed.
When it has, the entire slaveholding South has formed the unit of
analysis (e.g. Blassngame 1972; Genovese 1974). It is questionable
whether these discussions are applicable to slavery everywhere in the
Old South.
Although the slave cultural system which developed in the South
was characterized by general behavioral patterns (Gutman 1977), local
variations of this tradition were inevitable. Evidence of this has
been indicated in many localized Afro-American traditions, some of
which have survived until the present day. The coastal areas of
Georgia and South Carolina have been focal areas of African retentions
(Kerskovits 1958:120), and in the development of distinctive Afro-
American traditions, particularly in the decorative arts, language,
cuisine, and music.
Certainly the quality of slave life must have varied regionally.
Although this assumption remains untested, indications from archaeologi
cal resources are that the material standard of slaves in coastal
Georgia was better than that of slaves documented elsewhere in the
Old South (MacFarlane 1975; Otto 1975).
Further, a recent study of


29
People who are heterzygous for the sickle-cell trait are evidently
better able to survive and reproduce in a malarial region (Weiss and
Mann 1975:356). In the United States, studies have shown that a high
incidence of the sickle-cell trait presently exists among the "Gullah"
Negroes of the South Carolina and Georgia coasts. This finding sug
gests that Afro-Americans in these former rice-growing areas have per
petuated this genetic characteristic (Gunn 1975:5).
Given the medical knowledge of the day, it is doubtful that rice
planters understood the sickle-cell trait. It appears, however, that
some planters observed fairly early that blacks were less likely to
suffer from malaria and other diseases prevalent in swampy lowlands
than whites (Poste!1 1951:74-75; Savitt 1978:22-24), which established
a preference for black labor in rice fields. Thus, it is clear that the
Negro's adaptability to the coastal environment partially contributed to
the tradition which defended Afro-American slavery on the basis that
blacks were more suitable to perform work whites were unwilling to do
(Gunn 1975:6).
Another possible explanation for the preference of African labor
in the cultivation of rice was the African knowledge of rice growing.
Historians have overlooked this possibility until recently. Primarily,
through the pioneering efforts of Peter Hood, a preliminary investiga
tion of this suggestion has been undertaken (see Wood 1974:55-62).
Rice planting and processing involved obstacles to Europeans,
since they were generally unfamiliar with rice culture. Unlike
Europeans, Africans from several regions in West Africa were successful
rice producers (Wood 1974:58-59). In fact, some West Africans were


Table 12 (continued)
Species
Common Name
Bone Weight
in Grams
Percent
Odocoileus
virginianus
Whitetailed
deer
12.0
.54
Bos taurus
Domestic cow
442.5
20.16
Unidentified
mammals
867.00
39.51
Total
2194.20
100.00
Table 13
Relative Weights of Domestic and Nondomestic Animal Food
from Butler Island
Food Animals
Weight
Percent
Domestic
653.3
59.85
Nondomestic
438.2
40.15
Total
1091.5
100.00
aRodentia, Cricitidae, and Squamata not included.


Artifact Class
# of Items
Clear
Bottle
51
Tableware
5
Aqua
40
Total glassware
448
Clothing, adornments, and personal possessions
Buttons
Porcelain
12
Bone
4
Brass
56
Beads, glass
Hexagonal, faceted
32
Roung, wire-wound
3
Beads, hematite
1
Total
35
Miscellaneous items
Shoe soles (reconstructured)
4
Parasol attachment
1
Comb tooth
1
Pocket knife
E
Scissors
1
Tobacco equipment
White clay pipes
Stems
236
Bowl s
Plai n
154
Decorated
251
Porcelain
1
Total tobacco equipment
642
Whole
82
Fragments
62
Furniture
Brass drawer pull
i
',* rw.iv,.'t-.'o#**


137
plantation was primarily experimental and never really became sophisti
cated. Therefore, it is likely that the very well made clay bricks
used in the construction of structure five were bricks obtained from the
Savannah River region.
Building haroware
Although more nails were recovered than any other artifact class
(Appendix 2). few other construction artifacts could be positively iden
tified. Most iron materials were severely corroded. These materials
were weighed but provided few clues as to their identity. Two hinges
and a small pintle were the only hardware recognized from structure one.
Fewer construction materials were found in association with structures
four and five. It is possible that some building hardware was salvaged
and reused on other buildings once these structures fell into disuse.
Additional testing, however, is needed to support this suggestion.
Of the 4,367 nails and nail fragments, only 4 could be identified.
All 4 nails were classified as hand-wrought nails, including 2 rose heads,
1 L-headed, and 1 headless. The rose head was a general purpose nail
while the L-headed and the headless were most often used for flooring
or trimming (Noel-Hume 1976:253). The identification of hand-wrought
nails cannot be used to assume that most nails used in the construction
of slave dwellings were of this type. Although nails were made by black
smiths at Butler Island, .machine cut nails were frequently purchased,
as indicated by the annual expense lists. Thus, chances are that a
large portion of the nails recovered were in fact machine cut.
Virtually no window glass was present. This finding supports the
general consensus that slave houses had few, if any, glazed windows.
Windows were usually closed by shutters.


112
set by the superordinate planter class (Corbett 1941:14). Although
regional differences were evident in architectural styles and construc
tion materials, plantation arrangements were characterized by broad
similarities (Otto 1975:24). It has been suggested that the presence of
these similarities expressed the social ideals of the planter elite
(Corbett 1941:13-14).
Yet, to a lesser extent, plantation settlement patterns may have
expressed ideals other than those of a generalized planter elite. In
some cases, planter ethnicity or for that matter slaves' ethnicities may
have influenced settlement patterns. It is a well acknowledged fact
that southern antebellum architecture represents a blend of ethnic in
fluences (Kniffen 1965; Glassie 1968). Recently, the Afro-American
contribution to this architectural style has been investigated (Fine
1973:11-26; Anthony 1976; Viach 1978:122-138). Ethnic influences are
also evident in plantation layouts. Many of the early colonial planta
tions were intended to resemble medieval manors (e.g. Phillips 1929:49,
photograph of Mulberry Castle). Some postcolonial plantations also
exhibit the owner's ethnicity. For example, the spatial arrangement of
the big house and outbuildings at the Choclat plantation at Sapelo Island,
Georgia, is similar to a French Chatelet.* Choclat, however, was
atypical among Georgia tidewater plantations. It was not engaged in the
^Morgan R. Crook, West Georgia College, Carrollton, Georgia,
personal coiniunication, 1980. The Choclat ruins were mapped in 1974 by
Crook, who later discovered that the arrangement of these buildings was
similar to a French Chatelet. The present author pursued this sugges
tion further and found that Choclat1s layout was not only characteristic
of a French Chatelet, but that it was similar to contemporary French
settlement plans, circa 1786-1800 in general (see Mauban 1945:plate 18).


82
included rhubarb, Spirits of Turpentine, red lavender, aloes, alum,
magnesia, camphor, peppermint, castor oil, opium, and others. These
medications were typical of the antebellum apothecary (Postell 1951:98-
99). Black folk medicine 'was discouraged, although some slaves sought
help from "negro doctors" despite objections (Roswell King, Sr.,
10 August 1815).
Improvements in slave diet were apparently the primary actions
taken to prevent illness. This was especially true for children. Meals
were sometimes prepared separately for them to insure that they were
properly fed:
There is no estate with so fine of crop of young Negroes
as this Butler Island. There are over 100 hominy eaters,
during the summer once or twice a week they are collected
and have an extra meal of fish or pork soup which make
each little negroe grow in the river swamp, I think.
(Roswel1 King, Jr., 13 May 1827)
From time to time improvements were also made on the diets of adult
slaves, such as increases in the quality or quantity of plantation
rations. Generally, the amount of plantation ration varied according
to the slave's occupation. Ditchers received the largest amounts of
meat and corn; next in line were the drivers and mechanics; then came
the field workers; and last were invalids and retired slaves who re
ceived the least amount of food (Butler Estate Papers, 1815b). Slaves
kept vegetable gardens and raised chickens at Butler Island, but these
were periodically destroyed by freshets (Roswell King, Jr., 23 July
1829). Hunting (Roswell King, Sr., 28 January 1829) and fishing also
supplemented the diet, which is indicated by archeological evidence as
wel 1.


235
Fairbanks, Charles H. and Sue A. Mullins-Moore
1980 The archaeology of slavery. Early Man, summer (in press).
Fancher, Betsy
1971 The Lost Legacy of Georgia's Golden Isles. Doubleday, Garden
City, New York.
Ferguson, Lei and
1980 Looking for the "Afro" in Colono-Indi an pottery. In
Archaeological Perceptivas on Ethnicity in America.
edited by Robert L. Schuyler, pp. 14-28, Baywood Publishing,
Farmingdale, New York.
Fine, Elsa Honig
1973 The Afro-American Artist. Holt, Rinehart, and Winston,
New York.
Flanders, Ralph B.
1933 Plantation Slavery in Georgia. University of North Carolina
Press, Chapel Hill.
Fogel, Robert W. and Stanley L. Engeriran
1974 Time on the Cross: The Economics of American Negro Slavery.
Little, Brown, and Company, Boston.
Fontana, Bernard L,
1973 The cultural dimensions of pottery: Ceramics as social docu
ments. In Ceramics in America, edited by George Quirnby, pp. 1-
18. University of Virginia Press, Charlottesville.
Gall man, Robert E.
1970 Self-sufficiency in the cotton economy of the antebellum South.
Agrieu1 turn! History 44:5-24.
Garrison, Ervan
1977 Modeling inundation effects for planning and prediction. In
Ccnservatlon Archaeology: A Guide for Cultural Resource
Management. Studies, edited by Michael B. Schiffer and George
Gutnmernia.n, pp. 151-157. Academic Press, New York.
Gejvall, Nils-Gustaf
1S63 Cremations. In Science in Archaeology: A Comprehensive Survey
of Progress and Research, edited by Don Brothwell and Eric Higgs,
pp. 468-479. Basic Books, New York.
Genovese, Eugene D.
1974 Roll Jordon Roll: The World the Slaves Made. Pantheon Books,
New York.
Georgia Writers Project
1940 Drums and Shadows: Survival Studies Among Georgia Coastal
Negroes. University of Georgia Press, Athens.
" ' -


