INDIA AS A PHILOSOPHICAL PROBLEM: MCKIM MARRIOTT AND THE COMPARATIVE ENTERPRISE.
Marriott's work, little known outside the anthropological bailiwick, has implications for anyone interested in the comparison of cultures--and in such current hot-button topics as whether one can or should even try to "read" the literature of another culture from within the confines of one's own. Marriott's work, still in formation, is notable for the rigor with which it attempts to state the principles involved in apprehending an "alien" culture, and deserves a wider audience, both among Indianists and philosophers.
Publication of this symposium of articles by Marriott and some of his students provides an occasion to review what has been tried so far--especially inasmuch as the lead article, by Marriott himself, summarizes, in the most general form yet attempted, the principles his work presumes. Much correspondence has taken place between Marriott and the author, in which Marriott has clarified his formulations in yet more comprehensive terms. It is to be hoped that this review-article, even as tentative as it is, and that correspondence will encourage further publication in this vein.
MCKIM MARRIOTT'S RECENT EFFORTS to develop an anthropology sensitive to local modes of understanding societies (in this case, Indian self-understandings) are consonant with intellectual tendencies prominent in the last third of the twentieth century--among Western academics, at least: sensitivity to the plurality of cultures; suspicion of monotonic ideologies thought to reflect "hegemonies"; and attention to the claims of the excluded "other" to speak in its own voice. Such matters have, of course, been debated widely, both by Indologists and those less gifted--the nature and function of karma and other seemingly "indigenous" Indian categories; the peculiarities of Indian social arrangements, notably caste;2 the unique Indian contribution (if any) to contemporary world culture, etc., usually framed in terms of a "spiritual" component.  In Marriott's case, these tendencies have been molded also in the peculiar self-reflective crucible of the University of Chicago, which has seen itself as a pioneer in deve loping contrastive approaches, both in teaching and research, to the study of the three major non-Western civilizations, approaches whose distinctive character lies in their combining traditional (or "orientalist") classicism with hands-on inquiry into current social and political realities. Marriott's proposals have been reviewed sparsely, almost entirely by other anthropologists, and, apart from his own circle, not all that favorably.  Given the evident contribution of classicists like J. A. B. van Buitenen and A. K. Ramanujan to Marriott's thought, and his effort to engage the thought of the other important Indianist who has sought to investigate the gap between "modern" social theory and indigenous or classical categories--Louis Dumont--it is somewhat surprising that Marriott's project has not been commented on from an "orientalist" perspective. The little that has been written in this vein can hardly be considered serious. A distinguished Indian colleague, for example, has chastened Marriott for allow ing into his discourse Sanskrit adjectives incorrectly formed according to the rules of Panini's grammar (read, "tamasa" not "tamasika").  It should be stressed at the outset that Marriott's project, which is still very much in process, is a serious effort to think through the problems of cross-cultural comparison in a rigorously responsible and self-reflective fashion. Whatever its problems--and there are many--Marriott's work stands apart (indeed, in the august company of Louis Dumont's) as a contribution to Western philosophical literature on the morphology of civilizations--a distinguished tradition of inquiry that goes back at least to Vico and Hegel (if not to Herodotus).
1. Marriott seems at times to shift his ground as his critics weigh in against this or that aspect of his stilldeveloping project: the absence of a comprehensive statement of his theory is a major drawback, it must be admitted, and its often fragmentary presentation invites caviling. It was his lengthy review of Dumont's Homo Hierarchicus  that seemed to bring into focus the need for an "ethnosociology" of India more adequate than Dumont's "flawed" theory--a need then given positive voice in an article Marriott co-authored with Ronald laden, "Toward an Ethnosociology of South Asian Caste Systems."  The review, which in sum was quite critical (in both senses) of Dumont's method and conclusions, has, of course, made it difficult subsequently to dissociate their two names--though Marriott tends to belittle Dumont's contribution to his own thought, and denies that there has been any real "dialectic" between them. Still, for our purposes, which, I repeat, are Indological rather than social-scientific, this controversy does still serve as a very useful entree into Marriott's views. There seem to be two levels of dissatisfaction expressed in his long review of Dumont's work, and subsequently: speaking empirically, Dumont's top-down, "ideological" analysis, while often coherent with (and indeed, based on) the textual tradition (therefore demonstrating a unified civilization--but the demonstration is spurious), cannot confront or account for the observed diversity of behavior at the micro-level that painstaking fieldwork discovers--the strictly personal level, often inconsistent from village to village; further, a theory based on "ideology" alone cannot, according to Marriott, discover an "Indian" sociology that starts from "Indian" categories-that is, categories Indians, in their concrete variety, recognize and use in realizing their own culture; at best, it will capture only an aspect of the complex--either the self-serving idealizations of a dominant group that have found their way into a "literature," itself, o f course, a province of the dominant; or, what is worse, it will simply reveal categories proper to the researcher's own "ideology" naively taken for granted.
Marriott's effort to transcend what he regards as the overly simple and "dichotomous" work of Dumont shows him to be more interested, as were the great philosophers of history mentioned, in a general theory of cultural diversity that will both be universal--capable of apprehending a variety of cultures globally, as variations on a theme that may be stated formally; and essential--capable of ascertaining the differentia (or differentiae) proper to or characteristic of a given culture, taken singly. By contrast, Marriott's multivalent analysis presents itself as capable of mapping almost infinite varieties of cultural expression-- an "interactive" model that Marriott has attempted to visualize through his "cube," which is defended as a source of questions and hypotheses relevant to the social scientist's investigations--a philosopher's stone, so to speak, providing "topics" of inquiry--rather than as an icon of some sort having intrinsic value.
Constructing a theoretical social science for a culture requires somewhat more than providing a meaningful cultural account: it requires building from the culture's natural categories a general system of concepts that can be formally defined in relation to each other; it requires developing words and measures that can be used rigorously for description analysis and explanation within that culture; and it especially requires developing deductive strategies that can generate hypotheses for empirical tests in terms that will be analytically powerful enough to define all the major parameters of living in that culture without violating the culture's ontology, its presuppositions, or its epistemology. (p. 4)
2. Let it be said at the outset that Marriott, like Dumont, is a comparativist in the very best (which is also a "Western") sense of the word. Impressionistic oppositions, whether the East be "spiritual"--or just inscrutable, are merely anecdotal and express little but the unreflective prejudices of the speaker. If comparison is possible, judgments must be grounded in explicit and universally intelligible criteria, open to anyone who considers the matter with an open mind. How Marriott reconciles this universalizing agenda with his announced design of discovering "Hindu categories" of thought is a major concern of the present essay.
Marriott, on the other hand, is not to be confused with the "culture theorists" of recent vintage, for whom the study of cultural difference is essentially a pretext to criticism of any and all social arrangements that sit ill with the usually liberal and "post-modern" presuppositions of the theorist; these arrangements, of course, are nearly always seen as the result of hostile or alien "domination"-- "colonialists" are still conveniently available in this role of "other"--if not, then any privileged subgroup will do, so long as it has succeeded in imposing its interests, by definition parochial, on the weak of whatever "community" is at issue. According to this late-Marxian critique, "privilege" is nearly always nefarious, and its "deconstruction" (unmasking) is thought to be the only serious business of the cultural theorist, whose work thus becomes a species of rhetoric. Historical scholarship is reduced to advocacy--with a leitmotif, always sotto voce, of decidedly contemporary concerns. Apart from this confusion of genre, such highmindedness often seems paradoxically to trivialize true "diversity," and is not inconsistent with the real leveling of cultures to their common economic denominator that is so obvious an aspect of "globalization." No "culture" may henceforth be admired that is not founded on the Western and very recent notion of abstract equality; such variety as remains becomes, in this benign form, merely decorative. The "other" must be "respected" and seen as oneself. Value judgments that tend to differentiate among men or their functions are, ipso facto, "hegemonic" and "discriminatory." One of the species of post-colonial history sees its role as elevating the "subaltern" to the dignity formerly denied him both by the great of the empire and their lackey historians. The disinterested study of cultures, to be the extent that they are taken as substantial or even decorative alternatives to one's own--"Orientalism" in its broadest sense--is thus also implicated in the agenda of domination.  This Manichean (if not Hobbesian) view, when confronted with the nearly inexhaustible variety of "others" in the range of human history known to us, is notable also for its reductive simplicity--if not its sanctimoniousness, for it resembles in many regards the myopic parochialism associated in times past with the well-intentioned missionary. The incipient and rather bland monoculture that emerges from such theorizing bears, on the one hand, curiously reflexive relationships to the ideals of late-twentieth century liberal academic culture, and suggests, on the other, a homology with what, in Indian terms, would be called the "Vedantic" view of religion: to the extent that they disagree, societies, like religions, are unreal; all religions, as all societies, subtend the same relation to the one higher truth.  It is, by contrast, the real and irreconcilable moral diversity of culture that constitutes the best ethical reason for immersing oneself in the study of cultures and their history, a study that can of ten serve as an antidote to overhasty pride in our own recently discovered rectitude. The notion of "devotion," implying essential gradation, comes to mind--a notion that is so clearly crucial to any understanding of traditional Indian social and religious practice, and contrasts so strikingly with the egalitarian self-hype that must, it seems, express itself in all varieties of our "post" culture, whether economic or academic (an egalitarianism, however, that is not inconsistent with the pretensions of the theorist to speak on behalf of all men and all cultures). Marriott's work serves, like Dumont's, as a "counter-cultural" beacon in this regard. But, by the same token, both can be easily dismissed as anachronistic and "essentialist"--perhaps Marriott more than Dumont, for the latter, at least, seems to recognize the inevitability of the Western "monoculture," if not its inherent legitimacy.
Though they share many perspectives, Dumont's analysis of Indian civilization is, for Marriott, altogether too oppositional: it not only recognizes that civilization's essential uniqueness, but formulates it as a categorical alternative to "our" civilization--as though the two exhausted all the kinds of civilization possible. Marriott's view, founded in part on statistical models that attempt to account realistically for observed variations in behavior, is that any such account must involve at least three, if not more, variables, operating simultaneously and independently. A civilization is rather the product of infinitely complex and personal "transactions" that are variously motivated and codified, within various frames of reference, some of which are universal at the biological, some at the human, some at the social, and some (even) at the "civilizational," levels. Dumont's reductive "ideological" dichotomy of homo hierarchicus vs. homo aequalis both fails to exhaust the possibilities and imposes the rese archer's perspective on the subject. Still, Marriott begins his inquiry with the same basic question that animates Dumont's work: what is it that makes India a "distinctive" civilization, with its "own categories" of thought and its own unique way of characterizing the human experience?
