Panentheism.  Revisionism.  Anarchocapitalism.


David Ray Griffin

[link to CV]


Essays by Me

Essays by Others

From Journal of the American Academy of Religion 68/1 (March 2000), 99-125. See his rejoinder to responses from Professors Preus and Segal.


Religious Experience, Naturalism, and the Social Scientific Study of Religion

David Ray Griffin


In two recent books, Religion and the Social Sciences (1989) and Explaining and Interpreting Religion (1992), Robert A. Segal has portrayed an antithesis between the “social scientific” approach to religion, which he favors, and the approach of those he calls “religionists,” meaning thinkers who defend the truth and thereby the “religiosity” of religion (1989:1; 1992:1).  Likewise, in Explaining Religion (1987, reprint 1996) J. Samuel Preus draws an equally strong opposition between “the modern study of religion,” which he advocates, and all “theological” or “religious” approaches.  For both authors, the difference is that the approach that they advocate explains religion in “naturalistic” terms, whereas religious approaches resist naturalistic explanations.  These “religious” approaches are of two fundamental types.  One type rejects causal explanations altogether, saying that religion is a phenomenon to be “interpreted,” not explained.  The other type of approach, while agreeing with Preus and Segal that an explanation is needed, offers a religious explanation.  The form of religious explanation primarily targeted by Preus and Segal is theological explanation, which explains religion, or at least theistic religion, in terms of a genuine religious experience of a holy actuality distinct from the totality of finite things.  But Preus and Segal also reject all other religious explanations, such as those that explain religion in terms of an innate religious sense and those that explain nontheistic religions in terms of a genuine experience of an impersonal religious ultimate (which might be called Nirvana, Emptiness, or Nirguna Brahman).

Although Preus speaks of the “modern” or “naturalistic” study of religion and Segal of the “social scientific” study of religion, they both are clearly advocating the same approach.  For example, they illustrate the approach said to be appropriate for the academic study of religion with the same exemplars, such as Comte, Tylor, Durkheim, and Freud.  Also, Segal says that Preus’s book traces “the emergence of the social scientific, or naturalistic, explanation of religion” (1992: 123), thereby showing that he sees Preus’s “naturalistic” approach to be equatable with the “social scientific” approach that he advocates.  And Preus, regarding Hume as the thinker who completed the paradigm shift in the study of religion to a thoroughgoing naturalism (xi), says that he prefers “to label Hume as the founder of the scientific study of religion” rather than of the philosophy of religion (84 n. 1).  Therefore, although Preus does not characteristically use the term “social scientific,” he is, no less than Segal, saying that the academic study of religion should be equated with, or at least based on, the social scientific study of religion.1

Thus understood, their common position involves three distinguishable claims: 1) The academic study of religion should not avoid the task of explaining the origin and persistence of religion. 2) Any explanation, to be acceptable in the academy, must be a social scientific explanation. 3) Any explanation, to be considered a social scientific explanation, must be “naturalistic” in their sense of the term.  Any of these three claims can, of course, be challenged.  The first claim can be rejected by those who believe that the task of the academic study of religion is to interpret religion rather than to explain it.  This position may be held on the basis of a methodological distinction between the Geisteswissenschaften, which properly interpret their phenomena, and the Naturwissen-schaften, which properly explain their phenomena, and this distinction may in turn be based on the assumption that “explanations” are necessarily reductionistic.  Or one might simply believe for some other reason that the attempt to give causal explanations for religion is somehow inappropriate, perhaps sacrilegious.   In any case, although this division—on whether the study of religion should focus primarily on interpretation or explanation—is probably the major division within the academic study of religion at present, I will not here directly discuss this complex issue,2 instead simply registering my agreement with Preus and Segal that the task of attempting to explain the origin and persistence of religion should not be avoided.  While agreeing on the need for explanation, however, I reject their view, as expressed in the second and third claims, as to the only academically acceptable type of explanation.3

This Preus-Segal view seems to be rather widely accepted among those engaged in religious studies. That is, although many religion scholars reject the task of explaining religion or at least do not devote their own energies to this task, the negative implication of the Preus-Segal position seems to be widely assumed—namely, that insofar as causal explanations are proffered, academicians involved in the study of religion cannot, by definition, argue for the truth of religion in the sense of holding that the best explanation for the existence of religion is that it arises from a response to something holy in the nature of things.  One can remain neutral on this question—by not raising it or by endorsing methodological agnosticism.  Or one can advocate a negative answer—as do Preus and Segal.  But one cannot, at least in one’s academic work, argue for the truth of religion in the aforementioned sense.4 Although a more complete critique of this position would need to consider both theistic and nontheistic understandings of the “something holy,” I will here limit my discussion to the theistic understanding.

My argument against the shared position of Preus and Segal develops as follows.  In the first section, I lay out the Preus-Segal claim that academic explanations of religion, being naturalistic, are necessarily in opposition to theistic accounts of religion.  In the second section, I point out that although Preus uses the term “naturalism” as if it had a single meaning, his discussion actually employs the term with eight distinguishable meanings.  The question thereby raised is whether all of these meanings of “naturalism” are really implied by the rejection of supernaturalism.  I answer this question in the negative, denying in particular that the rejection of supernaturalism requires what I call naturalismsam, meaning the sensationistic, atheistic, materialistic version of naturalism.  In the third section, I argue that the only valid reason for advocating the employment of this version in the academic study of religion would be a philosophical conviction as to its truth.  In the fourth section, I argue that the acceptance of this form of naturalism is not philosophically warranted.  The overall purpose of my argument is to undermine the widespread assumption, expressed especially clearly by Preus and Segal, that the academic study of religion necessarily excludes all theistic explanations of religion.


1. Naturalistic versus Theological Explanations

The academic study of religion, both Preus and Segal argue, should take a “naturalistic” and thereby antitheological approach.  Preus, speaking of an “essential contradiction” between “theology” and “the study of religion” (xvi), locates this contradiction in the former’s being supernaturalistic, the latter, naturalistic.  The naturalistic approach, he says, “is the decisive feature which distinguishes the study of religion from theology” (205).  In referring to the theological approach as supernaturalistic, he refers not only to the confessional type of theology grounded in a supposedly infallible revelation, but to any nonreductionistic approach.  “[N]o decisive line of demarcation could be drawn between theology and the study of religion,” Preus says, “as long as a supernatural (or some objective, live, transcendent) ground of religion was assumed as the really existent referent and generative source of religious language” (xvi).  Although this statement might seem to allow for a distinction between a “supernatural” and an “objective, transcendent” ground of religion, Preus in effect equates the two.  Identifying “the study of religion” with a “naturalistic approach,” he contrasts it with “theology” and a “religious approach” (xi). And, referring to “two rival paradigms,” he equates “the naturalistic paradigm” with “an altogether nonreligious point of view,” which rejects “the absolutes of confessional and natural theology” (xi, xiii, xiv).  The naturalistic approach, he indicates, is “reductionistic,” claiming that “transcendence” is not the best answer to the question of the cause of religion (ix n. 2, xx).  This naturalistic, reductionistic approach has “its own explanatory apparatus” in which the categories of “transcendence” and “the sacred” are absent (xxi).  He clearly excludes, therefore, the possibility that religion might have a ground that is “objective” in the sense of being “transcendent” to one’s experience but that would not be properly called “supernatural.”