42
fiber carefully. Harvesting required 10 to 12 pickings, whereas the
short-staple only required three (Gray 1941:735). After it was picked,
the cotton was placed on a platform to dry in the sun and later stored
in the barn to await ginning.
The famous "Whitney saw gin," invented for the removal of seed
from short-staple cotton, was not used for ginning sea-island cotton;
instead, "roller gins" were used. In the early days these were hand-
powered mills derived from the Indian churka, Later improvements to
this apparatus led to the invention of gins powered by animals or
steam (Hammond 1897:72). After the cotton was ginned, it was "mated"
that is the residue from the broken seeds was removed. The clean
cotton was then hand-packed in bales, since hand-packed bales were less
injurious to cotton and consequently brought higher prices than those
packed by a press (Gray 1941:736).
Both acreage-per hand and yields per acre were severely limited
because of the laborious cultivation and processing methods. Where cul
tivation relied exclusively upon hand labor, the average number of acres
cultivated per laborer was 3 to 4. Where the plow was substituted for the
hoe, the average number of acres per laborer was 6 to 7. The average
yield per acre ranged from 125 to 400 pounds, although yields recorded as
high as 2,000 pounds per acre were known (Gray 1941:737; Phillips 1969:
275).
Only superior long-staple cotton commanded very high prices, but
its production was limited to a small group of planters (Allston 1854:
5S6). For the rest of staple, price fluctuations were considerable.
The average price range during periods of good to high prices was $.40
to $.80 a pound, and at periods of depressed prices $.14 to $.25 a


9
slave nutritional adequacy indicates that slaves in the coastal areas
of South Carolina, Georgia, and northeast Florida were adequately
nourished (Gibbs et al. 1980). These studies demonstrate that regional
approaches in the study of slavery are needed.
It is possible that certain economic, environmental, and demo
graphic conditions in coastal Georgia favored a distinctive develop
ment of slave behavior. In Georgia, the largest percentages of blacks
to whites were found on the coast. In addition, the most valuable farms,
the largest slaveholding, and the most livestock were located there.
Another factor was the value of lands and buildings per farm. This
was highest on the coast than in any part of Georgia (Flanders 1933:81).
Ecologically, the region provided abundant plant and animal resources
which were exploited for both domestic and industrial purposes. Employ
ment of the task slave labor system on most coastal plantations allowed
slaves more time for recreation and to improve their material lots
than the gang labor system employed elsewhere in the Old South (see
Otto 1975). For the most part, these conditions were not present in the
interior. It is likely that variations in these conditions produced
variations in slave behavior.
Historic factors were also important. Unlike their counterparts
of the short-staple cotton belt, most planters that came to coastal
Georgia were experienced slaveholders from the West Indies or South
Carolina (Vanstory 1970:74-79; Davis 1976:132). They were presumably
acquainted with the problems of slave management. Conceivably, they
attempted to establish the best possible conditions for a "stable
regime within which their slave could live" (Genovese 1974:6). Perhaps.


74
jble 4(continued)
Name
Agt
Task Rate0
Clarinda
Violet
Fanny
Molly
Harry
Lilly
Charlotte
Temple
Total
Captain (driver)
Dorcas
Lagette
Jenny
Cate
Dick
Charity
Morris
Melinda
Peter
London
York
Judy
Affa
Lilia
Mi ra
Violet
Pompe
Oncy
Joe
Juba
Betty
George
Lender
Pena
Philo
Elce
Silva
15
20
17
37
44
28
23
52
57
25
21
19
29
35
37
43
40
1
16
25
29
42
42
17
12
37
33
30
22
19
20
25
25
24
1/4
1/4
1/2
1/2
1
]
25 1/2
1/2
3/4
1/2
1/4
1/4
1/2
3/4
1 o ta i
1/2
22 3/4


132
Figure 24. Hearth of Chimney.
(Clay bricks: straight edges;
tabby bricks: rounded edges)
- '


ABBREVIATIONS USED IN REFERENCES
BFPC
Butler Family Papers Collection #1447
GHS
Georgia Historical Society, Savannah, Georgia
MDCC
Margaret Davis Cate Collection
PHS
Pennsylvania Historical Society, Philadelphia, Pennsylvania
WFPC
Wister Family Papers Collection (Butler Section)
228


44
%
Clearly, the conservatism toward the use of hand labor, manures, and
crop rotation had its origins in rice culture. Manuring was not essen
tial to rice growing since silt loam from irrigation aided soil fer
tility (Cole 1927:601). Similarly, hand labor was vital to rice cul
ture because of the difficulties of plowing. Perhaps an early implemen
tation of plowing and manuring in the cultivation of long-staple cotton
would have resulted in lower production costs and higher yields than
those actually experienced by most coastal planters. At least, the
success of the sea-island cotton industry in Florida was partially re
lated to the greater use of the plow (Gray 1941:737). Unfortunately,
the failure to employ crop rotation methods by both cultures resulted in
early soil exhaustion, which had disastrous effects upon post-bellum
agriculture in the area.
Likewise, labor practices were very similar. Here again, the
task system which developed from rice culture was later used for cotton
culture. Under the task system, field hands were classified as full,
three-quarter, half, and one-quarter hands (Flanders 1933:143-144).
Generally, each full laborer was expected to work a task of 1/4 to 1/2
acre everyday, whether in rice fields or cotton fields. This system
was particularly advantageous if the worker completed the task by mid
afternoon. Upon completion of the task, he or she had the remainder of
the day to tend to personal needs. Conversely, under the gang system,
the other major labor system for plantation slavery, workers labored in
the fields from sunrise to sunset (Phillips 1929:279-280; Stampp 1956:
53-56).


ACKNOW!. EDGMENTS
Many persons made this effort possible. Several faculty members
at the University of Florida have been central in fiel ping me develop my
research and writing skills, particularly Drs. Fairbanks, Jerald T.
Milanich, Maxine Margolis, Prudence Rice, Samuel Proctor, and Ronald
Foreman. Dr. Fairbanks has been a mentor and friend throughout my gradu
ate career and I have enjoyed studying with him. I also would like to
thank Dr. Von Mering for reading the final draft for my defense.
Dr. Kathleen A. Deagan, my first instructor in field methods of histori
cal archaeology, has always been supportive and interested in my career
goals.
Very special thanks go to Dr. Lewis Larson, State Archaeologist
for Georgia. I am very pleased that he asked me to undertake the Butler
Island Project. Also, his staff deserve special mention, particularly
Linda Tiironings, for undertaking all of the bookkeeping responsibilities
for the project.
I am very grateful to the field crews of summer 1978, spring 1979,
and summer 1979 for working along with me in the Butler Island marsh.
I think we are the only ones who can truly appreciate what slave life
must have been like at Butler Island. I am particularly grateful to
Ernest Despainge for the detailed drawings of the chimney; Bill Fisher
for helping out with the laboratory analysis and for asking his father,
Dr. Richard Fisher, to undertake the soil analyses; and Lorraine McCoske
for his assistance with the photographs.


130
The function of the irregular pattern of postholes located out
side of structure one could not be determined (see Figure 23). Although
only one side of this feature vas uncovered, it is suggestive of a pen
or coop. Historic sources indicate that the Butler slaves raised both
pigs and chickens. Possibly, these postholes represent a portion of an
enclosure to keep pigs or chickens fenced in. Postwar photographs of
rice coast slaves or freedmen houses frequently show fenced yards.
One of the posts of the presumed enclosure intruded into another
feature, which was possibly a fire pit. Charred bone and wood frag
ments were recovered from this feature. Such a fire pit could have
been used for a number of purposes, including to ward off bugs, burn
trash, wash clothes, or melt lead for the making of fishweights or buck
shots.
Chimney constructions
The most detailed information regarding the construction of the
cabins was derived from the chimneys. With the exception of structure
five, all of the chimneys appear to have been constructed of tabby
bricks. Even the clay brick chimney was found to have tabby bricks in
its lowest course.
Tabby, a construction material used extensively along the Atlantic
Seaboard in the eighteenth and nineteenth centuries, has been defined
as a concrete construction-like material made from equal or nearly equal
portions of lime, shells, water, and sand (Gritzner 1978:9). Like
adobe and modern concrete, tabby bricks are not fired in the manufac
turing process. Instead, the semi-fluid mixture was poured into brick
molds and left to harden for a period of 6 to 18 months. Close
.


SM 83- 82- 2685 Si
(S) Copyright 1980
by
Theresa A. Singleton
'


CHAPTER 2
PLANTATION AGRICULTURE IN THE GEORGIA TIDEWATER
Environmental Summary and Introduct i_o n
Tidewater refers to the geographic area of the coastal plain,
which is effected by the influence of the tides. In Georgia, this
region includes the barrier islands and extends approximately 75 kilo
meters inland to the point where extensive areas of pine forests begin
to occur (Crook 1978:28-30) (see figure 2). Three major subdivisions
characterize the Georgia tidewater: the strand section, the salt marsh
.and lagoon .section, and the delta section. These are discussed in de
tail elsewhere (Larson 1970; Johnson et al. 1374), and are briefly sum
marized here.
The strand borders on open ocean and is composed of beach and
dunes. It is the most rapidly changing of the three sections. Vegeta
tion is restricted and most animal resources which occur there are found
more frequently in other habitats along the coast.
Of the three subdivisions, the salt marsh and lagoon section is
the most ecologically diverse. It is composed of four habitats: marsh,
tidal creeks, hammocks or highlands, and freshwater swamps. As a
result of its diversity, this section is rich in both plant and animal
resources. The food chain begins in these salt marshes. Several
species of fish, shellfish, turtles, and waterfowl thrive in the marsh
The hammock forest community includes several
20
and tidal creeks.