It is an anomalous fact that the social sciences used in India today have developed from thought about Western rather than Indian cultural realities. As a result, although they pretend to universal applicability, the Western sciences often do not recognize and therefore cannot deal with the questions to which many Indian institutions are answers. (p. 1)
3. In this context, Marriott resorts to what has seemed to some a straw-man argument: he advances the notion that Dumont's treatment of India, because it proceeds essentially in terms of "dichotomous" concepts, is, ipso facto, locked into a "Westem"--because dualistic--model; therefore, Dumont's project of capturing the characteristic "India" is doomed to self-contradiction and failure. Indeed, his "structuralism," which owes much to Levi-Strauss, does lay stress on polarities, such as the obvious "purity-pollution." For Marriott, structuralism, it seems, is essentially "Western," in that it proceeds from such polarities: hot-cold, raw-cooked, nature-nurture, etc.
The Western ethoodisciplines themselves reflect the historic circumstances of their origins. The Reformation split 'spirit' from 'matter', 'value' from 'fact', and the 'humanities' (Geisteswissenschaften) from the 'natural sciences' (Naturwissensehaften); the Enlightenment further split 'mind' from 'body'... 'subjective' from 'objective'...; liberal capitalism split 'markets' from 'government' and 'economics' from political science'.... (p. 2, n. 4)
But it is hard to see why "dualistic" thought is more ethnically biased than Marriott's favored three-sided "cube." The West has, after all, not been uninfluenced by such thinkers as Hegel and Aristotle--hardly "dualists"; and India has monistic Advaita and pluralistic Nyaya beside the dualistic Samkhya, paradoxically so dear to Marriott. Both civilized traditions appear equally to have exhausted the possibilities of ontological assertion--from nihilism to pluralism. Is this a question that really divides civilizations?--or just sociologies of civilization? Marriott's claim is doubtless intended to be made from the angle of behavior: but do Indians indeed behave trichotomously? do we somehow behave dichotomously?
Uncharitably, several commentators have pointed out that Marriott cannot himself escape being labeled "Western," for, among other things, his project is quite securely lodged in the recent Western anthropological tradition--and, indeed, the very project of comparing cultures is demonstrably modern and Western. The real disagreement between Marriott and Dumont is met on a higher level, for each, in some sense, seeks to transcend the parameters of his intellectual culture by a process of reflecting on those parameters. Marriott's aims are even more ambitious than Dumont's: avoid a terminology that is culturally biased; discover a terminology native to the subject matter; thus invent a new "science" that is proper to the culture studied.
Thus, a positive impetus for this volume comes from the hope that more fully developed Indian ethnosocial sciences may take their place beside the Western ethnosocial sciences. Together with the ethnosciences of other lands they may provide better bases for the future claim of an expanded, multicultural set of sciences to have that 'universal significance and value' which Weber in 1904 [ref. given] prematurely reserved for rational social thought in the West. (p. 3)
The linguistic analogy here is patent--we might be speaking of "grammars" of civilizations. Levi-Strauss' structuralism, too, saw itself as discovering the fundamental functional categories that give pattern to the everyday life of the world's kaleidoscope of societies. These categories were thought to be reflected in the mental universe of the indigene, though no structuralist would have advanced the dubious proposition that they "emerged" from that universe or were "native" to it. They were, in fact, part of our universal mental baggage, but selected or particularized in this concrete universe--just as "language" is both a human universal, but is realized only in concrete variation. Marriott's approach seems to make the additional assumption that the categories themselves must also be discovered, and that, to be genuine, they must be "native" to the culture, and are, therefore, incommensurate with, untranslatable into, those of other cultures--are ambiguously relative, in a Wittgensteinian sense. "Science," at least "human science," no longer deals with or formulates analytical categories themselves universal, but is replaced by a multitude of "ethnosciences" that are essentially self-referential and present to each other only ambiguities. To press the linguistic analogy further, Marriott's argument with "structuralism" seems a somewhat anachronistic replay of American formalism's response--with its disinterest in questions of universal grammar--to the Chomskyan repositioning of linguistics in just such questions. For Marriott, the only "science" that can lay claim to universality is a tertiary demetaphorization of these ambiguous "ethnosciences" that is not imbricated in any "culture," and which no given culture will recognize as its own--not so much a "universal grammar' as a "mathematical logic." It resolves into formal patterns that resemble the terms of algebraic arguments, into which any content can be placed--just as medieval logic functions just as well with any proposition, whether or not "true," or ev en possible: "All unicorns are green." So, paradoxically, as we become more expert comparativists, our interest is refocused on this abstract "demetaphorical" level--and away from the "categories" that make Indian "science" unique.
Neither of these two ethnosociologies [American and Indian] makes sense of the other's society, but together both may contribute to a quaternary level of theory that does not yet exist--a systematic repertoire and synthesis of alternative, possible human social sciences. 
4. Marriott attempts also to marry Indology and sociology, to reconcile an anthropological and social-scientific method with matters typically debated only at the level of texts and in reflecting on texts, by refocusing our attention on the Indian "person," whose "nature" is constantly shifting, revealed in and modified through the "transactions" of social living. This "person" has little in common with the conventional Western person, whose nature is fixed both legally and ideologically by "inalienable" rights, thus giving substance to our notion of the "individual" as that out of which society emerges and whose ultimate interests it is the business of society to serve--just as questions of the "rights of man" now even begin to govern international relations. It is these "transactions" of the Indian person that Marriott seeks to map with his cube, wherein the "fluidity" of that person is represented in and through his relations, both human and natural, rather than as preexistent and absolute.
'Individuals' are indivisible, integrated, self-developing units, not normally subject to disjunction or reconstitution. Given such units, interpersonal influences, inequalities, and changes have to be brought in as external factors or pathologies. . . . The Hindu postulations. . . emphasize that persons are composite and divisible (what one might better call 'dividuals') and that interpersonal relations in the world are generally irregular and fluid, if not entirely chaotic. (p. 17)
We [Inden and I] did not think adequate Dumont's Durkheimian reading of Hindu theory to mean only that persons are socially ascribed to caste and varna boxes, and thus dedicated to the pursuit of just certain ends in life. We read varnas rather as transactional composites . . . used to characterize certain multifaceted roles of style of life, but not to contain all persons or fix their duties. We understood purusarthas as relevant to everyone. . . . We saw persons as "permeable, composite, partly divisible, and [their properties as] partly transmissible." 
On the other hand, Marriott does seem to accept Dumont's characterization of India as a culture that "ranks," rather than "levels." Village India, however, displays many different kinds of ranking behavior (that is, there are many competing kinds of "hierarchy," grounded on different principles, not just one, grounded on notions of "purity"); none, including the brahminical preoccupation with purity, can explain every kind of observed behavior, interpersonal or intercaste. How then do we then elicit the Indian categories in terms of which Indians think, and organize their social space?
5. In purely formal terms, the best way of complicating a view judged too simple because too dichotomous is to increase the number of independent variables in the ensemble. Starting with "analysis of variance" methodology and asking statistical questions such as "how many variables are needed to account for all the distinctions the natives are observed to make?" Marriott quickly passed through the stage of a second variable, and reached a more stable state of three variables, which better represent both Indian behavior on the ground and Indian views of behavior in the texts. 
. . . three appears to be the irreducible number of properties or components with which Hindus will comfortably think about human affairs. Thinking about constituted things in dualities is often condemned. At least three terms are always present, always combined. (p. 8)
Although he decries this overly simple formulation, it might be said that to Dumont's one hierarchical variable "purity-pollution," which he terms somewhat opaquely, "unmarking-marking," Marriott has added two others: "absorption-abstraction" and "disorder-order." "Purity-pollution" appears to him as one protocol by which Indians "mark" each other--as higher or lower in a scale of prestige, and which may not always be religiously based, as it was for Dumont. "Marking," in fact, is a concept borrowed from studies of animal behavior, and clearly implies subordination, as of a "territory"--rather than reflecting its perhaps more familiar linguistic use, with inverse implication.
Calling this process 'marking' is here meant to evoke the image of a substance (such as a sediment or a pigment) moving from a marker to a marked object. . . , as some property (tangible or intangible) is so often thought to do in Hindu interpersonal relations. . . . Hindu marking shares with linguistic marking the notion that unmarked (neutral). . . entities are more inclusive and thus taxonomically 'higher' than are the more marked and specialized entities included under. . . them. (p. 20)
But it seems counter-intuitive to deem the "unmarked" brahmin--who, in any case, takes such care to distinguish himself by markings of various sorts--as "including" the untouchable. Marriott does seem to agree with Dumont that it is characteristic of India that differences be stated rather than that similarities be stressed. "Equality," the similarity of all, though theoretically possible, is an anecdotal accident of the system, rather than its principle. Here both Marriott and Dumont (so acknowledged explicitly by Dumont) follow Hegel in considering the first principle of Indian social reality to be "differentiation," rather than the abstract "unity" that characterizes, e.g., Chinese society.  Castes, of course, give an explicit, and to us disconcerting, representation to this principle, whereby social differences are treated as facts of nature (i.e., one's "persona" is determined by the conditions of one's birth). Even "individuals" (the word is badly chosen--Marriott prefers "dividuals") are able to po sit themselves (it is said) only by "marking" those below them, or suffer being "marked by" those above them: they are differential arguments in a set of relations, rather than "entities" with "inalienable" status.
6. But to this principle, whose desacralization is made evident by calling it "marking," are adjoined two others. The second, a principle of absorption, is in a sense the corollary of "marking"--for if there are no absolutes in the system, then it is as false to assert absolute difference as it is to assert absolute identity. We are indeed different only to the extent that we also participate in the reality of those we differentiate ourselves from--a fact signaled in Indian daily life by the great concern that affects the sharing of objects, especially food, and every Indian's preoccupation with personal space, seen metaphorically as environing "substance" or extension of his "body"--which must be positioned "auspiciously" vis-a-vis the hours of the day and the constellation of the planets--and reaches into social notions, such as lineage, caste, and one's exposure to birth and death. By entering into relationships, "substances" acquire their character. This notion, which Marriott calls "mixing," appears to represent the natural basis to which ritual behavior responds, inasmuch as pollution (at one extreme) and purity (at the other) are, for individuals, conditioned by certain "mixings"--from bodily or animal refuse, at one extreme, through the caste-inflected transactions of daily life, to mediating fire and holy Ganga, at the other.
