M. G. SMITH
X Extra-Mural Department
UNIVERSITY COLLEGE OF THE WEST INDIES
THIS VOLUME HAS BEEN
BY THE UNIVERSITY OF
A FRAMEWORK FOR CARIBBEAN STUDIES
THE EXTRA-MURAL DEPARTMENT
UNIVERSITY COLLEGE OF THE WEST INDIES
t. A. 1
jf D F.C-
Introduction .. 5
Context . 6
Afro-American Research 11
The Folk-Urban Continuum .. 26
Psychological Studies & Interpretations 32
Stratification & Differentiation .. 38
Colour in the British Caribbean .. 51
Analytic Models for Caribbean Societies 57'
Conclusion .. .. .. 65
SReferences ... 66
Systematic social study of the British West Indies is a recent
development, hence the slenderness of our sociological literature,
and its dependent character. This dependent character reflects the
fact that hitherto most of the researches in this area have been
conducted by visiting social scientists from America or Britain,
and have been guided by theories and themes of interest developed
in studies of societies and cultures outside the British Caribbean.
The resultant diversity of approaches has undoubted value for the
systematic study of British West Indian society, as this diversity
directs attention to a wide range of problems and aspects of local
life. On the other hand, these researches have an ad hoc, explora-
tory character, and require careful sifting and collation if they
are to form the background of a systematic programme of area
studies. Yet it is patent that to build soundly and quickly, we
must use the old foundations, testing them first, and then assimilat-
ing all that proves useful and valid into the newer structure. The
present paper is intended as a partial contribution toward that
It has not been entirely fortuitous that social studies of the
British Caribbean reflect theories and themes of interest devel-
oped in the study of societies outside this area. This derivative
character of Caribbean sociology partly reflects the comparative
character of social science in general, and partly the fact that the
British Caribbean has many elements and patterns which have
been found and studied in a wider area. It is therefore necessary
to consider those studies conducted within this wider area which
have special significance for social research in the British Carib-
From the outset, Caribbean research is faced with problems
of frames of reference. These are implicit in the dual bases of
affiliation already mentioned. In the first place we have to face
the problem of the appropriate geographical frame of reference
for such studies; that is to say, we shall have to delimit the area
which forms the natural comparative context of Caribbean social
research, and to define its most important characters. In the
second place, we have to develop a system of concepts, orienta-
tions, and hypotheses, that is, a theory, which can act as an ap-
propriate frame of reference for research in this area. In building
this framework of theory, there is an obvious advantage in re-
viewing those studies carried out within the appropriate geo-
graphical frame of reference which have the most direct relevance
for Caribbean research. We must begin therefore by delimiting
the geographical frame of special comparative value to Caribbean
studies, directing attention to its more significant features from
our point of view. We shall then have to consider the various
bodies of research and theory developed within this wider area
which are of most importance in the present stage of Caribbean
studies. On the basis of such a review of the literature, rwe shall
then attempt to indicate the type of theoretical frame which seems
appropriate for Caribbean social research.
General historical processes define the regions from Brazil
to the United States as the wider context of direct relevance for
Caribbean area studies. This does not mean that all research
into social and cultural conditions in this wider area, or even the
majority of such work, possesses significance for the understand-
ing of West Indian problems, but only that some studies in this
wider area have already had considerable influence on Caribbean
research by virtue of their local relevance, quite apart from their
more general theoretical or methodological interests. This type
of thing can be expected to continue.
The historical conditions which define the area from Brazil
to the United States as the broad comparative context of Carib-
bean studies are well known. They consist in the expansion of
Europe to the New World, the common historical patterns of con-
quest, colonization, peonage or slavery, and the development of
multi-racial and multi-cultural societies throughout this area.
Regional differences of a contemporary or historical nature are of
obvious significance for comparative work within so vast a frame
of reference. For present purposes the differences of habitat,
economy, population composition, political history and status are
the most useful general guides in a preliminary subdivision of this
The Northern United States forms one region with an over-
whelmingly high proportion of Whites to Negroes in its popula-
tion, a temperate continental habitat, high degrees of urbanisation
and industrialization, independent political status, and Anglo-
Saxon Protestant affiliation, The Southern United States forms
another region, having a higher ratio of Negroes in its population,
lower degrees of urbanisation and industrialization, greater reli-
ance on agriculture, a sub-tropical habitat, and a political history
and status differentiated from that of the North in certain respects,
notably, of course, by defeat in the American civil war of 1865-7.
The Middle American states of Mexico, Guatemala, Hon-
duras, Costa Rica, Nicaragua, Panama, Columbia and Venezuela
can be treated as a group having certain common characteristics,
namely, populations composed principally of Whites and Amerin-
dian elements, with minor representation of Negroes, tropical,
continental habitats, low degrees of urbanisation and industriali-
zation, economies based on agriculture and mining, independent
political systems of similar types, Spanish and Catholic affiliations.
Cuba and Puerto Rico resemble these Middle American re-
publics in certain features, while differing in others. These two
islands have been in closer association with the United States
than the mainland countries, and occupy a politically dependent
position. They are also distinguished from the mainland republics
by the absence of significant Amerindian elements from their po-
pulation, and by the presence of Negroes in some numbers, al-
though clearly as minority groups.
Haiti differs from other provinces in this geographical frame
especially in its combination of French Catholic affiliation, poli-
tical independence achieved long ago through a revolt of slave
and free coloured, the absence of Whites or Amerindians from
the native population, and its recent political association with the
United States. Other Haitian characteristics include low urbani-
sation, low industrialization, tropical habitat, and agricultural
economy, and the island-wide border with its neighbour, the Do-
minican Republic, which resembles the Central American repub-
lics in certain respects.-
Within the vast sub-continent of Brazil, as in the U. S. A.,
several regional sub-types are distinguishable (1). These regional
differences primarily reflect significant variations of climate and
habitat, racial population distributions, urbanisation, industriali-
sation, and agriculture. A general description of that country in
terms of our present interest would note the White majority in its
population, the variable racial distributions with Indian elements
predominating over Negroes in the interior, while Negroes are
numerous in such coastal areas as Maranhao, Bahia, and Sao
Paulo. The political independence and Portuguese Catholic af-
filiation of Brazil are also distinctive as a combination of char-
acters within this region.
(1) Lynn Smith, & Marchant, 1951, passim.
This brings us to the Caribbean area proper, a region char-
acterised by political dependency, and consisting mainly of small
island territories, within which Amerindians are little represented,
White elements form a numerical minority, and the overwhelming
majority of the populations are of Negro descent. A general
summary of this kind at once calls attention to diversity within
this area. We have somewhat arbitrarily separated Haiti, the
Dominican Republic, Cuba and Puerto Rico from the rest of the
Caribbean, because of their distinguishing characteristics. Even
so, our Caribbean area as it stands includes British Honduras, and
the Guianas, British, French, and Dutch, all of which are conti-
nental territories with significant Amerindian elements in their
population. Trinidad and British Guiana are also distinguished
from the remaining British territories in the area by the presence
of large East Indian populations. Surinam differs from other
Caribbean colonies partly in its more varied ethnic composi-
tion, partly through the recent constitutional changes which have
accorded that country an increased share in the management of
its internal affairs.
The past and present associations of Caribbean territories
with different metropolitan powers are clearly important for com-
parative work within the area. Present effects of previous asso-
ciation rules out the treatment of this aspect of Caribbean differ-
entiation purely in terms of the contemporary distribution of
territories among British, Americans, French or Dutch. American
St. Thomas still reveals the influences of its former masters, the
Danes. Within the British Caribbean, islands such as Trinidad,
Grenada, Dominica, and St. Lucia differ as a group from certain
other territories by their continuing affiliation to Catholic tradi-
tion, a pattern laid down in earlier days by French or Spanish
masters. The St. Lucia folk probably have more in common
linguistically with French colonies such as Martinique or Guade-
loupe, or with the former French possession, Haiti, than with is-
lands such as Barbados or Antigua. If we attempt to classify
Caribbean colonies in terms of their present association with
metropolitan powers, we must therefore keep in mind present cul-
tural variations and continuities within and across these divisions
which reflect historical factors of various kinds. Within the
British colonies, the major distinctions reflect differences of racial
population ratios and composition, Protestant or Catholic affilia-
tion, insularity or its opposite. Together with the Caribbean
colonies of other nations, these British territories share a multi-
Sracial composition, from which Amerindian elements are largely
absent, dependence on agriculture, low levels of industrialization,
and low urban ratios.
It is clear that whatever the common patterns which the
British West Indies share with other Caribbean territories, or with
countries outside this Caribbean region, these British colonies
nonetheless form a separate area for social research, on the ground
of their present political relations as well as history. Yet the pat-
terns common to these British territories and other countries of
the wider area delimited above are often of an order which can-
not be ignored in the definition of Caribbean research problems
except at one's peril. Perhaps within the geographical frame des-
cribed above, the Middle American republics, with their Spanish
and Amerindian populations, political independence, continental
situation, and low population densities, have least in common with
the British colonies. Yet it is clear that conditions in these coun-
tries provide useful comparisons with those of the British area,
the populations of which differ from them in history as well as
composition. The comparison of British Caribbean and Middle
American conditions illuminates the study of either milieu separ-
ately by stimulating a variety of questions, hypotheses, and lines of
investigation. Without delaying over this point unduly, we may
mention such questions as the following which invite this com-
parison: what continuities of social structure obtain in multi-racial
societies which vary in their individual racial constitution as these
do? With what structural variations does this continuity coexist?
What differences are associated with the fact that the subordinate
race was settled in the area on the one hand, while it was brought
into it on the other? Or with the parallel fact that the dominant
group settled in the one area, and tended to remain expatriate
within the other? Or with the fact that subordination involved
slavery in one case, and peonage in the other? Or with the devel-
opment of plantation economies in one set of societies, and their
absence in the other? Or in the different Catholic and Protestant
traditions of the two areas? Or their different political histories
and conditions? It will be clear that Caribbean studies may gain
greatly from adopting initial orientations which include these and
parallel problems within a single comparative frame.
Racial and cultural intermixture and blending have gone on
in both Indian-White and Negro-White populations alike. Among
the peoples of Middle America, this has given rise to a section of
the population known as Mestizos (mixed-bloods), or, as in Guate-
mala, Ladinos. Within the Caribbean islands, it has produced a
hybrid group of mixed heritage and colour. The same applies I
broadly to the 'United States and Brazil. What then are the
similarities and differences of relations between these racially and
culturally differentiated groups in the various societies of our com-
parative frame? What significant implications do these continui-
ties and variations present for social research in the British terri-
tories? At what levels are they expressed, in what ways, in terms
of what structures, and with what variable types of function? These
are only a few of the problems which arise when Caribbean re-
search is conceived in its natural comparative context.
Linkages between Caribbean studies and researches con-
ducted within the broad geographical frame delimited above will
become clear from a brief survey of such work over the past
thirty years. We can date the development of current sociological
interest in the British West Indies from 1924 when Martha Beck-
with's first studies of folk-lore and life in Jamaica were published
(2). Two years later, Puckett, from the same American university,
published his definitive study of folk beliefs among the Negroes of
Alabama, Mississippi, and Louisiana (3). During these years also
Nina Rodrigues and Arthur Ramos in Brazil, and Melville and
Frances Herskovits in the United States were being led from initial
studies of folk-lore and physical characteristics of the Negro popu-
lations of their respective societies into the broader problems of
Negro contributions to Brazilian and American culture, and their
adaptation to these environments (4). Herskovits' interest in this
research field was greatly stimulated by an early visit with his wife
to Surinam (5). The broad problem of acculturation which Ramos
and Herskovits formulated was approached by Redfield working
among the Maya of Yucatan from a different point of view, and
led to the formulation of his dichotomy between folk and urban
societies (6). Within the U. S. A. studies of the assimilation of
American Negroes by Franklin Frazier and Lloyd Warner were
underway; and these approaches have also influenced research by
later workers in other regions of our geographical area (7).
Although Dollard's examination of caste and class in a South-
ern town of the U. S. A. acknowledges a debt to Warner and his
school, it also marked a significant departure from previous re-
search in this field by its thorough application of psychological
analysis to the social conditions and relations in the community
which he studied (8). This psychological approach was also com-
bined with an interest in cultural forms and development by
Powdermaker in her independent study of the same community
(9). Psychological studies and interpretations of West Indian so-
cieties have since been published by Campbell, Simey, Hadley,
Kruijier, Cohen, Rhoda Metraux, and Madeline Kerr (10). Mean-
while those interested in the administrative and political aspects
(2) Beckwith, 1924, 1929. (3) Puckett, 1926.
(4) Ramos, 1939. Herskovits, M. J., 1930, 1937, 1938, 1941.
(5) Herskovits, M. J., & F. S., 1934, 1937.
(6) Redfield, 1940, 1941, 1947, 1953.
(7) Warner, L., Junker, B. H., and Adams, W. A., 1941. Davis, A., Gardner, B. B.,
& M. R., 1941. Hill, M. C., & MacCall, H. R., 1950: Frazier, F., 1939. Simey,
T. S., 1946., Matthews, B., 1953.
(8) Dollard, 1937, 1939. (9) Powdermaker, H., 1939, 1943.
(10) Campbell, A. A., 1943: Simey, T. S., 1946: Hadley, C. V. D., 1949: Kruijier,
1953: Cohen, Y., 1953, 1954: Metraux, R., 1952: Kerr, M., 1952.
of Caribbean society were attracted by Macmillan's concise ac-
count, and were moved by the riots and strikes of 1937-8 to re-
consider the colonial situation, and to develop new policies and
programmes of social reconstruction (11). Since the end of the
Second World War, there has been a sharp increase in the volume
of social research in the British Caribbean, which has now come
to be recognized as a separate unit for a programme of area re-
search. Clearly the guiding ideas of future studies will reflect the
interests and approaches already developed in greater or less de-
gree; and it is to the separate consideration of these relevant orien-
tations severally that we must now turn our attention.
Afro-American Research merits consideration first, not
merely because it was the first major development with a direct
Caribbean reference but also because it has been largely developed
on the basis of materials from Caribbean societies, and therefore
has a direct and obvious relevance for us. Afro-American studies
owe a very great deal to Professor Herskovits, who has not been
content to study African survivals in the New World, but has also
sought to fill gaps in the ethnographic knowledge of parent so-
cieties on the West Coast of Africa by his own field-work. Socio-
logical literature on the Caribbean has also been enriched greatly
by the accounts of Professor Herskovits and his wife of their field-
work in Surinam, Trinidad and Haiti; and by other researches of
like orientation, falling within this area or its broader context (12).
Briefly, Afro-American researches consist in the study of
changes or persistence of African tradition and cultural forms
which have marked the historical association between Whites and
persons of African origin or descent in the Americas. Culture
here connotes the total body of learned and transmitted behaviour
which characterises a population, and distinguishes it from others.
Thus Afro-American research is focused on the problem of 'ac-
culturation'; or cultural change in a situation of contacts between
carriers of different cultures. In particular, such research studies
the processes by which the African immigrants and their descen-
dants have retained, lost, or adapted elements of their initial Afri-
can cultures within the contact situation provided by association
with Whites in the New World. The method employed in pursuit
of this enquiry is a combination of history and ethnology; and the
general object is to contribute to the study of cultural persistence
and change by unravelling some of the factors and processes of
acculturation, through detailed studies of Negro-White contacts
in the New World.
(11) MacMillan, W. M., 1936: Olivier, Lord S., 1936: Blanshard, P., 1947: Simey,
T. S., 1946.
(12) Bascom, W. R., 1941, 1952a, 1952b: Carr, A. T., 1952: Eduardo, O. D., 1948:
Pearse, A., 1955: Simpson, G. E., 1940a, 1940b, 1942, 1946, 1948, 1951, 1952:
Taylor, D. M., 1951.
Now clearly the determination of results of this culture-con-
tact must precede the investigation of the processes by which these
effects developed. Hence, Afro-American research is initially
concerned with an examination of contemporary Afro-American
cultures to discover survivals, retentions, syncretisms, or reinter-
pretations of African cultural elements obtaining within them. As
the results of such examination accumulate, they also raise
problems about the processes of acculturation, especially with re-
gard to the differential survival of African cultural elements of
various kinds, and in differing environments. A useful tool devel-
oped to facilitate such comparative analysis is the concept of a
Scale of intensity of Africanisms, which permits a classificatory
comparison of New World cultures from Brazil to the U. S. A.,
distinguishing between Africanisms in economic, social, religious,
and aesthetic life (13). Such a scale shows a greater concentration
of Africanisms in such fields as folk-lore, music, and religion, than
in technology and economic life; but there is a notable absence
of political and governmental institutions from this comparison of
Africanisms. The differential intensity of Africanisms in these
various fields invites some explanation. Analysis of the problems
which this differential intensity presents is undertaken with the
aid of various hypotheses and concepts, the most important of
which distinguish the focal aspects of cultures as those most
tenacious in situations of contact, and of greatest interest to the
populations concerned; and conceptualise cultural persistence in
terms of survivals or retentions on the one hand; syncretism and
reinterpretation on the other. Survivals have greatest direct
resemblance to original forms. Syncretisms involve a combi-
nation of parallel forms from the cultures in contact; while
reinterpretations adhere to the substance or content of the original
culture, although departing from its initial forms. Finally the con-
cept of cultural imponderables connotes that category of African-
isms which are clearly not included in the cultural focus, but none-
theless have a high level of intensity in New World populations.
It appears that values and automatic motor patterns constitute
the bulk of these surviving cultural imponderables. Within this
frame of research the ethnohistorical method has hitherto been
employed principally to determine the cultural provenience from
which Africans were recruited for the various New World slave-
states, and the types of condition to which they were subjected
under the slave regime. In the attack on problems of differential
intensities of Africanisms, that is, the variable effects and pro-
cesses of acculturation, the ethnohistorical method has played a
less prominent part than the development of the hypotheses and
concepts just mentioned.
(13) Herskovits, M. J., 1941, 1946.
This seems to constitute one of the major weaknesses of cur-
rent Afro-American studies. Instead of undertaking the examina-
tion of those processes of acculturation which forms the stated
object of such research, by the more vigorous employment of the
ethnohistorical method which documentary materials, autobio-
graphical and other studies permit, with a view to the unravelling
of detailed processes in limited fields, the tendency has been rather
to develop and systematise a set of concepts which taken together
obscure the problems of process rather than otherwise. We can
illustrate some aspects of this weakness by a brief examination of
these concepts and their inter-relations, ignoring for the moment
questions of variable distributions of African elements within
each of the populations concerned in these researches, or the na-
ture of Africanisms as such.
