GERALD D.

BERREMAN

Race, Caste, and Other

Invidious Distinctions in
Social Stratification
A society is socially stratified when its members are divided into categories
which are differentially powerful, esteemed, and rewarded. Such systems
of collective social ranking vary widely in the ideologies which support
them, in the distinctiveness, number, and size of the ranked categories,
in the criteria by which inclusion in the categories is conferred and changed,
in the symbols by which such inclusion is displayed and recognized, in
the degree to which there is consensus upon or even awareness of the
ranking system, its rationale, and the particular ranks assigned, in the
rigidity of rank, in the disparity in rewards of rank, and in the mechanisms
employed to maintain or change the system.
For purposes of study, such systems have been analysed variously
depending upon the interests and motives of the analyst. One of the most
frequently used bases for categorizing and comparing them has been
whether people are accorded their statuses and privileges as a result of
characteristics which are regarded as individually acquired, or as a result
of characteristics which are regarded as innate and therefore shared by
those of common birth. This dichotomy is often further simplified by
application of the terms ’achieved’ versus ’ascribed’ status. Actually,
what is meant is non-birth-ascribed status versus birth-ascribed status.
The former is usually described as class stratification, referring to shared
statuses identified by such features as income, education, and occupation,
while the latter is frequently termed caste or racial stratification or, more
recently, ethnic stratification, referring to statuses defined by shared
ancestry attributes of birth.
or

Regardless of its characteristics in a particular society, stratification has

been described as being based upon three primary dimensions: class,
status, and power, which are expressed respectively as wealth, prestige,
and the ability to control the lives of people (oneself and others).’ These
dimensions can be brought readily to mind by thinking of the relative
advantages and disadvantages which accrue in Western class systems to
persons who occupy such occupational statuses as judge, garbage man,
stenographer, airline pilot, factory worker, priest, farmer, agricultural
GERALD D. BERREMAN is Professor of Anthropology at the University of California,
Berkeley. This article was written while the author was a Fellow of the John Simon
Guggenheim Memorial Foundation.
386

labourer, physician, nurse, big businessman, beggar, etc. The distinction

between class and birth-ascribed stratification can be made clear if one
imagines that he encounters two Americans, for example, in each of the
above-mentioned occupations, one of whom is white and one of whom is
black. This quite literally changes the complexion of the matter. A
similar contrast could be drawn if, in India, one were Brahmin and one
untouchable; if in Japan one were Burakumin and one were not; if in
Europe one were Jew and one were Gentile; or if, in almost any society,
one were a man and one a woman. Obviously something significant has
been added to the picture of stratification in these examples which is
entirely missing in the first instance-something over which the individual
generally has no control, which is determined at birth, which cannot be
changed, which is shared by all those of like birth, which is crucial to social
identity, and which vitally affects one’s opportunities, rewards, and social
roles. The new element is race (colour), caste, ethnicity (religion, language,
national origin), or sex. The differences in opportunities and behaviour
accorded people as a result of these criteria are described by such pejorative
terms as racism, casteism, communalism (including especially ethnic and
religious discrimination), and sexism. To be sure, the distinctions are
manifest in class, status, and power, but they are of a different order than
those considered in the first examples: they are distinctions independent
of occupation, income, or other individually acquired characteristics.
While the list includes a variety of criteria for birth-ascription and rank
with somewhat different implications for those to whom they are applied,
they share the crucial facts that: 1. the identity is regarded as being a
consequence of birth or ancestry and hence immutable; 2. the identity
confers upon its possessor a degree of societally defined and affirmed
worth which is regarded as intrinsic to the individual; 3. this inherent
worth is evaluated relative to that of all others in the society-those of
different birth are inherently unequal and are accordingly adjudged
superior or inferior, while those regarded as being of similar birth are
innately equal. The crucial fact about birth-ascription for the individual
and for society lies not so much in the source of status (birth), as in the
fact that it cannot be repudiated, relinquished, or altered. Everyone is
sentenced for life to a social cell shared by others of like birth, separated
from and ranked relative to all other social cells. Despite cultural differ-
ences, therefore, birth-ascribed stratification has common characteristics
of structure, function, and meaning, and has common consequences in the
lives of those who experience it and in the social histories of the societies
which harbour it.
The specific question motivating the present discussion is this: is social
ranking by race absolutely distinctive, not significantly distinctive at all, or
is race one criterion among others upon which significantly similar systems
of social ranking may be based ? While identifying the last of these as
’correct’ from my perspective, I shall insist that the answer depends entirely
upon what one means by ’race’, and by ’distinctive’, and what one wishes
to accomplish by the inquiry. No satisfactory answer can be expected
without comparative, cross-cultural analysis encompassing a number of
systems of social differentiation, social separation, and social ranking,
based on a variety of criteria, embedded in a variety of cultural milieux,
analysed by reference to various models of social organization, and tested
against accounts of actual social experience. The attempt to do this leads
 387

to a number of issues central and tangential to the study of stratification

and race, some of which have been overlooked or given short shrift
in the scholarly literature, while others are well-discussed in particular
disciplinary, regional, or historical specialities without necessarily being
familiar to students of other academic domains to whose work and thought
they are nevertheless relevant.
There is not space here to present ethnographic and historical documen-
tation for particular instances of birth-ascribed stratification. I have done
so briefly in another paper, citing five societies on which there is fortunately
excellent published material vividly exemplifying the kinds of social sys-
tems I refer to in this paper, and their implications for those who comprise
them: Ruanda, India, Swat, Japan, and the United States. I recommend
those accounts to the reader.22

MODELS FOR ANALYSIS

In the course of scholarly debate concerning the nature and comparability

of systems of collective social ranking, a number of models and concepts
have been suggested, implied, or utilized. A framework can be provided
for the present discussion by identifying some of these and analysing
whether and to what extent each is relevant and applicable to all or some
systems of birth-ascribed social separation and inequality, with special
attention to the five societies cited above.
Stratification
By definition, stratification is a common feature of systems of shared
social inequality-of ranked social categories-whether birth-ascribed or
not. Where membership in those categories is birth-ascribed, the ranking
is based on traditional definitions of innate social equivalence and differ-
ence linked to a concept of differential intrinsic worth, rationalized by a

myth of the origin, effect, and legitimacy of the system, perpetuated by

differential power wielded by the high and the low, expressed in differential
behaviour required and differential rewards accorded them, and experienced
by them as differential access to goods, services, livelihood, respect, self-
determination, peace of mind, pleasure, and other valued things including
nourishment, shelter, health, independence, justice, security, and long life.
Louis Dumont, in Homo Hierarchicus, maintains that the entire socio-
logical notion of stratification is misleading when applied to South Asia,
for it is of European origin, alien and inapplicable to India. He holds
that the term implies an equalitarian ideology wherein hierarchy is resented
or denied, and that it therefore obscures the true nature of India’s hier-
archical society, based as it is on religious and ideological premises
peculiar to Hinduism which justify it and result in its endorsement by all
segments of Indian society. Stratification, he maintains, is thus a ’socio-
centric’ concept which cannot cope with the unique phenomenon of
Indian caste.3 My response to this is twofold: first, the caste hierarchy
based on the purity-pollution opposition as Dumont insists, is well within
any reasonable definition of stratification, for the latter refers to social
structure and social relations rather than to their ideological bases; and
second, Dumont’s description of the functioning of, and ideological basis
for, the caste hierarchy is idealized and similar to the one commonly
purveyed by high caste beneficiaries of the system. Few low caste people
would recognize it or endorse it. Yet their beliefs and understandings are
388
as relevant as those of their social superiors to an understanding of the
system. The low caste people with whom I have worked would find
Dumont’s characterization of ’stratification’ closer to their experience
than his characterization of ‘hierarchy’.44
Use of the stratification model focuses attention upon the ranking of two
or more categories of people within a society, and upon the criteria and

consequences of that ranking. Often, but not inevitably, those who use
this concept place primary emphasis upon shared values and consensus,
rather than power and conflict, as the bases for social ranking and its
persistence. This emphasis is misleading, at best, when applied to systems
of birth-ascribed ranking, as I shall show. It is obvious, however, that
while many systems of stratification are not birth-ascribed, all systems of
birth-ascribed ranking are systems of social stratification, and any theory
of social stratification must encompass them.

Ethnic Stratification
Probably the most recent, neutral, and non-specific term for ascriptive
ranking is ’ethnic stratification’. ’An ethnic group consists of people who
conceive of themselves as being alike by virtue of common ancestry,
real or fictitious, and are so regarded by others’,5 or it comprises ’a distinct
category of the population in a larger society whose culture is usually
different from its own [and whose] members ... are, or feel themselves,
or are thought to be, bound together by common ties of race or nationality
or culture.’6 Undoubtedly the systems under discussion fit these criteria.
Use of the adjective ’ethnic’ to modify ’stratification’ places emphasis upon
’

the mode of recruitment, encompassing a wide variety of bases for ascrip-

tion, all of which are determined at birth and derive from putative common
genetic makeup, common ancestry, or common early socialization and are
therefore regarded as immutable. This commonality is held responsible
for such characteristics as shared appearance, intelligence, personality,
morality, capability, purity, honour, custom, speech, religion, and so
forth. Usually it is held responsible for several of these. The ranked evalua-
tion of these characteristics, together with the belief that they occur differ-
entially from group to group and more or less uniformly within each group
serves as the basis for ranking ethnic groups relative to one another.
Van den Berghe has held that ’ethnic’ should be distinguished from ’race’
or ’caste’ in that the former implies real, important, and often valued social
and cultural differences (language, values, social organization), while
the latter are artificial and invidious distinctions reflecting irrelevant
(and sometimes non-existent) differences in physiognomy, or artificial
differences in social role.’ This is a useful point. In the recent sociological
literature, however, ’ethnic’ has increasingly been used to refer to all
social distinctions based on birth or ancestry, be they associated with race,
language, or anything else. This is the usage adopted here. Moreover, as
I shall elaborate in discussing pluralism below, race and caste entail the
kinds of cultural distinctions cited by van den Berghe as diagnostic of
ethnic diversity, for the social separation implied by those systems ensures
social and cultural diversity. For example, van den Berghe’s assertion that
’notwithstanding all the African mystique, Afro-Americans are in fact
culturally Anglo-American,&dquo;’ has been countered by ample evidence that
the African origin, social separation, and collective oppression of blacks
in America has resulted in an identifiable Afro-American culture.99
 389
All systems of ethnic stratification are thus based on ancestry, approxi-
mating a theory of birth-ascription, and if the definitions set forth by
advocates of this term are accepted, most systems of birth-ascribed stratifi-
cation can properly be designated ethnic stratification. Perhaps the only
recurrent exception is sexual stratification, wherein inherent, birth-ascribed,
and biologically determined characteristics which are independent of
ancestry are the basis for institutionalized inequality. This instance,
exceptional in several respects, will be discussed separately below, and hence
will not be alluded to repeatedly in intervening discussions although most
of what is said applies to it also.

Caste
A widely applied and frequently contested model for systems of birth-
ascribed rank is that of ’caste’, deriving from the example of Hindu India
where the jati (almost literally ’common ancestry’) is the type-case. Jati
in India reiers to interdependent, hierarchically ranked, birth-ascribed
groups. The ranking is manifest in public esteem accorded the members
of the various groups, in the rewards available to them, in the power they
wield, and in the nature and mode of their interaction with others. Jatis
are regionally specific and culturally distinct, each is usually associated with
a traditional occupation and they are usually (but not always) endogamous.

