THE STRUCTURAL AND CULTURAL CONSTRUCTION OF RACE
IN THE HANDLINE FISHING INDUSTRY ON
SOUTH AFRICA'S WESTERN CAPE COAST
JAMES F. GATES
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENT FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
For Heather-I wish she were here to share this with us.
I have many to thank and none to blame but myself for the pages that follow. This
dissertation would not have been possible without the assistance and support of many great
people. First and foremost I am deeply indebted to the fishers in this study for their trust, their
patience and their time. I can only pray that I have done their stories justice. Their indulgence
was rivaled only by their hospitality. Special thanks go to Jonathan and Gerry for their
willingness to tie me into the network of handline ski-boat skippers and the long hours of
incessant questioning they endured. My gratitude also goes out to Andy Johnstone for his depth
and breadth of knowledge and experience among artisanal fishers. I have the historian Dr. Lance
van Sittert, then at the University of Cape Town, to thank for inspiration early in my fieldwork. I
also have Dr. Marc Griffiths and the researchers at Marine and Coastal Management to thank for
supporting data and their unique perspective on the fishing industry as a whole. Special thanks to
Dr. John Sharp, then chair of anthropology at the University of Stellenbosch, for providing
support and an exciting academic community.
I owe the greatest debt for any success I may have in the writing of this dissertation to
Michelle, my wife. I am humbled by her love, support and careful attention to detail. Editor,
lover and friend, she cleared my often fuzzy thoughts, created space in our life for me to write
and put bread on the table for most of my graduate career. I could not have done this without my
parents, Chuck and Judy Gates; they have been waiting a long time for me to finally get my first
I owe my deepest personal and professional gratitude to my advisor, professor Brian M.
du Toit. He provided an unequalled depth of experience as an anthropologist and, as a South
African, added significant personal knowledge to my subject. He took far more than a
professional interest in my work, nudging carefully when I strayed down fruitless paths and
pushing even harder when needed. Even though he never did let me beat him on the tennis court,
he went far above the call of duty and took a strong personal interest in me and my family. My
heartfelt thanks go out to the other members of my doctoral committee-professors R. Hunt
Davis, Anthony Oliver-Smith and Suzanna Smith-for their patience and wisdom in guiding my
interests and my writing. My gratitude also belongs to professors Christopher McCarty and H.
Russel Bernard without whom much of the network analysis would not have been possible.
Professor Charles R. Gailey at Nazarene Theological Seminary deserves my heartfelt thanks for
getting me interested in anthropology and for his love and support along the way.
I am deeply indebted to the Pew Charitable Trust and the Pew Younger Scholar's
Fellowship committee for their financial and moral support for the research that led to this
dissertation. My gratitude also goes out to the University of Florida's Department of
Anthropology for their support.
Finally, I thank Charisa and Anthony, my children, for their patience with me during my
years of research and writing; I pray that what has been written here will one day help them to
understand their world better and inspire them to make it a better place to live.
TABLE OF CONTENTS
A C KN O W LED G M EN TS ............................................ .......................................................... iii
LIST O F TA B LES .............................................................................. ............................. viii
LIST OF FIGURES .. .......................................................................ix
A B STR A C T .................................................................................. ......................................... x
S INTRO DUCTION ...................... .. ................................................................
Reflexive Reflections: A Day in the Life of an Anthropologist .......................................
Sum m ary of C chapters ........................................ ......................................................... 1
2 THEORETICAL EXPLORATIONS ..................................................................... 14
Race and Ethnicity as Tools for Constructing the Other ............................................... 14
Anthropology and the Construction of Race and Ethnicity ...............................14
Competitive Versus Paternalistic Race Relations .............................................22
The Anthropology of Fishing .................................................. ... .......................27
Economic Anthropology and Its Application to Fishing ...................................27
Maritime Anthropology: A Review of Recent Literature .................................33
N otes .................................................................................. .........................................4 0
3 RESEARCH METHODOLOGY:
ANTHROPOLOGY AND THE STUDY OF DIFFERENCE ........................................41
Understanding the Ethnographic "Other" ......................................................................41
A M ethodological Journey .......................................... ......................... ....41
Preparing for the Trip .............................................. ...............................43
Vehicle Inspection ...................................... ... ......................45
Fruitful Paths and Rocky Roads ......................................... ..........................47
Navigating Between Scylla and Charybdis ........................................................53
A Case for Relative Objectivity .....................................................................57
Road Signs ............................................................... ................................60
Steering C lear ........................................ ................ ................................61
Analyzing Race via Social Networks .................... .. ...............................61
Research Statem ent ......................... .. ....... ........................ ....61
Constructing Race and Ethnicity .............................................. ..................63
Interpreting Social Structure ...................................... ................ ...............67
Research Questions and Hypotheses ........................... ..............................70
Inform ant Selection ........................................ ... ..... ....................... ....72
D ata C collection .............................................................. .. ........... ...............75
Name Generators ......................... .. ... ... .........................76
Name Interpreters ...................... ...... .............. ........................78
The Structured Interview ....................... ........................ ....................79
N otes ............................................... ...................................... .... ......................83
4 THE "SO-CALLED" PEOPLE:
COLOURED IDENTITY IN THE NEW SOUTH AFRICA.................................84
Identity in the Context of Oppression ................................... .......................84
Stereotypes .............................................. ............. ....................... .......................86
Race M ixing ........................................................ .......... .................................................88
O origins .................................................. .. ................. .................................................89
Politicization of Identity ............................................................................................. 93
Language, Race and Politics: Afrikaans and the Coloureds of South Africa ...............101
C conclusion ................................................................................. ...............................105
N otes .................................................................................... ..................................10 5
5 RACE, CLASS, GENDER AND SOUTH AFRICAN FISHERS ..................................107
Im age-ining C lass and Race .........................................................................................107
The Cultural Image of the 'Cape Coloured Fisherman' ....................................107
Race, Class and South Africa's Fishing Industry ............................................ 113
Handline Fishermen Construct the "True Fisherman" ......................................117
The Myth of the Colour-Blind Fisher ..............................................................126
Protesting Rights, Constructing Identity: Who is the True Fisher? ...................130
The Construction of Gender in South Africa's Handline Fishing Industry ..................135
Gender Relations in Fishing ............................... .... ..................................135
Gender and the Patterning of Social Networks .............................................. 137
Perceptions of Gender in the Skipper Fraternity ................................... ...139
6 THE HISTORICAL AND SOCIO-ECONOMIC CONTEXT
FOR HANDLINE FISHING............................................................ ......................146
A Snapshot of South Africa's Fishing Industry ..................... ......................146
Introduction ............................................. ................................................. 146
Prior to 1652 ..................................... ... ............... ............................... 152
D utch Settlem ent ................................... ....... .. ........................ ....154
British Colonization ..................................... ...............................154
The Early Decades of the Twentieth Century ....................................... ....167
Post World War II and the Beginnings of the Apartheid Era ..........................170
Recent Developments in the Management of South African Fisheries .............180
A Historiographical Footnote .......................... ......................................183
N o te s ............................................................... ....................................................... 18 5
7 CHASING A LIVING: COMMERCIAL HANDLINE SKI-BOAT FISHING
ON SOUTH AFRICA'S WESTERN CAPE COAST............................. ............ 186
Introduction ............................................ ...... ...............................................186
Race, Class and the Handline Fishermen ................................ ..... ....................... .... 188
Personal Histories ............................... .... ....... .....................188
W ork H stories .................................................................................................. 192
A D ispersed Sense of Place .......................................................................................... 197
The Mobility of the Roaming Handline Ski-Boat Fishermen .........................197
Rural vs. Suburban Fishers .............................................................................199
M embers of the Inform al Economy ....................... ............................................ .... 202
The Technology of Handline Fishing ......................... ............ .................203
T he R source ........................................................ .. ........................... 211
The Informal Marketplace ................................... ..............................214
C rew ........................................................ .................................................. 2 16
W eekenders ............................................ ............................................... 216
Langaaners ............................................. ............ .. .........................218
Managing the Linefishers ...................... ... .. .........................221
Fishing Association ................................... ..... ........................224
N otes ........................................ ...................................... ..................................2 32
8 THE IMPACT OF RACE
ON THE SOCIAL NETWORKS OF HANDLINE FISHING SKIPPERS ..................233
Introduction ............................. ........................ ..................................................233
G lobal N network Representation ............................................ ............................. .... 237
Introduction ................................................ .... ............................. 237
Global Network Properties ................................. ........................239
C liques .................................................. ... .....................................24 1
Network Centrality ........................ ..........................244
Race and the Structure of Ego Networks .............................. ..........................247
9 C O N C LU SIO N ................................................................................. ........................253
Broader Streams of Influence ...........................................................253
Implications for Marine and Coastal Management ...................................................256
G L O S SA R Y .................. ..........................................................................................................263
APPENDIX A INTERVIEW SCHEDULE FOR SKIPPERS ........................................... 265
APPENDIX B NAME TABLE ................................... ...... ......................275
APPENDIX C RESEARCH ASSISTANT JOB DESCRIPTION AND CONTRACT......276
APPENDIX D INFORMED CONSENT................................................277
APPENDIX E MARINE LIFE MENTIONED IN THIS STUDY...................................... 278
R EFER EN C ES C IT E D ........................................................................... ..............................279
BIOGRAPHICAL SKETCH ............................................... ..........................................296
LIST OF TABLES
1. Population Group by Province (Percentages)............................... ............. ................. 100
2. Percentage of South African National Languages Spoken by Province........................ 102
3. Skippers and their Alters: Percentage of Alters by Gender........................................... 138
4. Skippers and their Alters: Percentage of Kin Relations by Gender............................... 138
5. Nominal Commercial Catches (Tons Nominal Mass) by Fishery and Species ..............148
6. Nominal Commercial Catches (Rounded Mass) and Wholesale Values.......................149
7. South Africa's Imports and Exports of Fish in 1998.................................... .......150
8. Quota Distribution in Some Key Fisheries in South Africa in 1996.............................186
9. Chi Square Distribution for Previous Jobs by Race................................................. 193
10. Skipper's Race and the Jobs they Held Before Entering the Fishing Industry................ 195
11. Residence of Ski-boat Skippers Interviewed.................................. ............................. 199
12. Classification of Fishing Companies by the South African Government......................203
13. Chi Square of Skipper's Boat Ownership by Race...........................................206
14. Electronic Technology On-Board Commercial Ski-Boats............................................209
15. Sample Matrix Representing Skipper x Skipper Network .............................................238
16. Single-Link Hierarchical Clustering........................ ..... .........................243
17. Skippers with the Highest Degree Centrality ....... ......................................................244
18. Skippers with the Highest Betweenness Centrality...................................... ..........245
19. Skippers Closeness Centrality ..................................................................................246
20. Percentage of Same-Race Ties in Skipper's Social Networks ......................................250
21. Average Closeness of Ties in Skipper's Social Networks ............................................ 251
LIST OF FIGURES
1. Handline Fisherman Pulling Snoek into the Ski-Boat...................... .....................38
2. Handline Fisher Bringing a Snoek into the Boat.............................. .......................67
3. Population Group by Province (M illions).................................................... ............... 100
4. A Somerset W est Fisherman .................................................. .. ............. ..........108
5. Handline Ski-Boat Crew on their Way to the Fishing Grounds....................................119
6. Protesters March on Parliament for Local Fishers' Rights......................................131
7. A Diverse Crowd Marches in Solidarity with Local Fishers......................................... 133
8. The Southern African Snoek Thyrsites atun .................. ............... ......................... 155
9. The Snoek Seller .......................................................................... .......................156
10. Fishing Vessels in Operation in the Cape between 1880 and 1990............................... 159
11. Gantries for the Fishers of Kalk Bay, 1905....................................................................165
12 P urse Sein ing ................................. ............................................ ....................... 17 1
13. The Traditional "Chakkie" Used for Handline Fishing..............................................177
14. Education Levels by Population Group (Percentage)................................................ 188
15. Towns of Residence for Skippers in the Handline Ski-Boat Fraternity........................198
16. Fishing Vessels Operating in the Western Cape ....................... ........................204
17. A Fisherman's Hand, Protected by a "Vingerlappie," Casting a "Lood"......................210
18. Primary Habitat for Snoek Thyrsites atun.............................. ........................212
19. Historical Landings ofLinefish Species............................ ......................213
20. Hawkers Bidding on Snoek at Hout Bay................................ .......................219
21. Commercial Handline Ski-Boat Fishers Returning from a Day at Sea.........................235
22. Sociogram for Sample Skipper x Skipper Network.......................................................238
23. Multi-Dimensional Scaling Scatterplot of Proximity of Skippers in Network................239
24. A G raph and Its C liques ................................................................ ......................241
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE STRUCTURAL AND CULTURAL CONSTRUCTION OF RACE IN
THE HANDLINE FISHING INDUSTRY ON SOUTH AFRICA'S WESTERN CAPE COAST
James F. Gates
Chair: Professor Brian M. du Toit
Major Department: Anthropology
The primary aim of this dissertation is to describe how race relations are socially
constructed and historically situated through an analysis of the relationships between coloured
and white fishers in South Africa's handline fishing industry. In-depth interviews with 102
handline ski-boat skippers on the Western Cape coast serve as the core data for this analysis. My
primary aim was achieved by interpreting the images skippers use to describe themselves and
others, and analyzing of the skippers' perceived social networks. This dissertation moves beyond
common political and macro-economic analyses of race relations to in-depth exploration of the
ways individuals structure social relations by race in a historically specific context. I show how
fishing is embedded in a host of social and informal economic relations. Socioeconomic relations
off the boat directly impact social relations on the boat.
"Race" and "ethnicity" are historically specific and socially constructed categories of
group differentiation. "Race" and "ethnicity" are compared as analytical tools in the
methodological tool chest of social analysts. Specific attention is paid to the role that
anthropology played and continues to play in the use of these concepts as analytical
categories. I argue that in the South African context, both past and present, race remains the most
potent construct of social categorization. Specifically, I sketch the history of the ideology and
politics involved in the development of "coloured" as an identifiable group in South Africa.
Descriptions of the structural development of this category are balanced by the role of human
agency. I review the historical roots of those categorized as coloured, the common stereotypes
associated with being coloured, the challenging issue of race mixing, the role that intra-racial
politics played and the role that language played in the debate over coloured identity.
Anthropology as a discipline is in danger of being rent in two by the tensions between
those who practice it as a scientific enterprise designed to explain human variation and those who
practice it as an experience designed to be shared and interpreted. I explore the possibility of a
via media between the extremes of positivist nomothetic theorizing and the interpretive solipsism
of postmodern deconstruction. The verbal images and metaphors skippers use to describe
themselves and others reveal deeply rooted racial stereotypes and prejudices. An analysis of
these skippers' perceived social networks reveals to what extent these racial stereotypes and
prejudices structure perceived interaction.
Reflexive Reflections: A Day in the Life of an Anthropologist
Two-thirty in the morning and I was wide-awake, sitting in my bakkie, waiting for the
men to arrive. It wasn't supposed to be that cold in late January, but I didn't even notice it, much.
I felt conspicuous walking into the little petrol station convenience store dressed like I was. I
looked like the stereotypical "skollie." But who was going to care at that time of the morning? I
remember wondering if the shop attendant and the security guard were suspicious of why I was
there, since they kept looking in my direction while they continued their conversation. I was
dressed in a pair of dirty tekkies, some old sweatpants, two shirts and a warm jersey. They told
me to wear layers. I thought the wool knit cap I borrowed from my neighbor was probably a little
much, but I wanted to at least look the part. I later learned just how much 1 would need that cap.
I looked over at the white gumboots that were two sizes too big for me and hoped that I wouldn't
be too clumsy in them. The oilskins I borrowed from the same neighbor who lent me the wool
cap were a little snug, but they would have to do. I checked the LCD clock on my dashboard as it
slowly changed shape, letting me know that only one more minute had inched by. It was very
difficult getting up at a time when I normally had not yet gone to bed, but it was time to begin the
most important phase of what I had been preparing for the previous three years, maybe the
previous eight. I had it all planned out, written out, hypothesized, theorized and historicized. I
had attended lectures, helped to organize conferences, presented papers, passed my exams,
begged for funding, defended my ideas and made the long trip to get over there. I spent the
previous couple of months in the libraries, interviewing government officials and academicians,
driving to communities separated by more than 400 km at the extremes, spreading my name
around to those who needed to know, driving around with people I knew could introduce me to
the right people, making the contacts necessary to get the information I needed. But somehow, as
I sat in my bakkie with the engine running, trying not to become impatient, I felt totally
unprepared for what I was about to face.
Jonathan told me to meet him sometime around 2:15-2:30 AM at the Engine Garage on
the northern edge of the city. He knew that it would be another hour and a half or so to drive the
80 km up the West Coast road to get to Yzerfontein, and he wanted to be one of the first ones in
the queue. He said that Charles and Mailie would be riding up with him from the southern
suburbs, but that Balie, Talliep and Albert would be finding their own way to the Garage. He
wasn't sure if Albert was coming. Albert had only been with him for two weeks, and he
mentioned that he didn't know if he was reliable or not. I had wondered the day before why they
wouldn't just all ride up there together. I didn't know at the time that the crew lived spread out
across so many different parts of the city. Jonathan told me the day before that Albert had to
make it there all the way from Khayelitsha, known as the poorest black suburb in Cape Town. He
thought we might have to wait on him. As it turned out, Albert stayed overnight with a friend
much closer, in Jo Slovo Park. Although I didn't recognize him until Jonathan later introduced
us, Albert was the first to arrive. Charles, or "Boertjie," as they called him, lived down south of
the City in Muizenberg, a mostly white, middle-class suburb of Cape Town. He rode up with
Jonathan and Mailie because he lived close to both of them, and they had arranged to meet down
there ahead of time. Balie and Talliep arrived together; they both made their way up from a large,
sprawling complex of historically coloured suburbs usually lumped together and identified as
It was 3 in the morning and still no white bakkie pulling a six-meter ski-boat behind it. I
had my cell phone on me, and I knew that Jonathan would be carrying his, as most handline ski-
boat skippers did. But I didn't want to call just in case I had gotten the time wrong. After all, we
had just spoken the night before. But when 3:30 AM rolled around, my impatience outgrew my
sense of tact. I called Jonathan only to find out that some of the guys, he wouldn't say who, had
overslept and that he was on his way. If he wanted to go to sea for the day with a full crew, more
often than not he had to rouse them.
Jonathan was the skipper. He was the only one among the group that really spoke
English at home and with any regularity. Balie, Talliep and Mailie all spoke a variant of
Afrikaans that they referred to as "Kaapse Afrikaans." Albert spoke more Afrikaans than
English, but neither were one of the three other languages he spoke at home. The rudimentary
Zulu I had learned as a kid didn't get me very far, and Albert and I spent the rest of the time
speaking in a language that was native to neither of us. I had spent the previous couple of months
brushing up on the Afrikaans I had learned while living for most of the 1980s in what was
formerly known as the Transvaal. I learned quickly that I spoke what the fishermen termed
"suiwer" Afrikaans; they could tell I had learned my Afrikaans in the heart of white, conservative
Afrikaans country. No one really minded explaining the colloquialisms that this Yankee didn't
understand; after all, most were amazed that someone from the United States could speak
Afrikaans at all, let alone with a "boere" accent. They knew that I was the student there to learn
The diversity of Jonathan's crew and the cordial relations between them made me want to
believe what I was being told about race relations time and again by skippers in the handline
industry. They insisted that there is not, nor has there ever been, any racism at sea. Coloured and
white skippers alike would argue that the sea was the great equalizer, that the environment and
the hard worked leveled the playing field and created a meritocracy like none experienced
elsewhere in South Africa. "A fisherman is a fisherman," they repeated as if recited when they
earned their skipper's ticket. But this was my first day to go to sea, and I wasn't quite ready to
wrap up my conclusions from what I had heard thus far.
Four o'clock rolled around and everyone had finally arrived. I learned on the ride up the
West Coast road that it wasn't unusual to schedule a meeting time with the expectation that no
one would really be there until at least forty-five minutes later than scheduled. Fortunately this
time, at least, I wasn't on my way to somebody's house for dinner, somebody who expected the
same courteous "tardiness." I think my impatience was far more a sign of my excitement than my
commitment to punctuality. After all, I was about to go on my first trip out on the ocean to
interact with the very people I wanted to get to know over the coming months. I had finally
identified a group of fishers that I wanted to interview for my research, and I was on my way to
sea with one of them. I was nervous about the uncertainty of it all. I was nervous about how well
I would be accepted. I was nervous about how well I would be able to communicate. I was
nervous about asking the right questions, about making a good impression, about not sounding as
ignorant as I felt. So I decided to be honest with them about my ignorance. After all, they were
now my teachers.
