The Religion & Culture Web Forum
November 2006
Featured Essay
Justification and Truth, Relativism and Pragmatism:
Reflections on Indian Philosophy and Its Lessons for Religious Studies
By Dan Arnold
Assistant Professor of the Philosophy of Religion
University of Chicago Divinity School
View a printable PDF version of this article (Requires the free Adobe Acrobat reader)
This essay contains occasional special characters which may not display properly in all browsers.
In Buddhists, Brahmins, and Belief, I elaborated and engaged the epistemological contributions of three trajectories of Indian philosophy, attempting to understand not only how their positions were developed in conversation with one another, but also what each might contribute to philosophical debates that are still very much alive. I concluded that the conceptual tools I had found useful in characterizing these various positions in play in first-millennium India could be usefully developed, as well, to address what seem to me some recurrent issues in the field of religious studies—and that those presently engaged more generally in the academic study of religion might therefore be (as I have been) instructed by some attention to the philosophical debates recorded in centuries-old Sanskrit texts. I argued, in particular, that reflection on these debates can give us some conceptual tools for addressing what is surely among the most pressing questions for students of religion: what sense are we to make of the fact that a great many seemingly rational persons have ardently held, as really true, religious beliefs that are often mutually exclusive? That question becomes especially acute when we realize that our answer to it cannot consist in simply jettisoning the idea of truth (and therefore, of rival claims as at least possibly true); for that is an idea that itself is necessarily presupposed by our work as scholars.
Insofar as I developed my positions on these matters in conversation with various Indian philosophers (and by taking as a guiding example the debates between them), it is useful to frame my proposal with reference to my readings of some Indian philosophical debates. The first of the Indian traditions that I considered was the foundationalist strand of Buddhist thought initiated by Dignāga (c. 480-540 CE) and Dharmakīrti (c.600-660 CE)—two thinkers who decisively influenced the course of Indian philosophy, and whose thought defined, for many traditional and modern interpreters alike, “the Buddhist position” in matters philosophical. I next took up what was arguably the most orthodox of the various Brahmanical schools: that of Pūrva Mīmāṃsā, whose constitutive concern was the interpretation and application of the Vedic literature—and particularly of the earlier parts of that corpus that are concerned with the performance of ritual sacrifice. The key figures here were Kumārila (roughly contemporaneous with Dharmakīrti) and his commentators, all of whom had compelling philosophical and ideological reasons for opposing the philosophical program of the Buddhists Dignāga and Dharmakīrti—and whose thought, on the interpretation that I defended, has some striking affinities with the contemporary trend of “reformed epistemology.”
Finally, I took up the Madhyamaka (“middle way”) school of Buddhist thought that stems from the pregnant but enigmatic and logically elusive works of Nāgārjuna (fl. c. 150 CE)—here, in the form advanced by the commentator Candrakīrti (also roughly contemporaneous with Dharmakīrti—this was an exciting period in Indian philosophy!). While Candrakīrti’s works seem to have represented something of a philosophical dead-end in the history of Indian philosophy, he now enjoys great renown, partly in proportion to the extent to which his works were later claimed by ascendant Tibetan traditions of Buddhism as uniquely authoritative—indeed, as advancing the definitive interpretation of Madhyamaka philosophy, which is almost unanimously thought by Tibetan scholars to represent the pinnacle of Buddhist thought. On my reading, though, it is especially interesting that Candrakīrti’s works should thus have risen to such prominence in Tibet; for while it is characteristic of some of the most influential of the Tibetan appropriations to wed Madhyamaka to the philosophical program of Dignāga and Dharmakīrti, I argued that Candrakīrti’s philosophically principled objections to that approach run so deep as to be integral to his thought.
In arguing that the Buddhist Candrakīrti, like the Mīmāṃsaka Kumārila, was thus a philosophical critic of his co-religionists Dignāga and Dharmakīrti, I was interested in appreciating the logically distinctive character of these different critiques of a foundationalist approach. My contention was that while the Mīmāṃsakas elaborated something like a “reformed epistemology,” Candrakīrti instead argued in terms that are best characterized as metaphysical—that, in particular, he ventured what can be understood as transcendental arguments, such as were meant to show that the epistemologist Dignāga’s demands for justification were themselves intelligible only given the truth of Candrakīrti’s claims.