They are grouped into more inclusive, pan-Indian ranked categories called
vaina which are frequently confused with the constituent jatis by those

using the term ’caste’. The rationale which justifies the system is both
religious and philosophical, relying upon the idea of ritual purity and
pollution to explain group rank, and upon the notions of right conduct
(dharma), just deserts (kai-t77a), and rebirth to explain the individual’s fate
within the system. As an explanation of caste inequalities this rationale
is advocated by those whom the system benefits, but is widely doubted,
differently interpreted, or regarded as inappropriately applied by those
whom the system oppresses.
Many students of stratification believe that the term ’caste’ conveys an
impression of consensus and tranquillity that does not obtain in systems
of rigid social stratification outside of India. That notion, however, is no
more applicable to, or derivable from, Indian caste than any other instance
of birth-ascribed stratification.100
If one concedes that caste can be defined cross-culturally (i.e., beyond
Hindu India), then the systems under discussion here are describable as
caste systems. That is, if one agrees that a caste system is one in which a
society is made up of birth-ascribed groups which are hierarchically
ordered, interdependent, and culturally distinct, and wherein the hierarchy
entails differential evaluation, rewards, and association, then whether one
uses the term ’caste’, or prefers ’ethnic stratification’, or some other term
is simply a matter of lexical preference. If one requires of a caste system
that it be based on consensus as to its rationale, its legitimacy, and the
legitimacy of the relative rank of its constituent groups, then none of the
examples mentioned here is a caste system. If one requires social tranquillity
as a characteristic, then too, none of these is a caste system. If one allows
that a caste system is held together by power and the ability of people
within it to predict fairly accurately one another’s behaviour while dis-
agreeing on almost anything or everything else, then all of these systems
will qualify. If one requires a specifically Hindu rationale of purity
390
and pollution and/or endogamy and/or strict and universal occupational
specialization, then one restricts caste to India and to only certain regions
and groups within India at that. If one requires for castes, as some do, a
tightly organized corporate structure, then too one would exclude even in
India some jatis and other groups commonly called ’castes’. (This, how-
ever, does seem to me to be the structural criterion which comes closest
to differentiating Indian jati from other systems of birth-ascribed stratifica-
tion such as that of the United States. Corporateness evidently emerges
as a response to oppression and as a mechanism for emancipation even
where it has been previously minimal, e.g., in Japan, Ruanda, and the
United States. Thus, the corporateness of Indian jatis may represent a
late stage of development in caste systems rather than a fundamental
difference in the Indian system.)
Jati in Hindu India and the equivalent but non-Hindu quom organization
in Swat and Muslim India, are each unique, yet both share the criteria
by which I have defined caste, as do the tri-partite system of Ruanda
and the essentially dual systems of Japan and the United States, and all
share in addition (and in consequence, I believe) a wide variety of social
and personal concomitants. Caste is a useful and widely used term because
it is concise, well-known, and in fact (as contrasted to phantasy), the
structural, functional, and existential analogy to Indian caste is valid for
many other systems.

Race
Systems of ’racial’ stratification are those in which birth-ascribed status
is associated with alleged physical differences among social categories,
which are culturally defined as present and important. Often these differ-
ences are more imagined than real, sometimes they are entirely fictional
and always a few physical traits are singled out for attention while most,
including some which might differently divide the society if they were
attended to, are ignored. Yet systems so described share the principle that
ranking is based on putatively inborn, ancestrally derived, and significant
physical characteristics.
Those who use this model for analysis generally base it upon the negative
importance attached by Europeans to the darker skin colour of those they
have colonized, exterminated, or enslaved. A good many have argued that
racially stratified societies are sui generis; that they are unique and hence
not comparable to societies stratified on any other basis.&dquo; There is often
a mystical quality to these arguments, as though race were an exalted,

uniquely ’real’, valid, and important criterion for birth ascription, render-
ing it incomparable to other criteria. An element of inadvertent racism
has in such instances infected the very study of race and stratification.
In fact, as is by now widely recognized, there is no society in the world
which ranks people on the basis of biological race, i.e., on the basis of
anything a competent geneticist would call ’race’, which means on the
basis of distinctive shared genetic makeup derived from a common gene
pool. ’Race’, as a basis for social rank is always a socially defined phenom- .
enon which at most only very imperfectly corresponds to genetically trans-
mitted traits and then, of course, only to phenotypes rather than genotypes.
Racists regard and treat people as alike or different because of their group
membership defined in terms of socially significant ancestry, not because
 391
of their genetic makeup. It could not be otherwise, for people are rarely
geneticists, yet they are frequently racists.
To state this point would seem to be superfluous if it were not for the
fact that it is continually ignored or contested by some influential scholars
and politicians as well as the lay racists who abound in many societies.
To cite but one well-known recent example, Arthur Jensen, in his article
on intelligence and scholastic achievement, maintains that there is a genetic
difference in learning ability between blacks and whites in the United
States. 12 Nowhere, however, does he offer evidence of how or to what
extent his ’Negro’ and ’White’ populations are genetically distinct. All
of those, and only those, defined in the conventional wisdom of American
folk culture to be ’Negro’ are included by Jensen, regardless of their genetic
makeup, in the category whose members he claims are biologically handi-
capped in learning ability. Thus, large numbers of people are tabulated
as ’Negroes’, a majority of whose ancestors were ’white’, and virtually all
of Jensen’s ’Negroes’ have significant but highly variable percentages of
’white’ ancestry. Although, also as a result of social definition, the ’whites’
do not have known ’Negro’ ancestry, the presumed genetic homogeneity
of the ’whites’ is as undemonstrated and unexplored as that of the ’Negroes’.
In short, there was no attempt to identify the genetic makeup or homogen-
eity of either group, the genetic distinctiveness of the two groups, or
whether or how genetic makeup is associated with learning ability, or how
learning ability is transmitted. This kind of reasoning is familiar and
expectable in American racism, but not in a supposedly scientific treatise-
a treatise whose author berates those who deplore his pseudo-science as
themselves unscientific for failing to seriously consider his ’evidence’.
The fallacy in Jensen’s case is that he has selected for investigation two
socially defined groupings in American society which are commonly
regarded as innately different in social worth and which as a result are
accorded widely and crucially divergent opportunities and life experiences.
Upon finding that they perform differentially in the context of school and
test performance, he attributes that fact to assumed but undemonstrated
and uninvestigated biological differences. Thus, socially defined populations
perform differently on socially defined tasks with socially acquired skills,
and this is attributed by Jensen to biology. There are other defects in
Jensen’s research, but none more fundamental than this. 13 One is reminded
of E. A. Ross’s succinct assessment of over fifty years ago, that ’ &dquo;race&dquo;
is the cheap explanation tyros offer for any collective trait that they are
too stupid or too lazy to trace to its origin in the physical environment,
the social environment, or historical conditions’.14
The point to be made here is that systems of ’racial’ stratification are
social phenomena based on social rather than biological facts. To be sure,
certain conspicuous characteristics which are genetically determined or
influenced (skin colour, hair form, facial conformation, stature, etc.) are
widely used as convenient indicators by which ancestry and hence ’racial’
identity is recognized. This is the ’colour bar’ which exists in many
societies. But such indicators are never sufficient in themselves to indicate
group membership and in some instances are wholly unreliable, for it is
parentage rather than appearance or genetics which is the basis for these
distinctions. One who does not display the overt characteristics of his
’racial’ group is still accorded its status if his relationship to the group is
known or can be discovered. The specific rules for ascertaining racial
392

identity differ from society to society. In America, if a person is known to

have had a sociologically black ancestor, he is black regardless of how
many of his ancestors were sociologically white (and even though he looks
and acts white). In South Africa, most American blacks would be regarded
as ’coloured’ rather than ’black’. Traditionally, in a mixed marriage, one
is a Jew only if one’s mother is a Jew. In contemporary India, an Anglo-
Indian has a male European ancestor in the paternal line; female and
maternal European ancestry are irrelevant. In racially stratified societies,
phenotypical traits are thus never more than clues to a person’s social
identity.
As Shibutani and Kwan have noted, ’a color line is something existing
in the presuppositions of menu ’... what is decisive about &dquo;race rela-
tions&dquo; is not that people are genetically different but that they approach
one another with dissimilar perspectives. ’16 Van den Berghe makes a
similar point: ’Race, of course, has no intrinsic significance, except to a
racist. To a social scientist, race acquires meaning only through its social
definition in a given society. 117
This is illustrated by the title of DeVos and Wagatsuma’s book, Japan’s
Invisible Race, dealing with the hereditarily stigmatized and oppressed
Burakumin. The Japanese believe that these people are physically and
morally distinct, and their segregation and oppression are explained on
that basis when in fact they are not so at all. Instead they are recognizable
only by family (ancestry), name, occupation, place of residence, life style,
etc. The Burakumin thus comprise a ’race’ in the sociological sense of
Western racism, but an ’invisible’ (i.e., not genetic or phenotypic) one.
The authors subtitled the book, Caste in Culture and Personality, shifting
the analogy from that of race (in the West) to that of caste (in India).
The book could as well have been entitled: Caste in Japan: Racial
Stratification in Culture and Personality.
The Japanese example brings up a point which needs to be made about
the alleged uniqueness of ’racial’ stratification. All systems of birth-
ascribed stratification seem to include a belief that the social distinctions
are reflected in biological (i.e., ’racial’) differences. That is, caste and other
ethnic differences are said to be revealed in physical makeup or appearance.
Associated with these supposed natural and unalterable inherited physical
characteristics are equally immutable traits of character, morality, intelli-
gence, personality, and purity. This is the case in Japan, where no actual
physical differences can be detected; it is true in India and Swat where
physical stereotypes about castes abound but actual differences are minimal;
it is true in Ruanda where the ranked groups all are black but are said
to differ in stature and physiognomy as well as in culture; it is true in the
United States where the physical differences are commonly and erroneously
thought to be absolute. Cultural factors have to be relied upon in addition
to whatever biological ones may be present, in order to make the important
discriminations upon which ranked social interaction depends, and even
then mistakes are frequently made. Throughout the world, people who
look distinctive are likely to be regarded as socially different; people who
are regarded as socially different are likely to be thought to look distinctive.
They are also likely to be required to dress and act distinctively.
I suggest that, just as societies frequently dramatize the social differences
among kin groups (e.g., sibs, clans, phratries) by giving to them totemic
names and attributing to them characteristics of animals or plants, thereby
 393

identifying the social differences with biological species differences,&dquo;,

so also, societies with birth-ascribed status hierarchies dramatize and
legitimize these crucial social differences by attributing to them innate
biological, hence ’racial’, differences. As a result, the concept of mis-
cegenation arises, based on an ideology of innate difference contradicted
by a persistent and recurrent perception of similarity by people of opposite
sex across social boundaries.19
Thus, organization and ethnic stratification include racism; racial
caste
stratification is congruent with caste and ethnic stratification. Their
ultimate coalescence is in the imputation of biological differences to explain
and justify birth-ascribed social inequality. In this regard, sexual stratifica-
tion can be seen to be a phenomenon of the same order.
This universality of racism in birth-ascribed stratification can be under-
stood in the fact that physical traits not only dramatize social differentia-
tion, but can also explain and justify it. The effect of such explanation is
to make social inequality appear to be a natural necessity rather than a
human choice and hence an artificial imposition. Social distinctions are
man-made and learned; what man makes and learns he can unmake and
unlearn. What God or biology has ordained is beyond man’s control.
The former may be defined as artificial, unjust, untenable, and remediable;
the latter as inevitable or divinely sanctioned. This is important because
birth-ascribed stratification is widely or universally resented by those whom
it oppresses (at least as it affects them), and advocated by those it rewards.
Both categories share the human capability of empathy, and it inspires envy
and resentment in the one and fear or guilt in the other. Racism-the
self-righteous rationalization in terms of biology-is a desperate and
perhaps ultimately futile attempt to counteract those subversive emotions.
In sum, ’race’, as commonly used by social scientists, emphasizes
common physical characteristics (as does ’sex’); ’caste’ emphasizes com-
mon rank, occupational specialization, endogamy, and corporate organiza-

tion ; ’ethnic stratification’ emphasizes cultural distinctiveness. These are

real differences in meaning, but the degree of empirical overlap in systems
so described, and the commonalities in the existential worlds of those who
live within them are so great as to render the distinctions among them
largely arbitrary, and irrelevant, for many purposes. Individual cases
differ, but as types of social stratification, they are similar. With equal
facility and comparable effect, they utilize as evidence of social identity
anything which is passed on within the group: skin colour, hair form,
stature, physiognomy, language, dress, occupation, place of residence,
genealogy, behaviour patterns, religion. None is wholly reliable, all are
difficult to dissimulate. In any case, strong sanctions can be brought to bear
to minimize the temptation to ’pass’ among those who might be capable
and tempted. As the case of India suggests and Japan confirms, social
criteria can be as rigid as physical ones.
.; ...’ -~..).......’I,}.