The scale of intensity of Africanisms shows a variable ten-
acity of African cultural elements distinguished in terms of tech-
nology, economy, social organization, non-kinship institutions,
religion, magic, art, folklore, music and language. The problem
of variation in these different fields leads, after a discussion of
syncretism, retention and reinterpretation, to the following formu-
lation by Herskovits: "Even under the compulsions of the domin-
ant culture of the Whites, Negroes have retained African religious
beliefs and practices far more than they have retained economic
patterns. But when we examine the patterns of African cultures,
we find that there is no activity of everyday living but that it is
validated by supernatural sanctions. And consequently, these
figure far more in the total life of the people than does any other
single fact of the culture such as those matters having to do with
making a living, or family structure, or political institutions. This
weighting of the concerns of a people constitutes the focus of their
culture.... The role of cultural focus is of such great importance
in situations of cultural contact that a further hypothesis may be
advanced to the effect that more elements which lie in the area
of focus of a receiving culture will be retained than those apper-
taining to other aspects of the culture, acceptance being greater
in those phases of culture further removed from the focal area.
Where a culture is under pressure by a dominant group who seek
to induce acceptance of its traditions, elements lying in the focal
area will be retained longer than those outside it, though in this
case retention will of necessity be manifested in syncretisms and
reinterpretations." (1946, p. 352).
We can summarise the relations between the conceptual sys-
tem and the scale of intensity in the following terms; greatest in-
tensity of cultural survivals occurs in the area of cultural focus, if
cultural imponderables are excluded. The cultural focus is the
area of greatest tenacity in cultural retentions, consequently, as-
suming equal pressures on all fields of cultural activity from out-
side, it will show the highest degrees of purity in these retentions,
and will also contain the last elements to disappear. Purity of
retention diminishes as it passes from direct survival to syncretism,
and so to reinterpretation. The relative proportions of these
particular modes of persistence in different cultural spheres is re-
flected in the scale of intensity; and this illustrates or defines the
focus. So the circle is completed; and the variable persistence of
cultural elements of different kinds is simply restated in the form
of a system of hypotheses and concepts, ostensibly developed to
further the analysis of this variability, only to be canvassed there-
after as its explanation.
Even so the distinction between cultural focus and periphery
in terms of tenacity and persistence is not borne out by the scale
of intensity on which it is based. Thus, Herskovits' comparison
of folk-lore with magic or religion in terms of levels of intensity
of African elements of these kinds for the fifteen areas concerning
which materials were then available shows that folk-lore elements
persists on average to a higher degree than do either magic or re-
ligion. Similarly, Africanisms in music have a higher level of in-
tensity than those in any other field, in terms of this scale (Hers-
kovits, 1946, p. 352). Now it is possible to exclude music from
the comparison effectively by treating it as a pattern of motor be-
haviour of the type which is liable to persist in a very marked de-
gree, even though marginal to the culture focus. But this treat-
ment cannot be extended to cover folk-lore. Nor is it useful to
define folk-lore, religion, and magic coterminously, although this
would remove the problem of peripheral elements showing a
higher degree of tenacity and persistence than focal ones. On the
other hand, if folk-lore is included among the cultural imponder-
ables on grounds of the value-systems which it often expresses,
the question arises whether all other departments of culture are
not equally open to similar treatment, notably of course, religion,
and kinship. In such a case, the notion of cultural focus as dis-
tinct from periphery would cease to be of much use.
There is always grave danger in exaggerating the relative im-
portance of one aspect of culture at the expense of others; and
this difficulty is involved in the concept of cultural focus. The
type of conclusion which emerges from the most thorough study
of any African society yet published is relevant here. "To study
Tale kinship institutions apart from the religious and moral ideas
and values of the natives would be as one-sided as to leave out
the facts of sex and procreation. On the other hand, our analysis
has shown that it is equally impossible to understand Tale religious
beliefs and moral norms apart from the context of kinship. A very
close functional interdependence exists between these two cate-
gories of social facts" (Fortes, 1949, p. 346). Similarly Forde
finds from a review of African cosmologies, that "belief and ritual
tend, in other words, to mirror the scale and degree of social in-
tegration" (Forde, 1954, p. xvii). Fortes and Evans-Pritchard
reach a parallel conclusion from their review of African political
systems. "Myths, dogmas, ritual beliefs and activities make his
social system intellectually tangible and coherent to an African
and enable him to think and feel about it. Furthermore, these'
sacred symbols, which reflect the social system, endow it with
mystical values which evoke acceptance of the social order."
(Fortes & Evans-Pritchard, 1940, p. 17). These observations in-
dicate the very great need for caution in the classification of cul-
tural elements as focal or otherwise, and direct attention to their
close interdependence. Clearly, insofar as religion and kinship
are essential to the understanding or practice of one another, their
separation and ranking in terms of cultural priorities is liable to
do violence to these relations.
We can perhaps more usefully and easily distinguish foci in
the culture-contact situation itself than in the cultures themselves.
Herskovits makes frequent reference to this variability in the pres-
sures of the contact situation, but does not systematise concepts
to treat it. In the historical situation of Afro-American culture-
contact these foci of contact reflected the interests of the domin-
ant group in its control of the subordinate as slaves. Consequent-
ly the social organisation, technological, and economic practices
of the subordinate Negroes were subject to pressure of a kind
without parallel for intensity, and continuity, in such other fields
as religion, music, or folk-lore. Language, the essential mode of
communication between the dominant and subordinate groups,
occupied an intermediate position in this variable pressure of cul-
tural elements between the two groups. The advantages of con-
ceiving the contact-situation as a field of variable pressure over
time as well as at any moment are many and varied. It directs
attention consistently to the study of social relations between and
within the two culturally differentiated groups as the matrix of
these acculturation processes, and thereby invokes employment of
historical and sociological research together to relate these pro-
cesses to the structures and situations through which they matured.
It allows relatively simple and precise determination of the focal
and peripheral fields of culture-contact on the basis of document-
ary analysis. It thereby permits the study of persistence of dif-
ferent categories of elements and in differing degrees of intensity
or purity to proceed without supplementary postulates about focal
and peripheral sectors within the cultures themselves which are
hardly verifiable. It directs attention to the fact. that purity of
form in survival might simply indicate marginality within the ac-
culturative situation rather than any central significance of the
elements retained to the original culture on the one hanc; wnne
relative impurity of form in persistence might simply indicate the
relative intensity of pressure on the elements concerned, rather
than their marginality to the original culture. A good many pos-
sible fallacies are ruled out at once by such conceptualisation of
the contact continuum, and effort is thereby redirected from the
development of broad classificatory conceptions such as retention,
reinterpretation, and syncretism, or of imprecise and unverifiable
hypotheses such as that of cultural focus, toward the formulation
of more limited propositions capable of being tested against his-
torical materials on the contact situation within the particular
fields for which they are separately developed.
Hitherto we have been discussing certain aspects of Afro-
American research which focus on the processes by which accul-
turation proceeded among New World Negroes. We must now
turn to consider its conceptual system with reference to the classi-
fication and study of the forms produced by such processes. Here
we are mainly concerned with the precision or generality of the
principal concepts, although we must commence with the problem
of attribution. Clearly, Afro-American research can only yield
tentative ascriptions of provenience to contemporary custom, to
the degree that the African centres from which New World
Negroes and their ancestors were recruited are unknown, or to
the degree that parallel European practices, or the measure of in-
fluence exercised by these forms on the development of contem-
porary Africanisms are not fully determined. Systematic study
has shown that the areas from which most of the New World
Negro slaves were recruited lie along the densely populated West
African Coast. This reduces the problem of the provenience of
Africanisms, leaving only the question of their accretions of
European or Amerindian elements. Amerindians being largely
peripheral to the areas of Negro-White contact, they may be
ignored for general discussion.
Herskovits is careful to weigh the influence of European cul-
tural practice on African tradition, particularly where elements of
folk-lore are involved; but the issue is greatly obscured by his pos-
tulate of common denominators in European and African cultures
in the concept of the Old World as a single cultural province
(14). A concept of this level of generality is of dubious value.
It implies a division of the World into two cultural provinces, the
New and the Old. Yet this could easily be criticised, partly on the
evidence that points to the movement of Old World populations
(14) Herskovits M. J., 1941.
into the Americas to become its 'aboriginals'; more importantly
on the ground that absence of such elements as the wheel or writ-
ing from 15th century America on the one hand, or tobacco in
the Old World at that date, form an inadequate basis for such a
distinction, since cultural differences of a similar order have been
overlooked within the cultures of the Old World province itself.
Yet if this criticism was granted, it would place some strain on
the conceptual framework of Afro-American studies.
It is difficult to reconcile specific studies of Afro-American
acculturation with statements such as the following: "It is here we
must turn for an explanation of the seemingly baffling fact, so
often encountered, that given traits of New World Negro, and
especially of American Negro behaviour, are ascribable equally to
European and African origin. This may well be viewed as but
a reflection of the fact that deep beneath the differences between
these varied civilisations of the Old World lie common aspects
which, in generalized form, might be expected to emerge in situ-
ations of close contact between peoples, such as Europeans and
Africans, whose specialized cultural endowments are comprehend-
ed within the larger unity" (Herskovits, M. J., 1941, p. 18). Simi-
larly in comparing the wider persistence of Africanisms in magic
than in religion, Herskovits notes the advantages of magic in being
private and difficult of detection where pressures are brought to
prohibit both practices; ad concludes that Africanisms in magic
"persisted in recognizable form everywhere, particularly since the
similarity between African and European magic is so great that
the one cultural stream must have operated to reinforce the
other." (1946, p. 348). It is doubtful in what sense the predicated
similarity of European and African magic can be taken to contri-
bute to the survival of Africanisms in this field, as is clear from
a glance at such patterns in West Indian Obeah. As is well known,
a good deal of the magical rites of the Obeahman are taken in
whole or in part from imported literature such as the Sixth and
Seventh Books of Moses, the Black Arts, and the like. These
books describe techniques which are significantly different from
African practices, especially by use of cabalistic signs, writing,
foreign languages, and the like. They present a type of magic
which is distinguished locally in terms of its literary and learned
pretensions, as 'book magic'. In contrast, pure Africanisms in
Obeah rely mainly on the employment of herbal or animal sub-
stances, and the casting of spells in dialects or African tongues.
These two categories of Obeah, the literary and the strictly Afri-
can, are at least as much in competition as they are active in rem-
forcing one another. Thus acceptance of the one form may mean
displacement of the other. Something similar has been recently
reported from Africa also, by Nadel, who found the native pagan
C. A. 2
magic of Nupe being displaced by imported forms enjoying the
higher prestige of Islamic civilisation (15). Now, unless a careful
analysis of the magical systems of New World Negroes is made
to determine exactly what proportion and type of practice has
European or African provenience, the attribution of these forms
to African culture whether as syncretisms, reinterpretations or re-
tentions, is really begging the question of their origin. The con-
cept of a 'generalized form' which permits this type of attribution
is thus confusing rather than helpful to the analysis.
These observations direct attention to the levels of generality
on which the search for Africanisms and their attribution pro-
ceeds; and these levels vary widely indeed, from meticulous cor-
respondences between Haitian and Dahomean vodun, or Afro-
Cuban and Yoruba divinatory practices on the one hand, to such
conceptualisations as that which reduces 'matriarchal' family pat-
terns and loose mating associations among New World Negroes
to the level of 'reinterpretations' of African polygynous patterns by
the device of successive rather than simultaneous plural matings
on the other (16). Important questions concerning the levels of
generality in such conceptualisation of family-types, mating pat-
terns, and reinterpretations remain to be answered before the at-
tributions involved can be discussed profitably. But there is an
alternative approach to consideration of such problems as chang-
ing family structures present for acculturation studies, through
the comparison of parallel situations and developments in other
parts of the world.
Let us examine this matter of family forms for a moment,
since its handling by Afro-Americanists has promoted some con-
troversy, their opponents attributing the contemporary 'disorgani-
sation' of New World Negro family forms to "the historic condi-
tion of slavery", under which, as is well known, stable matings
among the Negro populations were inhibited by a variety of fac-
tors (17). Afro-Americanists, as we have seen, derive these
'deviant, disorganised' family patterns of New World Negroes
from African practice by reinterpretation. Much ink has already
been spilled on these conflicting ascriptions, and their antithesis
has directly affected the study of family patterns in the British
Since as we have seen the level of generality in the ascrip-
tion is imprecise, the only check on these competing theories is to
compare West Indian conditions with those elsewhere; yet, obvious
(15) Nadel, 1954.
(16) Herakovits, M. J., 1937, 1938b, 1941, pp. 169 ff.: Bascom, W. R., 1952a, 1952b.
(17) Frazier, 1939, 1953: Henriques, F. 1952, p. 24; 1953a 1953b; Simey, T. S., 1946:
Matthews. Dom B., 1953. cf. also, Herskovits, M. J., 1952b.
as this is, it has not yet been attempted. Among the South Afri-
can Bantu living in Native Locations on the outskirts of European
cities, "deviant, disorganised" family patterns strikingly similar to
those found among New World Negroes are common. Yet clearly
these South African Bantu families cannot be attributed to slavery
on the one hand, nor to the persistence of aboriginal Bantu pat-
terns by reinterpretation on the other, since there has been no ex-
perience of slavery among the urban populations presenting these
problems, and the new family types which are developing among
them are as deviant from Bantu norms as from European. (18).
A debate whether such family types among the urban Bantu are
reducible to persistent Africanisms which have been reinterpreted,
or to historical conditions such as slavery, would therefore have
little to recommend it; and by extension this applies also to the
debate about the derivation of the similar family patterns of New
World Negroes. A more realistic analysis than either school of
thought seems yet prepared to apply would involve the study of
mating and family patterns in terms of particular social and eco-
It is also unsatisfactory to attribute peculiar New World
Negro family types to "historic conditions of slavery" without
making a far more detailed comparative analysis of these than has
yet been attempted. Among the Muhammadan Hausa and Fulani
of Northern Nigeria who operated an institution of slavery similar
in many respects to the historic New World pattern, religious con-
version and marriage of slaves was an important obligation of their
owners, and the abolition of slavery in this area has left no after-
math of deviant and disorganised family types (19). What in fact
passes under the name of 'slavery' in the British West Indies and
the New World generally, includes a great many factors which are
not essentials of the institution. The Hausa sought to arrange
marriages for their slaves: the New World masters on the whole
did not. Slavery as an institution does not enjoin any particular
mating patterns, for slave or slave-owner. This institution, like
others, can only be fully understood by comparative studies of its
structure and function in different and similar contexts. It is
therefore just as unsatisfactory to reduce 'deviant' New World
Negro family types to slavery as to the reinterpretation of African
It is worthwhile to dwell a little longer on this problem of
family derivation as handled in Afro-American studies. If our
preceding observations are accepted, the problem is basically
(18) Mair, L.. in Phillips, A. (Ed.), 1953, pp. 28-45.
(19) Smith, M. F., 1954; Smith, M. G., 1954.
one of deviance, rather than a problem of cultural sur-
vival or reinterpretation. Similarly, if our observations on the
composite character of West Indian Obeah are valid, then Obeah
as a deviant pattern of behaviour is more extensive than the
specifically Africanist magical practices of West Indian popula-
tions. These distinctions between deviance and Africanism are
surely of critical importance to Afro-American research. The de-
viant mating patterns of various Latin American populations, as
implied by the statistics of illegitimacy, are obviously in some res-
pects parallel to those of New World Negroes, while in no way
being attributable to African culture. (19). The distinction be-
tween deviance and Africanism is however somewhat difficult of
construction in Afro-American studies, on three main grounds;
imprecision in the definition of traits as African or other, which
we have already seen to be partly related to the assumption of
correspondences between African and European cultures in the
Old World; wide variability in the level of generality of the con-
cept of reinterpretation, which together with the imprecise defini-
tion of Africanisms allows of extension equally to almost every
field of social life; and most importantly, the formulation of prob-
lems of Afro-American contact and research mainly in terms of
culture and acculturation, without corresponding emphasis
on the primarily sociological aspects of these processes.
Deviance, for example, is more easy to define in terms of social
norms than in terms of cultural form. On the other hand,
where the deviant practices are concentrated disproportionately
among a particular ethnic group, the culture of which forms a
direct object of enquiry, it is easy to conceive the differentiating
behaviour primarily or even entirely in cultural terms. This over-
simplifies the problem on the one hand, and poses questions of
acculturation and culture-change which may be of little -direct
relevance on the other. Yet because culture is the total transmitted
heritage of a population, deviance in any form also invites con-
sideration in terms of a theory of culture-change. Now this is
obviously a valid interest, but one liable to fail of attainment,
unless it is accompanied by a thorough study of the social struc-
tures and situations within which and with regard to which the
deviation has developed.
Another characteristic of Afro-American theory which merits
attention is the relatively great emphasis which it places on the
acculturative situation of slavery and the relatively marginal treat-
ment which is given to historical conditions following on eman-
cipation. For this reason it has been possible to debate the
derivation of deviant Negro family types in terms of slavery or
(19) Matthews, Dom B., 1953, p. 13.
African influences. Wittingly or not, in this matter as in certain
others, Afro-American studies give the appearance of opposing
African cultural conservatism and resilience on the one hand to
the culturally destructive practices and organisation of slavery on
the other (20). But the acculturative process is liable to serious
misconstruction if handled in terms of such a simple dialectic.
Many years have elapsed since emancipation in all parts of the
Western world, and this period may well have an importance
for the study of Negro acculturation equal to or greater than that
of the slavery which preceded it. If we compare the distributions
of Africanism among the Negroes of the Northern and Southern
United States on the one hand, and the Haitians whether urban
or peasant on the other, we find the scale of Africanisms in-
creasing in this order: Negroes of the Northern United States,
those of the South, the Haitian town dwellers, the Haitian
peasantry. Such a pattern of distribution directs attention to the
variability of acculturative situations and influences experienced
by these four groups in the post-slavery period. Despite the
relatively late emancipation of the American Negroes, they prac-
tise a culture which contains fewer African elements than that of
the Haitians, whose freedom was gained many years earlier.
Within either population moreover, the variation of African ele-
ments reflects differences of acculturative situation and exposure.
Thus, urban Haitians are exposed to foreign cultural influences
to a greater degree than the peasants, and the Northern American
Negroes enjoy an environment more favourable to their accultura-
tion than do those of the South. For some years after emancipa-
tion, Africans continued to arrive in the British West Indies as
free indentured labourers; and the same is true in various other
parts of the Afro-American area from Brazil to Haiti (21). In
Jamaica between 1834 and 1865, 11,380 free African immigrants
settled, a number well in excess of the indentured labourers im-
ported from India over this period (22). More so than any other
group of immigrants, these free Africans enjoyed cultural condi-
tions which were favourable to the survival of certain of their
practices. Quite often the decisive condition for this survival
consisted in group cohesiveness, as well as sympathetic attitudes
in the Negro section of the host society. We can illustrate this
process by a glance at the development and spread of the Shango
cult in the island of Grenada.