All of these traditions are recognizably a part of the same historical conversation, and share not only a great deal of Sanskritic learning and conceptual vocabulary, but a great many discursive strategies. All are concerned, for example, with questions relating to (what is the principal term of art in much Indian philosophy) pramāṇas (“reliable warrants,” or criteria of knowledge): which ways of knowing (perceptual, inferential, linguistic, traditional) are to be admitted as such, how they have the status they do, and whether or not the only justified beliefs are those that can be shown to have been engendered by one of them. All are basically scholastic traditions of thought, their arguments developed and conceptual problems addressed within the framework of commentaries on the authoritative texts of a received tradition. And all evince a preoccupation with characteristically Sanskritic analyses of language, with the arguments often turning on such matters as the definitions and etymologies of key terms, the rules of the Sanskrit grammarians, and the analysis of actions on the model of semantically complete verbal constructions.
These are, then, all recognizably Indic traditions of philosophy, and understanding the textual artifacts of these traditions is, to a large extent, a matter of Sanskrit philology. But understanding these is also a philosophical matter; that is, it involves understanding not only the Sanskrit utterances in which they are recorded, but the logic and cogency of their arguments. Indeed, neither can finally be done without the other, as anyone who has spent significant time reading Sanskrit philosophical texts can attest; for most such people will have had the experience of puzzling long over some recalcitrant passage, only to discover that what had seemed a grammatical difficulty turns out to have been a conceptual one—that one had all along “understood” what the sentence said, but that its intelligibility emerges only as the logic of the argument becomes clear.
Thus, while the three trajectories of thought I engaged are commonly shaped by the world of first-millennium Sanskrit learning, each also reflects commitments that are specific to a particular ethical and axiological framework, and the arguments of these various traditions are therefore to be understood as logical developments of their framing commitments. Recognizing this does not, moreover, preclude our assessing the philosophical success of the arguments—which is to say, assessing the arguments in terms not only of their coherence with the traditional commitments they are meant to develop, but in terms of other things we know or believe. By doing so, I hoped to suggest, we may learn more not only about these Indic traditions of thought, but also about our own philosophical commitments as scholars.
A recurrent concern in the works of these three strands of Indian philosophy (and in my analysis thereof) was with the production of knowledge and belief; these various thinkers wrestled, that is, with the question of how (or whether) the causal description of one’s forming a belief is relevant to the (normative, justificatory) question of whether the belief is one that anybody ought to hold. That there is an age-old philosophical issue here is clear if we briefly consider two broadly different kinds of intuitions about how best to account for the objectivity of our knowledge. Thus, it could reasonably be thought that there is no better way to account for this than to attend carefully to the points at which our thought bumps up, as it were, against “the world.” This is the broadly empiricist intuition that the world is most basically “given” to us in causally efficacious “impingements by the world on a possessor of sensory capacities” (in the words of philosopher John McDowell)—and that we ought, therefore, to see our knowledge of the world as having its surest “foundations” in the perceptual experience that can be described in such causal terms.
Though it will seem counter-intuitive to those who (as perhaps most people these days do) have empiricist predilections, the broadly rationalist (or as I would prefer to say, following Robert Brandom, inferentialist) rejoinder to this is compelling: In thus taking causally describable, perceptual sensations as foundational, one inevitably ends up basing the objectivity of knowledge on something finally subjective—specifically, the subjectively occurrent representations (what a tree looks like to me, what one looks like to you) that happen to be produced by the world’s “impingements.”1 Against this, then, the best way to account for the objectivity of our knowledge is by attending to something intersubjectively available—in particular, to the discursive sphere of concepts that allow us to give expression to commonly held judgments. Only thus can we ask, in Robert Brandom’s words, “a question Kant taught us to ask: does the experience (or whatever) merely incline one (dispositionally)? Or does it justify one in making a claim, drawing a conclusion?”2
These are the terms that I had in mind, then, in elaborating and critically engaging the Buddhist foundationalism of Dignāga and his philosophical heirs, whose philosophical project I take (not uncontroversially) to be a basically empiricist one. There are, to be sure, good Buddhist reasons why philosophers such as Dignāga should thus have been inclined to privilege the causally describable faculty of perception; for this can seem a natural way to warrant (what all Buddhist philosophers are most intent on showing) the belief that there is no self—there is, that is, no such thing as “what we really are,” no enduring substratum of all our experiences, there are only causally continuous series of sensations.3 Thus, if one can argue (as Dharmakīrti does) that the only things that really exist are causally efficacious particulars—things with determinate spatio-temporal locations, which pointedly does not include things like “concepts” (you can chop down some particular tree, but you can’t make lumber out of “the fact of being a tree”)—and can argue, furthermore, that perceptual experience is unique in being caused by such things (when you see a tree, the object of that perception is among the causes of that cognitive event); then the only finally warranted beliefs will be those that are caused by these uniquely particular and evanescent events—which are, therefore, the only things we are finally warranted in believing to exist. And with that, we have the generally Buddhist claim that is finally advanced by this whole philosophical program: all that is finally warranted by the kind of cognition that is uniquely in contact with really existent phenomena is the conclusion that there are sensations—which does not also warrant the (inferential) belief that these must be the states of a “self.”