’Race’ versus ’Caste’ i:::;¡ L.:l; &dquo;

Considerable controversy has surrounded the terms ’race’ and ’caste’
when applied outside of the contexts in which they originated and to which
they have been most widely applied: Western colonialism and Hindu
India, respectively. This is understandable because there are important
peculiarities in each of these situations, and to extend the terms beyond
them requires that those peculiarities be subordinated to significant
 394
similarities. Systems of birth-ascribed inequality are sufficiently similar,
however, to invite comparative study, and some general term is needed
to refer to them. ’Caste’ has seemed to me more useful than ’race’, because
it refers to social rather than allegedly biological distinctions, and it is the
social distinctions which are universal in such systems. If it were a catchier
term, ’ethnic stratification’ might replace both in the social scientific liter-
ature. Unfortunately it is not, so we must probably await a better term or
tolerate continuing terminological dispute and confusion. In any case, it
is the nature of birth-ascribed stratification-the ideas, behaviours, and
experiences which comprise it, the effects it has on persons and societies
and, quite frankly, the means by which it may be eliminated-in which I
am interested. The words applied to it are of little importance. When I

try to explain American race relations to Indians, I describe and analyse

America as a caste stratified society, with attention to the similarities
and differences in comparison with India. If I am trying to explain Indian
caste stratification to Americans, I describe and analyse India as a racist
society, with attention to the similarities and differences in comparison
to the United States. I do this as a matter of translation from the social
idiom of one society to the other. It is the most economic, vivid, and
accurate way I know to convey these phenomena to people whose experi-
ence is limited to one system or the other. I do not think Indian caste is
American race, or vice versa, but neither do I think that race stratification
in America is race stratification in South Africa, or that caste in India is
caste in Swat, or that caste in the Punjab is caste in Kerala. Neither do I
think racial stratification and racism are the same for blacks, Chicanos,
and whites in America, or that caste stratification and casteism are the
same for sweepers, blacksmiths, and Rajputs in Hindu India. There
are features in all of these which are the same in important ways, and by

focusing on these I think we can understand and explain and predict the
experience of people in these diverse situations better than if we regard
each of them as unique in every way.
Objections to the cross-cultural comparison of race and caste depend
either on an insistence that the two would have to be wholly identical to
justify such comparison, or more commonly, on misconceptions about one
or both of the systems being compared. It is worthwhile to identify and
comment upon some of these objections.
1. The most prevalent objection among experts on Western social stratifica-
tion is that caste status is accepted and endorsed by those in the system
whereas racial stratification is objected to and striven against by those it
oppresses. Thus Cox asserts that ’while the caste system may be thought
of as a social order in stable equilibrium, the domination of one race by
another is always an unstable situation.... The instability of the situation
produces what are known as race problems, phenomena unknown to the
caste system.’ I have contended with this claim in some detail elsewhere. 20
Suffice it to say here that anyone who has known low caste people in India
can affirm that this particular contrast is imaginary, as can anyone who
knows the history of religious conversion, social reform, and social
mobility striving in India, who has followed the reports of the Com-
missioner of Scheduled Castes and Tribes, or who simply reads the news
releases from India today.
2. It is often argued that caste in India is unique and noncomparable
.
because of the elaborate religio-philosophical rationale which underlies
 395

it, and that racial stratification is unique because it is based on colour.21

As I have pointed out, these contentions are questionable on empirical
grounds. Few would deny, for example, that caste exists in Swat as well
as India, and that it exists in Mysore; India, among both the Lingayats
and the Hindus, yet the rationale is very different in each of these cases.
On the other hand, few would deny that colour consciousness is an
important part of the ideology and metaphor of caste in India.
But on another level, these arguments are not necessarily relevant to the
issue of comparison. That phenomena are in some respects different
does not make them incomparable. They may still be similar in crucial
respects. Everything is, after all, unique, but there can be no science of the
unique. The claim for comparability can be refuted by showing that the
facts are in error, or that the interpretations of their significance are falla-
cious, not by citing differences extraneous to the argument being made.
Moreover, comparability (or incomparability) for one purpose or in one
context does not imply comparability (or incomparability) for another
purpose or in another context. The test must be whether or not and how
well explanation can be derived from a particular comparison.
If one knows that a society is characterized by birth-ascribed stratifica-
tion, he can predict a great deal about the experiences and attitudes of
people at different levels within it, about the social, political, and economic
behaviour they are likely to engage in under various circumstances, and
about the nature and consequences of social conflict.22 This, to my mind,
justifies the comparison.
3. A number of specific characteristics of caste in India or race in the West
have been cited as bases for non-comparability with other systems of birth-
ascribed stratification. (a) Endogamy has been cited as essential to caste.
However, what is crucial and universal in a caste system is not endogamy,
but birth-ascribed membership in a ranked category or group. Endogamy
is the most common method for achieving this in India and elsewhere, but
a firm patrilineal or matrilineal rule of status assignment will do as well,
and so will a rule which assigns to an individual the caste of his higher status
parent or of his lower status parent, or one which designates that an indi-
vidual whose parents are of castes A and B will be of caste C. All of these
rules of birth-ascription occur in South Asian castes.23
(b) Occupational distinctiveness has been cited as essential to caste. This
is indeed an almost inevitable concomitant of social separation, stratifica-
tion, and interdependence, be it birth-ascribed or not. But again, it is not
universal among castes even in India, nor is it lacking in racial and sexual
stratification. Some have maintained that caste, without a conspicuous
racial basis, cannot exist in a complex society simply because the anonymity
of such a society would make dissimulation and passing too easy to sustain
a caste system where no physical indicators of identity existed.24 It is
probably true that in the anonymity and mobility of contemporary urban
life, rigid ethnic stratification is increasingly difficulty to maintain when the
indicators of identity are learned, for learned characteristics can be
unlearned, suppressed, or learned by those to whom they are inappropriate.
To manipulate these indicators is often difficult, as the persistence of the
Burakumin of Japan makes clear, because the identifying characteristics
may be learned very early (language, gesture), and may be enforced from
without as well as from within (dress, deference, occupation), but it is
 396

possible, as instances of passing make clear.25 The more personal relation-

ships of traditional, small-scale societies, together with their formal and
informal barriers and sanctions against casual or promiscuous interaction
militate against the learning or expression of inappropriate status character-
istics and conspicuous indicators of status are largely unnecessary.
Reliable, immutable, and conspicuous indicators of identity are thus more
important to systems of birth-ascribed stratification in the anonymity and
mobility of the city than in the village, but the internal pressures of
ethnic pride combined with the external pressures of ethnic discrimination
and the vested interests which sustain it make such systems possible in
even the most unlikely-seeming circumstances.
(c) Some have argued that the corporate structure of many Indian
castes renders them incomparable to most instances of ethnic stratifi-
cation, a point which I have discussed above (p. 390) and will there-
fore not repeat.
(d) The comparison between black-white stratification in America and
Indian caste is occasionally contested on grounds that the former is a
dichotomy while the latter is a complex hierarchy. This is a real difference
between the two, but it does not make them incomparable. It is a differ-
ence in numbers of groups, not in race versus caste. For one thing, there
are other racial or caste-like groups in American society: Native Americans,

Chicanos, Puerto Ricans, Asians, etc. In India, on the other hand, most
interaction is not caste specific so much as it is specific to categories of
castes (categories such as ’untouchables’, ’twice-born’, the varnas, etc.),
thereby simplifying the interactional situation greatly. For example,
the Khas-Dom (twice-born versus untouchable) dichotomy in the mountain
area of my research 26 and the Brahmin-non-Brahmin dichotomy in South
India are quite comparable to the situation described as the colour-bar
elsewhere. One probable difference between dichotomous and multiple
hierarchies may be that in the former the oppressed characteristically
oppose the system as such. To rise within it would be to displace the
privileged-to reverse the hierarchy-and that is impracticable. To over-
throw the system is to erase their oppression and vice versa. They have
nothing to lose but their inferiority. In multiple hierarchies, objection may
more characteristically be made not to the system, but to the place of one’s
own group within it, the reason being that to overthrow the system is not
feasible for any one of the many groups within it, each of which is relatively
small, weak, and in competition with others. Moreover, successful
overthrow would result not only in equality with elites, but equality with
erstwhile inferiors. Therefore, it is regarded as more practicable and
more rewarding to attempt to rise within the system than to eliminate the
system itself. These remarks apply to systems of birth-ascribed stratifica-
tion regardless of the criteria used.
(e) Finally, it is sometimes held that racial stratification is an outgrowth
of Western colonialism, and hence India can be regarded as an example
thereof only in the relationship of Indians to the British, not in the relation-
ship among indigenous castes. I would maintain that caste in India is in
fact a product of colonialism and that its present manifestations are
closely analogous to the ’internal colonialism’ which Blauner, among
others, has described for the United States (see below).27 In India this is
an instance of what has recently been termed ’fourth-world’ colonialism,
.
i.e., exploitation inflicted by ’third-world’ (non-Western) people on their
 397
internal minorities, analogous to that they have often experienced them-
selves at the hands of ’first-’ and ’second-world’ colonialists (Western non-
communist and communist nations, respectively).
In short, the further one probes into the nature and dynamics of race
and caste, and into the experience of those who live them, the more it
becomes apparent that they are similar, comparable, phenomena.