We closed the canopy window as four of us crammed into the back of the bakkie with
seven pairs of boots, oilskins, lunch bags and fishing tackle baskets. I learned a lot more about
the men I was about to fish with on the way to Yzerfontein than I did the rest of the day. I had
planned to talk to the guys as much as I could throughout the day. 1 used the time we had on the
ride up to the harbor to share who I was, what I was doing and to get to know some background
information about each of them. Little did I know that I would not have much of an opportunity
to talk to any of them in any more detail the rest of the day. Much of our personal conversations
ceased once we reached the queue of boats at the harbor.
When we arrived at the harbor, I was disoriented. 1 had never been there in the dark, and
there were so many boats and people around that I didn't recognize it. I was initially
overwhelmed by the chaos. It was five-thirty in the morning, and it looked like we were at an
open-air bazaar. My first instinct was to marvel at how representative the racial demographics of
the crowd at the harbor were compared to the country as a whole. I learned later that the power
relations as they relate to race were also representative of the country as a whole, but the country
of not so long ago.
I was unaware of it at the time, but there was an important reason I was left alone as soon
as we arrived at the harbor. Each of the six crewmembers quietly got themselves dressed in
clothes very similar to the ones I had borrowed from my neighbor. As soon as they were done
getting their raingear on, they dispersed. I didn't see most of them again until we were all getting
on the boat when it was our turn to launch from the slipway. I saw Jonathan talking to some of
the other skippers but did not realize the importance of those conversations at the time. He called
me over and introduced me to some of the other skippers. Come to find out, these were some of
the skippers that Jonathan trusted the most and who trusted Jonathan the most. They were
sharing information about where they had found fish the day before and what direction they were
planning to head for the day. In the last couple of years the skippers had begun to rely on cell
phones to make these connections, but this morning Jonathan was using the oldest medium
available for this type of communication, a face-to-face conversation. The skippers were able to
share exactly where they had found fish the day before; even if they weren't well familiar with
the waters around Dassen Island, most of the commercial skippers had invested in a global
positioning system (GPS) in the past five years.
The thick fog that had slowed our traveling speed up the West Coast road hung on the
water until well after 11 o'clock in the morning. Those boats that had not invested in a GPS had
to follow someone who had. Jonathan told me there were plenty of skippers who routinely relied
on other boats to show them where the fish were biting. The better skippers called them
"holhangers" (lit. those that hang on my ass). If you were dependent on someone else to find
your fish, you weren't a very good skipper. I was told that the skippers who regularly worked on
the water had strategies for throwing the "holhangers" off their trail. Some would call out
coordinates or location on the public Very High Frequency radio (VHF) and then contact their
closer colleagues over the cell phone. Some of the skippers admitted to installing scramblers and
decoders on their VHF radio prior to the introduction of the cell phones. Catching shoaling fish
on the open ocean was a competitive environment, and the right information from the right person
gave the skipper the competitive edge.
On the ride out to sea from the slipway I quickly found out why that wool cap was so
important and why I was told to bring the extra clothing. The wind on the open sea cut through
every piece of clothing I was wearing. I wasn't told this, but I think there was another important
reason for the layers of clothing. The ride to and from the island was so rough, so bumpy that any
extra padding was more than welcome. As we bounced our way out to the island I gripped onto
the wood edging that lined the different "laaitjies" in the boat. Each commercial handline boat
was partitioned into cubicles, two for each man who fished on the boat. One of the laaitjies was
for standing in, the other for loading fish. It was important that each man have his own laaitjie so
as to keep track of each man's catch for the day. As they were all paid on the share system, i.e.,
for 50% of what they caught for the day, each crewmember wanted to make sure the skipper
could verify his particular totals.
The division of labor on the boat started even before we left the Engine Garage back in
Cape Town. Jonathan and Boertjie fueled the bakkies and the boat and planned their strategy for
the day while the others loaded the bait and the gear onto the boat. This was the first time I was
acutely aware of race relations on the boat. I was to find out later that Albert was one of the
exceptions among handline crew. Very few of the crew were identified as black. Most were
identified as either white or coloured. I wondered why the only white guy on the boat seemed to
be the skipper's right hand man. I was later to find out that it is quite odd for any white fishermen
to crew for a coloured skipper, particularly a coloured skipper that owned his own boat. Boertjie
was Jonathan's right-hand man and had been fishing with him for almost two years. He was
twenty-five years-old at the time, had a high school diploma and recently got out of the Navy.
Boertjie said he was fishing because he couldn't find work, an ironic statement given his sheer
physical output at the time. He claimed that he couldn't find a "real job" because he was now
disadvantaged by affirmative action in the new South Africa. Earlier one of the coloured
crewmembers from another boat jokingly jabbed at his white colleagues: "Ja, nou'sjulle die
kaffers, en ons die baase" ("Yeah, now you're the Kaffirs, and we're the bosses").
I was constantly aware of my own race and my status as a foreigner and wondered how
that would impact how people acted around me or responded to me. 1 found out later that I had
been very fortunate to be introduced to the network of skippers by Jonathan. Jonathan turned out
to be my closest and most reliable contact in the network. Jonathan was one of the most well-
respected and well-known skippers in the fraternity. He was well connected because he was
connected to a number of key persons from each of the main regional subgroups in the network.
Jonathan learned early on that I was very interested in asking tough questions about race and its
impact on the fishing industry. The fact that Jonathan trusted me meant that 1 had access to his
trusted network. Since he was well connected between both coloured and white skippers, he was
able to introduce me to subgroups that would otherwise have been difficult to contact.
The fact that I am white most likely influenced the kind and degree of responses I
received from each of the skippers I interviewed. More than a few times the racialized language
of the white skippers was so matter-of-fact that it was as if they assumed I understood the way it
worked, as if I should understand, as if it was common sense. I'm not sure if the white skippers
would have been so unguarded in their racialized language had I not been white and had I not
spoken Afrikaans in such a "suiwer" way. I noticed a similar assumption of familiarity when any
of the skippers, white or coloured, spoke of the women in their lives. Because I am a man, it was
assumed that I understood. Often when I asked them to explain what they meant or how things
work, a certain level of impatience was not uncommon. A number of times I was sharply
reminded of my race when a coloured skipper would politely pause for a caveat apology after
talking about "those whites."
When we finally laid anchor after what seemed like an eternity of pounding on the hull of
the boat, 1 had no idea we were only three soccer fields away from Dassen Island. The fog still
masked the area we were fishing. As we laid anchor the men baited their hooks quickly with the
sardines from the boxes they opened. Jonathan brought some of the "good stuff' along, the pike,
but they wouldn't use it unless the fish weren't biting well. Each of the guys took out the spool
with the thickness of gut that they felt would be most appropriate for how the fish were biting that
day. Before preparing their lines, each fisher put neoprene tubes, called finger "lappies," over
their fingers for protection against cuts from the lines. They each threaded hooks for bait onto
two lines, and what they called a "lood" or a stainless steel spinner on a third line. For the rest of
the day's fishing each man would tend three lines, baiting and throwing them back as fast as they
reeled them in. Jonathan let them know, from the information on his electronic fish-finder,
approximately how deep the fish were swimming and each man measured out his own estimation
of the fathom depth. Each fisher guesstimatedd" a fathom as the length of his own wingspan.
Everyone was quiet on the boat as we were heading out to sea. The noise from the twin
85 h.p. Yamaha outboard motors and the crashing hull made conversation impossible. Of course,
feeling the need to hold on for dear life did little to encourage intimate sharing. In fact, the
conversation on the boat for the rest of the day shifted away from the personal type of
conversation I was able to have on the ride to the harbor. The atmosphere on the boat shifted into
a high energy, jovial gear. Balie, the crew who stood closest to me once we started fishing, began
singing a risqu6 song about a woman he longed to be with. Much of the language on the boat was
what the skippers would later describe as "inappropriate for female company." Many of them
called it "vissermanstaal" (lit. fisherman's language) or simply "vloektaal" (lit. profane
language). It was understood that men could talk like that when they were on the water, since
rarely did anyone see a woman out on the sea. The language was particularly aggressive when
any boat passed within a few boat-lengths of another. Crews took offense when another skipper
"charged down" on their fish.
On the way out Jonathan had me stand with him by the console. When we anchored he
had me stand in the foremost laaitjie that he said no one else was using. Come to find out I had
taken the potential place of another crewmember for the day. Had 1 not come along for the day,
Jonathan would have picked up an extra crew from the many that were waiting for a "site" on the
harbor. At Yzerfontein, in contrast to the slipway at the Cape Town powerboat club, there tended
to be a surplus of"paloepas" available for skippers who needed crew. A guy was a "paloepa"
when he was seeking a site. A crew that bounced around too often from boat to boat received a
bad reputation as a guy with "rubber boots" and would reportedly be less likely to get a site in the
By taking one crew member's spot on the boat I not only took away the income for one
of the potential crew on the harbor but I reduced Jonathan's profits in the loss of the absent
crew's share. Between the constant activity on the boat, the lack of privacy in the context of a
boat full of listeners, the economic cost of an extra, non-fishing body, it turned out that the boat
was not the most appropriate place for doing interviews with each of the skippers. After about an
hour of taking notes and attempting conversation I decided to try handlining for snoek. I was
given a couple of"vingerlappies" and warned against the razor-sharp teeth for which the snoek
are famous. Balie, my teacher, would not let me take the first ten snoek off the hook for fear that
I would get all cut up. They could tell by my hands that I was not a fisher-that and the fact that
I probably left more behind in the water than I took out of it, both in terms of bait and the food
that I tried to eat all morning.
1 was told you can tell when you are in an area of high snoek concentration when you're
being bothered by the seals. The fishermen were constantly cussing at the seals on that day. We
remained anchored the whole morning, a decision that turned out to be the right one for that day.
By the time we headed back to the harbor we had over 400 snoek on the boat. I had caught 15.
The four main schools of snoek that swam under our boat caused a frenzy of activity. Lines got
tangled, tempers flared, but never were the hooks out of the water for more than a few seconds.
We could see other boats drifting in and out of our range of vision through the fog; they were also
catching. It was when the fog lifted sometime soon after 11 AM that I realized just how close we
were to the island and how many ski-boats were out chasing a living. We were some 300-400
yards away from the island, and it seemed like you could walk from one boat to the next to reach
the shore. I checked the slipway records later on and found out that there were 84 boats that
launched from Yzerfontein's single slipway that day. Yzerfontein's record for a single day
during the 1999-2000 season was 143 combined recreational and commercial ski-boats.
The ride back to the harbor was smoother than the ride out. We were loaded down with a
ton offish and were plowing our way back to the slipway. We passed a group of penguins, a pair
of dolphins and a whale in the distance on our trip back. I was too tired at the time to get excited.
Little did I know, but we had a full two hours of work ahead of us.
As Boertjie pulled the boat out of the water, Jonathan polled those on the docks for the
going price down at the fish market. By knowing the going price he could make sure he was not
underbid. At Yzerfontein the private powerboat club grounds are used for selling the fish and
cleaning the boats. Informal hawkers and fish shop owners come from as far as the southernmost
suburbs of Cape Town, tipped off by some of the skippers that there has been a good catch of
snoek for the day. Fish are sold to the highest bidder willing to take the entire load for that day.
Jonathan did the bidding; sometimes the skipper designates one of the crew for that task. Once
the fish are sold and offloaded, the crew cleans the boat and the equipment while the skipper
finalizes the transaction. We weren't finished cleaning the boat until close to 3 PM.
I arrived home more exhausted than I think I had ever been. I was impressed that these
guys would travel all that distance, return home, just to do the same thing the next morning. The
fish were biting, and they knew they had to take advantage. There were plenty of times when the
fish weren't biting, and they would have to make do. The work was physically demanding. The
hours were long and often monotonous. I was determined to explore the relationships that kept
these guys going. More than ever I wanted to understand how relationships in the handline
fishing industry were structured and constructed. I wanted to understand why race seemed to be
so central to the social networks of these roaming handline ski-boat skippers.
Summary of Chapters
Following the Introduction, Chapter 2 outlines the theoretical and disciplinary context for
the analysis that follows. "Race" and "ethnicity" are defined as historically specific and socially
constructed categories of group differentiation. "Race" and "ethnicity" are compared as
analytical tools in the methodological tool chest of social analysts. Specific attention is paid to
the role that anthropology as a developing discipline played in the use of these concepts as
analytical categories. As an example of the power that these categories have to influence society,
I proceed with a discussion of the difference between Competitive and Paternalistic Race
Relations. Chapter 2 continues by placing this study in the context of economic anthropology. I
particularly focus on the role that anthropology has played in the understanding of economic
activity as embedded in social context. The chapter ends with a review of the recent literature on
social relations in maritime communities.
Chapter 3 discusses the methodological concerns at the heart of this dissertation,
beginning with a discussion of the historical development of methodology in anthropology. This
chapter explores the possibility of a via media between the extremes of positivist nomothetic
theorizing and the interpretive solipsism of postmodern deconstruction. I analyze the strengths
and weaknesses inherent in various approaches and make a case for selective use of methods
from various camps. The chapter continues by outlining the major research questions to be
explored and hypotheses to be tested. Specific detail is given on social network analysis and its
use in understanding the construction of race among commercial handline fishers in South Africa.
Informant selection, data collection and the structured interview used are discussed in detail.
Chapter 4 is an historical analysis of the ideology and politics involved in the
development of coloured as an identifiable group in South Africa. Descriptions of the structural
development of this category are balanced by the role of human agency. I specifically focus on
the development of racial identity in the context of political oppression. I review the historical
roots of those categorized as coloured, the common stereotypes associated with being coloured,
the challenging issue of race mixing, and the role that internal politics played in the construction
of modern coloured identity. The chapter concludes with a look at Afrikaans as a central part of
the debate over coloured identity.
Chapter 5 builds on the foundational assumption that race, class and gender are socially
constructed categories that influence thought and behavior in historically and culturally specific
ways. This chapter explores how the issues of race, class and gender merge in the construction of
the "Cape Coloured fisherman." The influence of race and class in South Africa's fishing
industry provides the socioeconomic and political context for this construction. An analysis of
how handline fishermen construct the "true fisherman" and the "myth of the colour-blind fisher"
show how race, class and gender are woven into the fabric of human experience. An
ethnographic description of a fishers' protest march on parliament provides a transition to the
important discussion of the construction of gender in South Africa's handline fishing industry. A
general discussion on gender relations in fishing precedes an analysis of gender in the patterning
of the social networks and perceptions of skippers on the handline ski-boats.
Chapter 6 provides the historical and socio-economic context for handline fishing. After
a brief introduction to the modem South African fishing industry, I trace a selected history of
events important to understanding the modern handline industry. This historical sketch begins
with what is known of fishing in South Africa prior to 1652, proceeds through its lack of
development during Dutch settlement, the easing of restrictions on fishing with British
colonization and the birth of the modern industrial fisheries in the early decades of the twentieth
century. The sketch continues with post-World War II developments and the beginnings of the
apartheid era as they marked the significant beginnings of the ski-boat industry. An analysis of
recent developments in the management of South African fisheries provides the political context
in which handline fishers operate. The chapter ends with a discussion of the limitations of the
historical sources available for the handline fishing industry.
Chapter 7 is an ethnographic description of the roaming commercial handline ski-boat
fishers on South Africa's Western Cape coast. Following a brief introduction, an analysis of their
personal histories provides a sense of who these skippers are and what brought them to the
handline industry. The mobility of the roaming handline ski-boat fishermen is discussed with a
focus on how this mobility influences the skipper's sense of place. The second half of the chapter
deals with handline fishing as embedded in informal economic relations. Variations in the
technology of handline fishing and the fish sought after are described in socioeconomic context.
The chapter continues with an analysis of the important socioeconomic relations off the boat that
directly affect social relations on the boat. The chapter concludes with a description of the
influence that government fisheries management has on the linefishers.
Chapter 8 deals with the impact of race on the social networks of handline fishing
skippers. The theory behind social network analysis as structural analysis is applied to the case of
the roaming handline ski-boat skippers. Both global and ego-network representations are
explored. Specific attention is given to the cliques that emerged and reasons for the centrality of
particular individuals in the group. The chapter concludes with an analysis of race and the
structure of ego networks. Chapter 9 consists of concluding remarks that point to the future
potential of this research.
Race and Ethnicity as Tools for Constructing the Other
Anthropology and the Construction of Race and Ethnicity
Prior to the twentieth century many terms for the categorical variation among humans
were conflated. Race, ethnic group, nation, culture and people were commonly used
interchangeably as meta-categorical labels on human difference. Even though the term ethnic has
been traced at least back to Victorian England, the term ethnicity did not make it into English
dictionaries until the 1950s (Hutchinson and Smith 1996). Well into the twenty-first century, race
and ethnicity have often been used as synonymous labels to distinguish between the broadest
possible categories of human variation. A historical glimpse of how race and ethnicity came to
label these categories provides the broadest base possible from which to understand race and
ethnicity as tools in the construction of reality. Understanding the historical variations in the use
of the concept helps to understand the many permutations of the concept read in contemporary
literature. As Karl Mannheim said (1936), all ideas have an address. Knowing the address of
concepts as broad and diversely used as "race" and "ethnicity" helps provide a roadmap for their
While ancient conceptions of human difference were not labeled "race" or "ethnicity,"
the sense of kinship, group solidarity and common culture that these terms refer to is older than
the written record. Historians place the development of ethnicity in the context of urbanization,
nomadic conquest and enslavement, long-distance trade, endemic disease and the need to
replenish urban labor forces. Long distance trade, for example, led to permanent communities of
aliens in major urban centers. "These trade and skill diasporas, like ancient slavery, attained legal
definition from very early times, as the rights of merchants prescribed by the laws of Hammurabi
show" (McNeill 1996:107).
The rise of universal religions also played a homogenizing role in macro-group
Beginning about 500 B.C., the rise of portable and universal religions-i.e., Buddhism,
Judaism, Christianity, and Islam together with some less successful faiths like
Manicheanism-provided an effective cultural carapace for trade diasporas, insulating
them from their surroundings in matters of faith and family as never before. Portable and
universal faiths, in fact, permitted followers of a religion that differed from that
prevailing in the environing society to maintain a corporate identity indefinitely,
generation after generation. (McNeill 1996:107-108)
Territory, religion and language heavily influenced group identity further away from the
Middle Eastern center of Western civilization. In the far east, for example, the development of
relatively homogeneous people groups varied. While the relative homogeneity of Japanese ethnic
identity tended to be established from the beginning of the written record, China did not exhibit a
similar isolation. Except for the ancient gulf between the Ainu and other residents of the Islands
of the Rising Sun, "whatever ethnic diversity initially existed among those ancestors disappeared
before the historic records begin" (McNeill 1996:108). While there was a sense in which Chinese
civilization maintained a greater cohesion than was true of European, western Asian and Indian
civilizations, the imperial aspiration of the Han dynasty, for example, ensured contact with groups
that are familiar to Westerners as Turks, Tibetans, Mongols, Manchus and Koreans.
Modern notions of ethnicity have their most immediate roots in the rise of the nation
state. John Armstrong emphasized the essential distinction between sharp identities maintained
for centuries and those that have developed as a result of the more recent diffusion of nationalist
ideology (Armstrong 1996). He highlighted the development of ethnic diasporas. Jews and
Armenians are particularly well-known examples of relatively homogenous communities who
have settled in trading enclaves after being forced to leave their homelands for economic or
political reasons. Decentralized religious organizations, sacral language, myths, texts and
liturgies played an important part in socialization for these groups. "The intensity of identity
produced by older sacral myths-based on but not always coextensive with distinctive
religions-has never been exceeded by modem secular myths" (Armstrong 1996:121).
The term "ethnos" comes from the ancient Greek term for distinguishing separate people
groups, particularly those who were not Hellenic. At the time Herodotus was writing, while the
Greek peninsula and the islands surrounding it were solidly Greek in identity, there were small
communities tied deeply to that identity scattered throughout Asia minor, Sicily and the alien
worlds of geographically proximate foreign powers. Except when forced to do otherwise by
foreign powers, each of these scattered communities had their own government, coinage, calendar
and laws, and its own temples and cult, most often tied to the traditions of the Hellenic state.