The naturalistic approach to religion, in other words, begins with the assumption that there is no genuine religious experience, in the sense of an experience of a divine reality distinct from the totality of finite causes.  The naturalistic analyst of religion, says Preus, rejects the religious participants’ own account of their religious experience, “because the analyst does not believe their explanation that mysterious transcendent powers beyond the realm of natural causation . . . really create this experience” (174).  The same point is made by Preus’s succinct statement of the central question for the academic study of religion: “if ‘God is not given,’ how is one to explain religions—that is, their universality, variety, and persistence until now?” (xv).

As these statements show, what Preus means by the academic study of religion excludes not merely confessional theology but any position that is “theological” in the sense of explaining the origin and persistence of religion in terms of genuine religious experience of, and thereby causal influence by, a divine reality distinct from the totality of finite causes.  To be academic, the study of theistic religion must consider it essentially false.

Segal makes essentially the same point.  In his more recent writings, to be sure, Segal has come to focus his critical attention more on those religionists (as he calls them) who deny the propriety of explaining religion in favor of “interpreting” it than on those who offer religious explanations for it.  Indeed, he even criticizes Preus’s “concentration on a religious cause” as “outdated” (1992:124f.).  The fact that Segal now thinks the major battle to lie elsewhere does not, however, imply any retraction of his view that the social scientific and thereby the academic study of theistic religion, to be such, must consider it false.

For example, assuming that to attribute religion to “the experience of god” is to speak of a “supernatural origin” (1989:76, 81), Segal says that every social scientific explanation of religion involves “a naturalistic rather than divine origin” (1992:19). In one place, in fact, Segal says that “any naturalistic explanations . . . are social scientific ones” (1989: 78), thereby seeming to say that being naturalistic is a sufficient as well as a necessary condition for an explanation to count as (social) scientific.  Should it not, one might ask, also have some other characteristics, such as self-consistency and reasonable adequacy?  In any case, the social scientist’s naturalistic (read “nonreligious”) explanation of religious experience, says Segal, intends to be a sufficient explanation, obviating the need for any divine cause (1989:82). The naturalist, Segal says, argues that “believers never encounter God” (1992:71).  Social scientific explanations thereby deny the genuineness of religious experience.

Given my exposition of the positions of Preus and Segal thus far, it would seem that they completely equate the academic study of religion with a naturalistic approach and this with a wholly nonreligious approach.  And that is, indeed, their overwhelmingly dominant tendency.  In a few passages, however, they both allow for a more nuanced understanding.  Segal grants that some (unconventional) social scientists have given a “social scientific affirmation of the truth of religion” (1992:15, 107).  And Preus points out that some early members of the naturalistic tradition gave theistic explanations of religion (1992:23-28, 206, 211).  If one can be a theist and yet be a naturalistic social scientist, “naturalism” must be multidimen-sional, with some dimensions being more essential than others.  I turn now to this issue.


2. Various Dimensions of “Naturalism”

A careful reading of the writings of Segal and Preus, especially the latter, reveals that their advocated naturalism, which at first sight seems to be a simple notion that one must either affirm or reject wholesale, is actually a very complex notion with many dimensions, not all of which are entailed by all the others.  It might be possible, accordingly, to affirm some aspects of their naturalism while rejecting others, with the result that an approach that is theological as well as naturalistic might be possible. 

I will carry out this analysis in terms of Preus’s treatment of naturalism, which is far more extensive. The basic distinction to be made is between the epistemic and ontological dimensions of the full-fledged naturalism his discussion advocates.  Each of these dimensions, furthermore, I contains four distinguishable doctrines.  Preus himself does not draw attention to these distinctions, which may help account for his simple division between naturalistic and supernaturalistic approaches.  The distinctions, nevertheless, are there to be seen.  I will lay out the four epistemic meanings of “naturalism,” then the four ontological meanings.

1. The most obvious meaning of epistemic naturalism is the denial of epistemic supernaturalism, which holds that some ideas are to be accepted because of their alleged mode of origin in an infallible revelation, in which a divine being supernaturally overrode the fallibility that characterizes most human thought, thereby making the message error-free.  The denial of this doctrine can be called epistemic naturalismnsr (with “nsr” indicating “no supernatural revelation”).  Preus indicates that acceptance of this view puts one within the naturalistic tradition, at least partly, by saying that the work of Jean Bodin and Herbert of Cherbury represented “a paradigm shift—for although they stayed firmly within a religious mode of explaining religion . . ., they put reason above the alleged revelations as the norm for religious truth, and they rejected the authority of confessional groups” (206). Preus further illustrates the importance of epistemic naturalismnsr in referring to Tylor’s criticism of “the refusal of theologians to give up the fight for a theological or revelational scheme to answer questions that belong to natural science” (145). Preus is surely right to say that the academic study of religion must be naturalistic in this sense, a contention that is no longer a matter for serious debate.

2. More relevant to current controversies is what can be called epistemic naturalismdu (for “domain uniformitarianism”), which is the insistence that religion is to be explained in terms of the same causal categories that are used in other cultural domains (x, xvi). “[T]he religious element in culture,” says Preus, is not to be exempted “from the processes that produce all the rest of it” (153).  Preus illustrates this dimension by pointing out that Vico’s “explanation [of religion] was naturalistic except for one remaining theological residue—a divinely implanted, universal religious sense” (206) and by referring to Tylor’s rejection of “the wish of theologians to be excused from the universal process of evolution by claiming an exemption for religion, or a particular form of religion, in the name of an entirely autonomous supernatural realm of law and causality” (153).  In rejecting the assumption that religion has a “privileged status” (xix), epistemic naturalism in the sense of domain uniformitarianism rejects not only the use of different causal categories to explain it but also the view that religion should be exempted from causal explanation altogether (81).  For example, in relation to Eliade’s statement that “the ‘sacred’ is an element in the structure of consciousness, and not a state in the history of consciousness,” Preus says that “in an academic setting where other scholars are struggling with the evolutionary emergence of our species . . . one legitimately wants to know how Eliade’s remarkable ‘element in the structure of consciousness’ might have gotten there” (xix). Preus is entirely right about all this.  The proper approach, as he says, is one that “pushes explanation to the limit” (210), and in this effort we should work towards a uniform set of explanatory categories for all features of the world.

Preus’s discussion of domain uniformitarianism, however, is problematic.  Although he rightly says that we need a “unified science of [humanity]” with “a single web of relationships, none of which is autonomous” (203), he confuses two possible meanings of “autonomy.”  It is surely proper to reject complete autonomy, according to which religion would be “a law unto itself, entirely self-determined” (203).  But his denial of the claim that “‘transcendence’ is an independent variable” (203) implies the rejection of even partial autonomy.  It is now customary to take the political, the economic, the sociological, psychological, and the cultural realms all to involve independent variables, so that none is wholly reducible to any or all of the others. Preus, however, is suggesting that religion, usually taken as central to the cultural realm, is completely reducible to one or more of the other realms.  Indeed, his rejection of the claim that “religion is [even partly] a response to transcendence” (204) is simply the reverse side of his view that “psychosocial causes” provide a sufficient explanation for religious experience (161, 197).  This notion reflects Preus’s conviction that Freud and Durkheim have provided the most complete theories of religion thus far, and that a fully adequate theory will need somehow to combine their insights (158, 161f., 196, 209).  As Preus’s own account shows, however, neither Freud nor Durkheim came close to providing an adequate account of religion, and Preus himself does not even try to show how their radically different views could be combined into an adequate psychosocial theory. His confidence that psychosocial causes will prove sufficient for explaining religious experience, accordingly, seems to be based on what Freud declared to be the basis of religion: wishful thinking.