Colonialism
The concept of colonialism has gained popularity in recent years for the
analysis of racism and racial stratification in the West. 28 It therefore merits
further discussion. This model focuses on the history of Western expansion
and the exploitation of alien peoples, emphasizing notions of the superiority
of the dominant, Western, white society whose members arrogated privi-
lege to themselves through the exercise of power (usually technological,
often military) to dominate, control, exploit, and oppress others. Racism
has been an integral aspect of this process, for there usually have been
differences in colour between the colonizer and the colonized which were
used to account for the alleged inferiority in ability, character, and men-
tality which in turn were used to justify colonial domination. Colonialism
has been most often described as the result of overseas conquest, in which
case the colonizing group has usually comprised a numerical minority.
Less often colonialism has included conquest or expansion across
national boundaries overland, but the results are the same, if the romance
is less. These phenomena have recently come to be termed ’external
colonialism’, in contrast to ’internal colonialism’ which refers to similar
domination and exploitation, within a nation, of an indigenous, over-run,
or imported minority. This distinction directs attention to the locus
of colonial domination whereas the distinction between third-world and
fourth-world colonies cited above, directs attention to the sources of that
domination.
While it has not been much easier to gain acceptance of the colonial
model for analysis of American race relations than it has been to gain
acceptance of the caste model, it is clear that here again, the problem is
semantic rather than substantive. Some of those who argue persuasively
the cross-cultural and multi-situational applicability of the colonial
model, deny such applicability for the caste model and in so doing use
precisely the logic and data they deplore and regard as faulty when their
intellectual adversaries deny applicability of ’colonialism’ outside of the
classical overseas context.29
Colonialism, external and internal, is a process which has occurred
repeatedly, in many contexts, with many specific manifestations and
many common results. It long antedates the recent period of European
and American expansion. Caste stratification, racial stratification, ethnic
stratification, and ’pluralism’ have been its recurrent products.3°0 The
point can be made with specific reference to caste in India. Rather than
regarding colonialism as an antecedent condition which excludes tradi-
tional India from the category of racially or ethnically stratified
societies, it can well be used as a basis for assigning India historical priority
among such societies, in the contemporary world. That is, traditional
India may represent the most fully evolved and complex post-colonial
society in the world. It is easy to obtain explanations of caste from
398
informants or books in India which refer directly to the presumed early
domination of primitive indigenes by advanced invaders. There is little
doubt that the present caste system had its origins some 3,000 to 3,500
years ago in a socio-cultural confrontation that was essentially colonial.
Low status was imposed on technologically disadvantaged indigenes by
more sophisticated, militarily and administratively superior peoples who
encroached or invaded from the north and west, arrogating to them-
selves high rank, privileges, and land. The large number of local and ethnic-
ally distinct groups on the sub-continent were fitted into a scheme of social
hierarchy which was brought in or superimposed by the high status
outsiders, culminating in the caste system we know today.31 Social
separation and social hierarchy based on ancestry became the essence of
the system; colonial relations were its genesis. Even today, most tribal
people-those who are geographically and economically marginal and
culturally distinct-are incorporated into Hindu society, if at all, at the
bottom of the hierarchy (except in those rare instances where they have
maintained control over land or other important sources of income and
power).
If one were to speculate on the course of evolution which ethnic stratifi-
cation might take in the United States in the context of internal colonialism,
of rigid separation, hierarchy, and discrimination which are part of it,
and the demands for ethnic autonomy which arise in response to it, one
possibility would be a caste system similar to, though less complex than
that of India. The historical circumstances may be rather similar despite
the separation of many hundreds of years, many thousands of miles and a
chasm of cultural difference. Actually, development of the degree of social
separation common in India seems at this point unlikely given the mass
communications and mass education in the United States, its relative
prosperity, and the rather widespread (but far from universal) commit-
ment to at least the trappings of social equality. But surely if anything is
to be learned from history and from comparison, the case of the Indian
sub-continent should be of major interest to students of American race
and ethnic relations, social stratification, and internal colonialism.
In sum, colonialism is as inextricable from caste and race as caste and
race are from one another. There may be instances of colonialism where

birth-ascription is or becomes irrelevant, but every instance of caste, race,

and ethnic stratification includes, and relies for its perpetuation upon,
the kind of ethnic domination and exploitation that defines colonialism.

Class
Closely associated with each of the models discussed here is that of
social class. Class is a matter of acquired status rather than of birth-
ascription, and is in this respect distinct from race, caste, and ethnic
stratification, with different social consequences. In a class system, one is
ranked in accord with his behaviour and attributes (income, occupation,
education, life style, etc.). In a birth-ascribed system, by contrast, one
behaves and exhibits attributes in accord with his rank. In a class system,
individual mobility is legitimate, albeit often difficult, while in ascribed
stratification it is explicitly forbidden. Systems of acquired rank-class
systems-prescribe the means to social mobility; systems of ascribed rank
proscribe them. As a consequence, a class system is a continuum; there
are individuals who are intergrades, there are individuals in the process of
 399
movement, there are individuals who have experienced more than one
rank. Miscegenation is not an issue because there are no ancestrally
distinct groups to be inappropriately mixed. A birth-ascribed system is
comprised of discrete ranks on the pattern of echelon organization,
without legitimate mobility, without intergrades; the strata are named,
publicly recognized, clearly bounded. Miscegenation is therefore a social
issue. In a system of acquired ranks, the strata may be indistinct, imper-
fectly known, or even unknown to those within the system. In fact, there
is considerable debate among students of stratification as to whether or not
awareness of class is essential to a definition of class. Some hold that social
classes are properly defined by social analysts who use such criteria as
income to designate categories which may be entirely unrecognized by
those in the society.
In a class system individuals regard themselves as potentially able to
change status legitimately within the system through fortune, misfortune,
or individual and family efforts. In a birth-ascribed system, individuals
know that legitimate status change is impossible-that only dissimulation,
revolution, or an improbable change in publicly accorded social identity
can alter one’s rank and hence life-chances.

Despite these differences, class is in no way incompatible with birth-

ascribed systems. In fact, in so far as it is a term for categories of people
ranked by income, occupation, education, and life style, it co-occurs with
them. Low castes, despised races, ethnic minorities, and colonized people
comprise economically and occupationally depressed, exploited classes
who are politically and socially oppressed; high castes, exalted races,
privileged ethnic groups, and colonizers comprise economically and
occupationally privileged, power-wielding, elite classes who live off the
labour of others. In this respect, class differences pervade and reinforce
systems of birth-ascribed stratification. Furthermore, it is not unusual
to find significant class differentials within a caste, racial, or ethnic group or
within a colonized or colonial group.32 That is, class, in the conventional
sense often occurs conspicuously within such groups, and may also bridge
their boundaries without obscuring them. But it is not possible to analyse
birth-ascribed stratification solely in terms of class, for no amount of class
mobility will exempt a person from the crucial implications of his birth
in such systems.
Those who have sought to identify the positions of European immigrants
to America such as the Poles, Italians, and Irish, with the position of blacks,
Native Americans, Chicanos, and Asians have failed to discern the
essential fact that racism is the basis of American caste, and that it bestows
upon those who experience it a unique social, political, and economic stigma
which is not bestowed by class or national origin. Second generation
white Europeans can meet all of the criteria for acceptance into the
American white race-caste for they are regarded as being only culturally
different. A fifteenth generation American black, or a fifteen-hundredth
generation American Indian cannot, for their differences are regarded
as innate, immutable, and crucial. Equalitarianism has produced no
’American dilemma’ among racists, as Myrdal believed, simply because it
is an equality for whites only, and its extension to other groups has moved
slowly, painfully, and with vehement opposition, even where it has moved
at all.
 400
Systems of collective social rank, whether ascribed or acquired, are sys-
tems for retaining privilege among the powerful and power among the
privileged, reserving and maintaining vulnerability, oppression, and want
for those upon whom it can be imposed with minimal risk while retaining
their services and their deference. In this way they are similar. In the prin-
ciples of recruitment and organization by which that similarity is effected
and in the individuals’ prospects for mobility they differ, and those
differences have important consequences for individual life experience
’

and social processes in the societies which harbour them.

Pluralism
Pluralism is a model which has been applied to socially and culturally
diverse societies since the writings of Furnivall on South-East Asia.33
Cultural pluralism obtains when ’two or more different cultural traditions
characterize the population of a given society’; it is ’a special form of
differentiation based on institutional divergences’ .34 Systems of birth-
ascribed stratification are inevitably systems of social and cultural pluralism
because they are accompanied by social separation. In a caste system,
’Because intensive and status-equal interaction is limited to the caste, a
common and distinctive caste culture is assured. This is a function of the
quality and density of communication within the group, for culture is
learned, shared and transmitted. ’35 The same is true for any system of
racial or ethnic stratification. M. G. Smith has noted, ’it is perfectly
clear that in any social system based on intense cleavages and discontinuity
between differentiated segments the community of values or social relations
between these sections will be correspondingly low. This is precisely the
structural condition of the plural society.’3g And I have noted elsewhere
that,
... castes are discrete social and cultural entities.... They are maintained by
defining and maintaining boundaries between castes; they are threatened when
boundaries are compromised. Even when interaction between castes is maximal
and cultural differences are minimal, the ideal of mutual isolation and distinc-
tiveness is maintained and advertised among those who value the system. Simi-
larly, even when mobility within, or subversion of the system is rampant, a
myth of stability is stolidly maintained among those who benefit from the system. 31

Mutual isolation of social groups inevitably leads to group-specific

institutions (an important criterion for pluralism according to Furnivall),
because members are excluded from participation in the institutions of
other groups.
Caste, race, and ethnic stratification, like all plural systems therefore,
are systems of social separation and cultural heterogeneity, maintained
by common or over-riding economic and political institutions rather than
by agreement or consensus regarding the stratification system and its
rationale.38 This does not deny consensus, it only defines its nature:

In caste systems, as in all plural systems, highly differentiated groups get along
together despite widely differing subjective definitions of the situation because
they agree on the objective facts of what is happening and what is likely to happen
-on who has the power, and how, under what circumstances, and for what
purposes it is likely to be exercised. They cease to get along when this crucial
agreement changes or is challenged. 39
 401
The constituent social elements of plural societies need not be birth-
ascribed, and they need not be (and sometimes are not) ranked relative to
one another, although by Furnivall’s definition, one element must be dom-
inant. In fact, unranked pluralism is the goal many ethnic minorities choose
over either stratification or assimilation. But a system of birth-ascribed
stratification is always culturally, socially and hence institutionally hetero-
geneous, and thus pluralistic.

Hierarchy as Symbolic Interaction

I have elsewhere described the universality among social hierarchies of
patterns of interaction which symbolize superiority and inferiority.4o
Social hierarchy, after all, exists only in the experiences, behaviours, and
beliefs of those who comprise it. Interpersonal interaction becomes
the vehicle for expression of hierarchy: for asserting, testing, validating,
or rejecting claims to status. Almost every interaction between members
of ranked groups expresses rank claimed, perceived, or accorded. When
the hierarchy is birth-ascribed, the membership of its component groups
is ideally stable, well-known, and easily recognizable. In such systems
people are perceived by those outside of their groups almost wholly in
terms of their group identity rather than as individuals. They are regarded
as sharing the characteristics which are conventionally attributed to the

group and they share the obligations, responsibilities, privileges, and

disabilities of their group. In intergroup relations, therefore, one individual
is substitutable for another in his group, for all are alike and interchange-
able. This is the setting for prejudice, discrimination, bigotry, chauvinism,
and is an ideal situation for scapegoating. These attitudes and their
behavioural consequences are designated and deplored by such terms as
racism, casteism, communalism (referring to ethnic chauvinism of various
sorts), and recently, sexism. They are characterized by domination,
deprivation, oppression, exploitation, and denigration directed downward;
obedience, acquiescence, service, deference, and honour demanded from
above. They result in envy, resentment, dissimulation, and resistance
arising from below, balanced from above by fear, guilt, and that combin-
ation of arrogant self-righteousness and rationalization which is found in
all such systems. Maya Angelou has aptly characterized the result in Ameri-
can race relations as ’the humorless puzzle of inequality and hate’;
’the question of worth and values, of aggressive inferiority and aggressive
arrogance’41 which confronts and exacts its toll not only from black
Americans, but from the denizens of all those jungles of inherited inequality
I call caste systems. It is this quality of interpersonal relations rather than
any particular event or structural feature which struck me most vividly,
forcefully, and surprisingly as similar in Alabama and India when I
first experienced them for over a year each within a period of five years.42
For me, this is the hallmark of oppressive, birth-ascribed stratification.
A specifically interactional definition of caste systems applies equally to
all systems of birth-ascribed stratification: ’a system of birth-ascribed
groups each of which comprises for its members the maximum limit of
status-equal interaction, and between all of which interaction is consistently
hierarchical.’43 The cultural symbols of hierarchical interaction vary; the
presence and importance of such symbols is universal and essential to
racism, casteism, and their homologues.... _ , . _ _ . ......
402