38
rice sold for $.80 to $2.23 per bushel or 2 to 4.75 cents per pound
(House 1954b:78).
Rice, like all other plantation crops, was marketed through fac
tors in major port cities. The factor served as the "middleman" be
tween the planters and the rice merchants. He was tine banker and stock
broker for the planter and provided him with credit in meeting opera
tional costs. Since the planter depended upon the returns from his
crop for cash, he was often in debt to the factor when rice sales were
low (Easterby 1941). Prior to 1840, most Georgian rice was sold through
factors based in Charleston. After that date, Georgian planters in
creasingly sold their rice from Savannah. By 1859, Savannah had 82
factor establishments. Savannah experienced a rapid growth as a rice
port. Its population increased; a thriving trade developed with ports
in the West Indies, Mobile, and New Orleans; and it became an important
social and cultural center (House 1954b:70-79).
The rice industry on the Carolina-Georgia costs was seriously
beginning to decline on the eve of the Civil War, although large-scale
rice production did not vanish completely from the area until the end
of the nineteenth century. Competition from rice-growing states in the
southwest and the rising cost and scarcity of slaves (Cole 1927:600-
601) were the major causes for this new crisis facing the rice planter.
Nonetheless, the industry managed to survive because of large crop
yields. After the Civil War, however,
prosperity to the industry were futile,
increased flooding from freshets caused
attempts to restore the former
The new labor system and
by accelerated erosion in the
interior (Trimble 1969) made rice cultivation unprofitable. Gradually,


204
Two tools, the function of which could not be determined, are
shown in Figure 40. The first item is a thick U-shaped object. The
other appears to have been used as a mold for making small rounded ob
jects, possibly buckshot or fish net sinkers, but the function of the
screw holes on each side is unknown.
It is interesting that most of the tools and implements viere
uncovered within the house. At structure one the distribution of some
of the tools has been plotted (see Figure 22). The presence of tools
within the slave cabin structure strongly suggests that slave dwell
ings doubled as draft workshops and tool sheds. Apparently, tool sheds
were not available at the site.
Although woodworking tools were identified, no carved wooden ob
jects were found. Thus far, very little evidence of slave crafts has
been revealed archaeologically in coastal Georgia, yet the area is well
known for its postbellum Afro-American crafts (Georgia Writers Project
1940; Vlach 1978). There are, however, some items at Butler Island
which may be suggestive of slave craft activity, for example, the
carved bone fragments. Also, a hugh hand-hewn cypress log found close
to settlement #4 was possibly smoothed down for the construction of a
canoe or other object. The log was approximately 25 feet long (7.47
meters) and was found in the vicinity of the old antebellum dike. It
would be difficult to demonstrate that this log was intended to be
a canoe, but it appears to have been hand-hewn, and someone had begun
hollowing out the inside. Assuming that this is an antebellum artifact,
it is likely that slaves were responsible for it.
More convincing evidence of slave crafts is provided by the
occurrence of three fragments of a crudely made ceramic (Figure 41).
£
t.


148
Undecorated white ceramics. The most frequently occurring ceramics
were undecorated white fragments, possibly including creamware, pearlware,
and whiteware, with most appearing to be whiteware. Those classed as
whiteware were defined as such on the basis of a white overall glaze tint
and by the absence of the blue puddling characteristic of pearlware
(Price 1979:13-15) or by the creamy white color and greenish-yellow
puddling found on the creamwares (Noel-Hume 1973:217-254). The undeco
rated whitewares are found in a variety of vessel forms. The Butler
Island collection includes chamber pots, bowls, mugs, platters, and
plates.
Transfer printed pearlware and whiteware. The surface decorations
on transfer printed ceramics were mass produced by the use of copper plate
engravings. These ceramics are one of the most frequently occurring
at late eighteenth- and nineteenth-century sites (Noel-Hume 1976:129-
130). At slave sites, however, they appear to be minority wares, and are
frequently associated with high status households. Only 17 sherds were
recovered, but three vessel forms are represented: a platter, a plate,
and a bowl or cup (see Figure 28).
Handpalnted pearlware and whiteware. Blue-on-white and later
polychrome floral designs were painted on plain white bodies. These
ceramics seem to have been particularly popular among poorer classes.
Beginning in 1835, bright floral designs were stenciled on these ceramics
(Noel-Hume 1976:129).
Edged wares. White bodied vessels with embossed feathers, shells,
or fish scales placed on the edge were usually painted in blue or green.
Typical vessel forms are plates and platters, but other vessels are
known (see Watkins 1970, Figure 3a),

:> isscbftaw* -'


72
An Approximate
Table 4
List of Slaves by Age and Task Rate at Butler
1803-1845: Field Hands
Island,
Name
Age
Task Rate3
Sambo (driver)
33
Pender
31
1
Robi n
25
1
Cloa
26
1
Eve
23
1
John
20
1
Lewi s
18
1/4
Robin
25
1
Cornelia
30
1
Joe
43
1
Hannah
44
1
Limas
23
1
Amy
18
1
Charlotte
15
1
Molly
15
1/4
July
46
1
i
Nancy
26
1
Renter
56
i
FI ora
53
1
Mary
32
1
Li stern
29
1
Molly
25
1
Billy
49
1
Quacer
35
1
Mary
35
1
Toby
34
1
Adam
35
1
Minder
35
1
Total
25 1/2
Jack (driver)
57
__ _
London
28
I
Dido
21
I
John
29
1
Beck
16
1/4
Harry
29
1
Nancy
18
1/2
J1 mmy
35
i
l
Toby
37
1
Nancy
52
1
Constant
16
1/4
Tinsh
14
1/4


slave 1 i fe ways of the two cash crop regions. Archaeological!./, the
differences are primarily discernible in slave community organizations,
the natural resources exploited, and slave crafts. Similarities are
evident in most household artifact patterns: plantation food rations,
food preparation equipment, personal possessions, and leisure-time activi
ties. It is suggested that the differences in slave life reflect the
dissimilarity between the habitats where rice and long-staple cotton were
produced. On the other hand, the similarities may reflect general labor
management practices adapted for the production of coastal staples.


¡able 7
Mortality Rates
of Infants and
Children of
Slaves on
the Butler Esta
te, 1819-1834
Number
Ini
rants
Chi1dv
~en (1-4)
Children
(5-14)
Total
Year
(Total deaths)
#
%
It
0/
io
#
%
Jf
ii
0/
10
1819
19
9
47
16
3
15
15
79
1820
16
7
43
2
13
0
0
9
56
1821
27
3
11
5
18.5
3
11
11
40.5
1822
22
3
14
7
32
3
14
13
60
1823
14
12
86
0
0
0
0
12
86
1 £24
18
9
50
2
11
0
0
11
61
1825
28
9
32
C
j
18
i
3
15
53
1826
23
15
65
i
4
2
9
13
78
1827
39
24
62
5
13
2
5
31
80
1328
25
2]
84
3
12
1
4
25
100
1829
16
8
50
3
19
2
12
13
81.5
1830
25
10
40
3
12
2
8
! 5
60
1831
n ^
au
15
75
r'
£.
10
1
5
18
90
1832
29
12
41
5
17
1
3
18
61
1833
34
21
62
r*
0
14
0 .
0
26
76
1834
34
17
50
5
15
1
3
23
68
Total 389
200
51
56
15
25
5
281
72
CO


Figure 38.
Porcelain Pipe Bowl.


17
in slave life from two cash crop sites within a narrowly defined region.
It became apparent in the field that the time needed to investi
gate the first objective was insufficient for the time and funds
available for fieldwork. As a result, attention was then directed
toward uncovering archaeological evidence associated with the slave
community plan. Presumably, settlement and certain subsistence data
would provide archaeological evidence for the special adaptations of
slavery to rice cultivation. At the same time, comparison of the
Butler Island data with long-staple cotton plantations would supply
additional evidence of slave life which exceeded adaptations to cash
crops. In other words, the comparison would provide data specifically
related to the economic function of the plantation, on one hand and
data which were not so related on the other hand. Further, it is
hypothesized by the author that similar artifact patterns occurring in
both crop loci are indicative of the material conditions of slave life
and slave behavior patterns in a larger coastal Georgia area.
The factors which are believed to
have influenced a
riistinctive
development of slavery in coastal Georgia have been briefly discussed.
It is important, however, to define more accurately their role in the
development of coastal slavery. The demographic, economic, and
historic factors (see pages 9 and 10} are all related to plantation
management practices. These management practices developed out of an
adaptation to the coastal environment of Georgia. Although the habitats
for the cultivation of rice and long-staple cotton are very dissimilar,
both habitats are part of a larger ecosystem (Johnson et al. 1974).
Thus, this is a study of adaptation on two levels: adaptation to