Preus’s wishful thinking here seems to have two aspects, one of which is simply the wish to have religion understood in some purely reductionistic way.  But another aspect seems to be the wish to belong, to have the study of religion fully accepted as part of the academy.  For example, after saying that theology in all its modes “fails as a unifying paradigm because its peculiar insistence that religions are ‘manifestations of the sacred’ isolates it from the presuppositions that are otherwise operative in the humanities and social sciences” and that this insistence “excludes fruitful intellectual interaction with people in other disciplines,” Preus makes the amazing statement that “[t]he issue is not whether ‘transcendence’ refers to something extramentally real, but whether the study of religion wishes to enter as a full partner in the study of culture” (210).  It is disappointing at the end of Preus’s long book to learn that it may have been motivated less by the concern for truth than by the concern to belong.

We can, nevertheless, assume that Preus really does believe that “transcendence” does not refer to something extramentally real, so that a true account of religious experience would be a reductionistic account.  We are still left, however, with no good reason for accepting this view.  I have argued thus far that this view is not entailed simply by the rejection of epistemic supernaturalism combined with the acceptance of domain uniformitarianism.

3. A third dimension of Preus’s epistemic natural-ism is indicated by his statement that the theories of Durkheim and Freud were crucial for the naturalistic paradigm “because their allegation that psychosocial causes are at the root of religious experience was subject to testing through scientific methods of observation and correction” (161f.).  This doctrine, which can be called epistemic naturalismeg (for “empirical groundedness”), is used by Preus to differentiate the scientific from the religious or theological perspective: “The critical difference is that the vitality and fruitfulness of the Freudian-Durkheimian approaches is due to the reciprocal relation that holds between theory and data” that allows these approaches to have “progressive and self-correcting results” (195).  By contrast, “theology is controlled,” Preus says, “by no reality principle” (195).

One can agree with Preus that any academic theory of religion should be experientially grounded, however, while doubting that this criterion necessarily rules out theological theories.  Various forms of theology, some of which have used the label “empirical theology,” have insisted that all talk of a Divine Reality be rooted in immediate experience. Preus pauses to acknowledge that some examinations of experience have led to theological revisions, but he dismisses these efforts derisively, saying that “one wonders whether that experience is “religious experience” as much as it is the experience of implausibility and the demand for apologetic projects aimed at legitimating religion itself in the modern world” (196).

4. This dismissal, by seeming to rule out a priori the possibility of genuine religious experience, points to the fourth and decisive dimension of the epistemic side of Preus’s naturalism, which can be called epistemic naturalisms (for “sensationism”).  This doctrine further specifies the previous one, insisting that to be naturalistic a theory of religion must be based not simply on empiricism but on sensate empiricism, according to which we can have no experience of anything beyond ourselves except through our physical sensory organs.  That this is Preus’s meaning is suggested by his statement that “naturalistic theories of religion” were implied once revelation and innate ideas were rejected, so that “religion had to be accounted for as a response to experience” (54f.)—which seems to mean that religion had to be accounted for as a response to sensory experience.  This assumption seems to be common to all the figures in Preus’s historical account of the naturalistic paradigm for interpreting religion.  For example, this paradigm was, according to Preus (xiv-xv, 84), first fully expressed in Hume, and “sensate empiricism” is virtually synonymous with “Humean empiricism.”  And Durkheim, one of those said to have brought this tradition to its culmination, is cited as saying that the task of the scientific study of religion is “explaining ‘the sacred’”—that is, explaining why religious people see the world in terms of the distinction between the “sacred” and the “profane,” even though “nothing in sensible experience seems able to suggest the idea of so radical a duality to them” (168).  The suggestion that the idea of the sacred might arise in “nonsensible” experience is not seriously entertained: although distinctively human faculties have often been explained theistically, Durkheim rejects this explanation of such faculties with the comment that it would “attach them to some superexperimental reality. . . whose existence could be established by no observation” (165), a comment that Preus quotes without demurer.

The centrality of sensationism for Preus’s understanding of naturalism is also suggested by the fact that William James is not included in his canon of major theorists of religion.  James, of course, rejected Humean empiricism in favor of a “radical empiricism” that allowed for nonsensory perception. Indeed, one of James’s most famous statements, that it takes only one white crow to prove that not all crows are black, was aimed against the dogma that there is nothing in the intellect that was not first in the senses (1986:131).  It was this nonsensationist form of empiricism that allowed James to take religious experience seriously, as possibly revelatory of a “larger power” (1961:407).  It is probably precisely because of this fact that James, who is included in most lists of major theorists of religion, is not included in Preus’s book.  It is, in any case, this dimension of Preus’s epistemic naturalism that lies behind his conviction that the existence and persistence of religion must be explained on the assumption that “God is not given.”  This conviction is, however, also supported by the ontological side of his naturalism, to which I now turn (and to which the latter four of Preus’s meanings of “naturalism” belong).

5. The most obvious meaning of ontological natu-ralism, which can be called ontological naturalismns (for “nonsupernaturalism”), is the rejection of the belief in a supernatural being that can interrupt the world’s normal causal processes.  Naturalism in this sense is the most important aspect of ontological uniformitarianism, the insistence that no causes operated in the past other than the kinds of causes operating today.  This doctrine is exemplified in Preus’s discussion (132, 134, 135) of E. B. Tylor. Saying that the uniformity of civilization can largely be ascribed to “the uniform action of uniform causes,” Tylor rejected all “considerations of extra-natural interference.”  Against Archbishop Whately’s idea that the emergence of civilization out of savagery required a supernatural revelation, Tylor said that such a revelation would be “an impossibility according to the present course of nature.”  Preus’s endorsement of ontological naturalismns is also exemplified in his statement that Vico’s explanation of religion “was naturalistic except for . . . a divinely implanted, universal religious sense” (206).

The idea that the academic study of religion should be naturalistic in this sense seems to me entirely correct.  But ontological naturalismns does not necessarily exclude theistic explanations, because many theologians believe that theism should be naturalistic in this sense.  Preus does note that the incommensurability of naturalistic with theological paradigms is today “much less clear” than it was in “the days of traditional theology,” especially given the existence of “[theological] campaigns against ‘super-naturalism,’” but he seeks to overcome this difficulty for his general thesis by mocking this type of theology, repeating Freud’s view of refined concepts of God as “pitiful remnants of the mighty Father of tradition” (196).  Preus here simply resorts to one of the standard moves of those who wish to dismiss theism in all its forms: Theologians, to be real theologians, are required to hold the supernaturalistic idea of deity, which the critic can then reject as obviously false.  It is as if a physicist, to be a real physicist, would have to hold all of Newton’s ideas.  In any case, Preus does believe that naturalism excludes theism of any form, because his ontological naturalism has dimensions other than simply the rejection of supernatural interruptions.

6. The dimension most directly pertinent to the subject at hand can be called ontological naturalisma (for “atheism”).  It could also be called ontological naturalismfc (for “finite causation”), because it is the doctrine that there are no causal powers beyond the totality of finite causes.  This doctrine was already illustrated in our earlier discussion of Preus’s naturalism, with its reductionism, its denial that religion is rooted in any “objective, transcendent ground,” and its denial of any “mysterious transcendent powers beyond the realm of natural [read ‘finite’] causation” (xvi, xx, 174).  The fact that Preus’s atheism is ontological, not simply epistemic and methodological, is shown by the fact that he rejects the “absolutes” not only of confessional theology but also of natural theology (xiv).  It is also reflected in his statement that “Hume produced a thoroughgoing naturalistic critique of all available theological explanations of religion, whether rationalistic or revelational” thereby “undercutting] all appeals to supernatural or transcendent causes of religion” (xv; emphasis added).