At the turn of the present century, the representative African
cult of the Grenadian folk was a type of ancestor ritual known
locally as the Big Drum, the Nation Dance, or simply saraca
(20) Herskovits, M. J., 1941, pp. 110-142.
(21) Leyburn, J. G., 1941, pp. 46-50, p. 86: Pierson, D., 1942.
(22) Hall, D. G. H., 1954.
(sacrifice) (23). The ritual dances of this cult normally lasted
three days, from Wednesday to Saturday, and contained several
elements which have since been assimilated by Grenadian Shango;
but Big Drum rites were not associated with spirit-possession,
which was not then practised. Shango in Grenada was originally
the ritual of certain closed communities of Africans and their
descendants at Munich, Concorde, and La Mode. These Africans
came from Ijesha in Yorubaland after slavery had been abolished,
over 1,000 of them in 1849. On completion of their indentures
they settled in the three communities named above. Within
these communities, Yoruba was the spoken language, and
numerous elements of Yoruba culture were preserved, including
kinship elements, and the basic concepts and rites of Yoruba
poytheism. Later, when the Africans and their descendants
started to move out from these communities, Creole Grenadians
showed great receptivity to their cult, and its spread outwards
from these three centres was marked by syncretisms of form and
content, numerous traits being taken over from the Nation Dance
as well as from Catholicism, until Shango is now the representa-
tive form of African ritual among the Grenadians.
Here we see clearly the importance of group cohesion for
the persistence and survival of a trait or complex of traits.
Shango with its priesthood, has displaced the Nation Dance,
which lacked formal group organisation, in Grenada island, al-
though Shango was a late arrival competing with a widely held
cultural form. But Shango remained quite unknown till 1953 to
the population of Carriacou, a Dependency of the Grenada gov-
ernment, and only about 23 miles away. No groups of post-
emancipation immigrants from Africa had settled in Carriacou;
and the Big Drum cult still flourishes as the representative folk
ritual in that island.
To devote insufficient attention to the sociology of accultura-
tive situations or the role of organised and persistent groups in
the preservation or transmission of culture is methodologically
perilous, since it is clear that the group as a carrier of culture is
a natural unit far superior to the individual; and also that the
structural relations holding between as well as within each of the
culturally differentiated groups involved in the acculturation pro-
cess might well be of the utmost significance in understanding its
development and effects. It seems likely for example that the
Dahomean contingent imported to Haiti by Christophe after
slavery and maintained by him as a unit with high prestige may
have contributed to the present persistence of Dahomean patterns
(23) Pearse, A., 1955.
in Haiti in a degree disproportionate with their relative numbers.
(24). The point to note here is that if acculturation is to be
studied in terms of specific social contexts, then the implied op-
position of slavery and African cultural conservatism must be
replaced by a continuous study of the contact situation both within
and since slavery, and particular attention must be devoted to
the historical reconstruction of the social situations within which
these developments occurred. This extension of interest could
well entail major revisions of the method, theory, and conceptual
equipment of Afro-American research.
However, this conceptual equipment in any case needs some
overhauling. Let us briefly consider first, the principle of re-
interpretation, and then the concept of Africanisms, for example.
"Where it is not possible to set up syncretisms, the force of
cultural conservatism seeks expression in the substance, rather
than the form, in psychological value rather than in name, if the
original culture is to survive at all. Here the importance of
resemblance of the old element to the new is again involved.
Though to a lesser degree than in the instance of syncretisms,
reinterpretation also requires that some characteristic of the new
cultural element be correlated with a corresponding part of the
original one by those to whom it is presented, before the mechan-
ism can operate effectively" (Herskovits, M. J., 1946, p. 351).
Much of this is perilously like reification; and one of the dangers
of acculturation studies which are not balanced by continuous
examination of and reference to social process consists precisely
in this tendency toward the reification of cultures and their com-
ponent forces or parts. This danger is inherent in acculturation
studies which are undertaken without adequate sociological em-
phasis, since their field of interest is therefore defined purely in
terms of two or more cultures, or bodies of tradition, in contact.
But cultures do not carry themselves, nor do they of themselves
have contacts, although their human carriers do.
The other notable point in this definition of reinterpretation
is its inclusiveness, and its functional reference. Retention and
syncretisms are the conceptual categories which focus directly on
the survivals of cultural form. With reinterpretation however we
are concerned with the survival of substance, that is content and
function, rather than with cultural form. Thus reinterpretation
demarcates a category of persistence which cannot be defined or
recognized in formal terms, but involves functional or valuational
correspondences. From this point of view, it is somewhat difficult
to set any bounds, other than the purely formal one already men-
tioned, to the ramification of the reinterpretative mechanism which
(24) Leyburn, op. cit., pp. 46-50, 86.
expresses 'the force of cultural conservatism' within any process
of cultural exchange. Herskovits is well aware of this, but fails
to face its implications. "This, of course, raises one of the most
difficult problems in the entire field of cultural dynamics-whether
any element of culture is ever taken over without some degree of
reinterpretation, however free the borrowing." ibidd, p. 351),
The point is that, as defined, reinterpretation can apply to every
cultural item observed among persons initially classified as bearers
of any particular culture on grounds of race, descent, nationality,
or otherwise. All that is necessary for this ascription is that
there should be no formal parallels with practices of the 'original'
culture and also that t h e r e should be a general tendency of
human cultures to show some correspondences on the planes of
function or value. Categories of such generality are liable to
mean everything and nothing at once, and to invite questions as to
whether particular reinterpretations may not be as validly attri-
buted to a writer as to the African population itself.
The key concept of Africaiism is similarly ambiguous. It
applies equally to any particular trait, however minute, such as
the word 'ere', to the thing or condition with which it is as-
sociated and to the system within which the trait is found. It
applies equally to relatively pure retentions on the one hand, or
reinterpretations which have dubious value on the other, and to
minute elements or large behavioral systems, such as family-pat-
terns, throughout all ranges of persistence. It does not seem to
permit easy attention to competing Africanisms, such as have been
illustrated above from the development of folk ritual in Grenada,
nor to compound formations of African tradition which develop
through such processes, though here also the concepts of survival,
syncretisms, and reinterpretation might apply. Taken together
with the postulate of cultural conservatism and the general con-
cept of reinterpretation it allows easy admission to a variety of
possible types of deviant behaviours as Africanisms, although they
may have no African provenience or parallels at all. In concert
with the postulate of common denominators among the Old
World cultures, it can also be applied to such institutions as the
West Indian 'wake' or 'nin' night', which Walter Scott reported
among his countrymen. Similarly, although Haitian vodun is
clearly African in inspiration and detail, there are elements in it,
such as the pantheon of Creole gods headed by Dom Petro,
which have different, perhaps Amerindian, provenience. (25).
These Creole elements are clearly African in a different sense
from the more strictly Dahomean parts of the complex, and can-
not adequately be handled in terms of syncretism or reinterpreta-
(25) Deren, M., 1953, pp. 61-71, 82-5, 270-286.
tion separately. The problem of independent developments
among the New World Negroes, such as the Calypso music and
steel bands of Trinidad, must also be faced. In some sense these
new cultural forms are Africanisms, but little is to be gained by
classifying them as retentions, syncretisms, or reinterpretations.
Pearse's approach to the study of these and other musical forms
in the South Caribbean in terms of their institutional settings, par-
ticipants, and pattern, offers a useful lead by its precise formula-
tion of research problems and definitions which could also be
applied to certain other areas of Afro-American cultural ex-
Studies of acculturation which are inadequately balanced by
studies of the social situation, processes, and structures involved
in such change, can hardly be expected to produce agreement,
where social structures are the cultural traits involved. Careful
reconstructions of past states of the societies with which Afro-
American studies are concerned on the basis of indexed docu-
mentary materials, undertaken with the object of defining the
structures, contexts, and functional characteristics of units and
institutions in these past systems, are perhaps the only ways of
determining how past social conditions and processes may have
guided acculturation and contributed to the development of cur-
rent social and cultural forms. Comparative studies of Afro-
American societies focused on the detailed analysis of their
structural and functional characteristics will also give rewarding
leads about the role of social factors in the present as well as the
past acculturative processes. Together such studies might serve
to clarify the nature of cultural exchange and evolution, the types
of context within which particular cultural developments occur,
the distribution of Africanisms within delimited populations, and
the relation of this distribution to contemporary social and
economic factors on the one hand and historical process on the
The lack of any systematic study to date of the degrees and
types of acculturation associated with different s o c i a and
economic positions in Negro-White populations of the New World
constitutes a serious weakness of Afro-American research. Until
these studies of social and cultural differentiation within the
populations from which Afro-American materials are drawn have
been made, problems posed by the variable distribution of African-
isms within these populations, as well as between them, cannot
even be defined, let alone receive attention; nor can the critical
relations of social structure and acculturative process be analyzed.
Yet variability in the incidence of Africanisms within a popula-
(26) Pearse, A., 1955.
tion is clearly an aspect of the internal differentiation of that
unit, and implies the concurrence of acculturative processes be-
tween the differentiated sections, as well as of acculturative pro-
cesses separately within each. Until such studies of internal dif-
ferentiation are made, the representativeness or significance of the
Africanisms reported from such populations remain open to ques-
tion, and the core of the acculturative process cannot be exposed.
The Folk-Urban Continuum.
Robert Redfield, who has also been interested in the problem
of acculturation, has attacked it on a different front-in Central
America-and from a different point of view. Redfield worked
in Yucatan, studying an important urban centre, Merida, and
three rural Maya communities which were situated at different
distances from the town. He found that differences in the cultural
and social life of these rural communities seemed to form a
developmental series which was matched by their relative distance
from the common urban centre, or, conversely, by their immunity
and isolation from modern influences. Redfield therefore raised
the question whether in fact the significant feature of this accul-
turative process did not consist in a change of societal type; and
on this basis he formulated an ideal-type dichotomy of folk and
urban societies (27). In terms of this polarity, the most isolated
of his rural communities, Tusik, most nearly approximates the
definition of the Folk Society, while Merida is representative of
Urban Society in Yucatan, and other intermediate communities
such as Chan Kom occupy a position on the cultural scale cor-
responding to their geographical situation between these two
poles. This ideal-type antithesis allows Redfield to present a
systematic comparative analysis of cultural changes to be found
as one moves from Tusik towards Merida in terms of increasing
disorganisation, secularization, and individualisation; and then to
enquire into the functional relations between these three aspects
of the acculturation processes to which modern urbanisation sub-
mits folk societies.
A brief comparison of Redfield's approach with that of
Herskovits may be useful at this point, since both sets of re-
searches fall into the common field of acculturation studies, and
these two workers once co-operated in producing an early memo-
randum on the study of acculturation (28). Both writers employ
comparative procedures in their studies of culture change or per-
sistence, but whereas Herskovits devotes little attention to the
variability of acculturation levels within a population, and com-
pares societies on the basis of the distribution of African items
(27) Redfield, R., 1940, 1941, 1947, 1953.
(28) Redfield, R., Linton, R., & Herskovits, M. J., 1936.
between them, Redfield's emphasis is rather on the comparison
of different levels of acculturation within a single continuum, and
on the processes associated with acculturation as such. Unlike
Redfield who takes the contemporary situation' as his historical
base-line, and then proceeds to compare four differently situated
communities in terms of certain selected indices and characters,
Herskovits regards the initial situation of contact between the
carriers of different cultures as the appropriate base-line for his
studies of Afro-American acculturation. Herskovits is therefore
committed to ethnohistorical research, whereas Redfield's con-
cern is with the incidence and processes of contemporary accul-
turation. Here it must be pointed out that Redfield's analysis of
the cultural continuum of Yucatan is considerably weakened by
the insufficient attention which he gives to the effects and pro-
cesses of suppression of the great Maya rebellion in that area, as
well as to the general history of Indian-Spanish relations in
Mexican politics and government.
Redfield's approach directs greater attention to the structure
of relations within and between groups as well as to impersonal
media of communication than apparently does that of Herskovits;
and this difference of orientation is reflected in the different con-
ceptual systems of the two writers. Thus Redfield analyses the
acculturation continuum from Merida to Tusik in terms of a
societal polarity, and consequently employs concepts and cate-
gories which focus on the social aspects or concomitants of cul-
tural change, such as disorganisation, individualization, or secular-
ization. Herskovits, on the other hand, approaches problems of
process through the initial study of forms, and therefore utilises
categories such as survival, syncretism, or reinterpretation, which
reflect degrees of purity and independent persistence of the
original cultural forms. Herskovits, in other words, has placed
a primary emphasis on the cultural frame of analysis, whereas
Redfield has tended to dovetail his studies of acculturation with
the analysis of changing social forms.
Behind these differences of Herskovits and Redfield loom
the historical differences of the areas with which each worker
is concerned. The subordinate Maya of Yucatan were inducted
into the wider society of their conquerors as communities, and
their original social organisation persisted in this new situation
as far as the two were consistent. The African populations of
the New World, however, were recruited on a basis which, in
large measure, destroyed their original social organisation, and
were then subjected to a mode of social reorganisation in slavery
which tended to continue that process. Consequently, the study
of acculturation among the Maya can hardly be developed with-
out devoting equal attention to social process and changing social
forms, while the same study of acculturation in Afro-American
populations would encounter many obstacles if it was initially
conceived in terms of the concurrent study of social change.
These important differences of milieu largely account for the
hitherto divergent interests and procedures of the two sets of
studies; but they do not by any means rule out the combination
of their approaches within the framework of a general theory and
conception of sociocultural processes which may hold for both
Redfield's approach, like that of Herskovits, has stimulated
considerable interest and debate, and will probably continue to
do so for some time to come. Tax has shown that Redfield's
ideal-type dichotomy is not mutually exclusive as suggested, even
in nearby Guatemala (29). Julian Steward has pointed out
that the concept of levels of organisation within a developmental
continuum applies to the materials which Redfield uses, and
therefore makes the classification of any community in terms of
the folk-urban continuum a purely definitional problem (30).
Herskovits has brought West African data to bear on the con-
tinuum (31). Bascom, quoting Schwab, notes that some Yoruba
cities are 'urban' in terms of form, but 'folk' in terms of process
(32), and it has often been pointed out that urban communities
frequently contain folk elements, and that folk societies also may
exhibit urban elements. The position of certain societies such
as the Hausa of Northern Nigeria in this continuum will vary
according to the particular differentia of the polar ideal-types
which are under immediate consideration (33).
It seems, however, that the principal criticism of Redfield's
ideal-type analysis and approach to the study of cultural contact
and change may consist in the incommensurability of the units
between which his comparisons were actually made. One term
of Redfield's dichotomy is given as the Folk Society, but it is
only communities that he actually investigated, and there are
.crucial differences between communities and societies, particularly
where the former are subordinate rural administrative units of a
modern state whose governmental and economic life is centred
in the urban areas. (34). Steward has recently called attention
to this difference and to the problems of comparability which it
presents. In terms of Redfield's ideal-types and empirical re-
search, the urban unit, Merida, more closely realises the condi-
tion and status of a society therefore, than do any of the rural
communities which are under its effective administration. For
this reason also, the variation in degrees of urbanisation, dis-
(29) Tax, S., 1939. (30) Steward, J., 1950, p. 106 ff. (31) Herskovits, M. J.,
1948. n. 606. (32) Bascom, W. R.. 1955. (33) Smith, M. G., 1955.
(34) Steward, J., ibid, p. 20: Arensberg, C. M., 1954.
organisation, and the like, which are exhibited by the several
rural communities which Redfield studies, are more significantly
analysed in terms of the effective subordination and inclusion
of these communities within the wider administrative unit, than
simply as a function of linear distance from the town.
The root of this difficulty seems to lie in the failure to
distinguish sufficiently between social systems and societies. Al-
most any group structure or activity can be conceived of as a
social system of some particular kind or other, and this applies
to schools, societies, and communities alike. But the society
as a system is distinct from other social systems in many im-
portant ways. (35). The society is a system which includes all
other types of social system as parts of itself. It is self-recruiting,
theoretically self-sufficient, and is self-determining within the
context of its relations with external societies. Communities
differ from societies particularly with respect to inclusiveness,
self-determination, and self-sufficiency. Communities also vary
in the degree or level to which they control their own regulative
institutions as self-sufficient units, and this variation corresponds
directly with their subordination to and dependence on govern-
mental, religious, and economic institutions and processes which
regulate local life and integrate local groups within the broader
framework of a wider society. It follows, therefore, that corres-
ponding variability obtains between communities with respect to
the levels of their integration as local units on the one hand, and
their integration within the society on the other. This variability
forms an important aspect of the processes of secularization, dis-
organisation, and individualisation, which Redfield noticed in
the Yucatan communities which he studied. But it is clear that
of themselves community studies provide an inadequate basis for
the construction of a societal typology, since variability in the
local control of regulative institutions, and hence of local inte-
gration, corresponds to variation of levels and degrees of integra-
tion within the societies of which communities are parts.
The significance of Redfield's work for British Caribbean
sociology can hardly be overestimated. Beckwith, Edith Clarke,
Cohen, Herskovits, Madeline Kerr, Matthews, Taylor, and R. T.
Smith have all been concerned with problems of folk culture and
its relation to the dominant traditions of various British Carib-
bean territories in one way or another (36). Hadley and Mac-
millan have generalised distinctions on this basis, and applied
them as exploratory categories in formulating general models of
(35) Levy, M. J., 1952.
(36) Beckwith, 1924, 1929: Cohen, 1953, 1954: Clarke, 1953: Herskovits, M. J., 1941:
Herskovits, M. J., & F. S., 1947: Kerr, 1952: Matthews, 1953: Taylor, 1951:
Smith, R. T., 1953.
British West Indian society (37). The Editors of Caribbean
Quarterly have also directed attention to this cleavage, and Broom
has indicated the necessity for specific studies of urbanisation
within the area. (38). On the other hand, analyses of society
in Jamaica and Trinidad have been presented in terms of class,
caste and colour concepts, without reference to the polarities of
'folk' and 'non-folk' (39); and Henriques has challenged Kerr's
statement that a cultural conflict obtains between folk and elite
cultures in Jamaica (40). It seems clear that differences of
definition, research problems, and theory may all be involved in
this conflict of analytic models, and that some conceptual and
theoretical clarifications are requisite at this stage in the develop-
ment of British Caribbean social research.