Long, F. L.
1958 Chemical properties and the fertility requirements of some
brackish marsh soils of Georgia. Unpublished M.S. Thesis.
University of Georgia, Athens.
Lovell, Carole Couper
1932 Golden Isles of Georgia. Little, Brown and Company, Boston.
Ludlum, David M.
1963 Early American Hurricanes, 1492-1878. American Meteorological
Society, Boston.
Luscomb, Sally C.
1975 The Co11ector1 $ Encyclopedia of Buttons. Crown Pub1ishers,
New York.
MacFarlane, Suzanne
1975 The ethnoarcheology of a slave cabin community: Couper
Plantation site. Unpublished M.A. Thesis. University of
Florida, Gainesville.
Mauban, A.
1S45 L1Architecture Francaise.
De Jean Maviette, Vanoest, Paris.
McClane
1978
A. J.
Field Guide to_ Freshwater Fishes of North America.
Rinehart and Winston, New York.
Hoi t,
McGimsey Charles R. III and Hester A. Davis
1977 (Editors) The Management of Archaeological Resources: Airlie
House Report. Special Publication of the Society for American
Archaeology, Washington, D.C.
McIntosh County Records
MS Deed Books. H-52, 1-20, 1-160.
Courthouse, Darien, Georgia.
On file in McIntosh County
Miller, Elinor and Euguene D. Genovese
1974 Piantation, Town,_and County: Essays in the Local History
of American Slave Society. University of Illinois Press,
Urbana.
Sue A.
An archaeological survey of Butler Point, St. Simons
Island, Georgia. Unpublished Report on File. Department
of Anthropology, University of Florida.
M u 11 i n s Mo o re, S u. e A.
nd The Antebellum Barrier Island Plantation: A Search for an
Archaeological_ Pattern. Ph.D. Dissertation. University of
Florida, Gainesville (In preparation).
Tul Iins,
1978


126
Slave Dwel 1 J ngs
Unfortunately none of the slave dwellings at settlement #4 were
completely excavated. Testing at structure one and the historic photo
graphs of slave houses at settlement #1 (Figures 18 and 20) provide
evidence of architectural style and some specific construction details.
All of the slave structures at Butler Island were evidently duplex
dwellings with central chimneys. This "two pen, saddle-bag" house type
(Kniffen 1965:556) was commonly used for slave housing throughout the
South. Single-pen dwellings, however, were equally as common. Regional
differences to this basic architectural style were most evident in con
struction materials. In the piedmount and interior coastal plain, slave
houses were frequently built with timber logs. Frame, brick, and tabby
constructions were typical of the tidewater (Otto 1975:104-105).
All indications are that the slave dwellings at Butler Island were
of frame construction. Contemporary accounts of nineteenth-century
visitors to Butler Island (Kemble 1961 (1863):67; Lane 1973:194), his
toric photographs, and archaeological resources support this finding.
These structures were evidently made of cypress. The walls were pre
sumably vertical board-and-batten as indicated in Figure 18. This is
not surprising because cypress, a very durable building material
(Panshin 1964:489), was abundant in the rice swamps. In fact, all
wooden remains uncovered at Butler Island have been identified as
cypress. At settlement #4, structure one provided evidence of cypress
construction. In Figure 23 the dotted line represents the outline of a
large fragment of hand-hewn cypress, possibly the remains of the floor.
Additional fragments were found scattered around the chimney.


104
mentis
Figure 13. Contour Map of Settlement #4. (Note five chimney falls
in lower left.)


Virtually little or no information is contained in plantation records
regarding the building materials used in slave housing, the size of
these dwellings, or the wild food resources utilized by slaves, and many
other details. These items of slave life are discussed in the remainder
of this Study.
Archaeological Resources
Archaeological investigations were undertaken at Butler Island
in the summer of 1978 under the direction of Dr. Lewis H. Larson, of
West Georgia College, and in the spring and summer of 1979 by the author.
The cyclical inundation practices of the Waterfowl Management Program
necessitated that all fieldwork be conducted in the spring and summer
months. This schedule, which involves the flooding of most of Butler
Island in the fall and gradually removing the water from late winter to
early spring, prohibited fieldwork before April. Even in April, many
areas were still very damp, and some were not completely dry for field-
work activities until June or July. Thus, the time spent at any one
site depended, to a great extent, upon the degree of moisture present at
the site during the time allocated for fieldwork. Fieldwork activities
included surface reconnaissance by foot and by boat, mapping of sites,
and limited excavations at slave settlement #1, the southernmost tidal
mill, and slave settlement #4 (see Figure 7).
Site Assessmen
t,T>
The sites designated in Figure 7 are presumably antebellum period
sites. These have been located and identified by use of a combination
of resources including historic maps, photographs, records, aerial
photographs, informants, and archaeological assessments. Recent land


of these immigrants, and he became one of the wealthiest slaveholders
in the South. In fact, the Georgia coast became known as a region of
wealthy planters. Possibly, this wealth affected the quality of slave
life. At least this seems to have been the case with the Butler estate.
The Butler slaves are documented to have received items (alcohol, pipes,
tobacco, and ceramics) which were apparently not supplied to slaves at
many other plantations.
Regional interpretations of slave lifeways have been offered for
the entire coastal Georgia area. Yet variations are likely to have
existed. Variations resulting from habitat specific differences have
been discussed. The proximity of plantations to towns or cities may
also have been an influential condition. It has been suggested that
the higher frequences at Butler Island of beads, pipes, and stylish
ceramics may have been related to the fact that the Butler Island slaves
were able to trade more often with nearby merchants than were slaves on th
barrier islands. Slave life styles in towns and cities have been docu
mented to be very dissimilar to those on rural plantations (see Miller
and Genovese 1974:337-452; articles on slavery in town and city). This
finding, however, awaits archaeological investigation. In coastal
Georgia, no urban or town slave sites have been archaeologically inves
tigated, and none of the urban-orientated Savannah River plantations
has been excavated. Therefore, it is not known whether the regional
interpretations offered here extend to those situations. At the
southern extreme of the coast, variations to the ceramic patterns dis
cussed in Chapter 5 have already been noted in the St. Mary's River


As at other coastal slave sites, a variety of ceramic styles were
found to be present at Butler Island. These assemblages seem to indi
cate that slaves received an assortment of unmatched ceramics. In the
Butler Island collection some vessels appear to have been "wasters,"
those which have no market value (Noel-Hume 1974:170). Generally the
wasters were found to be poorly fired, warped, or with a sloppily applied
glaze. Apparently, ceramics were not regularly supplied to the Butler
Island slaves, as these items do not appear on the annual expense lists.
However, they may have been provided occasionally, and there are two ref
erences to the "purchase of 50 crates of crockery" (Butler Estate Papers,
1811; Roswell King, Jr., 18 March 1929). The size of this purchase sug
gests that these ceramics were for the slaves. How often ceramics were
supplied to the slaves could not be ascertained, but it appears to have
been infrequent. Most likely, slaves purchased their own crockery with
money they earned from selling produce. Merchants in Darien frequently
advertised the sale of ceramics, and indications are that these were
perhaps special wares obtained exclusively for slave use (Otto 1975:
160).
While a variety of vessel forms was recognized, bowls and cups
predominated (see Table 8). The frequency of bowls and cups strongly
suggests that most slaves' meals were one-dish soup, stew, or porridges.
A high frequency of bowls has also been observed at other Afro-American
sites (e.g. Baker 1978). Baker suggested that the high frequency of
bowls may be not only an artifact of poverty, but also an indicator of
Afro-American cuisine (1978:81).
Of the decorated ceramics, the annular wares occurred in very
high frequencies. At Butler Island, these ceramics comprised over 52
percent of all the decorated ceramics (see Table 9).


145
Table 8-
-(continued)
Ceramic Style
Vessel Form
Number
Mocha
Whiteware
Serving bowl
1
Yellow paste
Serving bowl
1
Martelized
Bowl and cups
2
Sponge decorated
Serving bowl?
1
Stonewares
Alkaline glazed
Storage jars
2
Brown salt glazed
Storage jars
2
Low fired unidenti
tied
Storage jar
1
Porcelain
Indeterminate
1
Total
59
Grand Total
Bowls or cups
32
Plates/platters
11
Storage jars
6
Mugs
2
Chamber pots
4
Other
4


resources. Even taking into consideration the grave limitations of the
Butler Island faunal collection, freshwater fishes and turtles were
very abundant resources in the delta and possibly were exploited most
often by the slaves.
Table 15
Relative Weight of Identified Nondomestic Food by Animal Class
Class
Weight
Percent
Fishes
137.1
31.1
Turtles
285.6
64.7
Bi rbs
2.0
.5
Mammals
14.0
3.7
Total
438.2
100.00
Fishes and turtles
The three predominant fishes identified include bowfin (Amia calva),
gar (Lepiososteus sp.), and freshwater catfish (Icaturidae), all found
most frequently in turbid, stagnant waters.'
The bowfin is a primitive fish of a large family, most of which
now exist only as fossils. The bowfin has a unique air bladder which
enables it to live in water unsuitable for other fish. It is a bottom
feeder and usually inhabits weedy lakes and sluggish streams. Seldom is
it found in fast currents (McClane 1978:178-179).
Gar is another primitive fish. It is primarily found in fresh
water but has a high tolerance for saltwater. Like the bowfin, gar has
a modified air bladder which it can use for breathing in very stagnant


47
Butler Island and Environs.
Figure 4


Kelly, Marsha C. S. and Roger Kelly
1980 Approaches to ethnic identifications in historical archeology.
In Archaeological Perceptivas on Ethnicity in America, edited
by Robert L. Schuyler' pp. 133-143. Baywood Publishing,
New York.
Kemble, Frances A.
1882 Record of a Later Life. Henry Holt, New York.
1961 Journa1 of A Residence on A Georgian Plantation In 1833-1839
(18637 edited by John A. Scott. Alfred A. Knopf, New York.
Ketchum, William C., Jr.
1971 The Pottery and Porcelain Collectors Handbook: A Guide to
Early American Ceramics From Main to California.'" Funk' &
Wagnalls, New York
Kidd, Kenneth and M. A. Kidd
1970 A classification system for glass beads for the use of field
archaeologists. In Occasional Papers in Archaeology and
History #1. Ontario, Canada.
Kill on, Ronald and Charles Waiter
1973 (Edi to rs) Slavery Time When I Was Chi Hun Down on Master's
Plantations. Beehive Press, Savannah, Georgia.
King, Roswell, Jr.
1819- Manuscript. (Letters to Frances Butler, Major Butler, and
1835 Thomas Butler). BFPC, boxes 10.11, 12 and WFPC, boxes 33,
40, PHS.
1828 On the management of the Butler estate, and the cultivation of
sugar cane. Southern Agriculturalist 1:523-529.
1844 Daybook. (Estate of John and Pierce Butler in Georgia).
MDCC, folder 205, GHS.
King, Roswell, Sr.
1803- Manuscript. (Letters to Major Pierce Butler). BFPC, boxes
1819 9, 10, PHS.
King, Thomas, Patricia Parker Hickman, and Gary Berg
Anthropology in Historic Preservation:___ Caring for Culture's
Clutter. Academic Press, New York.
Knapp, S. A.
1900 Rice culture in the United States. U.S. Department of
Agriculture Farmers Bulletin jfT!0_. Government Printing Office,
Washington, D.C.
Kniffen, Fred
1965 Folk housing: Key to diffusion. Annals of the Association
of American Geographers 55(4):549-577.
''y --
, : i t \ ....
:V'ns-f -w,..