7.  A third dimension of Preus’s ontological position can be called ontological naturalismm (for “materialism”).  This doctrine, which denies not only theism but also the idea of a soul or mind distinct from the physical body, is especially relevant to Tylor’s equation of religion with “animism,” understood as belief in spirits, both creaturely and divine (139).  This relevance is shown by Preus’s summary of Tylor’s definition of religion as “belief in spiritual beings (that do not exist)” (139).  “In Tylor’s view,” Preus says, “the world is now divided between only two completely opposed doctrines: animism and materialism” (144).  One implication of the materialistic denial of animism was Tylor’s rejection of “an animating, separable, surviving entity” (143). Besides making life after death inconceivable, however, the rejection of an anima also provides support for the sensationist doctrine of perception: if what we call the mind is not distinct from the brain, it would be natural to assume that the mind has no means of perception other than the brain’s sensory system.  Preus’s materialism thereby provides additional support for his assumption that there is no genuine religious experience.

8. The Tylor-Preus rejection of animism has yet another implication: the rejection of personal causation.  In Tylor’s view, says Preus, “Science has gradually deanthropomorphized the universe, repla-cing the idea of personalistic causality, life, and will, with the impersonal notions of natural laws and forces” (143).  Preus’s endorsement of this idea reflects a final dimension of his naturalism, which can be called ontological naturalismnl (for “natural law”), because it holds that everything, from the behavior of subatomic particles to the operation of the human mind and the course of history, is governed by natural laws.  Preus describes Tylor’s uniformi-tarianism as his belief in “the universal extent of the operation of natural laws” (152), and he quotes Tylor’s view that “the history of mankind is part and parcel of the history of nature, that our thoughts, wills, and actions accord with laws as definite as those that govern the motion of waves” (133).  The fact that Preus accepts this view is shown by the fact that he criticizes Tylor for not holding to it consistently.  Following the lead of and Marvin Harris’s “cultural materialism,” Preus points out that Tylor continued to assume that “mind determines itself,” that [b]elief itself is a ‘cause,’” and that “reasons are not merely rationalizations but the actual causes of actions and events,” so that “causes need not be sought beneath reasons” (148, 149).  Preus, by contrast, agrees with H. P. Rickman that “in the human sphere, as well as the physical,” we need explanation in the sense of “the discovery of general laws that govern the connections between events” (79).  This feature of Preus’s materialism adds yet another reason for religion scholars to reject the idea of divine influence and the idea of a distinct mid (which could conceivably experience such influence), because both ideas would run counter to the assumption that all things are determined by impersonal causes.

In any case, now that we have before us the various dimensions of what Preus considers a fully developed naturalism, we can ask whether all these dimensions are properly presupposed by the academic study of religion.  Because our ultimate question is whether the academic study of religion necessarily excludes all theological interpretations of religion, we can approach this question by asking which dimensions of Preus’s naturalism might be compatible with a theological perspective.  The first and fifth doctrines, namely, epistemic naturalismnsr and ontological naturalismns, would, of course, be incompatible with supernaturalistic theologies, which affirm the possibility of supernatural interruptions, including supernatural revelations.  There are, however, forms of theism that reject ontological supernaturalism and thereby epistemic supernatural-ism.  These naturalistic theisms, furthermore, would not necessarily be incompatible with the second dimension of epistemic naturalism, namely, domain uniformitarianism.  Some forms of naturalistic theism, in other words, say that, insofar as the category of divine influence is needed to explain religious experience, it is also needed to explain other types of human experience and perhaps all kinds of events whatsoever.  A naturalistic theism, furthermore, would also probably agree with the third meaning of epistemic naturalism, that an explana-tory theory should be experientially grounded, so that the theory can be tested against observations. There would be a conflict, however, between naturalistic theism and epistemic naturalisms, because the sensationist doctrine of perception rules out the possibility of genuine religious experience of a divine actuality.  Accordingly, a naturalistic theism, having rejected supernatural revelation, could have no empirical grounding.

Besides being incompatible with sensationism, a naturalistic theism would also be in conflict with the latter three doctrines of Preus’s ontological natural-ism—that is, with its atheism, its materialism, and its insistence that all things are determined by natural laws.  Because the last of these doctrines, ontological naturalismnl, can be regarded simply as an aspect of ontological naturalismm, we can reduce the problematic doctrines to three: sensationism, atheism, and materialism.  This threefold form of naturalism can, therefore, be abbreviated as natural-ismsam. (The fact that Preus’s nickname is “Sam” is, of course, purely coincidental.)

The question as to whether the naturalism proper to the academic study of religion necessarily excludes a theological explanation of religion is, therefore, the question of whether the study of religion should presuppose naturalismsam.  That is, given the acceptance of doctrines 1, 2, 3, and 5—that is, of ontological naturalismns and all the meanings of epistemic naturalism except sensationismm—we would have a “minimal naturalism,” meaning the naturalistic presuppositions that are arguably necessary for participation in academic, including scientific, conversations.  The Preus-Segal proposal, however, is that those wanting to be “full partners” in the academic study of religion should also affirm naturalismsam.  It is to this issue that I now turn.


3. Should the Academic Study of Religion Presuppose Naturalismsam?

One basis for a positive answer to the question serving as the heading of this section would be the assumption that “naturalism” as currently under-stood is all of a piece, so that the rejection of supernatural interruptions entails sensationism, atheism, and materialism.  We have seen, however, that one could accept the other dimensions of Preus’s naturalism without accepting naturalismsam.  The agreement that we need to have an approach that is minimally naturalistic, as distinct from supernatural-istic, does not settle the question as to whether naturalismsam, which rules out all theistic explana-tions of the existence of religion, should be accepted.

One reason why some thinkers prefer natural-ismsam is, of course, precisely the fact that it does rule out all theistic explanations.  But this is not a valid reason.  For example, Preus, complaining about “hidden apologetic intentions” (xxi), rightly says that answers to the question of the true causes of religious belief cannot be limited to those that support religion, because the goal of the academic study of religion “is not to legitimate religion but to explain it” (xx).  The Preus-Segal program, however, would limit the answers that could count as academic explanations of religion to those that delegitimate it. To his credit, Preus points out that the field is “still struggling with the question of how to address these problems in a more unified way without creating a new dogmatism that would undermine pluralism” (210).  Segal, likewise, speaks in favor of “multiple programs,” saying that “competition is indispensable to staving off dogmatism” (1992: 39). However, the present proposals of Preus and Segal, if accepted, would involve a dogmatic exclusion of theistic accounts, even if such accounts otherwise exemplified the naturalism presupposed by the academy.  In any case, negative apologetic intentions do not, any more than positive ones, provide a valid basis for deciding upon an appropriate conceptual framework for the study of religion.

Another reason for assuming that a nontheistic framework is needed for the academic study of religion, at least in state-supported schools in the United States, is the widespread belief that theistic explanations would violate the First Amendment of the U. S. Constitution.  As I have argued elsewhere (1991), however, such explanations are not excluded if they are based on rational-empirical grounds, rather than on confessional (in the sense of epistemically supernaturalistic) grounds.  Further-more, the Constitution enjoins the state not only from supporting religion but also from showing hostility to it.  Therefore, given the fact that antitheistic interpretations are allowed in the academy (as in the advocacy of the neo-Darwinian theory of evolution), it is the exclusion of all theistic interpretations that would actually be unconstitu-tional.