It is an indication of the general resemblances between
British Caribbean societies and those of the wider area initially
delimited that similar ambiguities about the appropriate analytic
models for communities in this wider area should also obtain.
Thus Haitian society has been analysed in terms of class and
caste stratification, as well as cleavages between the urban and
folk cultures (41). Brazil and the Southern United States have
received similar treatment (42). So has Central American
society (43). The point at issue here is not simply the superior-
ity of these competing models-stratification, and folk-elite cul-
tural differences-since each may well be more appropriate for
particular problems. Rather we are concerned with their com-
bination to produce a richer and more balanced analysis that can
be expected from researches conducted within either of these
frames of reference separately. Where, as in Haiti, stratification
coincides with marked cultural differences, there is almost no
problem of the type we are now considering, and comparability
of the research with other studies in the wider area of our
concern is assured, whatever analytic model is employed. But to
the extent that stratification and cultural differentiation are not
patently coincident, then considerable problems of interpretation
and conceptualisation invest the analysis, and competition rather
than comparability of models and materials may result.
At present, the debate over Redfield's theory seems to have
shifted towards the question of the cultural status of the alterna-
(37) Hadley, 1949: MacMillan, 1936.
(38) Caribbean Quarterly, 1953, vol. 3, no. 1. p. 3; cf. also, Mayhew, F. ibid, pp. 13-24.
(39) Braithwaite, 1953, 1954: Henriques, F., 1951, 1953b.
(40) Henriques, F., 1953a.
(41) Simpson, 1941, 1951, 1952: Herskovits, M. J., 1937: Leyburn, 1941: Lobb, 1940.
(42) Pierson, 1942: Eduardo, 1948: Lynn Smith & Marchant, 1951; Wagley, 1952,
1953.. Puckett, 1926: Davis, Gardner & Gardner, 1941: Dollard,
1937: Powdermaker, 1939; Park. 1950. (43) Redfield, 1941: Gillin, 1948, 1952:
Tumin, 1952: Wagley, 1941, 1949.
tive poles of his dichotomy. The problem here is not
whether the use of ideal-type constructs is valid or fruitful, but
whether the particular constructs offered by Redfield are the most
appropriate for field research. The criticism has been made
that the culture of communities such as Chan Kom is most ap-
propriately handled in terms of the culture of the inclusive society,
and hence may be regarded as the sub-culture of a community
within that unit. Such a view recognizes the significance of cul-
tural and political autonomy in theoretical typologies, and in
comparative analysis of these types of data.
Recent work in Puerto Rico promises to clarify some of the
problems raised by Redfield, together with a good many other
themes in Caribbean sociology. Steward and Manners, working
from an ecological basis, have indicated the existence of at least
three types of regional culture in rural Puerto Rico, associated
with the cultivation of sugar, coffee, and tobacco respectively
(44). Mintz has directed attention to the plantation as a type
of rural social unit with urban or industrial characteristics, and
there are important differences between plantations themselves
(45). i Clarke and the team of the Jamaica Social Survey
hav pund responding regional variations in culture, social
organisation, an ecology. Accounts of Cuba and Haiti indicate
regionaldifferen es there also; and it is clear that the study of
r eonalvari ins in terms of ecological factors has much to re-
commen methodologically, and can be expected to yield signi-
ficant comparable results.
Nonetheless it is as well to call attention to the problem of
conceptualisation involved in these studies of regional culture
variants. This problem is largely implicit in Redfield's distinc-
tion between urban and other cultures, since localisation in a
particular type of community, the city, characterises a different
kind of culture and society in Redfield's dichotomy. The point
however is that culture may vary on a regional basis while en-
tailing no significant differences of kind or of societal type. Such
regional culture variants must be distinguished sharply from the
type of cultural variation which entails possible differences of
societal type, and which are linked with horizontal stratification,
where they obtain within a society culturally segmented on a
regional basis also. If the term 'sub-culture' is applied to these
regional variants, then to avoid confusion between such local
differences, and the other type of horizontal cultural differentia-
tion, some other term must be adopted for these (46). If we
choose to describe these horizontal strata as classes or 'castes',
(44) Steward, J., & Manners, R., 1953.
(45) Mintz, 1953a, 1953b.
(46) Steward, J., 1950, p. 116, 140-3.
or in terms of folk and elite, and if they are found to exhibit
cultural differences of an order sufficient to distinguish one cul-
ture from another, we could then apply terms such as class-cul-
ture, caste-culture, folk and elite cultures, to distinguish their
ways of life. Clearly, where the two are found together, this
type of cultural differentiation is of a different type from regional
variation, and it invites confusion to treat them as equivalent
(47). Such an equivalence does violence to social space, re-
ducing its contemporary dimensions from three to two. Thus
problems of substance which are involved in this differentiation
of vertical and horizontal cultural variations have been some-
what obscured in Redfield's formation of the folk-urban
continuum; but terminology should aid rather than obscure their
Psychological studies and interpretations.
Problems of cultural and social differentiation, and of their
handling, can be appreciated from a review of the psychologically
oriented literature on the Caribbean and adjacent areas. Since
a good deal more has been published on the psychological con-
tent and aspect of social relations among the Negro-White popu-
lations of the United States and the Caribbean, than about the
mixed Indian-White communities of Middle America, we shall
confine our discussion to psychological studies and interpreta-
tions of mixed Negro-White societies. From the outset it is
important to stress the distinction between psychological studies
on the one hand, and interpretations which employ psychological
concepts but do not represent the results of psychological field
research on the other. A brief summary of the writings about
to be examined will direct attention to this difference.
Since the pioneer studies by Dollard and Powdermaker of a
cotton-producing area in Mississippi, formal studies of social
psychology have been made in several Caribbean territories, in
St. Thomas by Campbell, in Jamaica by Madeline Kerr and
Yehudi Cohen, working independently, and apparently by Rhoda
Metraux in Haiti. Psychological interpretations lacking bases of
formal psychological research have been applied to St. Martin
and St. Eustatius by Kruijier, to the British West Indies en bloc
by Simey and Hadley, and to the Trinidadian middle-class by
Braithwaite (48). Clearly these two sets of work rest on dif-
ferent foundations and require different treatment.
The geographical spread of these studies is almost matched
by their differences of psychological orientation. Dollard and
(47) Cumper, G., 1954.
(48) Dollard, 1937; Powdermaker, 1939. 1943: Campbell, 1943: Kerr, 1952: Cohen,
1953, 1954: Metraux, R., 1952; Kruijier, 1953: Simey, 1946; Hadley, 1949;
Powdermaker gave detailed treatment to the psychological con-
tent of race and caste relations without attempting to formulate
overall personality configurations for the population studied.
Lloyd Warner in an early study of Chicago Negroes developed
a specialised concept of 'social personality types' from the con-
sideration of the individual life-changes of American Negroes,
and found thirty-two such types among his sample (49). Braith-
waite applied the concept of the Authoritarian Personality to one
social class in Trinidad. Campbell, Kerr, and Kruijier on the
other hand, have attempted to define 'basic personalities' in the
populations which they studied, using methods and concepts for-
mulated by Kardiner. Simey and Hadley have both written
about British West Indian 'basic personality' types, while em-
ploying the frustration-aggression theories of Dollard and others
in their formulation. Campbell has also supplemented Kardiner's
methodology by constructs derived from Lewin. Cohen's analysis
of interpersonal relations within a Jamaica community represents
yet another approach.
Problems of social and cultural differentiation and their
handling are of central importance in the consideration of these
psychologically oriented studies, and have received the clearest
formulation in Warner's work. Since the determination of social
structures, groups, and classes must precede their psychological
investigation, it is thus necessary to consider the treatment of
social differentiation first, before proceeding to discuss the psy-
chological interpretations proposed for such patterns. In this
respect, the first point to note is that the studies listed above
were made in several different types of society. Chicago, we are
told, operates a system of informal caste. In Mississippi caste
is traditional, highly formalised, and ubiquitous. In St. Thomas,
the Americans have recently introduced their caste pattern to a
society formerly organised otherwise in terms of wealth, colour,
and culture. In Haiti, however much writers may dispute the
nature of stratification in terms of caste or class, all agree that
cultural and social differentiation coincide. In discussing strati-
fication in Trinidad, Braithwaite describes it in terms of caste,
semi-caste, and class, while also assigning a prominent place to
colour differences. Kerr, whose psychological studies were main-
ly of the Jamaican folk, distinguishes three classes in Jamaican
society, and points out that these are correlated with cultural and
colour differences, such that the conflicting cultural requirements
which obtain are productive of personality disorientations and
disorganisations under certain conditions. As mentioned above,
this coexistence of diverse cultural traditions in Jamaica is ques-
tioned by Henriques, who submits no evidence to the contrary
(49) Warner, L., et al, 1941.
C. A. 3
however, and in his conceptualisation of Jamaican culture in
terms of 'syncretism' would seem to indicate that Kerr's observa-
tions may reflect concrete conditions (50).
Within the small community of 270 souls studied by Cohen,
three classes were distinguished, primarily in economic terms.
Cohen also adds that even the upper class of this community
would "occupy a lower class status in the urban areas" (51).
Simey conceived of West Indian social stratification in terms of
three classes, but described one "basic West Indian personality
type", and has been criticised for this by Hadley on the basis of
data from St. Vincent which indicate the existence of three major
strata with marked cultural differences, and personality types
postulated to correspond (52). It would seem therefore that
outside the United States with its formal or informal caste
systems, there is little unanimity about the nature and significance
of differentiation within the societies examined.
A brief consideration of the two finest studies now under
consideration, those by Dollard and Powdermaker, will serve to
bring this issue into sharper focus. Both these workers conduct-
ed independent field researches in the same community of
Mississippi, and attempted psychological analyses of the social
milieu. Both recognized the importance of social differentiation
within the community, and the necessity for determining its form
and extent as a preliminary to the psychological study of social
and race relations. The 'caste' system was clearly indisputable,
but within either caste both writers report differently the existence
and extent of class divisions. Dollard regards the white popula-
tion of the community as being mainly middle-class, with no local
representatives of the Southern upper class, and only Poor
Whites or 'red-necks', who live outside the township studied, in
the white lower class. Powdermaker finds representatives of the
white upper class present also. Among the Negroes Dollard
finds two major divisions, a middle class, and a lower class, while
Powdermaker reports the existence of a Negro upper class also,
and implicitly subdivides the Negro middle and lower classes.
It seems that whereas Dollard gives his classification a primary
regional reference, Powdermaker's was primarily local; but there
are other differences in the two concepts of class with which these
writers worked. Dollard's classes seem to have the same refer-
ence or connotations on both sides of the caste line, whereas it
appears that the class subdivisions which Powdermaker employs
have different meanings or referents on either side of the caste
(50) Henriques, 1953b, 1953c, p. 172. (51) Cohen, 1953, Ms. Ph.D. thesis. (52)
Siapy, 1946, pp. 90-105, esp. p. 99: Hadley, 1949.
These alternative schemes naturally have significant implica-
tions for the psychological analyses which these writers present.
Thus Dollard defines the race problem as presented primarily by
the patterned relations among and between middle-class Whites
and lower-class Negroes, although he also discusses the situation
of middle-class Negroes. Powdermaker on the other hand finds
a more diffuse, variable, and less dramatically articulated set of
patterns, corresponding to her initial view of the greater degrees
of internal differentiation of both groups.
This comparison is made here simply to highlight the critical
importance of differentiation and stratification for social-psycho-
logical studies of communities, and to indicate that even after
careful investigations by able workers carried out in an area
of highly formalised social differentiation disagreements concern-
ing the nature and type of social differentiation remain which have
significant influences on the psychological analyses themselves.
Even admitting that different emphases or objectives may pre-
dispose workers to favour different systems of classification for
the same population, we are simply left with the question of the
expository, heuristic, or substantive nature of the categories in-
volved, and their relations (53).
Another aspect of these researches of Powdermaker and
Dollard which is worthy of mention reflects the nature of the
social system in which they worked. This system, that of the
Southern United States, presents a rigidly organised set of race
relations, and equally rigid definitions of racial membership. In
an important sense therefore, Dollard and Powdermaker were
carrying out studies of 'race relations' under conditions of their
precise and elaborate articulation which are not usually found
in other populations of the wider area which forms the natural
frame of reference for Caribbean social research. On this ground
alone it would seem quite likely that the conclusions reached by
Dollard and Powdermaker in Mississippi would be unlikely to
hold generally elsewhere in this area, and that their application
to societies with less rigid systems of racial differentiation would
be somewhat incautious. In fact, however, Simey has relied
heavily and uncritically on Dollard's formulation of Southern
Negro frustration-aggression reactions to their racial situation,
and has applied this in a generalised form to the British Caribbean
at large, with little attention to variations between or within
colonies. This is simply one instance of the growing number
of loose applications of concepts and theories developed by care-
ful psychological studies of quite distinct field conditions to the
description of West Indian social life. Social and psychological
research will both suffer if this vogue should continue unchecked.
(53) On this point, see also, Gross, L., 1949.
The fact that the frustration-aggression hypothesis lends
itself to such facile application as that made by Simey is of itself
sufficient to raise questions concerning the degree of precision
in the definition of key terms, or in the delimitation of processes
and effects, covered by that theory. Simey, for instance, argues
that submissive behaviour is a form of aggression (54). Powder-
maker also writes of a young Negro girl who is "aggressive in
her determination to hold her head high and break away from
the lowly position to which the whites would condemn her" (55).
Thus both self-respect and its opposite, self-abnegation, are in-
stances of aggressive behaviour in similar contexts; and one is left
to wonder what isn't. Key terms which are used helter-skelter in
such a fashion are clearly of dubious analytic value.
Divergencies in the psychological interpretations of West
Indian social patterns as well as in the procedures of their inves-
tigation are probably as great as any community of method or
findings among the psychologically oriented writers on this area.
It seems to be generally assumed, probably on the basis of the
two Mississippi studies just discussed, that aggressiveness charac-
terises a great many areas of social relations in the Caribbean,
though relatively little attention is usually paid to the frustrations
required by theory as correlates of this. Hadley and Campbell
however point out that aggressiveness varies in its expression,
incidence and intensity according to the status of the individuals
concerned. Kerr makes little use of this concept in her study of
the Jamaican basic personality, by which she perhaps means the
'peasant' personality type, and which she finds to be extra-puni-
tive (56). In his study of a small Jamaican community Cohen
presents a picture of a group of persons markedly hostile, in-
secure, dependent, and fearful. Braithwaite, discussing authori-
tarian characteristics of the Trinidad middle-class in a quasi-psy-
chological manner, also stresses anxiety and insecurity, together
with drives for power and submission. Rhoda Metraux traces
Haitian individualism in political and social life back to certain
traumatic adjustment situations of infancy and childhood. Camp-
bell employs Lewin's concept of aspiration and achievement levels
to subdivide the population of St. Thomas into three groups,
those with high aspiration and achievement, those with low aspira-
tion and achievement, and those with high aspiration and low
achievement, the last being a situation productive of both frustra-
tion and aggression. This attempt to extend the analysis beyond
the circular frustration-aggression reaction by delimiting the types
of situation under which frustration develops is highly commend-
able as an aim, though unsatisfactory in performance. It is not
(54) Simey, op. cit., p. 96.
(55) Powdermaker, 1939, p. 334.
(56) Kerr, op. cit., pp. 165-174.
easy to demonstrate that persons enjoying favourable conditions
are any whit less aggressive than those who do not, although
their frustrations may be less immediately obvious. Nor is Camp-
bell's omission of the fourth category of Lewin's conceptual system,
those with low aspiration but high achievement, entirely satis-
factory. Social systems being what they are, such a category is
not likely to be altogether devoid of empirical reference, and in-
deed, if aspiration is relative to the individual's situation, and if
achievement is measurable in terms of social recognition, (how
else?), then on Campbell's data, both the immigrant American
and Creole groups of St. Thomas contain elements classifiable in
those terms. On the whole, however, it would appear that such
highly general categories could be applied a priori to any human
or animal population whatsoever, yielding similar results, and
thus saving a good deal of research.
As a group these psychological studies and interpretations
are therefore subject to two main criticisms, the generality and
vagueness of their various theoretical and conceptual systems, and
the ambiguities in their classifications of the populations studied.
It is also of interest to note how variously these students stress
different aspects or stages of social life. Dollard and Powder-
maker, for example, give the psychologically primordial infancy
situation scant attention. Cohen deals with the early period
most methodically and effectively, but largely ignores the school,
which Miss Kerr finds to be a fairly efficient agency for personality
disorganisation in Jamaica. Hadley regards the past and present
situations of the British Caribbean populations as of such mo-
ment for understanding contemporary personality patterns that
he devotes half of his paper to these topics, the remainder being
a brief summary of the three main personality types as revealed
by this survey. Simey indeed seems to have made little direct
or systematic study of the West Indian situation with which his
book deals, but was content to apply concepts drawn from studies
in America to the Caribbean area in a manner not strikingly dif-
ferent from the application of stereotypes. In somewhat similar
fashion, Kruijier has simply adopted Campbell's work on St.
Thomas Negroes as a convenient frame for the discussion of St.
Martin and St. Eustatius, while Braithwaite has done the same
for Trinidad's middle-class with the concept of the Authoritarian
Personality. The apparent ease with which these psychological
concepts lend themselves to wholesale projections in alien areas
without prior research strongly suggests that their formulations
may be defective in rigour, or else are not fully understood by
those applying them; and it also suggests that such concepts and
interpretations may possess more significance as an expository
device in the hands of certain writers, than as substantive state-
ments about the populations referred to. Bearing this in mind,
it is interesting to note the agreement obtaining among these
students who lay such different stresses on different phases of the
psychogenetic process, or on different situations expressive of
personality traits, in the definition of West Indian personality
Stratification and Differentiation.
Commencing with a review of studies which focus on cul-
tural contact and change, we have moved gradually to the other
plane of analysis offered by sociology and have now to deal
directly with the problems of internal social differentiation. As
a glance at the literature will show, these are among the most
complex phases of social reality, and a great many conflicting
views are held about their nature, interpretation, and appropriate
methods of investigation (57). However, we have found that
without some adequate conceptualisation of this phase of socio-
cultural reality, systematic comparative studies within the Carib-
bean or its geographical context are hardly possible on the one
hand, while even the intensive studies of small districts and com-
munities have unclear general implications and references on the
other. Cultural variability and psychological problems gain in
significance and precision from their definitions in terms of par-
ticular situations, structures, and groups. Thus, although alter-
native interpretations of social life might be offered by cultural,
sociological, and psychological studies, the competition of these
analytic planes, and of the models constructed within their sepa-
rate frames, is ultimately more apparent than real; and in the
last resort, we shall only approach an adequate and refined
analysis of human problems or relations by a combination of
these and other approaches. The need for this combined ap-
proach is well known of course; what differences of opinion exist
on this subject mainly reflect divergences about the form or
method of such interdisciplinary work.