The problem, though, is that these causally precipitated moments of awareness are not themselves beliefs at all; beliefs are what they are (they have, as philosophers say, the content they do) only in virtue of their inferential relations to other beliefs—only, that is, in virtue of relations (such as “one thing’s being warranted in light of another,” in John McDowell’s phrase) that cannot be exhaustively described in causal terms.4 One can, then, only be said to be engaged in the activity of justifying a belief—only, that is, offering reasons why you ought to hold this belief, as opposed to explaining, for example, why I happen to5 —if what one is talking about is something that is at least possibly true. What is needed, then, is a way to understand that the beliefs at issue are about something more than just what’s appearing in my head—that they are, rather, about something in the world.
In the book, I made this point by characterizing Dignāga and Dharmakīrti’s program as entailing a finally epistemic conception of truth—a conception that takes the truth of beliefs to be somehow related to the fact of their being known. That is, Dignāga and Dharmakīrti’s appeal to perceptions as uniquely “constrained” by reality is ultimately an appeal simply to what happens to appear most “clear and distinct” to us. But while the clarity and distinctness (the vividness) of a conviction may very well explain why we happen to hold it, it does not necessarily represent a good reason why anyone ought to; for (in the words of John Henry Newman) “a proposition, be it ever so keenly apprehended, may be true or may be false.”6
To say this approach entails an epistemic conception of truth is not, to be sure, to say that perceptual cognitions cannot count towards the truth of beliefs, that they can’t be among our reasons for believing. Indeed, if the Mīmāṃsakas (as I understand them) are right, perceptions must (lest the whole world be blind) intrinsically confer justification to the same extent that any other pramāṇa does. It is surely a matter of empirical fact that we are justified in holding a great many beliefs about which we know nothing more than that we have them, and whose arising can be explained causally. The point, rather, is that Dignāga and Dharmakīrti cannot coherently claim that it is finally only to perception that we must appeal in (actively) justifying beliefs. That is, we may be entitled to consider a great many of our beliefs (including perceptual ones) really true; but what we are doing when we try to show our entitlement is something other than merely showing how we happen to have arrived at the belief.
It is with these arguments in mind that I characterized the Mīmāṃsaka and Mādhyamika critiques of Dignāga as compatible, unlike Dignāga’s program, with a realist conception of truth—one, that is, according to which something’s being true is logically independent of how or whether anyone happens to know it.7 In both cases, it is because of a suspicion particularly of Dignāga’s privileged category of perception that these other Indian philosophers were led to critique his Buddhist version of empiricist foundationalism. For proponents of the orthodox Mīmāṃsaka school of thought, the problem is that Dignāga’s ultimate appeal finally to the subjectively occurrent representations yielded by perception poses a particular threat to the Mīmāṃsaka vision of Vedic religion. For Mīmāṃsakas, the quintessentially religious goal of “dharma” consists in what Francis Clooney has characterized as a “decentering” of the human—one such that once a sponsor “has undertaken a certain sacrifice, the action is no longer governed by his viewpoint and desires. He himself is now part of a larger event not totally dependent on him.... The transcendence ‘occurs’ when the performer finds himself in a world which accounts for his viewpoint without making him the center of the world.”8
This overriding concern with the world as the locus of ritual action inclines Mīmāṃsakas to be particularly concerned to avoid the subjectivism that arguably follows from empiricism. Moreover, the characteristically Mīmāṃsaka account of the objective world involves a particular focus on the Vedic text as an eternal (hence, authorless) part of that world. Clearly, though, the Mīmāṃsaka confidence in the unique value of Vedic textual injunctions to ritual action is challenged by the empiricism of the Buddhist foundationalists; for a typical Vedic injunction (like “one desirous of heaven should perform the agnihotra sacrifice”) is not likely to generate much confidence to the extent that it is thought to require perceptual corroboration. Accordingly, in order for Mīmāṃsakas to advance their concerns on an epistemological front, it is important to undermine the view that perception is uniquely reliable—and indeed, to argue generally that there is no kind of cognition which can, simply in virtue of its being that kind, uniquely confer justification.