Greeks living in the heartland or in the periphery had a common label for all other peoples:
barbaroi. Calling other groups barbarians "was a clear signal of the qualitative differentiation,
commonly but not always with a pejorative implication" (Finley 1996:112). Ethnos was a term
used to distinguish between the various politically organized barbaroi.
Through the Middle Ages right down to the late eighteenth century, educated Europeans
accepted without question that the universe was organized as a Great Chain of Being. This meant
that everything in the universe, including people groups, was organized into an immense number
of hierarchical links, beginning with the most insignificant and rising to divine perfection itself.
The notion of the Great Chain of Being shines through in Dante's Divine Comedy as a general
moral hierarchy and was applied to all aspects of life, including the classification of nations.
Belief in the Great Chain of Being lead in part to the hierarchical and paternalistic world of the
Elizabethan England as revealed in Shakespeare's MacBeth. It was only the power of the
scientific revolution in the methodology and discoveries of Bacon, Galileo and Newton and the
dualistic and/or materialistic philosophies of Descartes, Locke and Hobbes that brought about the
eventual collapse of this hierarchical metaphysical outlook.
Corrosion in the concept of the Great Chain of Being did not necessarily lead to the
evaporation of hierarchical thinking. New ways to conceptualize difference were developed in
modernity that looked strangely like the old, except this time it was the canons of science rather
than the canons of the church that dictated reality. With the new scientific paradigm,'
conceptions of human difference in language and customs were tied to the idea of human
progress. People groups would still be ranked with European cultures as the ideal.
With the rise of the social sciences and other sub-categories of the academy in the
nineteenth century, classification systems and nomenclature became more specific, more precise.
But the early ancestors of anthropology themselves did not make a careful distinction between
race and ethnicity when describing general categories of human variation. We see, for example,
in L. H. Morgan's Ancient Society (1871) a ranking of human cultures (with his own at the top,
of course) but no explicit discussion of the relationship between race and ethnicity. Herbert
Spencer's social evolutionary approach in the late nineteenth century conflated biological and
cultural variation, applying Darwin's models of biological selection to cultural variation. Early
physical anthropologists like Hrdlicke promoted the classification and ranking of people groups
based on phenotypic differences. At the beginning of the twentieth century, social
anthropologists studying South Africa were categorizing and ranking groups based on physical
types. The San, for example, were ranked low on the evolutionary scale because of their
"pedomorphic" facial features and protruding "steatopygia" (see Saul Dubow's Illicit Union:
Scientific Racism in South Africa (1995)).
The earliest discussions about the relationship between race and ethnicity centered
around the relationship between the cultural and the biological aspects of human variation. In the
early part of the twentieth century, Boas challenged the association between biology and culture,
pointing to the need to understand individual cultures in their respective historical contexts. One
of Boas' main concerns was to address the overt racism inherent in much of the social analysis of
his contemporaries. Although he was largely successful in pulling the anthropological
community in his direction, he, too, did not carefully distinguish between ethnicity and race as
analytical categories. This is particularly surprising as one of his contemporaries, whom he heard
lecture and whose writing can be seen reflected in Boas' own, was perhaps the most explicit
about race as the central marker of difference. In 1903 W.E.B. Du Bois prophetically claimed
that the color line would be the problem of the twentieth century. A phrase like "color line"
clearly expresses an understanding that phenotype is somehow involved as the major marker of
difference, but it does not tell us how nor to what extent.
At the end of the second World War the analysis of race had escalated as a politically
charged issue. The horrors resulting from the racial categorization of the Nazi eugenicists
convinced social scientists to shy away from racial theorization. In 1941 a prominent Harvard
anthropologist, Ashley Montagu, published a work that later typified anthropological dogma for
more than halfa century. In Race: Man's Most Dangerous Myth (1974) Montagu successfully
argues against the necessary link between the biological and the cultural, with specific critique
directed against biological determinism. It was his leadership that led to a post-war UNESCO
statement arguing for the use of "ethnic group" as a concept more heuristic for analysis than race.
For the researchers of the time, ethnic group carried less historical baggage and could be used to
discuss cultural variation without specific reference to phenotype.
For political conservatives, as Manning Marable has pointed out in his discussions of
affirmative action, a shift toward the language of ethnicity and, consequently, the removal of the
language of race as a key component in understanding social inequity allowed for the
conversation to shift away from the historical connection between race and oppression (1996).
He believes that this shift was in part responsible for the current attack on race-based affirmative
action in the US because of how it shifted the focus from compensation for historical wrongs to
an emphasis on diversity for diversity's sake. Many more groups could now be the beneficiaries
of a set of programs originally intended to primarily benefit racially oppressed groups. William
Darity's work clearly shows that, even with the affirmative action programs of the last thirty
years in the US, African-Americans in general have still not been given a level playing field.
South Africans have only just begun to struggle with the need to address reparations and the
appropriate mechanisms to carry out such justice.
Anthropologists who did not want to see discussions of biological variation pushed to the
margins of anthropological discourse began using the fluid concepts of genetic variation,
preferring to speak of dines and probability estimates rather than static categories. They were
beginning to build on work done by the human biologists that confirmed that there was more
variation within than between generally defined "racial" groups.
In her Annual Review of Anthropology article on the status of race as a category of
analysis in anthropology, Faye Harrison (1995) argues that most anthropologists from the mid-
twentieth century on moved to a "no-race" policy. This was effectively, a "no discussion of race"
policy. What was helpful in this period was the growth of the sophistication of research on the
social construction of human categories of variation under the auspices of "ethnic studies." What
was not helpful, as confirmed by Mukhopadhyay and Moses (1997) and the authors in the
February 1999 issue of the American Anthropologist, is that the singular focus on ethnicity-based
principles of classification and organization could not adequately explain the persistent power of
racism and its impact on those affected by it. Their work gives evidence of the dangerous decline
in the usage of race as a valuable concept in anthropological textbooks and other works. To
ignore race as an important marker of social difference is to deny the experience of those most
oppressed by it.
Many current works have moved beyond the identification of race with biological or
phenotypic types to a social-constructionist perspective. Gregory and Sanjek's (1994) work
explores race as a socially constructed and contested domain. Gert Oostindie and colleagues built
on Harry Hoetink's concept of racial distance in describing racial and ethnic relations in the
Caribbean, describing a continuum of racial experiences (Oostindie and Hoetink 1996). Kevin
Yelvington (1993; 1998) shows the complex relationship between race, class and gender,
particularly for working women in Trinidad. Edmund Gordon (1998) shows how confusing the
analytical differences between race and ethnicity can become, using the Creoles of Nicaragua as
examples. The evidence from each of these works shows that to ignore race, to ignore blackness,
for example, or even shades of blackness, is to ignore the core identity for the groups studied. To
ignore race is to ignore the historical experiences of those most affected by race as an organizing
As David Roediger (1991), Noel Ignatiev (1995), Karen Brodkin (1998) and Ruth
Frankenberg (1993) have shown us recently, it is just as important to historicize the construction
of what it means to be white. If George Fredrickson (1981, 1995) is correct, and I think he is,
South Africa's current racial categories must be understood in the politicized light of the struggle
between white supremacy and black liberation. In discussing coloured identity in South Africa it
is then that much more important to understand the categories against which colouredness is
In describing the racial and ethnic relations in Colombia, Peter Wade tells us of the
pressure for blanquemiento (whitening) in a culture that promotes itself as a country of mestizos
(mixed race) (1995). Wade's work also shows the danger of excluding race in our understanding
of social difference. In this case the dominant acceptance of mixed race as the ideal has
marginalized those stereotyped as the bottom of the scale, i.e., black Colombians. He shows that,
while there is variation in the experiences of black Columbians of different classes, the ladder of
social mobility is more difficult to climb because of their race.
In her study of Cape Verdean immigrants to the US, Marilyn Halter shows just how
complex the relationship between race and ethnicity can get. She chooses the term racial-ethnic
group to describe a people who do not fit neatly into the prevailing definitions. In Between Race
and Ethnicity (1993) she shows how difficult it has been for some Cape Verdean immigrants to
be incorporated into the dichotomous black-white category of US racial thinking. Unfortunately,
her new term leaves us with no less opaque a concept for analytical purposes.
In exploring systems and structures of social inequality in South Africa, the dominant
contemporary tendency is to substitute the language of ethnicity for any discussion of race.
Racial rhetoric is so politically sensitive in South Africa that most have simply avoided its use. In
a recent study done on the identity of those historically called Coloured, Brigit Pickel (1997)2
avoided any discussion whatsoever of race in either her historical or theoretical analysis.
However, she implicitly incorporates apartheid racial (not ethnic) categories in her questionnaires
and analysis. While the language of ethnicity may be more politically acceptable, it is less than
historically accurate. The dominant categories of human variation in South Africa remain
predominantly racially constructed.
Currently, there is much debate over the use of the term coloured as a designation for any
particular group of people in South Africa (Adhikari 1994, Pickel 1997, Morris 1992, James and
Caliguire 1996, Du Pre 1994). The well-known anthropologist Micheal Whisson (1971) was
arguing in the early 1970s that coloureds do not even constitute a group but a residual category of
persons whose sole common feature is negatively defined. Despite the analytical and political
problems with this category, its common use in government statistics, the press, television, radio
and private conversations reveals its persistence as a marker of difference. With this in mind, I
have chosen to retained the term "coloured" as a referent to a heterogeneous and ill-defined group
of South Africans whose experiences differ significantly from other South Africans. Even though
there may be more intra-category than inter-category variation, to say that all citizens should
simply be called South Africans is to deny the historically peculiar situations people of various
sub-groups have experienced and continue to experience. Particularly in the rural fishing
communities of the Western Cape coast, the ascription of coloured identity, both by self and
other, is common. Not only is it common, but for many it is assumed to be natural. I use
coloured here with a lower case "c" to distance myself from the politically ascriptive use of
"Coloured" under apartheid3. I retain the South African spelling to distinguish between this
group in South Africa and a large body of literature on the "people of color" in the US and other
parts of the world. In short, I see coloured as primarily a racial referent with ethnic subdivisions.
Broad categories of human variation, like race, ethnicity, class and gender function best
when understood as historically conditioned, culturally contextualized categories. As descriptive
categories they are always open to change, dependent on the historical variation of the groups
under discussion. For those who want to use these categories for describing the causes of present
or past behavior or as prescriptive tools for predicting future behavior, they must consider the
variations in all of the dimensions that these broad categories encompass. Given that diversity
exists and that named groups exist, the focus should shift beyond a debate over the labeling of
such difference to exploring the content of such difference. To such an end, any useful definition
of race as a marker of social difference should be broken down into at least the following
dimensions: 1) what is the historical depth under consideration, 2) what is the spatial scope of the
distinctions, 3) how discrete need the variables be before we can consider one group different
from another, 4) for how long does a group have to exist before it can legitimately be called a
racial, 5) how much movement is there of people between the categories, 6) is it possible to avoid
evaluating one group as superior or inferior if such a classification is made, 7) what are the
relations of power implied by the distinctions being made and 8) is there more variation between
or within the defined categories?
Competitive Versus Paternalistic Race Relations
"Race" has always been used as a characteristic in the competition between "us" and
"them." Although Du Bois is correct in prophetically stating in 1903 that the color line is the
problem of the twentieth century, the color line has divided societies since the beginning of
recorded history. St. Claire Drake's work has shown how race has been an organizing principle
for societies throughout history, from the Babylonian, Syrian, Persian and Egyptian empires,
through the Greek and Roman empires into the middle ages and beyond (1987). A keen example
of how race was used to justify oppression in the Middle Ages is found with the Christian
Crusades at the turn of the first millennium. Although popularly justified as acts of religious
conversion, the history records have shown that political and economic dominance was at the
core of what motivated the crusades (Latourette 1975). Racial ideology was an integral part of
the crusader's rallying cry: "Death to the Moors," a common battle cry. The conception of
difference was specifically associated with phenotypic distinctions.
Pierre van den Berghe (1978) has spoken of a transition between what he calls
paternalistic systems of race relations to competitive systems of race relations. A system of
paternalistic race relations exists when one socially defined racial group positions themselves as
somehow superior to other socially defined racial groups. The group that perceives themselves as
superior uses race as a primary characteristic to define who the "other" group is. Attitudes do not
become a system until they have been fleshed out in ideological and material constructs in
defense of prevailing racial attitudes. While race was an integral part of the way people
constructed the world prior to the middle ages, the best example we have where race relations
were paternalistically systematized is the advent of European colonialism.
The transition van den Berghe refers to between paternalistic and competitive systems is
a relatively recent transition. It is not until the twentieth century that race relations have been, in
the majority of cases, anything but paternalistic. Paternalism, the way van den Berghe and other
contemporary writers use the term, intensifies and is systematized in context of European
modernization and expansionist rhetoric. A paternalistic system of race relations grew out of the
combination of Western confidence in the idea of progress and the imperial (colonial) expansion
into new worlds. If Cartesian doubt can be credited with starting the era of modern scientific
thought, then Newtonian physics must be credited with providing the impetus for an ideology of
progress that is still with us. The Enlightenment brought with it a confidence in human evolution,
both personal and social. With the industrial revolution, confidence ran high that Western
civilization was more advanced than other societies have ever been. Technological sophistication
took on a universalizing evaluative quality. People groups were classified and ranked according
to European notions of the ideal.
Some of the earliest modern social analysis shows how race was swept along the growing
stream of modernization theories. in the eighteenth century the social theorist Jean-Jacques
Rousseau argued for a ranking of the races and, while implicitly assuming his culture to be the
ideal, made a case for the admirability of other cultures, the "noble savage." The evidence for the
paternalistic categorization of the races was perhaps nowhere better on display than at the 1893
World Fair, that international celebration of Western civilization and technology. Here the
"savages" were put on display with the same glamour as the latest technology.
It was not until well into the nineteenth century that writers were starting to distinguish
between race and culture or ethnic group or nation or people group in general. The paternalism of
Western nations is evident from much of the literature on nineteenth and twentieth century
Colonialism. Crawford Young (1994) and Mahmood Mamdani (1996) show the surviving effects
of European colonialism on modem Africa. Africans were not effective in throwing off the
shackles of paternalistic colonial systems until well into the twentieth century.
To state the obvious, competitive race relations can only exist when there is the
possibility for competition. One group cannot so dominate the other that there is no room for
competition. When one group retains material and ideological hegemony over another (see
Godelier 1986), there is little possibility for competitive race relations. Competition is squelched
by the same state apparatus that supports the inequitable distribution of resources.
In From Savage to Negro (1998) Lee D. Baker echoes the transition to which van den
Berghe refers. Here the author traces the role Anthropology played through the 1950s in the
construction of Western conceptions of the "other" or, more specifically, those of African
ancestry. Part of what allows for competitive relations to emerge is the valuation of the life and
experiences of those defined as "other," where the other is no longer outside the dominant
conversation of what is to be valued. Conversation here refers to more than ideological or verbal
communication. Making the "other" part of the conversation also means that institutions are put
in place (or removed) in order to effectively ensure the possibility for competition. In the case of
European colonialism in Africa, the adoption of independent African governments was the first
major political step in creating the context for competitive relations to exist.
It is much easier to provide examples of paternalistic race relations than competitive race
relations. However, a few examples of relatively competitive race relations can be raised.
Consider, for example, the way in which racial categories have been institutionalized in the
census categories of countries like the United States, Brazil and South Africa. As Anthony Marx
shows (1998), these three countries have relatively recently provided the institutional
infrastructure that makes race-based competition for resources possible. Robert Price (1997)
argues that South Africa has virtually guaranteed a continued focus on the racial categories of the
past by instituting redistribution programs that are intended to compensate via the same hardened
categories. Increased competition is guaranteed by the hardening of these categories. Kevin
Yelvington (1995) gives a good example of how the competitive nature of race relations in
Trinidad is further complicated by variations in class and gender among those defined in each
If we follow Timothy Keegan's work on the effect of colonial expansion on South
Africa's racial paradigms, van den Berghe's schema fits. Keegan (1996) shows how the racial
paradigm developed over the course of Dutch and British (and later Afrikaner) colonial
expansion. He argues that some of the harsher racial categorization and paternalistic systems of
racial relations came from the British, not the Boers. The prevailing thought of mid-twentieth
century historians of South Africa was that the Cape Liberal tradition provided room for
competitive race relations to develop. Since the Cape Liberal tradition espoused equal rights for
all true citizens, it was theoretically possible for any person in the Cape to gain access to political
power. After all, the constitution of 1853 did not exclude de jure non-whites from voting.
However, there were property requirements attached to the franchise, requirements that made it
extremely unlikely that too many Khoi, San, Xhosa or any other non-white could participate. By
the 1870s the property requirements were strengthened and by the 1890s race became an
exclusionary qualification. Wilmot James and Mary Simmons (1991) also outline how race
became paternalistically institutionalized during the history of the western Cape.
But there is little doubt that, when the National Party came to power in 1948 and the
whole apartheid edifice was erected, any possibility for competitive race relations was virtually
wiped out. An entire political and economic infrastructure ensured that the transition to
competitive race relations would not happen. The most restrictive of these actions had to have
been the series of Acts that were passed around 1950 that formed the foundation for the apartheid
system: the Population Registration Act, the Group Areas Act, the Immorality and Indecency
clause and later, the Prohibition of Mixed Marriages Act.
It was not until the late 1980s that the true possibility for competitive race relations in
South Africa became a reality. This is not to say that many or even most South Africans did not
carve competitive space for themselves in an extremely restrictive environment, but the
infrastructure for true mass competition was not there. In the late 1980s the structures of
apartheid were collapsing, the unions that fought so hard in the previous decades had made
significant strides and the educational institutions were opening their doors to begin making true
It is difficult to make a judgement regarding the applicability of van den Berghe's
analysis to South Africa's history outside of the context of European expansion. If we consider
some of the new work being done on the Zulu expansion (Gump 1994), it is difficult to argue for
its applicability, although for different reasons. Senzagakona certainly developed hegemony over
a vast area of South Africa via his impi army, but I am not sure we can describe the rhetoric that
must have been used as racial, if our category by definition includes a phenotypic dimension.
An interesting example of how restricted the competitive nature of race relations could be
for at least one black South African sharecropper throughout the first eight decades of this
century is found in Charles van Onselen's work (1996). Van Onselen shows how Kas Maine
negotiates a space for himself and his family in the midst of a century of changing race relations.
He never fully surrendered his right to compete with the white farmers. This was expressed in the
central metaphor of the book: The plow may be his, the oxen and the shares may be his, and the
seed may be his, but the land is not.
There is no doubt that South Africa is now transitioning to a more competitive system of
race relations. A major part of making this transition a reality will be the attention given to the
redistribution of resources though programs like the Reconstruction and Development
Programme and other affirmative action initiatives. While the ties that bind its citizens to the
paternalistic system of race relations are far from severed, it is slowly but surely being
transformed. What remains to be seen, however, is if the news ways of constructing social
relations in South Africa will continue to rely on old systems of classification. To date, these old
systems of classification remain socially, economically and politically significant.
The Anthropology of Fishing
Economic Anthropology and Its Application to Fishing
According to Stuart Plattner, economic anthropology is the study of economic institutions
and behavior done in anthropological places in an ethnographic style (1989:1). "Anthropological
places" is a concept increasingly difficult to define. Historically, anthropology has left the study
of technologically advanced state societies to their sister sociologists. A discipline born the child
of imperialism and colonial expansion, anthropology's great quest has been to understand how
lifestyles of the remotest "other" relate to what is more commonly known and done in our own
society. Ultimately, the anthropologist is interested in comparing the range of human experience
over the course of human history, making anthropology perhaps the most ambitious discipline.
What makes anthropology specifically ethnographic is the focus on research methods (like
participant observation and in-depth personal interviews) that target individual levels of thought
and experience, while at the same time attempting to connect these thoughts and experiences to
the broadest possible categories of human experience.
Anthropology was once in the salvage business, attempting to save primitive societies
before they were destroyed by the advancing waves of modem, Western thought and technology.
"The ethnographer arrives on the scene of a world on the wane and salvages it in texts before it is
lost to modernization" (Marcus 1994:45). Consider the primary subject matter of economic
anthropology: Hunters and Gatherers, Horticulturalists, Trade and Markets in Precapitalist States,
Peasants, non-Western Markets and Marketplaces, the Impact of Industrial Agriculture, the
Informal Economy, Women and economic institutions, to name the more prominent themes. But
the salvage business was in a sense a step away from the paternalism of the anthropologist as
culture broker, interpreting the world of the primitive, the conquered, to the outside world.