We have seen that studies of cultural process and change
lose a great deal when conducted without equally intense examina-
tion of the social milieu within which these developments occur,
and to which they give form; also that societal differentiation
cannot be handled apart from equally intensive studies of the
forms and processes of cultural differentiation which it implies;
and finally that psychological analyses of social systems depend
for their utility in large measure upon adequate prior analyses
of the structure of such systems, if they are to avoid errors of
(57) Beals, R., 1953a, 1953b: Bendix, R., & Lpset, S. (Eds) 1953: Blanshard, 1947;
Braithwaite, 1953: Davis, Gardner & Gardner, 1953: Goldschmidt, W., 1950:
Gross, 1949: Henriques, 1953c: Hill, M., & MacCall, L., 1950: Hughes, E. & H.,
1952: Levy, 1952: Pfautz, H., 1953: Parsons, in Bendix & Lipset, (Eds) 1953;
Centers, 1949: Park, 1950: Warner, W. L., Meeker, M., & Bells, K, 1949.
reference as well as interpretation. On these grounds it would
seem that the order of objectives in a combined interdisciplinary
study of society would give high priority to the determination
and analysis of the social structure, defining its major units and
their interrelations in a systematic fashion which would facilitate
the integration of cultural and psychological studies. Clearly
enough, the alternative to this procedure is the method of hit or
miss, which has already enjoyed an unhappy monopoly for too
long. But there is little agreement about the nature, field, and
content of social stratification and differentiation. Since these di-
mensions are basic to the formulation of structural models of com-
plex societies of the type with which we are concerned, lack of
agreement here has been largely responsible for delay in the de-
velopment of an integrated interdisciplinary approach to these
Social stratification is nowadays conceived of as a continuous
hierarchic ranking of social positions and roles (58). It is some-
times also regarded as an inherent attribute of all societies; but
this view is not empirically tenable. Such a postulate is only
essential to general theories which seek to 'explain' stratification
as inherent and implicit in the concept of societies as systems of
a particular type. If we look behind such theories to the prob-
lems which led to their formulation we shall arrive at a better
understanding of both.
The recent history of thought about social stratification can
be said to commence with the work of Karl Marx. He defined
classes in terms of their relations to the means of production, that
is, as economic categories within the society. Marx also pos-
tulated a continuous conflict between these economic classes,
which thereby implied their existence as solidary and self-con-
scious social units in greater or less degree. That is to say, Marx
commenced with an abstract operational classification of persons
on the basis of economic criteria, and concluded that these cate-
gories were substantive social units. Marx's procedure illustrates
one of the critical problems in this field of study, the problem of
distinguishing between operationally defined 'classes', and actual
social classes; or of ascertaining their correspondence.,
In the past forty years, many alternative formulations of the
class concept have been offered, some of which are given in the
book by Bendix and Lipset referred to above. At the same time,
other dimensions than those selected by Marx of importance to
the development and study of class systems have attracted notice;
and the problems of the interrelation of these dimensions have
occupied a central place in theory and research alike. Teimino-
(58) Levy, op. cit., p. 164 ff. Parsons, op. cit.,p. 93.
logical shifts have accompanied this refinement of the field of
study; and nowadays one speaks of economic or social or political
classes, of elites, estates, and status, of prestige classes and par-
ticipation classes, of reference groups, and the like. Each of
these terms and concepts reflects certain criteria relevant to the
existence or study of classes, and these criteria are often closely
related, so that the attribution of any single order of weighting
and priority among them raises many problems.
This summary suggests the type of problems which lie be-
hind certain current approaches to the study of stratification.
There are problems inherent in the notion of classification itself,
as an analytic process. Are social classes simply analytic pos-
tulates of various writers? Do they have any substantive exist-
ence, and, if so, how? There are further problems concerning
the constitution of class-concepts themselves, which centre upon
the relation of the various criteria recognized as somehow asso-
ciated with class organisation. There are problems of the boun-
daries of social classes, of their functions as well as bases, and
of their definition by objective indices of various kinds on the
one hand, or by subjective identifications on the other. Above
all, there is the question of their existence, which is implicit in
disagreements about their definition and nature. One does not
question the objective existence of lineage relations in an African
tribe, however one disputes their interpretation, simply because
these lineage relations recur constantly and in regular patterns
during studies by different investigators. One does doubt the
reality of the numerous human 'instincts' postulated by various
psychologists simply because their lists show so little agreement.
Similar difficulties arise in the study of class, and the layman can
rightly ask the sociologists, "How are you all so certain of the
existence of class, when unable to agree among yourselves in what
The type of theory which Parsons advances seeks to meet
these problems among others, by differentiating between value-
systems in a way which is consistent with, and to some extent
reflected in, the conflicting interpretations or models of class or-
ganisation that currently obtain. Thus Parsons' four primary
value-standards or systems allow of coexistent conflicting inter-
pretations of the system of stratification within any society, to the
degree that these value-standards are themselves tightly or loosely
integrated. Different rank-orders may therefore obtain according
to which of these standards are under consideration. But Parsons
also holds that these standards themselves are always ranked in
some order of priority, and that the distribution of power accords
with this ranking between them, and with the prestige of persons
or groups which also reflects this ranking. In other words, such
a theory seeks to accept and explain differences of opinion about
class boundaries and bases by postulating a system of multiple
valuations, which is nonetheless itself hierarchic in structure. And
this rank-order of the primary value-standards is conceived on an
analytic ground as inherent in the notion of social systems as
such. "The principal criterion of priority of evaluation of func-
tions, hence differentiated sub-systems, is strategic significance for
system-process." (Parsons, op. cit., 1953, p. 110, his italics).
This simply means that insofar as societies are systems they must
each have a continuous rank-ordering of roles and.positions.
As already observed, however, this generalisation is not al-
ways borne out, so that either some societies are not systems, or
they are systems of a different type from Parsons' models. In
certain tribal societies for example, half the population, women,
are often excluded from ranking which obtains among males;
and this ranking frequently, as among the Hausa, or in certain
lineage systems, accords equal value to positions discharging dif-
ferent roles, as well as the contrary. Empirical materials of this
type may be insufficient objection to Parsons' theory on either of
two grounds firstly, that such societies are not systems of the type
or in the sense of his discussion; secondly, that, although they are,
their analysis has failed to define their four primary value-stand-
ards, as well as the content of each, or their interrelation. To
these possible rejoinders, there are two replies. Firstly, by what
criteria other than field study itself can one establish whether any
society is a system, or a system of the particular type under dis-
cussion? Secondly, by what criteria, other than the system of
stratification itself can one establish the rank-ordering of the
various value-standards, or indeed their separateness and content?
For clearly, if the theory which purports to 'explain' certain phe-
nomena assumes but cannot predict them, then it simply consists
in their description from one particular point of view.
This is not the place for a detailed discussion of Parsons'
theory of social stratification. It has been quoted merely to il-
lustrate current thinking about these problems, and the com-
plexity which they present. But before proceeding to discuss
empirical researches, three observations are necessary. Firstly,
Parsons discusses stratification in terms of roles of performances,
which accords with the action frame of reference he adopts, but
may not always be appropriate to or consistent with ranking as
it goes on in empirical societies, some of which at least accord
priority in rankings to positions, rather than roles, defining these
consistently or otherwise in terms of one or more criteria. Be-
cause Parsons defines stratification problems in terms of roles,
and ranks roles in terms of their "strategic significance for
system-process", he implicitly assumes that stratification, besides
being continuous with the society, unidimensional and integrative,
also has positive functions essential to the system. None of these
imputations is easily demonstrable by empirical materials, al-
though each of them may be controverted by such data.
Secondly, Parsons' model of stratification develops from an
abstract consideration of social systems as equilibrium systems
with particular levels and types of integration. It seems however
that levels and types of integration are properly a matter for
empirical research, and that they cannot be postulated in any
precise sense for societies at large, or for any single society at
different points of time. Such postulation would lead to the treat-
ment of a system in rapid and violent change, for example, during
a period of revolution and counter-revolution, as exhibiting condi-
tions of equilibrium and integration at all instants equally, and
would also permit the entire process to be conceptualised in these
terms. Now clearly, predicates of this level of generality and
imprecision are of dubious value as guides to empirical studies or
their analysis. And the same charge of generality and imprecision
may be levelled at each of the value-standards in turn, not as for-
mal categories, but with regard to their content and substance in
any empirical society. Traits and functions classifiable in terms
of each of these four standards in one context may be classifiable
differently in another, and within the same society, as well as be-
tween societies. Such a condition simply reflects the multifunc-
tional nature of institutional activities, the multiplicity of their
goals, and the manifold of their inter-relations. To conceptualise
these diverse aspects in terms of phases reflecting basic differences
of value-orientations involves abstractions of an order which are
both so highly general as to be of doubtful value, and at the same
time so constructed as to mutilate the organic structure and
character of institutional activities.
Finally, the point must be made that the existence as well as
the nature of a single continuous system of stratification must be
established by field research in any particular society, rather than
predicted as an implication of theory. Until such a single series
is demonstrated, it is wiser to conceive the field of relations which
involve ranking and distinctions in terms of social differentiation,
which does not imply any particular form for the structures under
study, admits the possibility of a single hierarchy, reduces the
theoretical problems and overtones which invade such -studies,
and also keeps empirical research alert to the many factors and
aspects which are or may be involved.
Empirical studies of social stratification in America have
been conducted on the, basis of one or more indices of an objec-
tive kind, such as occupation, income, house-type, residential
area, or association and group membership, as well as by the use
of subjective evaluations and identifications of their own class
position, or those of others, as given by informants. These ap-
proaches are sometimes combined, as in the work of Warner and
his associates (59). The criteria selected in these studies indicate
the two primary orientations of the class concept, objectively to-
wards an economic and social base, and subjectively, through
identification and other references, as an expression of social
psychology. Hence the frequent definition of such classes in terms
of prestige, and the phenomenon of their ranking, as more or less
prestigious. Naturally, as one would expect, even the most sub-
jective classifications have a certain consistency in their objective
referents; but this is by no means a uniformity, and cannot be re-
duced to invariant relations between any of the various dimensions,
such as wealth, occupation, social participation, style of life,
family history, power, or the like, which are involved in the general
As regards the relation of class-difference to culture, there
has so far been little discussion. And this itself is one of the
most revealing aspects of stratification studies. Class in other
words assumes adherence of the groups which it differentiates to
common institutions, and thereby, in the last resort to a common
system of values. Classes are thus differentiated culturally in res-
pect of non-institutionalized behaviours, such as etiquette, stand-
ards of living, associational habits and value-systems which may
coexist as alternatives on the basis of the common values basic
to the class-continuum. Classes differentiated in this manner
reflect, besides their hereditary aspects, educational, economic,
and other differences also. But they are not definable in terms of
adherence to different systems of social institutions, since that
would imply their equation with societies, as distinct cultural
Now in the U. S. A., the criteria used in studies of stratifica-
tion among Whites have also been applied to Negroes. The sharp
racial lines and frequently deviant patterns of American Negroes
notwithstanding, there appears to be sufficiently general consensus
about the validity of such studies, to indicate that in America we
have two parallel social segments, one white, the other black,
linked together in various ways, among them by formal adherence
to common institutions. The nature of this racial division itself
calls for some consideration. It varies between regions and be-
tween states, but includes legal prohibition on inter-marriage, oc-
cupational differentiation to a fair degree, bans on commensality
in much of the area, provision of separate public facilities for the
two racial groups, and the definition of a Negro as any person
with one drop of Negro blood, or more. In terms of this defini-
(59) Warner, Meeker, Eells, 1949.
tion Negroes are a residual social category, and as could be ex-
pected, under such conditions, pure Negroes form a minority
within the American Negro group (60). Warner and his fellow-
workers have labelled this type of race-differentiation, 'colour-
caste', a term which directs attention simultaneously to its parallels
with caste systems as for instance obtaining in India, and its dif-
ferences. Caste implies uniformity in the ranking of units, and
thus uniformity in ranking members of such groups, which does
not always obtain in America. Thus across the American colour-
caste line, the poor White occupies an inferior social-class position
to the middle or upper-class Negro in many, though not in all, res-
pects. Under a caste system moreover, occupational inheritance
obtains which is also lacking in America. Individual social mobil-
ity is ruled out by caste, theoretically even where migrancy occurs,
and the system of caste stratification enjoys the sanction of re-
ligion. Caste is conceived of, in other words, as a sacred struc-
ture, and the disobedience to caste norms and inter-relations is
not merely unlawful, but also sinful. These latter features of the
true caste-system are lacking from the American scene, and con-
sequently Warner's terminology seeks to reflect this.
Within the American Negro group, as many researches de-
monstrate, differences of colour and shade are widespread and
of great significance in personal relations. This means that the
number of criteria adequate for a study of class-structure among
American whites must be increased to include the colour factor
in parallel studies of American Negroes; and the correlation be-
tween colour-ranking and other forms of status-determinant also
requires investigation. Readaptations of this sort clearly reflect
the ideological patterns associated with American race relations
on the one hand, and the continuities of economic, cultural, and
other factors which obtain between American Whites and Negroes
on the other.
It is clear that such conditions are distinctive of America
within our area. It follows therefore that conceptualisation of
social differentiation in other racially-mixed societies of this wider
field in terms of American patterns may obscure factors of greater
significance in the societies under study. Pierson's pioneer study
of Bahia, Brazil, which was oriented towards the problem of race
relations between the white, coloured, and black groups of that
city, illustrates the point very well (61). Pierson's particular
interest lay in the contrast between Brazilian and American race
relations. This led him to treat social differentiation in terms of
race, with little attention to the problems of cultural differentiation
which are prominent in Bahia, although race differentiation or its
absence are themselves cultural facts reflecting other cultural facts.
(60) Klneberg, (Ed.) 1944: Myrdal, O., 1944.
(61) Pierson, D., 1942.
Pierson's application of American concepts of stratification
and differentiation to Latin American conditions is not unique, as
Beals' review article makes clear. Beals concludes that "the use
of strictly economic or economic and political criteria for class
analysis of Latin America is the least useful approach" (62). He
directs attention to the differences in stratification and differentia-
tion between North and Latin America, and especially to the cul-
tural differences between the Indian and Mestizo or Ladino
groups within these Latin American societies. Redfield also has
only touched on these problems of social differentiation within a
community indirectly, so we shall have to look elsewhere for their
The Guatemalan village of San Luis, studied by Gillin and
Tumin, is a suitable example (63). Both workers were in the field
together, but carried out independent researches, which, although
overlapping a good deal, were focused on different aspects of
local life. In San Luis, two-thirds of the population are Indians,
and the remainder, of mixed Spanish-Indian descent, are known
as Ladinos. Within San Luis, Indians are subordinated to Ladi-
nos, individually and as a group. Beals, in the article just cited,
observes that "In Guatemala, an anomalous situation exists, for
horizontal movement is often possible for the Indian when he
leaves his community; within his community, however, movement
into the mestizo (here called Ladino) group is regarded as vertical
movement, and is virtually impossible" (op. cit., p. 338). Gillin
and Tumin describe this system of stratification in terms of
'caste', and both writers indicate that its posited racial basis is
largely without objective foundation. Although Ladinos and
Indians may not intermarry or eat together within the village, they
sometimes do so beyond its perimeter. There are few other fea-
tures at San Luis of the type usually associated with 'caste' in com-
parative sociology, except the uniformity of the dominance-sub-
ordination relations. On the other hand it is abundantly clear
from both accounts that very marked differences of social status
and culture obtain between Ladinos and Indians, and that the
type and degree of their social differentiation is itself a cultural
fact which is linked with and reflects these other cultural dif-
In America for example, colour-caste obtains between Whites
and Negroes independently of cultural or class similarities or dif-
ferences (64). In San Luis, caste difference seems to consist in
the social expression of cultural dissimilarity, and seeks to per-
petuate this. The Haitian situation is structurally similar to that
of San Luis. Although disagreements occur among students of
(62) Beals, 1953, p. 339. (63) Gllin, 1951: Tumin, 1952.
(64) Warner, et al., 1941.
Haiti concerning the definition of Haitian stratification in terms
of caste or class, there is marked consensus on the existence of
two different cultures in that country, and also on the definition
of Haitian social stratification in terms of this cultural cleavage.
In effect this means that cultural cleavages of a particular order,
if associated with a rigid and traditional pattern of subordination,
may give the appearance of caste, when really of a different na-
ture. It also means that in populations characterized by wide cul-
tural variability, the analysis of social differentiation must proceed
together with the study of the cultural differentiation, in the same
way that the study of acculturation must proceed also by the study
of the social structures involved. Unless this co-ordination of
cultural and sociological study obtains, then acculturation study
is deprived of its social reference on the one hand, and stratifica-
tion is inadequately conceived in terms of race, caste or class on
Few explicit studies of social differentiation have yet been
made for British Caribbean societies. On the other hand there
are several accounts of these populations which interpret them in
terms of social and cultural divisions. Thus Martha Beckwith's
study of Jamaican folk-culture implicitly recognizes an important
cultural cleavage within that society by its exclusion of a consider-
able section of the local population from its reference. Miss Kerr's
study reveals similar assumptions, and she explicitly states
that this cultural cleavage, expressed as conflict, inhibits the
development of healthy personality types among many Ja-
maicans. For Dutch Guiana and Trinidad, the researches of
Herskovits into folk-life have similar implications, many of which
are explicitly formulated in terms of acculturation processes.
Clearly any acculturative situation presupposes the contact and
coexistence of two or more cultural traditions. Observations by
other students of Caribbean societies, such as Campbell, Reuter,
Gordon, Hadley, Pearse, Kruijier, Cohen, Clarke and Matthews,
indicate that this condition is general among Caribbean societies
(65). In Central America comparable levels of cultural differen-
tiation are reported by Redfield, Tumin, Wagley, Gillin, and
others, although interpretations vary from the folk-urban con-
tinuum of Redfield to the caste constructs of Tumin (66).
Writers with primary political or administrative orientations
also face this problem of interpreting Caribbean social structures
and culture. Blanshard, Proudfoot, Simey, and MacMillan direct
attention to the complexity of their internal differentiation in both
these fields, though differing about the significance of this condi-
(65) Campbell, 1943: Reuter, 1946: Gordon, 1950: Hadley, 1949: Pearse, 1952, 1955:
Kruijier, 1953: Cohen, 1953, 1954: Matthews, 1953: Clarke. 1953.