Kumārila and his philosophical heirs did this by taking the task of epistemology to be not the normative one of “raising the bar” that cognitions must clear, but that of describing what must be the case in order that we can have such knowledge as we generally believe ourselves already to be justified in claiming—what must be the case, as Kumārila’s commentators put it, if we are to avoid concluding that the whole world is blind. For if it is thought that we count as knowing something only if we can show (by pointing to putatively indubitable perceptions) how we know, then we would turn out to “know” precious little.
On my reading of Kumārila’s epistemology (which I developed by commending one commentator’s reading of his texts as better, on both interpretive and philosophical grounds, than another’s), the basic argument is one to the effect that justification is all the more that any cognition can confer—no cognition, that is, can coherently be thought to yield not only justification, but also truth; for one could advance such a demonstration only by appealing to further cognitions, none of which could reasonably be thought to confer something fundamentally different in kind than what was provided at the start.
But it is precisely insofar as a realist conception of truth virtually consists in recognizing the logical independence from one another of justification and truth—the logical independence, that is, of facts about our epistemic perspective from facts about what is true—that this strategy turns out in fact to be better able to accommodate the idea of truth than can the foundationalism of Dignāga and Dharmakīrti. Thus, against those for whom it is a problem to urge that the Mīmāṃsaka epistemology develops an understanding only of justification as contra one of truth (as J. N. Mohanty does when he carps that Kumārila’s “theory, I should think, is talking about truth and not merely of truth-claim”),9 we can rejoin that to be justified just is to be entitled to think one’s belief really true; and what more could we want than thus to be entitled to judge our beliefs true? Even if (counterfactually) one could demonstrate that one’s justified belief was also true, nothing further would be added, since having been justified was already to be entitled to think so—unless, perhaps, it is thought that showing a belief additionally to be “true” could consist in compelling the assent of all rational persons (and I take it as uncontroversial to say that this does not occur).
I argued, finally, that while the Buddhist Candrakīrti would surely never have countenanced the idea of enlisting the orthodox Brahmanical philosopher Kumārila as an ally, the foregoing arguments can be brought on board to inform what I understand as Candrakīrti’s transcendental arguments for the characteristically Mādhyamika elaboration of Buddhist commitments. To characterize Candrakīrti’s as transcendental arguments for a properly metaphysical claim—to understand him, that is, as arguing that any denial of the truth of his claims would be self-contradictory—is to fly in the face of the many interpreters for whom Madhyamaka is best understood as a radical sort of skepticism. But that understanding, I think, misses something important about the logically distinctive character of Candrakīrti’s arguments, and about the sense in which those are best understood as serving claims that are proposed as really true.
To unpack that a bit: the characteristically Mādhyamika claim is that “all dharmas are empty”—empty, that is, of an essence or “independent existence” (svabhāva). This is a claim that must be understood relative to (because it is constitutively framed against) earlier Buddhist talk about “dharmas.” In the schools of Buddhist philosophical thought that take their bearing from the so-called “Abhidharma” literature (which is arguably to say, most Buddhist schools), dharmas represent the “ontological primitives”—the most basic or (as Buddhists say) “ultimately existent” constituents of reality to which such merely “conventionally existent” things as selves can be reduced. Against this idea, Mādhyamikas like Nāgārjuna and Candrakīrti can be understood as arguing that the set of “ultimately existent” things is an empty set—a point that Mādhyamikas characteristically press by insisting on the “emptiness” not only of wholes like “persons,” but also of the analytic categories (dharmas) to which these are reduced in Abhidharma literature. Hence, the claim that “all dharmas are empty.”