To say that 'mixing' (rather than unmixing) is a general property of the Hindu world is to assert the rarity of reflexivity, the improbability that any entity in that world can relate only to itself, even by a relationship of equality or identity. Mixing thus suggests the probability that any entity will be found non-self-sufficient, incompletely related to itself, not even equivalent to itself--being to a greater or lesser degree open and dependent for its qualities and processes upon exchanges with others. (p. 19)
A perfectly "unmixed" person is, in the Indian world, nearly self-contradictory; nevertheless that vector is represented, in extremis, by the Indian sadhu, the "renouncer" of all normal social contacts, one therefore "dead" to his former self--which means here, his "mixed" self. The relation between the sadhu and the untouchable is for Marriott one of opposites, which, like opposites generally, coincide in rejecting the social middle. The sadhu's exceptional "unmixing" derives from an act of will; the exceptional "mixing" of the untouchable makes of him the social problem par excellence. Though this overstates the actual social isolation of the sadhu, who is only rarely the lonely yogin in his Himalayan cave (and, as well, of the untouchable, who fulfills essential, though liminary, social functions), it is still the case that even those renouncers who maintain structured institutional relations with the rest of society (as gurus, heads of asramas, etc.) are characteristically (or even solely) regarded as "im partial" by their acolytes --unless, of course, they be regarded as frauds, which brings up issues not relevant to questions of principle. One index of the vast gulf between Indian and Western social reality is that, by this logic (which Dumont in another way also expresses), the "ideal" individual of the West--possessed of inalienable rights and existing absolutely so that he may (according to Rousseau) by a mere act of will create a society--is, in the Indian world, not only rare, but when identified, considered incapable of any constructive social relation. 
The third variable, "(un)matching," is temporal, rather than spatial, as are the first two. When time is added to the mix, a principle of disorder, of entropy (a view fully in accord with Indian notions of time) serves to disintegrate the "markings" and the "mixings" that, themselves, condition stable social life.
'Unmatching' (which could alternatively be called 'messing' or 'nixing'). . .summarises. . .the disorder which is anticipated in Hindu postulations. . .that 'darkness' (tamas) and 'incoherence' (adharma) are expectable conditions of the universe generally. (p. 20)
One begins to see here, at least for Marriott, the meaning of the term dharma--and why it is, though valued, always at risk. The transactional "order" so laboriously worked out in terms of the first two principles is never fixed and absolute: an essential aspect of dharma is that it must be constantly reinvented. (It is, however, not the case that the words dhatu and dharma are "of one etymology" [p. 8]!) For Marriott, it would seem, it is rather adharma that is the principle, dharma, its mere colollary: he prefers the negative here: "unmatching," disintegration--the "involuntary" destiny that stands opposed to and is fully a match for man's petty will. The world of the third principle is that of the Mahabharata, rather than the dharmasastras.
To give some examples of how these variables are used to characterize Indian "types," a brahmin priest may be said to occupy the intersection of vectors "unmarked, unmixed, matched"; a ksatriya, by contrast, is "unmarked, mixed, matched"; a vaisya "marked (relatively), unmixed, matched." The "cubal" opposite of the brahmin priest, the "untouchable," is "marked, mixed, unmatched." Other social "types" can also easily be accommodated: the "demon," the ascetic, etc. (These statements may be decoded by reading into them the appropriate glosses for each term; compare the various "cubes" diagrammed on pp. 10, 25.) The exemplary power of Marriott's analysis is seen in the way it can account for a variety of non-standard and fluctuating social personae that are only with difficulty compressed into Dumont's two-dimensional model.
What is "Indian" about the mode of analysis? Marriott prides himself on having finally abandoned "Eurocentric" social science, whose oppositional categories are imbued with the ideology of the sovereign individual. Mutatis mutandis, this "Indian" sociology would seem to make the "individual" of the West into a conundrum nearly impossible to clarify--he is off the social map, so to speak. But perhaps this sociology should not be applied generally, and especially in the West! (And Marriott's case is thereby made!) How one could "be" anything for very long in Marriott's scheme is open to question. To Dumont's "simple" opposition of two ideologies, of two notions of "man in society," the "hierarchical" and the "egalitarian," each providing for a modicum of stability--and the former, perhaps even more than the latter--Marriott opposes his complex of "dividuals," who vary along scalar vectors of infinite differentiation and whose "types" are statistical extremes: stability is valued here perhaps because it is impo ssible to achieve. What is at issue, rather, is "flowing" or "liquidity," and man's rather puny efforts to arrest or channel it. One begins to appreciate the Buddha's emphasis on "fixing" the mind, on "mindfulness." And, perhaps somewhat ironically, the Advaitin's discovery of an absolute "non-dual" (which he dare not express in the affirmative, as "one") beyond all possible conceptualization!
7. In sum, Marriott's Indian social universe is not a place in which it would be easy to make one's way--bereft of a lasting "identity" (the great bugaboo, interestingly enough, of American pop culture--everyone is summoned to find out, once and for all, "who he is"); adrift in a flux of ever-shifting relations; menaced by an evil that is omnipresent and that cannot be overcome even at the end of time (for even time has no beginning and no end). Marriott's claim is that he is better able to represent and understand the behavior he observes in terms of this relativistic grid of three variables--which he has attempted to visualize in the form of the often (to judge by others' comments) less than well understood "cube."
8. Marriott's effort to link this anthropological analysis (or at least one founded on observation of Indian behaviors) with Indian "ways of thinking" has led him to align these three variables also with three commonplaces of the Indian learned tradition--the three "qualities" (guna) of the classical Samkhya, notions as nearly universal as any in the varieties of Indian speculation--and, like Marriott's three, understood as inseparable components of any and all reality, yet varying infinitely in their particular dosages.
To "(un)marking," according to Marriott, corresponds sattva 'clarity, intelligence'; to "mixing," rajas 'energy, vigor'; and to "(un)matching," tamas 'darkness, the inchoate'. Each of these is further aligned with a physical principle, the three "elements" water, fire, and air, selected this time from the Samkhya pentad of elements; and with a moral principle, derived from the three original purusarthas, or "aims of man": artha 'advantage, profit'; kama 'passion, love'; dharma 'duty'. Other analogies are also posited, but it is impossible to enter into the details here. Indian "reality" is thus not only shown to come in threes, but the various triads are asserted to be "metonymic" (p. 16 et passim) with one another, and their "merging" (p. 12) to ground Marriott's own more abstract "ethno-scientific" triad. In this rather ingenious way, he can claim that his analysis has been anticipated by the Indian doctors themselves, and indeed reflects, in its depth and multiple assonances, an "Indian" analysis of a rea lity identifiably "Indian." While some of these "metonymic" associations are quite plausible, especially from the vantage point of Marriott's own ideas, they do often appear ad hoc to the textual scholar--and, in fact, do not always match the parallels that Indians have also drawn--with the result that, though the terminology looks "Indian," the notions are definitely "Marriott." He makes subtle adjustments, and is determined to read the texts in his own way, better to support his thesis--and this has occasioned some very pointed criticism from, among others, Gerald Larson,  who points out that, despite its emphasis on the three gunas, the Samkhya system is universally understood as asserting an ontological dualism (of purusa 'inert spirit' and prakrti 'active matter'), which both echoes earlier-rejected, and supposedly "Western," dualistic structuralism and is difficult to reconcile with Marriott's otherwise cubal notions. Some of the other quibbles that may be raised are these: why privilege three physi cal elements from the five of the Indian system, setting aside "earth" and "ether"? What is the basis for the counterintuitive association of sattva 'purity, intelligence', with ap 'water' (rather than with fire) and (especially) with artha 'profit' (rather than with dharma, as is habitually done in the Indian manuals)? Note also that the trivarga does not associate the ksatriya with kama; this "end" belongs to the procreative vaisya. Some genuinely perverse associations arise: e.g., air, fire, water with tamas, rajas, sattva (and these with "grasping, walking, excreting"!),  which suggests that the categories are sometimes manipulated mechanically, in disregard of their dissonances and genuine discontinuities. (But it is also the case that the Indian texts in places themselves seem to indulge in ad hoc parallels: cf. Samkhya Karika 24, 28, where the five karmendriyas are merely listed, in an order paralleling the five buddhaindriyas: while "speech" seems to go well with sabda cum ether, the other end of the scale seems quite arbitrary: "evacuation" with taste and water; "intercourse" with smell and earth! [the reverse seems more plausible!]. In any case, the parallelism of the text is anecdotal; no exact parallelism is proposed of member to member.) Note especially that the usual ordering of the gunas--sattva, rajas, tamas, which obviously reflects the "hierarchy" brahmin, ksatriya vaisya--is itself disputed by Marriott: is it too Dumontian? This has to do with the logic Marriott sees implied in "marking" and its contrary. "Marking," in fact, is not for Marriott the dominant notion among his three variables--I have misrepresented him above on this point better to bring out the parallelism with Dumont; it is rather "mixing" that deserves prior emphasis, in accordance with Marriott's view that "hierarchy" in the Indian world is itself a derivative notion, and that the primary conditional determinant of the notion of the Indian "person" is that his body has no precise boundaries. So Marriott is led to invert th e traditional order, and apparently, importance, of the three gunas, which now become "rajas, sattva, and tamas" (pp. 14-15).
Although he has sought to distance himself from these simple tri-partite analogies in his later writings (see below), let me expatiate on the ambiguities that derive from Marriott's attempt to privilege three among the five elements: fire, water, air; for this illustrates, inter alia, the problems met in "appropriating" texts. Beyond the fact that this triad as such is certainly exceptional in the Indian materials, the reduction effaces one of the truly distinctive aspects of Indian physical theory: the fiveness of the elements, as opposed to our usual four: earth, air, fire, water. Thus the parallelism with the five senses is missed, along with the probable rationale for the fifth (for the Indians, first) element, akasa 'ether'--the Veda that is sabadamatra, i.e., "etherial." Also truncated is the numerological progression implicit in the elementsense parallelism: the "simplest" element is ether, for here sound/hearing alone functions. "Air" reposes on a "dualism," for, in addition to sound/hearing, touch/f eeling is operative: we also feel the wind. Similarly, "fire" adds a third sense: color/sight: we see, feel, and hear fire. By the same token, "water" superadds a fourth: tongue/taste. And in conclusion, the Samkhya numerology clearly identifies "earth" as a plenum of sorts, for here all five senses are at play--the fifth and last, proper to "earth" itself, being nose/smell. Not only is this plenum missed in Marriott's tri-partite analysis, and also the element that to brahmanical India is arguably the most "important" of all, sound/ether, but even the middle three of the pentad are presented in an order: first fire, then water, then air, that confounds the Samkhya--an order that the Samkya regards as integral to their conception, for it reflects notions of increasing complexity and distinctive levels of concreteness. But what is even more striking is the total interdependence, for Samkhya of the pentad--there seems no obvious reason to select stages two, three, and four for special treatment (unless one is w illing to risk being accused of imposing "alien" values on the subject matter).