Anthropologists were once relatively unconcerned with the impact of history and macro-
social forces on their subjects. Ethnographies often were written in the "ethnographic present,"
assuming that societies were static, integrated, self-sustaining systems. An attempt was made to
describe "primitive" or "peasant" cultures, before Western contact changed them, as timeless
cultural constructs untouched, pristine and isolated. "Anthropologists could believe in the static
'ethnographic present' only when we were unaware of the extent to which local cultures are
products of world history" (Kottak and Colson 1994:398).
We were once interested in native rationality, in describing the seeming bizarre "other" as
sensible and intelligent once the local context is understood. Assuming the psychic unity of
humankind, we debated whether the categories of economics as understood in the West could be
applied as pan-human categories. "Formalists" analyzed choice as the product of rational
decisions in the context of limited means. "The crux of the [formalist] approach is the
assumption that individuals in every culture exercise rational choice in a means-ends, constraints,
and opportunities framework" (Plattner 1989:13). The formalists, lead by Raymond Firth (1970)
saw all relationships as exchange relationships, as utilities to be maximized.
The "substantivists," lead by Karl Polanyi (1957) and George Dalton (1961), argued that
non-western societies were qualitatively unique and that the application of Western meta-theories
to these societies unjustly distorted their reality. Substantivists saw the use of the classical
economic paradigm as ethnocentric; such theorizing was either wrong or too simplistic and
irrelevant. They argued that scarcity is not a necessary condition of human society but a
historically derived condition. Scarcity of options grows out of the scarcity of wealth, which is a
direct result of the extractive penetration of Western capitalism. The rational choice paradigm
cannot be universalized because the differences between cultures are too great. With no labor
market and no money, reciprocity and redistribution were considered analytical tools far more
important than market exchange. Instead of focusing on monetary value, anthropologists focused
on the value of production for use rather than production for exchange. Substantivists argued that
"interest," "capital," and "credit" are inappropriate categories for describing band and tribal
Evidence that this debate was never resolved lies in the fact that both of these strains of
thought survive in anthropology today. Critique of nomothetic theorizing lies at the heart of a
good deal of current feminist, Marxist and postmodern anthropology, while stalwarts of cultural
materialism and sociobiology continue the search for appropriate macro-theories. Rational-
choice theorists in anthropology, however, continue to be committed to the neo-classical model of
economics as derived from Smith (1905) and Keynes (1936). Rational choice theorists focus on
prescriptive theories specifying how people should act if they want to make efficient economic
decisions. The fundamental assumption here is that people know what they want and should
maximize their energy to get it (Schneider 1974). People are assumed to be calculating beings
who understand their own values and act with forethought. People are also assumed to have the
necessary knowledge concerning costs, income and yield with respect to their options. It is also
assumed that people have the ability to calculate the maximal choice. So, rational-choice
theorists will speak of opportunity costs or the loss of alternatives due to choices we make. Or
they will refer to diminishing marginal value, where the quantity of input eventually outweighs
quantity of output. Employer-employee relationships are described in purely economic terms.
With confidence in free market capitalism, it is assumed that self-interest will eventually make
the world a better place for all (Friedman 1953).
A strong reaction to the confidence in theories of Western capitalism came from those
committed to the historical materialism of Marx. Anthropologists who borrow from Marx focus
on how history and ideology operate to determine the distribution of wealth and power in a
society. The central focus is on the control over the means of production and the various
stakeholders that are affected by this control. Anthropologists influenced by Marx analyze class
conflict and face head on the contradiction between the ideal that all should be equal and the
reality that society is stratified. Following the work of Maurice Godelier (1986), an important
trend has been the analysis ideology that upholds special interests while appearing to benefit
general concerns. The focus on hegemony shifted the focus in anthropology to issues of politics,
power, hierarchy and inequity.
Godelier's notion of hegemony has allowed social relations to become the center of
economic analysis but not without its critics. Immanuel Wallerstein (1974; 1979) criticized
Godelier and the Marxists for their lack of grounding in history. He argues that individual
economies and cultures are not disconnected from others and that, in many cases, there are direct
and dependent relationships directly responsible for the welfare (or its opposite) of the other. Eric
Wolf showed that power relationships have a face, that specific historically significant people and
institutions are responsible for the underdevelopment of many other parts of the world (1982).
With its focus on the micro and the local, anthropology was spiraling into the self-referential
chaos of historical particularism. For those who were caught in this trap, Wolf and Wallerstein
helped raise the scope of the anthropological lens to view the global interconnectedness of culture
Nowhere is the importance of this global perspective in anthropology more obvious than
in the study of fishing communities. And nowhere is anthropology's role in interdisciplinary
conversations more clear. The anthropological analysis of fishers and their communities
embodies the tensions in the history of economic anthropology itself. The smallest of the
traditional fishing communities are by trade dependent on a globally significant resource. Fishing
is often romanticized as a primitive, local trade, yet even the most remote fishing communities
are bound to the consequences of global fishing efforts. Fishing is often atomized as work, and
the embeddedness of this work in the host culture is often ignored.
Anthropologists have become central participants in discussions about how to manage
public natural resources. Researchers committed to understanding the connection between
ecology and culture have developed an interdisciplinary field around the issues involved in the
local and global control over natural resources. James Acheson (1981, 1987, 1989, 1990, 1994),
Eleanor Ostrom (1990) and Bonnie McCay (1987) have lead the way in focusing anthropology's
lens on an issue that forces us to be dependent on multiple disciplines with strands of evidence.
The central theoretical issue in what has come to be known as Common Property Resource
Management stems from Garret Hardin's "Tragedy of the Commons" argument (1968). The
argument stems in part from the Malthusian assertion that unconstrained freedom to produce
children will result in a population disaster for the world. Hardin used the parable of a pasture
held in common property by a community of herders to show that coercion is necessary to
prevent the destruction of the resource. He tried to show that the need for each person to
maximize the personal gain necessarily leads to the destruction of the common grazing area.
Hardin argued that coercion can be mutually agreed upon but need not be just. Arguing that
injustice is preferable to total ruin, Hardin advocated the use of draconian state intervention into
The Tragedy of the Commons theory assumes that the users of common property
resources are individualistic profit maximizers driven by economic goals to overexploit the
resources on which their livelihood depends, despite the best interests of the society as a whole
(McCay and Acheson 1987). It assumes that the users of these resources have the technical
capacity to exceed the biological maximum renewal rate of the resource. It also assumes that
those using common property resources and the local communities they live in cannot or will not
erect effective institutions to protect the resources they live on. Naming it the Free-rider problem,
theorists challenged the presumption that people would act for the collective if it was known that
such action could bring collective benefit (Olson 1965). Finally, theorists committed to the
Tragedy of the Commons model deduced that the exploitation of collectively owned resources
can be halted only by instituting private property or by the government taking action. Private
property was thought to result in a more efficient use and conservation of resources and greater
increases in wealth than do less exclusive forms of property (Ostrom 1990).
Economic anthropologists have important challenges to make to the assumptions of the
Tragedy of the Commons model. Ostrom argues that what makes these models so powerful is
that they fit many situations around the world. What makes them dangerous is when they are
used as proscriptives, where the huge assumptions are taken on faith as being fixed (Ostrom
1990). Anthropologists have shown that there are institutions that effectively limit exploitation.
Assets are rarely open-access (Berkes 1985). Communally owned property (as opposed to open-
access property) is not automatically subject to overexploitation. This is common in many parts
of Africa where grazing lands are owned by the community, clan or tribe. In many fishing
communities, ownership rights are established formally or informally over "fishing space." The
case of the lobster fishers of Maine has proven how communally "owned" property resulted in
less exploitation and greater economic benefit for those involved; access is far more controlled in
perimeter-defended fishing areas than in nucleated fishing areas (Acheson 1988). Durrenberger
and Palsson's study of Icelandic fishers (1985) show how state and local level resource
management can operate together effectively. The Tragedy of the Commons focuses too narrowly
on property rights. Problems associated with open-access property rights are more closely related
to political economy (issues of population growth, industrialization and the expansion of the
capitalist system and markets). As is the case in South Africa, poverty, underdevelopment and
overpopulation push conservation priorities way down the list on the national agenda.
Maritime Anthropology: A Review of Recent Literature
In the past twenty-five years there have been three pivotal publications that deserve
attention for any student of contemporary maritime anthropology. The first is M. Estellie Smith's
edited volume: Those Who Live from the Sea (1977). Published in the late 1970s her work
included a summary of key debates in the study of fishing communities from an anthropological
perspective. The articles focused on themes such as the organization of life around occupation,
the political economy of small commercial fishing communities, technological change and its
impact on local fishers and the relationship between captain and crew in the context of changing
fisheries and markets.
A second work of pivotal importance is Antonius Robben's Sons of the Sea Goddess
(1986). This work is particularly important because it is one of the first works to clearly
articulate the embedded nature of fishing. While Robben clearly focuses on the business of
fishing in small communities in Brazil, he is more concerned with the social relations involved in
the communities that fish. He shows the interdependence between producers and consumers,
between various stakeholders in the industry and between the fishers and their families.
The third publication of pivotal importance in the history of maritime anthropology is the
journal titled Maritime Anthropology. This journal was published out of the Netherlands and was
discontinued in the early 1990s. It helped to draw together works from diverse perspectives that
were each concerned with fishers, families of fishers and fishing communities. This journal was
in part responsible for promoting a field some called Maritime Anthropology. It provided a good
overview of the debates that dominated the literature in the seventies and eighties. Included
among these debates were the analysis of captain-crew relations, debates over the validity of the
"Tragedy of the Commons" model for commercial fishing, the impact of social change on
commercial fishers and the lessons to be learned about commercial fishers from archeological
More recent works that have become pivotal to our understanding about the organization
of commercial fishing communities and/or have pointed in new directions include works by
Matthiessen (1988), Acheson (1988), Acheson and McCay (1989), Eleanor Ostrom (1990) and
Garrity-Blake (1994). Matthiessen (1988) showed implicitly how power relations are structured
between men in the fishing industry. Although for the most part uncritical in his construction of
gender, Matthiessen highlights the fishing industry as work that deeply resonates with traditional
male roles and aspirations. He is particularly clear about how much male fishers value their
independence and freedom. James Acheson (1988) has shown us how local fishers take control
over the management of their territory and fishing rights. He tells us of a "limited entry" system
developed by local lobstermen to limit competition, protecting the resource and their own
livelihoods. Their limited entry system was reinforced by locally constituted "gangs" and backed
up by threat of violence for violators.
Even more recently Barbara Garrity-Blake's analysis of the North Carolina/Virginia
menhaden industry broadens the scope of analysis (1994). She focused her analytical lens on
historical changes in the industry and the impact of these changes on social conditions. With a
study spanning a hundred years she provided the depth of insight lacking from many earlier
works. Her work also added a new twist in that it focused on the construction of race and the
power relations associated with race. She showed how changing race relations were and were not
reflected in the captain-crew relations in the industry.
Two final pivotal works to be mentioned are Acheson and McCay's edited volume The
Question of the Commons (1989) and Eleanor Ostrom's Managing the Commons (1990). While
both focus on theoretical issues beyond fishing, both are directly focused on the central debates
that have dominated the maritime anthropology literature in the last two and a half decades. Both
discuss the strengths and weaknesses of different approaches to what has come to be known as
Common Property Resource Management (CPR). The central concern of CPR is the ecological
health of a region that contains a resource available to anyone. CPR researchers have examined
fisheries, forests, grazing lands, water and air under their microscopes.
Much of the maritime anthropology literature has focused on fishing as a lifestyle,
something more than a job. No work represents this theme better than Antonius Robben's
(1984). His work shows the embeddedness of economic relations. Others have focused on the
culture of fishing in their search for recording the history and emphasizing the heritage of the
commercial fishing lifestyle (Ram 1991, Turner 1991, Smith 1993, Johnson 1995). As an
example, the story of Cortez, a central Florida fishing community, is told by Ben Green (1985).
Another central issue the maritime literature has recently explored is the "skipper effect."
Those interested in the skipper effect devise ways of testing whether the knowledge and skills of
the captain have a significant impact on their success as often defined by the size and frequency
of their catch, independent of other variables such as technology, weather, fishing region, etc.
Durrenberger has worked on this problem and, as with most of the literature so far on this topic,
has concluded that the answer to this problem lies in future research (1993).
Garrity-Blake's work (1994) provides a good example of an anthropological study that
includes an analysis of how technological change affects social relations in the industry. She
shows how the electric spool adopted after World War II not only increased production but
decreased the size of crew needed and the skill needed of those who were hired. This had a
particularly harsh impact on the African-Americans working in the industry. Work yet to be done
in the menhaden industry includes an analysis of the impact of spotter planes on the industry.
Anyone familiar with commercial fishing industries knows that it is an extremely
dangerous occupation. Popular literature and film has recently reminded us of the immediate
dangers of an unpredictable ocean (Junger 1997; this book was also made into a motion picture).
Some social scientists have used this context to gain understanding in how humans approach and
cope with risk. Pollnac and Poggie's work (1995, 1998) on the Icelandic fisheries measures what
people are willing to do, how much risk they are willing to take and how high the benefits have to
be for the different levels of risk. An interesting observation in their findings is that the benefits
from fishing for which people are willing to risk heavy stakes are not necessarily material. They
include personal motivations like a sense of worth or purpose.
Some of the most recent work has focused on the gendered division of labor in the fishing
industry. Works like Matthiessen's ( lu u would lead us to believe that the fishing industry is
dominated by men. Thompson (1985), Nadel-Klein and Davis (1988), Smith and Jepson (1993)
and Smith, Jepson and Lee (1993) have shown us the complex division of labor common in
commercial fishing families. They have shown that for women there is, in part, a complex blend
of traditional gender roles (e.g. mother, teacher, nurturer, housekeeper) and non-traditional roles
(marketing, sales, breadwinner, head-of-household, business manager). Further work needs to be
done in this area to see how and why these roles vary, how they vary with different types of
fisheries and how changes in the fisheries are reflected in changing gender roles in the home. But
it is also important to study how each gender constructs expectations for the other.
Gaps in the knowledge base of maritime anthropology include the following issues:
* What are the boundaries of a fishing community?
Can this be determined by economic or industrial relations alone?
-Can community be better defined by a less-geographically centered analysis?
What is the connection between the fishers and the local community?
How do fishers construct their own sense of community?
* How do race and ethnicity play a role in structuring relations among commercial fishers?
What impact do race relations play in structuring the networks of commercial fishers?
How do the small entrepreneurial businesses reflect the racial stereotyping and biases of
the society in general and what effect does this have on access to the industry?
How is race/ethnicity used as a tool in the competition over fishing rights?
* What role does the construction of gender play in organizing social relations in the industry?
How and why do men construct fishing as a particularly male form of work?
What do gender relations in the fishing industry tell us about the classical distinction
between the formal and informal economy?
One of my objectives is to address the latter two of these gaps in the maritime anthropology
Very few anthropological or sociological studies have been done on the fishing
communities in South Africa, and even fewer on the activity of fishing itself. Most studies in the
1990s developed out of the need to understand the redistribution of fisheries resources in light of
the changing political landscape. The most notable study was De Wet Schutte's (1993) study of
thirteen fishing communities on the Western Cape coast. Schutte was commissioned by the
Department of Environmental Affairs, the Welfare Department and the African National
Congress to collect baseline data with an eye towards improving pending community
development projects. One of his most interesting findings was the diversity of the fishing
communities on the Western Cape coast. He challenges the notion of "the West Coast fishing
community," preferring instead to speak of "those in the fishing industry on the West Coast" or
"those fishers who live in coastal communities" (1993:107). He found that the homogenized
image of fishing communities did not accurately reflect the stark differences he found between
different communities; each had a unique identity and unique day-to-day problems.
Another significant finding in Schutte's work relates to race relations in the coastal
communities. He found that the further the communities along the coast were from Cape Town,
the worse the relationship between local white and coloured residents. When making
recommendations for the implementation of community development programs he recommended
that the local political dynamics in each community be taken into account, including the racial
dynamics as they differ from community to community.
In the last few years there have been some promising developments in research on fishers
and their communities. Two notable master's theses, Isaacs (1998) and Lindsay (1999) analyzed
how local fishers perceived and responded to their social, political and natural environment.
Lindsay's qualitative study investigated and described the perceptions of the fishery resource held
by commercial and recreational fishers in the town of Struisbaai. Though her research employed
a range of methods, including informant interviewing, participant observation, and the use of
secondary data, her findings were based primarily on a series of in-depth interviews with
individual fishers in Struisbaai. Her paper described the ways in which commercial and
recreational fishers in Struisbaai "perceive, conceptualize, and ultimately use the fishery
resource" and explores "the fishers' perceptions of the resource as they are informed by
Apartheid's psychological and social legacies" (1999:i).
Fgure I HjndJl, n FI i, rn ,i n Pulbnh Sn-.rekl nic. ihr Ski-Boat
source: Stibbe and Moss (1998)
Moeniba Isaacs' master's project (1998) took on the form of a report to the University of
the Western Cape's School of Government. She highlighted the conflicts facing the fishing
industry in South Africa exemplified in the fishing communities of Ocean View, Hout Bay and
Kalk Bay. In analyzing the various stakeholders such as the government, the fishing industry,
organized interests (unions) and the unorganized interests (local communities), her research
stressed the conflicts of maintaining stability versus redistribution (how to achieve
empowerment), empowering fisher folk (share holding schemes versus local capacity building),
and managing the marine resource (National fisheries management versus local community
management). She specifically referred to theories of co-management and a historical overview
of the fishing industry in South Africa that sketched the background to the processes leading to
the White Paper up to the passing of the Marine Fisheries Bill (later to become the Marine Living
Aside from the political transformation of the fishing industry, the most publicized
activity in the fishing industry during the late 1990s is the poaching of abalone and rock lobster.
Significant work has been done on studying these activities and the impact of these activities on
local coastal communities (Hauck 1999; Hauck and Hector 2000). Supported by the University
of Cape Town's Institute of Criminology, Hauck and Hector have gone to considerable lengths to
understand how local fishers conceptualize the management of marine species and their
willingness to participate in such management. The strength of their research is in its
ethnographic detail. They move the theoretical discussions of co-management and other fisheries
management paradigms from the classroom to the communities where fishers live.
An important piece of research was funded by the government's Department of Marine
and Coastal Management and involved a survey of subsistence fishers and fisheries all along the
South African coast (Clark 2000). Although lacking in ethnographic detail, the scope of this
project was impressive and provided conceptual tools useful for future research. Subsistence
fishers all along the South African coast were surveyed about their socioeconomic status,
resource harvesting techniques, activities and aspirations. The study was intended to provide
MCM with a baseline to understand subsistence fisheries in order to know how to better
implement subsistence fisheries management plans. A key finding in this research was an
identification of the diversity of activities that can be defined as subsistence fishing. Researchers
involved in this project found that there was too large of a conceptual gap between "subsistence
fisher" and "commercial fisher" and recommended an intermediate category, the small-scale
commercial fisher. It is the fishers in this intermediate category, the small-scale commercial
handline fishers on the roaming ski-boats, who take center stage in the analysis that follows.
' See Thomas Kuhn's The Structure of Scientific Revolutions (1970) for a description of this
transition in history.
2 For a critique of this book, see Gates (1999).
3 Throughout the dissertation I place categorical racial labels in the lower case, except where used
as part of a title. Given my position that these categories are socially constructed and historically
situated, the lower case indicates that these labels are neither proper nouns nor deserving of
codified titular status.
ANTHROPOLOGY AND THE STUDY OF DIFFERENCE
Understanding the Ethnographic "Other"
A Methodological Journey
Despite the ever-shifting boundaries and elusive meanings of"postmodernism," it is
possible to point out some of the more helpful and some of the more dangerous paths down which
influential "postmoderist" thinkers are moving and, more specifically, the value of this
movement for writing ethnography. In its recognition that all representations of the "other" are
inherently political and in its emphasis on the provisional nature of cultural interpretation,
postmodernism leads down a fruitful path. This journey down fruitful paths continues as
postmodernists bring to light, together with feminist influences in anthropology, the need for
reflexivity in writing ethnography. When postmodernists lead us down the road of subjectivist
epistemological obscurity and ill-defined moralism, the road gets all too rocky. In their denial of
relative objectivity as a goal for ethnographic description, postmodernists take us further down
the road to an analytical black hole. Postmodernism also leads down the rocky road of
narcissistic solipsism when the sole purpose, or even the dominant purpose, of the ethnography is
the reflexive examination of the author's own position in the research. By focusing on questions
raised by postmodern anthropologists, it is possible to show that the centripetal forces of
postmoderism and the centrifugal tendencies of the scientific enterprise have the potential for
The anthropologist steps off the plane into the exotic land of the "other." Social
philosophers and researchers have long been concerned with understanding and communicating
the ideas and behaviors of people groups different (often radically) from their own. The history
of cultural anthropology as an academic discipline is the history of research that helps the "other"
become familiar. But as the world is made smaller through mass communication, rapidly
advancing communication technology and the increasing ease of international transportation, the
"other" is no longer so unfamiliar. The cultural diversity that the anthropologist once had to
bring home in his briefcase is now an integral part of our own neighborhoods. The boundaries
between the "we" that study and the "they" that are being studied are breaking down while, at the
same time, new social categories are being formed. I believe it is with the analytical tools that
cultural anthropology provides that we can best come to understand the dynamics of these
changing social categories.