One basis for Preuss advocacy of naturalismsam seems to be his conviction that for the study of religion to overcome its current “identity crisis” (xvii), it should return to “the path of critical inquiry that characterized the study of religion in its vital formative period—namely, the proposal of alternatives to the explanations that the religious offer for religion” (xx).  However, even if that was heretofore the most vital period in the study of religion, vitality is not necessarily indicative of truth, which should be our goal.  (Vitality, furthermore, is likely to be engendered today by employing a new approach, not by returning to the program of yesteryear.)

Preus’s advocacy of naturalismsam also seems to be influenced by his view that the Enlightenment’s aspiration to explain everything, including religion, is “fully operational” only when “every aspect of religion [is] regarded as amenable to nonreligious explanation” (xvi).  That, however, is mere prejudice.  As we saw earlier, Preus (rightly) advocates an approach that “pushes explanation to the limit” (210).  Many thinkers seeking to do this, however, have found that they have been led to theistic explanations.5  It would be arbitrary to allow the assumptions of some Enlightenment figures to determine what types of ultimate explanations are permissible in today’s academy.

Yet another motive for Preus’s advocacy of naturalismsam is his desire for domain uniformity, so that religious studies can be “a full partner in the study of culture” (210).  Preus, as we have seen, indicts theology because “its peculiar insistence that religions are ‘manifestations of the sacred’ isolates it from the presuppositions that are otherwise operative in the humanities and human sciences” (210).  He is right to say that this isolation needs to be overcome, that the study of religion should employ the same explanatory categories that are used elsewhere in the academy.  But he evidently simply assumes that the only way to achieve this goal is for the study of religion to accommodate itself to the currently dominant “rough consensus of the academy at large” (210)—which would imply the acceptance of naturalismsam.  Another way to achieve domain uniformity, however, would be for religious philosophers and theologians to convince the remainder of the academy that a categorial revision is needed in order to achieve more adequate explanatory theories in all domains, a revision that might involve the rejection of materialism and the addition of the categories of nonsensory perception and divine influence.  This program might seem like a pipe-dream, but whether it should be attempted rests on a judgment as to which form of naturalism seems the closest approximation to the truth, and that is a philosophical judgment.

A possible argument against this program might be based on the widespread assumption that naturalismsam simply is the “modern scientific worldview,” so that it is necessarily presupposed in the social sciences and thereby in the academy in general.  As I have argued elsewhere (2000a), however, the current association of science with this worldview is due to highly contingent developments. At the root of these developments was the theologically motivated adoption, in the latter half of the seventeenth century, of the sensationist doctrine of perception and the mechanistic-materialistic doctrine of (physical) nature for the purpose of supporting a dualistic doctrine of human beings and a supernaturalistic doctrine of the universe as a whole. Naturalismsam resulted simply from decapitating that early modern worldview, lopping off its human and divine minds—not from looking at all the evidence and rationally deciding that a sensationist, atheistic, materialistic worldview could best do justice to it.

So, although it is indeed sociologically the case that this worldview is currently dominant in scientific circles, the crucial issue, which has not been adequately debated, is the philosophical question of whether this worldview should continue to remain dominant.  We should seek to explain the existence of religion from the perspective of this worldview only if an affirmative answer to this question seems warranted.  The only valid reason for advocating naturalismsam for the academic study of religion, in others words, would be the philosophical argument that this form of naturalism is superior to all other forms, providing a more adequate framework for interpreting all the evidence of human experience, including scientific experience.  Otherwise, there would be no reason for an a priori rejection of theistic interpretations of religion, at least if they embody naturalism in the minimal sense.

Although Preus and Segal both seem to realize that the issue is finally philosophical, Segal has evidently so fully accepted the identification of science with naturalismsam that he sometimes suggests that the question whether religion is to be understood religiously or nonreligiously is a (social) scientific question, as distinct from a philosophical, issue.  For example, against the view that the ultimate nature of religion is knowable a priori, Segal says that “[t]he issue must be settled by research” (1992:7).  Also, after saying that Freud and Marx deny that overtly religious data are most deeply religious, Segal says that “they do so nondogma-tically: by demonstrating how the subject matter is better categorized psychologically and economically” (1992:46). The terms “research” and “demonstra-ting” suggest that the issue is primarily empirical.

Elsewhere, however, Segal points out that for those classical social scientists such as Tylor, Frazer, Marx, and Freud, who declare religion to be false, “religion is false on philosophical, not social scientific, grounds.”  They, in other words, “do not argue [the falsity of religion] on the basis of their social scientific findings” but instead “argue for the secular origin and function [of religion] on the grounds of the falsity of religion,” which, they assumed, had been established philosophically (1992:16). (This judgment, incidentally, fits with Tylor’s statement, cited by Preus [142], that Hume’s Natural History of Religion “is perhaps more than any other work the source of modern opinions as to the development of religion.”)  It is hard to see, therefore, how Segal can suggest that Marx and Freud demonstrated the (entirely) secular or nonreligious origin of religion through social scientific research.  Segal’s alleged conflict between “religionists” and the “social sciences” is really, by his own account, simply a conflict between religious and nonreligious philosophies.

The same ambiguity shows up in the different questions he assigns to philosophy and the social sciences.  “The assessment of [the truth of] religious belief,” says Segal, “is the task of philosophy” (1989: 75).  The social sciences, however, have the task of assessing the origin of religious belief and thereby “the truth of religious explanations of religious belief” (1989: 75).  We have, therefore, an apparent division of labor: “Philosophy determines whether the reasons believers provide for believing are sound reasons for believing.  The social sciences determine whether the reasons believers provide for the coming to believe are truly the reasons for which they have come to believe” (1989:75).  It turns out, however, that the assessment of truth is not left to philosophy, because “a social scientific explanation of the origin of religious belief has [some] bearing on the truth of the belief” (1989: 75).  This is because, as we recall, Segal insists that social scientific explanations are, virtually by definition, nonreligious explanations. Therefore, “A social scientific explanation, once accepted, renders the truth of religious belief improbable” (1989: 79).

On this point Segal is entirely correct against those who dismiss, as instances of the “genetic fallacy,” all arguments based on the origins of belief. If a (nontheistic) social science account provides a sufficient explanation of that origin, then a theistic account of that origin is, as Segal says, superfluous (1989:78).  And if divine influence is thereby shown not to be even a partial cause of the origin of religious experience, then “the social sciences undermine the genetic justification for the truth of religious belief” (1989:82).  Segal is right, furthermore, to consider this genetic justification to be so important that, if it is undermined—so that we would have no reason to think that religions originated even partly through divine influence or, otherwise put, through genuine religious experience—then the truth of religious belief in a divine reality would be undermined, because it would be implausible to assume this belief just coincidentally to be true.