In proceeding thus, Mādhyamikas can be understood to think that the ontologizing impulse of Abhidharma compromises the most important insight of the Buddhist tradition—which is, on the Mādhyamika reading, that all existents are “dependently originated” (pratītyasamutpanna).10 More precisely, Mādhyamikas can be said to have recognized that the ontological primitives posited by Abhidharma could have explanatory purchase only if they are posited as an exception to the rule that everything is dependently originated; that is, dependently originated existents could only be ultimately explained by something that does not itself require the same kind of explanation. But it is precisely the Mādhyamika point to emphasize that there is no exception to this rule; phenomena are dependently originated all the way down, and it is therefore impossible to specify precisely what it is upon which anything finally depends. Hence, there can be no set of “ultimately existent” things.
But that fact itself is proposed as ultimately true; “essencelessness” (niḥsvabhāvatā), as Candrakīrti often says, is itself the essence of things. To embrace this paradox, as Candrakırti clearly does, just is to say it is really true that “all dharmas are empty.” And it is the intelligibility of saying as much that I tried to advance by reconstructing Candrakırti’s as transcendental arguments in support of a constitutively metaphysical claim. Thus, against Dignāga’s demand that Candrakīrti show his claims to be warranted by some empirical means of justification (some pramāṇa),11 Candrakīrti argues that such a means of justification cannot coherently be demanded with respect to Candrakīrti’s claims regarding emptiness. This is because Dignāga’s project gains purchase only given its peculiarly technical use of ordinary words (words like “perceptual”), with this transformation of conventions serving a systematic re-description of our cognitive practices—serving, that is, an attempt to explain conventions by terms that are not themselves conventional.
But the whole point of Madhyamaka is that there is nothing that is not itself subject to the same constraints as our conventions—nothing, that is, that is not dependently originated. To the extent, then, that Dignāga’s demand for justification can reasonably be considered a demand precisely that we adduce something not dependently originated, that demand is itself a further example of precisely the problem to be overcome; hence, Candrakīrti argues instead by showing Dignāga’s demand itself incoherent, insofar as it must presuppose the very conventions that exemplify the truth of Candrakīrti’s claims. And (here supplementing this reconstruction with an appeal to the lesson gleaned from considering the Mīmāṃsakas) among the discursive conventions that Candrakīrti can thus think are necessarily presupposed is that there is a difference between “what we think” and “what is true.”
But despite our necessarily presupposing this distinction, it is one that is partially overcome—indeed, overcome to the only extent possible for those of us who are not Buddhas—in being justified; for to be justified just is to be entitled to think that the belief in question really is true. As Candrakīrti says (invoking a famously recurrent Buddhist scriptural reference), all dharmas are empty whether or not there is a Buddha to be aware of that fact12 —that is, quite independent of the fact of anyone’s knowing it. But in that case, the difference between Dignāga and Candrakīrti is not best understood in terms of one’s having beliefs and one’s not; rather, it is that the content of Candrakīrti’s beliefs is such as to require a different kind of justification—and that precisely in virtue of their being proposed as really true.
***
Attention to the distinction between truth and justification—the appreciation of which just is what distinguishes a realist from an epistemic conception of truth—is, I think, helpful not only in expressing what I hope are some promising ways to think about the philosophical contributions of Madhyamaka and Mīmāṃsā, but also in situating this discussion within the larger context of the field of “religious studies.” I would like to suggest, then, that getting clear on this distinction can dispel some important confusions that recurrently surface within the field of religious studies (as in the humanities more generally)—by suggesting that, in the phrase of the pragmatists, the difference between truth and justification is one that makes a difference. In particular, I would commend the realization that many of the theoretical and philosophical projects that have influenced the field of religious studies are to be understood as concerning only justification—and that their possible contributions are compromised to the extent that they are taken instead to concern (often by arguing against the relevance or possibility of) truth.
Thus, we can argue, with Bruce Lincoln, that constitutively “religious” discourse aims above all to efface its own origins in the interests of particular people, “giving an historical intention a natural justification, and making contingency appear eternal”13 —and that we ought therefore to be alert to the ways in which religious discourse is eminently a matter of power relations. Or we can argue, with George Lindbeck (whose work owes much to Wittgenstein), that “the proper way to determine what ‘God’ signifies ... is by examining how the word operates within a religion and thereby shapes reality and experience rather than by first establishing its propositional or experiential meaning and reinterpreting or reformulating its uses accordingly”14 —that, in other words, being religious cannot be thought to consist simply in assent to propositional claims, insofar as the latter are only intelligible to those who already know the “grammar” of the faith. Or we can ask, with Talal Asad, how “(religious) power create[s] (religious) truth,” emphasizing St. Augustine’s view that “coercion was a condition for the realization of truth, and discipline essential to its maintenance.... It was not the mind that moved spontaneously to religious truth, but power that created the conditions for experiencing that truth.”15 Or we can, as students of Indian philosophy are to be commended for having recently begun to do, follow Pierre Hadot in appreciating that for many if not most pre-Enlightenment philosophers, the point of engaging in philosophical discourse was “not so much to inform the reader of a doctrinal content but to form him, to make him traverse a certain itinerary in the course of which he will make spiritual progress”16 —and that philosophical arguments will be understood rather differently if they are thus taken as the artifacts of a “way of life.”
All of these theoretical projects can surely be thought helpfully to describe real aspects of the epistemic situations relative to which humans have been and are apt to form their beliefs (and to be justified in holding at least some of them). They describe, that is, aspects of the nexus of power and contestation, psychosis and fear, love and hope that shape us as holding the beliefs we do, and that give us (for better or for worse) our intuitions about what count as good reasons and arguments for belief. None of these projects, however, is rightly understood to preclude consideration of the possible truth of the beliefs thus formed—none of these, that is, renders unintelligible the distinction between what we think and what is true. It may indeed be the case, for example, that the characteristically Mīmāṃsaka appeal to the “transcendence” (apauruṣeyatva) of the Vedas functions to “naturalize” Brahmanical claims to authority and power—just as it may indeed be the case that one must have ritually concluded a period of celibate study with a bath before one can properly have the “desire to know dharma” (dharmajijñāsā) that motivates Mīmāṃsaka study,17 or that Mādhyamika teachings regarding emptiness can finally be understood only by those who have first wept at the thought of the Buddha’s fathomless compassion.18
But these various facts (if such they be) concern only the justification of belief—only, that is, the various circumstances in which a person might be constituted as someone for whom certain beliefs are rationally held. How and why the beliefs in question were thus developed is, however, logically independent of whether or not they might be true—a distinction that is given up only at the cost of denying that we understand the difference between “what is said or thought and what it is said or thought about.”19 Just as acknowledging something as “true” does not make it thus, so, too, a belief is not necessarily false (and the question of truth not superfluous) simply because its acceptance is, in any of the ways described by these and other theories, historically contingent—otherwise, we could not be said to hold any true beliefs at all (not even the belief that at least one of these theories is right), since of course the discovery of any truth can only take place in history.
It is important to note, though, that thus to say the question of possible truth is not precluded is not to say that all of the various beliefs that may be justified are true20 —only that whether or not they are is a logically distinct question. To recognize the distinctions I am commending, then, is not to forfeit the possibility of judging some beliefs true and some false—or even of arguing that some are more rationally held than at least some others. The importance of these points emerges, I think, from consideration of a provocative article recently published in Harper’s by Stanley Fish.21 Characteristically venturing into the heart of controversy, Fish engaged the question of whether, as some had argued, the events of September 11 advanced any position in the “Culture Wars”—and whether, in particular, these events undermined the cogency of characteristically “postmodern” thought by showing the moral impossibility of relativism. Surely, that argument goes, here were events so monstrous that nobody could ethically or reasonably judge them to have been guided by any sort of rationality.
Fish argues, among other things, that the perpetrators of such events are not helpfully judged to be simply “irrational,” and that any condemnation of them necessarily issues from the perspective of some commitments. Significantly, though, Fish states his point in this regard conditionally: “if by ‘a reliable condemnation’ [of a rival perspective] you mean a condemnation rooted in values, priorities, and a sense of right and wrong that no one would dispute and everyone accepts, then there is no such condemnation, for the simple reason that there are no universally accepted values, priorities, and moral convictions. If there were, there would be no deep disputes.”22 Fish’s point, though, is to deny the antecedent of this conditional—this is not what we ought to mean by “condemnation,” which must instead be understood always to presuppose commitments that are at least in principle disputable.
This can, however, seem like an expression of morally vacuous relativism only if we ignore the following, highly significant clarification:
I am not saying that there are no universal values or no truths independent of particular perspectives. I affirm both. When I offer a reading of a poem or pronounce on a case in First Amendment law, I do so with no epistemological reservations. I regard my reading as true—not provisionally true, or true for my reference group only, but true. I am as certain of that as I am of the fact that I may very well be unable to persuade others, no less educated or credentialed than I, of the truth so perspicuous to me. And here is a point that is often missed, the independence from each other, and therefore the compatibility, of two assertions thought to be contradictory when made by the same person: (1) I believe X to be true and (2) I believe that there is no mechanism, procedure, calculus, test, by which the truth of X can be necessarily demonstrated to any sane person who has come to a different conclusion (not that such a demonstration can never be successful, only that its success is contingent and not necessary). In order to assert something and mean it without qualification, I of course have to believe that it is true, but I don’t have to believe that I could demonstrate its truth to all rational persons. The claim that something is universal and the acknowledgment that I couldn’t necessarily prove it are logically independent of each other. (34)
Fish’s point, in the terms I have been suggesting, is simply that truth is logically independent of justification—recognizing which, we can judge the beliefs of others to be false (and indeed, can condemn them strenuously), while nevertheless appreciating that their holding them does not, ipso facto, show them to be irrational—while appreciating, in other words, that we may fail to persuade them.
To think otherwise—to think, for example, that the “objectivity” of true beliefs consists in the likelihood of their compelling the assent of all rational persons—is to forfeit a realist conception of truth; for this just is to think that what causes a belief is at the same time what makes it true. As Frege recognized, though, that way lies solipsism; for all manner of subjective facts (psychological, socio-economic, and neurological facts specific to the situation of the knower) engender beliefs. The objectivity of beliefs has to do, rather, with their being intersubjectively available—with their being, in other words, framed in language, and at least possibly expressed and tested for their inferential consequences in the eminently social game of exchanging reasons. Indeed, objectivity virtually consists in such intersubjectivity. That is, the “conditions of the possibility” of being justified are, as Wittgenstein effectively urged in arguing against the possibility of a private language, never simply willed by individual agents—we do not choose what reasons will be found compelling (even by ourselves) in any context.
But if we appreciate the distinction between truth and justification, there is room to recognize that there is yet a further element of objectivity to our beliefs; for the possibility that beliefs thus circumstantially and socially justified might also be true brings into play something (truth) altogether objective (that is, independent of our perspective as knowers)—even though our epistemic situation in this sublunary world will allow nothing more than that we ever be justified in thinking our beliefs really true, and never “knowing” them to be so. But in that case, it is only reasonable to think that “knowing” consists in justifiably thinking one’s beliefs true; for otherwise, nobody could be said to “know” anything, and we would be left without a use for a perfectly ordinary word—and without a use for a concept (truth) that we presuppose whenever we so much as think it possible to be mistaken about something.
These points have been lucidly developed by Jeffrey Stout, who appropriates the work of Robert Brandom particularly for its value in religious, theological, and ethical inquiry. Stout recognizes, then, that justification is context-sensitive: “...affirming that many of us are justified in holding some of the (nontrivial) moral beliefs we hold is not the same thing as affirming that somebody has established a set of (nontrivial) moral beliefs that any human being or rational agent, regardless of context, would be justified in accepting.”23 Stout further recognizes, with the Mīmāṃsakas, that there is a crucial distinction “between being entitled to a belief and being able to justify that belief to someone else” (87). He elaborates these points particularly with reference to the American tradition of pragmatism, emphasizing the extent to which justifying a claim is an activity. As such, its success is to be gauged in terms of the difference it makes: “In what, then, does the success of a justification consist? In eliminating relevant reasons for doubting that P. What reasons for doubting P are relevant and what suffices for their elimination? That depends on context, in particular, on the people to whom the justification is addressed” (234-35).
But Stout recognizes, with Brandom, that while it thus affords rich resources for explaining the circumstantial character of justification, pragmatism fails as a conception of truth, which is logically distinct from the question of how we may know it:
Truth pertains to the conceptual content of a claim, not the epistemic responsibility of the person who accepts or asserts it. Truth, or accuracy, is an objective status as well as a normative one.... whether our beliefs and claims actually enjoy the status of being true is not up to us. Believing that someone has a particular obligation, right, or virtue does not make it so. Truth-talk has a place wherever we take the subject matter under discussion—and not simply the evidence pertaining to it—as the object of our inquiry. By engaging in truth-talk, we implicitly view our subject matter as something we might get wrong, despite our best cognitive efforts.24
Recognizing this makes it possible to commend what is surely the honorable impulse behind relativism: viz., the belief that there is no single way of looking at the world that is self-evidently more rationally held than all others. But this can now be rightly understood as a point about the circumstantial character of justification. We can retain, then, the relativist’s recognition that many different (even mutually exclusive) beliefs might alike be rationally held—but only if we also recognize that this point becomes incoherent if understood as concerning the truth of beliefs. A relativist conception of truth, Stout rightly urges, “erases disagreement among groups rather than making it intelligible” (2004: 238). Only by making disagreement intelligible is it possible to respect the beliefs of others enough to appreciate that they are considered really true; and to appreciate this is, ipso facto, to disagree with them where they seem to contradict one’s own commitments.25
To urge that we necessarily have recourse to talk of “truth” is, though, not to claim to be in possession of the truth, which is something that committed relativists seem typically to suspect whenever the word “truth” so much as rears its head; indeed, quite the opposite. Thus, critiques of the idea of a “transcendental perspective” (of a “God’s eye view”) can be recognized and acknowledged even, for example, by those who offer transcendental arguments—provided that we understand such critiques to pertain to our epistemic situation, to the circumstances of our being justified, and not to truth. Recognizing the validity of such critiques—the validity, for example, of the pragmatist account of the context-sensitivity of justification—the proponent of a transcendental argument can acknowledge that the fact of her offering the argument is not understood to reflect her being in possession of the “transcendental truth,” or her occupying a transcendental perspective. The argument is, rather, only a way of justifying the belief that such and such a condition really does obtain. The content of the most important beliefs thus credited as true may, indeed, itself be precisely such as to relativize our perspectives as knowers; “the rhetoric of a higher law is little more than an imaginative embellishment of the gap between the concepts of truth and justification, between the content of an ideal ethics and what we are currently justified in believing.”26
Recognizing the distinction between truth and justification gives us the conceptual resources to describe religious people as (among many other things) thinking their beliefs really true (and correspondingly, of thinking contradictory beliefs really false)—and to appreciate, moreover, that the possibility that their justified beliefs are really true may never finally be eliminated, whatever other explanatory or theoretical interests we may have. What this distinction gives us more generally, as human persons with beliefs and commitments, is a way to explain the possibility of saying people are wrong, without necessarily judging them thereby to be irrational—and insofar as we can only understand those commitments that, as rational, are possibly intelligible to us, it is surely imperative that we be in a position to do this. It is, then, as we should expect that while the arguments of Kumārila and Candrakīrti represent cogent critiques of Dignāga’s foundationalist demand for justification, they should not necessarily compel our acceptance of their beliefs. Certainly, it is indisputably the case that, as an empirical matter, no arguments in history have given all rational persons good reason to adopt their conclusions, at least if we are to judge by whether or not they have succeeded, rhetorically, in persuading all who have received them.
But to say this is not to make a point that can properly be held against these (or any other) arguments.27 One can only fault arguments for failing to persuade people (or, conversely, fault believers for not having arrived at their beliefs as the result of assent to arguments) given the view that what arguments are for is to produce beliefs. Against this, we should understand arguments as meant, rather, to justify beliefs. We can then conclude without contradiction that Mīmāṃsakas like Kumārila and Mādhyamikas like Candrakīrti have cogently argued that their beliefs are rationally held, and that they are entitled moreover to consider those beliefs true—and yet, just as rationally opt not to adopt them as our own.
Adapted from Buddhists, Brahmins, and Belief: Epistemology in South Asian Philosophy of Religion by Dan Arnold. Copyright © 2005 Columbia University Press. Reprinted with the permission of Columbia University Press.
The Religion and Culture Web Forum comes from the Martin Marty Center at the University of Chicago Divinity School. Persons seeking to copy or use this material must obtain permission from the Divinity School. 1025 East 58th Street, Chicago, IL 60637. (773) 702-8230.
Read the
invited commentary from Michael Kremer and Ronney Mourad
Home | A-Z
Index | Search | Directories
| Contact Us | The
Divinity School | UChicago
1025 E. 58th St., Chicago, IL 60637
tel: 773-702-7049 fax: 773-702-8223