None of these "unmatchings" is in itself very important, but together they risk giving the impression (to the classicist) that Marriott is at least as interested in dressing up his notions in Indian garb as he is in discovering how Indians "think." We might be better served by a straightforward presentation of the "cube" as a set of abstract principles that illumine Indian behavior as observed (a method that would have much in common with Spengler's--see below), with less emphasis on a terminology that sometimes obscures profound differences the Indian world has regarded as important. Perhaps there is more than one "Indian" way of thinking, more than one "Indian" view of the "body" or the "person"--just as there are many views of moksa 'liberation'. Indeed, Marriott is himself open to this possibility: "The model outlined above is undoubtedly biassed in the direction of its sources, which are mostly Hindu, more north Indian than southern, more learned than popular, more of samkhyayoga than of any other darsa na, more ayurvedic than astrological, more orthodox than devotional, more high caste than low, and more male than female" (p. 32). And who would be surprised at such multivalence who had attended to Richard McKeon's meticulous analysis of the many and ambiguous meanings of the term "freedom" in the Western tradition? 
Marriott seems at times to overstate his case, to want to find a single paradigm into which all the pieces will fit-- to discover the veritable Hindu "ethnoscience"! He, in this fashion, might be seen to be pursuing the task begun by the Santana Dharmikas, who attempted also, earlier in the twentieth century, to construct a coherent whole out of bits and pieces of "Hinduism"--but for purposes of ethical and religious instruction! 
9. Marriott, it is true, has attempted to meet these kinds of objections, sometimes even disputing the classicists reading of certain key texts; he asserts, for example, that no ordering of the three gunas is canonical, citing Manu 2.2, where kama seems to take precedence over dharma--perhaps echoing passages such as Rgveda 10.129.4. Marriott also defends privileging rajas (alias kama, srngara), inasmuch as (irreflexive) "mixing" must be logically prior to (asymmetrical) "marking"--which he identifies with sattva. Reasons for the selection of the interior triad of elements are also offered--essentially, it is these three that are found utilized more frequently in Ayurveda (though Ayurveda, in principle, also adopts the Samkhya pentad--see, e.g., Caraka, su. 1.64). The three better match the three gunas, etc.--which are, unarguably, crucial to Indian views of reality, for they have been grafted onto the formulations of many competing systems, even Vedanta.
In the work under review and in later writings, Marriott has proposed a complication of his own system, partly to accommodate the two missing elements, above, and such typical Indian "metaphysical" ideas as the atman and the fourth goal, moksa.  The two elements are accounted for in terms of what now appears as a fourth variable, which Marriott calls "grossening": "ether" and "earth" constitute limits between which other variables have "existence." We have now adduced an existential principle, in addition to spatial and temporal ones. This "fourth dimension" would seem, on the face of it, to cast into some doubt the fundamental character of the triad of variables so patiently worked out in the "cube." On the other hand, the thoroughly philosophical, rather than anthropological (or, geometrical), character of Marriott's analysis also hereby becomes clearer. With "grossening," and its counterpart, represented by "ether," Marriott can now adduce terms like "consciousness" in his discourse, and even register the Indian preoccupation with "liberation," moksa, which expresses ethically the "etherial" extreme. It should be stipulated, however, that Marriott's account of these elements appears to conflate two distinct senses of the "etherial" and the "gross." It is not so much that "earth" (prthivi) is "grosser" than ether (akasa)-it is indeed, as we have seen, more complex, according to the Samkhya account; rather, both ether and earth (as elements external to us) are considered "gross" (both are material, or bhuta forms) vis-a-vis the "subtle" (tanmatra) forms-- "sound," "smell," etc.-- of the same elements that we apprehend in dichotomous parallelism. He has, in any case, traveled far from "village India," and his earlier stress on behavior, although it is admittedly enlightening sometimes to view even metaphysical notions in a behavioral matrix. It is, also, very true that in our modem world, there is no reason to stop with three dimensions--what could be more Indian, or Buddhist, than the notion of an infinity of worlds? It just makes dichotomization more dangerous, as Nagarjuna never tired of telling us. There is, to date, no "final" version of Marriott's matrix--which makes this and any commentary tentative.
10. Gerald Larson, addressing hints in Marriott's earlier writings, has called attention to the fact that a mathematical model can be seen to underlie Marriott's "cube"; Larson may perhaps have provided an impetus for Marriott's subsequent use of such a model as a further "level" of analysis.  In any case, the problems detailed above, and many others, have also encouraged him to attempt a more abstract and generalized statement of his theories. According to this latest version, the three axes of the cube resolve into the three conditions that, when satisfied, define what is called an "equivalence relation" in Boolean set theory. Such a relation may be stipulated among a set of items when (a) each item is related to itself univocally [a R a]; (b) two related items may be reversed without falsifying their relation [if a R b, then b R a]; (c) three items are so related that if a relation holds for any two pairs, it will also hold for the third pair [if a R b and b R c, then a R c]: these properties are term ed reflexivity, symmetry and transitivity. The significance for sociology derives from the extrapolation of this "set theory" onto the standard definition of the "constituent person" of modern Western societies: he is "individual" (self-reflexive, self-authorizing), "equal" (a symmetrical relation holds among any two persons, who become, ipso facto, "citizens"), and "coherent" (such relations apply transitively to all citizens, relationships that can then be formulated abstractly as "laws' which, by definition, cannot be ad hominem). Such is the Western "individual"--in Marriott's terms, ummixed, unmarked, and ever matched. These principles are recognizable in the three catchwords of the French Revolution: liberte, egalite, fraternite. As Marriott's terminology indicates (and it has partly been formulated with a view to expressing this difference), he views the Indian social matrix as implying as its ideal the contraries of these terms--as privileging "anti-equivalence" relations: the Indian "person:" is "div idual" (multi-faceted, deriving his being from an infinity of relations to others--not "reflexive"), "hierarchical" (the relations between any two persons is irreversible--not "symmetrical"), and "chaotic" (no universal law governs all; relations are necessarily particular--"context sensitive," as Larson puts it--not "transitive"). The typical Indian "person" is therefore, according to Marriott, essentially "mixed," predominantly "marked," and rarely "matched." Though the "cube" permits the definition of other "types," including the "Western," we note that the "unmixed" person is often so by private decision and that, though not everyone is "marked," those unmarked are able to assert their status by actively "marking" others.
Although the analysis, as far as it goes, seems quite subtle and, in fact, improves upon Dumont's strictly oppositional strategy, the terminology is very cumbersome and sometimes self-defeating --for, among other things, it seems to point, when all is said and done, to a particular "type" of Indian person, characteristically different from "Western" man, just as does Dumont's analysis; at the same time, the great range of variation made possible by altering one parameter at a time seems to impute to Indian reality the possibility of realizing any "type" of person, even the Western. (The reverse, it must be allowed, cannot be said: the Western "type" is fixed and admits of no variation.) Marriott takes advantage of his variables to generate many Indian "persons": sadhu, king, student, demon, priest, etc., all of which vary in some particular from the "typical." The analytical framework, at this level of abstraction, is no longer India-specific, but belongs to a universal sociology--which result does not s it well with Marriott's announced intention of developing an "Indian" sociology. The best we can say is that, just as for Dumont (but he does this in a straightforward and entirely intentional way), the Indian "case" serves Marriott really as a pretext, to formulate a universal theory of culture that is far from "culture-specific," though capable of representing cultural variations. For both, the "Western" variety is perhaps most notable for its "extreme" and largely invariable monotype.
11. Another cognitive problem that arises from Marriott's increasing reliance on mathematical models comes from the mathematical side. Though the mathematical study of "equivalence relationships"--those satisfying the three criteria of reflexivity, symmetry, and transitivity--is well founded (and is the basis of most set theory and its practical applications, including, most recently, computer logic), it is not all that clear that the study of those systems that fail to satisfy one or more of the announced criteria--"anti-equivalence relationships"--is anything more than a theoretical curiosity. Two considerations arise: one has to do with the obviously metonymical application of these mathematical principles--which have precise mathematical definitions--to the amorphous world of social relations. Reflexivity, for example, is usually understood as the application of the procedure ("relation") that defines the "set" to the term itself such that the result is within the set. If the set is "those numbers whose d ifference is an even number," then the condition of reflexivity is satisfied, because (a - a) = 0: the "difference between a and itself" is zero, zero being interpreted as an "even" number. (This set, incidentally, is also symmetrical and transitive, therefore equivalent.) As sociologists use the term, however, it appears to have a much more commonsensical meaning: the unit that is seen as self-defining, self-validating--not requiring justification through another. No "function" is evident here: it is not clear in what sense the "individual," for example, "operates" on himself to arrive at any "result" whatever. This usage is tautological, in that it simply posits the individual (like Bertrand Russell's minimal postulates?) absolutely, distinguishing him only from his nonexistence. Marriott's "reflexivity" thus seems to amount to the "law of identity" (A is A), an axiom of most scientific discourse, starting with Aristotle: terms whose meaning changes within the range of investigation are not subject to scien tific treatment. If the Indian really is a "dividual," he would seem to satisfy the formula (a R -a), and thus be indeterminate in that sense, and not subject to scientific treatment, at least without further analysis. Mathematical "reflexivity," on the other hand, is a function--a relation--applied (perhaps tautologically) to an entity already assumed to exist.
If the terms seem ambiguously applied to "Western" social discourse, what are we to make of their inverted application to "non-Western"? Not only are "dividuals" amorphous, but Marriott's logic suggests that Western mathematics, with its emphasis on set theory and "equivalence relations," may also lie within the category of "culturally bound" knowledge--a position carried to its logical extreme in the writings of Oswald Spengler, according to whom each civilization has, among other cultural expressions, its own proper "mathematics." But is scientific mathematics just "ethno-mathematics"? Will a properly "Indian" mathematics, then, if we may be permitted to use the term, operate then in terms of "anti-equivalence relations," and thus be bizarre, indeed--though not unthinkable? One might suppose that its postulates would be: (a R -a); (a R b [greater than] b R a); (if [a R b] and [b R c], then [a X c])-which might well also formulate Marriott's "Indian" person. Curiously enough, these "anti-equivalence" criter ia appear to be illustrated, in part or wholly, by many mundane relations that few would regard as either Western or Indian: e.g., the relation of "inclusion," which fails to satisfy the condition of symmetry (unless, of course, the two sets are identical, in which case, the symmetry is trivial); or "difference," which is symmetrical, but fails the other two conditions. And then there is "fatherhood"--which fails to satisfy all three conditions! I am not my own father; if I am the father of X, then X cannot be my father; and, certainly, I am not the father of my son's sons. Does this explain why "fatherhood" is more congenial in the Indian universe of discourse? (Before we jump to this conclusion, it should be noted that the same considerations apply to "motherhood.") In fact, it would be fairer to say that "equivalence relations"--though they may be at the heart of certain aspects of mathematical theory--by no means exhaust even the "Western" mathematical universe. They constitute, indeed, an important kind of relation--among others that have other, though perhaps not equally "applicable," implications. 