(66) Redfield, 1941: Gillin, 1949, 1951, 1952: Wagley, 1941. 1949.
tion (67). Olivier writing about Jamaica gave less attention to
these aspects of local society, than to the community of social and
cultural patterns laid down in its history. But publication of this
work was followed by major social upheavals in the island, which
contrast sharply with the implications of Olivier's study (68). On
the other hand, MacMillan's alternative interpretation of Jamaican
society, which directed attention to the instability and internal
tensions implicit in its differentiation, was borne out to a high
degree by events of the year which followed its publication.
Explicit studies of social differentiation in British Caribbean
societies have recently commenced with the work of Braithwaite,
Henriques, and Broom (69). All three writers are concerned
primarily with the Creole populations, Braithwaite discussing
Trinidad, the others Jamaica. Henriques and Braithwaite con-
ceive the field in terms of stratification, but Broom prefers the
more inclusive concept of differentiation. All three writers
direct attention to ethnicity and colour as criteria of importance
in the social structure, but give little explicit attention to cultural
differences within the population, or the relation of these factors
to the patterns of stratification or differentiation. These three
studies are therefore similar in their attack on a common problem
to the extent that they omit direct treatment of the cultural aspects
of this problem.
Beyond this, they frequently differ, despite concentration on
similar problems, and the use of common criteria such as colour,
occupation and wealth. Henriques defines Jamaican stratification
in terms of three 'colour-classes', while Braithwaite speaks of a
similar number of strata in Trinidad as castes, semi-castes, and
classes. Henriques divides Jamaican society among these classes
as follows; lower class, 85%; middle-class, 10%; upper class
5%; and concludes with the candid observation that, "The use of
the class divisions, upper, middle, lower, is a necessary methodo-
logical device, and does not indicate the actual divisions in the
society" (70). In other words his analysis may well have no ob-
jective reference. Braithwaite's position is very similar; and he
simply dismisses the problem of objective reference in a 'note on
numbers', consisting of license totals of various kinds, and income
tax assessments, which cannot themselves be equated directly with
the type of variable, colour, on which his analysis is principally
based (71). In other words for both these writers the empirical
boundaries of social classes, and thus their constitution, presents
little problem for stratification analysis.
Both writers agree that the system under discussion is one of
(67) Blanshard, 1947: Proudfoot, 1954: Simey, 1946: MacMillan, 1936.
(68) Olivier, Lord, 1936.
(69) Braithwaite, 1953: Henriques, 1953c: Broom, 1954.
(70) Henriques, 1953c, pp. 105, 161. (71) Braithwaite, 1953, pp. .102-2, f. p. 75.
continuous stratification, though one within which individual
movement across the principal class or caste boundaries is ex-
tremely difficult. Both urge that this system is continuous for the
population by virtue of the common valuations which it represents.
Thus Henriques summarises the Jamaican situation as follows: "to-
day the whole colour-class system is dependent upon the almost
complete acceptance by each group of the superiority of the white,
and the inferiority of the black" ibidd, p. 41). Braithwaite's view
of Trinidad is essentially similar: "here again, we see that the key
to the unity in the diversity of judgments is the acceptance of the
upper class as the upper class. In this case, however, we have the
main common values shared by all the groups in the society"
ibidd, pp. 52-3). Now this really consists in explaining social
stratification in terms of itself after first assuming that it forms
the general framework of these social structures, without detailed
empirical examination of this question. Agreement upon these
points however, does not rule out differences of their interpreta-
tion. Thus Henriques emphasises the disnomic or disbalanced
condition of Jamaican society, whereas Braithwaite sees Trinidad
undergoing a process of change from adherence to a paramount
common value system of a particularistic ascriptive type towards.
one stressing universalistic achievement. But Braithwaite notes.
cautiously that "the change may be expected to lead to a certain
amount of conflict within the system; and the contradictions be-
tween the rival systems of values will be likely to lead to tenden-
cies towards disintegration within the social system" ibidd, p. 170).
This can only be taken to indicate that the "common value-
system" is not truly common at all; and such a view is supported
by the data which Braithwaite provides illustrating the divisive
effects of the caste, colour, class patterns and values obtaining in
Trinidad. Given these divisions, the type of consensus which is
presupposed by common values has marginal significance com-
pared with the type of differentiation which obtains. It may well
be the case that recent changes in the contemporary situation of
Trinidad place primary emphasis on universalistic standards, and
even that certain elements of Trinidadian society would gain
by these; but such conditions refer to the context of Trinidad, and
do not directly warrant inferences about their acceptance by the
Henriques' conception of Jamaica as disnomic directs atten-
tion to the low level of integration within Jamaican society. This
can only mean that the hierarchy of social divisions as a
ranked series of white, brown and black strata, has a primarily
divisive rather than integrative function. Such a view does not
conform to those theories of stratification which define it in terms.
of integrative functions. Now since social integration is an aspect
of adherence to common institutions, such an association of strati-
fiction with disnomic conditions as Henriques observes in Ja-
maica would seem to indicate that institutional divergences char-
acterised the different strata in greater or less degree. And this in
turn would indicate that stratification is definable to a large ex-
tent in terms of associated levels of cultural differentiation under
such conditions. However we find Henriques demurring to Kerr's
conception of conflicting cultures in Jamaica, and describing Ja-
maican 'culture' in terms of synthesis and syncretism (1953 c, p.
172; 1953b., p. 62).
The question of importance here concerns the form
and level at which this synthesis or syncretism obtains. We
can illustrate this point from the work of Herskovits. When
Herskovits speaks of 'the Old World cultural province' as a unit,
he refers to certain generalized forms which obtain throughout
that area. In a sense therefore, he presents a synthesis of Old
World cultures of a very abstract character under the concept of
a single cultural province. Of course, Herskovits is quite aware
of the abstract and classificatory nature of such a synthesis. The
question'raised by Henriques' account of Jamaica is whether the
cultural 'synthesis' of which he speaks is not a conception of this
kind, a classificatory, methodological device, subsuming or ob-
scuring many diverse elements. On the evidence which he pre-
sents, this would seem to be the case. "Assimilation of European
culture for the middle and upper classes has been successful....
The problem for the lower class in the elaboration of its institu-
tions has been to endeavour to find avenues of expression denied
it by the greater society." ibidd, p. 172). This can only mean that
the institutions of the lower class differ from those of the upper
and middle classes, in which case the Jamaican cultural synthesis
is a heuristic taxonomic device of the student, rather than a con-
crete observable field datum.
Similar problems of the level of generality and abstraction
invest the concept of cultural syncretism, as used by Henriques.
Cultural syncretism connotes the identification of elements
from two or more different traditions. Defined in this way the
concept has objective reference to defined complexes and traits,
and is clearly valuable, as for instance in the work of Herskovits
(72). In contrast with this definition in terms of specific cultural
patterns, Henriques sees to extend the notion of syncretism to
all aspects and forms of cultural process simultaneously. Now
even if this use of the concept was accepted, such a syncretism
would have to be demonstrated in any particular case. But Hen-
riques makes no effort to do this, and in fact, as can be seen from
the quotation just given, recognizes the institutional differentia-
(72) Herskovits, M. J., 1946, p. 350.
C. A. 4
tion characteristic of Jamaican society. It is clear therefore that
when he postulates cultural syncretism in Jamaica he does so on
grounds which do not derive from field materials; and these
grounds can only be theoretical or methodological in nature. Pos-
sibly these two conditions are integrated, and require the concep-
tion of a single cultural field, whether as synthesis or syncretism,
if the type of analysis Henriques offers is to be possible at all.
"Complete disnomia is difficult to conceive as such a state implies
an entire lack of order in which a society could not exist". ibidd,
Now the type of analysis which Henriques offers assumes a
single continuous system of stratification. This in turn implies
some level of consensus about values among the population, some
integrative functions of the stratification, and some level of cul-
tural synthesis or syncretism. The questions of interest here are
whether these assumptions are liable to empirical demonstration;
whether they derive from empirical materials or deductive theo-
rems; whether in fact the nature and form of social differentiation
in Caribbean colonies may not be more complex than such
analytic schemes can handle adequately. Leonard Broom's view
of Jamaica would seem to imply that this is the case; "Social
stratification in Jamaica cannot be understood as an uninterrupted
continuum of status positions. No matter what empirical criteria
are employed, gross discontinuities are to be found. Given the
historical forces briefly outlined, this fact should cause no surprise,
but the extreme character of this status cleavage affects all facets
of Jamaican society". (1954, p. 119).
Both Henriques and Braithwaite devote great attention to
colour differences in their studies of stratification. But neither
appear to handle this problem consistently or systematically. Thus
Braithwaite identifies his classes with castes, and defines both in
terms of colour factors; but his diagram of the colour distributions
obtaining in the contemporary stratification of Trinidad society
indicate considerable departures from these definitions. (op. cit.
pp, 47, 156). Henriques similarly defines his strata in terms of
colour-class, and notes that 'this colour-class division is not at all
rigid' (1953c, p. 42), wealth being an important variable; but
he gives a good deal of evidence which suggests rigidity in several
spheres. "Each colour category has a series of stereotypes for
all the other categories, and for itself" ibidd, p. 52); and he
concludes that "in fact colour can be said to pose the whole
problem of 'cultural' values in the Caribbean" ibidd, p. 168).
If this is so, then a systematic analysis of the colour complex
would appear to form an essential preliminary to the develop-
ment of any adequate models for the study of Caribbean societies.
And it is to this analysis that we must now turn.
Colour in the British Caribbean.
In most mixed Negro-White populations, the concept of
colour is critical and pervasive, hence we can expect on general
grounds that it may have several distinct though overlapping
referents. A systematic analysis of the colour concept therefore
consists in the isolation of these different meanings, and the
determination of relations between them.
As normally used in the British Caribbean, the term 'colour'
connotes a combination of physical characters, such as skin-colour,
hair-type, form of facial features, prognathism or its absence, and -
so forth. This is the sense in which Braithwaite and Henriques
use the term, and will be referred to here as phenotypical colour.
The phenotypical colour of an individual is simply his or her
racial appearance. In British West Indian colonies there is a
clear overt rank-order of different phenotypical colours in terms
of a prestige scale, which places white phenotypes at the highest
and black phenotypes at the lowest points. The phenotypical
colour of an individual is therefore a factor of importance in his
status placement in these societies.
But the matter does not rest there. Often enough, someone
who is phenotypically black will claim that he has 'white blood',
while someone who could be mistaken for white is known to
have a mixed ancestry. These conditions direct attention to the
biological variation of phenotype associated with varying degrees
of racial mixture, and thus to the difference between genealogical
colour-the biological status of an individual as defined in terms
of his purity or mixture of racial descent-and phenotypical colour ct
or appearance. The classic instance of status ascription on the
basis of genealogical colour in our field is of course the American
system of colour-caste; but although British Caribbean societies
do not formalise status ascriptions solely in terms of this aspect
of the colour concept, it receives attention in all of them. Two
phenotypically equivalent individuals may be dissimilar in terms
of genealogical colour, while two genealogically identical in-
dividuals, such as brothers, are frequently phenotypically distinct.
This variability in the association of phenotype and genealogy
guarantees that, where colour values loom large in social organi-
sation, attention will be directed to both these conditions, even
although they may not be generally distinguished with any con-
sistency or precision. Between them, these two categories define
and exhaust the genetic aspect of colour.
By themselves, however, phenotypical and genealogical
colour can have no direct or necessary significance in society.
This will be apparent immediately the lack of role differentiation
on grounds of colour in such racially mixed Muhammadan
societies as Hausa-Fulani Zaria is brought to mind. In Zaria,
light skin-colour is regarded as an attribute of beauty, but colour
differences as such form no basis for role differentiation or status
ascription (73). This example alone is sufficient to show the
fact that status and role allocation on a basis of colour difference
is subject to cultural determination. It also suggests that the
general concept of colour may be an inadequate or misleading
term of analysis unless its social and cultural components and
their significance are clearly distinguished.
The social aspect of colour differences refers to the part
which they play in individual and group association. It is thus
to be determined by empirical study of formal and informal
associations within a society, and can therefore be expected to
vary from one society to another. We can most briefly isolate
this dimension of the colour complex under the term associational
colour. The associational colour of an individual is simply the
expression of his associational habits in terms of the colour of
the persons with whom he typically associates on terms of
equality, familiarity, and intimacy. Thus a white man who
habitually associates with black persons is associationally black,
while a black man who habitually associates with white persons
is associationally white. Now there is room for misconception
of the associational colour of an individual, but this reflects mis-
conceptions about the nature and type of his associational habits.
Frequently enough, of course, an individual misconceives his own
situation in this way, and this can be expected especially of
mobile individuals. The point to note is that the associations
which an individual forms, and in terms of which his associational
colour, or "class-colour," is estimated, may vary at different times
of his life, within limits which reflect the influence of other factors.
For instance, as sometimes happens, a person whose fortunes
change for better or worse may seek to change his associates
correspondingly, and by this social mobility either acquires a new
associational colour, or has to reconcile himself to a position of
relative isolation. The concept of associational colour therefore
permits a classification of individuals in terms of upward or
downward mobility on the one hand, and static or isolated posi-
tions on the other. That is to say, besides allowing a formal
classification of the population in terms of associational colour
categories, this concept focuses attention on the dynamics of
such systems. It goes a long way towards making possible the
direct and systematic study of differences between the observable
phenotypes of persons and their classification in colour terms
within such societies. For, where colour differences have social
significance, although the colour concept reflects this in its com-
(73) Smith, M. F., 1954: Smith, M. G., 1954.
plexity, there is normally a ranking of persons in terms of colour,
a sort of chromatic status scale, and it is in these terms that the
equation between an individual's associational and phenotypical
colour is defined.
So far nothing has been said about behaviour, although
associational patterns have behavioral content and aspects. Yet
behaviour per se, the institutions, mores, conventions, and value
patterns, which together constitute the culture of a population,
besides giving definition and cohesion to the social form, also
defines the colour complex itself. Hence behavioral patterns
are clearly significant to the definition of individual and group
statuses within society, and where social differentiation in terms
of relativistic colour concepts obtains, can be expected to in-
fluence or correspond with this differentiation in greater or less
The relevance of this point of cultural identity or difference,
real or assumed, for our present analysis can be readily appreciated
by consideration of the lack of status ascription on colour bases
in multi-ricial Islamic societies which for religious reasons are
intolerant of internal cultural differentiation. In contrast, role
and status ascription on racial and colour bases are characteristic
of multi-racial societies where the racial divisions are, or until
recently, have formed, culturally distinct groups. Under condi-
tions of the latter type, the alternative to the complete disnomia
which puzzled Henriques consists in an hierarchic organisation of
these culturally distinct groups, and the domination of one of
these racial or cultural groups over the others provides the basis
for a general prestige scale of their distinctive characters, pheno-
typical or other, such as we are now concerned to dissect. This
simply means that, where culturally divergent groups together
form a common society, the structural imperative for maintenance
of this inclusive unit involves a type of political order in which
one of these cultural sections is subordinated to the other. Such
a condition derives from the structural requisites of society on the
one hand, and the condition of wide cultural differences within
some populations on the other. It is under such conditions that
differences in race and colour acquire general social significance.
We can illustrate the relation between race and culture
briefly by considering the two opposite types of limiting case.
Muhammadan Zaria presents an instance of the type in which
cultural uniformity obtains within a multi-racial society, and race
and colour differences do not provide bases for status determina-
tion or role allocation. Although the governmental structures
are hierarchic in type, office is open to members of all races within
the society. Under conditions such as these differences of race
and colour lack social significance and hence are not socially
systematized. Cultural community overrides these racial dif-
The other type of limiting case obtains where culturally dif-
ferentiated groups of the same race form a common society,
characterized by the domination of one group by the other, and
the expression of this cleavage in racial terms. This seems to
be the position in contemporary Guatemala, since racial differences
which lack biological foundations are postulated between Ladino
and Indian. Similar conditions obtained in Britain after the
Norman Conquest, and persisted until the Tudor Period, when the
cultural differences of Norman and Anglo-Saxon having lapsed,
together with the form of feudalism in which they were expressed,
the concept of a general English culture, and of the Englishman
as distinct from either Norman or Anglo-Saxon, developed.
Later, similar racial differences were mooted as an aspect of the
cultural cleavage between the British and the Highland Scots, but
attention to these racial factors also lapsed with the destruction
of Scottish Highland culture in the years after 1745.
These instances show that race-differences may be predicated
even where of marginal character, provided that cultural cleavages
of a certain order exist within a society. They also show that
race-differences lack formal expression where cultural uniformi-
ties obtain between the two groups within a single society. It is
thus clear that hierarchic race relations reflect conditions of cul-
tural heterogeneity in the societies in which they
obtain, and that they tend to lapse or lose their hierarchic char-
acter as cultural uniformity increases. It follows that multi-
racial societies which invest racial or colour differences with social
status significance display cultural heterogeneity; and also that in
such societies, the dominant culture will have high prestige, the
For societies such as those of the British Caribbean, this
conclusion implies the existence of a scale of cultural colour, in
which 'white' and 'black' cultures provide the poles. Moreover,
it implies that all the members of such societies will be ranked
in terms of their behavioral conformity to one or other of these
traditions. Thus we arrive at the notion of cultural or behavioral
colour, that is the extent to which an individual's behaviour con-
forms to the norms associated with one or other of the hier-
archically ranked cultural traditions of the society, as these norms
themselves are associated with colour-differentiated groups. Thus,
in Jamaica for example, there is a set of expectations which
define the behaviours of whites, browns, and blacks. In terms
of these expectations, a white person whose behaviour conforms
more closely to that bf the brown population is culturally brown.
Similarly, a pure Negro may be culturally white or brown, which
simply means that his behaviour is analytically homologous with
that distinguishing these colour groups as cultural groups.
This type of colour concept is implicit in much of the
American literature, but requires explicit formulation for the
analysis of societies in which colour 'can be said to pose the
whole problem of cultural values'. Powdermaker illustrates the
concept when she introduces her discussion of Negro acculturation
in Cottonville with the following remark, "some Whites deplore
the process, and question how far it can go, believing the Negro
incapable of becoming 'sociologically White'." (1939, p. 354).