Figure 41
Slave-Made Cerami


Figure 5. The Butler Georgian
Estate.
(Adapted from Scott 1561.)


25
Under royal government, the colony experienced an accelerated
growth. A significant source of immigration to the coastal area was
the South Carolina planter. South Carolinians were aware of the rice
growing potential of the Georgia tidewater areas, end they migrated to
Georgia with their families and slaves (Wilms 1972); rice soon became
Georgia's most important crop. Later, the introduction of indigo gave
the plantation system an added boost (Spalding in Coleman 1977:52).
Planters were attracted to the coastal lands, but good lands were
not abundant. Moreover, the development of these lands was very costly.
An eighteenth-century observer in coastal Georgia estimated that the
initial expense of settling a rice estate was about $10,000, and for
cotton or indigo $9,000 (Thayer 1957:81). Consequently, only large,
wealthy planters could afford to acquire and cultivate these lands. The
small planter unable to compete settled further in the interior. Thus,
coastal Georgia became a region of a few, wealthy planters shortly
after the emergence of plantations to the area (Calloway 1948:47-49).
The Revolutionary War brought disastrous setbacks to the coastal
plantation economy. The bounty on indigo formerly received from the
British government ceased and the'exportation of rice declined. Pros
perity was restored, however, with the successful introduction of long-
staple cotton in 1786. Planters of rice plantations acquired land and
equipment for cotton production. Thus, rice was temporarily dis
placed as a major staple of the tidewater (Flanders 1933:55-57),
By the mid-nineteenth century, the reliance upon sea-island
cotton gradually shifted to rice. Apparently, this change roughly
coincided with periods of depressed long-staple cotton prices (Gray


Figure 11,
Eighteenth-Century Slipware Fragments
from Settlement #1.


230
Baker, Vernon G.
1978 Historical archaeology at Black Lucys garden, Andover,
Massachusetts: Ceramics from the site of a 19th century
Afro-American. Papers of the Robert S. Peabody Foundation for
Archaeology #8. Phillips Academy, Andover, Massachusetts.
1580 Archaeological visibility of Afro-American culture: An example
from Black Lucy's garden, Andover, Massachusetts. In Archaeo
logical Perceptives on Ethnicity in America, edited by Robert
Schuyler, pp. 29-37. Baywood Publishing, Farmingdale, New York.
Barker, Sandford W.
1841- Ledger. Barker Plantation Records. (Experiments on Rice
1844 Culture). South Carolina Historical Society Manuscript
Collection, Charleston, South Carolina.
Bassett, John S.
1968 Southern Plantation Overseer as Revealed in His Letters. Negro
Universities Press, New York.
Berry, Brian
1964 Approaches to regional analysis: A synthesis. Annals of the
Association of American Geographers 54:2-11.
1968 A synthesis of formal and functional regions using a general
field theory of spatial behavior. In Spatial Analysis: A
Reader in Statistical Geography, edited by Brian J. L. Berry
and Duane F. Marble, pp. 419-430. Prentice-Hall, Englewood
Cliffs, New Jersey.
Binford, Lewis R.
1964 A consideration of archaeological research design. American
Antiquity 29:425-441.
Blassingame, John W.
1972 The Slave Community: Plantation Life in the_Antebel1um South.
Oxford University Press, New York.
1976 Status and social structure in the slave community: Evidence
from new sources. In Perceptives and Irony in American
SI avery, edited by Harry P. Owens, pp. 137-152. University
of Mississippi Press, Jackson.
Boney, F. N.
1977 Part three: 1820-1865. In A History of Georgia, edited by
Kenneth Coleman, pp. 129-187. University of Georgia, Athens.
Bonner,
1964
James C.
A History of Georgia Agriculture. 1732-1860.
Georgia Press, Athens.
University of


223
number of findings were uncovered that were not only absent in the docu
ments, but also, in some cases, were contrary to them. Clearly, archaeo
logical data need to be incorporated with historical sources to investi
gate the slave subsistence patterns. It may be that slave artifact
pattern recognition should begin with the establishment of artifact pro
files from well-documented sites. Possibly, the artifact profile as well
as the community organizations at Butler Island can serve as models for
future investigations of rice coast slave sites.


Figure 36. Parasol Attachmen


CHAPTER 4
SETTLEMENT PATTERNS OF COASTAL SLAVE COMMUNITIES
The Determinants of Plantation Sett!ement Patterns
The determinants of settlement patterns have been defined as those
"classes of factors that interact with each other to produce the spatial
organizations of a social group" (Trigger 1968:53). These determinants
can be conceived of in terms of three levels of settlement: first, the
individual structure; second, the arrangements of structures within a.
community; and third, the manner in which communities are distributed
over the landscape (Trigger 1968:55).
At slave villages, what were the factors that determined spatial
arrangements? In order to answer this question, it may be necessary to
examine the plantation settlement as a whole. Certainly, topographic,
economic, and social factors influenced, if they did not determine, the
spatial arrangements of most plantations. Precisely what these factors
were at coastal plantations and the ways these were manifested at Butler
Island are briefly explored in this chapter. To date, very little has
been written on the subject of plantation settlement pattern. Archaeo
logical studies are especially few in number (see Otto 1975:18-28; Lewis
1979). The attempt is made in this discussion to synthesize from
a v a Hable settle me n t
lished works in anth
pattern data obtained from plantation plats; pub-
ropology, history, and geography; and the archaeo
logical resources uncovered
Sutler Island.
109


LIST OF TABLES
¡ble
Page
1. Ownership of Butler Island, 1790? to the Present 50
2. Rice Production at Butler Island, 1802-1833 64
3. Annual Births and Deaths of Butler Slaves 71
4. An Approximate List of Slaves by Age and Task Rate at Butler
Island, 1803-1845: Field Hands 72
5. An Approximate List of Slaves, by Age and Occupation, at
Butler Island, 1803-1845: Plantation Specialists 76
6. 1830 Census of Slaves at Butler Island 77
7. Mortality Rates of Infants and Children of Sieves on the
Butler Estate, 1818-1834 81
8. Vessel Forms of Ceramic Styles 144
9. Frequencies of Decorated Tablewares 155
10. Application of the Mean Ceramic Date Formula to Tablewares 156
11. Styles of Glass Bottles Represented at Butler Island,
Settlemen
t h 4
158
12.
Identified Fauna from
Slav
e Site #4, Butler Island
171
13.
Relative Weights of Domest
from Butler Island
ic and Nondomestic Animal Food
1 72
14.
The ReTat
ive Weight of
Domestic Species at Butler Island
174
15.
Relative
Class
Weight of Ide
nti f
led Nondomestic Food by Animal
179
15,
Artifact
Profiles for
F o u r
Slavery Sites in Georgia/Florida
214
17.
Adjusted
Artifact Prof
lie
for Kingsley P1 antation
215
18.
The Slave
1 Artifact Pat
te rn
in Coastal Georgia/FIorida
216


62
for Butler Island, King sought the counsel of two drivers at Hampton
Plantation (Roswell King, Sr., 15 November 1818). Several years later,
Roswell, Jr., praised the work of a driver by saying: "The hands appear
to be fit enough to attack another crop, which can only be attributed
to Sambo's management in varying the work as much as possible" (Roswell
King, Jr., 24 September 1826). The death of a good driver was con
sidered great loss to the estate, as well as to the entire planter
community: "Driver Bram . was worth a great deal to the estate .
his death is regretted by every white person who know him" (Roswell
King, Jr., 1 February 1825). But problems with drivers were not un
common. One at Butler Island was involved in stealing large amounts of
corn from the warehouses and selling the corn to merchants in Darien
(Roswell King, Sr., 22 November 1806). Another driver, also at Butler
Island, apparently was an alcoholic (Roswell King, Jr., 29 August 1806).
Occasionally, a driver's cruelty to the field hands required that he be
removed:
I have been obliged to displace Morris at St. Anne.
He was always tyranical and brutal, which I gener
ally kept under, but lately he had become intemperate,
in fits of that kind, and has used great severity to
some of the negroes. (Roswell King, Jr., 16 July
1826)
Despite occasional problems, most drivers were good at their jobs and
to the field hands (see Van Deburg 1979,for a reassessment of the slave
driver). Generally, drivers "strove to mediate between the Big House
and the quarters" (Genovese 1974:378). At Butler Island, drivers
were particulalry vital to the management of a plantation which saw
few whites.