To appeal to such notions as "reflexivity" and the rest in an attempt to disambiguate culture types, is perhaps to mistake the "cultural" import of such abstractions. It is certainly not the case, in this battle over the modern "individual," that the West has canonized a "reflexive" principle while the "East" has not; it is rather--as McKeon teaches us--a question of what indeed is considered "reflexive." From what "subject matter" are the principles elicited? The human individual, in India, may not be taken as a reflexive principle, to be sure, but "self-defining" categories do nevertheless get advanced--for the civilization is, after all, highly "self-aware," "self-conscious." (Or, at least, this has been the business of the brahmin for two millennia or more.) As we are well aware of by now, many modes of traditional Indian intellectuality affirm the illusoriness, or the non-entity, of the atomic individual--vis-a-vis an unchanging "self," or a flux that alone is "real." But what is real, then, is that lar ger self, be this interpreted as atman/brahman (atman is the reflexive pronoun, after all), or as the self-sustaining village (Dumont), or as the caste dharma that reincarnates without end (Krsna), or as the karmic process itself, entirely self-referential and eternal (Buddha). Such conundra point again to the difficulty of aligning abstractions of any sort with terminology that pretends to be culture-specific. "Native" terminology, of course, is itself "abstract"--and the claim must either be made that there are culture-specific abstractions, which appears to go against the generality inherent in abstraction itself, or that abstractions are not part of the culture-specific universe of discourse, which seems to imply that the natives either do not think or are not "natives" insofar as they do.
12. We have referred several times in what precedes to Oswald Spengler, whose work offers many interesting parallels to Marriott's project. Spengler is one of the more uncompromising exponents of cultural relativism. "The history of humanity has no meaning whatever" (Decline of the West, 2: 44). With antecedents in the great nineteenth-century romantics-from Hegel he derived his taste for comprehensive solutions, and from Nietzsche, his disdain for Western civilization--Spengler saw himself as the final prophet of the downfall of the West, a senescent civilization exhausted by its own achievements. Unlike those of the biblical prophets, Spengler's was not a moral denunciation; but was, rather, a commentary on the ineluctable destiny of peoples, of which destiny he had discovered the key. The perspective is Jovian: Spengler's philosophy is itself a "reflexive" sign of the very decay that saps Western culture--it presents itself as a wholly intellectualized and complete system of thought that renders action un necessary--illogical, even: typical of the "final phase" of any culture.
What interests cultural relativists, though, is not the Gotterdammerung sung for Western civilization, but the elaborate formal methodology that buttresses the doom-saying. The Hegelian view that the West consummates the world-historical process is rejected in favor of a radical relativism that attributes to each civilization an existential and a moral autonomy--extending to all spheres of its spiritual life, architecture, politics, philosophy, and, yes, even to mathematics, which too is proclaimed relative to the conditions of its origin: ours is that of boundless space, the mathematics of infinity--calculus; that of the Greeks, of the bounded body, the mathematics of surfaces--geometry; that of the Arabs, of the "cave," the mathematics of the contextual variable--algebra. These mathematics are not, strictly speaking, comparable, because they derive from different "prime symbols"--infinite space, the finite body, the bounded but empty cave--are, in other words, different realizations or representations of t ime and space. The relations between civilizations is "morphological," not substantial--each civilization will have a mathematics, a concept of the "state," an ethics, etc.-but the relations between them can never be ascertained by direct conflation or confrontation, but only by apprehending the place of each in the network of formal possibilities that define each culture as a self-sustaining whole. (Spengler was contemptuous of Western colonialism, and, despite his bourgeois German roots, he broke with the Nazis on the question of whether Jews were "inferior.") The apprehension of that "morphology" was, of course, Spengler's great achievement, and from it he was able to "deduce" the "destiny" of peoples. To his critics, of course, the pretentious methodology rested on a rather simple formal analogy--that of the life-cycle of the individual with that of a civilization: Spengler's "morphology" was that of the biological process itself, applied rather naively to the vast organisms (if they are that) we call "cu ltures"--each of which will be seen to pass through "stages" and to elaborate "forms" of growth and response that define appropriately the life of the culture--at its birth, in its youth and maturity, and eventuate in its inevitable death.
Spengler doesn't say too much about India, though the Buddha figures as a "universal" reformer typical of the late stages of a culture. It's not even clear what "prime symbol" Spengler attaches to India--although certain stray remarks suggest that he thought of India in terms not unlike Hegel's "negativity"--the great Indian invention, after all, was "zero." If so, we might suppose that the prime Indian symbol was that of "being elsewhere," or "not being a body" (dare we say "otherness' or Nagarjuna's sunyata?)--not to speak of Marriott's "mixing"? One might also interestingly point to one of the important Buddhist theories of "definition," apoha: what a thing is, is the sum total of its not being other things. Although Spengler's main attention is directed to the three civilizations already mentioned, he does opine that Egyptian and Chinese also have much in common--for both have as their prime symbol a "path"--in the Egyptian case, narrow, unidirectional, bound in stone (life is a form of premature death, and is perfected in mummification); in the Chinese, a "wandering" (tao) through a "landscape" of incommensurate opposites (yin and yang), in endless and ambiguous cycles of life and death (the ancestors remain within the family and immortality may be had in the living body). And, again, in an aside that we wish had been developed more fully, the Mayan is said to have much in common with the Greek.
The architectural and spatial metaphors that undergird the distinctiveness of civilizations invite comparison with Marriott's fixation on "dimensional" representations of culture: the "dichotomous" West; the "cubic" East. One of the lacunae in Marriott's argument--at least in its present form--is its "polarity"; it is fixed, like Dumont's, on the contrast between India and "us," it pays little heed to the other major civilized traditions. In this respect, at least, Spengler's perspective is more comprehensive: he does not simply assert in the abstract the possibility that each culture is unique, and then go on to discuss just two cultures. Marriott has begun to respond to this problem by advancing a variety of new metaphors that will guide the analyst of cultures other than India and the modem West. Japan, we are now told, conceives of the "person" in pneumatic terms, whereas in Morocco, the person is more akin to a beam of light.  Just how these additional metaphors jibe with the triadic dichotomies is not immediately clear. And so, as the metaphors multiply, the Spenglerian analogy becomes more troublesome, and Marriott's ultimate thesis becomes more elusive.
13. In order to address more directly the claim made several times already in this essay--sotto voce--that the real significance of Marriott's enterprise should be judged in terms of its philosophical, rather than its strictly anthropological, implications, we might ask ourselves the question whether there is any way of understanding this enterprise other than as participating in a "controversy"--with Dumont and others (including a broad spectrum of thinkers extending back from Hegel or Spengler to the anonymous authors of the Samkhya system)--where "the resulting reformulations... are placed in opposition as if they were univocal and as if the choice between them were a simple problem of logic involving little more than the resolution of contradictions."  Certainly, the greater utility of Marriott's formulations--for the rest of us, at least--will not be discovered at the level of resolving disagreements over "facts" or "models" of interpretation, but rather in the way new methodologies of thinking abou t India and, in general, the variety of cultures are thereby authorized and developed. To this end, I will attempt a very coarse "disambiguation" of Marriott's theses in the context provided by Dumont and Spengler, using a version of McKeon's problematic method, as outlined in his essay "Philosophical Semantics and Philosophical Inquiry."
If Spengler's method represents a variety of McKeon's "dialectical" method,  it is no doubt because of its developmental and comprehensive character: the basic biologic metaphors of birth, growth, and decay are applied to wholes that are "civilizations." Contradictions become an essential aspect of the civilization's working out of its "destiny"--its course being marked by a necessary transit through opposed "moments": youth, maturity, senescence. But by "living through" its natural course, a civilization in fact becomes "complete"--the stage Spengler saw in contemporary Western civilization: complete, and therefore "dead" (or dying: no longer open to the vibrant, "creative" innovations of youth).
The "historical" variety of civilizations, however, constitutes Spengler's problem: he is able ultimately to discover their unity, their essential oneness, but only in their process; civilizations are also unique, in that each reveals a unique "form"--itself derived from a different understanding of "space," the essentially formless, a difference that permeates all organic aspects of the civilization, its architecture, philosophy, social relations, its mathematics. In this way, one might say, the modern notion of the "individual," both as a life form and as an irreducible value, has in one grand sweep been attributed to the protean (and otherwise formless) patterns we call "civilizations." The "objects" to which the dialectic method is applied are not those "proper" to dialectic, for each has an essential character, and is irreducible to the others--with the result, of course, that there is not one "dialectic" of history, but variously many, unified by form alone. "Essentialistic" interpretations, says McKeo n, "resolve problems by seeking properties and causes which are natural functions or acquired conditionings,";  indeed, the "kinds" of civilization are limited only by the number of basic "morphs" that are available in some sense for exploitation.
And, finally, it is these "kinds," or "forms" themselves, that function as Spengler's principles, for they make possible the objects that the method discloses. But their arbitrary character surely suggests that they belong properly to an "operational" mode of thought: they are hypotheses--both of the author and of the civilization--that have no rationale other than adumbrating a "different" Weltanschauung. They exhaust their utility in so defining possible "life forms"--none is necessary (for civilizations are different) and none is of "value" (for no hierarchy of forms is admitted); they are purely existential--in Sartre's sense: "free" acts ex nihilo.