We are here at the heart of what Segal considers the “confrontation” between religionists, who defend the religiosity of religion, and the social sciences.  But is there really such a confrontation?  In one place Segal “den[ies] that the social sciences always, let alone must, deem religion false” (1992:27).  Also, as we saw earlier, he indicates that a person can be a social scientist, even if an “unconventional” one, while making a “social scientific affirmation of the truth of religion” (1992:15, 107).  It would seem, therefore, that it is not the social sciences as such that are antithetical to religious interpretations of religion but only conventional social scientists, those who are in continuity with the “classical” social scientists, such as Marx, Tylor, Frazer, Comte, Durkheim, and Freud.  And, as Segal himself repeatedly says (1992: 16, 31, 123; 1989: 80 n. 15, 83), these classical figures did not come to regard religious beliefs to be philosophically false because of scientific investigations into their origins.  Rather, they developed these “scientific” accounts on the basis of their prior philosophical convictions as to the falsity of religion.  Segal’s view, incidentally, corresponds with the position of E. E. Evans-Pritchard, who wrote:” [T]he persons whose writings [on primitive religion] have been most influential have been at the time they wrote agnostics or atheists . . . . Religious belief was to these anthropologists absurd. . . . But some explanation of the absurdity seemed to be required, and it was offered in psychological or sociological terms” (1965:15). Given this Pritchard-Segal view, then, the confrontation is not between religionists and the social sciences, as if the latter were necessarily antireligious, but only between religious and anti-religious philosophies.  The issue must, therefore, be adjudicated in terms of philosophical criteria.

It might be argued, to be sure, that, although the motivation to look for secular causes of religious beliefs first came from a prior philosophical conviction as to the falsity of religion, these secular explanations, once provided by these social scientists, do show the falsity of religious belief by revealing that it originates in purely secular causes. This would indeed be the case if these account really showed religion to be rooted in purely secular causes.  It would not be the case, however, if they only show that any concrete religion is significantly shaped by psychological, sociological, and economic factors.  Segal’s recognition of this point is reflected in his statement that the (conventional) social scientist claims to be offering a sufficient cause of religious experience (1989:82) along with his statement that every social scientific explanation involves “a naturalistic rather than a divine origin” (1992: 19; emphasis added).  This same recognition is also reflected in the fact that after saying “surely religion can have an origin that is partly sociological and partly religious,” Segal quickly dismisses this possibility as “unlikely” (1992: 9).

We must ask, however, what the basis is for assuming this partly/partly view to be unlikely.  What is the basis, in other words, for assuming that the evidence that religious beliefs are always shaped by psychosocial factors also shows that they are not also partly rooted in genuine religious experience? No empirical evidence could provide a sufficient basis for such a sweeping negative claim.  The basis, it seems clear, is nothing other than a prior conviction of the truth of naturalismsam, which implies that we have no capacity to perceive anything other than material objects and that, even if we did, there is no divine reality to perceive.  The basic question, then, is whether naturalismsam is, as widely assumed, the most adequate philosophical position available.


4. Is Naturalismsam Philosophically Warranted?

Both Preus and Segal show that, besides the other reasons they may have for favoring naturalismsam, they presuppose that its endorsement by the scientific community is justified, so that theistic interpretations of religion can be excluded as antiscientific and thereby false.  Preus says, for example, that “Tylor rightly saw religious explanations of the world as relics” (209; emphasis added).  Preus’s justification of this stance, however, is remarkably brief, consisting of a one-sentence argument: saying that his naturalistic approach does not take religious explanations seriously, Preus adds: “for, as Occam held, explanatory entities (such as ‘transcendence’ . . .) must not be multiplied beyond necessity” (211). This argument, of course, begs the basic question, because it simply assumes that a causal power transcending the totality of finite causes is not necessary to explain religion and perhaps much else besides (which Occam himself certainly did not assume).  Although simplicity should be a desideratum, adequacy is surely the decisive criterion.

Segal, besides also appealing to this argument from simplicity (1989:78; 1992: 71), provides some other arguments for the superiority of “a naturalistic, social scientific explanation” to a religious explanation, which he assumes would necessarily be a “supernatural explanation” (1989: 78).  One of these arguments is that a (nonreligious) social scientific account of religion, “however inadequate, is more adequate than a religious one” (1992:70). Segal’s reason for this contention is that social scientists “provide a host of processes and entities like projection, wish fulfillment, complexes, collective representations, and symbols to account for how and why religion originates and functions,” whereas religionists, such as Schleiermacher, Otto, and Eliade, “provide nothing”—except “the litany that religion originates as a response to the transcendent” (1992: 70).

This argument calls for several comments.  First, Segal is right that most philosophers of religion and theologians fail to give any explanation of how a divine reality might be experienced, especially an explanation that would pass the test of domain uniformitarianism.  It is false, however, to suggest that nothing by way of explanation has been provided by any religionists, even if we restrict our view to the three that Segal mentions. Schleiermacher, for example, certainly took great pains to explain how the (nonsensory) “feeling of absolute dependence” could arise.  Segal’s contention that Schleiermacher provided no explanation whatsoever perhaps simply reflects a belief that nonsensory perception is unintelligible. Other religious philosophers, furthermore, have arguably done better than Schleiermacher.  Second, in saying that a social scientific account, however inadequate, is better than a religious one, Segal is evidently forgetting that he has set a far higher standard for a (nonreligious) social scientific account to be adequate: it must provide a sufficient explanation for the existence and persistence of religion.  The mere fact that a (nonreligious) social scientific account provides something, therefore, does not necessarily obviate the need for a religious account.  Third, religious explanations, by contrast, need not claim to provide sufficient explanations of religion.            

Needing only to account for the religious nature of religions, they can rely on various psychosocial explanations to account for most of the concrete  details of the various religions.  Therefore, unless naturalismsam is already  presupposed, antecedent probability would arguably favor an explanation that combines religious and nonreligious factors.                                              |

Segal’s third argument, however, claims antecedent probability for a purely nonreligious account, saying: “[A] social scientific explanation of religious belief has a higher prior probability than a supernatural one.  For it is linked to natural science, which has provided the most persuasive explanation so far of the physical world. Social science . . . extends to the study of humans the nonsupernatural framework of natural science” (1992:79).  This argument, which illustrates especially well the importance of distinguishing among the various meanings of “naturalism,” seems to involve the following points:                                                                        

1.    Natural science has provided the best explana-tions of the physical world.

2.    Natural science employs naturalistic rather than supernaturalistic explanations.

3.    Social science, studying human beings, also employs naturalistic rather than supernatural-istic explanations.

4.    Being naturalistic, social scientific explana-tions of human religiosity are, therefore, antecedently more probable than religious explanations, which are supernaturalistic.

To assess the soundness of this argument, we would need to unpack ambiguities and other problems in each thesis.  For example, the first thesis—that natural science has provided the best explanations of the world—evokes the question, “Compared with what?”  That is, natural science, which was previously called “natural philosophy,” is by definition the enterprise devoted to giving explanations of the physical world.  (The only intended contrast could be with “scientific creationism,” but it is seldom taken seriously by religious philosophers and theologians in the academy.)  So the statement that it has provided the best explanations of the physical world is tautological.  The important question is how adequate the kinds of explanations allowed by naturalismsam are.  Insofar as they are inadequate, the need for a better set of explanatory categories is suggested.  I will return to this question below.

The second thesis, which asserts that natural science uses naturalistic explanations, is not true of modern natural science from the seventeenth century through much of the nineteenth century, which occurred within a supernaturalistic, even if increasingly deistic, framework.  So the scientific successes of this period cannot be credited to a nonsupernaturalistic, let alone fully atheistic, worldview.