Hierarchy as Ideology
Dumont has emphasized the point that Indian caste is unique in that it
is based on an ideology of hierarchy defined in terms of ritual purity and
pollution.44 He regards other systems of hierarchical social separation as
non-comparable because of the inevitable differences in the ideologies
supporting them. In the comparative framework which I advocate, I
maintain simply that the Hindu rationale is one of several ideologies (cf.
those of Islamic Swat, of the South Indian Lingayats to whom purity is
irrelevant, of Ruanda, of Japan, and the United States) which can and do
underlie and justify systems of birth-ascribed social hierarchy. Each is
unique to the culture in which it occurs; each is associated with remarkably
similar social structures, social processes, and individual experiences. I
believe that anyone who has experienced daily life in rural India and the
rural American South, for example, will confirm the fact that there is
something remarkably similar in the systems of social relations and atti-
tudes. I believe that anyone who has experienced daily life in an urban
slum, a public market, or a factory in India and the United States would
come to the same conclusion. That similarity is generated by birth-ascribed
stratification and it is not concealed by differential ideologies.45
Contrary to another of Dumont’s assumptions (shared with Cox), there
is nothing incompatible between an ideology which underwrites a hierarchy
of groups and a notion of equality within each group. This combination,
in fact, is found not only in the United States where it accounts for the
above-mentioned absence of a real ’American dilemma’ in race relations,
but also in each of the other systems described here. Members of each
ranked group are inherently unequal to those of each other group and are
by birth potentially equal to those of their own group. More importantly,
the existence of an ideology of hierarchy does not mean that this ideology
is conceived and interpreted identically by all within the system it is
presumed to justify or even that it is shared by them. Acquiescence must
not be mistaken for concurrence. Dumont’s assumption to the contrary
is the most glaring weakness in his analysis of Indian caste. 46

Sexual Stratification
Finally, in my discussion of models for analysis, I turn to the controver-
sial and sociologically puzzling matter of sex as a basis for social separation
and inequality. The special problems which the sexual criterion poses
for the student of stratification are both academic and substantive. The
academic problems derive from the history of the study of stratification.
Although the role of women in various non-Western societies has been
discussed by anthropologists (including prominently Margaret Mead),
and the position of women in European societies has been discussed by
some social historians, the sexual dichotomy rarely appears in sociological
works on stratification. That this criterion has been largely ignored or
dismissed by stratification theorists is attributable to several factors not
the least of which is no doubt that members of the privileged sex have
authored most of the work and to them such ranking has not been a
problem and hence has not been apparent. Also, their culturally derived
biases have been such that this kind of ranking was taken for granted as a
manifestation of biological differences. ’Many people who are very hip
to the implications of the racial caste system ... don’t seem to be able to
see the sexual caste system and if the question is raised they respond
 403
with: &dquo;that’s the way it’s supposed to be. There are biological differences.&dquo;
Or with other statements which recall a white segregationist confronted
with integration. 141 The biological rationale-what Millett refers to as the
’view of sex as a caste structure ratified by nature’4g-recalls also the
justification offered for all birth-ascribed dominance-exploitation relation-
ships be they caste in India, Burakumin status in Japan, sexual roles, or
any other. In each instance the plea is that these are uniquely real,
significant, unavoidable, and natural differences, and therefore they must
be acted upon. Thus, in an interview about their new book, The Imperial
Animal, which is said to claim that males have dominated human history
because ’the business of politics ... is a business that requires skills and
attitudes that are peculiarly male,’ anthropologists Robin Fox and Lionel
Tiger were reported to have vehemently denied that their theory about the
reasons for women’s roles might be a sexist theory. ’ &dquo;These are the facts,
don’t accuse us of making up the species,&dquo; Tiger said.’ And again,
’
&dquo;Because this is a racist country, people relate sexism to racism.&dquo; But
these two reactions are actually different because while there are no
important biological differences between races, there are very important
differences between the sexes.’49 Whether the differences are real or not
(and who would deny that males and females differ in important ways ?),
the sociological and humanistic question is whether the differences require
or justify differential opportunities, privileges, responsibilities, and rewards
or, put negatively, domination and exploitation.
Birth-ascribed stratification be it sexual, racial, or otherwise, is always
accompanied by explanations, occasionally ingenious but usually mundane
and often ludicrous, as to why putative natural differences do require and
justify social differences. Those explanations are widely doubted by those
whose domination they are supposed to explain, and this includes increas-
ing numbers of women.
The substantive issues which becloud the topic of sexual stratification
have to do with the mode of recruitment, the socialization, membership,
and structural arrangements of sexually ranked categories. First, there is
the fact that while sex is determined at birth, it is not contingent upon
ancestry, endogamy, or any other arrangement of marriage or family,
and is not predictable. It is the only recurrent basis for birth-ascribed
stratification that can be defensibly attributed solely to undeniably physical
characteristics. Even here there are individual or categorical exceptions
made for transvestites, hermaphrodites, homosexuals, etc., in some
societies as in the case of hijaras in India.5° The significance (as contrasted
to the fact) of the diagnostic physical traits-of sexual differences-is,
however, largely socially defined, so that their cultural expressions vary
widely over time and space. Second, as a concomitant to the mode of re-
cruitment, males and females have no distinct ethnic or regional histories.
It must not be overlooked, however, that they do have distinct social
histories in every society. Third, the universal co-residence of males and
females within the household precludes the existence of lifelong separate
male and female societies as such, and usually assures a degree of mutual
early socialization. But note that it does not preclude distinct male and
female social institutions, distinct patterns of social interaction within and
between these categories, or distinguishable male and female subcultures
(in fact the latter are universal) including, for example, distinct male and
female dialects.
404

Partly as a consequence of these factors, the nature and quality of

segregation of the sexes has not been defined by sociologists as comparable
to that of the other ascriptive social categories discussed here. Neverthe-
less, most of the characteristics of birth-ascribed separation and stratifica-
tion (racial, caste, ethnic, colonial, class, and pluralistic characteristics),
and virtually all of the psychological and social consequences of inborn,
lifelong superiority-inferiority relations are to be found in the relationship
of males and females in most societies. These stem from similar factors in
early socialization and from stereotypes and prejudices enacted and
enforced in differential roles and opportunity structures, rationalized
by ideologies of differential intrinsic capabilities and worth, sustained and
defended through the combination of power and vested interest that is
common to all birth-ascribed inequality. I have elsewhere contrasted some
of the consequences of these assumptions and behaviours in the United
States and India as reflected in the political participation of women in the
two nations although this is dwarfed by Millett’s more recent work on male
domination, its sources, and manifestations in the West.51
If we agree with van den Berghe that ’race can be treated as a special
case of invidious status differentiation or a special criterion of stratifica-

tion,’52 I think we are bound to agree that sex is another.

CONSEQUENCES OF INHERITED INEQUALITY

Assuming that there are significant structural and interactional similarities

among systems of birth-ascribed stratification, the question can still be
legitimately asked, ’so what?’ Is this merely a more or less interesting
observation-even a truism-or does it have some theoretical or practical
significance ? My answer would be that it has both, for such systems
have common and predictable consequences in the individual lives of those
who live them and in the cumulative events which comprise the ongoing
histories of the societies which harbour them.

Caste systems are living environments to those who comprise them. Yet there is
a tendency among those who study and analyse them to intellectualize caste,
and in the process to squeeze the life out of it. Caste is people, and especially
people interacting in characteristic ways and thinking in characteristic ways.
Thus, in addition to being a structure, a caste system is a set of human relation-
ships and it is a state of mind. 53

Their ’human implications’ are justification enough for studying and

comparing systems of birth-ascribed stratification. There are neither the
data nor the space to discuss these implications fully here, but I will
suggest the nature of the evidence briefly, identifying psychological and
social consequences. I am well aware that many features of such systems
are found in all sharply stratified societies. Some are characteristic of all
relationships of superordinationand subordination, of poverty and
affluence, of differentialpower. Others are found in all societies made up
of distinct sub-groups whether stratified or not. It is the unique combina-
tion of characteristics in the context of the ideal of utter rigidity and
unmitigable inequality which makes systems of stratification by race,
caste, ethnicity, and sex distinctive in their impact on people, individually
and collectively. ,.. , , , . _ ~ &dquo; &dquo; .
&dquo; ,
_
.
 405

Psychological Consequences
Beliefs and attitudes associated with rigid stratification can be suggested
by such terms as paternalism and dependence, noblesse oblige, arrogance,
envy, resentment, hatred, prejudice, rationalization, emulation, self-doubt,
and self-hatred. Those who are oppressed often respond to such stratifica-
tion by attempting to escape either the circumstances or the consequences
of the system. The realities of power and dependence make more usual
an accommodation to oppression which, however, is likely to be less passive
than is often supposed, and is likely to be unequivocally revealed when the
slightest change in the perceived distribution of power occurs. Those who
are privileged in the system seek to sustain and justify it, devoting much
of their physical effort to the former and much of their psychic and verbal
effort to the latter. When these systems are birth-ascribed, all of these
features are exacerbated.
Kardiner and Ovesey conclude their classic, and by now outdated, study
of American Negro personality, Mark of Oppression, with the statement :
’The psycho-social expressions of the Negro personality that we have
described are the integrated end products of the process of oppression. ’54
Although it is appropriate to question their characterization of that
personality in the light of subsequent events and research, there is no
doubt that such oppression has recurrent psychological consequences for
both the oppressor and the oppressed, as Robert Coles has demonstrated
in Children of Crisis and subsequent works. 55
Oppression does not befall everyone in a system of birth-ascribed in-
equality. Most notably, it does not befall those with power. What does
befall all is the imposition by birth of unalterable membership in ranked,
socially isolated, but interacting groups with rigidly defined and con-
spicuously different experiences, opportunities, public esteem and, in-
evitably, self-esteem. The black in America and in South Africa, the
Burakumin of Japan, the Harijan of India, the barber or washerman of
Swat, the Hutu or Twa of Ruanda, have all faced similar conditions
as individuals and they have responded to them in similar ways. The same
can be said for the privileged and dominant groups in each of these
societies, for while painful consequences of subordination are readily
apparent, the consequences of superordination are equally real and
important. Thus, ethnic stratification leaves its characteristic and indelible
imprint on all who experience it.
The consequences of such stratification include many of the attitudes and
responses vividly described in literature on black-white relations in the
United States. Immediately to mind come accounts of the black experience,
such as James Baldwin’s Notes of a Native Son, Nobody Knows My Name,
and Go Tell it On the Mountain; Claude Brown’s Manchild in the Prorrcised
Land; Eldridge Cleaver’s Soul on Ice; Ralph Ellison’s Invisible Man;
Richard Wright’s Native Son and Black Boy; and Malcolm X’s The
Autobiography of Malcolm X. Outstanding among those dealing with the
white experience in relation to blacks are W. J. Cash’s The Mind of the
South; Lillian Smith’s Killers of the Dream and Strange Fruit. The psychia-
trist Robert Coles has provided insights into both sides of the American
caste barrier in his works cited above.
Corresponding literature on other caste-like systems include, for India,
Mulk Raj Anand’s Untouchable; Hazari’s An Indian Outcaste: The Auto-
biography of an Untouchable; Harold Isaac’s India’s Ex-Untouchables;
406