31
the difficulties posed by the enslavement of Indians, and the insuf
ficient supply of indentured whites (Wood 1974:37-48). Despite the
fundamental role these factors contributed to the existence of African
enslavement in the New World, it is suggested that the practical advan
tages offered by African physical adaptability and African knowledge of
rice growing fostered the Carolina-Georgia rice industry.
Methods of Irrigation and PI anting Rice
Two different species of rice were cultivated in the Southern
United States: lowland rice, Oryza sativa, and upland rice, Oryza mutica.
Lowland rice, a marsh plant, required an abundant supply of water for
ripening (Austin 1893:9). Of its many varieties, "goldseed" rice and
"white" rice were cultivated in the lowland areas of the Atlantic Sea
board (Knapp 1900:5). Upland rice was grown by dry culture in various
hilly areas of the cotton belt. It was not a commercial crop, but
grown essentially for home use (Gray 1941:723).
Lowland rice was probably first grown in moist soils without
irrigation (Gray 1941:229). Around 1724, irrigation was employed in
what were called "inland swamps." These were small fields adjacent
to freshwater streams. The rice fields were irrigated with water-
stored in reservoirs formed by dams. After irrigation, the water was
drained through ditches into streams. Because the water supply in the
reservoirs was dependent upon rainfall, the crop was subject to either
frequent floods or drought, often making yields small and uncertain
(Heyward 1937:12-16; Wi1 ms 1972:50-51 ).
By the late eighteenth century or perhaps earlier (see Wilms
1972), rice planters began to recognize the advantages of cultivating
-
- ,


193
Mls cel 1aneous I terns
Included within miscellaneous items are a brass object, possibly
a parasol attachment (Figure 36); a glass lens (Figure 37); a hard
rubber tooth, presumably of a comb; and a bone handled pocket knife.
The brass object appears to have been part of the apparatus used
to open and close an umbrella belonging to a female. This attachment
and the little china buttons were uncovered from structure one, which
provided convincing evidence of a female's presence at that household.
The glass lens is likely to have been part of a pair of glass
spectacles, but its magnification is presently very slight. If not an
eye piece, it could have been a lens for an unknown object.
The size or type of comb represented by the uncovered hard rubber
tooth could not be ascertained, but it provides evidence that the
Butler Island slaves had a few amenities.
The pocket knive is the only personal possession that has also
been uncovered at other slave sites. Its occurrence at Butler Island
and other slave sites may indicate that pocket knives were typically
owned by slaves.
Summary
The previously described personal items and those mentioned else
where in this study are very significant. Whether these items were
obtained through purchase, stealing, or gift, they indicate that slaves
made efforts to improve upon their material conditions. Assuming that
the items were obtained through purchase, the task labor system is
likely to have been a crucial factor for making the purchases possible.


gure
Page
22.
Stratigraphic Profile of Drainage Ditch.
124
23.
The Extent of Excavation at Structure One, Settlement #4.
123
24.
Hearth of Chimney.
132
25.
Chimney Brickwork.
133
26.
Arch-Shaped Hearth Support.
135
27.
Red Coarse Earthenwares.
147
23.
Transfer Printed Pearlware Fragments.
150
29.
Annular Ceramics.
153
30.
Annular Ceramics.
153
31.
Eating Utenc.ils.
154
32.
Carved Bone Fragments.
164
33.
Food Procurement Artifacts.
168
34.
Clothing and Personal Adornments.
189
LO
CO
Shoe Soles.
192
36.
Parasol Attachment.
195
37.
Eye Lens.
197
33.
Porcelain Pipe Bowl.
201
39.
Woodwork!ng Tools.
203
40.
Unidentified Tools.
206
41.
Slave-Made Ceramics.
208
xi
s-'r:::-: -:.Vx


the funds available were insufficient for the time needed to map and
test these sites.
Ultimately, the investigations of slave villages at Butler Island
centered upon slave settlement #4. This site was evidently the best
preserved of the four settlements. Visually apparent above the ground
were five chimney falls. The clearing of the chimneys revealed that
these were H-shaped hearths, presumably of duplex slave dwellings with
central fireplaces. A contour map of this site was made in the summer
of 1S78 (see Figure 13), and it was decided at that time to return the
following spring and summer to investigate the site further.
The field strategy devised for settlement #4 during the 1979 field
season was twofold: first, two or more of the slave dwellings would be
tested for structural information and, second, a subsurface testing pro
cedure would be developed to locate other activity areas hidden beneath
the ground surface. It was hoped that outbuildings, concentrated refuse
dumps, wells, and privies could be identified from subsurface testing.
The testing scheme developed utilized a systematic random sampling pro
cedure. At 10 meter intervals, a row of posthole tests was made ev
ery 5 meters. In the beginning, a hand posthole digger was used, but it
was soon replaced with a pointed shovel, the latter being better suited
for the clayey soil. Although the shovel did not yield tests of a con
sistent size, most tests ranged from 25 to 33 centimeters. A total of
222 tests were made. These tests, however, did not provide useful in
formation for delineating areas of refuse dumps, wells, or privies.
Artifacts recovered from the tests were concentrated in and around slave
dwellings. The exception to this rule was the uncovering of structures


2.22
district.* Perhaps these variations are related to the fact that planta
tions in Camden County or along the St. Mary's River were exposed to
more of a frontier environment than those of the Altamaha estuary. The
St. Mary's district was settled considerably later and it is further
away from the port cities of Savannah and Charleston than is the Altamaha
district. It is possible that gradations in slave material conditions
existed from the more urban-centered plantations along the coast to
more rural ones. Thus, even along the narrow belt of the Georgia coast
variations in slave lifeways existed.
Recently, it has been suggested that the unit of analysis for the
study of slavery be limited to the county (Otto 1979). The county may in
fact be the maximal areal extent for any regional interpretation of
slavery. Still it has been possible here to tentatively suggest that
regularities existed in slave life which exceeded both habitat specific
influences or county lines. To suggest that these patterns extend beyond
the limits of the Georgia tidewater needs to be investigated. Hope
fully, future regional studies of slavery in Georgia and elsewhere will
help to refine the suggestions offered here.
Perhaps the greatest contribution of the Butler Island study is
that it has further demonstrated that archaeological data can provide
information not available in historic documents (see Fairbanks 1977).
Butler Island is a very wel1-documented historic site. In many ways
the documentary resources outshine the very limited archaeological data
recovered. Yet, in spite of the limitations of the archaeology, a
*Chad Bradley, Assistant Archaeologist, Kings Bay Project of
Anthropology, University of Florida, Gainesville, personal communica
tion, 1980.


Figure.37. Eye Lens.


155
Table 9
Frequencies of Decorated Tablewares
Ceramic Style
Fragments
Percent
Edged ware
53
20.78
Transfer printed
17
6.66
Handpainted
48
18.82
Sponge decorated
2
.80
Annular
135
52.94
Total
255
100.00
The high frequencies of banded ceramics at Georgian slave sites suggest
that these were inexpensive ceramics obtained for slave use.
Dating the ceramics from But!er Island
One method to determine slave usage of outmoded ceramics would be
to apply the Mean Ceramic Data Formula (South 1977:201-230) to those
uncovered at settlement #4. Although the application of the formula to
nineteenth-century ceramics has not been refined, it has been used to
demonstrate the presence of out-of-date ceramics in another slavery con
text (see Fairbanks 1974).
The formula was applied only to ceramic styles of the Butler
Island collection for which South had determined median dates of manu
facture (South 1977:210-212). A median date of 1835.01 was derived
(see Table 10). This date closely approximates the actual median date
of antebellum occupation, assuming that the site was occupied between
1804 and 1861.


24
Al tamaha Rivers. The trustees desired to establish a colony with small
settlements of independent farmers (Bonner 1964:2). To obtain this
objective, the trustees prohibited large land grants and Negro slavery.
The plantation system was outlawed for several reasons. First, the
colony was purportedly founded for debtors, although debtors made up
fewer than a dozen of the settlers during the entire trusteeship period
(Spalding in Coleman 1977:18). A large portion of tfie early settlers to
Georgia, however, were small businessmen, farmers, tradesmen, and
unemployed laborers (Coleman 1976:36-54), who could not be expected to
compete with a planter class. Second, Georgia was selected by the
trustees to produce commodities needed by England, such as silk, which
was not adapted to plantation agriculture. Third, the colony was to serve
as a military buffer to protect Carolina from the Spaniards in Florida.
Plantations would spread the people over the land instead of concentrat
ing them into settlements. Moreover, a large slave population could sub
ject the colony to slave insurrection, making it vulnerable to Spanish
attack (Potter 1932:116-117; Calloway 1948:30-31).
From the very beginning some colonists objected to the restrictions
placed upon Georgia. These "malcontents" were particularly displeased
with the prohibition of slavery. They were convinced that the use of
a cheap labor force was the only way to develop the colony. As dis
satisfaction grew, many colonists abandoned Georgia. On the brink of
economic distress, it became apparent to the trustees that if the
colony was to succeed it must be made attractive to potential colonists
(Davis 1976:126). Finally, in 1750, slavery was made legal. But,
discouraged by their failures, the trustees surrendered their charter
to the British Crown in 1752 (Spalding in Coleman 1977:44).