Regarding the second, third, and fourth theses: insofar as the scientific community has had a better explanatory framework since the latter part of the nineteenth century by virtue of rejecting supernaturalistic explanations, this fact does not entail that any of the scientific successes of this period have been due to the acceptance of sensationism, materialism, and (complete) atheism. The assumption of naturalismsam by some social scientists, accordingly, provides no reason for expecting their explanations of religion to be superior to explanations that, while being nonsupernatu-ralistic (because they accept ontological naturalism ns and epistemic naturalismnsr), would not presup-pose naturalismsam.                                          

Indeed, I suggest, we should expect the opposite—that social scientific        explanations will be inadequate precisely insofar as they do presuppose naturalismsam, because—to return to the discussion of the first thesis—scientific explanations in general have been inadequate insofar as they have presupposed naturalismsam.  This is obviously a huge topic.  I can here only mention a few issues, pointing in the references to my discussions of them elsewhere.  I will begin with the epistemological dimension of naturalismsam, its sensate empiricism.

Those in the naturalistic tradition endorsed by Preus and Segal, as we have seen, reject the theistic interpretation of religious experience.  That is, although it has seemed to countless numbers of people that they have had genuine religious experiences in the sense of experiencing a divine or holy power distinct from all worldly powers, these naturalists declare that these experiences were not what they seemed to be.  The epistemological assumption behind this a priori dismissal of the experiencers’ self-interpretations seems to be the assumption that the sensationist doctrine of perception, which rules out such religious experiences, is perfectly adequate for science and common sense.

This, however, is not the case.  In the first place, one of the scientific disciplines, parapsychology, has over the past century been providing increasingly incontrovertible evidence for extrasensory percep-tion (see Griffin 1997: Chs. 1-2).  The fact that the ideological leadership of the scientific community has been committed to sensationism has created the sorry spectacle of scientists, who are supposed to value empirical evidence above all else, denouncing the empirical results of fellow scientists in the name of an a priori dogma (Griffin 1997:117-119, 283-287; 2000a: Ch. 7).

However, it is not only or even primarily in relation to the controversial science of parapsychology that the sensationist doctrine of perception has created problems for science.  More serious is the fact that it makes impossible any empirical grounding for many ideas that are inevitably presupposed in all our practice, including our practice of science.  Since the time of Hume, for example, it has been widely recognized that sensate empiricism leaves us with no explanation of our apparent knowledge of causation and of the external world.  With his famous phrase “solipsism of the present moment,” Santayana added our knowledge of the past and thereby of time to the items rendered mysterious by this form of empiricism.  The assumption that we have no nonsensory perceptions and therefore no intuition of moral norms, furthermore, has led either to the conclusion that all moral beliefs are noncognitive or to various unsuccessful attempts to explain how they can involve genuine knowledge.  The sense that truth is important is also left without an explanation.  It is emphatically not the case, therefore, that the sensationism of naturalismsam supports science over against religion.  It is as threatening to the rationality of science as it is to the rationality of religion.

It is highly arbitrary, accordingly, to employ it against the reality of religious (and perhaps telepathic and moral) experience while not using it to cast doubt on the reality of causal efficacy, the external world, the past, and time.  In other words, the fact that sensationism cannot allow for experience of a divine reality says nothing about the possibility of such experience, given the fact that sensationism also cannot allow for experiential knowledge of all these other things.  It would seem to be sensationism that is in trouble, not genuine religious experience.  Indeed, sensationism’s inability to allow for genuine religious experience becomes, from this perspective, simply one more sign of its general inadequacy.

The materialism of naturalismsam also creates its share of problems.  Some of these follow directly from that feature of materialism that puts it most in conflict with religious beliefs, its doctrine that the mind or soul is identical with the brain.  This identism, as we have seen, supports sensationism, so the problems for sensationism are thereby problems for materialism.  But this materialistic identism is also used to deny the possibility of life after death by denying that there is any soul or spirit that, being distinct from the brain, could conceivably exist apart from it.  The problem created for science by this doctrine is that it is in tension with a growing collection of well-documented evidence for out-of-body experiences that is being provided not only by parapsychologists but also by members of the medical profession (Griffin 1997: Ch. 8).  Having an experience of seeming to be out of one’s body, to be sure, does not necessarily mean that one really is, and many students of these experiences reject the extra-somatic interpretation of them.  The empirical evidence, however, is far more consistent with this interpretation, whereas the only support for the intrasomatic interpretation is the philosophical assumption that it must be true (Griffin 1997: 256-261).

The main difficulties created for science by materialism, however, involve more commonplace features of the mind-body relation.  The most obvious problem is that of explaining how conscious experience could have emerged in the evolutionary process, given materialism’s assumption that the ultimate units of nature are wholly devoid of experience.  An increasing number of philosophers are declaring this problem to be insoluble in principle (Griffin 1997: Ch. 3; 1998a: Intro., Chs. 1, 7). Given the fact that the existence of conscious experience is that of which we are most certain, it is hard to imagine a failure more damning to a philosophical position than the inability to explain how conscious experience could exist.

Equally problematic, however, is the twofold problem of accounting for mental causation and freedom.  We all inevitably presuppose in practice that we and other people have beliefs, motives, and plans, and that these mental states causally influence our behavior.  As illustrated by Preus’s criticism of Tylor, however, a fully consistent materialistic theory must deny that they do.  Mental states must be regarded always as concomitants or epiphenomena of bodily states, never the cause thereof.  This conflict with the inevitable presuppositions of our practice, including our scientific practice, constitutes a reductio ad absurdum of materialism, as a leading materialist philosopher, Jaegwon Kim, has virtually admitted (Griffin 1998a: Ch. 10).  Even more materialist philosophers agree that materialism cannot allow for real human freedom, that is, that the mental causation is based on an element of self-determination.  And yet, as some of these philosophers, including Thomas Nagel and John Searle, admit, we cannot help presupposing in practice—again, including our scientific practice—that we and other people are partly responsible for our actions because we have a degree of genuine (self-determining) freedom (Griffin 1998a: 38-40,163-70). This contradiction implies another reductio.

Although materialism creates yet other insoluble problems, such as the problem of making sense of the reality of time prior to the time (!) at which experience is assumed to have emerged (Griffin 1986; forthcoming), I will move on to problems created for scientific explanation by the atheism of naturalismsam.  Perhaps the best known difficulty is that of explaining the order of the universe.  One special form taken by this difficulty in our time is the problem of accounting for the apparently “fine-tuned constants” that evidently existed at the outset of our universe, a problem that has led a number of scientists to affirm the existence of a cosmic intelligence.6

Not only cosmic evolution but also biological evolution, furthermore, has features that are difficult to explain apart from the assumption of a cosmic power transcending the totality of finite, local causes.  One such feature is the evident directionality and progress of the evolutionary process, which must be counter-intuitively denied by staunch neo-Darwinists (see Nitecki).  Another problem, created by the atheism of neo-Darwinism in conjunction with its materialism, is the inability to do justice to the real novelty that has emerged many times, novelty that is obviously more than simply the complex rearrangement of elementary parts.  (The emergence of animal experience and then of distinctively human experience are simply two of the most recent illustrations of this more general emergence of real novelty.)  Yet another problem is that neo-Darwinism, because of its atheism, entails that evolution must occur very gradually, in the sense that each step is quite tiny, whereas both the fossil evidence and conceptual considerations suggest that all the major transitions must have involved rather sudden jumps (Griffin 2000a: Ch. 8).

Still more problems are created by truth, beauty, and goodness.  The world seems to have far more beauty than can be explained by the utilitarianism of neo-Darwinism, as some of its defenders admit (e.g., Weinberg: 250).  And the problem of accounting for the objectivity that we inevitably ascribe to truth and moral norms, which was discussed in epistemological terms earlier, is intensified by the denial that the universe contains a locus in which truth and normative values could subsist and by which they could be given causal efficacy (see Griffin 1997:273-287; 1998a: 203-209).