The typical product of the cultural anthropologist's research takes on the form of an
ethnography. An ethnography is a text in which a portrait, sometimes even a masterpiece, of a
particular group of people is painted. Alongside other functions, writing ethnography (as
dissertation, professional articles or monograph) serves as a form of academic capital, a rite du
passage to the scholarly world of anthropology, an expected entry on the curriculum vitae of the
aspiring tenure track professor. As such, the ethnography becomes an intellectual product on the
publishing assembly line.
Any discussion of methodology (epistemology) cannot be wrested from the academic
context in which anthropologists are trained. Sally Cole notes that she "went about the task of
producing an ethnography that would pass the examining board that would license [her] as a
professional anthropologist" (1992:118). The anthropological dissertation, "typically a
straightforward analytical and descriptive account from fieldwork, is the ethnography that most
anthropologists must write. Since the granting of professional credentials has depended on its
evaluation, it has tended to be a conservative exercise" (Marcus 1986:265).
Taking its cue from contemporary literary theory, cultural anthropologists have had to
reexamine their discipline in light of critiques increasingly being labeled postmodern. But for
one in the process of deciding which methodologies prove to be most useful, questions raised by
postmoderists are cause for examination, not reexamination. To discuss postmodernism is like
taking a journey with no prescribed destination, no directions and a thousand back-seat drivers
(pardon the hyperbole). No one knows quite where it is going, how one will get there or who is
giving directions. The multiplicity of interpretations of postmodernism is evidence that it is a
highly contested theoretical domain. Postmodernism is like an amoeba with flexible boundaries
and often directionless movement. Those who claim to be postmoderists are often willing to
ingest any theoretical or methodological tidbit that may advance their cause. However, despite
the ever-shifting boundaries and elusive meanings of"postmodernism," it is possible to point out
some of the more helpful and some of the more dangerous paths down which influential
"postmodernist" thinkers are moving and, more importantly, the value of this movement for
Preparing for the Trip
As with any other academic discipline, cultural anthropology is involved in the
production of knowledge, and the ethnographic text is our most common academic product.
Anthropologists are in a continuing dialogue over the goals for the ethnographic representation of
culture. Should ethnographic description aim at the interpretation of the individual's symbolic
representations of her world, or should ethnographic description aim at providing valid
generalizations about cultural phenomena? Are these two goals mutually exclusive?
Furthermore, anthropologists are in disagreement over what the relationship should be between
the researcher and the subjects studied. How explicit should the details of this relationship be
made in the ethnography, and to what extent should the ethnography reflect the dialogue between
the anthropologist and his informantss? These are questions of methodology (epistemology),
i.e., questions regarding the assumptions behind the process of writing ethnography, questions
that reveal how we know what we say we know.
Russ Bernard (1994) makes a broad distinction between two epistemologies dominant in
cultural anthropology: interpretivist and positivist. He makes an important distinction between
epistemology (methodology) and methods. "Positivists and interpretivists may disagree on
matters of epistemology," Bernard says, "but when we talk about methods at the level of strategy
and technique, methods belong to all of us" (1994:169). One's methodology consists of the body
of assumptions that informs and directs the methods (strategies and techniques) we use in
research. The primary questions that drive our research both inform and are informed by these
assumptions. Although placing himself firmly within the logical positivist tradition, Bernard
all anthropologists need a thorough grounding in the various approaches to knowledge
that have characterized our discipline. This means exposure of all students, whatever
their initial predilections, to the philosophical foundations of structuralism, symbolism,
interpretivism, hermeneutics, phenomenology, positivism and empiricism. (1994:174)
Although the dominant debate in the discipline throughout the last three decades of the twentieth
century has been between positivists (e.g. Kuznar 1997) and interpretivists (e.g. Geertz 1973,
1988), the debate has escalated where the battle lines are drawn between those committed to a
scientific, empiricist epistemology and those intent on deconstructing the foundations for the
scientific enterprise itself, i.e., a postmodernist epistemology.
Michael Agar envisions the current debate in anthropology between the postmodernists
and those committed to the scientific, empiricist method as a war in which there is, as with most
wars, more destruction than anything else. In The Professional Stranger he notes that it is "high
time for some peace negotiations" (1996:4), and I would have to agree. My work is not a
comparison and contrast between scientific method and postmodernist critique. This chapter is
not intended to give a full documentation of the scientific epistemology followed by a full
articulation of the postmodernist critique. This sketch is less ambitious. Rather than facilitate the
destructive polarization often reflected in such generalized comparisons, I will aim a more
constructive lens at postmodernism.
Postmoderism includes a set of critical epistemological or methodological tools that
direct the writer (and reader) to challenge established ways of representing, or writing, culture.
More specifically, postmodernism emphasizes the radical heterogeneity of culture and the
"decline of ideological hegemony in politics and social life" (Dickens and Fontana 1994:4).
Regarding cultural anthropology, postmodernism aims its strongest critique at its most popular
contribution to academic inquiry: the ethnography as a re-presentation and interpretation of
culture, or more specifically, the ethnographer as re-presenter and interpreter ofculture. Yet
there are more helpful (fruitful) paths and more dangerous (rocky) roads down which
postmodernism can lead cultural anthropology. For my own research, I will take the more fruitful
paths and steer clear of the rocky roads.
I will discuss these fruitful paths and rocky roads in this essay. In its recognition that all
representations of the "other" are inherently political, postmoderism leads down a fruitful path.
The re-presentations of the ethnographer both contribute to and are affected by relations of
power. The power relations that affect the writing of ethnography occur in the academic world of
the anthropologist, in the socio-political world of those being studied and in the interchange
brought about by the contact of these two worlds. In its emphasis on the contextualized nature of
cultural interpretation, postmodernism leads down another fruitful path. When postmodernism
leads down the road ofsubjectivist epistemological obscurity and ill-defined moralism, it leads
down a dangerous road. The radical subjectivism of the deconstructionist agenda ultimately
denies the potential for comparison among groups; in its attempt to highlight ambiguities,
difference and discontinuity, postmoderism forgets that there are also regularities, patterns and
discernable intracultural continuity.
As the Ecclesiastical philosophers recognized, there is nothing new under the sun
(Ecclesiastes Chapter 1). This proverb also applies to anthropology as a discipline: "In
anthropology we are continuously slaying paradigms, only to see them return to life, as if
discovered for the first time" (Wolf 1994:220). According to Roy Rappaport:
Two traditions have proceeded in anthropology since its inception. One, objective in its
aspirations and inspired by the biological sciences, seeks explanations and is concerned
to discover causes, or even, in the view of the ambitious, laws. The other, influenced by
philosophy, linguistics, and the humanities, and open to more subjectively derived
knowledge, attempts interpretations and seeks to elucidate meanings. (1984:154)
Citing continental philosophers, social theorists and literary critics such as Paul Ricoeur (1971),
Jacques Derrida (1972, 1978), Michel Foucault (1980a, 1980b), Jean Baudrillard (1981), Jean-
Francois Lyotard (1984), Pierre Bourdieu (1987), and Jurgen Habermas (1987), postmodernists
reflect this second tradition. In postmodernism we hear the strong echoes of Robert Lowie's
famous statement that culture (civilization) is a "thing of shreds and patches" (1920:441).
Subjectivity is prized over objectivity. The particular is given primacy over the general.
Postmodernism "remains more a socio-cultural theory than a set of epistemological and
discursive principles" (Agger 1992:109). The distinctions between post-structuralism and
postmodernism are difficult to determine. "One would have to engage in purposely simplifying
taxonomy of names and their intellectual contributions in order to map the terrain of post-
structural and postmodern cultural studies adequately-and even then the map would blur some
crucial points of difference" (Agger 1992:109). Such a map is beyond the scope of this essay.
However, as Dickens and Fontana highlight, not all who consider themselves postmodemists (i.e.,
critics of modernism) would fully embrace Derrida's style of deconstructionism (Dickens and
Fontana 1994). Yet, in its emphasis on culture and, in fact, all of reality as text, deconstruction is
inseparable from the postmodernist agenda. Deconstruction involves:
the attempt to take apart and expose the underlying meanings, biases and preconceptions
that structure the way a text conceptualizes its relation to what it describes. This requires
that traditional concepts, theory, and understanding surrounding a text be unraveled,
including the assumption that an author's intentions and meanings can be easily
determined. (Denzin 1994:184)
Deconstruction provides the critical tools for postmodernism to challenge representations of the
"other" in ethnographic texts.
Fruitful Paths and Rocky Roads
One helpful path down which postmodernism steers anthropology is the recognition that
all representations of the "other" (as the product of research) are inherently political, which in this
case is defined in the broadest sense as the power to influence decision making. Representations
of culture are inherently political because culture itself is a contested domain. We can no longer
speak of "the Nuer" or "the Andaman Islanders" without acknowledging the intracultural
variation, the diverse and often competing voices each claiming a stake in what it means to be
Nuer or Andaman. Deeply indebted to Foucault's notion that knowledge and power exist in a
dynamic, interactive relationship (1980b), postmodernism recognizes the need to deconstruct
hegemonic representations of culture that operate to effectively oppress certain portions of
society. These power relations are commonly divided along the lines of age, class, gender and
race, among others. The search for statistically significant patterns of thought and behavior (the
core agenda of positivist anthropology) has its limitations in that it focuses on dominant
representations, the most politically influential persons, hegemonic rather than subaltern forces,
the core rather than the periphery. If one is intent on deconstructing and redefining the core-
periphery debate, statistical probabilities can only take you so far. Postmodernism points to the
need to create a space for under-represented voices. And as Agar so insightfully recognizes,
Ethnography is populist to the core, in this sense-skeptical of the distant institutions that
control local people's lives; certain of the fact that the best society is built from the
participation of its members in decisions that affect them; aggravated by injustices
caused by distant institutions that force people to live in worlds not of their own making.
A second fruitful path down which postmoderism has redirected' cultural anthropology
is an emphasis on the provisional nature of cultural interpretation. As all culture is constantly
changing, representations rely on changing contexts. Both the ethnographer and the people who
are studied are influenced by historically (politically, socially, economically) situated
circumstances as, for example, Evans-Pritchard's research for writing The Nuer was influenced
by the support he gained from the Sudanese government, or simply the fact that he was an upper-
class, white male British researcher, or the reality of pending civil war in 1939. Postmodernism
highlights the need to recognize the historical contingency of cultural interpretations.
Postmodernism challenges the "definitiveness" of the ethnographer's representation and
deconstructs the role the ethnographer in "writing culture" (Clifford and Marcus 1986).
In an attempt to practice what they preach, postmodernists promote reflexivity in
ethnographic writing. The ethnographic "other" has been the under-represented voice in the
classical ethnographies (e.g. Radcliffe-Brown's The Andaman Islanders (1922) or Evans-
Pritchard's The Nuer (1940). Sally Cole believes that contemporary ethnography should be a
"collaboration" between her and the groups) with whom she is working. She is an advocate for a
"contemporary ethnography wherein anthropologists self-consciously attempt to acknowledge
their presence and integrate their personal experience or political consciousness in the writing of
ethnography itself' (Cole 1992:115). This acknowledgment and integration is called
"reflexivity." Explicit reflection on the observer-observed relationship is an integral part of a
reflexive ethnography. The anthropologist not only places those studied in their historical
(economic, political and social) context but analyzes their own historical context and how their
position relates to the ones being studied. Reflexive ethnographers often include verbatim
dialogue with particular informants in an attempt to place interpretive power in the hands of the
informants. They are known for their use of the first person narrative in the life histories
(individuals or groups) and for the personal narratives of anthropologists, where the
anthropologist is studied as subject (Crapanzano 1980, Dwyer and Muhammad 1982, Behar
1993, McClaurin 1996).
Reflexivity can be taken to an extreme. Consider the soliloquized comments of Malcolm
Ashmore, a sociologist being reflexive about reflexivity.
In order not to be scientific, one must be outside science; but to study science or anything
else from the outside is to be scientific. Therefore, in order to study science
unscientifically one must abandon objectivity and study it from the inside. But to be
inside science means to be scientific. And therefore... (Ashmore 1989:109)
His point is not to destroy the goal of objectivity in science but to problematize it. He concludes
not that we should throw out objectivity but that "in the study of science (and knowledge
practices generally) the student cannot avoid being inside and outside at the same time"
(Ashmore 1989:109). The danger of hyper-reflexivity is that we will be like the sixth century
monastic Hesychasts, believing that navel-gazing will bring us closer to our goal. Consider one
of the anthropologists representative of the postmodern ethnographers, James Clifford. "The
other for Clifford is the anthropological representation of the other" (Rabinow 1986:242).
Although it is essential to critique the way we represent others in our ethnographies and to
acknowledge our own position in the ethnography, the ethnography is not an autobiography. I
want my ethnography to say something important about someone else.
Another of the dangerous roads to which a postmodern perspective leads anthropology
stems from postmodernism's marriage with Derrida's style of deconstruction. "A deconstructive
cultural studies does not linger very long in the land of nomenclature, taxonomy, glossaries or
conceptual refinements as if these events could somehow take place outside of the con-texts
within which everything is subsumed under the rule of undecidability" (Agger 1992:110).
Derrida built on de Saussure's insight that language consists of a system of relations among
arbitrary signs whose meanings are defined by the differences that set them apart from one
another. Derrida challenged all claims to knowledge that do not recognize the constantly shifting
nature of representations of reality, since representations of reality are all that we can
approximate (Dickens and Fontana 1994:1-24; 183-202).
Ben Agger (1992) explores the contributions of"poststructuralism" (deconstruction) and
"postmodernism" to the emerging tradition of postmodernist cultural studies. A critique of this
deconstruction, or postmoderism in general, is at one level futile given the self-proclaimed
aversion to theoretical constructs and the distrust of authoritative definitions in postmodern
cultural studies. As Agger notes, "Deconstructive cultural studies... is especially capable of
theorizing its shifting, evolving nature, resisting fixed definitions at every turn" (1992:109). Yet,
it is still possible to judge the internal consistency between various claims made by those
promoting postmodern cultural studies. One question that needs to be asked of the postmodern
agenda is this: are the methods of Derrida's deconstruction consistent with the political aims
claimed central to a postmodern cultural interpretation?
In order to construct postmodern cultural studies as something more than deconstruction,
Agger proposes a constructive political aim for postmodernism. Agger hopes to do more than
highlight hidden assumptions and inconsistencies. He wants to shift "the cultural-political
balance of power" (1992:101). The political agenda of Agger's postmodernism is clear: to
destroy the authority of "positivist cultural empiricism" (1992:94) or "mainstream positivist
sociologists of culture" (1992:97). But if Agger's aim is solely destructive, the end result would
be nihilism. Agger attempts to posit a constructive aim for postmodernism, i.e., "restoring the
value of marginalia" and non-hegemonic interpretations of culture. But is deconstruction
consistent with this aim?
Cultural study "intervenes in the cultural field... as a subversion of unchallenged authorial
privilege" (1992:97). But cultural study, as interpreted by Agger, cannot ultimately make use of
deconstruction to meet its political aims. If, by exposing hidden forms of power, the critic is to
provide voice and power for those left out of the "mainstream," they soon become part of the
power center themselves. The center may have shifted, but hegemony itself continues. In
attempting to "restore the value of marginalia" (1992:107), deconstructive cultural criticism
creates a new center. This center may be formed of novel ideas and may empower persons
formerly marginalized, but the marginalized soon become the hegemony. Consider the "gods" of
postmodernism themselves. Although they would certainly deny the status of authority for
themselves, persons like Derrida, Foucault, Bourdeau, De Saussaure and Adorno are consistently
invoked in "postmodern" literature for their critical acumen (dare I say theoretical position?).
Although Agger states that "poststructuralism and postmodernism oppose their
methodologization" (1992:94), he reflects a contradiction inherent to postmodernism when he
states that "Derridian cultural studies.., has devoted much more attention to the development of
critical method than to the building of substantive social and cultural theory that functions
politically" (1992:108). Deconstruction itself is methodological while claiming no methodology.
It is theoretical while claiming an aversion to theory-building. It centers discussion on the
continual process of analysis of text while claiming decentralization as its aim. Deconstruction
leaves a "vacancy of meaning that needs to be filled" (1992:102) but leaves us not only without
the tools to fill the vacancy but ultimately without the possibility of filling the vacancy. In trying
to fill this vacancy with a constructive political aim, if Agger is not open to the criticisms of
Derrida himself, then he contradicts the deconstructive methodology. The alternative is to remain
in a continual process of deconstruction upon deconstruction upon deconstruction. The methods
of Derridia's deconstruction are inconsistent with the political aims claimed central to a
postmodern cultural interpretation.
Derrida's style of deconstruction extrapolates on the reality of historical conditioning,
mentioned above as a fruitful enterprise, to the extreme of epistemological obscurity, a rocky road
too dangerous to travel. Following the logic of deconstruction, since no knowledge or
understanding of the "other" is possible, i.e., only interpretations upon interpretations, the best
that ethnographers can do is to deconstruct the variables that influence (or bias) their own
interpretation of reality and their interpretations of the "other." The tendency of this intense
subjectivity is to degenerate the study of culture to the level of psychoanalysis, negating any real
possibility for comparative research. While deconstructionists like Mark C. Taylor (1984) are
enjoying the infinite wandering of exploring interpretation upon interpretation (a process he calls
trace), the cultural anthropologist needs to move on to explore the differences and similarities
between groups of people. "The new ethnography turns a blind eye courtesy of its own ideology
here, I think. They're so committed to the complexity, ambiguity, subversion and relativity of
any given moment that they lose sight ofpatternized variation that regularly occurs" (Agar
There is another dangerous road down which postmodernism leads the ethnographer. In
their eagerness to emphasize the subjective nature of reality and to describe culture as a contested
domain, postmodernists have lost sight of (and often attacked) objectivity as a goal for research
(Clifford and Marcus 1986). The goal should not be to destroy the desire for validity and
reliability in research (i.e., the two key components of scientific objectivity) but to reexamine
who determines what is valid and reliable and for what purpose. Even those who most fervently
preach the gospel of a positivist commitment to the scientific method do not claim access to
absolute TRUTH but rather emphasize provisional truths. Striving for explanations, or
interpretations, that others can agree upon does not negate the possibility for amplifying the
voices of the marginalized.
Geertz believes that there is no substitute for local knowledge,
But maps and charts may still be useful, and tables, tales, pictures, and descriptions, even
theories, if they attend to the actual, as well. The uses of ethnography are mainly
ancillary, but they are nonetheless real; like the composing of dictionaries or the grinding
of lenses, it is, or would be, an enabling discipline. And what it enables, when it does so,
is a working contact with a variant subjectivity. It places particular we's among
particular they's, and they's among we's; where all, as I have been saying, already are,
however uneasily. (1994:463)
"Objectivity means becoming aware of one's biases, and transcending them, not the lack of any
biases.... Striving for objectivity is important even if perfect objectivity is unobtainable"
(Bernard 1994:172). In taking anthropology down the road of particularist, subjectivist
descriptions, postmodernists may lead down a road they themselves wish to avoid. Ethnography
can be the great enemy of ethnocentrism, "of confining people to cultural planets where the only
ideas they need to conjure with are 'those around here,' not because it assumes people are all
alike, but because it knows how profoundly they are not and how unable yet to disregard one
another" (Geertz 1994:463).
A final rocky road to avoid when exploring the usefulness of postmodernism for cultural
anthropology is its tendency to moralize without articulating clear moral principles.