Table 4-(continued)
Name
Age
Task Ratea
Frank (driver)
42
Elce
23
1
A1bert
52
1
Lender
44
1
Brewster
33
1
Lookey
35
1
Molly
13
1/4
Jerry
16
1/4
Bruths
36
1
Jenny
34
1
K.egew
39
1
Jack
21
1
Tom
37
1
Amy
29
1
Hagar
22
i
Ti na
25
1
Mingo
17
1/4
Justice
32
3/4
John
33
1
Phoebe
32
1
Betty
35
i
Mary
41
3/4
Garrett
25
1
Ley
25
7
\
Rose
42
1
Patty
18
3/4
Esther
15
1/4
Cate
29
1
Cooper Nero
29
1
Mary
29
1
i
Total
--
26 3/4
Grand Total
122 1/2
Note. Total number = 156.
U1 = full time, 1/2 half time, etc.


of cholera, dysentry, diarrhea, typhoid, hepatitis (Savitt 1978:59).
By far the greatest sufferers among the slave population at Butler
Island were infants and chi 1 dren. The infant mortality rate was higher
there than at any of the other plantations in the Butler estate:
You will perceive that the mortality among infants is
very high particularly at #6 this place [Butler Island]
the state of the atmosphere is certainly injurious to
infants. The proportion of young people at this place is
far behind St. Simons. (Roswell King, Jr., 5 January 1829)
Deaths among infants at the Butler estate often comprised 50 percent or
more of the annual deaths as indicated in Table 7. Infant mortality
rates such as these were typical of the late antebellum period (Steckel
1979:95-96). Several conditions generated high levels of slave infant
mortality, but recent studies have suggested that the predominant causes
of slave infant deaths were Sudden Infant Death Syndrome (SIDS), or crib
death (Sutch 1976:283-292; Savitt 1978:122-127), tetani-related
diseases (Savitt 1978:121-122), and occasionally infanticide (Savitt
1978:127-128). Although no definite cases of infanticide were reported
at Butler Island, carelessness or neglect on the part of mothers was
given as the cause of death for some infants (Roswell King, Sr.,
16 June 1816; Roswell King, Jr., 17 September 1826). Most deaths among
children between the ages of 2 and 10 years were believed to be due to
worm infestation (Roswell King, Or., 10 September 1835). Worms were a
general health problem among black children in the Old South (Savitt
1978:128).
The treatment of diseases was usually handled by the Kings. In
cases of major emergencies, a doctor was called upon to tend to the
The frequently used medications listed on annual expense lists
slaves.


116
ITW ft
?* > ; V-- 1 " ,; ?
- : mk
. -
Figure 19. Steam Mill Ruins.
Figure 20. Linear Arrangement of Slave Dwellings.


1 9?.


168
'S,'39r-


215
To correct for the difference in the architectural frequency of
Kingsley, a correction factor was added to this artifact group. The
corrected artifact count and frequency for the architectural group in
dicated in Table 17 were derived by adding approximately the number of
nails recovered, if the Kingsley structure had been frame, to the
actual architectural count indicated in Table 16. The approximation for
the number of nails recovered from a frame dwelling was derived from
taking the average number of nails recovered from the three slave sites
with frame structures. The adjusted artifact profile for the Kingsley
plantation is given below. Admittedly, the adjustment of the Kingsley
Table 17
Adjusted Artifact Profile for Kingsley Plantation
Artifact Group
Artifact Count
Frequency
Architecture
3950
73.23a
Ki tchen
1385
25.66
Furniture

0.00
Clothing
18
.34
Guns
10
.18
Personal
5
.09
Tobacco
15
.28
Activities
12
. 22
ocal
5394
ICO.00
Corrected
artifact count and frequency.


129
The hearth level of the chimney for structure one indicates that
it was elevated off the ground approximately 69 centimeters or 2 1/4
feet. No evidence of brick footing piers was uncovered, but the round
posthole approximately 3.72 meters from the center of the chimney (see
Figure 22) may have functioned as a timber building pier. If this
posthole represents the outer wall of the structure, then the width of
this dwelling was approximately 7.44 meters,or 24 feet. The two small
square posts to the left of structure four's chimney (see Figure 21)
indicate that it was approximately the same width. Given a width of 24
feet, the length must be close to 43 feet. These dimensions, 24 by 48
feet are slightly larger than the dimensions given for most duplex slave
houses. Dimensions of slave houses at the other Butler plantations
appear to have been 20 by 40 feet (Roswell King, Jr., 6 Feburary 1835,
1 February 1835). Moreover, Olmsted noted that the duplex family tene
ments on the Georgia rice coast were 21 by 42 feet (1968 (1856):466).
Both Olmsted's and King's dimensions suggest that the slaves at Butler
Island had a little more space available to them than the average slave
family on the Georgia coast.
In 1811, 86 slaves were listed on the provision list for
settlement #4. Assuming that eight duplexes were at that time on the
premises, the average number of slaves to one duplex was 10.7 or So
slayes per family unit. These figures correspond remarkably with the
5.2 figure recently suggested as an estimate for the number of slaves
living in a single family structure (see Fogel and Engerman (1974:115).
These figures also adhere to those given by Fanny Kemble for Butler
Island (1961 (1863):67) and those of Olmsted for other rice plantations
(1968 (1856):422).


91
historic resources. The site east of settlement #1 is purported to be
a slave cemetery. But surface evidence of a burial ground in the form
of headboards, tombstones, or distinctive Afro-American burial patterns
(Combes 1974) is lacking. Because of the dense tree cover, it is pos
sible that subtle evidence of a graveyard was overlooked. Attempts to
locate coffins by probing with a metal rod proved to be unsuccessful.
Although no superficial evidence indicated that this area was a grave
yard or, for that matter, a site, its size and location strongly suggest
that it was a site of some type. Because of the lack of time and
resources, its function could not be ascertained. Additional historical
research and fieldwork are needed to determine its function.
The site located between slave settlements #3 and #4 appears to
have been a mill or other industrial site. No structural features are
evident, but a nearby impounding pond and the site's layout appear to
be similar to that of the excavated tidal mill. More specifically,
historic evidence suggests that this was the site of a sugar mill.
Roswell King placed the "sugar works" at Butler Island halfway between
slave settlements #3 and #4 (Roswell King, Sr., 20 April 1816). This is
the only location in that area indicative of a site. Only through
future excavations can this suggestion be confirmed.
Like the previously described site, it is believed that the site
in the northwest corner of Butler Island is also a mill or an industrial
site. The evidence pointing to this assessment is more conclusive than
for the other sites of unknown function. Again, an impounding pond
similar to that of the tidal mill is present at this site. The recovery
of wooden fragments identical to the excavated waterwheel further


Figure 28
Transfer Printed Pearlware Fragments.


-s;vsjssrw^ %¡ i
CHAPTER 3
HISTORICAL AND ARCHAEOLOGICAL INVESTIGATIONS
AT BUTLER ISLAND
Butler Island is one of several "brackish marsh islands" (Long
1953:11} within the Altamaha drainage in McIntosh County, Georgia (see
Figure 4). These islands are termed brackish because they become
covered with saltwater when tides are pushed by high wind and are in
undated with freshwater when rivers overflow as in the case of freshets
(U.S. Dept, of Agriculture 1961: 21), The land consists of sediments
washed down from streams flowing out of the piedmont and the coastal
plain. Soils vary but are primarily clayey. Generally, these marsh
soils are poorly drained and can be extremely acidic, particularly on
unreclaimed marsh (Long 1958:13).
During the nineteenth century, most of these islands were con
verted for rice culture. Later in the twentieth century, truck farming
replaced rice culture. Most recently, these marshlands have become
managed breeding grounds and refuge for waterfowl. Since 1954, several
of these islands have been acquired by the Game and Fish Commission of
the Georgia Department on Natural Resources for the development of a
waterfowl management area. Butler Island and adjacent Champney Island
were the first to be acquired for the program (Anonymous nd). Today,
Butler Island has become the headquarters for the Altamaha Waterfowl
Management Area.
46


1941:38). During this time of renewed rice interests, Carolina planters
intensified their activities in Georgia. They not only increased their
acreage in the old rice producing areas along the Savannah (see Clifton
1978) and Altamha Rivers, but also expanded further southward to include
the Satilla and St. Mary's Rivers (see Figure 3). In the 1850s, cotton
prices were on the upswing (Gray 1941:739) and rice cultivation was at
its peak (Hilliard 1975:65). Planters producing both crops were enjoy
ing sizeable profits on the eve of the Civil War
Devastation from the
Civil War prevented the successful revival of either culture in the
post-war years. The reign of rice and long-staple cotton as the dominant
cash crops of the coast had come to an end.
The Rice Industry
Early experiments with rice cultivation were attempted in
colonial Virginia, but rice was first established as a staple in the
South Carolina low country (Gray 1941:277). Although the details of its
introduction remain unclear, apparently early promoters of the colony
encouraged rice cultivation shortly after the founding of Carolina in
1670. By 1695, South Carolina's rice industry had begun (Salley 1919).
Rice and Slavery
si
Methods of rice cultivation during most of the plantation era were
entirely dependent upon hand labor. As a result, large-scale rice pro
duction required an enormous labor force. By the mid-nineteenth century,
approximately 15 laborers were needed to cultivate 100 acres of rice
land or 6 to 7 acres per hand (House 1954a:151). Presumably, colonial
practices demanded greater labor requirements. For example, the average


Figure 32. Carved Bone Fragments. (Note scoring in upper fragment.)