In sum, the point I made above with regard to sensationism in particular can be made with regard to naturalismsam as a whole: the fact that it is incompatible with genuine religious experience arguably says less about religion than it says about naturalismsam.  That is, given the realization of how inadequate this version of naturalism is to a wide range of indubitable and other well-grounded facts, we can take the incompatibility of religious experience with this version of naturalism as simply one more sign of the latter’s inadequacy to serve as the framework for the natural sciences, the social sciences, and the academy in general.  At the very least, we can say that, given the fact that naturalismsam has not proven itself adequate in relation to a wide range of phenomena, including many inevitable presuppositions of scientific activity itself, it would be precipitous to consider this form of naturalism a standard to which any theory of religion, to be considered scientific or even academic, must conform.


5. Summary and Conclusion

The sciences in particular and the academy in general should, I agree, presuppose naturalism in the minimal sense—meaning primarily the rejection of ontological and epistemic supernaturalism.  This minimal naturalism does, of course, rule out religious belief of a certain type.  But it does not necessarily rule out all significant religious beliefs, including the belief that religions exist and persist partly because human beings have genuine religious experiences. This and other significant religious beliefs are, however, ruled out by the sensationistic, atheistic, materialistic version of naturalism that is now widely regarded as the “modern scientific worldview.”  If this version of naturalism were truly adequate for the natural sciences and the other social sciences, it would make sense, given the desirability of domain uniformity, to say that this version of naturalism should also be presupposed by the study of religion. This version of naturalism, however, is adequate for only a very limited range of purposes.  Far from allowing us to push explanation to the limit in all areas, it threatens to leave a wide range of our inevitable presuppositions and other well-grounded facts in the category of permanent mysteries.  The fact that this worldview is currently dominant among the ideological leaders of the scientific community, therefore, does not imply that it should be adopted by those engaged in the study of religion—assuming, of course, that the concern for truth should take priority over the wish to belong.  But we could move toward domain uniformity in another way: A form of naturalism that is truly adequate for scientific explanations might also allow for a nonreductionistic explanation of religion.  Insofar as my argument is sound, this would indeed be the case, because such a naturalism would move beyond sensationism, atheism, and materialism (but without returning to dualism, the problems of which led to materialism).  I have argued elsewhere (Griffin 2000a; 2000b) that such a naturalism is available.

Here I must rest content with the negative point that the kind of naturalism required by the sciences, both natural and social, does not rule out the possibility of genuine religious experience and thereby the appropriateness of theological explanations of religion.  It should not be assumed, accordingly, that accounts of religion are more scientific or academic simply by virtue of rejecting such explanations.  Just as advocates of religious accounts should not argue that attempts to explain religion reductionistically are to be excluded a priori, advocates of reductionistic accounts should not argue that attempts to provide religious and even theistic accounts are to be excluded a priori.



1 I am uncertain how much they would want to limit the “academic” study of religion to the “social scientific” study of religion, as distinct from simply insisting that the former should include, at least in the sense of being compatible with, the latter. Insofar as I use the two terms interchangeably in discussing the Preus-Segal position, I mean this usage to be open to either understanding.

2 My essay does indirectly address this issue, however, in articulating a framework in which scientific explanations need not be reductionistic and even need not, in fact, exclude the category of divine causation.  My position overcomes, therefore, the fundamental assumption behind the widespread conviction that the methods and categories of the Geisteswissenschaften in general, and especially the study of religion in particular, must be different in kind from those of the natural sciences.

3 Some philosophers and theologians would challenge the second claim, that all academically correct explanations of religion must be social scientific explanations, from a supernaturalist standpoint, saying that the origin of religion, or at least true religion, can only be correctly explained in terms of supernatural intervention.  The second claim could also be challenged, however, by one who 1) affirms a version of naturalistic theism that allows for variable divine influence in the world while 2) accepting the conventional view that scientific explanations by definition methodologically exclude any reference to divine influence.  Although I could endorse such a position, my preferred position is to challenge the third claim by arguing that the very nature of science can and should be reconceived so that scientists, qua scientists, could refer to divine influence, naturalistically understood, as a causal factor.

4 Some religion scholars speak of religion as based on true beliefs, of course, by defining “truth” in a completely relativistic way, so that whatever a community believes is regarded as true.

5 Alfred North Whitehead, for example, endorses “the belief that clarity can only be reached by pushing explanation to its utmost limits” (153).  But when he began trying to find the most complete explanation possible (in moving from the “philosophy of nature,” understood as a limited enterprise, to “metaphy-sics,” understood as including the knower within the system), he found himself led to give up his longstanding atheism, or at least agnosticism, for a (naturalistic) type of theism (see Griffin PPR Ch. 5).

6 Although the establishment of these finely tuned cosmic constants is usually presumed to require a creator with coercive omnipotence and thereby a supernaturalistic theism, I have briefly argued otherwise (Griffin 2000a: Ch. 8).



Evans-Pritchard, E. E., Theories of Primitive Religion. Oxford: Clarendon Press, 1965

Griffin, David Ray, “Introduction: Time and the Fallacy of Misplaced Concreteness.” In Physics and the Ultimate Significance of Time: Bohm, Prigogine, and Process Philosophy, 1-48. Ed. by David Ray Griffin. Albany: State University of New York Press, 1986.

Griffin, David Ray, “Professing Theology in the State University.” In Theology and the University: Essays in Honor of John B. Cobb, Jr., 3-34. David Ray Griffin and Joseph C. Hough, Jr., eds., Albany: State University of New York Press, 1991

Griffin, David Ray, Parapsychology, Philosophy, and Spirituality: A Postmodern Exploration. Albany: State University of New York Press, 1997

Griffin, David Ray, Unsnarling the World-Knot: Consciousness, Freedom, and the Mind-Body Problem. Berkeley and Los Angeles: University of California Press. 1998a

Griffin, David Ray, forthcoming  “Pantemporalism and Panexperientialism.” In The Textures of Time. Ed. by Paul A. Harris. [Time and Uncertainty, edited by Paul A. Harris and Michael Crawford, appeared in 2004 but did not contain any essay by Griffin.]

Griffin, David Ray, Religion and Scientific Naturalism: Overcoming the Conflicts. Albany: State University of New York Press, 2000a

Griffin, David Ray, Reenchantment Without Supernaturalism: A Process Philosophy of Religion. Ithaca, NY: Cornell University Press, 2000b

James, William, Varieties of Religious Experience. New York: Collier Books, 1961

James, William, Essays in Psychical Research. Cambridge: Harvard University Press, 1986

Nitecki, Matthew H., ed., Evolutionary Progress. Chicago: University of Chicago Press. 1988

Preus, J. Samuel, Explaining Religion: Criticism and Theory from Bodin to Freud. New Haven: Yale University Press [reprint ed., AAR Texts and Translation Series, 16. Atlanta: Scholars Press, 1996], 1987

Segal, Robert A., Religion and the Social Sciences: Essays on the Confrontation. Atlanta: Scholars Press, 1989

Segal, Robert A., Explaining and Interpreting Religion: Essays on the Issue. New York: Peter Lang, 1992

Weinberg, Steven, Dreams of a Final Theory. New York: Vintage Books, 1994

Whitehead, Alfred North, Process and Reality. Corrected edition. Ed. by David Ray Griffin and Donald W. Sherburne. New York: Free Press, 1978

David Ray Griffin Page