Postmodemism provides a moral agenda: to alleviate oppression. Yet the objects of oppression
are not clearly identified; who is it that is oppressed and in what ways? As mentioned above, a
strength of the postmodern perspective for cultural anthropology is its recognition of the need to
deconstruct hegemonic representations of culture that operate to effectively oppress certain
portions of society. Race, class and gender are invoked as core variables that contribute to this
oppression (or hegemony). But postmoderism has the dangerous potential of becoming the very
thing it critiques. It has the potential of becoming the arbiter of the hegemonic discourse, rather
than the critic. When creating space for under-represented voices, what prevents the under-
represented voices from becoming the hegemonic voices? Postmodernism provides critical tools
to deconstruct the composition (definition) of power relations but has yet to define guidelines for
reconstructing power relations in a manner that will ultimately result in a more even distribution
of power. Moreover, it is not clear whether equity is even an important goal for postmodernists
to pursue; for some the whole academic enterprise is nothing but an academic interpretive game.
Navigating Between Scylla and Charybdis
The discipline of cultural anthropology is in crisis over how to define its core product, the
ethnography. As an attempt to apply the critiques brought to the writing of ethnography by
postmodernism, consider this comparison of a classic ethnography, E.E. Evans-Pritchard's The
Nuer, with one example of an ethnography influenced by postmodernism: Vincent Crapanzano's
Tuhami. More specifically, attention is given to their differing approaches to the relationship
between the general and the specific, between theory and text, between generalizations and the
phenomena of experience. An argument is made for the necessarily interactive nature between
both the theoretical poles (theory and text) and the relational poles (anthropologist and
informant). This comparison is intended to neither oversimplify nor dichotomize differences
among ethnographic approaches. A comparison and contrast between Evans-Pritchard's The
Nuer (1940) and Crapanzano's Tuhami (1980) will reveal some of the fundamental issues with
which current graduate students intent on ethnographic production must wrestle.
As mentioned above, cultural anthropology is involved in the production of knowledge,
and the ethnographic text is our most common academic product. Anthropologists continue to
dialogue over the goals for writing ethnography. Strong lines of preference are drawn around
whether ethnographic description should aim at the interpretation of the individual's symbolic
representations of her world or if ethnographic analysis should aim at providing valid
generalizations about cultural phenomena. Furthermore, anthropologists are in disagreement over
what the relationship should be between the researcher and the subjects studied. There is
considerable debate over how explicit the details of this relationship should be made in writing
ethnography, and to what extent the ethnography should reflect the dialogue between the
anthropologist and his informantss. A comparison of The Nuer and Tuhami should help
illustrate these particular issues.
Doing his research in the 1930s, Evans-Pritchard was at the cutting edge of British social
anthropology, following in the footsteps of his mentor, A.R. Radcliffe-Brown. Radcliffe-Brown,
and consequently Evans-Pritchard, saw the goal of ethnographic research as the systematic
analysis of the structures of society and how such structures functioned in the context of other
such social structures. Radcliffe-Brown was famous for his analysis of kinship structures and
how such structures could predict behavior. Reacting to what he perceived was an extreme
particularism in the Boasians, Radcliffe-Brown sought comparative sociological laws that could
explain similarities among cultures and account for differences. Evans-Pritchard believed in his
mentor's approach and set out to prove the effectiveness of his theories through extensive field
research. His goals are articulated in the introduction of The Nuer "We first describe the
interrelation of territorial segments within a territorial, or political, system and then the relation of
other social systems to this system" (1940:4). Evans-Pritchard believed that social structures
existed and sought to fill carefully reasoned structural-functional categories with the necessary
ethnographic content, thereby adding validity and reliability to the theory. The goal of The Nuer
was not only to describe "the ways in which a Nilotic people obtain their livelihood, and their
political institutions" (1940:3) but also to provide research that supported a structural-
functionalist theoretical paradigm. He was specifically intent on providing valid generalizations
about cultural structures.
Evans-Pritchard's work was controlled by a deductive analysis of social situations. Intent
on providing an understanding ofNuer culture, the specific was subsumed under the general, the
individual understood only in the context of the group. Rather than describing the experiences of
specific, identified individuals, Evans-Pritchard attempted to explain Nuer culture by defining, for
example, the structure of kinship and providing examples to support the definitions. Although in
the introduction he related some of the struggles he had in finding regular, consistent informants,
Evans-Pritchard masked the voices of his informants in the omniscient ethnographic "we,"
relating information about the Nuer from an authoritative position. In order to obtain the kind of
quality detail that he did, Evans-Pritchard had to have had numerous strong informants. But the
identity of these informants remains beyond confirmation from the text provided.
Writing four decades later than Evans-Pritchard, Crapanzano reflects the influence of
such anthropological icons as Clifford Geertz, Nancy Munn, James Clifford and George Marcus.
Disillusioned with the attempt to fit cultural phenomena into preconceived theoretical constructs
(constructed by academics), Crapanzano's mentors seek to deconstruct the relationship between
anthropologist and informant in an attempt to provide a voice for the subjects of their research.
These theorists are interested in highlighting the active role researchers play in the representation
of the "other." The central goal of Crapanzano's ethnography is to examine the way in which the
researcher and subject interact to produce a "negotiated reality" (1980:x). Taking an explicitly
psychoanalytic approach,23 Crapanzano examines in great detail the way in which Tuhami uses
symbols in his creation of reality.4
The fulcrum of analysis for Crapanzano is the subjective experience of the individual.
He attempts an inductive ethnography, starting from the phenomena of mutual experience
between himself and Tuhami and building generalizations about Tuhami and his cultural context.
In contrast to Evans-Pritchard, where details were provided in support of generalizations,
Crapanzano attempts to show generalizations can be built from a particular life history
(autobiography). He believes that Tuhami's tale "carries implicitly, if not explicitly, the
Moroccan values, interpretational vectors, patterns of association, ontological presuppositions,
spatiotemporal orientations, and etymological horizons that are embedded in his idiom" (1980:7).
At times Crapanzano speaks with informative authority and at times allows the evocative voice of
Tuhami to come through relatively uninterpreted (1980:14, 27-72, 91-130, 155-172).
The Nuer is a classic example (although a bit outdated) of the ethnography that attempts
to make valid generalizations about cultural phenomena. The relationship between the
ethnographer and the "informants" remains implicit. Evans-Pritchard chose to include neither
lengthy quotations from informants nor direct dialogue with them in this ethnography. Tuhami is
an example of an ethnography that is intent on interpreting the life of an individual as symbolic
representations of reality. The relationship between the ethnographer and the individual studied
is explicitly stated throughout the book.5 The former ethnography moves from the general to
understand the specific, from the theory to understanding the data, from generalizations about
livelihood and political institutions to the details of Nuer existence. The latter ethnography
moves from the specific to the general, from one man's life to an understanding of life in
Morocco, eschewing theoretical formulations.
Unfortunately, neither approach is entirely satisfactory. Evans-Pritchard's ability to
present the dominant, patriarchal political and economic systems is admirable. However, his lack
of attention to intra-cultural variation and to the representation of minority voices must be
criticized, albeit as much a product of his generation of anthropology as a weakness in his
method. In this work I get an understanding of how resources were controlled at the macro level
yet remain unconvinced that the structures presented represent the majority of"the Nuer." Little
attention is given to the structural roles for women or for those without access to political and
economic resources. Evans-Pritchard falls into the trap of forcing the round pegs of Nuer
experience into the square holes of structural analysis. And what I learn about the ethnographer
from this work, I learn from reading between the lines (or the very short introduction). This
approach can be improved upon by increased attention to under-represented forces and by making
more explicit the relationship between the ethnographer and the ones being studied.
Tuhami: Portrait of a Moroccan is just that: an interpretation of one Moroccan's unique
existence. One strength of this book is its ability to represent the multiple levels of reality from
which Tuhami makes sense of his life in Morocco. From this I learn a little about a man named
Tuhami and a little about an anthropologist named Crapanzano. The reflexivity of this work
stands as not one of many agendas but perhaps the central agenda. I learn very little about
Morocco or Moroccan culture and quite a lot about Crapanzano, the anthropologist. Cultural
interpretation is reduced to the most subjective form of psychological inquiry: psychoanalysis.
What generalizations are made are not done so by providing evidence but by reference to the
conversation between the ethnographer and the individual. Other than such references, the reader
must rely on the generalizations and interpretation given by Crapanzano himself. Crapanzano's
attempt to place the ethnographic authority back in the hands of the "other" has collapsed in on
A Case for Relative Objectivity
Objectivity, reliability and validity are laudable goals for the social researcher to target,
even if they are ultimately unattainable. Kirk and Miller (1986) believe that systematic
quantitative and qualitative social research can provide analytical tools that, while they are not to
be taken as absolute criteria, are tools that get at greater approximations of truth in research. This
provisional truth is ideally the kind of knowledge that is open to public scrutiny, testability and
falsifiability. Objectivity "refers to taking an intellectual risk-the risk of being demonstrably
wrong" (Kirk and Miller 1986:10). But if someone wants to prove the researcher wrong, he
would have to do so on the terms as defined by the community of scholars. If validity and
reliability are not absolute, someone must be responsible for defining the acceptable levels. If the
community of scholars is responsible for setting the standards that data must meet, then the
authority of valid and reliable (i.e., objective) research findings rests in the hands of the
community of scholars, not simply some abstract methodology we call science.
One of the things that makes cultural anthropologists different from tourists is the attempt
to collect information about people systematically. The social researcher wants to do more than
observe and enjoy, although both observation and enjoyment are hopefully a part of the research
process. Cultural anthropologists record their observations, interpret their observations in light of
others who also study issues similar to those they have studied (i.e., a community of scholars) and
attempt to communicate their findings in a language best understood by this community of
scholars. Irrespective of other forms in which anthropologists want to communicate their
findings (e.g. as reports to development agencies, political petitions for local community activists,
screen plays, historical archival records, etc.), academic anthropologists need to communicate
their findings in a form conducive to academic inquiry. This means in part that researchers need
to agree on a common language. Researchers need a common language that contrasts data and
noise, what is important and what is not important.
Kirk and Miller (1986) argue that the search for objective data is an essential component
of a common research language for social scientists. By "objectivity" they do not mean some
pure form of knowledge or absolute truth; they prefer to leave the search for this type of
knowledge to philosophers and theologians. Objectivity is based on the assumption that there is a
world out there that can be empirically known. But knowledge of this empirical world is always
provisional, based on theoretical assumptions that the anthropologist must be able to articulate.
"'Truth' (or what provisionally passes for truth at a particular time) is thus bounded both by the
tolerance of empirical reality and by the consensus of the scholarly community" (Kirk and Miller
1986:12). This truth is always provisional because it must remain testable and falsifiable by those
interested in studying the same issues.
The search for historically conditioned "objective" data necessarily involves the
realization of as much validity and reliability as possible. "Reliability is the degree to which the
finding is independent of accidental circumstances of the research, and validity is the degree to
which the finding is interpreted in a correct way" (Kirk and Miller 1986:20). But to define what
is truly "independent of accidental circumstances" and to determine when a finding is interpreted
in a "correct" way remain open to debate. Alternative views are tested against the standards set
by the intellectual community over time. Acknowledging the influence of a scholarly tradition
allows for alternative views to challenge historically specific arguments in the contexts they were
argued. As others join this community, they have the opportunity to change and reshape the
definitions that form the foundation for quality research.
In the study of culture there should be a dialogical relationship between the general and
the specific, between our understanding of the individual and the groups) to which that
individual belongs, between text and theory. Crapanzano joins the Boasians in his attempt to
allow the historical particularity of Tuhami's experience to speak for itself, but one cannot
understand individual experience void of historical and structural context. Interpretations of an
individual's symbolic representations of the world must be placed in their social, political and
economic context (con-text: literally that which goes "with the text"). This context necessarily
involves a level of abstraction, a generalization from the individual to the social. The two are not
Furthermore, the relationship between the researcher and the subjects studied should be
made explicit in the ethnography. It is important to know whose perspective of culture is re-
presented in the ethnography and, if generalizations are made, on whose authority generalizations
are made. While the relationship between the ethnographer and the informants should be made
explicit, this relationship need not overwhelm the study. The reflexivity of the researcher can be
taken to an extreme where all one sees is a portrait of the author and not the ones studied.
Like Odysseus on his perilous journey, the anthropologist must avoid the Scylla of
sacrificing text to theory and the Charybdis of sacrificing theory to text. If the theoretical
construct of the ethnographer does not take into account the complexity and dynamic nature of
intracultural variation, the ethnography may be destroyed by the rocks of ethnographic validity.
If all the ethnographer can offer is the uniqueness of individual experience, the ethnography will
be sucked down by the whirlpool of nihilistic subjectivity.
Where postmoderism is helpful to cultural anthropology is first in its recognition that
culture is a contested domain in both its construction and representation. Amplifying voices from
the margins is important, if not central, to the writing of ethnography. Postmodernism is also
helpful in emphasizing the historical contingency of ethnographic representations. Both the
ethnographer and the people being studied are conditioned by historical (social, political and
economic) motivations. Furthermore, as postmodernism encourages constructive reflection on
the process of writing ethnography and on the relationship between the ethnographer and those
studied, it takes us down another fruitful path.
If anthropologists are intent on answering comparative questions, they cannot buy
wholesale into Derrida's style of deconstruction. On the one hand, deconstruction is a helpful
methodological tool for analyzing the process of writing ethnography, for analyzing power
relations between anthropologists and their "others" and for recognizing the contested nature of
cultural interpretations. On the other hand, to follow all the way down the deconstructive road is
to buy into a subjectivism that destroys any real chance of understanding anything other than the
internal conversation with oneself, a solipsism dangerous to one's sanity. Postmodernism should
aim its critical faculties at the construction of objectivity, at whom it is that defines the
"objective" perspective and for what purpose. The goal of objectivity itself does not have to be
destroyed in the process. Furthermore, anthropologists are not precluded from taking a moral
stance (see Schepper-Hughes 1987, 1995), but writers must carefully articulate by what moral
principles the ethnographer should operate.
One way to picture the debate between postmodernists and those committed to the
scientific exploration of the social world is by means of a metaphor from modern physics.6
Postmoderism acts as a centrifugal force on anthropology, compelling us to take seriously the
particular parts of our world in all their uniqueness and ambiguity. The scientific approach acts
as a centripetal force, forcing us to take seriously the structures and patterns of culture that tie the
world together. If either the centrifugal tendencies of postmoderism or the centripetal forces of
science claim exclusive right to the driver's seat in cultural anthropology, it will destroy the
dynamic character of anthropology itself.
By maintaining a dialogue between these opposing forces, checks and balances are
provided that keep the discipline from either collapsing in on itself or exploding into chaos. If the
anthropologist can only talk to and about herself, her solipsistic wanderings will be of no use to
anyone else. If the anthropologist has no tools with which to organize the chaos, no one will
Analyzing Race via Social Networks
Research over the past thirty years has shown that social networks are of crucial
importance in understanding the procurement and retention of employment (Granovetter 1973,
1974, 1982; Lin and Dumin 1986; Zimmer and Aldrich 1987). A strong interest has also grown
in the roles that race and ethnicity play in constructing and maintaining social networks,
particularly as each relates to employment (Cobas et al. 1993, Cobas and DeOllos 1989, Light
and Bonacich 1988, Du Toit 1998, Ooka and Wellman 1999). The role of inter- and intra-racial
social networks has been shown to be particularly central in the procurement and retention of
resources for entrepreneurs (Sanders and Nee 1996). While some have shown the importance of
intra-racial ties for success in entrepreneurial development, little work has been done to study the
importance of inter-racial ties on such success or lack of success (see Ooka and Wellman 1999
for an exception). This study provides insight into the structure and content of personal and
professional networks of commercial fishing skippers and crew on South Africa's Western Cape
coast. Particular emphasis is on the role that the construction of racial and ethnic identities plays
in developing and maintaining these social networks. Special interest is taken in understanding
the networks of those fishers classified as "Coloured" under apartheid and how new constructions
of that racial and ethnic identity operate in everyday work life.
As much of cognitive anthropology and social psychology has shown, people develop
implicit models of whom they know and respect in their heads (and how these people are
connected), and these models in turn influence with whom they choose to associate. The choice
of associates becomes particularly important in the context of employment, where the structure
and content of personal relationships often determine the success in the development and
maintenance of a livelihood (Granovetter 1973, 1974; Lin and Dumin 1986; Sanders and Nee
1996). In particular, small-scale entrepreneurial enterprises thrive and survive on the size and
quality of social networks (Light and Bonacich 1988). These networks help to establish ties
instrumental to the success of the enterprise, ties that minimize the uncertainty associated with
limited institutional infrastructure. Add to this uncertainty the high-risk environment of
commercial fishing, where captain and crew depend on each other not only for a living but for a
safe return home, and social networks become crucial for personal and professional survival.
Through quantitative and qualitative analysis of these social networks I explain: 1) what kinds of
individuals are more likely to be involved in social networks of greater racial heterogeneity, 2) to
what extent individuals use inter- as opposed to intra-racial ties for procuring and sustaining
employment, and 3) what characteristics of the networks themselves lead to the greater likelihood
of full-time employment (i.e., less dependence on other sources of income).
In understanding racial and ethnic relations in South Africa, little attention has been paid
to the structure and content of social networks. With few exceptions (see De Jongh 1995)
researchers have focused on social networks as metaphor, as a heuristic device, or predominantly
in the context of formal kinship studies. But advances in methodological techniques for the
systematic study of social networks have made it possible to move beyond an understanding of
social networks as metaphor to an analysis of the structural properties of personal and
professional relationships (Scott 1991, Wasserman and Faust 1994, Johnson 1994, Degene and
Forse 1999). Yet social network analysis techniques can be even more powerful when combined
with the analysis of cultural content (Emirbayer and Goodwin 1994). While social research in the
past half century has tended to split into social structural and cultural analytical branches (among
others), these branches are still part of the same tree. A narrow focus on social structure runs the
danger of disguising human agency and obfuscating human creativity. A myopic focus on
individual experience runs the danger of ignoring larger social processes that shape human
thought and behavior. In an attempt to understand the effect of race and ethnicity on making a
living in South Africa, this project navigates between the Scylla of structural determinism and the
Charybdis of interpretive solipsism via a synthesis of social network analysis and what Clifford
Geertz called "thick description" (1973). Theoretical synthesis will emerge out of
Constructing Race and Ethnicity
For decades Anthropologists have challenged static and naturalistic views of race and
ethnicity (Sanjek 1971; Drake 1980, 1987; Brodkin 1989; Eller and Coughlan 1993; Gregory and
Sanjek 1994). Racial and ethnic identities are socially constructed in particular historical
contexts, malleable, and often overlap with other kinds of social identity (Roosen 1989). In the
right context, some people also have the ability to assume various racial or ethnic identities in
different situations (Okamura 1981). Fredrik Barth's (1969) focus on ethnic boundaries was a
necessary correction to primordial definitions of ethnicity but limited in its ability to describe
ethnicity as a product of individual consciousness and social interaction. Recent debates over the
salience of ethnicity as a social force have concentrated on how ethnic relations function at the
communal level, e.g. the relationship between ethnic conflict and the modern state (Anderson
1993, Erickson 1993, Vail 1993). Most of the latter studies define ethnicity in instrumentalist
terms (Cohen 1978). This approach is helpful to the degree that it emphasizes ethnicity as a
social, political and cultural resource to be called upon when the opportunity presents itself.
There are two main weaknesses in the instrumentalist literature: 1) the inability to describe how
ethnicity functions as a flexible form of individual identity and 2) the limited ability to describe
the structural constraints that racial and ethnic categories place on human interaction.
Instrumentalists tend to be overly dependent on studies of the elite and over-confident in the
power of rational choice (Hechter 1986).
It can be argued that much like ethnic categories, racial groupings are "imagined
communities" (Anderson 1993) in the sense that they are social inventions not fabrications. But
there are considerable differences in the styles in which different racial and ethnic groups are
imagined. Classifying the "races of man" was the raisonne d'etre of early social science.
Essentialized categories conflated biology and morality. The Boasians limited the definition of
race to the biophysical and morphological characteristics and divorced it from the learned
behavior of language and culture. But the Boasian anti-racism was based largely on
assimilationist assumptions (Baker 1998). Recent works have moved beyond these assumptions
and reinvigorated research on race in anthropology (Harrison 1995; Wade 1995; McClaurin 1996;
Mukhopadhyay 1997). One of the strengths these works share is their ability to describe racial
categorization as an historical process, deeply rooted in local culture and constructs of the
The current study is expressly concerned with the structural and cultural meaning of race
and ethnicity, concerned with how individuals conceptualize and utilize identity in relationship to
others. With one eye on the historical development of"coloured" as a socio-economic and
political category (Marais 1939; Venter 1974; Goldin 1987; Lewis 1987), the central focus of this
dissertation is on how this racial and ethnic identity is socially constructed and made useful in the
context of social networks. As we learn from the social psychological approach of Donald
Horowitz (1985) and DeVos and Romanucci-Ross (1995), ethnic identity is grounded in
individual experience and played out in social relationships. By careful observations of the
language, the dominant idioms and the stereotypes perpetuated in the context of the social
networks of commercial handline fishing captains, this study will help paint a picture of
contemporary inter- and intra-ethnic relations in South Africa.