and, from the British side, E. M. Forster’s Passage to India; and in Burma,
George Orwell’s Burmese Days. On Japan, we can refer to the contributors
to DeVos and Wagatsuma’s Japan’s Invisible Race, and, cited therein,
Ninomiya’s ’An Inquiry Concerning the Origin, Development and Present
Situation of the Eta ...’, and Shimazaki Tbson’s Hakai (Breach of Com-
mandment). Works on South Africa include Alan Paton’s Cry the Beloved
Country and Too Late the Phalarope; Albert Luthuli’s Let My People Go;
and van den Berghe’s Caneville and South Africa: A Study in Conflict.
Telling analyses of colonial situations include Franz Fanon’s Wretched
of the Earth, and Albert Memmi’s The Colonizer and the Colonized, both
deriving from Algeria; O. Mannoni’s Prospero and Caliban, from Mada-
gascar ; J. C. Heinrich, The Psychology of a Suppressed People, on India.
This listing does not do justice to the relevant literature, but is suggestive
of it.
Research on the psychological consequences of racism in the United
States is well-known and voluminous, though restricted almost exclusively
to black-white relations. Although this literature has followed many blind
leads and has been distorted by many subtle (and some not-so-subtle)
biases, stereotypes, and prejudices, it nevertheless suggests that the public
and self-disparagement or aggrandizement and the differential opportuni-
ties which go with birth-ascribed status, low and high, result in characteris-
tic psychological problems and resort to characteristic psychological
mechanisms.56 Comparable literature on non-Western societies is scanty
but increasing.5’
The consequences of birth-ascribed stratification are self-fulfilling and
self-perpetuating, for although low status groups do not adopt views of
themselves or their statuses which are consistent with the views held by
their superiors, they are continually acting them out and cannot avoid
internalizing some of them and the self-doubts they engender, just as high
status groups internalize their superiority and self-righteousness. The
oppression of others by the latter serves to justify and bolster their super-
iority complex and to rationalize for them the deprivation and exploitation
of those they denigrate. ’Once you denigrate someone in that way,’ say
Kardiner and Ovesey, ’the sense of guilt makes it imperative to degrade
the subject further to justify the whole procedure. ’58 Gallagher notes
that in the southern United States,

By the attitudes of mingled fear, hostility, deprecation, discrimination, amused

patronage, friendly domination, and rigid authoritarianism, the white caste
generates opposite and complementary attitudes in the Negro caste. It is a touch
of consummate irony that the dominant group should then argue that the
characteristics which exhibit themselves in the submerged group are ’natural’
or ’racial’. 59

The products of oppression are thus used to justify oppression.

Change and Emancipation
The self-reinforcing degradation described above combines with greed
and fear of status-loss or revolt to comprise a dynamic of oppression which,
in birth-ascribed stratification, probably accounts for the widespread
occurrence of pariah status or untouchability. Elites characteristically
justify oppression by compounding it; they enhance their own rewards
by denying them ever more stringently to social inferiors, and they strive
 407

to protect themselves from challengesto status and privilege from below

by rigidifying the status boundaries, reinforcing the sanctions which
enforce them, and increasing the monopoly on power which makes the
sanctions effective. This assures increasing social separation and hier-
archical distance between groups until such time as it generates rebellion,
reform, or disintegration.
The fact that social order prevails most of the time in any given instance
of inherited inequality does not mean that all of those in the system accept
it or their places within it willingly, nor does it mean that the system is
either stable or static. It most often means that power is held and exercised
effectively by those in superordinate statuses, for the time being. Such
systems are based on conformity more than consensus, and are maintained
by sanctions more than agreement. Nevertheless, change is inherent,
resistance and mobility-striving are universal, and effective challenges to
such systems are probably ultimately inevitable because the response they
elicit from those they oppress is subversive. The possibility of acting out the
subversion depends largely upon the balance of power among the stratified
groups and the definitions of the situation their members hold. The pro-
cesses of change and patterns of conflict which lead to them are major areas
of commonality in such systems.so
The history of every caste system, of every racially stratified system, of
every instance of birth-ascribed oppression is a history of striving, conflict,
and occasional revolt. That this is not generally acknowledged is largely a
result of the fact that most of these actions occur in the context of over-
whelming power and uncompromising enforcement by the hereditary
elites and are therefore expressed in the form of day-to-day resentment and
resistance handled so subtly and occurring so routinely that it goes un-
remarked Even conspicuous manifestations are likely to be quickly
and brutally put down, confined to a particular locality or group, and
knowledge of their occurrence suppressed by those against whom they
have been directed. These phenomena often can only be discovered by
consulting and winning the confidence of members of oppressed groups,
and this is rarely done.
Only the most spectacular instances of resistance, and the few successful
ones are likely to be well-known. Immediately to mind come such martyrs
to the cause of emancipation of oppressed peoples as the Thracian slave
Spartacus, who led a rebellion against Rome; the American slave rebellion
leaders Gabriel and Nat Turner, the white abolitionist John Brown, and
the contemporary leaders of black emancipation in America, Martin Luther
King, Medgar Evers, and others (too many of them martyred) among their
fellow leaders and supporters, black and white. No doubt there are many
more, most of them unknown and unsung, in the history of all groups
whose members society condemns by birth to oppression. In the folk
history of every such group, and in the memory of every member, are
instances of courageous or foolhardy people who have challenged or
outwitted their oppressors, often at the cost of their own foreseeable and
inevitable destruction.
Better-known and better-documented than the individuals who led and
sometimes died for them, are the emancipation movements which have
occurred in most such societies-movements such as those for black power
and black separatism in the United States, anti-casteism and anti-untouch-
ability in India, Hutu emancipation in Rwanda and Burundi, Burakumin
408

emancipation in Japan, and anti-apartheid in South Africa. All have

depended primarily upon concerted efforts to apply political, economic,
or military power to achieve their ends. They have comprised direct
challenges to the systems. Most have followed after the failure of attempts
less militant, less likely to succeed, and hence less threatening to social
elites-attempts towards assimilation or mobility within the systems such
as those of status emulation.
The movements among black Americans (and more recently among
other ethnically stratified groups in America such as Chicanos and Native
Americans, and among women), are too well-known and well-documented
to require enumeration here.~2 Anyone who reads the American press and
especially the black and left press, cannot avoid being aware of the main
currents in this area.
Emancipation movements outside of the West are less known to Euro-
pean and American readers, but they are more numerous than can be
indicated here albeit quite poorly documented. Any list would have to
include the following obvious examples:
India: Escape from the consequences of caste stratification has been a
primary appeal of every religion and social reform movement to gain
adherents in India from the beginnings of Jainism and Buddhism in the
sixth century B.C., through Islam which became a significant religion in
India from the eleventh century A.D.; Sikhism from the early sixteenth
century; Christianity, dating from the ’Syrian Christians’ of the first
or sixth century A.D. in Kerala, followed in the sixteenth century by con-
verts of Portuguese and French Catholic missionaries and in the nineteenth
century by Anglican and Protestant converts; various Hindu reform move-
ments including the Brahmo Samaj and Arya Samaj of the early and late
nineteenth century, respectively; the post-Independence resurgence of
’neo-’Buddhism among low castes as an explicitly anti-caste phenomenon
under Dr. B. R. Ambedkar, the revered leader of Indian untouchables for
many decades and founder of the Scheduled Castes Federation and its
successor in the later 1950s, the All-India Republican Party.
Regional movements have been many, among the best-known of which
are the Maharashtrian anti-Brahmin movement under the leadership of
Jotirao Phule in the 1870s, a largely middle-caste Maratha movement, and
in the same area an emancipation movement among untouchable Mahars
which matured into the above-mentioned Scheduled Caste Federation
under Ambedkar.s3
Innumerable local and regional movements, primarily in urban centres,
have arisen whose aim has been to emancipate the members of a caste or
cluster of castes, to force public recognition of higher status for a caste,
or to ameliorate their caste-based disabilities through solidarity and

political action .64 There can be little doubt that contemporary revolution-
ary parties and movements in India depend to a significant extent on their
appeal to oppressed castes whose members see them as vehicles to emanci-
pation.
In addition to these sophisticated efforts at emancipation, upward
mobility movements among low castes are endemic to all regions of India;
and have evidently been so throughout its history. These movements entail
a claim to high status that has not been recognized. They co-ordinate and
enforce among the members of a caste emulation of the behaviours and
attributes of high castes in the hope that this will result in public recognition
 409
of the claim. This process has been termed ’Sanskritization’, in recognition
of the fact that the behaviours adopted are often those prescribed for high
castes in the sacred Sanskrit literature. 65 Emulation alone, however, no
matter how successfully done, is not enough to confer status, for status
in a caste system is not based on behaviour but on birth. To change the
societal definition of a caste’s status requires a concerted, sustained, and
powerful group effort, and it is most often unsuccessful.
In addition, as Harold Isaacs has documented in India’s Ex-Untouchables,
individuals occasionally escape the consequences of their caste status into
rewarding caste-free occupations in cities, often as a result of education and
sometimes by the successful concealment of their caste identities (passing).
In rural areas, some people emigrate to cities to escape the disabilities of
their caste status; some cluster together in low-caste communities to
avoid daily contact with, and humiliation by, their caste superiors; a
few are able to acquire a degree of exemption at the cost of conventional
family life through adoption of non-priestly religious roles or resort to
various socially deviant identities. 66
Japan : Publicly recognized emancipation movements among the Bura-
kumin of Japan have been many, militant, and frequently violent following
the official Edict of Emancipation of 1871. They have been so well docu-
mented and conveniently summarized by Totten and Wagatsuma that to
repeat the information would be superfluous. Wagatsuma has carried the
chronology of political militance through the post-World War II period,
and has also documented non-political, religious, and educational ap-
proaches to amelioration of Burakumin oppression.67 The similarities
to emancipation efforts in India and the United States are little short of
uncanny. The occurrence and problems of ’passing’ among Burakumin,
and the limited rewards to be acquired through educational, occupational,
and residential mobility are closely parallel to those reported for India
and the United States. 68
Ruanda: In Ruanda and nearby Urundi, the dominant Tutsi seemed until
fifteen years ago to be in firm control, with the subordinate Hutu and
Twa relegated to dependent economic and political roles .6 But this proves
to have been a false calm. In 1957, while Ruanda was still part of the
Belgian trust territory of Ruanda-Urundi, the Hutu issued the Bahutu
Manifesto which initiated an emancipation movement. Opposing political
parties then arose, advocating Hutu emancipation on the one hand and
Tutsi supremacy on the other. The brutality of the latter unleashed a
successful Hutu revolution in 1959, in which most of the Tutsi were driven
from the country. The resultant Hutu government was confirmed when the
emancipation party won overwhelmingly in a plebiscite in 1961. That
party was in power when Ruanda became independent Rwanda in 1962,
and has remained so despite frequent incursions by Tutsi from across the
Rwanda borders. In the adjacent kingdom of Burundi (formerly Urundi
and part of the same trust territory), the Hutu emancipation movement
was suppressed by the Tutsi-dominated government after independence
in 1962, and inter-caste tension thereafter increased. When the Hutu
won a majority of governmental seats in the election of 1965, the king
refused to name the majority government. This lead to an unsuccessful
Hutu coup, put down by the army, after which most of the Hutu leadership
was shot. These Hutu emancipation efforts were perhaps a surprise to the
Tutsi overlords, but they came as no surprise to any serious student of
 410
birth-ascribed oppression, for such systems are always fraught with tension
and resentment which await only a belief in the possibility of success for
drastic change to be attempted from below.
Henry Adams characterized the slave society of Virginia in 1800 as
’ill at ease’.10 This seems to be the chronic state of societies so organized-
the privileged cannot relax their vigilance against the rebellious resentment
of the deprived. That such rigid, oppressive systems do function and
persist is a credit not to the consensus they engender any more than to the
justice or rationality of the systems. Rather, it is a tribute to the effective-
ness of the monopoly on power which the privileged are able to maintain.
When in such systems deprived people get the vote, get jobs, get money,
get legal redress, get guns, get powerful allies, get public support for their
aspirations, they perceive a change in the power situation and an enhance-
ment of the likelihood of successful change in their situation, and they are
likely to attempt to break out of their oppressed status. These conditions
do not generate the desire for change, for that is intrinsic; they merely
make it seem worthwhile to attempt the change. Sometimes the triggering
factor is not that the deprived believe conditions have changed so that
success is more likely, but rather that conditions have led them to define
the risk and consequences of failure (even its virtual certainty) as accept-
able. Resultant changes are often drastic and traumatic of achievement,
but they are sought by the oppressed and by enlightened people of all
statuses precisely because of the heavy individual and societal costs of
maintaining inherited inequality and because its inherent inhumanity.
An important difference between the dynamics of inherited stratification
and acquired stratification results from the fact that in the latter, power and
privilege accompany achievable status, emulation is at least potentially
effective, and mobility and assimilation are realistic goals. Therefore
energies of status resentment may rationally be channelled toward mobility.
Most immigrant groups in the United States, for example, have found this
out as they have merged with the larger society after one or two genera-
tions of socialization. But in a system where inherited, unalterable group
identity is the basis for rewards, emulation alone cannot achieve upward
mobility, and assimilation is impossible so long as the system exists (in
fact, prevention of assimilation is one of its main functions). Only efforts
to destroy, alter, or circumvent the system make sense. In the United
States, blacks, Chicanos, and Native Americans have found this out.
Only in response to changes in the distribution of power is such inherited
status likely to be re-evaluated and the distribution of rewards altered.