It may, indeed, be interesting (and not a mere coincidence) that Spengler and Marriott (who also, as we will see, relies on what McKeon terms "actional" principles, which are related to the "operational" method) ultimately seek to express their understandings of civilizational "difference" via spatial metaphors. The principles though are quite different--those of Spengler are "static"; those of Marriott, "dynamic" (if we are to judge by the "transactional" principle that characterizes Indian social intercourse); nevertheless, the notion of a basic "metonymic" form infusing all manifestation is common to both. And these are not Plato's "forms"--ideas more real because more universal; they are forms of "appearance" only. For Spengler this is evident: "space" has no being as such: it is its representations alone that give it "reality." For Marriott, too, the Indian "person," characterized, especially, by "mixing," is no "universal man," either by manifestation or hierarchical evolution--but represents merely a specific index of values whose range, and therefore possibility, is abstractly defined by the grid, or "cube." Both are, in that sense, arbitrary and causeless--though certainly "definable."
Opposition and contrast are throughout the leitmotifs of Dumont's method. The principles to which he appeals, in Homo hierarchicus, are themselves only two in number: that of equality and that of hierarchy. And one should note also that the "problem" which these principles respond to is their own supposedly absolute character. Relativization is thus essential to the method adopted. As a method, we have approached the area of McKeon's "debate"--but here of "value systems," whose confrontation (comparison) is essential to the acquisition of any useful view of them-- "apprehender intellectuellement d'autres valeurs."  (Each, by taking itself as absolute, renders statements about itself circular and "petitiones principii"--for which reason Dumont is dismissive of those anthropologists who treat merely of "social stratification," a way of reducing Indian values to ours.) These are hallmarks of the "operational" method. The study, too, proceeds in terms of oppositions, notably "purity" and "pollution," which, in India, have been disentangled from the competing hierarchy of power and thus ground a social system unlike any other known, "religious," but in a different sense. Dumont's insistence that the caste system is "religious" betrays his need to portray it not only as an absolute in its own terms, but as involving "values" that cannot simply be dismissed as "aberrant" versions of "our" values. But, as to what are" purity and pollution, we find only that they have no "essence." The indices of hierarchy are arbitrary, relative to one another only--they frequently concern regulation of "natural" functions, chiefly reproduction and alimentation--but also social functions, such as housing and employment. But why should pre-puberty, as such, function as a sign of marriage eligibility for those of high status? Or why should eating only vegetables serve a similar social function? What is important is not the custom itself or the act, but that a differentia be established: those lower in the scale marry their daughters a fter puberty; they also eat fish and meat. Of course, emulation constantly disturbs these differentiae (for they have no absolute rationale-- and their inversion could easily be presumed), which leads to ever more differentiation--in this case, to younger and younger brides, or to more and more finicky eating habits--affecting not only what is eaten but whom one eats with. The floating character of the indices, as well as their arbitrary character, signify that they have no importance in se, but only in context. To the question, "what is pure?" we respond only that "it is what Brabmins do." Again, the "clarifying of terms" is an essential aim of "debate." This again sounds quite "operational"--as method: the "knower makes knowledge."  In addition to the fact that such polar oppositions are characteristic of the "operational" mode, there is the additional factor that this opposition has no basis in "fact," but exists only reciprocally: that is, "pure" can be defined only as what is "not polluting," and vic e versa. So, if Dumont's method is, in McKeon's terms, "operational," where do we find his "principles"?
For Dumont, "comparison" is crucial, and it turns out that comparisons can be effected at very abstract levels-- that of the "idea" itself, the "principle" that enlivens a "civilization"--and in the case of modern Western civilization--distinguishes it from all precedent "civilizations." This, indeed, begins to sound Aristotelian, to the extent that contrasts of this sort define exclusive subject-matters, whose "nature" is clear only when projected against alien or non-conforming subject-matters; for the underlying principles can be disambiguated only in their problematic employment. In "Physics" is clarified the subject-matter of "Art," in that each develops the notion of "cause" differently. We then see that it is not the "bed" that the physicist studies, but the "wood," for it is only this that has the "efficient cause" within itself. Qua bed, there is no "object" for the physicist to study--no way for him to make causal sense of "it."  Dumont also finds his problem in ambiguous subject matters--two o pposed systems of value--"egalitarian" and "hierarchical" civilizations, neither of which can make sense of the other in its own terms--that each considers itself absolute and exhaustive. And yet their incompatibility points to the truth that neither can be final. "Reflexive" principles, associated with McKeon's "problematic" mode of thought, have his character: they "resolve problems into a plurality of wholes formed by principles which are reflexively instances of themselves." 
The aim of such comparison, says Dumont, is to construct a science, which he terms "sociology," that will be able to situate the "unity" of mankind and its social systems. This science is instrumental only, however; its rationale is not conceived as that of offering a final "truth" (for each of the systems it studies already offers that); it is rather that such a science frees us from the parochialization that is inherent in "being born in a given time and place." Only in that kind of "intellectual" realization is "unity" found. And it is a unity that has as much to do with the way we can see ourselves as it does with the way we see others. Presumably, although Dumont does not say this as such, the recognition that one's own values are relative will have actional consequences as well, at least of an eirenic sort. The values themselves will not, as such, be modified, however. This "(re)interpretation" of and by the partners in the debate, in fact, makes "debate" possible--and replaces the "dialogue des sourds " that precedes it. But what, in fact, do we learn thereby about ourselves and about others (or, "the other"--for there is just one partner here)?
The "system" of castes is a hierarchy, to which "ranking" is essential: activities and types are ranked on a scale that, grosso modo, may be seen to run the gamut from "divine" (or above) to "animal" (and below). Those activities that associate man with the divine are "pure" (worship, study, visiting holy places, obligatory rituals, etc.); those that characterize him as an animal (birth, eating, reproduction, death, etc.) are "polluting." ("Occupations," as such, including "ruling," are more or less neutral, except as they are involved in such activities.) The universe thus "constructed" is exhaustive and involves "values" at every turn and level. But this has more the character of an "interpretation"--made possible or likely by the method followed (which is certainly not "empirical"): there is no self-evident link between any given act and its purity locus: the main thing is the place it occupies vis-a-vis the "opposites" in the ranking system. This is certainly made clear in the way castes interact with ot hers above or below them in the hierarchy--and goes a long way to explain the notion of "untouchability": by this is expressed the necessary distance between functions at either end of the hierarchy--both being absolutely essential to (indeed, defining) the social whole. If the "untouchable." were truly alien, there would be no reason to consider him "untouchable." Such "constructions" appear to involve "entitative" interpretation, associated with the "logistical" mode of thought: "knowledge" is here constructed "from the elements of the knowable."  "Knowledge," in this case, means the ensemble of rules and protocol that both define and enable the notion of a social "whole"--which Dumont locates in the village and its immediate context ("India," in a Dumontian sense also, is an accumulation of such village universes--also an aspect of a "logistical" construction from "least parts": but India is a "whole" in a less interesting sense than its villages, for it really has no "functional" status). What we come to "know" is the village as such a universe. The "elements of the knowable" are, of course, the arbitrary differentiate of rank that define the "village" as a functioning unit, both ritually and economically. The "entities" in such a system are, indeed, "wholes" and their "parts" are seen always sub specie unitatis--but are not absolutes. The "whole" thus constructed is as arbitrary as its parts, though it appears to be based on or reflect "natural" gradations. "Untouchability" is , however, made necessary only by the need for definition itself. Dumont's "operational" method thus sits somewhat uneasily with his "entitative" interpretations: the indices of rank from which the wholes are constructed only appear to be "natural"--but, of course, at the same time derive their authority from that conflation. From the point of view of the "interpreting" mechanism, what we thus acquire is a "sociology," which, though entirely conventional, is unique to the Indian case--and, through which, mutatis mutandis, we see the possibility of a unique "sociology" for the modern West--or, to put it in another, slightly more "Roman" way, one in which rhetoric finds its proper place among the civilizing and unifying functions of society.
Despite constant shifts in perspective, there is a constancy of direction to Marriott's thought, perhaps best appreciated in his hostility to Dumont's program. This criticism has many levels, but in its most general terms, Marriott accuses Dumont of misrepresenting "India"-- on the one hand, falling prey to an excessively textual, "brahmanical," world view; on the other, adopting, unreflectively, "Western" categories of analysis (or, perhaps, transposing "brahminical" categories into "Western" dichotomies?--the exact interface of these seemingly contradictory positions is not clear), which he then attempts to "impose," top-down, on the real society. There are two issues: it is somewhat paradoxical to claim that the intellectual representatives of a culture are themselves not valid witnesses to what the culture is about; in any case, a "top-down" model is exclusively "ideological," and does not capture the sense of what motivates Indians to behave as they do--around the village well, so to speak. Put together , these criticisms seem to say that brahmins are unreliable (probably because self-interested) students of extra-brahmin culture. It seems also to assert that "disinterested," truly "Indian" categories of analysis are not only "hidden" from view, but must be elicited by an agent alien to the "biases" of brahminism. To a certain extent, this is not to denigrate the brahmin. He has, throughout Indian history, been the chief force promoting a coherent and organized Indian "culture"--doubtless, in part, representing his own "interests" (as hierarchically superior, etc.), but also himself constituting the only cement in a fissiparous system otherwise devoted to fragmentation.  Dumont's analysis of "India" places great weight on this factor; Marriott's downplays it. Oddly enough, by ignoring this "political" dimension of brahminism, Marriott's "bottom-up" approach misses the main "social" force of the brabmin--and goes on to replace it with a construct of notions chosen seemingly ad hoc from the classical lexic on, which Marriott sees as including and completing brahminical "ideology." It is perhaps significant that much seems to be drawn from the Samkhya-Yoga system, itself probably "non-brahminical" in origin (or at least "noncaste-brahminical").