The archaeological evidence of slavery, however* may not be as
difficult to recognize as are other Afro-American sites. Recently,
an excellent attempt was made to identify an undocumented site as a
slave site at Spiers Landing in Berkley, South Carolina (Drucker and
Anthony 1979). Although Handler and Lange (1973:228) conclude that
archaeological data do not identify the slave status or slavery,
the Spiers Landing example refutes this. To suggest that slavery sites
can be identified entirely from archaeological remains is very pre
mature at this stage of research. This objective, however, may be
realized in the future through the use of regional approaches to Afro-
American sites, particularly the utilization cf pattern recognition
techniques.
Finally, in their discussion of ethnic identification in his
torical archaeology, the Kellys recommend the use of regional research
designs (1980:135-6). They suggest that research designs geared toward
the recognition of ethnic cultural patterns within a defined region may
aid in the delineation of geographic variations of a broad cultural
tradition (1980:130). The delineation of a geographic and temporal
variation of a broad Afro-American cultural tradition is precisely the
objective of this study.
Problems and Hypotheses
From the available reports, it appears that slave sites are known
better archaeological!:/ in coastal Georgia than in any other region
of the Old South (see Figure 1, for,the location of these sites). De
spite this knowledge, these studies have suffered from three major


APPENDIX 1
SUMMARY OF SOIL ANALYSIS FROM SETTLEMENT #4, BUTLER ISLAND*
Provenance
Zone A (Figure 15)
Zone B (Figure 15)
Zone C (Figure 15)
Structure six
Drainage ditch
(adjacent to structure four)
10YR 5/1** (grey) with/7.5YR 5/6
(strong brown) flecks; silt loam; crumb
structure; slightly hard dry, friable
moist, slightly sticky wet (soil acidity)
4.9; P (phorphorus) content 21 ppm
(parts per mi Ilion).
1OYR 5/1 (grey) with/7.5YR 6/8 (redish
yellow) flecks and charcoal; clay loam;
blocky structure; hard dry, firm moist,
slightly sticky wet; pH 5.1; P content
38 ppm.
10YR 6/1 (grey) with/7.5YR 5/8 (strong
brown) mottles; silty clay; massive
structure; very dry, very firm moist,
sticky wet; pH 5.3; P content 27 ppm;
contains, old cypress roots.
1OYR 3/2 (very dark grey brown); loamy
coarse sand;'massive, cemented with
iron oxide and organic matter (organic
matter 12%); hard dry, friable moist,
non-sticky; pH 5.9; P content 14 ppm;
spodic horizon (iron hard pan) with
old roots that have largely decayed or
old earthen floor with decayed straw.
7.5YR 3/2 (dark brown) with/7.5R 3/8
(dark red) pigment chunks (possibly old
paint pigment); silty clay; massive
structure; slightly hard dry, friable
moist, slightly sticky wet; pH 5.5;
P content 135 ppm; organic matter 20%;
high P and organic matter suggest old
privy but test for human colioform
bacteria was negative.
*Kobert F. Fisher, School of Forest Resources and Conservation,
University of Florida, Gainesville, written communication.
**Moist Munsell colors.
224


63
Of all the properties in the estate, Butler Island was the largest
in both acreage and in production. More slaves were utilized in the
plantation operation than at any of the other properties. As many as
300 to 400 slaves 1 ived and worked at this rice island and they occupied
four slave settlements or villages. Rice production, however, was not
the major interest of the estate until the 1820s. Prior to that time,
long-staple cotton dominated the planting interest (Scott 1961:xxxviii-
xxxix). Most of the cotton was planted at St. Simons and Little
St. Simons, but some was also cultivated at Butler Island, apparently
for crop rotation purposes (see Coulter 1941:104). Despite the interest
in cotton, considerable quantities of rice were produced before 1820.
As indicated in Table ?., the average annual production of rice was 712
tierces between the years 1802 and 1820. From 1821 to 1833, the
average annual production (1827 and 1829 are not included) was 1,246
tierces. Thus, in later years, rice production increased by approxi
mately 43 percent. Years of low production coincided with losses due
to storms, as well as to years when the acreage planted in cotton may
have exceeded that of rice.
The rice cultivation methods utilized at Butler Island were
generally typical of the region. As a rule, the most sophisticated
methods and equipment were employed. As early as 1802, rice fields were
manured, possibly with lime and marl (Barker, 1841-1844) to enhance soil
productivity (Roswell King, Sr., 17 September 1803). Fields were
planted with rice one year and then rotated with cotton to keep down
the growth of grasses (Coulter 1941:104). Occasionally, the fields
were prepared with plows, but hand labor predominated. In the processing


73
Table 4~(continued)
Name
Age
Task Rate
a
Quash
36
Will
13
FI ora
32
Auber
18
Jeffrey
37
Dally
45
Dago
26
Jean
34
Hagar
48
Sarah
18
Lisbon
24
Flu
2.5
Tamfsor
22
Mari a
42
1
1/2
i
i
3/4
1
1
i
i
1
1
1
Total
Ley (driver)
30
Peggy
31
Sarah
19
Randal 1
45
Hector
24
Peggy
17
Sam
33
Libby
27
Provamy
23
Toby
19
Peter
16
Hagar
52
Le i ra
52
Lei pi a
26
Ccsmba
42
Quanri n
24
Hagar
17
Robi n
16
Frank
29
May
22
Abram
42
Cootnba
43
Billy
19
Lew
20
Lucy
29
Jimmy
47
Rose
20
1
1
1
1
i
1
1
1
3/4
1/4
1/2
1/4
1/2
1/2
3/4
1/4
1/4
1
1
1
1/2
3/4
3/4


141
b. Food procurement activities (hunting and fishing)
c. Food resources
1. Domestic
2. Nondomestic
2. Personal posessions
a. Clothing
b. Accessories and ornaments
c. Household items
3. Lei sure-time activities
a. Tobacco equipment
b. Games and toys
c. Miscellaneous objects
4. Farming and specialized crafts
Obviously many of these groups crosscut categories, which is
generally a problem with the establishment of functional categories.
But to discuss these artifacts on the basis of material characteristics
alone would be meaningless to this study. These artifact groups very
closely resemble those established by South (1977:95-102) for the defi
nition of artifact patterns, and his classes are used in Chapter 6 to
derive a slave artifact pattern. Counts of artifacts recovered from
slave settlement #4 are listed in Appendix 2.
Kitchen Artifacts
According to historic sources, food
kitchens in each of the slave villages at
was prepared at central
Butler Island (Kemble 1961
(1863):55).
'et it is curious that kitchen artifacts form the second
highest artifact class next to building hardware. Although most of


61
selected to replace him upon recommendations from Roswell, Sr., and
John Couper of Cannon's Point on St. Simons Island (Couper to Major
Butler, 15 February 1818). Roswell, Jr., presided as overseer until
1839. It was young King's resignation which resulted in the trip Pierce
Butler, Jr., and iris wife, Fanny Kemble, made to Georgia. Butler had
come to Georgia to hire a replacement for Roswell, Jr. Thomas Olden,
formerly the manager at James Hamilton Couper1s Hopeton plantation
(Anonymous 1965?), was hired to managed the estate. Olden died in 1841,
and Roswell, Jr., once again became involved with the management of the
Butler estate, but this time in the capacity of a steward. As a
steward, he supervised two overseers and managed the account hooks.
Roswell. Jr., held this post until the nrid-1850s (Scarborough 1966:168).
Unfortunately, most of the plantation records for the Butler
estate, particularly the overseers' records, are between the years 1803
and 1835, prior to Fanny Kemble's visit. Within these 32 years, however,
it is possible to obtain a very clear picture of planting and labor
management activities. After 1835, it becomes difficult to document
management practices, but it may be reasonable to assume that earlier,
established practices prevailed in later years.
Under the management of the Kings, the Butler estate was operated
efficiently. Both Kings had taken an active interest in planting. With
regard to labor management, runaways were few, and slaves received treat
ment comparable.to other slaves in the area. To a great extent, their
success as managers may have been related to the dutiful drivers of the
estate. Most of the drivers were loyal and reliable, and the Kings felt
a high regard for them. For example, in the selection of a new driver


Zone
Zone
Zone
Zone
N 120
WATERTABLE
(=:=! MOTTLED SREY/BROWN/YELLOW
SILT
T| GREY SILTY LOAM
I METER
c MM GREY CLAYEY LOAM
0U77\ GREY CLAY WITH RED FLECKS
Figure 22. Stratigraphic Profile of Drainage Ditch.


18
specific crop requirements and adaptation to a more generalized coastal
environment.
The purpose of this dissertation is to test the following
hypotheses and their implications:
Hypothesis A: Slave community and household patterns in coastal
Georgia reflect adaptations to the specialized habitats where tidewater
staples were produced. Archaeologically this will be identified in the
following:
1. Patterned placement of slave village sites relative to
habitats.
2. Community plan of spatial arrangements
a. Specialized buildings and material culture
b. Nonstructural features
3. Slave dwellings
a. Size and available space
b. Construction details and materials
4. Farming implements and specialized crafts
5. Nondomestic plant and animal resources exploited
Hypothesis B: Slave material conditions and behavior patterns
ref1ect management
cally these will be
practices adapted for coastal Georgia. Archaeologi-
identified in the following household artifact
patterns:
1. Food-reated sctivi ties
a. Food preparation equipment and techniques
b. Food serving equipment
c. Domestic plant and animal food resources


232
Cate, Margaret D.
1925 Photographs. (Butler Island slave cabins, rice mill chimney,
tidal mill and boat landing, overseer's house). MDCC,
folder 206, GHS.
1960 Mistakes in Fanny Kemble's Georgia Journal. Georgia Histori
cal Quarterly 44(1):1-17.
Cate, Margaret D. and 0. Wrightman
1955 Early Days of Coastal Georgia. Gallery Press, New York.
Cathey, Cornelius
1956 Agricultural Developments in North Carolina, 1783-1820.
University of North Carolina Press, Chapel Hill.
Chance, David H.
1977 South: Method and theory in historical archaeology (a review).
Historical Archaeology 11:126-129.
Childs, St. Julien Pxavenel
1940 Malaria and Colonization in the Carolina Low Country, 1526-1696.
John Hopkins Press, Baltimore, Maryland.
Clark, J. D., H. G. Baker, and W. B, Morgan
1562 The history of sub-Saharan food crops. Journal of African
History 3:211-239.
Clifton, James
1978 Life and Labor on Argyle Island: Letters and Documents of
a Savannah River Rice Plantation-, 1833-1867. Beehive Press,
Savannah, Georgia.
Cole, Arthur H.
1927
The American
advantages.
rice growing industry: A study of comparative
Quarterly Journal of Economics 41:495-643.
Coleman,
1976
t
Kenneth
Colonial Georgia: A History.
Charles Scribner's Sons, New
York.
1977
(Editor) A \
Athens.
listory of Georgia.
University of Georgia Press,
ombes, John D.
1974 Ethnography, archaeology, and burial practices among coastal
South Carolina blacks. Conference on Historic Site Archaeciocjy
Papers 7:52-61.
Michael F.
A preliminary study of the planter aristocracy as a folk level
of study in the Old South. Unpublished H.A. Thesis. Univer
sity of North Carolina, Chapel Hill.
1941