Broom makes a similar point in his recent analysis of Jamaican
social differentiation. "Certainly the differential statuses which
are all too apparent in the 1943 census are reinforced by the
selective perception of census takers. For example, a pheno-
typically black civil servant of the upper categories is most likely
to be classified as 'coloured'. A dark-coloured peasant is most
likely to be classified as 'black'." (1954, p. 117). In Grenada
the folk use the term 'African' to denote persons who adhere
to African cultural traditions, as for instance, ritual forms, such
as the African Dance (Shango), whatever their racial or colour
characteristics. At the same time, Grenadians describe wealthy
black persons who practise the same culture as that of the Whites,
Where two cultural traditions coexist within a society, and
their hierarchic relation does not closely approximate to caste,
then social mobility and acculturation proceed simultaneously
with sufficient volume to produce an intermediate cultural tradi-
tion, a sort of hybrid culture, which, although approximating
more closely to one of the two homogeneous original traditions,
contains elements of both, and by virtue of its hybrid nature
occupies a middle position in the cultural prestige scale. Where
the two terms of this cultural scale are identified with white and
black, this intermediate culture, the product of fusion, accultura-
tion, and syncretism, is classified as coloured or brown.
There is a tendency for genealogical and phenotypical colour
to correspond. There is also a tendency for associational and
cultural colour to correspond. This latter correspondence simply
reflects the fact that people tend to associate with those whose
behaviour conforms to the norms which they themselves hold,
rather than with those who hold different norms. But this latter
convergence is at one and the same time more open to the in-
fluence of other factors and more variable, than is the associa-
tion of phenotype and racial composition, which once given re-
mains unalterable. The nature, basis and range of variability
in this association of cultural and associational colour forms an
important aspect of the typological definition of the social systems
in which such multiple stratifications obtain. The same point
holds for the association between the cultural and social colour-
scales on the one hand, and the biological scales on the other.
In conditions such as American colour-caste, cultural uniformi-
ties between Whites and Negroes are not yet recognized as an
adequate ground for their equal association, although there is a
trend towards revision of the caste line as acculturation proceeds,
and in terms of our preceding discussion, we can expect 'caste'
to disappear as cultural uniformity extends to all levels of insti-
tutional behaviour. Where colour-caste does not obtain, cul-
tural colour, subject to certain other factors, such as wealth on
the one hand, and genealogy, or phenotype on the other, tends
to provide a basis for individual and group association of various
It is methodologically useful as well as revealing to assume
initially that there is a complete correspondence of cultural with
associational colour, and of both with the biological dimension,
in non-caste societies; and then to proceed to the empirical
examination of this assumed correspondence, to determine the
extent to which it in fact obtains, the conditions under which
it does and does not obtain, and the variability in its actualisa-
tion and associated conditions at different levels of the system.
An empirical study of this type, focused on colour-scale corres-
pondences and divergences, will direct attention to the fifth and
final dimension of the colour concept, structural colour. The
structural dimension consists in those factors and aspects of
social process, and the relations between them, which give the
society its distinctive form as an arrangement of units and pro-
cesses. Thus the structural dimension is an abstract analytic
category reflecting the distributions and types of power, authority,
knowledge, and wealth, which together define and constitute the
social framework. When we speak of structural colour, we
imply an allocation of these variables among colour-differentiated
groups which obtains presently and reflects historical conditions.
Thus structural colour connotes the empirical distribution of
these variables among the colour-differentiated population. If
we like, we can distinguish between contemporary and historical
scales of structural colour, and for certain purposes, such as the
analysis of a rapidly changing situation, a distinction of this type
may be essential. To the extent to which it is unnecessary, the
society under study conforms to conditions of stationary equili-
brium, under which the expected distributions of power, wealth,
authority, and knowledge among the population are empirically
confirmed. In such a scale, the structural colour of an individual
expresses his equation with the expected or empirical distribution
of these controls among the population classified in a colour scale.
Thus a black man possessed of wealth, and other structural criteria
normally associated with the white population in such a scale is
structurally white. Similarly the 'Poor White' as the term im-
plies, is not structurally white, and for this reason, although he
may adhere to white institutions completely, ceases to be asso-
ciationally White, and forms a separate community of his own,
where conditions permit. The analysis of discrepancies and con-
sistencies between the empirical and the assumed correspondence
of the first four colour-scales will thus expose the fifth dimension
of this complex, and at the same time lay bare the social dynamics
to concrete analysis. For the reasons given above, it will be
apparent that any society which is characterized by role ascrip-
tion on a colour or cultural basis is also characterized by a scale
of structural colour.
In the analysis of such social systems, we are therefore
presented with the problem of determining by examination of the
correspondences or lack of correspondences of the various colour
scales, the bases and conditions of superordination and subordina-
tion, and the relative significance of these forms and factors at
different points of the system, by an inductive empirical study of
deviant and typical cases together. This means simply that the
definition of such social structures as a particular set of relations
holding between certain critical variables, such as power, wealth,
authority, and knowledge, and reflecting their distribution among
the population, can only be approached after the position of a
sufficient number of individuals within a composite colour-scale
representing the four dimensions of phenotype, genealogy, asso-
ciation, and institutional adherence, has been determined; and
when the correspondences, or lack of correspondences, between
these several distributions have been inductively analysed in terms
of the structural variables.
Analytic Models for Caribbean Societies.
The type of theoretical model which emerges from this
examination of colour in the Caribbean has several interesting
features. It is eclectic in the sense that it seeks to analyse and
relate the parts played by such factors as race, culture, associa-
tion, and power in the social system without predicating any
initial priority or order among them. It is eclectic also in the
sense that it provides a common framework for the integration
and co-ordination of the various lines of research which we have
been reviewing, and attempts their synthesis. It has heuristic
values as a hypothesis about such societies; but the model is
merely a guide to field studies, and its revision, elaboration, or
abandonment will surely follow from the empirical researches
which it may influence or stimulate. In other words, it is a goad
to investigation, not a narcotic; and it defines the methods of field
research as quantitatively and qualitatively inductive, their focus
in terms of social and cultural differentiation, and their results
and analyses as both the purpose and test of theory, and the
ground for new, superior hypotheses. A brief consideration of
such topics as social structure, social mobility, or acculturation,
will serve to define the model more precisely, and illustrate its
general comparative character.
Multi-racial societies vary in the social significance which
they ascribe to racial difference. In those societies where racial
factors play a prominent part in status placement and role alloca-
tion, contemporary or historical conditions of cultural differentia-
tion obtain. Thus such societies are characterized by a tradi-
tion of cultural plurality. To the extent that this condition per-
sists as a contemporary fact, then such societies approximate the
model of plural societies. Thus the first problem of field research
is the determination of the levels and areas of institutional cul-
tural differences or continuities within the population. Plurality
is an aspect of cultural differentiation; not a finite thing, but
a dimension, in terms of which some societies are more or less
plural or homogeneous in their culture, and may be so variably in
different fields. It is therefore essential in the comparative study
of institutions and other general cultural forms, to ascertain their
uniformity or difference, formally and functionally. Since the
major institutions, such as family, mating, kinship, religion, law,
education, government, property, and the like, constitute the basic
complexes and units of culture and society alike, the study of
cultural and social differentiation must focus on the variability
of their definition and the distribution of their different forms
throughout the population. The analysis of these results will
show the degree to which and the fields in which the society under
study approximates the extreme condition of cultural pluralism.
However, a word must be said about the structural limit
of extreme differentiation. This structural limit is itself inherent
in the nature and concept of society as an order or system of
relations, that is, a type of unit. The unity of a population
characterized by extreme differences of institutional culture is only
possible where one of these culturally differentiated sections con-
trols the destinies of the total unit. The alternative to this is
disunity, that is, the existence of several societies as separate units
having external relations with one another. Thus, within units
characterized by cultural heterogeneity, governmental institutions,
such as law, administrative, political and military systems, are by
definition under the ultimate or exclusive control of one or other
of the culturally differentiated groups. And in this way the
maintenance of social unity is interdependent with hierarchic sec-
tional relations of dominance and subordination. It follows,
therefore, that within plural societies plural political institutions
of a formal character equivalent to their other institutional diver-
gences cannot obtain. Such political pluralism really connotes
the existence of different societies, each possessing internal auto-
nomy in greater or less degree. Thus, the Hausa states of
Northern Nigeria which have continued to be administered under
the British policy of Indirect Rule through their own chiefs and
political institutions are not part of an Anglo-Hausa plural society,
but form separate societies of their own. Where cultural plurality
obtains within a single polity however, it follows that the formal
political institutions and organisation of subordinate cultural sec-
tions of that unit have been repressed as a condition of the poli-
tical unity of the total society under the control of the dominant
We can describe that part of the population of a culturally
heterogeneous society which practises a distinctive and uniform
system of institutions, that is to say, a separate culture, as a
cultural section. The boundaries of such a section, and the
definition of its culture, are both matters for determination by
field studies. Theory guides such investigations by directing at-
tention to the problem of institutional variations, and their distri-
bution within a population. It also provides a frame for the
analysis of systems characterized by institutional variability of
different types and orders; but does not predicate these orders or
types. It is important to grasp this point, as it contains the
difference between an ideology, which is often unverifiable by
nature, and may well hinder rather than help research, and a
frame of hypotheses which defines field problems, systematizes
their study, and is itself revised by their results.
Now, as we saw from our discussion of colour, associational
patterns may obtain between persons who practise different cul-
tural traditions within plural units, and may also serve to dis-
tinguish persons who practise an identical culture. It is inherent
in the nature of the hierarchic relations between culturally
differentiated sections, which express and maintain the political
order of a plural society, that the status divisions between these
cultural sections should be of an extreme kind. Consequently
the intimate and habitual association of members of different
cultural sections on terms of equality will normally be marginal
within static pluralities, absent in homogeneous societies, and of a
changing character and range under dynamic conditions of struc-
tural change in plural societies. We can therefore conceive
of plural societies in terms of a series of associational continue,
between which marked status gaps obtain, and within each of
which, though not between them, to the extent that these continue
correspond with cultural divisions, class stratification may obtain.
We can describe these associational continue as social or
status sections, and the marginal areas resulting from their bi-
section of cultural boundaries can be classified as active or in-
active margins, in terms of their relation to the processes of struc-
tural change of such societies. Active margins will consist of
groups associating primarily with adherents of different cultural
traditions. Inactive margins are formed by groups associating
primarily with members of their own cultural section, but also
with members of active margins. The diagram which follows
illustrates one possible set of relations between these structural
categories in a plural society of three cultural and social sections.
Despite its schematic character and oversimplification, such a
Cultural Sections, A, B, C.
Social Sections, 1, 2, 3.
Status divisions, .
model has the virtue of directing attention to certain characters of
decisive importance in the comparison and analysis of plural
systems. These characters include the nature of the status gap,
the nature of the cultural diversity, the number of social and
cultural sections, the correspondence between them, the margins
of their overlap, and the stability, change, and conditions asso-
ciated with these margins.
In terms of such a model we can compare societies accord-
ing to the number and arrangement of their cultural and social
sections; and in this comparison the variations in the corres-
pondence and relations of these two dimensions of internal differ-
entiation are clearly of critical importance. On this basis we
can distinguish between those pluralities within which associa-
tional divisions are defined without respect to cultural differentia-
tion on the one hand, and those in which they are defined pri-
marily in cultural terms. In the first of these two types of
plurality, associational sections will normally have rigid bound-
aries and sharp definition, in the second type of society less so.
Where associational discontinuities are uniformly instituted, ir-
respective of cultural similarity or difference, they will normally
be defined in terms of some single and rather obvious variable
such as language or race. In certain American communities this
is the type of condition represented by colour-caste. Haiti and
Guatemala which give the appearance of conforming to this
American model are substantially different, in that the association-
al divisions are primarily cultural in base. The distinction be-
tween American colour-caste and Haitian or Guatemalan cul-
tural and social stratification consists in the vertical division of
American society on the one hand, and the horiz~ntI aT visions
of Guatemalan and Haitian societies on the-oter-i-the
American system rigidity with regard to racial definition is con-
sistent with cultural continuities across this line, as well as dis-
continuities within each racial segment. In Haiti and Guatemala,
the caste-like structure, for all its posited racial and linguistic
bases, really consists in the cultural cleavage of which it is the
social expression, and could not obtain without this condition.
We find therefore that there are three principal alternative forms
of social differentiation within societies characterized by cultural
heterogeneity; the American pattern, under which social differ-
entiation is defined rigidly without respect to cultural variables on
the one hand, and relativistically within its racial segments in
cultural terms on the other; the Guatemalan or Haitian pattern
in which social and cultural differentiation correspond complete-
ly; and the type of pattern represented by our diagram, in which
cultural and social differentiation vary to some extent independ-
ently of each other, although within limits set by other aspects of
their inter-relation within a single structural unit. Where cul-
tural differentiation provides the general and historical basis for
associational differentiation, but overlaps and margins occur,
then a system of multiple criteria provides the basis of status
ascription, and the relativism of individual ranking, which con-
trasts with the rigidities of American or Guatemalan 'caste', re-
quires for its examination the type of concept and method out-
lined above in our discussion of Caribbean colour.
We have therefore to deal with culturally homogeneous or
plural societies on the one hand, and with relativistic or rigid
systems of social differentiation on the other. It will at once
be apparent that both forms of social differentiation may co-
exist in either type of society. Thus, a homogeneous society
may contain two rigidly differentiated categories of persons, within
each of which ranking by multiple criteria of similar or different
kinds reflects relativism. This would appear to be the American
position, looked at from a national point of view. A plural
society may have similar structure. In the British Caribbean,
for instance, it is possible that Barbados, Antigua and St. Kitts,
which share a common ethnic and cultural composition, are
examples of this structural type. Trinidad and British Guiana,
which have a more complex ethnic and cultural composition,
also belong in this category. In plural societies of this type, the
racial variable, as culturally defined, provides a rigid uniform
basis for the social differentiation of particular ethnic groups,
within each of which relativistic ranking may obtain on similar
or different lines, reflecting their internal cultural diversity.
In effect, therefore, we emerge with six possible structural
types, looked at from our present point of view: plural or homo-
geneous societies having either rigid or relativistic ranking systems
on the one hand, or those containing both types on the other.
Clearly the co-existence of these two modes of differentiation
within a society is only possible where a rigid division of the
spheres and conditions in which each has significance for the
other obtains. A similar point applies to the distinction between
homogeneous and plural societies, which can be illustrated by
consideration of the ambiguous position of Brazil and the U. S. A.
There is a good deal of evidence of pluralistic conditions in
the U. S. A. for instance, though these conditions occupy a
minority status in terms of national American culture. This
means that the significance of localised cultural plurality is mar-
ginal for the classification of American national culture or society
in terms of homogeneity or plurality. Despite the sharp differ-
ences of Brazilian and American race relations, these two
societies occupy a similar position in that both contain within
their national framework localised pluralities having a minority
status. This similarity between Brazil and America reduces to
three related conditions. In both societies the dominant culture
which also enjoys the highest prestige is practised by the over-
whelming majority of the population. Divergent cultures there-
fore have true minority status at the national plane, although
their local significance is often great. It is very probable that the
concept of a homogeneous society containing culturally pluralistic
elements is only applicable under the combination of limiting
conditions which characterises Brazil and the U. S. A. It is
notable that different patterns of race relations may obtain in
such conditions, without in any way distinguishing between them
at the present level of discussion.
Granted the presence of two or more cultural traditions in
Caribbean and Middle American societies generally, that is to say,
their pluralistic character, then after the determination of these
traditions, the first task in their comparison consists in the study
of their various systems of differentiation and stratification, that
is, the comparative analysis of their intersectional and intrasec-
tional frameworks. Such study consists largely in an examina-
tion of the interrelation of cultural and social patterns within
and among these various populations; and it focuses on the
definition of relations holding between each of these systems
separately, and both together, especially as these relations and
their variability are essential to the understanding of particular
The appropriateness of such an approach is apparent from
a consideration of social mobility in the Caribbean. In a plural
society, social mobility has two forms, individual mobility, and
sectional mobility. Sectional mobility is initially expressed
through changes in the sizes and directions of intersectional mar-
gins, and is indicative of general structural change, such for
instance as may now be underway at different rates in the various
British Caribbean colonies. Individual mobility in a plural
society may be upward, downward, or lateral. Vertical individual
mobility is of course found in many homogeneous societies, but
both sectional and lateral individual mobility cannot by definition
develop within them. Lateral mobility occurs when a person,
usually already marginal in some degree, alters his behavioral
pattern to conform with a culture different from that which he
formerly practised without effecting any corresponding alteration
in his social position. In terms of the total society, the marriage
of 'peasants' after years of 'faithful concubinage' is an instance of
lateral mobility, even when associated with changes in other in-
stitutional patterns, such as religion, land-tenure, and the like.
It is this difference between cultural movement and social mobility
which to a large degree accounts for the failure of acculturation
programmes instituted by the dominant section of a plural society
to win much acceptance from subordinate sections.
Individual upward mobility presents some interesting prob-
lems in the study of British Caribbean societies. It throws a good
deal of light on the relative significance of components in the
structural colour scale, and also on the variability of their rela-
tions at different structural levels. Let us briefly explore a set
of hypothetical cases in which differently placed individuals all
return to their homeland with equal increases of wealth. We
have then a situation in which one factor, wealth, is constant,
while others vary. Assuming a relativistic ranking system, our
problem consists in the relative mobility of individuals drawn
from different sections of the totality; for example, three brown
men, one of whom is culturally white, another culturally inter-
mediate or brown, and the third culturally black; or three white
men, distinguished similarly; or three black men; or one white,
one brown, and one black man, all belonging to the same cul-
tural section. And so on. It is clear that the rate, range, and
type of upward mobility will vary in respect of these differences.
Similarly, if we considered the probabilities of downward mobility
for such persons, consequent on an equal loss of wealth, similar
differences of type, rate, and range would be noticeable; and the
systematic character of the relations between colour, cultural ad-
herence, wealth, and individual mobility in such societies could
be determined and subjected to predictive tests.
Acculturation as an aspect of process in the life of Middle
American and Caribbean societies has received extensive atten-
tion. It has provided a field for such studies as those of Hersko-
vits and Redfield. Of itself, this fact indicates the relevance of
our pluralistic concepts for the analysis of such societies. As a
process occurring within populations sharing common political
institutions, acculturation implies a high degree of their internal
cultural differentiation, that is, their pluralistic character. Thus
such acculturation processes define the societies in which they
occur as pluralities of greater or less degree. Even where such
societies show marginal plurality on the national plane, as for
instance, is true of Brazil and America, the condition of accul-
turation indicates pluralism of significance at the local or com-
munity level. Thus, despite their differing orientations and prob-
lems, both Afro-American and Folk-Urban studies reflect and
imply as their field of investigation, a type of social system con-
taining culturally differentiated sections, that is to say, a condi-
tion of pluralism.