In April of 1996 The Economist reported that people in South Africa who were classified
as coloured under apartheid feared being lost between the economic power of whites and political
power of blacks. At a time when the nature of coloured identity itself is being questioned and
redefined, those who have historically been known as coloured wonder about their place in the
new South Africa. Where do those categorized as coloureds fit into attempts to create equal
opportunity in employment for all South Africans? In the fishing industry, as with many other
industries, management has been historically white and the labor historically non-white. In the
small-scale handline industry, "non-white" specifically refers to those classified as coloured.
As a contribution to the sub-discipline of economic anthropology, this research explores
the roles race and ethnicity play in the structuring of professional networks (Gladwin 1989).
Economic activity is embedded in a complex web of social relations and institutions (Appadurai
1986). It has become a truism in anthropology that ethnic identity and racial stereotyping enable
or restrict economic success (Gregory and Sanjek 1994; Lipuma and Meltzoff 1997). Yet neither
the construction of racial or ethnic identity nor the development of economic activity can be
divorced from the relational context in which both occur simultaneously. "We should expect that
different social groups, situated in different objective conditions as to their capacity to earn a
livelihood, will have distinct experiences giving different meanings to a cultural concept that at
first might appear homogenous" (Narotsky 1997:222-223). Therefore, the use of systematic
social network analysis tools, combined with the cultural detail gathered from in-depth interviews
and participant observation, will serve to move the debate over employment discrimination
beyond the rhetoric of individual experience. At the same time, the ethnographic nature of the
other methods employed will add qualitative depth to research that, thus far, has concentrated
predominantly on aggregate labor statistics (Jiobu 1990; Uchendu 1995; Lipuma and Meltzoff
As a study of the relations of production in the commercial fishing industry this project is
forced to move beyond traditional understandings of employer-employee relations and typical
employee-wage structures (Smith 1977; Robben 1986; Bailey 1991; Smith and Hanna 1993).
The skippers and crew of small handline boats, i.e., the primary subjects in this study, do not fit
neatly into predefined economic categories. The skippers, in most instances, are self-employed
entrepreneurs. They are dependent on a wide variety of uncertainties such as the physical
environment (weather, ocean currents, tides), the condition and availability of the resource (i.e.,
fish stocks), ever-changing government regulations, large fluctuations in market demand for the
product and the health of up to twelve crewmembers. Captain-crew relations approximate
traditional employer-employee relations but rarely contain any formalized contracts.
Crewmembers are dependent on the same conditions as the captain but with the added variable
that they do not control when they go fishing. They are dependent on the captain for initiating
Figure 2. Handline Fisher Bringing a Snoek into the Boat
source: Stibbe and Moss (1998)
As an analysis of social relations in the commercial fishing industry this study also
contributes to the expanding literature on the relations of productions among workers involved in
natural resource exploitation (Palsson 1991; Durrenberger 1992, 1993; Smith and Jepson 1993;
Garrity-Blake 1994). Particular themes from this literature that will be explored include: local
participation in fisheries management (Miller 1979; McCay 1987; Acheson 1990; Ostrom 1990;
Pollnac 1991; Smith and Jepson 1993), variations in the use of technology to exploit natural
resources (Bernard 1987, Zerner 1991), the risk and uncertainty involved in fishing for a living
(Pollnac, Poggie and Vandusen 1995; Pollnac, Poggie and Cabral 1998) and the question of
fishing as a marker of identity (i.e., a fishing culture, or "more than ajob" (Smith and Jepson
1993; Eacker 1994, Garrity-Blake 1994).
Interpreting Social Structure
If there is to be a synthesis of structural and interpretive approaches to understanding
significant social constructions (such as race and ethnicity) and their impact on lived experiences
(such as job procurement and retention), then one way to begin that synthesis is by a synthesis of
research methods. There should be an intimate connection between the answers sought in a
research project and the types of questions raised. "If substance ("data," "findings," "facts") are
products of the methods used, substance cannot be considered independently of method; what the
ethnographer finds out is inherently connected with how she finds it out" (Emerson 1995:11).
This project will proceed by combining research priorities from the interpretive literature and
tools from social network analysis.
"Network analysis is a recent set of methods for the systematic study of social structures"
(Degenne and Forse 1999). In describing the broad streams in the types of network approaches
used in anthropology, Jeff Johnson notes that there is a major distinction in the literature between
those who use the concept of social networks metaphorically (Walsh and Simonelli 1986;
O'Conner 1990) and those who use social networks as a formal/analytical research tool (Johnson
1994). This study intends to move beyond understanding social networks as a heuristic concept
to a more detailed and systematic approach to collecting and analyzing information about these
networks. By following this analytical approach to understanding social networks this project
stands with one foot in a tradition that can be traced back to the works of Durkheim (e.g. the
concept of organic solidarity) and the mid-century structural-functionalism of A.R. Radcliffe-
Brown. Network analysis owes a more immediate tribute to the works of John Barnes (1951;
1969; 1972), Clyde Mitchell (1969), Elizabeth Bott (1971) and Ron Burt (1982), to name a few.
Network analysis is used as an inductive attempt to identify patterns of relationships and the
behaviors or thoughts that correlate with those patterns. "Then it sorts out aposteriori and
identifies the concrete constraints of structure on behaviour at the same time as it uncovers
constraints on structure from group interactions" (Degenne and Forse 1999:2-3). The primary
focus is not on the attributes of individuals but on the relationships between individuals.
Social network studies can be divided into two basic types: ego-centered (also called
personal or partial networks) and whole network (also called global networks) approaches.
Whole networks are an "abstraction of the overall, 'global' features of networks in relation to a
particular aspect of social activity" (Scott 1991:31). For a review of anthropologists studying
whole networks see Johnson (1994). The present study is concerned to a limited degree with
whole networks. More centrally this study concentrates on the networks that are "anchored
around a particular individual so as to generate 'ego-centered' networks of social relations of all
kinds" (Scott 1991:31). Studying the entire network of handline skippers will allow for the
analysis of group formation. While ego-network data cannot provide accurate descriptions of the
overall social structure of a population, this approach "gives representative samples of the social
environments surrounding particular elements and is compatible with conventional statistical
methods of generalization to large populations" (Marsden 1990:438). It is my intention that the
summary measures on the ego-network data and the interpretive ethnographic analysis will
complement one another for a fuller explanation of the effect of race on social relations in the
One of the weaknesses in this study is its inability to measure how networks change over
time. By nature of its cross-sectional design this study is forced to deal with the charge of static
bias. Social structures such as personal networks change over time, but the description and
prediction of such change is beyond the scope of this project. This project can serve as a baseline
for future diachronic analysis. However, for analysis of how current conditions relate to
conditions in the past I will rely on generalized comparisons with the historical literature on race
relations in South Africa (James 1991; Bekker 1993, Vail 1993; Fredrickson 1995; Pickel 1997).
A sample of first order zone cognitive network ties served as a proxy for understanding
the typical social relations in which commercial skippers were involved. "First order zone" refers
to a person's direct contacts, someone they interact with personally (Barnes 1969). The specific
focus will be on cognitive networks or the way people construct their social universe in their
heads. The measurement of actual exchanges is beyond the scope of this project. Although key
research has shown that informants are relatively inaccurate in detailed recall of their exact social
interactions (for a summary see Bernard et al. 1984), the bias in informant inaccuracy is towards
typical interaction (Freeman and Romney 1987, Freeman et al. 1987). In-depth interviews will
make it possible to add historical detail to such typical interactions (e.g. how long an alter is
known by an ego and in what capacity/ies).
Research Questions and Hypotheses
The following research questions and hypotheses structure the content for this dissertation.
1) What kinds of individuals are more likely to be involved in social networks of greater racial
HI: Education will have no significant effect on the racial heterogeneity of social networks.
H2: Age will have no significant effect on the racial heterogeneity of social networks.
H3: People with a racially heterogeneous friendship network will be more likely to also have
a racially heterogeneous professional network.
H4a: People categorized as historically disadvantaged will be more likely to have racially
heterogeneous professional networks.
H4b: People categorized as white/European will be less likely to have racially heterogeneous
H5: There will be no significant difference in the racial heterogeneity of friendship networks
between different racial categories.
The racial heterogeneity of a friendship network will be measured as the proportion of same-
race ties as determined by the skippers' answers to the friendship name generators 1-3. If they
categorize 40% or more of their close friends as racially different from themselves, their
friendship network will be considered heterogeneous (which is slightly more liberal than Ooka
and Wellman's use of 50%) (Ooka and Wellman 1999). A person's professional network will be
considered racially diverse if 40% of the groups of professional contacts they have are diverse.
Each group of contacts will be considered racially diverse if ego judges 40% or more of those in
the group as racially different from themselves. Professional groups for skippers include: Factory
Ownership (FO), Factory Management (FM), Factory Labour (FL), Boat Owner (B01 & BO2*),
Boat Skipper (BS 1 &BS2*), Boat Crew (BC 1 & BC2*), Administration Local (AL, ministry,
staff), Administration National (AN, ministry, staff), Administration Research (AR), Other (0).
The differentiation denoted by is: 1 =own boat, 2=other boat.
One of the strengths of the ego-centered approach to network analysis is the ability to
combine the analysis of individual traits with analysis of the kinds of relations that exist between
individuals. It is often assumed that education has a liberalizing effect on perceptions of
difference and a consequent increase in inter-racial relationships. This assumption will be
challenged by analyzing how education relates to the racial diversity of friendship and
professional networks, proving that, for the range of education common among commercial
fishers, education will not impact the racial heterogeneity of social networks.
Works on the ethnic solidarity of migrants have shown older migrants are more likely to
build densely knit, tightly bounded social networks (Light and Bonacich 1988). But following
Portes (1995) it will be argued here that age itself will not be a significant determinant of the
racial heterogeneity of social networks. It will also be argued that those whose friendship
networks are racially heterogeneous will be more likely to have racially diverse professional
networks. The reverse is not assumed to be true. Historically disadvantaged racial and ethnic
groups in South Africa (e.g. black or coloured) still face systemic racism. Therefore, persons
from these groups will be more likely to use diverse networks for their success. Conversely,
persons from historically advantaged communities (i.e., white) will be more likely to use densely
knit, tightly bounded social networks to maintain benefits they already have. I believe that
friendship networks are more socially conservative than professional networks and will thus be
more homogenous regardless of how ego classifies himself and others.
2) To what extent do individuals use inter- as opposed to intra-racial ties for procuring and
H6a: Members of historically low status racial groups will be more likely to use inter-racial
ties than intra-racial ties for obtaining employment.
H6b: Members of historically high status racial groups will be more likely to use inter-racial
ties as opposed to intra-racial ties for obtaining employment.
H7: Tenure in the fishing industry will be positively associated with the use of inter-racial
Building on the work done by Ooka and Wellman (1999), it is assumed that people make use
of both inter- and intra-racial ties for finding (and keeping)jobs. "The advantages (or
disadvantages) of working in an ethnic economy or ethnic niches depend on the resources that
particular ethnic groups can mobilize through their co-ethnic networks" (Ooka and Wellman
1999:3). Inter-ethnic ties have been shown to be advantageous, particularly for developing
entrepreneurs (Cobas et al. 1993). Butjob seekers and entrepreneurs can also increase their
opportunities by making use of important ties outside "co-ethnic networks," particularly with
persons of higher status ethnic groups.
Network analysis name generators and interpreters formed a major part of an in-depth
interview schedule administered to 102 commercial handline skippers that live and travel from
Struisbaai (south east of Cape Town) to Lambert's Bay (north west of Cape Town), a geographic
range determined by the residence of those in the network. Each captain was interviewed at the
place of their choosing, most often in their homes. Where possible, private time was reserved for
the lengthy interview with the skipper alone. The interviews with the skippers lasted a minimum
of one hour and more often than not ran longer than three hours. Consent was verbally requested
from each skipper prior to the formal interview (see Appendix D).
The method of subject selection for the study is drawn from the social network analysis
literature. The research design is cross-sectional where "data are collected at one point in time
from a sample selected to describe some larger population at that time" (Babbie 1990:56). This
design will be used to provide for in-depth description of current social relations. It will also
allow for the determination of the relationship between individual demographic characteristics
and the racial composition of social networks and for the determination of the degree to which
certain types of individuals use inter- as opposed to intra-racial ties for procuring and sustaining
employment. Analysis of how these conditions relate to conditions in the past will rely on more
generalized comparisons with the historical literature on race relations in South Africa (James
1991; Bekker 1993, Vail 1993; Fredrickson 1995; Pickel 1997).
The sampling frame used for the study is a subset of a list of all fishing boats registered
on South Africa's Western Cape Coast. Significant data collection and analysis on the
commercial fishing industry are published annually by the editors of the Fishing Industry
Handbook: South Africa, Namibia and Mozambique (Warman 1999) in cooperation with the
national fisheries regulation body, Marine and Coastal Management (formerly Sea Fisheries).
Information in this data set includes the boat names, owner names and contact details, length of
the boats (among other detail), number of crew on each boat and an independently researched
category for the type of fishing done on the boat. Information from this list proved helpful in
making initial contacts, but personal confirmation of the information from these sources
highlighted the inaccuracies in the national database. The lists could not distinguish between full-
time and part-time commercial fishers, nor did it carefully distinguish between those operating
mobile ski-boats and those operating the less mobile "chakkies". Using the government
information as a starting point, informants were selected by snowball sampling from confirmed
cases. Those finally selected were all full-time, commercial handline ski-boat skippers that
traveled the coast following the fish and were actively fishing between September 1999 and
August 2000. Skippers of the more traditional handline chakkies were omitted due to the
profound difference that geographical range makes on social networks. Chakkies remain moored
in their port of origin and are not transported from harbor to harbor in search offish. From
preliminary interviews, subsequently confirmed by the ski-boat skippers interviewed, the social
networks of skippers on the chakkies and those on the ski-boats do not display much overlap.
The networks and experiences of active commercial skippers of handline vessels served
as the fulcrum of analysis for this study. The type of boat and fishing is held constant so as to
control for variations associated with the type of fishing done. The range of the boat length and
number of crew will also be limited to increase the likelihood of comparable units. As expected,
crew size correlates with boat length (86% for the west coast boats). The study was limited to
skippers of handline boats with twelve crew or less. These skippers serve as the primary units of
analysis. As will become evident, part of the analysis relies on the skipper and crew as a unit.
Handline boats tend to be relatively small, independently owned operations dependent on
informal social and economic ties for their success. This lack of formal restrictions on a skipper's
choice of associates increases the likelihood that the associations are voluntary.
The initial cut in the data for selection of a sampling frame involves limiting the
geographical range. Subjects were selected by snowball sampling (Babbie 1990), beginning with
preliminary interviews of the skippers on commercial fishing boats registered in the ports ranging
from Lambert's Bay in the north to False Bay in the south (Warman 1999:243-284). The reasons
for this first cut are both theoretical and practical. The commercial fishing industry of the western
Cape coast serves as the regional and industrial context for this study in part because this industry
has historically been dominated by coloured workers (Marais 1939; Venter 1974; van Sittert
1992; Stibbe and Moss 1998). As was evident from statistics gathered during the 1994 elections,
over 50% of voters living in the Western Cape Province identified themselves as coloured (Pickel
1997). The effects of the changes in South African race relations are of pressing concern to these
voters (James 1996). Relationships in the commercial fishing industry are highly likely to involve
employees who identify themselves as coloured.
From preliminary personal interviews with key informants it seemed, and was later
confirmed, that many of the commercial handline skippers who were registered in the ports of
interest concentrated their range of fishing between Lambert's Bay and False Bay. This is a
reasonable range for skippers at the geographical extremes to travel in search of prime fishing
(approximately two hours of drive time from Cape Town in either direction along the coast).
This range also makes sense in light of the need for proximity to the lucrative Cape Town
markets and the putative heavier concentrations of snoek7, the primary species for commercial
handliners on the western coast (Van der Elst 1981, Griffiths 2000).
Harbor masters, factory owners and managers, national local Marine and Coastal
Management officials and other key informants were used to confirm the identities of the
appropriate skippers in each region. Over the course of the first five months of initial research,
from September 1999 through January 2000, 1 identified approximately 155 skippers of what
were identified as "full-time commercials" or active handline ski-boats in the study area. I
continued to add to this list when one of the skippers mentioned full time handline ski-boats not
already on the list. It became my intention to interview the skippers from each of these 155 ski-
boats. After sorting through some errors in informant recall, boats that had been sold or were no
longer in service, and skippers who had moved, 118 active full-time commercial handline skipper
remained. Of these, 102 were interviewed, four declined when asked to participate, and twelve
could not be reached for logistical reasons. I confirmed the status of each of the 102 interviewees
as full-time commercial handline ski-boat skippers with at least three independent persons, at
least two of which were other skippers. As there were significant tensions between full-time and
part-time/recreational fishers, skippers quickly categorized others in one of the two camps.
Where there was uncertainty, I gained confirmation from the skipper himself.
The typical procedure for collecting ego-centric network data is to elicit alters (persons
tied to ego) via one or more name generators. Name generators are specific questions designed to
elicit a list of individuals with whom the respondent has direct ties of a specific kind. Additional
information about the alters is then generated through name interpreters. Marsden categorizes
three types of name interpreters: "(a) reports on attributes of persons or alters enumerated (e.g.
age, education, race/ethnicity); (b) reports on properties of the tie between respondent and alter
(e.g. frequency of contact, duration of acquaintance, intensity); and (c) reports on the intensity of
ties between pairs of alters" (1990:441). Research has shown that the demographic
characteristics of alters can be reported with substantially greater accuracy than the attitudes of
alters (Bernard et al. 1985). Projection plays a part in responses to questions asking for proxy
reports on attitudes, particularly for more distant ties (e.g. friends as opposed to spouses) (Wilcox
and Udry 1986). Therefore, respondents were not asked to divulge what they thought any of their
alters thought or believed.
The name generators used in this study were designed to elicit the perceived
characteristics of alters. They also focused on the perceived strength of ties between ego and
alters. In terms of the properties of ties between ego and alter (e.g. frequency of contact, duration
of acquaintance, intensity), respondent reports are generally in concordance with alter reports,
particularly for close ties and reasonably general types of interaction (esp. frequency of contact,
duration, kinship and intensity of relationship) (Hammer 1984). A combination of the intensity,
frequency and duration of ties between pairs of alters has been shown to be the strongest indicator
of tie strength (Marsden and Campbell 1984).
1) From time to time, most people discuss important matters with other people, people
they trust. The range of important matters varies from person to person across work,
leisure, family, politics, whatever. The range of relations varies across work, family,
friends and advisors. If you look back over the last six months, who are the four or
five people with whom you discussed matters important to you?
2) Consider the people with whom you like to spend your free time. Over the last six
months, who are the three or four people you have been with most often for
informal social activities such as having a potjie or braai together, having drinks
together, going to films, visiting one another's homes, and so on?
3) Who would you say are your closest three or four friends? This may or may not be
the people you spend the most time with.
4) Of all the people working in the commercial fishing industry on the western Cape
coast, who are the five or six people who have contributed most to your success in
fishing-i.e., your most valuable work contacts?
5) Suppose you had a friend who wanted to do what you are for a living. Who are
the most important people (that you know personally) you would introduce them to
who would give them the best information and advice? Who do they really need to
get to know?
6) Are there any individuals you regard as a mentor, someone who has taken a
strong interest in how well you do as a fisherman and has provided you with the
opportunity or means to do better?
7) Of all the people you know in the industry, who has made things most difficult for
you to do well in what you do? Remember that all names are coded and kept
confidential and will not be released from my research except as combined statistics.
8) If you decided to find a job in the commercial fishing industry, who are the two or
three people with whom you would most likely discuss and evaluate your job
options? These could be people who work with you now, or people from other than
where you work now such as friends, family, people who work on other boats or other
people in the fishing industry.
9) When it comes to information you need that will enable you to catch more fish or get
higher prices for your fish, who are the skippers that you contact for important
information (e.g. where the fish are biting, what prices hawkers are paying on the open
market, recommendations for new crew members, etc.)?