’
’ ’
’ ’

:~ CONCLUSION .

’Race’ asthe term is used in America, Europe, and South Africa, is not
qualitatively different in its implications for human social life from caste,
varna, or jati as applied in India, quom in Swat and Muslim India, the
’invisible race’ of Japan, the ethnic stratification of Rwanda and Burundi.
Racism and casteism are indistinguishable in the annals of man’s inhuman-
: ity to man, and sexism is closely allied to them as man’s inhumanity to
woman. All are invidious distinctions imposed unalterably at birth upon
whole categories of people to justify the unequal social distribution of
power, livelihood, security, privilege, esteem, freedom-in short, life
chances. Where distinctions of this type are employed, they affect people
 411
and the events which people generate in surprisingly similar ways despite
the different historical and cultural conditions in which they occur.
If I were asked, ’What practical inference, if any, is to be drawn from the
comparative study of inherited inequality-of ascriptive social ranking?’
I would say it is this: There is no way to reform such institutions; the
only solution is their dissolution. As Kardiner and Ovesey said long ago
’there is only one way that the products of oppression can be dissolved, and
that is to stop the oppression’. 71 To stop the oppression, one must eliminate
the structure of inherited stratification upon which it rests. Generations of
Burakumin, Hutu, blacks, untouchables, and their sympathizers have
tried reform without notable success. Effective change has come only when
the systems have been challenged directly.
The boiling discontent of birth-ascribed deprivation cannot be contained
by pressing down the lid of oppression or by introducing token flexibility,
or by preaching brotherly love. The only hope lies in restructuring society
and redistributing its rewards so as to end the inequality. Such behavioural
change must come first. From it may follow attitudinal changes as
meaningful, status-equal interaction undermines racist, casteist, communal-
ist, and sexist beliefs and attitudes, but oppressed people everywhere have
made it clear that it is the end of oppression, not brotherly love, which
they seek most urgently. To await the latter before achieving the former is
futility; to achieve the former first does not guarantee achievement of the
latter, but it increases the chances and makes life liveable. In any case,
the unranked pluralism which many minorities seek requires only equality,
not love.
To those who fear this course on the grounds that it will be traumatic
and dangerous, I would say that it is less so than the futile attempt to
prevent change. Philip Mason spoke for all systems of inborn inequality
when he called the Spartan oppression of the Helots in ancient Greece a
trap from which there was no escape.
It was the Helots who released the Spartans from such ignoble occupations as
trade and agriculture.... But it was the Helots who made it necessary to live
in an armed camp, constantly on the alert against revolt.... They had a wolf
by the ears; they dared not let go. And it was of their own making; they had
decided-at some stage and by what process one can only guess-that the
Helots would remain separate and without rights forever. 72

That way, I believe, lies ultimate disaster for any society. A thread of hope
lies in the possibility that people can learn from comparison of the realities
of inherited inequality across space, time, and culture, and can act to
preclude the disaster that has befallen others by eliminating the system
which guarantees it. It is a very thin thread.