What does this disagreement tell us about Marriott's methodological stance? Where are his principles? It seems that they are induced from observation of a variety of behaviors--at least three oppositions being necessary to account for the decisions Indian themselves make as to how they act: three, instead of Dumont's single opposition, "purity-pollution" (which grounds his notion of "hierarchy"). While Dumont stresses the single contrast that suffices to characterize "India" in relation to the modern West, Marriott seeks to downplay the influence of "Western" modes of thought of any sort (even "comparison") in an effort to discover the principles Indians themselves employ. "Marking" is Marriott's term covering those aspects of behavior that most preoccupy Dumont (and which he formulates as "hierarchy"): yet, even here, Marriott uses the term in ways opposed to Dumont's views. The "marked" category is the lower, hierarchically speaking; the higher, "unmarked": for Dumont, certainly, it is the higher category that is characteristic and specific; it would be perverse to consider the "natural" or the "animal" in us "marked." "Purity" is the special state, designed to be induced and preserved by all the "rules"--no rules are needed to accede to the claims of "nature," which is rather the arena of "pollution." "Mixing" is more important for Marriott--not only because it represents an aspect of Indian behavior missed by Dumont, but also because here is grounded the notion of the Indian "person." His is a "fluid" universe in which matters, qualities, even ideas are "shared"--much as "substances" that, in the classical view, underlie and sustain different surface forms. The "person" then is not the isolate of Western thought, intrinsically self-maintaining and source of all validation, but, potentially at least, a relational nexus whose "other" is the universe itself. An Indian, then, is ialways a "such and such": qualified as to birth, language, occupation, residence, age, even time of day--and, of course, caste. But it is to be noted that these qualifications do not so much serve to separate the "individual" from his fellows, as they link him with correspondingly defined, highly "fluid," classes of "fellows," often greatly different from each other, given the different aspects of the "person." For Marriott, even the rules of purity and pollution cannot be understood without positing this interpenetration of "substances"-- for a person is "pure" or "impure" depending on his contacts and context, and these states are maintained only relationally. By a different route, Dumont, of course, reaches a similar conclusion--his "homo hierarchicus" is the anti-individual corresponding to the self-validating individual of Rousseau, "homo aequalis." Marriott's third principle, or rather, anti-principle, is "(un)matching," a principle of indeterminacy that serves as antidote to the efforts to "clarify" by ranking. This is the universal "chaos" so feared by Indian cosmologists.
These "principles," if such they are, do not appear to be "reflexive," in the Dumontian or Aristotelian senses. They do not serve to establish self-referential subject matters; rather, they function ensemble to constitute a "map" whereby the Indian agent isolates issues and characterizes choices, much as do Aristotle's "topics" for the logician and rhetorician. And, certainly, all three principles are found embodied simultaneously and inseparably in any Indian person or situation. These "principles" then have rather the character of scientific "laws," or at least "functions," and their tendency to abstraction, their omnipresence, as well as their interpenetration in the "real," are thus explained. We are then wrong to consider them "principles," for they have rather the character of formulations, conclusions; they already represent "interpretations" and suggest that the "method" employed here--which aims at their discovery and formulation--is a variety of McKeon's "logistical." This explains, among other thi ngs, why the number of "laws" in the system can be expanded at will or as necessary: the addition of "grossening" is no defect, but indicates the method's probative power. As interpretations, however, the conclusions appear to be "essentialistic," associated with the "problematic" method--that is, they "resolve the problems encountered in the known in theory, practice, and production, to reconstitute the known in a new form."  What, indeed, are the "principles" with which the method "begins"? Here, we seem to meet the almost constant refrain that "India" needs to be understood in terms of "Indian" categories. The "problem," for Marriott is the false universality of "Western" modes of thought. (Note that this differs significantly from Dumont's notion that "Western" and "Indian" constitute self-referential universes--their "problem" is rather that they do not "communicate" usefully.) Principles are then to be found in the recognition of cultural specificity, which, in the first instance, both verifies and demotes "Western" thought. This perhaps accounts for the importance to Marriott of Dumont's work, as well as for the criticism he persistently levels against it--for Dumont is the cas typique of illegitimately generalizing "Western" thought. Since Marriott also represents a "Western" mode of thought, the negative, or self-destructive character of the principle is rather too evident: "reflexivity" is, in part, the answer--if the self-analysis leads to the recognition of "self-referential" principles. (It is the principles that must be self-referential, not the "universes," as with Dumont.) Marriott is elusive on these points, but it is his discovery and valorization of "mixing" that seems to resolve these issues for him--not the three terms we initially termed "principles" (but which were, at least in their functional shapes, conclusions)--but the single term "mixing": the notion of "fluidity" that for him serves absolutely to differentiate the Indian view of the "person" and to define the Indian world. (The o ther notions are "derivative" in the sense that they are borrowed from other accounts and that they logically presume "interpenetrability" as their rationale. Note also Marriott's insistence that rajas [scil., "mixing"] is prior to sattva.) This notion not only relates the Indian "person" to (in principle) the universe, but it seems to propose that the universe is not statable except in terms of an infinity of vanishing points, each related to all. (We are reminded of Hua-yen Buddhism.) The "universe"-- and it may well be the village--is in any case a summation of such "units."
The "mixed" person, then is the "simple" whose combinations suffice to generate "reals," both in se and extra se, that are always complex (for "existence" itself is thinkable only in terms of such a plurality) and whose modes of combination the method seeks to understand and formulate. "Transfers," "transactions," then, become the only entree into this "fluid" world, where "substance" becomes "quality" and "quality," "substance." This hypothetical "person," who truly exists only in relation, is not "atomoic" in the sense of Democritus or Leibniz; the simple here is already characterized as a relation; though in this sense "simple," the "principle" here is that typically associated with the operational mode: "actional principles provide beginnings in uninterpreted terms set in fixed but undefined relations by thesis or postulation before they are interpreted to produce knowledge and values."  The ambiguity of method and principle is brought out in Marriott's case by the fact that the "simple" principle on which the logistical method depends for its beginnings is interpreted here as a "relation," a "person" who is already a "non-person," and who is thus implicitly already the universe of such persons that the method seeks to formulate.
Viewed in this way, Marriott's thought reveals also an unusual combination of McKeon's modes: an "operational" principle used in a "logistic" method to produce a conclusion typical of the "problematic." For Dumont, an "operational" method is linked with "problematic" principles and "logistical" interpretation. Both eschew the "dialectic" mode in any of its incarnations; each of the other three modes appears, but the ensemble is arranged differently. Structured comparisons with Spengler or Hegel, for instance, will therefore have only limited heuristic value.
14. What can this analysis do to help understand or resolve the differences between Marriott's and Dumont's theories? One suspects that the core difference is that of method. A "logistic" method is one of synthesis, which finds its problems in combinations of given "simples," on the one hand, and looks to the formulation of abstract universals, on the other; an "operational" method is grounded in the universality of "debate"; it finds its problems in the arguments, conclusions, of others, and looks to the identification of "simples"--values, interests--that is, "terms," found among the oppositions that the method "postulates" and which most clearly justify the strategy chosen. To Marriott, the Dumontian approach looks both irrelevant and ad hoc; to Dumont, the method of Marriott seems limited by a naive faith in "realia" that are "given" to the method--not admitting that there are no terms apart from those supplied by the discourse itself.
The "selection" of problems is not the same: for Marriott, the Indian "universe" has no preordained character, but is a fecund source of terms proper to it: it is a unique object of study: the "principles" thus selected are those that uniquely characterize, make complete, the unique whole that is "India"--"mixing' etc., as we have seen. For Dumont, problems are found in the conclusions of other scholars, with whom "debate" suffices to isolate issues, that is, "theses," having to do with the "formulation" of the individual: the "context" of the debate is marginal--"India" as such is a pretext, a side-issue; Dumont's own work in India is simply a "preparation." (This is even clearer in his subsequent work, developing the theme of Homo aequalis, which engages in a debate with other theorists "from Mandeville to Marx," the subject of which is European ideology.  Principles are not so much discovered, as they are "postulated" via the inherent logic of ideas--for no amount of "fieldwork" will reveal the self-v alidating "individual" of the West, or the interreferential, graded "agent" of the pre-modern East. These are, in fact, "revealed" by the opposition itself--logically complete and exhaustive.
Marriott's "science"--the ensemble of "interpretations"--is again incommensurate with Dumont's: the increasingly abstract "laws" of Marriott's later work signal the view that a "science" aims ultimately at universals that hold invariably for all relevant "data." It is only in such "knowledge" that universality is to be had--"theses" grounded on observable facts are irremediably biased by the conditions of their origin--which is, for Marriott, not a sign that they are not "true," for they are as true as they ever can be: they are "relative" at that level--as are Marriott's own beginnings: they suffice to supply "problems." The "science" of Dumont, on the other hand, is hardly "universal": it is as "particular" as it can possibly be--the sets of terms that state and justify the postulated oppositions that underlie the argument. I think "purity" and "pollution" figure here, for if one "understands" these terms, and their reciprocal reference, the structures and modes of behavior of Indian society devolve of the mselves. Even the sense of these terms is "particular" to the Indian context--but that just reinforces the point that the Indian universe of discourse imposes its own definitions--as does the Western.
Looked at, then, from the angle of the discourse itself, Dumont's has the unusual feature of substituting reciprocal or reflexive "theses" for the "actional" theses normally associated with the "operational" mode; his "interpretations," on the other hand, are "entitative," rather than "existential"--which we have taken to mean that "purity," pollution," etc., are seen rather as "secondary qualities... appearances" that are constructed "from a nature that underlies phenomena" --the "nature" at issue being, as we said above, the notion of hierarchy and its necessary indices.
Marriott, on the other hand, substitutes "actional" principles where "simple" principles are expected; and his interpretations are "essentialist," rather than "entitative," as, again, might be expected in the "logistic" mode: they "resolve problems by seeking properties and causes which are natural functions or acquired conditionings" --Marriott's three intersecting "variables" that render Indian behavior "around the village well" both intelligible and justifiable.
15. This exercise in "problematic" may have permitted us to see somewhat more clearly not only why Marriott and Dumont were unlikely to "agree" on the major issues broached--despite selection of a subject matter and conception of problem highly similar. Philosophical disagreements, in this sense, are both inevitable and fecund. The disambiguation of their positions may also, in McKeon's terms, help to further discussion by directing it away from useless confrontations and toward more careful consideration of the methodological format that lends cogency to their arguments. McKeon's problematic also helps us understand why striking similarities are found in the otherwise quite dissimilar projects of Marriott and the "dialectical" Spengler, and how to phrase those "similarities" in such a way as to confront the logics on which they depend. This otherwise superficial resemblance is also worth some comment for Marriott can he said to have addressed an aspect of the history of "civilizations" that had been a provi nce of the dialecticians, but who approaches the questions of civilizational character and morphology from the other methodological extreme-- attempting to "build" his wholes (now, complexes) from an inventory of simple indices in which all possible variation is contained. That McKeon, too, saw his problematic as useful in comparing cultures and "fashions" generally is indicated throughout his essay, "Freedom and History," which begins by adducing a telling similarity between recent American (and English) philosophy and that of the Roman silver age, particularly Cicero's, in their tendency to approach philosophical problems from the angle of--indeed, reduce them to--problems of language and rhetoric.  Though "categories of analysis"--the philosophical method and "mode of thought" adopted--in this way both determine the problems treated and the possibilities of solution offered, it is not the case that McKeon's problematic forces us into a bland relativism of mere "points of view"; for it is precisely the disambiguation it makes possible that allows refocusing our attention on the perennial strengths and timelinesses of certain perspectives, and away from the empty carping on each other's "weaknesses" that so dominates each point of view's account of its alternatives and of itself.