10) If you wanted to have a friendly but business related sit down chat with the head
of Marine and Coastal Management, who would be the people you contacted to set it
up for you?
11) Look over this list of names. Can you name anyone else whom you would
consider as important to you as the people you see on this list?
The preceding name generators have been adapted for this study from the works of Fisher
(1982), Burt (1984, 1985, 1997), Rook (1984), Leffler, Krannich and Gillespie (1986), Kochen
(1989), Wegener (1991), Podolny and Baron (1997). One of the influences in developing these
name generators is the finding that multiple name generators from multiple categories of relations
elicit a broader social network (Burt 1997). By asking about personal and professional ties I was
able to look for evidence of the overlap between the two types of networks. If social capital, for
example, is a product of network ties, then "the strongest evidence of social capital occurs when
personal and corporate relations together define the network used to measure social capital" (Burt
Name generators one through three are focused on the individual's personal network,
irrespective of the social context. Name generators four through ten focus attention more
specifically on contacts related to employment. The first of these name generators is taken
directly from Podolny and Baron's adaptation of the network name generator on the 1985 General
Social Survey (GSS) (Burt 1984, 1985; Podolny and Baron 1997) and is designed to sample ego's
discussion network. The generators that follow are designed to elicit ego's #2) socializing
network (Burt 1997:359), #3) intimate friendship network (Fisher 1982), #4) most valued work
contacts (Burt 1997:359), #5) general employment info and advice (Podolny and Baron
1997:691), and #6) mentorship networks (Podolny and Baron 1997:692). The seventh generator
is included to elicit conflictive ties, since some studies of social support suggest that the absence
of non-supportive ties is more crucial than the presence of supportive ties (see Rook 1984;
Leffler, Krannich and Gillespie 1986). Name generator eight is designed to elicit contacts for
ego's alternative options in the industry. Name generator nine is to elicit ties that provide specific
strategic information for ego's success as a skipper from other skippers in their category. It is
designed specifically to tap co-worker networks, people in similar positions. Name generator ten
is designed following the first step in a small world experiment. It is designed specifically to
measure ego's proximity to the most important decision makers in the regulation of their industry
(see the contributions in Kochen 1989). Name generator eleven was created because of Burt's
findings that his respondents, having reviewed their name lists, added names to the end of their
list but did not place these names in any of the generator categories (1997:359). Name generator
twelve was generated to delve into the skipper's knowledge and perspective of his crew.
Name Interpreters (see Appendix B):
sex = male (0); female (1)
age = actual age in years
race = ego's perception of what they would have been categorized under apartheid
reside = the town/neighborhood they live in (smallest geographical unit known)
langl = primary language spoken at home:
lang2 = secondary language spoken at home
()= How often do you communicate with this person: 1) Daily; 2) Weekly; 3) Monthly;
4) Less than Monthly
duration = How long have you had a relationship with this person?
role = What does this person do in the fishing industry, if any?
Factory Ownership (FO); Factory Management (FM); Factory Labour (FL); Boat
Owner (BOl & BO2*); Boat Skipper (BSI &BS2*); Boat Crew (BCI & BC2*);
Administration Local (AL; ministry; staff); Administration National (AN; ministry;
staff); Administration Research (AR) Other (0) <* l=own boat; 2=other boat>
family = How is this person related to you, if at all?
(spouse, mchild, child, hmparent, hfparent, wmparent, wfparent, sibling,
wsibling,, hmparentsib, hfparentsib, wmparentsib, wfparentsib, grandparent,
wgrandparent, other adult, other child)? (m=male; f=female; h=husband; w=wife;
intimacy = How close do you consider this person to you?
1) Distant (avoid contact unless it is necessary) ; 2) Less close (Ok to work with, no
desire to develop friendship); 3) Close (close, but not one of the closest contacts); 4)
Especially close (one of your closest contacts)
religion = Do any of the members on your list attend the same religious congregation you do?
job = Rank which of these people were most responsible for you getting the job as
skipper? (top three)
In addition to important demographic characteristics such as sex, age, putative race, area
of residence, marriage status and language preference, each skipper was asked to name
characteristics of their relationship to the alters they name. Frequency, duration and intimacy are
all indicators of tie strength (Marsden 1990). Role and family are included as traditional
structural categories to be correlated with tie strength indicators. These are often found to be
strong predictors of tie strength (Marsden and Campbell 1984; Marsden 1990). Religion is
included as a measure of socialization.
The Structured Interview
The central tool for primary data collection in the project was a detailed interview
schedule consistently applied across all interviewees (see Appendix A). The structured interview
included both closed and open-ended questions, tables, Likert scales and important follow-up
probes. At no time was the interviewee asked to fill out information for himself. The questions
were organized into the following categories: Demographic and Personal History, Household, On
the Job, Networks, Women in Fishing, Gear, Annual Round, Crew, Regulations/Licensing, Future
of Fishing, Job Attachment and Job Satisfaction.
Demographic and Personal History questions were designed to elicit general sociological
information for comparison within group and to other populations. Question four was
particularly central to the research project as it asked the interviewee to identify under which race
or ethnic classification he was identified during apartheid. Due to the contested nature of racial
and ethnic identity in South Africa, particularly for those once classified as coloured, I asked
directly about this categorization on two different occasions. Later in the interview, question
thirteen, I asked the interviewee to explain what the classification they were once assigned meant
to them in the present.
Questions about the household were primarily targeted at the contribution that fishing
makes to the household income. After a few preliminary interviews and advice from others who
had interviewed in the fishing industry I decided not to ask for gross or net income figures. With
the informal nature of these businesses, skippers were constantly wary of direct financial
questions. The government was in the process of formalizing their sector of the fishing industry.
Despite the consent protocol that clearly identified who I was, what my intentions were and how
they were protected, I was constantly queried as to whether I was representing either Marine and
Coastal Management or the tax auditor.
I followed questions about the household with questions designed to elicit details about
their experiences on the job. One of the more important questions in this section was the first; it
asked skippers to describe the typical fisherman. This question was designed to elicit the
attributes that fishers have of themselves and others like them. It was also designed to give the
interviewees an opportunity to project their own cognitive models of who is who in the industry.
This section included questions about their history in the fishing industry.
The network questions formed the middle third of the interview schedule, and
approximately a third of the interview time was spent on these questions. The questions are
discussed in detail above. Names were first generated on the Name Table (see Appendix),
numbered by the name generator used. If names were repeated, they were written again,
preceded by the corresponding name generator. No strict limit was given on the number of alters
each skipper could mention in relation to any of the name generators. After names were
generated from all twelve generators, the interviewer and interviewee went back over the list
together to answer the name interpreter questions for each of the alters mentioned. Interpreter
information was filled out for the first time the alter's name appeared on the list.
In preliminary interviews handline fishing was constantly described by men and women
alike as "a man's world," "not a place for a woman." Women were clearly involved in the
handline fishing industry in bookkeeping, marketing, sales, household management, running
errands, preparing food, and often providing a consistent income and benefits with formal
employment, but the network analysis design of this study precluded an in-depth analysis of
women's perspective on or input into handline fishing as a business. The questions about women
in fishing were designed to explore these skippers' stereotypes of women with an eye toward
understanding how and why men construct fishing as gendered work. If women's voices are
muted in this study, it is because these voices are muted in the conversations of the all-male
The size and technological sophistication of the gear used for fishing is a good indicator
of the capital each owner has available to invest in their handline ski-boat business. I assumed
there would be much more variation in technology than what I actually found. Questions in the
Annual Round section were designed specifically to inquire about the type of fishing engaged, the
formal or informal seasonal patterns of fishing, and the geographical movement of this highly
mobile class of fishermen.
Open-ended questions about a skipper's perspective of his crew were designed to allow
interviewees to project their own categories of class and race onto those who work for them. As
most were quick to make a categorical distinction between skippers and crew, these questions
became important in determining how skippers viewed relations of power, specifically with
relation to class and race.
Local and national political struggles over the access to fishing licenses and quotas were
on the top of everyone's agenda for discussion. I intentionally left questions about the
government management system until near the end of the interview so as not to allow such topics
to dominate the entire interview. Responses to these questions tended to be long and passionate.
With the passage of the new Marine Living Resources Act just two years prior and the
implications of that act for handline fishers still not legislated, quotas and the boat licensing
system were pressing concerns.
I ended the interview by having skippers assess where they had come from and what
future they saw in fishing. Optimism for the future, job attachment and job satisfaction all
contribute to mental health. Questions here were designed to explore what skippers thought of
the future of the fishing industry and their place in it. I also wanted to know how attached they
felt to the kind of work they were doing and whether they had searched out other options.
Finally, I was concerned with how satisfied these skippers were with important aspects of what
they did for a living.
With the time-intensive nature of the detailed interview schedule and the geographical
scope of the social network, it became necessary to hire research assistants to increase coverage
of the network. Three research assistants were hired from June through August 2000. Each
assistant was trained in interview techniques and the specific issues important to the project.
Each of the three assistants was an experienced interviewer in fishing communities. They were
referred to my project by the coordinators of the Subsistence Fisheries Task Group (2000), a
research project completed in February 2000. After listening to samples of completed interviews,
I held three follow-up sessions with the assistants to discuss their progress and to clarify
interpretation of the questions. See the Research Assistant contract (Appendix C) for details of
SI use the term "re-directed" to acknowledge postmodernism's debt to earlier methodologies
such as what Harris calls the Historical Particularists or the Boasians (Harris 1968).
2 "The reader will recognize too, especially in my questions, a psychoanalytic orientation that I
have found impossible to eliminate, so embedded is this orientation in contemporary Western
thought" (Crapanzano 1980:10).
3 See, for example, Crapanzano's interpretation of Tuhami's psychological servitude to the
female spirit "A'isha Qandisha" (1980:71-72).
4 An example of an ethnography where this reflexivity is taken to an extreme is Ruth Behar's
Translated Woman (1993).
5 As a metaphor, the picture drawn here should appeal to the postmodernist. As an analogy from
modern physics, the picture drawn here should appeal to the scientist.
71 am indebted to Marc Griffiths of the Sea Fisheries Research Institute in South Africa for all of
the information on this important southern African species. He generously provided a pre-
publication copy of his latest work entitled: Life history of South African snoek Thyrsites atun
(Pisces: Gempylidae): a pelagic predator of the Benguela Ecosystem (Griffiths 2000). Specific
information on snoek throughout this dissertation has been taken from this paper.
THE "SO-CALLED" PEOPLE:
COLOURED IDENTITY IN THE NEW SOUTH AFRICA
Identity in the Context of Oppression
In order to do justice to the complexity of issues surrounding coloured identity, it is
important to clarify the language used. Language is an integral part of culture, of cultural change,
of ethnic identity and of identity politics. The words we choose both reflect and affect our
perceptions and attitudes. Language colors our understanding. Our words often reflect the power
relations in society. The same term can be used for directly opposing purposes. This is
particularly true of terms used to distinguish one people group from another. The term
"coloured," as used at present when referring to a particular people group in South Africa, is, on
the one hand, saddled with the static baggage of South Africa's apartheid past and, on the other
hand, is a category fluid enough to include a host of new identities in a rapidly changing social
landscape. The ambiguity of the issues surrounding coloured identity has lead many academics,
journalists and popular writers to euphemistically refer to this group of people as the "so-called
coloured," invariably with "coloured" in quotation marks.
Race and ethnicity are both historically conditioned, socially constructed attempts to
distinguish one group of people from another by reference to each group's unique interaction
between ideological commitments and material conditions. Self-identification with an ethnic or
racial group involves an ever-changing process of identity formation in which certain individual
psychological and material needs are met (e.g. sense of belonging, inheritance of land) and
whereby collective (social) mobilization can take place to further a common interest. Persons of
the same ethnic or racial group may identify themselves (or have been identified by someone
else) with particular physical characteristics. Inclusion in a particular ethnic or racial group
usually involves a shared sense of history or identification with a common past. The survival of
the group depends on a commitment to preserving a certain level of distinctiveness of the group.
Primarily due to its use as a tool of oppression and domination, many South Africans
have rejected the label "coloured." Those who have been classified as coloured by successive
South African governments have a history of diverse origins, lack of social, economic or
phenotypic homogeneity and an increasing heterogeneity of residence, religion and political
affiliation. Legislative definitions, as exemplified in the Nationalist's Population Registration
Act, relied on arbitrary and subjective definitions of physical appearance. The two major
government commissions of inquiry into the "Coloured Question," the Wilcox Commission in
1938 and the Theron Commission in 1976, failed to agree on a substantial definition (Goldin
1987; Lewis 1987). Some will argue that there is a separate coloured culture. Some will argue
that "coloured" is solely a term used by the white-supremacist state to preserve racial purity in
support of its social evolutionary ideals. I will make the argument that the history of coloured
identity in South Africa involves a very complex set of variables, a complex interplay between
exclusion and cooptation, between collaboration and resistance, between centripetal and
centrifugal forces from within and without. In the rural fishing communities on the Western Cape
coast, and in most of the suburbs of Cape Town, questioning "coloured" as a viable category of
identity is purely an academic exercise.
Ian Goldin expressly focused on the political mobilization of coloured identity, i.e., the
"articulation and representation of coloured identity in organizations and institutions" (1987:xv).
Although he acknowledged that a full understanding of coloured identity would require a larger
study that would include studies of "religion, culture, language and other complex psychological
manifestations of identity" ( IU .. I h ii work reflected the dominant stream of literature,
placing ethnic identity primarily in its political and macro-economic context. Goldin noted that
he was not interested in exploring "the "lower levels" of coloured identity, which relate to often
inarticulate and hidden expressions of identity" (1987:xv). Although political and macro-
economic factors cannot be ignored, few attempts have been made to connect the political and
economic implications of ethnicity to these "lower levels." This dissertation is one such attempt.
Richard van der Ross, a self-identified coloured, founder of the Labour Party and scholar
of South African history, defended the use of the term "Coloured" as a word distinguishing a
particular South African social group. However, it is with strong reservations that such self-
identification was made. He believed that "Coloured People" was the least objectionable term
and the one least likely to lead to confusion as it dominated the historic literature. To him the
terms were not as important as the meaning behind the terms. Van der Ross attempted to flush
out some of the destructive and erroneous meanings associated with coloured:
1. All Coloured people have the same origin.
2. Coloured people are easily recognizable.
3. Coloured people have their own culture.
4. All Coloured people are "the same," so that they "belong together."
5. Coloured people prefer to be together.
6. Even if they do not prefer to be together, it is better that they are together
so that the "better class" can uplift the others.
7. Their own identity must be protected at all costs. Intermarriage and
improper contact with other population groups should be avoided and
8. Coloured people are a separate nation, or a nation in process of becoming
9. The natural consequence of their being (or becoming) a nation is that
they should have their own "homeland."
10. Coloured people must be "protected" against any tendency to pass for
White, or to marry or cohabit with Blacks.
11. It is right and good that Coloured people should have their own identity.
12. Unless this identity is accepted, protected and developed, Coloured
people will not be fully developed or gain their rightful place in South
13. The alternative to the theory of Coloured identity is integration with
either Whites or Blacks, and both these alternatives are unacceptable.
14. For economic purposes it is absolutely essential that Coloured people
accept the concept of their own identity, or they will suffer economic
ruin due to labour competition from Blacks, and the entrepreneurial
superiority of Whites and Asians. (Van der Ross 1979)
Van der Ross' primary purpose in writing Myths and Attitudes was to expose the misconceptions
surrounding the "myth of Coloured identity" (1979).
Marais reported that the Khoi were readily conscripted into the labor force of the settlers,
readily traded their cattle for copper, beads and tobacco and were particularly vulnerable to the
"tot system." The tot system was an attempt by land owners to "encourage" their laborers by
supplying them with wine, or preferably brandy. Marais then traced this practice to perceptions
of the coloured people of the late 1930's: "The habit of drinking to excess, implanted from
generation to generation, is still one of the besetting sins of the coloured People" (1939:3).
Vernon February was concerned with highlighting and challenging the stereotypes of coloured
identity found in South African literature. He noted in particular how the definition of"coloured"
was most often expressed in the negative, i.e., by what it did not mean. "The stereotype of the
present-day 'coloured' draws, I venture to say, on a fairly continuous tradition starting with the
depictions of Khoi in literature" (February 1981:23). February examined one of the first dramatic
works to be written in Dutch-Afrikaans: De Temeperantisten written in 1832 by E. Boniface.
February argued that in Afrikaner culture the stereotypes placed on the "Hottentot" were often
automatically transferred to the coloured. "In general, then, writers portray the 'Hottentot'
characters as care-free, comical, witty, loud-mouthed, fond of liquor, and prone to fighting easily"
(February 1981:26). It is difficult to tell, however, which has been more damaging to coloured
people and their identity: the liquor or the stereotype.
Stereotypes were also appropriated by coloured people themselves. The struggle over
coloured identity created complex internal social relations. For some who came to accept the
coloured identity as their own, the boundaries of coloured identity often became less fluid. They
saw those who were trying to "pass themselves off as whites" and called them "play-whites."
February refers to the practice of"venstertjies kyk" (literally: looking in the windows), which
refers to what happened when coloured friends or relatives see the "play-whites" approaching.
They pretend to be window-shopping in order not to embarrass the person or relative in question
(February 1981:198). Fredrickson argued that, at the end of the nineteenth century, this tendency
to "pass" hindered the efforts of those who wished to politically mobilize coloureds. "It was so
easy for successful and relatively light-skinned people with nonwhite ancestry to pass over into
the European population that it was difficult for group-conscious leaders to emerge" (Fredrickson
1995:46). Yet early in the twentieth century such a consciousness did emerge, institutionalized
by the founding of the African Political Organization (APO) in 1902.
The controversial issue of miscegenation (race mixing) cannot be ignored as a factor in
the development of coloured identity. The terms 'mixed race' and 'mulatto,' employed often by
American media and academics, are often considered offensive by South Africans because these
terms imply miscegenation (Morris 1992). The myth of a pure "race" has always marginalized
people of mixed heritage. Although primarily a biological category, miscegenation has
historically been associated with illegitimacy. The struggle for coloured identity is a struggle for
legitimacy; it is the struggle for a self-affirmation of legitimacy, a struggle for legitimacy in the
eyes of all South Africans and in the eyes of the world. Yet the history of coloured identity
reflects, to some extent, the development of all cultures. The development of coloured identity
and its complex history challenge static notion of culture and point toward the radical
interconnectedness of all people, despite phenotypic distinctions.
Comparative historians have argued that a general strategy for managing race mixture
usually develops early in a multi-racial society. As George Fredrickson put it, "The anarchic
nature of the human libido has always created serious problems for guardians of ethnic
boundaries and privileges" (1981:94). The earliest legislation passed in both the United States and
South Africa that discriminated based on ancestry included laws restricting inter-racial sex or
marriage. Legislators then had to deal with complex decisions over what should be done with the
offspring of inter-racial unions. The United States tended to develop a binary, polarized system
whereby descendents of mixed unions were classified with their black progenitors, often
monitored by an arbitrary assumption of biological impurity referred to as the "one-drop rule." In
other parts of the world, other post-colonial, post-slavery nations developed more complex
categories. It is difficult to prove that Brazil's well-known struggle over how to classify mulattos
has resulted in a polarized system of racial classification (Harris 1964, 1993). Although the
difference in socioeconomic status between mulattos and blacks in Brazil is insignificant in
comparison with the relative privilege of whites, gradations of racial discrimination remain
evident (Marx 1998).
The South African case does not fit neatly into a typology of nations that include bi-polar,
tri-polar or graded-scale racial orders. As in the United States, race mixing in South Africa was a
problem that required a solution. The solution ultimately came in the form of government
proclamations that varied over time and eventually never explicitly defined who was coloured
and who was not. For a time, white South Africans bolstered their political numbers with
coloured allies partially out of fear of being outnumbered. But eventually the threat of being
outnumbered was overcome with draconian laws protecting white privilege. Apartheid
eventually destroyed even the marginal advantages that coloureds had gained under the promise
One of the difficulties in determining the history of ethnic identity of the coloured is that
very few history books were written by those who would consider themselves coloured. If our
perceptions and understanding are influenced by the language we use, then history itself is
colored by the limited perspectives of the historians. Even if historians attempted to be as
"objective" as possible, they necessarily rely on documents that themselves have been written
mostly by non-coloureds. Although the standard histories of the past may not intentionally
contain false information, without significant input from coloured voices the histories remain
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E1YD69F1T_H7NVAY INGEST_TIME 2014-04-19T00:19:42Z PACKAGE AA00017636_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC