Radically Pluralist, Thoroughly Critical: A New Theory of Religions

This is pre-publication version of an article puglished in in the Journal of the American Academy of Religion, 60 n. 4 (Winter 1992) pp. 693-715.]

This paper is about a dilemma one faces in trying to construct a genuinely pluralist framework for the study of religions. On the one hand, at first glance it does not seem possible to have a critical theory of religions unless one assumes that there is one set of beliefs (at least a hypothetical set not yet in existence) which represents “the truth” about the subject matter dealt with by religions. If one does not assume, at least in principle, that the truth in this area is single, then on first sight it appears that religious claims cannot be treated as cognitive claims, and one cannot evaluate them as to their validity.
On the other hand, the assumption that religious truth is single has implications for the interpretation of religions that undermine a thoroughgoing pluralism. That is, on this assumption one must assume that, if Christianity and Buddhism for example are both at least partially valid religions, then a Buddhist trying to understand what is most valid in Buddhism and a Christian trying to understand what is most valid in Christianity must be converging toward a single set of truths. But on the other hand, it seems at least prima facie plausible that particular religions are like local cultures: What makes Irish culture a great culture is not necessarily determined by what Irish culture has in common with Russian culture. Similarly, it seems at least plausible that discovering what makes Christianity and Buddhism great religions might not be the same as discovering what Christianity and Buddhism have in common.
The difficulty then which this paper tries to tackle, is to formulate a general theory of religions that is both critical and pluralist. By “critical” I mean a theory that assumes religion to be cognitive. It is not like the rules of football. To be “mistaken” about the rules of football is only to be mistaken about what some other persons decide are the rules of football, not about what these rules “really are”, independently of such decisions. (There is no reality which could serve as a basis for criticizing common conceptions, or the decisions of some group, as to what the rules are. Human decisions are completely constitutive of the rules.) A critical theory of religion, on the other hand, assumes that knowledge of Tao or of Brahman is knowledge of some reality. People can be mistaken about Tao itself, not just about what someone thinks of Tao. One can criticize some given conception of Tao or Brahman, even if it is widely accepted by people devoted to Tao or Brahman, on the grounds that this conception is not adequate to their reality. A critical theory of this kind differs from neo-Wittgenstinian theories like that of D.Z. Phillips, as well as from Ninian Smart’s approach, which “brackets” truth-claims (20-23, 37-73). A critical theory must hold that something must be true in order for a given set of religious beliefs to be well-founded, and it must say specifically what it is that must be true and how one might know it to be true. It must be careful to avoid the charge which Proudfoot’s book Religious Experience makes against many modern theories of religion: that they are merely protective strategies allowing believers to make important claims while avoiding any real test of the warrants for the claims. A critical theory is like the Sachkritik advocated by Bultmann and Barth (Robinson: 30-34), except that here it proceeds on a pluralist basis.
A radically “pluralist” theory, on the other hand, is one that allows for the possibility that Tao and Brahman can each be genuinely real and valid ultimate foundations for religions, and yet (contrary to John Hick and W.C. Smith) be genuinely and ultimately different from each other. A thoroughgoing pluralism is one that does not require ahead of time the assumption that religious beliefs concern different manifestations of the same reality (Hick), or that they are attempts to represent a single reality which religious people can come to understand better through dialogue with each other (Smith).
Anyone can see that there are quite formidable theoretical difficulties in holding a theory of religions that is both critical and pluralist in the strong senses just described. It is these theoretical difficulties, at the most fundamental and abstract level, that I will be addressing in this paper. The fundamental topics which must be treated can be briefly outlined as follows:
1-If the truth is not single in some given area, can we still speak of constraints which reality places on belief? I will argue that we can, and point to some developments in modern mathematics as an illustrative example.
2-What is the nature of the constraints on valid religious beliefs? I will argue that these constraints do not derive from reality as it is in itself apart from human consciousness, but from phainomena, reality as it appears in human experience. In this respect the constraints on religious beliefs differ from the constraints governing thought in the physical sciences, and require a different mode of critical investigation.
3-This second point brings up the issue as to whether phainomena in general are only “creations of the human mind”, which can therefore make no claims on this same mind. I hope to present a plausible theory to the contrary, and to indicate some general lines along which such a theory could be defended.
4-Finally, there is the issue as to the “contextual” nature of all phainomena, including those that are the objects of religious belief — the fact that the essential content of most phainomena is determined in its essence by the place of these phainomena in a total cultural “system”. Here again the issue is whether one can accept a fully contextual theory of religions, while maintaining that religions are about ultimate norms that make a claim on us — they are not only “functionally” useful in fulfilling social or psychological ends external to themselves.
There are two limitations on the ambitions of this paper that must be emphasized. First, these proposals do not themselves constitute a defense of any particular set of religious beliefs, nor even of religious belief in general as opposed to “secular” worldviews. They are only meant to provide a fundamental framework for critical discussion of religions. While the framework aims to allow for the theoretical possibility that there can be more than one valid religion, it also aims to prejudice as little as possible the outcome of critical discussion within the framework. Thus the framework itself does not assume that there is any religion that is actually valid. (It does not assume even a minimal “confessional” stance of the kind rightly criticized by Donald Wiebe in his debate with Charles Davis.) It only proposes a very general theory as to what kinds of evidence and argumentation one would have to offer to support the validity of some given religion — or of any other view of ultimate norms.
The theory proposed is, broadly speaking, an “experiential” theory of religion. That is I do not take the task of critical thought to be the establishment of some “necessary” truths about religion (unlike David Tracy: 172-203, rightly criticized by F. Schüssler-Fiorenza: 298-300). Rather, I take experience to be absolutely primary in the origin of well-founded religious beliefs. Critical thought is strictly a second-order activity reflecting on experience. It aims only to examine the contents of the experiences that originate religious beliefs, and to see whether what is experienced actually provides an adequate basis for the beliefs in question, and especially for the conclusions which are drawn from these beliefs.
The second limitation of the paper has to do with its strictly “programmatic” nature. Given the space restrictions, and the controversial nature of the complex issues that need discussing, I cannot hope to provide fully persuasive argumentation for all the positions I will take. That is, I will not take it as my task actually to defend a critical-pluralist theory of religion. Rather, the task I am taking on is simply to set forth in detail a set of theoretical positions which would make possible what seems at first sight implausible: a theory which allows substantive criticism of religions while not requiring singleness of religious truth. I hope to propose positions that, if they could be defended, would solve the most serious foundational problems such a theory must face. I hope also to show that these positions are plausible, and indicate some general lines along which I think they could be defended. But their actual defense cannot be attempted in this short space, so I must beg the reader’s indulgence in advance for what will seem at times a somewhat dogmatic way of proceeding.

REALITY-CONSTRAINTS IN THE ABSENCE OF SINGLENESS OF TRUTH: THE CASE OF MATHEMATICS.
One of the main difficulties in the way of the formulation of a critical pluralist view in any field of knowledge today is that we are living now “in the shadow of the absolute”. We are living at the end of an age in which a primary ambition of almost all serious thinkers has been to provide some standard or criterion, by which a definitive decision could be made in favor of one particular belief among various competing beliefs on any major question. This ambition determined the horizon of thought for so long, that it came to seem obvious that the only alternative to providing such a definitive criterion or standard was a complete skeptical relativism in which literally anything goes — in which there can be no discrimination at all between competing beliefs. Part of the purpose of this paper is to sketch a third alternative. And to begin I would like to offer some observations about modern mathematics, as a case in which the collapse of belief in absolute foundations does not necessarily lead to a completely anarchistic relativism — a relativism in which everyone believes whatever she wants to believe, with no criterion whatsoever for discriminating between well-founded and ill-founded beliefs.
Euclidean geometry once formed the paradigm of completely certain knowledge: A single, unified system of theorems derived by indubitable logic from a small set of supposedly self-evident truths. The last two centuries has seen the gradual erosion of this faith in geometry, and in mathematics generally, as a single system of certain truths. The results of this erosion, what one can still reasonably hold about the nature of mathematics itself, seems still to be a matter of some debate For purposes of the present paper, I would like simply to imagine here what would be, for the traditionalist mathematical true believer, a kind of “worst case” version of the situation of mathematics today. Let us suppose the following:
Mathematics is not about the material world at all, but explores the logical properties of abstract objects. Abstract entities, and entire systems of mathematics, can exist which have no correlates in, or application to, the material world. Further, the abstract entities which mathematics investigates do not form a single, internally unified and consistent set which constitutes “the truth” which all mathematics must give an account of. All abstract entities have their existence relative to some system of logical terms and postulates, and there is no single set of terms and postulates which can claim to be more “self-evident” than the others, and can claim to be clearly superior to all others in generating a mathematical system free of contradictions. Mathematics consists then of arbitrarily positing various fundamental sets of axioms, without regard for their “true” or “self-evident” character, and exploring the properties of the various systems of theorems that can be generated by doing this.
One could then ask the question: Would this kind of mathematics be “nothing but the creation of the human mind”? What I want to argue here is that, while there is some truth to this statement, it is a gross exaggeration. The following observations of Stephen Barker, for example, would still apply:
“The view that mathematical objects such as numbers and sets are creatures of the mind, abstract entities brought into being by thinking, is a view that many people have found attractive… When the mathematician thinks to himself, “let it be postulated that there are numbers of such-and-such a kind,” he thereby brings them into being, his sovereign creative power being like that of an omnipotent Deity who creates out of nothing whatever he wills to be.
It would be too extreme, however, to imagine that the mathematician is entirely free of restrictions in this activity…. Mathematicians…are subject to the requirements of consistency, and cannot bring into being self-contradictions. For example, suppose someone attempts to postulate the existence of an entity answering to the description “A natural number which is the cardinal number of the set of natural numbers.” It might at first sight appear that a mathematician could postulate such an entity if he likes. Yet if someone tries to join the assumption that there is such a thing to the normal axioms for the natural numbers, inconsistency results… [The number invoked would have to be both finite and infinite.] An attempted creative fiat like this would be unsuccessful at bringing its object into the world.(72-73)
If something is completely a creation of the human mind, then there is nothing one can genuinely “discover”, nothing that can resist the mind’s free decisions, nothing one can be mistaken about. But this would not be true even in the kind of completely “foundationless” and pluralist mathematics described above.
First of all, not just any set of primitive terms and fundamental axioms will give rise to a consistent system of theorems. There are even more constraints if one wants to generate a system which is logically interesting: a system which generates a complex array of theorems, whose logical relation to each other is exhibited with great clarity by the relatively small number of axioms from which all can be derived. In Barker’s words again, such a system needs to “employ a comparatively economical set of primitive terms and axioms and select these so that an appropriately rich array of theorems can be deduced…If the axioms are too few…then the theorems deducible…will be insufficient to make the system interesting…[But] a system that lacks economy cannot give us much insight into the logical connection of its sentences.”(25). One cannot bring such systems into being by fiat. Some foundations will actually work and some will not. Which ones will work better and which will work not so well is a matter of discovery, not free creation.
Secondly, once primitive terms and fundamental axioms are chosen, what system of theorems follows from them, and what entities and operations have a valid existence in the system, is not a matter that the creator of the system can freely decide (or even initially know about in very many cases).
The case of “division by zero” in ordinary arithmetic is an interesting example for reflection. This is an operation that is “not allowed” in arithmetic. One can say: It has no existence in the system of ordinary arithmetic. Although the symbols are available to formulate the sentence “five divided by zero”, this statement represents a case of Wittgenstinian “language on a holiday”. No abstract entity (or complex of entities) exists to which it could refer.
What serves as evidence for the existence/non-existence of this operation? In an earlier age one might have said that it is not allowed because one could not imagine any manipulation of objects in the real world that would correspond to dividing something by zero. In the mathematics described above, this is not a legitimate argument. Operations and concepts are allowed that have no real-world correlates. The crucial test is whether allowing division by zero would lead to large-scale ambiguities and/or contradictions. A mathematical operation has to be defined in such a way that its result is uniquely determined — definite, precise, and certain, not a matter of opinion. And it has to fit into a given system of mathematical objects and operations in such a way that it does not lead to large-scale contradictions.
Division by zero cannot meet these two tests together (let us at least suppose this for the present argument), without fundamental alterations in the entire system of arithmetic. It is on these grounds that one could say that “division by zero” is a “meaningless” phrase in arithmetic.
Consider on the other hand positive cases, cases in which mathematicians have successfully introduced new concepts, such as Georg Cantor’s introduction of the concept that there are ascending orders of infinity (infinite sets of various sizes. See Kline:199-204; Wilder: 111-148). Support for this notion could not come from intuited certainties or from real-world experiments. But neither could the notion be introduced into mathematics simply by fiat. Cantor convinced his mathematical colleagues by producing rigorous logical arguments in support of this concept, (for example by showing that the infinite set of real numbers could not be placed in one-to-one correspondence with the infinite set of natural numbers, and concluding from this that the former is larger than the latter). He also had to work out an arithmetic for these variable-sized infinities which could quantify the relations between them, and produce unambiguous and non-contradictory results.
Did Cantor mentally “create” infinite sets of variable sizes, or did he “discover” them? Were they “out there”, “existing” before he thought about them? As thing-like entities spacially “out there”, of course not. Nor (according to the mathematics proposed above) do such abstract entities exist in all possible mathematical systems. But clearly, within the logical postulates within which Cantor was working, certain logical facts concerning variable-sized infinities were true before he discovered their truth — and these truths are what warrants treating ascending orders of infinities as valid objects in mathematics. Cantor did not just create a new set of symbols and assign them a meaning which he arbitrarily decided upon. Unlike the case of division by zero, something independent of Cantor’s mind was there to serve as the meaning of his notation. Even though no one knew this before Cantor’s discoveries, it turns out that one is not logically free to admit the postulates within which Cantor worked, and deny the existence of ascending orders of infinity.
The foundationless and pluralist mathematics described above would not invalidate the great “discoveries” of mathematics that have taken place at the cost of great labor in the last three centuries. It would not reduce them to arbitrary inventions which anyone could easily do on their back porch of a Sunday evening. On the contrary, if the picture of mathematics set out here has any plausibility, it is only because rigorous adherence to mathematical epistemology — to the reality-constraints proper to mathematical reasoning — may have turned out (contrary to the intentions of mathematicians) to undermine the belief in singleness of truth in mathematics.
. CATEGOREAL REGIONS.
One could generalize from these observations as follows: The mathematics described above would still have an epistemology, a set of assumptions as to what counts as good evidence and good argumentation for and against given mathematical theses. Something has to be shown to be true in order to show that ascending orders of infinities exist. This epistemology is by and large the epistemology actually used by modern mathematicians in their discussions and debates on particular issues.
These observations about mathematics could be made, mutatis mutandis about modern physics. One could pose again the possibility that a complete impasse has now occurred, preventing the carrying out of the ambition once aroused by Newtonian physics – a single and complete system giving us “the truth” about the physical world. Suppose it can be shown that theories in physics are inescapably and seriously underdetermined by empirical data. There will always be several theories that are equally supported by the data, not only in specific areas, but regarding physics as a whole. Or, to pose a related possibility equally undermining of the Newtonian ambition: Suppose, as many now hold, there are no uninterpreted “data” in science. What counts as data, and the most primitive terms describing the data, are relative to the theory one is operating with. And there is no single, unified theory which can embrace all of physics and be shown to be superior to all the rest.
Again, one who held this view of the present situation in physics would be tempted to draw radically skeptical conclusions, and to say that a modern physicist’s choice of fundamental categories is completely arbitrary. But this is not true. Physics has a basic epistemology: That is, theories in physics must be able to predict unambiguously the observable results that will follow from specific alterations in physical systems, and to trace in detail the causal networks and sequences that lead from the initial alteration to the results (Taylor:49-54). Some sets of categories and theories are far more suitable than others for this purpose. It is obvious, for example, that the table of elements in modern chemistry represents a much better set of categories for this purpose than the set of categories used by medieval alchemists. Again, abandoning the position that there is one single correct set of categories or theories for thought about causal relationships does not automatically mean that reality places no constraints at all on the categories which are suitable for this kind of thought.
There is an epistemology in physics, reality-constraints which it operates under. Something must be shown to be true in order to postulate the existence of particles without place, causation at a distance, etc. And this epistemology can discriminate between less adequate and more adequate sets of categories and theories in physics, even if it turns out that it cannot narrow the field down to one uniquely valid set.
Further, the constraints on our beliefs are different for physics than they are for mathematics. Even though modern physics is of course heavily dependent on mathematics, pure mathematics operates under different kinds of constraints than physics. Categories in mathematics are not subject to the constraint that they be useful in predicting cause-effect correlations in our world, so there can be valid mathematical categories and systems that have no physical application whatsoever.
But this is also a limitation on the range of application of mathematical theory. If it is unconstrained by the physical world, then neither can it, by itself, yield truths about the physical world. For example: When mathematics was still thought to yield necessary truths about the physical world, thinkers puzzled about the use of negative numbers (Kline: 108-110) and infinite numbers (199-200) in mathematics — how could these categories be legitimate if negative and infinite quantities do not exist? If we say that mathematical thought takes place under different constraints, this difficulty disappears.
But this is also a limitation on the meaning of mathematical terms, and the valid range of application of mathematical theories. A completely formalized mathematics must give up its claim to deliver to us, by itself, necessary truths about the real world.
These observations can be generalized in terms of a theory of “categoreal regions”, which can serve as a substitute for now widely questioned “metaphysical” theories about necessary, universally valid sets of categories and truths that would obtain in all possible worlds. There may be no “necessary” set of mathematical categories and fundamental axioms, but there are constraints on the contents of categories used in mathematics, deriving from the epistemology of mathematics and the realities with which it deals. Only categories with certain kinds of contents can be given meaning in mathematical thought, the kinds of content that can be made parts of some given mathematical system through adherence to the epistemology of mathematics. This is what constitutes mathematics as a “categoreal region”.
Likewise there are certain reality-constraints operative in physics, and certain kinds of category-contents that are in accord with these constraints. Insofar as the constraints, and the correlative allowable category-contents, are different in physics from what they are in mathematics, physics constitutes a separate “categoreal region”.
A key principle related to this notion of categoreal regions is that the character of the reality-constraints on categories and theories determines both their meaning and the valid range of their application. And conversely the intended meaning and the desired range of application of a theory determines the kind of reality-constraints to which it is subject. The “range of application” of a theory has to do with questions about the valid conclusions one can draw from that theory: If we know that x theory is true, what else do we know? For example, if one wants to know truths about what causes what in the real world — if this is the desired “range of application” of one’s theories — then one’s categories and theories must be constrained by tests showing powers of predicting observable outcomes.
This principle of the correlation between constraints and range of application in any categoreal region, makes explicit the limitations on the knowledge to be gained in any particular field of inquiry. To illustrate: The category “substance” has become problematic in modern science. Categories used in connection with scientific experiment must describe properties that will make a difference in the observable output of causal networks. Any part of a category’s content which cannot be operationalized in this way cannot be confirmed or refuted by scientific experiment. In my terms, this part of its content has no place in the “categoreal region” dealt with in physics. “Substance” may be an example of just such a category. It may be that the observed interactions of causal networks can be completely accounted for by speaking wholly in terms of “properties”. It may be that it makes no difference, in the description of any given causal network, whether we think of these properties as inhering in substances, or of a substance as a nexus of properties, or if we eliminate substance from our account entirely. If all this were true, it would not show that substances do not exist. It would not show that we can have no reliable knowledge at all about substances. But it would show that scientific experiments focused on causal networks cannot as such and in themselves answer any questions about them. It cannot help us decide between alternate conceptions of what the content of the category “substance” ought to be, or if this category corresponds to anything in reality at all. Such questions are beyond the range of application of scientific theories, due to the character of the constraints operating in scientific experiment.
. THE CATEGOREAL REGION OF PHAINOMENA.
The above discussions are not intended to “prove” anything one way or another about the viablity of critical pluralism in religious studies. Rather, first, they provide two plausible models, outside religious studies, of how in general one can have critical theory without singleness of truth. Secondly, they introduce the notion of different “categoreal regions” which will be central to the remainder of my paper. That is, besides allowing for a pluralism of valid theories within a given field, this theory implies also an epistemological pluralism with respect to different fields of inquiry. Physics operates with a different epistemology than pure mathematics. And, I will hold, the critical study of religions ought to operate with a different epistemology than physics or mathematics.
My next step will be to introduce a third categoreal region to which, on the present view, belong the categories that make up the content of religious belief. The region itself is very broad — I will call it the region of “phainomena”, the Greek word for “appearances”, from which the modern “phenomenology”, derives. The world of phainomena is the world as delivered to us in involved human experience, apart from any explicit conscious reflection on or specialized theoretical inquiry into such experience. By “experience” here I intend the Erlebnis of continental philosophy (Gadamer: 58-63), rather than the “sense impressions” of British empiricism. People “experience” not only color, weight, smell, etc. They also experience love, insults, status, beauty, etc. The contents of experience includes not only objects and events. It also includes human meanings attached to them. We “experience” a handshake as a friendly gesture, and experience a slap as an insulting gesture. We experience the birth of a child as a meaning-filled event.
The experience-world of any given person has a determinate content. There is something there independent of free decisions, which one can discover, something which one can be mistaken about, and which one can correct one’s mistakes about. At this point I am not asserting anything about what such experience of phainomena might tell us about reality beyond phainomena themselves. I am speaking only of the actual content of the experience of phainomena. To say that the crucial beliefs of any given religion are beliefs about phainomena is to give up any attempt to verify or to falsify religious belief by trying to prove something about reality as it exists outside human experience. This move does of course make more problematic how it is one can say that religious beliefs are really “based on reality”. I will address this question below. At this point I would like simply to sketch out what it means to posit phainomena in general as a categoreal region separate from other categoreal regions, in which there can again be critical thought without singleness of truth.
Speaking of phainomena as a special categoreal region means that critical thought can take place which employs categories derived from human experience taken in the broad sense just outlined. We can and do make mistakes about phainomena. Someone makes a friendly gesture which we mistakenly experience as hostile; we think we ourselves are depressed when we are actually angry; we experience a promotion as a confirmation of our competence and worth when it happens to be a matter of luck, etc. To say that we can have critical thought about phainomena as such is to say that, in all these cases, critical corrective thought can take place using categories derived from experience. We do not have to reformulate questions in some relatively specialized and restrictive set of categories in order to think critically about them. For instance, in order to discover whether we are “really depressed” or “really angry”, we do not have to dismiss the meaning-content of “anger” delivered in our experience (anger as a phainomenon), and formulate what “being angry” would mean in the categories offered by the science of neurobiology. Both anger and depression are phainomena, and we can think critically about which one is occurring without going outside the categoreal region of phainomena altogether.
This part of my proposals is opposed, for example, to the atomistic empiricism of the kind advocated by Hume (Enquiry, Section II) — the idea that the first step in critical thought on any subject is to break down our experience into small, self-contained bits of sense data, verify each separately, and regard as verified only what can be regarded as agglomerates of atomic entities made known to us in this way. The contents of phainomena very often need to be defined not atomically but relationally. The experienced meaning of a handshake for example has to do with the way this gesture is related to a whole host of other experiences and cultural conventions. (This observation is of course even more true of the intensely meaningful experiences in which religious beliefs frequently originate.) The epistemology involved in Hume’s atomistic empiricism would eliminate in an a priori way much of the content of many categories derived from experience holistically conceived.
This proposal contrasts also with attempts to formulate a relatively technical set of philosophical or cosmological categories — the categoreal schemes, say, of Aquinas, of Whitehead, of early Heideggerian existentialism, etc. — and insist on reformulating all questions in these categories as a precondition for critical thought about them. So far as the study of religion goes, my objection here is that the crucial categories in any given religion derive their particular contents from experiences rather rich in relations to other very particular phainomena in the experience-world proper to that religion. In the case of Buddhism, for example, the content of Nirvana is crucially determined by the relation of Nirvana to other experientially derived Buddhist categories (dukkha, meditation, detachment, etc.). To insist on translating these categories into some preestablished different categoreal scheme is inevitably to eliminate in an a priori way some of the contents of Buddhist categories, some aspects of the phainomena which appear in Buddhist experience, and which get their content from their relation to other aspects of Buddhist experience. It has to regard these contents as in principle not subject to critical verification or falsification, and hence either nonessential or matters of completely groundless belief.
On the other side, this proposal about phainomena as a categoreal region must be distinguished from theories (e.g. some neo-Wittgenstinian theories) which would focus entirely on the actual language of a given culture or religion, and would preclude asking for example whether or not the categories offered by the language which certain communities use are adequate to any further reality which the language might be taken to represent. There is certainly a very close relationship between language and phainomena, but they are not the same thing. I would point, for example, to the relatively frequent cases where groups of people continue to use words and concepts which they have inherited from a religious or cultural tradition, when this language is to all intents and purposes a “dead” language for them. That is, the categories provided in this traditional language are grossly inadequate to the character of their actual experience of the world. There are cases in which people regularly use language which does not really mirror phainomena as they are actually experienced.
For any given person, there is a world of interrelated and mutually defining phainomena that constitutes the basic horizon within which all experience takes place, and which determines the content of all individual experiences. In the case of any given person, and for groups of people sharing a common way of experiencing things, this experience-world has a determinate character. It is something which other people can be mistaken about — as frequently happens when a person from one culture misinterprets the statements and experience of someone in another culture. It is also possible to be mistaken about phainomena given in one’s own experience. That is, there is a prereflective layer of experience which a person’s conscious thought can fail to accurately represent. (It is not however a completely pre-linguistic, “uninterpreted” layer, a point developed further on p. below.) As in the example given above, a person can fail to look closely into the way things actually do appear in her experience, and use relatively alien and inadequate categories derived largely from conventional social norms, dead religious images, etc
On this account, then, critical thought about phainomena must necessarily begin with a “hermeneutical” moment — a striving to make explicit the implicit contents of one’s own or of others’ experience in a way that remains as true as possible to the experienced contents of these phainomena. To this extent, the remarks above stand largely within the phenomenological tradition. One crucial place where I would differ however, especially from Husserl, is my proposition that here again critical thought can take place in the absence of singleness of truth. The critical thought about phainomena which I have proposed so far does not have as its object finding some absolute grounding for the categories given in anyone’s experience, in the form of some universal set of categories which necessarily shape everyone’s experience at some deep level. My thesis is that the phainomena appearing in any given person’s experience has a determinate character even in the absence of any such ground. There is nothing incoherent in saying that Heidegger giving a critical phenomenological account of the contents of his experience, and a !Kung tribeswoman doing likewise for the contents of her experience, might end up with different “phenomenologies”, even at the most fundamental level. (My point is not to reject a priori the possibility that there are cultural universals. This is a matter which should be decided by actual investigation. What I would question, first, is the privileged status one might want to accord to them simply on the grounds that they are found everywhere. I would resist, for example, regarding some aspects of Christianity as peripheral or inessential on the grounds that they happen not to be universal. Secondly, in any case my main point is that critical thought about phainomena is possible whether such universals exist or not.) I would point to a work like Calvin Schrag’s Experience and Being as laying some important groundwork for the pluralist phenomenology which would be required for carrying out the present proposals.
On this view, then, the study of a given religion ought to begin with a descriptive moment in which the aim is to grasp accurately the interrelated phainomena which the language and symbols of this religion attempt to give expression to. But the major work of critical evaluation in the study of a religion comes primarily when one comes to consider further one particular kind of phainomena thus brought to light, categories belonging to what I will call the categoreal region of “the good”. To this I now turn.
. RELIGION AND “THE GOOD”.
“Religion” is a broad term, which can refer to concrete social groups and institutions, as well as to an ideal set of beliefs, experiences, attitudes, and practices. It is religion in the latter aspect that I intend here as an object of critical inquiry. In this respect I take as a central given a certain “range of application” claimed for religious beliefs — the claim that religions make to describe what it is people ought to do with their lives, what is ultimately important. In the context of the present discussion, then, the question becomes: What kinds of reality-constraints ought to govern our conceptions about what we ought to do?
On the present view, conceptions concerning what we ought to do are all based on reality as it appears in human experience. Thus the categoreal region involved here is included in the more general region of phainomena. Categories depicting ideals, obligations, importance, deserving, etc., are a subset of the set of possible categories describing the contents of reality-as-experienced.
I will call this sub-region of phainomena the region of “the good” — though this term needs immediately to be qualified for the purposes of the present discussion. I will use “the good” here, to refer to whatever deserves our respect or our service for what it is in itself. Wittgenstein comes close to a general description of this region when he describes inquiry into it (which he calls “Ethics”) by listing typical examples: this is “enquiry into what is valuable, or, into what is really important…into the meaning of life…into what makes life worth living…into the right way of living” (641). To show that something is “good” in this sense, one would have to show that it can be grounds for respect or admiration, for an obligation we have toward another, or valid grounds for being proud of ourselves. So for example I think a good case can be made that Aristotle’s “happiness” (eudaimonia) is something “good”, whereas it would be much more difficult to make a case that mere “pleasure” (pleasurable sensations or euphoric feeling, taken simply in themselves) is “good” in this particular sense. From this point of view, the sense I intend is narrower than some philosophical usages. On the other hand, it is broader than any particular theory of “the good”, such as Plato’s. “The good” is a categoreal region, in which there are potentially many valid systems of categories (just as mathematics may be a region in which there are several different mathematics-systems). Finally, it must be emphasized that “the good” here is an abstract “category of categories”. It should not be taken, as it was in Neoplatonism, as the name of a particular reality which one might experience directly in itself, and which could serve as the object of religious devotion. One cannot be “devoted to the good”, rather than being devoted to God, Tao, virtue, etc. — any more than one can eat a vegetable rather than eating a carrot or a tomato.
Again, the alternatives traditionally posed about this area are: either there is a single correct set of categories and principles which ought to govern all human thinking about “ought’s” — or else there are no constraints at all on our beliefs in this area. On the present view, these are false alternatives. There is a third possibility.
Religious/ethical categories depict some realities or ideals that “make a claim” on us. It is true there are a great variety of realities and ideals which might make a claim on us. But it is not true that just anything can justifiably be regarded as making a claim on us. There is a certain way in which reality does constrain our beliefs about what we ought to do with our lives, and such constraints can exist in the absence of singleness of truth in this area. Not just anything can be posed as good grounds for deserving, for obligations, for claims to importance, etc.
An example of how this principle might work out in practice can be given by pointing to the classic case of Socratic definition of virtue-concepts. (This is only an illustrative example, chosen for its relative simplicity. I am not of course proposing the following method as a fully adequate basis for critical theory in religion.) In the Meno (71e), Anytus wants to define virtue (arete) in such a way that it is bound up with actual political success in governing. If Anytus is simply a philosopher defining a neutral term he wants to use, or if he is merely reporting what he thinks his contemporaries mean by the word arete, this is a defensible procedure. But if he wants to use arete as a validating concept, such that one will be able to say that if so-and-so has arete he has a genuine claim on our respect, then this places a constraint on the content of the concept itself. In this case Anytus’ concept can be criticized by simply pointing out that a malicious tyrant might achieve actual political success by force, without truly deserving anyone’s respect. If validation of someone’s claim to respect is to be included in the range of application of a concept, then there are certain constraints on the formation of the concept. (Ultimately, the epistemology of the good that I would advocate would involve a much broadened form of Rawls’ “reflective equilibrium” [Rawls: 46-53; I would follow a broadened account such as the one proposed in Schüssler-Fiorenza: 301-304]. We know what is good ultimately by an experiential “intuiting” of claims made on us, though no specific intuition in itself gives certain and incorrigible knowledge. The implicit contents of specific intuitions can be explicated, generalized, and subjected to rational critique, and can be also be corrected in the light of other concrete experiences.)
What is important is that the possibility of the above criticism of virtue-concepts in this way is completely independent of the proposition that there is one and only one correct answer to the question “what is virtue”? If one regards as crucial to this criticism only the kinds of constraints operative in it, it is theoretically possible for there be several conceptions of virtue (or courage, justice, love, etc.) that are more or less equal in their ability to withstand such criticism — as well as, however, a number of conceptions which cannot withstand it.
To say that religions need to be grounded in something “good” is not to say that religion can be “reduced to ethics” — especially if this is taken to mean that there is some single system of ethical principles in the light of which religions ought to be evaluated. Religions and systems of philosophical ethics (like different mathematical systems, or different explanatory structures in physics) are different systems of categories occupying the same region.
On the other hand, to place the crucial objects of religious belief in the region of the good is to insist that beliefs about what we ought to do with our lives cannot be grounded in some realities lying outside the region of the good. In other words: The “necessity” which we are under to anything that makes a claim on us is sui generis. It cannot be reduced to physical/causal necessity, nor to logical or “metaphysical” necessity. (It is in the nature of the good that we are always physically free in relation to it, and it is not necessarily illogical to refuse recognition to it. This is how I would translate, and accord partial validity to, Kierkegaard’s insistence in the Postcript on existential decision vs. what he calls the “objective” stance. I would of course not agree with him that such decisions are based literally on “uncertainty”, in the sense that no supportive reasons can be given for them.) The only crucial consequence of failing in the respect we owe to the good as such is… being in the wrong. The grounds of religious beliefs can be “transcendent” in the sense that they make claims that go far beyond relatively mediocre conventional (“worldly”) conceptions of the good. They cannot “transcend” the region of the good itself (or, consequently, the region of phainomena) without ceasing to make a claim on us.
On this basis, there can be three general kinds of criticisms of religious belief. First, it might be true that the grounds for holding certain beliefs do not justify the conclusions that are drawn from them. For example, I’m reasonably sure one could make a cogent case that the followers of Charles Manson and Jim Jones did not have grounds for their beliefs sufficient to draw the conclusions about purpose for their lives which they drew from these beliefs.
Secondly, believers in a good religion can be mistaken in the connections they suppose to exist between different facets of their beliefs, and in the conclusions they draw from their beliefs. For example, many Christians suppose that the authority of God — the valid claims he has on human obedience — rests on his physical power and the fact that he created the world. Correlatively, they think they can know “by faith” apart from any scientific evidence, that such a physically powerful being exists.
From the point of view of the present proposals, both these assumptions are false. On the first point, imagine a science-fiction scenario: A world of people has been created by an all-powerful but malicious being, whose purpose in creating this world and its people was to see people torture each other. It would be at the very least problematic for the people in such a world to take this being’s purposes as authoritative for their lives. Physical power, even on the largest imaginable scale, is simply not good grounds for moral obligations.
On the second point, we have to ask about the grounds of Christian beliefs about God’s causal power — whether these beliefs have been subject to those kinds of reality-constraints that would entitle us to draw conclusions from them about what causes what in the universe. “Faith”, I take it, is generally a response to an experience of being religiously moved by something (reading religious writing, undergoing some moving experience or witnessing some moving event, etc.). Insofar as such experiences can claim to be experiences of something which makes a very strong claim on us (something very “good”), they can be very good grounds for knowing what we ought to do with our lives. But can they serve as good grounds for knowing about the causal interactions that brought our present physical world into existence? On the present view, clearly not. There are certain kinds of constraints on beliefs from which one wants to draw conclusions of this kind. These are constraints of the sort commonly used in astrophysics. We have no good reason to suppose that experience of the good gives one any privileged information about physical causes in the world. It is the freeing of the physical sciences from this kind of influence that has brought about their undeniable successes in describing the causal structure of the world. Following the above definition of “faith”, then, one would have to say that religious faith as such, like pure mathematics, must give up its claim to give knowledge about what causes what in the physical world. Many believers do of course in fact regard such claims as both warranted by religious experience, and warranting the importance of what is given in these experiences. On the present theory, neither of these impressions is valid, if one takes “cause” in the strict sense it is given in modern science. This is why Proudfoot seems to me mistaken in emphasizing the issue of the causal explanation of religious experiences, as central to any test of their validity (209-221). This is connected with his advocacy of a causal theory of perception (177), with which I would also disagree.
A third kind of criticism of religious beliefs has to do with the particularity of any given person’s experience. (As such this criticism can only be fully practiced by an individual with respect to her own experience, though externally observable behavior also provides some more public grounds for evaluation.) That is, although on the present theory it is theoretically possible for there to be a plurality of equally well-founded notions of what we ought to do with our lives, this theory does not lead to the conclusion that every given individual is faced with a religious supermarket in which all choices of theoretically good religions are equally valid. This latter view is the result of the excessively “intellectualist” view of religious knowledge which is one of the chief targets of this paper: The view that, if our beliefs are not constrained by universally valid theoretical arguments, then they are subject to no constraint at all, but must be regarded as completely arbitrary. By contrast, the present proposals rest on a radically experiential theory of religion. Knowing that Christian salvation is the one thing necessary, or that Buddhist Enlightenment is the one thing necessary, is ultimately a matter of experiencing the overwhelming importance of one or the other these goals. One who doesn’t experience this can be “converted” to experiencing things this way. (I would accord partial validity to Karl Barth’s position in this way.) But genuine conversion is a thoroughgoing personal change effected by concrete encounter with something overwhelmingly good which one experiences as such — not through intellectual proof and consciously controlled choice. And on the other hand, such personal conversion is subject to criticism. A person can critically examine her experience of the overwhelming importance of salvation or enlightenment, and discover that the conclusion about overwhelming importance is ill-founded, that what motivates it is not something truly “good” (e.g. that it only represents a flight from the psychological discomfort of anomie). But here again, critical reason comes as a second-order critique of experience, capable of falsifying the deliverances of experience, not as first-order discovery establishing the one true set of beliefs which all must adhere to no matter what their experience.
. THE OBJECTIVITY OF PHAINOMENA.
Placing the realities which found religions in the region of phainomena and “the good” solves one problem which a critical-pluralist theory must face — it gives a plausible account of the general kind of constraint that must be operative in well-founded religious beliefs. But the solution proposed also brings up an important difficulty. This is the problem, famous since Kant, about the relation between things as they appear in human experience, and things as they are in themselves. It is obvious to all today, I think, that human consciousness is not simply passive in its perception of reality. “The world” of our experience is a world which is organized in the process of perceiving it, and this organization cannot be regarded as a property of things as they are apart from human consciousness. The most common interpretation of this situation is to say that the mind at least partially “creates” the world which it perceives. Thinkers of a positivist bent typically take this to mean that what the mind adds to reality is not really there, and does not deserve serious attention. Others in the “idealist” tradition like Hegel, Cassirer, and Sartre, exalt the “creative spirit” of man that does the organizing of the world, and urge us to still greater “creative” activity.
In either interpretation, the idea that phainomena are a creation of the human mind creates a major problem for a theory of religion that founds religion on phainomena of any kind. It makes religion sound like the rules of football. If they have their origin in the mind, how can they create real obligations for this mind? They only have whatever “authority” the mind wants to give them, and it can withdraw this authority in the same way we withdraw from a football game.
To begin to address this difficulty, it is important, first, to note that this problem arose in its modern form as a result of the rise of the exact sciences. It was this development which for the first time gave people a systematic way of investigating properties that things have independently of the way they are perceived in human experience. The results of these investigations were initially taken as a picture of “the world as it really is”. And soon philosophers (most notably the British empiricists) became aware that there were whole areas of reality which could not be verified to be part of the real world by these methods. Hence the question became, not whether this or that picture was beautiful or not, but whether “beauty” itself was a possible part of the real world or not. Since science could not verify it, beauty as a category came to seem represent a “mere appearance”.
But this view depends on taking the results of the exact sciences as a true picture of the world “as it really is”, which one can use as a contrast with the “appearances” which we experience, and which we can use to correct all mistakes we make about reality. We can now see that this is an oversimplification. In the beginning, Newtonian physics found an easy marriage with a certain kind of commonsense empiricism. The categories of physics were derived from certain categories of experience (weight, speed, force, etc.) At this early period we still find arguments about what is “necessarily” true about physical bodies and their interactions, based not on experimental evidence but on intuitions about the content of certain categories. (As for example, the discussion about actio in distans — whether causation at a distance, through no medium, is possible.) But as physics has progressed especially in this century, its connection with experiential-intuitive categories has grown more and more tenuous, so that physicists now regularly accept notions previously regarded as “unthinkable” (such as particles that have no position). And it has become more obvious that the categories it uses to describe physical reality at the most fundamental level, cannot be taken as a true picture of actual things as they are. What we are sure these categories accurately represent is not the nature of things, but the causal relationships that obtain between them. For example, a certain notion of “force” is implicitly defined by its use in scientific formulae. But, by itself, the fact that some particular definition “works” in this way only means that, taken with a large set of other concepts, it accurately represents causal relationships. The fact that it works in this respect says nothing about what a force “really is”, about the nature of an independent real entity called a “force”. This is beyond the valid range of application of what can be shown by scientific experiment.
On the present view, the physical sciences probably do tell us all we can know about material reality as it is apart from human cognition. On the other hand it is unlikely that causal relationships exhaust reality as it is apart from human consciousness. We have to say then simply that there probably is more to reality-apart-from-consciousness than causal relationships, but that it is beyond the range of scientific method as such to tell us anything about it. There appears to be no other way that reality-as-it-is can act as a constraint on our knowledge. Hence we ought to say that all other knowledge we have is a knowledge of reality as it appears in human cognition.
This might seem to lead to a slightly modified Berkleyan idealism: almost the entire world of our experience is a “creation of the human mind”. In answer to this, I would like to borrow an argument from Henry Allison (14-34), in a recent reinterpretation and defense of Kant’s position vis-a-vis Berkeley. (It is important to distinguish my proposals from Kant’s position in other respects, particularly from other attempts that have been made to defend religion on Kantian grounds.) On this view, Berkeley’s idealism is not radical enough. Berkeley’s idealism takes as its starting point the fundamental difference which we experience between two categories, “the human mind” and “the external world”. It is a theory about their relationship, saying that the latter really exists only “inside” the former. What is needed is a theory which situates more radically the very distinction between the human mind and the external world. One needs to realize that both of these categories, and the perceived distinction between them, are already “appearances”. Both are already “reality as perceived by human consciousness”, not “reality as it is in itself”. This applies to our perception or idea of the human mind no less than it does to our perception of external reality.
This same point can be stated in different terms as a distinction between two senses of the word “objective”. In one sense of the word, “objective” refers to the world perceived as “out there”, in contrast to the “subjective” reality of internal ideas, feelings, etc., perceived as “in here”. In another sense, “objective” refers to reality as it is in itself, entirely apart from the way it appears in human cognition. It is a mistake to conflate these two senses of the word “objective”, and assume that trying to understand the “objective reality” that we are confronted with in our ordinary lives, is the same thing as trying to find out what reality is like apart from human cognition. Science has brought about an awareness that many of the kinds of things that appear to be out there cannot be verified as properties of things as they are in themselves. Should we conclude from this that we are mistaken about their “objectivity” in the first sense?
On the present view, no. This would be a mistaken conflation of two different senses of “objective”, which need to be kept separate. In a world consisting of phainomena already structurally differentiated into “subjective” mental reality (a “self”), and an “objective” reality which this self encounters, it is inaccurate to think of the latter as a product of the former, to think of the reality objectively encountered as a creation of the human self. The human self is itself a phainomenon, whose perceptual content is largely defined relationally vis-a-vis other phainomena which it encounters “in the world”. Given this structure of phenomenal reality — the reality in the context of which all perception and thought takes place — there is no warrant for a wholesale removal of all reality from one phenomenal sphere (the sphere of the world), and placing it in another (the sphere of the mind). There are of course specific cases, like hallucinations, in which experience itself gives us reason for doubt. But to say that phainomena as a whole are not real would be to say that “the world”, in the sense we normally use that word, is not real. This is an incoherent view, because again what we usually mean by “real” is “really belonging to the world of our perception”, the only world we know or can really imagine in detail.
One has to say that there is indeed a sense in which this “real world” is partly constituted by an experiencing consciousness, which is not merely passive in its experience. If there were no human consciousness the “real world” that actually appears in our experience would not exist. (There would still be some existent world, but we cannot imagine it — to do so would be to bring it into the region of phainomena.) But the “human consciousness” spoken of here is not the same as the phenomenal self. The phenomenal self creates the rules of football, and it has a choice whether to do so or not do so, to formulate them in one way or another way, to play or not to play football. The “human consciousness” which is partially responsible for the constitution of the world of phainomena is not like this. It has done its work before “I” (the phenomenal mind) come on the scene, so to speak. “I” arrive on the scene always already too late to regard myself as the one who creates the scene.

This is a rather difficult (but crucial) point, so let me try to formulate it in one other way: Consider the assertions:

“Knowledge of phainomena is not knowledge of the objective world. It is largely a knowledge of mental creations”.
Now consider the way in which these statements would have to be understood in order to make them true. In the normal case, all words, including the terms “objective”, and “mental creation”, get their meaning in relation to reality as we experience it, i.e. in relation to the world of phainomena. But in order for the terms “objective” and “mental creation” in these sentences to have a meaning which would render them true, we would have to mentally erase from the contents of the categories “objective”, and “mental”, all relation whatsoever to this phenomenal world as such. In particular we would have to erase from the meaning of these terms all association with any experience of an “external world” contrasting with a “human mind”. There could be no association whatsoever to such experiences as hallucinations, dreams, having invisible thoughts, etc., from which we derive our normal understanding of the distinction between an “internal mind” and and “external world”. One could perhaps move in the direction of such erasure, by a very strenuous effort at abstraction, but it seems doubtful that one could ever fully abstract from the phenomenal world, and still end up with something that could be called “understanding a meaning” in these words. The conditions for understanding the statements, in a sense that would make them true, may be impossible of fulfillment.
On this view, then, there are many phainomena, and aspects of phainomena, which are non-verifiable by science, but which are still “objective” in the sense that they are an integral and necessary part of the world that we as individuals experience ourselves as facing. They are “givens” for our empirical ego, for the phainomenon which is our self. The naive impression that many phainomena, non-verifiable by the exact sciences, are “out there in the world” (in anything like the ordinary, non-technical sense of this expression) is not mistaken.
What is important in the above discussion of phainomena in the present context is of course that the category of reality I have called “the good” is a subcategory in the general categoreal region of phainomena. My thesis about the good as an aspect of the objective world, then, can be formulated as follows:
Contrary to Kant, there are no particular categories concerning the good that are a given and necessary part of reality for us. But there is something necessarily given about the region of the good: What is necessary is the fact that the external reality we experience has a dimension of good/not-good. Take for example the question of “life’s meaning”. The word “meaning” can refer simply to the fact that some proposition makes sense in some consistent scheme of concepts. (This is the
sense in which statements involving “division by zero” are said to be “meaningless” statements in ordinary arithmetic.) But to say that there are facts in the world having some “meaning” in the sense that they can be described in some consistent theoretical vocabulary, is not yet to say that the world is a “meaningful” place to live in in the more ordinary sense of that word. To say that “life is meaningful/meaningless” is to make a statement about “the good” in life. A meaningful world “deserves” our participation, it “makes life worth living”. A meaningless world does not.
In this context, the present proposals do not assert that the world is meaningful in some particular way, or even that it is meaningful rather than meaningless. On the view I’m proposing, “meaninglessness” is negative meaning. It is a category still belonging to the categoreal region of the good. What is given as part of the phenomenal world is simply that this world has a dimension meaningful/meaningless. It is possible for the world to be genuinely meaningful in very different ways for different people. And it should be considered within the realm of possibility that the world might genuinely be meaningless in someone’s experience of it — again in several different ways. What is not possible is that a human being experience the world in a way that is completely neutral in respect to meaningfulness, in which the dimension meaningful/meaningless is completely absent.
The meaningfulness of the world is something that one can be mistaken about. One can for example hold onto the idea that the world is meaningful simply out of fear of entertaining the opposite view. One can mistake a feeling of mere pleasantness for a perception of genuine meaning. Conversely, a person could be attracted to the view that the world is meaningless out of mere egocentric resentment that she has not been given a more important place in the scheme of things. These kinds of considerations make for “ill-founded” views about the question of the world’s meaningfulness. But, since meaningfulness is a category belonging to the general categoreal region of phainomena, one cannot go outside one’s experience to try to correct one’s mistakes. (An epistemology requiring impersonal objectivity precludes a priori the category “meaningful”.) One can only base such corrections on appeals to other experienced phainomena.
. RELIGION AND RELATIVISM.
A final fundamental issue that needs to be dealt with here is the well-known problem of “relativism”. Religion constitutes a way of organizing the world meaning-wise. The pluralist position holds that different religions constitute irreducibly different ways of organizing the world, genuinely different ultimate accounts of “the meaning of life” which assign different meaning to life experiences and actions.
Very often, the view that there is no one way the world is organized meaning-wise is taken to lead automatically to the view that individuals and groups impose a certain order on the world to “fulfill certain needs” which they have (for comfort, order, social cohesion, etc. See for example Berger: 3-51). From the point of view of this kind of relativism, religious beliefs are not really knowledge about ultimate realities. They stand rather to the fulfillment of such needs as means to ends, and in this sense the fulfillment of needs is more ultimate than they are. This results in a skeptical, “nihilist” relativism: Ultimately we believe only what it suits us to believe, and this is all that beliefs about life’s meaning can be. Again, such beliefs can finally make no more claims on us than the rules of football.
I would separate the proposals argued for here from such skeptical relativism, first, by repeating in slightly different form the point made above in relation to the objectivity of phainomena. That is, while it is possible that the world can be organized differently for different cultures and religions, meaning-organization as such is a given in all human experience. As modern criticisms of traditional (atomistic) empiricism have pointed out (see Morick’s anthology), there is no such thing as “uninterpreted” sense data. At whatever primitive level one wants to examine the contents of experience, one will never find isolated and self-contained bits of data. Everything is always experienced “under some description”, and the “descriptions” involved get their content from their place within some larger system of categories.
Kant already observed that there is “no percept without a concept”. Modern studies of cultural variations have expanded on this (in a way contrary to Kant’s foundationalism) with an insight given classical expression in Saussure’s “semiotics” (Culler: 30-39, 105-108): that there are no concepts with self-contained and “necessary” meanings which can serve as fixed points for defining the meanings of all other concepts (Derrida’s “transcendental signifieds”). All percepts get their content as parts of “systems of signs” that are completely mutually defining, and there is no “necessary” system.
From this point of view, the Sapir-Whorf principle that “language shapes reality” does not put the issue quite radically enough. One should say rather that “reality is linguistic”: The components which make up the world that we experience are always already organized “semiotically” — after the manner of a “system of signs” mutually defining each other’s content. There are no uninterpreted, non-semiotic, “raw data” for language to refer to. (See Eco’s argument [58-68] on this point. One should note that this thesis about the semiotic organization of reality is independent of the thesis of early structuralists like Levi- Strauss, that at some deep level there is a single set of structures universal to all human culture and experience. Categories whose essential meaning is derived from their relation to other categories can have determinate meanings whether or not such “universal” structures exist. Umberto Eco’s Theory of Semiotics provides a well-worked-out theory maintaining Saussure’s basic insight, but which neither depends on “universalist” structuralism, nor draws the skeptical conclusions from its rejection that Derrida sometimes seems to.)
Following these observations, one could say that it is indeed partially and sometimes true that the way the world is organized meaning-wise for a given individual or group is the result of conscious or unconscious desire to fill certain needs. But this is to be distinguished from a more fundamental meaning-organization which is always a given part of reality, prior to such need-fulfilling activity. The picture of a self consciously or unconsciously “organizing reality to fulfill its needs” already assumes such a more basic organization: Without this more basic meaning-organization there could be no “self” recognizing certain goals as “needs” and adopting “means” to fulfill them.
The importance of this point in the present discussion is that it makes possible a contextual conception of “the good” which still does not lead to nihilistic relativism. On this conception, the contents of concepts representing the good are always “relative” to some cultural context. But the fact that the “relativity” involved is semiotic rather than functional means that the good involved is still a “good in itself” making a claim on us for what it is in itself, rather than an “instrumental good”, only functionally useful as a means to achieving something else. To illustrate by analogy: A machine is a functional system in which parts interact to produce something outside themselves. A painting is by contrast a semiotic system. The meanings of its various elements are contextually defined in relation to each other (a red patch in the center of a predominantly blue picture has a different aesthetic meaning than the same red patch in the corner of a yellow picture.) But they don’t interact causally as “means” to produce something different from themselves. What the elements of a painting “produce” is a structured whole made up of these same elements. When categories related to “the good” are defined in relation to semiotic systems rather than functional systems, this allows for a different sense to be given to the phrase “good in itself”. That is, it allows for something to be “good in itself”, in the sense that it is not instrumentally good for something else, and yet not “good in itself” in the sense that it is good in all (semiotically construed) contexts.
On the present view religions, like paintings, are best regarded as “semiotic systems”, each element of which gets its essential meaning by its relation to other elements of the respective system. The “God” Paul believes in gets his essential meaning from his relation to other key Pauline concepts and experiences: Sin and fallen-ness, faith, salvation by Christ’s crucifixion and resurrection, the contrast with salvation by works, etc. — and from his relation to the rest of Paul’s culture and worldview. The same can be said of the Buddhist concept of Nirvana in relation to Buddhist concepts and practices, and to the Asian culture of some particular Buddhists. One can hold this and still hold that believers might entrust themselves to God, or strive for Nirvana, as a response to claims made by some realities “good in themselves”, not as a “means” to achieving something desirable on other grounds.
From this point of view, it is indeed possible that the beliefs of some individual or group about life’s meaning have been molded, consciously or unconsciously, to serve certain needs for comfort, self-esteem, social order, etc. But when this is true it is a contingent fact, to be supported by evidence in each case. It is not a blanket explanation which we know a priori to be the true explanation of all cases of religious belief. There is another possibility which ought to be considered. The way the world is meaningfully organized for any given person is partially dependent on that person’s interests. Among these interests can be concern for the good. It is then possible that particular set of religious beliefs might represent the way the world is organized meaning-wise for a person (a “converted” person in the sense described above, page ), whose perspective on the world is dominated by selfless concern for the good. (The same I think can be said for the worldview of Heidegger’s Being and Time. Heidegger’s “Being”, as something that makes a claim on us, belongs in large part to what I am calling the region of the good. In Heidegger an understanding of the “meaning of Being” requires a developed “conscience” (312-348), and an “authentic” response [see Zimmerman: 69], which belong to the moral sphere, however much one wants to distinguish them from what counts conventionally as “morality”, or from the subjectivist value-theory of German idealism from which Heidegger wanted to distance himself.)
From this point of view, when religious beliefs have indeed been molded by concerns other than the concern for the good, this is to be accounted a defect in them, grounds for just criticism. This is true even if one is able to show that, counting heads statistically among those who call themselves “religious”, concerns other than concern for the good have actually predominated in the formation of their beliefs.
At this point one can see the closeness of the critique of religions which the present proposals would lead to, and the internal critiques conducted by religiously advanced members within each religious tradition. That is, the critique which these proposals envisage, like that of more idealistic members of each religion, would tend toward an ideal interpretation of a given religion, even when such interpretation might be statistically not most common among actual believers in that religion. A Buddhist who construes the karma-reincarnation idea as an invitation to strive for personal rewards in the next life has a perspective dominated by pursuit of self-interest rather than concern for something that makes a claim on us, and this interpretation of Buddhism deserves religious criticism on these grounds (as it is actually criticized in early Buddhist writings, see Horner: 320). The same is true of course of a Christian who construes Christian afterlife-beliefs as an invitation to bargain with God. (Such belief would remain religiously defective even if one could show that there is an all-powerful being with whom one could bargain about afterlife rewards. Self-interested reward-seeking is not of course in all cases wrong, but it is something totally different from placing ourselves in service to something which makes a claim on us. So far as devotion to the good is concerned, self-interested bargaining with God is not different in principle from self-interested bargaining with a real-estate agent.)
This is the sense in which there are reality-constraints which ought to govern the contents of one’s beliefs about God, Tao, Brahman, Nirvana, etc. That is, when taken together with some entire system of Christian beliefs, the meaning that “God” has in the context of this system must be such as to represent something very “good” and deserving of single-hearted devotion. Criticized by this standard, some conceptions of God can fare much better than others, and some can be completely discredited as the basis for a valid religion. On the other hand there might be several different concepts of God which fare more or less equally well, especially when the way one construes the remainder of Christian beliefs are also altered, as has actually happened frequently in the history of Christian (and Muslim, and Jewish) belief. Similar things can be said of Tao, Brahman, etc.
As to actual criteria for evaluating religions, it appears to me that although there are fairly clear negative criteria by which certain religious conceptions can be ruled out, positive criteria vary between religions, and can only be fully developed by studying any given religion as a total “system” (hence the difficulty of treating this topic adequately in this paper). “God” represents a different kind of good than “Nirvana”, and the ideal relation in which the respective Christian and Buddhist believers stands to these two realities is also quite different. Hence the ways in which God and Nirvana might be shown to represent something “good” might differ.
On this view, then, “God”, “Nirvana”, “Tao” etc. have irreducibly different meanings. These are not meanings possessed by realities (or a single reality) that can be defined apart from the meanings. The meanings are constitutive of the being of God, Nirvana, and Tao as objects of religious belief — these meanings are the foundation of the valid claims made by these realities on the devotion of believers. FN1.AAThis statement implies a non-literal interpretation of such otherworldly realities: I.e. they are not realities whose existence can be conceived of, and verified, as isolated entities, as can trees and rocks and houses. (“Literal” existence I take to mean existence verifiable across all experiential contexts.) I hope my discussions above have given some grounds on which they can be regarded as “real” in spite of this. This is one facet of my theory which admittedly needs a much more extended defence than I can give it here. On the present view these meanings are potentially as much a part of reality as the rest of the world people encounter in everyday life. The theory proposed here would allow their actual reality to be critically substantiated only in concrete cases for specific individuals, and for groups sharing a common experience.

                     Reference List

Allison, Henry E.
1983 Kant’s Transcendental Idealism: An Interpretation
and Defense, New Haven: Yale.
Barker, Stephen F.
1964 The Philosophy of Mathematics. Englewood Cliffs,
N.J.: Prentice-Hall
Berger, Peter
1969 The Sacred Canopy. New York: Doubleday.
Cassirer, Ernst
1955 The Philosophy of Symbolic Forms. Vol. I:
Language. New Haven: Yale Univ. Press.
Davis, Charles
1975 “The Reconvergence of Theology and Religious
Studies”, in Sciences Religieuses/Studies in
Religion, 4 (1975), 205-21.
1984 “Wherein there is no ecstasy”, ibid. 13/4 (Fall
1984), 393-400.
Davis, Philip J., and Hersh, Reuben
1981 The Mathematical Experience. Boston:
Houghton Mifflin.
Eco, Umberto
1979 A Theory of Semiotics. Indiana: Indiana Univ.
Press.
Gadamer, Hans Georg
Truth and Method. New York: Crossroad. 1982.
Harries, Karsten
1967 “Martin Heidegger: The Search for Meaning” in
G.A. Schrader Jr. ed. The Existential
Philosophers: Kierkegaard to Merleau-Ponty. New
York: McGraw-Hill, 161-208.
Heidegger, Martin
1962 Being and Time. Tr. J. Macquarrie and E.
Robinson. Harper & Row.
Hick, John
1980 God Has Many Names. Philadelphia: Westminster.
Horner, I.B. tr.
1954 Middle Length Sayings Vol. I. London: Luzac.
Hume, David
1939 An Enquiry Concerning Human Understanding in
E.A. Burtt ed. The English Philosophers from
Bacon to Mill. New York: Random House.
Kant, Immanuel,
1966 Critique of Pure Reason. tr. F. Max Müller. New
York: Doubleday.
Kierkegaard, Soren
1941 Kierkegaard’s Concluding unscientific postscript
tr. David F. Swenson. Princeton: Princeton Univ. Press.
Korff, H. A.
1966 “Kant Romanticized by Fichte” in Thorlby, A. K.,
ed. The Romantic Movement. London: Longmans,
Green & Co, 107-112.
Lindbeck, George
1984 The Nature of Doctrine: Religion and Theology in
a Postliberal Age. Philadelphia: Westminster.
Margenau, Henry
1950 The Nature of Physical Reality. New York:
McGraw Hill.
Phillips, D.Z.
1976 Religion Without Explanation. Oxford:
Blackwell.
Proudfoot, Wayne
1985 Religious Experience. Berkeley: University of
California Press.
Rawls, John
1971 A Theory of Justice. Harvard: Cambridge Univ.
Press.
Robinson, James M.
1964 “Hermeneutics Since Barth”, in The New
Hermeneutic. New York: Harper & Row.
Sartre, Jean-Paul
1965 The Philosophy of Jean-Paul Sartre. R. D. Cumming
ed. New York: Random House.
Schrag, Calvin O.
1969 Experience and Being: Prolegomena to a Future
Ontology. Evanston Ill.: Northwestern University
Press.
Schüssler-Fiorenza
1985 Foundational Theology:Jesus and the Church. New
York: Crossroad.
Smith, Wilfred Cantwell
1981 Towards a World Theology: faith and the
comparative history of religion. Philadelphia:
Westminster.
Smart, Ninian
1973 The Science of Religion and the Sociology of
Knowledge: Some Methodological Questions. New
Jersey: Princeton Univ. Press.
Taylor, Charles
1970 “The Explanation of Purposive Behavior”, in
Borger, R., and Cioffi, F. eds. Explanation in
the Social Sciences. Cambridge: Cambridge Univ.
Press.
Tracy, David
1975 Blesssed Rage for Order: The New Pluralism in
Theology. Minneapolis: Winston Seabury Press.
Weibe, Donald
1984 “The failure of nerve in the academic study of
religion”, in Sciences Religieuses/Studies in
Religion 13/4 (Fall 1984), 401-22.
Wilder, Raymond L.
1968 Evolution of Mathematical Concepts: An Elementary Study. New York & London: John Wiley
Wittgenstein, Ludwig
1965 “Lecture on Ethics” Philosophical Review.
3-12).
Zimmerman, M. E.
1981 The Eclipse of the Self: The Development of
Heidegger’s Concept of Authenticity. London: Ohio
Univ. Press.