References
1
Max Weber, From Max Weber: Essays in Sociology, H. H. Gerth and C. W. Mills
trans. and ed. (New York, Oxford University Press, 1946); W. G. Runciman, ’Class,
Status and Power?’, in Social Stratification, J. A. Jackson, ed. (London, Cambridge
University
2
Press, 1968), pp. 25-61.
See for Ruanda: Jacques J. Maquet, The Premise of Inequality in Ruanda (London,
Oxford University Press, 1961); for India: F. G. Bailey, ’Closed Social Stratification
in India’, European Journal of Sociology (Vol. IV, 1963); Gerald D. Berreman, ’Caste :
The Concept’, in International Encyclopedia of the Social Sciences, D. Sills, ed. (New
York, Macmillan and The Free Press, 1968), Vol. II, pp. 333-9; André Béteille, Castes
412
Old and New (Bombay, Asia Publishing House, 1969); Louis Dumont, Homo Hier-
archicus (London, Weidenfeld and Nicolson, 1970); J. H. Hutton, Caste in India, Its
Nature, Functions and Origins (London, Cambridge University Press, 1946); Adrian
C. Mayer, ’Caste: The Indian Caste System’, in D. Sills, ed., op. cit., pp. 339-44;
M. N. Srinivas, Caste in Modern India and Other Essays (Bombay, Asia Publishing
House, 1962), and Social Change in Modern India (Berkeley, University of California
Press, 1966); for Swat: Fredrik Barth, ’The System of Social Stratification in Swat,
North Pakistan’, in Aspects of Caste in South India, Ceylon and North-West Pakistan,
E. Leach, ed. (London, Cambridge University Press, 1960), pp. 113-48; for Japan:
George DeVos and Hiroshi Wagatsuma, eds., Japan’s Invisible Race: Caste in Culture
and Personality (Berkeley, University of California Press, 1966); Shigeaki Ninomiya,
’An Inquiry Concerning the Origin, Development and Present Situation of the Eta
in Relation to the History of Social Classes in Japan’, The Transactions of the Asiatic
Society of Japan (Second series, Vol. 10, 1933); cf. Herbert Passin, ’Untouchability
in the Far East’, Monumenta Nipponica (Vol. 2, No. 3, 1955); for the United States:
Allison Davis, B. B. Gardner, and M. R. Gardner, Deep South: A Social Anthropological
Study of Caste and Class (Chicago, The University of Chicago Press, 1941); John
Dollard, Caste and Class in a Southern Town (Garden City, New York, Doubleday,
1957); Gunnar Myrdal, An American Dilemma: The Negro Problem in Modern Democ-
racy (New York, Harper, 1944); Alphonso Pinkney, Black Americans (Englewood
Cliffs, New Jersey, Prentice-Hall, 1969); Peter I. Rose, ed., Americans from Africa,
Vol. I: Slavery and its Aftermath and Vol. II: Old Memories, New Moods (New York,
Atherton Press, 1970). See also contrasts with South Africa: Pierre van den Berghe,
South Africa, a Study in Conflict (Berkeley, University of California Press, 1967); Latin
America: Marvin Harris, Patterns of Race in the Americas (New York, Walker, 1964);
Julian Pitt-Rivers, ’Race, Color and Class in Central America and the Andes’, Daedalus
(Spring, 1967); the Caribbean: M. G. Smith, The Plural Society in the British West
Indies (Berkeley, University of California Press, 1965). G. D. Berreman, Caste in the
Modern World (New York, General Learning Press, forthcoming).
3
4
Dumont, op. cit.
Gerald D. Berreman, ’A Brahmanical View of Caste: Louis Dumont’s Homo
Hierarchicus’, Contributions to Indian Sociology (New Series, No. V, 1972).
5
Tamotsu Shibutani and Kian M. Kwan, Ethnic Stratification: A Comparative
Approach
6
(New York, Macmillan, 1965), p. 572.
H. S. Morris, ’Ethnic Groups’, in D. Sills, ed., op. cit., Vol. 5, p. 167.
7
Pierre van den Berghe, ’The Benign Quota: Panacea or Pandora’s Box’, The Ameri-
can Sociologist (Vol. 6, Supplementary Issue, June 1971).
8Ibid., p. 43.
9 Cf. Robert Blauner, ’Black Culture: Myth or Reality?’, in Rose, Old Memories,
New Moods, pp. 417-43.
10
Gerald D. Berreman, ’Caste in India and the United States’, The American Journal
of Sociology (Vol. LXVI, September, 1960); cf. Berreman, ’A Brahmanical View of
Caste
11
...’; op. cit.
Oliver C. Cox, ’Race and Caste: A Distinction’, The American Journal of Sociology
(Vol. L, March, 1945); cf. Oliver C. Cox, Caste, Class and Race (Garden City, New
York, Doubleday, 1948).
12
Arthur R. Jensen, ’How Much Can We Boost I.Q. and Scholastic Achievement ?’,
Harvard Educational Review (Vol. 39, No. 1, Winter, 1969).
13
See the various articles comprising the ’Discussion’, of Jensen’s article in Harvard
Educational Review (Vol. 39, No. 2, Spring, 1969).
14
E. A. Ross, Social Psychology (New York, Macmillan, 1914), p. 3.
15
Shibutani and Kwan, op. cit., p. 37.
16Ibid., p. 110.
17 Pierre van den
18
Berghe, Race and Racism (New York, Wiley, 1967), p. 21.
Claude Lévi-Strauss, ’The Bear and the Barber’, Journal of the Royal Anthropological
Institute (Vol. 93, Part 1, 1963).
19
Winthrop D. Jordan, White Over Black (Baltimore, Penguin Books, 1969), p. 137-8.
20
Cox, Caste, Class and Race, p . 433, cf. Berreman, ’Caste in India and the United
States’, op. cit.; Berreman, ’A Brahmanical View of Caste ...’; op. cit.; Gerald D.
Berreman, ’Caste, Racism and "Stratification" ’, Contributions to Indian Sociology
(No. VI, December 1962); Martin Orans, ’Caste and Race Conflict in Cross-Cultural
Perspective’, in Race, Change and Urban Society, P. Orleans and W. R. Ellis, eds.,
comprising Urban Affairs Annual Reviews (Vol. 5, 1971).
21
Dumont, op. cit., Cox, Caste, Class and Race.
22
Berreman, ’Caste in India and the United States’, op. cit., Gerald D. Berreman,
’Caste in Cross-Cultural Perspective: Organizational Components’ comprising Chapter
14, ’Structure and Function of Caste Systems’, and Chapter 15, ’Concomitants of Caste
Organization’, in Japan’s Invisible Race, DeVos and Wagatsuma, eds., op. cit.; Gerald
D. Berreman, ’Stratification, Pluralism and Interaction: A Comparative Analysis
of Caste’, in Caste and Race: Comparative Approaches, A. de Reuck and J. Knight, eds.,
(London, J. and A. Churchill, 1967); Orans, op. cit.
 413
23
24
Cf., Berreman, ’Caste in Cross-Cultural Perspective ...’ op. cit., pp. 279-81.
F. G. Bailey, ’Closed Social Stratification in India’, op. cit., p. 113.
25
Cf., George DeVos and Hiroshi Wagatsuma, ’Group Solidarity and Individual
Mobility’ inJapan’s Invisible Race, pp. 245-8; Harold R. Isaacs, India’s Ex- Untouchables
(New York, John Day Company, 1965), pp. 143-9 et passim.
26
Gerald D. Berreman, Hindus of the Himalayas: Ethnography and Change (Berkeley,
University of California Press, 1972), pp. 200 ff.
27
Robert Blauner, ’Internal Colonialism and Ghetto Revolt’, Social Problems (Vol.
16, No. 4, Spring, 1969); Robert Blauner, Racial Oppression in America (New York,
Harper and Row, 1972).
28
Cf., Blauner, ’Internal Colonialism ...’, op. cit.; Stokely Carmichael and Charles
Hamilton, Black Power (New York, Random House, 1967); Frantz Fanon, The
Wretched of the Earth (New York, Grove Press, 1966); O. Mannoni, Prospero and
Caliban: The Psychology of Colonization (New York, Praeger, 1956); Albert Memmi,
The Colonizer and the Colonized (Boston, Beacon Press, 1967).
29
Cf. Blauner, ’Internal Colonialism ...’ p. 395-6.
30
Gerald D. Berreman, ’Caste as Social Process’, Southwestern Journal of Anthro-
pology (Vol. 23, No. 4, Winter, 1967); Blauner, Racial Oppression in America; S. F.
Nadel, ’Caste and Government in Primitive Society’, Journal of the Anthropological
Society of Bombay (Vol. 8, 1954); J. S. Furnivall, Colonial Policy and Practice: A
Comparative Study of Burma and Netherlands India (London, Cambridge University
Press, 1948); M. G. Smith, The Plural Society in the British West Indies (Berkeley,
University of California Press, 1965); James B. Watson, ’Caste as a Form of Accultura-
tion’, Southwestern Journal of Anthropology (Vol. 19, No. 4, Winter 1963).
31
Cf. Irawati Karve, Hindu Society: An Interpretation (Poona, Deccan College
Postgraduate and Research Institute, 1961).
32
Davis, Gardner, and Gardner, op. cit.; St. Clair Drake and Horace R. Cayton,
Black Metropolis (New York, Harcourt, Brace, 1945); Dollard, op. cit.; Marina
Wikramanayake, ’Caste and Class among Free Afro-Americans in Ante-bellum South
Carolina’, paper delivered before the 70th Annual Meeting of the American Anthro-
pological Association (New York, November 1971).
33
Furnivall, op. cit.; cf. Malcolm Cross, ed., Special Issue on Race and Pluralism,
Race (Vol. XII, No. 4, April 1971).
34
M. G. Smith, op. cit., pp. 14, 83.
35
Berreman ’Stratification, Pluralism and Interaction ...’, op. cit., p. 51.
36
M. G. Smith, op. cit., p. xi.
37
Berreman, ’Stratification, Pluralism and Interaction ...’, op. cit., p. 55.
38 Cf.
39
Fumivall, op. cit.
40
Berreman, ’Stratification, Pluralism and Interaction ...’, op. cit., p. 55.
Ibid., cf. McKim Marriott, ’Interactional and Attributional Theories of Caste
Ranking’, Man in India (Vol. 39, 1959).
41
Maya Angelou, I Know Why the Caged Birds Sings (New York, Bantam Books,
1971),
42
p. 168.
Cf. Berreman, ’Caste in India and the United States’, op. cit.
43
Berreman, ’Stratification, Pluralism and Interaction ...’, op. cit., p. 51.
44
Dumont, op. cit.
45
Cf. Berreman, ’Caste in India and the United States’, op. cit.; Berreman, ’Caste
in Cross-Cultural Perspective ...’, op. cit.; Gerald D. Berreman, ’Social Categories
and Social Interaction in Urban India’, American Anthropologist (Vol. 74, No. 3).
46
Cf. Berreman, ’A Brahmanical View of Caste ...’, op. cit.
47
Kate Millett, Sexual Politics (New York, Avon Books, 1971), p. 19.
48
Casey Hayden and Mary King, ’Sex and Caste’, Liberation (April, 1966), p. 35;
cf. Millett, op. cit.
49
Fran Hawthorne, ’Female Roles Examined by Rutgers Professors’, Daily Cali-
fornian (Berkeley, 6 October, 1971), p. 5. See also Millett, op. cit., p. 57, for a summary
of the common psychological traits and adaptational mechanisms attributed to blacks
and women in American society as reported in three recent sociological accounts.
50
Cf., G. Morris Carstairs, The Twice-Born (Bloomington, Indiana University
Press, 1958), pp. 59-62 et passim; Morris E. Opler, ’The Hijara (Hermaphrodites) of
India and Indian National Character: A Rejoinder’, American Anthropologist
(Vol. 62, No. 3, June, 1960).
51
Gerald D. Berreman, ’Women’s Roles and Politics: India and the United States’,
in Readings in General Sociology, R. W. O’Brien, C. C. Schrag, and W. T. Martin, eds.,
(4th Edition, Boston, Houghton Mifflin Co., 1969). First published, 1966. Cf. Millett,
op. cit.
52
van den Berghe, Race and Racism, op. cit., p. 22.
53
Berreman, ’Stratification, Pluralism and Interaction ...’, op. cit., p. 58.
54 Abram Kardiner and Lionel
Ovesey, Mark of Oppression (Cleveland, The World
Publishing Co., 1962), p. 387.
55
Robert Coles, Children of Crisis (Boston, Atlantic-Little, Brown, 1964); with Jon
Erikson, The Middle Americans (Boston, Little, Brown, 1971).
414
56
Berreman, ’Stratification, Pluralism and Interaction ...’, op. cit., Cf. Thomas
F. Pettigrew, A Profile of the Negro American (Princeton, D. Van Nostrand, 1964);
Stanley M. Elkins, Slavery (Chicago, University of Chicago Press, 1959); Mina Davis
Caulfield, ’Slavery and the Origins of Black Culture: Elkins Revisited’, in Americans
From Africa: Slavery and Its Aftermath, P. I. Rose, ed., J. H. Rohrer and M. S.
Edmonson, The Eighth Generation Grows Up (New York, Harper and Row, 1960).
57
Santokh S. Anant, ’Child Training and Caste Personality: The Need for Further
Research’, Race (Vol. VIII, No. 4, 1967); Santokh S. Anant, Inter-Caste Attitudes
(provisional title), (Delhi, Vikas Publications, 1972); K. K. Singh, Patterns of Caste
Tension (Bombay, Asia Publishing House, 1967); DeVos and Wagatsuma, op. cit.
58
Kardiner and Ovesey, op. cit., p. 379.
59
B. G. Gallagher, American Caste and the Negro College (New York, Columbia
University
60
Press, 1938), p. 109.
61
Berreman, ’Caste as Social Process’, op. cit.
Raymond Bauer and Alice Bauer, ’Day to Day Resistance to Slavery’, Journal of
Negro History (Vol. 27, October 1942); Douglas Scott, ’The Negro and the Enlisted
Man: An Analogy’, Harpers (October 1962), pp. 20-1; cf. Berreman, ’Caste in India
and the United States’, op. cit.
62
Cf. Carmichael and Hamilton, op. cit.; Eldridge Cleaver, Soul on Ice (New York,
Dell Publishing Co., 1968); E. U. Essien-Udom, Black Nationalism (New York, Dell
Publishing Co., 1962); Lewis M. Killian, The Impossible Revolution? Black Power and
The American Dream (New York, Random House, 1968); Martin Luther King, Stride
Toward Freedom (New York, Harper Brothers, 1958); C. Eric Lincoln, The Black Mus-
lims in America (Boston, Beacon Press, 1962); Louis E. Lomax, The Negro Revolt
(New York, New American Library, 1962); Raymond J. Murphy and Howard Elinson,
Problems and Prospects of the Negro Movement (Belmont, California, Wadsworth, 1966);
Pinkney, op. cit.; Rose, op. cit.; Charles E. Silberman, Crisis in Black and White (New
York, Random House, 1964); Stan Steiner, The New Indians (New York, Harper and
Row, 1968); Stan Steiner, La Raza: The Mexican Americans (New York, Harper and
Row, 1970); Howard Zinn, S.N.C.C., The New Abolitionists (Boston, Beacon Press,
1964).
63
Charles H. Heimsath, Indian Nationalism and Hindu Social Reform (Princeton,
Princeton University Press, 1964); Owen M. Lynch, The Politics of Untouchability
(New York, Columbia University Press, 1969); Gail Omvedt, ’Jotirao Phule and the
Ideology of Social Revolution in India’ (Dept. of Sociology, University of California,
Berkeley, 1971), mimeographed paper; cf. B. R. Ambedkar, The Untouchables: Who
were They and why They became Untouchables (New Delhi, Amrit Book Co., 1948);
B. R. Ambedkar, What Congress and Gandhi have done to the Untouchables (Bombay,
Thacker and Co., 1946); B. R. Ambedkar, Annihilation of Caste (Bombay, Bharat
Bhusan Press, 1945).
64
Orans, op. cit.; Martin Orans, The Santal: A Tribe in Search of a Great Tradition
(Detroit, Wayne State University Press, 1965); Lloyd I. Rudolph and Susan H. Rudolph,
’The Political Role of India’s Caste Associations’, Pacific Affairs (Vol. 33, 1960); James
Silverberg, ed., Social Mobility in the Caste System in India, Supplement III: Compara-
tive Studies in Society and History (The Hague, Mouton, 1968).
65
M. N. Srinivas, Caste in Modern India and Other Essays (Bombay, Asia Publishing
House, 1962), pp. 42-62; M. N. Srinivas, Social Change in Modern India (Berkeley,
University of California Press, 1966), pp. 1-45.
66
Cf. Berreman, Hindus of the Himalayas, Epilogue, ’Sirkanda Ten Years Later’.
67
George O. Totten and Hiroshi Wagatsuma, ’Emancipation: Growth and Trans-
formation of a Political Movement’, in DeVos and Wagatsuma, op cit., pp. 33-67;
Hiroshi Wagatsuma, ’Postwar Political Militance’, and ’Non-Political Approaches:
The Influences of Religion and Education’, in DeVos and Wagatsuma, op. cit., pp.
68-109.
68
George DeVos and Hiroshi Wagatsuma, ’Group Solidarity and Individual
Mobility’, in DeVos and Wagatsuma, op. cit., pp. 245-56.
69
70
Maquet, op. cit.
Henry Adams, The United States in 1800 (Ithaca, New York, Cornell University
Press,
71
1961), p. 98.
Kardiner and Ovesey, op. cit., p. 387.
72
Philip Mason, Patterns of Dominance (London, Oxford University Press for the
Institute of Race Relations, 1970), p. 75.