Other writers with less explicit theoretical interests than Red-
field and Herskovits provide impressive evidence of the nature
and extent of this cultural heterogeneity in the Caribbean
and Middle America, and sometimes distinguish clearly
between folk and elite within such societies (74). Such co-
incidence can hardly be accidental, and its theoretical implications
(74) See Table of References :or works by, Beckwith, Blanshard, Campbell, Carr,
Clarke, Deren, Gillin, Hadley, Kerr, Leyburn, Lobb, MacMillan, Pearse, May-
hew, Matthews, Pierson, Puckett, Simpson, Willems, Wagley.
cannot be ignored, considering the multiplicity of approaches
which such studies represent: administrative, political, anthro-
pological, folk-lorist, social psychology, and race relations. Their
convergence provides most impressive evidence of the condition
and nature of society in this region, and of the appropriateness
of the plural framework to the analysis of such units. Such con-
sensus is especially impressive in that it contains or reflects no
hint of an explicit conception of the societies and cultures con-
cerned in terms of plurality, as we have defined it. Yet we have
shown that the deficiencies of the various approaches to the study
of these societies all reflect an inadequate treatment of the inter-
related planes of social and cultural differentiation; and that some
conceptualisation of these relations in terms of plural systems
alone provides a basis for the systematic and detailed empirical
study of such conditions. In terms of such a programme of
studies, economist, historian, political scientist, anthropologist,
social psychologist, folk-lorist, and sociologist, can all contribute
to one another's understanding of the common field equally and
continuously. It is also worth mentioning here that concepts of
pluralism such as we have discussed above have already been
applied by Van Lier to Surinam, and by Beals to certain Middle
American societies (75).
Our analysis of the literature has been necessary and re-
warding. It has revealed the principal current approaches to the
social and cultural study of our area. These approaches consist
in Afro-American studies, the Folk-Urban theory, studies of strati-
fication, and psychological research which initially relies on con-
cepts drawn from cultural and social analysis. We have seen
how a competition of models tends to obtain, acculturation studies
presenting one framework, while stratification theory offers an-
other. We have seen that each of these separate models is in-
adequate for the systematic and comprehensive study of these
societies, and that both try to disguise their inadequacies behind
a screen of vague unverifiable assumptions and indeterminate con-
cepts. It has also been shown that when combined in terms of a
theory of plural societies and cultures, these competing approaches
provide a unified and refined frame of concepts which is of equal
use in defining the problems under study as a system of area re-
search, in guiding the investigations of psychologists, and in pro-
viding an integrated comparative framework for the study of this
(75) Van Lier. 1950. Beals, R., 1953b.
C. A. 5
Arensberg, C. M., 1954. The community-study method. American
Journal of Sociology, vol. LX, no. 2, pp. 109-124. Chicago.
Bascom, W. R., 1941. Acculturation among the Gullah Negroes.
American Anthropologist, vol. 43, pp. 43-50, Manasha, Wis.
Bascom, W. R., 1952a. The Esusu: a credit institution among the
Yoruba. Journal of the Royal Anthropological Institute, vol.
LXXII, pp. 63-70, London.
Bascom, W. R.. 1952b. Two forms of Afro-Cuban divination. Accul-
turation in the Americas. Edited by Sol Tax, pp. 169-179, Chicago.
Bascom, W. R., 1955. Urbanization among the Yoruba. American
Journal of Sociology, vol. LX, no. 5, pp. 446-454. Chicago.
Beals, R., 1953a. Acculturation. In Anthropology Today, Edited by
A. L. Kroeber, pp. 621-641, Chicago.
Beals, R., 1953b. Social Stratification in Latin America. American
Journal of Sociology, vol. LVIII, no. 4, pp. 327-339. Chicago.
Beckwith, Martha W., 1924. Jamaica Anansi Stories. Memoir of the
American Folk-lore Society, vol. XVII, New York.
Beckwith, Martha W., 1929. Black Roadways: A Study of Jamaica
Folk Life. University of North Carolina Press, Chapel Hill.
Bendix, R., & Lipset, S. M. (Eds.). 1953. Class, Status, and Power: a
Reader in Social Stratification. Free Press, Glencoe, Illinpis.
Blanshard, P.. 1947. Democracy and Empire in the Caribbean. New
Braithwaite, L., 1953. Social Stratification in Trinidad. Social & Eco-
nomic Studies, vol. 2, nos. 2 & 3, pp. 5-175. Jamaica, B.W.I.
Braithwaite, L., 1954. The problem of Cultural Integration in Trinidad.
Social & Economic Studies, vol. 3, no. 1, pp. 82-96, Jamaica, B.W.I.
Broom, L., 1953. Urban research in the Caribbean. Social & Economic
Studies, vol. 1, no. 1, pp. 113-119, Jamaica, B.W.I.
Broom, L., 1954. The Social Differentiation of Jamaica. American
Sociological Review, vol. XIX, pp. 115-123, Washington.
Campbell, A. A., 1943. St. Thomas Negroes: A Study of Personality
and Culture. Psychological Monographs, vol. LV, no. 5, Illinois.
Caribbean Quarterly, 1953. Editorial Note, vol. 3, no. 1, p. 3, Trinidad,
Carr, Andrew T.. 1953. A Rada Community in Trinidad. Caribbean
Quarterly, vol. 3, no. 1, pp. 35-54, Trinidad.
Centers, R., 1949. The Psychology of Social Classes. Princeton.
Clarke, Edith, 1953. Land Tenure and the Family in Four Communities
in Jamaica. Social & Economic Studies, vol. 1, no. 4, pp. 81-117,
Cohen, Y., 1953. A Study of Interpersonal Relations in a Jamaican Com-
munity. Ph.D., thesis, Yale, Dept. of Anthropology. Unpublished
Cohen, Y., 1954. The Social Organisation of a selected Community in
Jamaica. Social & Economic Studies, vol. 2, no. 4, pp. 104-137,
Cumper, G. E., 1954. A Modern Jamaican Sugar Estate. Social &
Economic Studies, vol. 3, no. 2, pp. 119-160, Jamaica.
Davis, A., 1945. Caste, Economy, and Violence. American Journal of
Sociology, vol. LI, no. 1, pp. 7-15.
Davis, A., Gardner, B. B., & Gardner, M. R., 1941. Deep South,
Deren, Maya, 1953. Divine Horsemen: The Living Gods of Haiti.
Dollard, J., 1937. Caste and Class in a Southern Town. New York.
Dollard, J., et al, 1939. Frustration & Aggression. New York,
Eduardo, O. D., 1948. The Negro in Northern Brazil: A Study in
Acculturation. Memoirs of the American Ethnological Society, vol.
Forde, Daryll, 1954. Introduction, in African Worlds, Edited by Daryll
Fortes, Meyer, 1949. The Web of Kinship among the Tallensi. London.
Fortes, M., & Evans-Pritchard, E. E. (Eds), 1940. African Political
Foster, G. M., 1953. What is Folk Culture? American Anthropologist,
vol. LV, pp. 159-173, Menasha, Wis.
Frazier, E. Franklin, 1939. The Negro Family in the United States.
Frazier, E. Franklin, 1953. Theoretical Structure of Sociology and
Sociological Research. British Journal of Sociology, vol. 4, no.
4, pp. 293-313.
Gillin, J., 1948. Mestizo America, in Most of the World, Ed. by Ralph
Linton, Columbia, New York, pp. 156-211.
Gillin, J., 1951. The Culture of Security in San Carlos. Pub. no. 16,
Middle American Research Institute, New Orleans.
Gillin, J., 1952. Modern Cultural Development and Synthesis in Latin
America, in Acculturation in the Americas, ed. by Sol. Tax,
Goldschmidt, W., 1950. Social Class in America A Critical Review.
American Anthropologist, vol. 52, no. 4, pp. 468-482. Menasha.
Gordon, Maxine, 1950. Cultural Aspects of Porto Rico's Race Problem.
American Sociological Review, vol. XV, pp. 382-391.
Gross, Llewellyn, 1949. The Use of Class Concepts in Sociological Re-
search. American Journal of Sociology, vol. LIV, pp. 409-421.
Hadley, C. V. D., 1949. Personality Patterns and Aggression in the
British West Indies. Human Relations, vol. 11, no. 4, pp. 349-362.
Hall, D. G. H., 1954. Jamaican Economic Development, 1840 to 1865.
Ph.D., thesis, London University, London. Unpublished MS.
Henriques, F., 1951. Colour Values in Jamaican Society. British
Journal of Sociology, vol. 2, no. 2, pp. 115-121.
Henriques, F., 1952. West Indian Family Organisation. Caribbean
Quarterly, vol. 2, no. 1, pp.16-24. Trinidad.
Henriques, F., 1953a. Letter to the Editor, Caribbean Quarterly, vol. 2.
no. 3, p. 56.
Henriques, F., 1953b. Review of 'Personality and Conflict in Jamaica'
by M. Kerr, in Caribbean Quarterly, vol. 3, no. 1. pp. 61-2.
Henriques, F., 1953c. Family and Colour in Jamaica. London.
Herskovits, M. J., 1930. The Negro in the New World: the statement
of a new Problem. American Anthropologist, vol. 32, pp. 145-155.
Herskovits, M. J., 1937. Life in a Haitian Valley. New York.
Herskovits, M. J., 1938a. Acculturation: A Study of culture contact.
Herskovits, M. J., 1938b. Dahomey (2 vols.), New York.
Herskovits, M. J., 1941. The Myth of the Negro Past. New York.
Herskovits, M. J., 1943. Education and Cultural Dynamics. American
Journal of Sociology, vol. XLVIII, pp. 737-749.
Herskovits, M. J., 1946. Problem, Method, and Theory in Afroamerican
Studies. Phylon, vol. VII, no. 4, pp. 337-354. Atlanta.
Herskovits, M. J., 1948. Man and His Works. New York.
Herskovits, M. J., 1948b. The contribution of Afroamerican Studies to
Africanist Research. American Anthropologist, vol. 50, no. 1,
Herskovits, M. J., 1952a. Introduction, in Acculturation in the
Americas, Ed. by Sol Tax, Chicago.
Herskovits, M. J., 1952b. Some Psychological Implications of Afro-
American Studies, in Acculturation in the Americas, Ed. by Sol
Herskovits, M. J., 1952c. Letter to the Editor, Caribbean Quarterly, voL
2, no. 2, pp. 44-5, Trinidad.
Herskovits, M. J., & F. S., 1934. Rebel Destiny: Among the Bush
Negroes of Dutch Guiana. New York.
Herskovits, M. J., & F. S., 1937. Suriname Folklore. Columbia Con-
tributions to Anthropology, vol. XXVII, N.Y.
Herskovits, M. J., & F. S., 1947. Trinidad Village, New York.
Hill, M. C., & MacCall, H. R., 1950. Social Stratification in Georgia
Town. American Sociological Review, vol. XV, pp. 721-9.
Hughes, E. C., & H. M., 1952. Where Peoples Meet. Free Press,
Kerr, Madeline, 1952. Personality and Conflict in Jamaica. Liverpool.
Klineberg, O. (ED.), 1944. Characteristics of the. American Negro.
Kruijier, G. J., 1953. St. Martin and St. Eustatius Negroes as compared
with those of St. Thomas. West-Indische Gids, XXXIV, pp. 225-
Levy, Marion J, Jr., 1952. The Structure of Society. Princeton.
Leyburn, J. G., 1941. The Haitian People. Yale, New Haven.
Lobb, John, 1940. Caste and Class in Haiti. American Journal of
Sociology, vol. XLVI, pp. 23-34.
Lynn Smith, T., & Marchant, A. (EDs.), 1951. Brazil: Portrait of
Half a Continent. N.Y.
MacMillan, W., 1936. Warning from the West Indies. London.
Mair, L. P., 1953. African Marriage and Social Change, in Survey of
African Marriage and Family Life, Ed. by Arthur Phillips, pp.
Matthews, Dom. B., 1953. The Crisis in the West Indian Family,
Mayhew, Frank, 1953. My Life. Caribbean Quarterly, vol. 3, no. 1,
pp. 13-23, Trinidad.
Metraux, Rhoda, 1952. Some aspects of hierarchical structure in Haiti,
in Acculturation in the Americas, Ed. by Sol Tax, pp. 185-194,
Mintz, S. W., 1953a. The Folk-Urban continuum and the rural Prole-
tarian Community. American Journal of Sociology, vol. LIX, pp.
Mintz, S. W., 1953b. The Culture-History of a Puerto Rican Sugar Cane
Plantation, 1876-1949, Hispanic American Historical Review, pp.
Mintz, S. W., 1954. On Redfield and Foster. American Anthropologist,
vol. 56, pp. 87-92, Menasha, Wis.
Myrdal, Gunnar, 1944. An American Dilemma. New York.
Nadel, S. F., 1954. Nupe Religion. London.
Nelson, Lowry, 1950. Rural Cuba, Minneapolis.
Olivier, Lord S., 1936. Jamaica, the Blessed Island. London.
Park, R., 1950. Race and Culture, Free Press, Glencoe, Illinois.
Parsons, Talcott, 1953. A Revised Analytical Approach to the Theory
of Social Stratification, in Bendix & Lipset, Eds., Class, Status,
and Power, pp. 92-128, Free Press, Glencoe, Illinois.
Pearse, Andrew, 1952. West Indian Themes, Caribbean Quarterly, vol.
2, no. 2, pp. 12-23.
Pearse, Andrew, 1955. Aspects of Change in Caribbean Folk Music,
International Folk Music Journal, vol. VII, pp. 29-36, UNESCO.
Pfautz, H. W., 1953. The Current Literature on Social Stratification:
Critique and Bibliography. American Journal of Sociology. Vol.
LVIII, pp. 391-418. Chicago.
Pierson, D., 1942. Negroes in Brazil. Chicago.
Powdermaker, H., 1939. After Freedom: A cultural study of the Deep
South. New York.
Powdermaker, H., 1943. The Channelling of Negro Acculturation by
the Cultural Process, American Journal of Sociology, Vol.
Proudfoot, Mary, 1954. Britain and the U.S.A. in the Caribbean.
Puckett, N. N., 1926. Folk Beliefs of the Southern Negro. Chapel Hill.
Ramos, A., 1939. The Negro in Brazil. Washington.
Redfield, R., 1940. The Folk Society and Culture. American Journal
of Sociology, vol. XLV, pp. 731-742.
Redfield, R., 1941. The Folk Culture of Yucatan. Chicago.
Redfield, R., 1947. The Folk Society. American Journal of Sociology,
Redfield, R., 1953. The Natural History of the Folk Society. Social
Forces, vol. XXXI, pp. 224-8.
Redfield, R., Linton, R., & Herskovits, M. J., 1936. Memorandum for
the Study of Acculturation. American Anthropologist, vol. 38,
Reuter, E. B., 1946. Culture Contacts in Puerto Rico. American
Journal of Sociology, vol. LII, pp. 91-101. Chicago.
Simey, T. S., 1946. Welfare and Planning in the West Indies. Oxford.
Simpson, G. E., 1940a. The Vodun Service in Northern Haiti. American
Anthropologist, vol. 42, no. 2, pp. 236-254.
Simpson, G. E., 1940b. Haitian Magic. Social Forces, vol. 19, no. 1,
Simpson, G. E., 1941. Haiti's Social Structure. American Sociological
Review, vol. VI, no. 5, pp. 640-9.
Simpson, G. E., 1942. Sexual and Familial Institutions in Northern
Haiti. American Anthropologist, vol. 44, no. 4, pp. 655-674.
Simpson, G. E., 1946. Four Vodun Ceremonies. Journal of American
Folklore, 59, pp. 154-167.
Simpson, G. E., 1951. Acculturation in Northern Haiti. Journal of
American Folklore, vol. 64, pp. 397-403.
Simpson, G. E., 1952. Discussion of Dr. Price-Mars' Paper, in Accul-
turation in the Americas, Ed. by Sol Tax, Chicago.
Smith, Mary F., 1954. Baba of Karo, a Woman of the Moslem Hausa.
Smith, M. G., 1954. Slavery and Emancipation in Two Societies.
Social & Economic Studies, vol. 3, nos. 3-4, pp. 240-290, Jamaica.
Smith, M. G., 1955. The Economy of Hausa Communities in Zaria
Province, H.M.S.O., London.
Smith, R. T., 1953. Family Organisation in British Guiana. Social &
Economic Studies, vol. 1, no. 1, pp. 87-112, Jamaica.
Steward, J. H., 1950. Area Research: Theory and Process. Social
Science Research Council, Bulletin, 63, New York.
Steward, J. H., & Manners, R., 1953. The Cultural Study of Contem-
porary Societies: Puerto Rico. American Journal of Sociology,
vol. LIX, 123-130.
Tax, Sol, 1939. Culture and Civilization in Guatemalan Societies.
Scientific Monthly, vol. 48, pp. 463-7.
Tax, Sol (Ed.), 1952. Acculturation in the Americas. Chicago.
Taylor, D. M., 1951. The Black Caribs of British Honduras. Viking
Fund Publications in Anthropology, no. 17, New York.
Tumin, M. M., 1952. Caste in a Peasant Society. Princeton.
Van Lier, 1950. The Development and Nature of Society in the West
Indies. Royal Institute for the Indies, Amsterdam.
Wagley, C., 1941. Economics of a Guatemalan Village. Memoir of
the American Anthropological Association, no. 58, Menasha, Wis.
Wagley, C., 1949. Social and Religious Life of a Guatemalan Village.
Memoir of the American Anthropological Ass'n, 71, Menasha,
Wagley, C., 1952. The Folk-culture of the Brazilian Amazon, in Accul-
turation in the Americas, ed. by Sol Tax, Chicago.
Wagley, C., 1953. Race and Class in Rural Brazil, Unesco, New York.
Warner, W. L., Junker, B. H., & Adams, W. A., 1941. Colour &
Human Nature. Washington.
Warner, W. L., Meeker, M., Eells, K., 1949. Social Class in America.
Willems, Emilio, 1949. Race Attitudes in Brazil. American Journal of
Sociology, vol. LIV, pp. 402-8. Chicago.
Willems, Emilio, 1952. Caboclo Cultures of Southern Brazil, in Accul-
turation in the Americas, ed. by Sol Tax, Chicago, pp. 231-243.
Printed by The Gleaner Co. Ltd., Harbour Street, Kingston, Ja.
L. fl"~4c ,. II I II
4 ; ir'r i ,
F". 21 $4%
*- i; ~ ~~,t:~
i ", ; -.1
ty THE U I- p,,ry OF