It is probably adventitious that each of the four writers I have discussed in this essay illustrates one of McKeon's four "methods"--McKeon himself undoubtedly illustrating his own "problematic" method--but it is precisely this contrast that demonstrates the richness of the "coin parative" problem, and its insolubility in the mode of confrontation: Marriott's logistic leads him to focus on characteristic behavioral units; Dumont's operationalism, on exhaustive ideological oppositions; Spengler's dialectic, on unique and inclusive wholes; and McKeon's problematic, on the various theories of comparison themselves, each "adequate" in itself. The latter is, as one might suspect, more commensurate with this problem, which has at least as much to do with one's perspective as with the "things" compared.
If any conclusion is possible to such an essay, it is that Marriott has come to occupy more of the territory of the Indologist and philosopher than his original agenda might have appeared to warrant. He has done yeoman service in again daring to raise the question of the integrity of cultures, at their most abstract and "civilizational" levels; he has, in these terms, reenlivened the old debate between "nature" and "nurture"--whether cultures, in their historical and regional diversity are essential or adventitious aspects of man's "nature." He has, for better or for worse, sought to present us with a "unified mode of thought" on this issue--and, in this, has certainly been more successful than the "cultural theorists," with their vague and essentially self-referential notion of "diversity." Even if there are no final answers, Marriott's work carries us beyond such jejune views and back to problems of cultural heuristics. For this, and for the intellectual challenge his work provides, we are in his debt.
This is a review-article of: India through Hindu Categories. Edited by MCKIM MARRIOTT. Contributions to Indian Sociology, Occasional Studies, vol. 5. New Delhi: SAGE PUBLICATIONS, 1990. pp. xvi + 209. Rs 185.
(1.) In the series of symposia devoted to "Karma and Rebirth," e.g.,... in Classical Indian Traditions, ed. W. D. O'Flaherty (Berkeley and Los Angeles: Univ. of California Press, 1980);...: Post Classical Developments, ed. R. W. Neufeldt (Albany: S.U.N.Y. Press, 1986).
(2.) Given a very pointed formulation in the work of Louis Dumont, notably Homo hierarchicus (Paris: Gallimard, 1966), but, of course, animating the work
of countless anthropologists and others, most notably McKim Marriott, the subject of the present essay.
(3.) Cf., inter alia, S. Radhakrishnan, Eastern Religions and Western Thought (London: Oxford Univ. Press, 1940); H. Zimmer, Philosophies of India, ed. J. Campbell (Princeton: Princeton Univ. Press, 1951), etc.
(4.) Some collected in the volume of Contributions to Indian Sociology, n.s., 24.2 (1990) devoted to Marriott's work. See, particularly, the hostile, even venomous, critique of Michael Moffatt (ibid., 215-36).
(5.) Ibid., 255 et passim (K. N. Sharma). But Gerald Larson is an exception: see his "A Samkhya Response" (ibid., 237-49). I will have occasion to return to Larson's work below.
(6.) American Anthropologist 71(1969): 1166-75.
(7.) In The New Wind: Changing Identities in South Asia, ed. K. David (The Hague: Mouton, 1977), 227-38.
(8.) The cultural solipsism of the theorists is sometimes carried to an extreme that in times past would be regarded as comical: according to H. von Stietencron, "Hinduism" itself may be considered a Western "construct" (in Hinduism Reconsidered, ed. G. Sontheimer and H. Kulke [Delhi: Manohar, l989], 11-27)-- and whether true or false, my point is made! The post-Foucauldian passion to translate all argument into "discourse" and all truth into "projection" (a reflexive phenomenology that, by the way, has much in common with the vijnaptimatra of the Yogacara) allows others to speak of "caste" as though it were largely an invention of those who first described "caste"--the English (or Portuguese) who were its first systematic witnesses! (Cf. Ronald Inden, Imagining India [Oxford: Blackwell, 1990], and especially, his "Orientalist Constructions of India," Modern Asian Studies 20 : 401ff.; for an acerbic "deconstruction" of Inden's "deconstructionism," see the review by David Kopf of Imagining India, JAOS 1 12 :674-77, complete with references; it is certainly ironic that Inden, who was one of Marriott's early and closest collaborators in their studies of caste, should now have so radically abjured, not only "caste," but his own not inconsequential work on it.)
(9.) Cf., e.g., the philosopher Martha C. Nussbaum's recent book, Women and Human Development (Cambridge: Cambridge Univ. Press, 2000)--one more attempt to solve an immemorial Indian social problem by appeal to "universal" (ch. 1) values, this time those of a late twentieth century (need we add, Westem) bourgeois feminism. If the values are "universal," it obviates any pressing need for particular reference to India's peculiar civilization or long history. Here we see the very inverse of the mentalite that finds expression in Marriott's work--not to speak of Dumont's, and marks them both as "essentialists"--along with, of course, all the other great philosophers of civilization in the Western tradition.
(10.) "On 'Constructing an Indian Sociology'," CIS, n.s., 25 (1991): 296. In this article, Marriott replies to comments made by others on the work under review that were published in CIS 24.2 (see n. 4, supra).
(11.) Personal communication, 1998. See also "Toward an Ethnosociology of South Asian Caste Systems," 233: "Persons and genera are thus conceived of as channeling and transforming heterogeneous, ever-flowing, changing substances." This idea is, as we will see, not only ubiquitous in Marriott's writings, but may be understood as a principle on which his work is based.
(12.) First discussed in Marriott's influential A.A.S. presentation, "The Open Hindu Person" (1980), a paper not yet available in an authorized published version.
(13.) See Philosophy of History, tr. J. Sibree (many editions): "Part I: The Oriental World," in three sections, "China," "India," and "Persia," representing the still implicit Absolute as "undifferentiated unity," as "differentiation," and as "light"--a single natural principle giving significance to infinite variety.
(14.) A point made well by Agehananda Bharati, to whose judgment disputes were frequently referred by his village hosts during his pilgrimage from Benares to Cape Comorin. (The Ochre Robe [Seattle: Univ. of Washington Press, 1962], 159ff.)
(15.) And so Marriott, speaking on behalf of himself and Inden: "We agree with him [Dumont] that a conception of the autonomous 'individual' does not validate Hindu ideas of society, Dumont stressing the containment of the individual in larger units, we the 'dividuality' of the person in the flow of social relationships" (JAS 36 : 190).
(16.) See note 5, above.
(17.) Cited from "Postulates for a Hindu Human Science," xerox of teaching materials, dated August 1996.
(18.) See "Freedom and History" (1952), recently republished in Freedom and History and Other Essays (Chicago: Univ. of Chicago Press, 1990), 160-241-but even more to the point is the succinct resume contained in the following essay (of which much will be made below), "Philosophic Semantics and Philosophic Inquiry," 247-49.
(19.) A goal patently underlying works such as Sanatana Dharma (Benares: Board of Trustees, Central Hindu College, 1903). The structure of this work certainly reflects the influence of M. M. Malaviya. Here we find Hinduism" presented as a coherent system of thought and ethics--a "brahmanical" reading of Hinduism strongly influenced by modernist reform movements who were seeking ways to respond to Western (scil., Christian) "systems" of thought. (The "Central Hindu College" was the precursor of Benares Hindu University.)
(20.) See, on these points, pp. 21-29; also, "Classes, Stages, and Others in Dharmasastra" and "The Five Pentads" (xeroxes of course materials, dated 1996 and provided by Marriott); some other quotations are selected from recent personal correspondence.
(21.) Adumbrated in the work under review, pp. 17ff., but developed more thoroughly in Marriott's reply to comments made on this work, in "On 'Constructing an Indian Sociology', 296ff.; also personal communications, 1997-98.
(22.) Indeed, some of the many applications of the "cube" in this volume are more than usually opaque--as when Marriott approves of Melinda Moore's ("The Kerala House as a Hindu Cosmos," pp. 169-202) location of the sacred "center" of the Hindu house at the exact center of the cube, where all the vectors are "in a perfect state of balance" (p. 29), at "(5, 5. 5)," graphically speaking, on a scale of 1 to 9. But since "purity" is one of the variables, this interpretation would seem to lead to the paradox that the "sacred" center is "half-pure"!
(23.) V. R. Potluri, Professor of Mathematics, Reed College (personal communication).
(24.) "On 'Constructing an Indian Sociology'," 298.
(25.) "Philosophic Semantics and Philosophic Inquiry," 247. (See n. 18, above.) No apology need be made for this apparent digression: McKeon himself observes: "The basic ambiguity of philosophic statement and discussion is not peculiar to philosophy. It is common to all discourse and to reflective inquiry in all fields. One of the tasks which has always been an inseparable or irresistible adjunct to philosophical speculation is the clarification of ambiguities" (ibid., 243).
(26.) "Freedom and History," 182.
(27.) "Philosophic Semantics and Philosophic Inquiry," 246.
(28.) Homo hierarchicus, 15. Italics in original. The citations in this and the following paragraphs, suitably translated, are taken from Homo hierarchicus, 13--17, of which these remarks purport to be a summary.
(29.) "Philosophic Semantics and Philosophic Inquiry," 250.
(30.) Physics 193a31 et seq.
(31.) "Philosophic Semantics and Philosophic Inquiry," 246.
(32.) Ibid., 250.
(33.) See note 19, above.
(34.) "Philosophic Semantics and Philosophic Inquiry," 250.
(35.) Ibid., 251.
(36.) From Mandeville to Marx: The Triumph of Economic Ideology (Chicago: Univ. of Chicago Press, 1977). (Bernard Mandeville [1670-1733] was one of the first to identify self-interest as the wellspring of economic activity.)
(37.) "Philosophic Semantics and Philosophic Inquiry," 246.
(38.) Ibid., 246.
(39.) "Freedom and History," 161. We may add here that this powerful "fashion" continues apace, as often noted above, both in the study of literature and history--witness the persistent drive to reduce the latter to "narratives" of one sort or another, and to treat the former exclusively as though it were an "activity" of the reader. Fortunately, "styles" change.
|Printer friendly Cite/link Email Feedback|
|Publication:||The Journal of the American Oriental Society|
|Date:||Jul 1, 2000|
|Previous Article:||COMMENTARY AND XIAOSHUO FICTION.|
|Next Article:||ASSYRIAN PROPHECIES, THE ASSYRIAN TREE, AND THE MESOPOTAMIAN ORIGINS OF JEWISH MONOTHEISM, GREEK PHILOSOPHY, CHRISTIAN THEOLOGY, GNOSTICISM, AND MUCH...|