No one who has the temerity to speak about the broad themes to which this series of lectures is devoted, the themes of “science, religion and human experience”, can hope to hide behind an academic façade of “professional expertise”. To be sure there are matters which inevitably come up in any such lecture which can benefit from being treated with scientific or philosophical sophistication, I believe – otherwise, what am I doing here? But the big issues: to believe in God or not to believe in God; to engage in such religious practices as prayer, attending services, studying religious texts or not to do so (I am not equating this with the issue of believing or not believing in God, by the way); to look for “proof” of God’s existence, if one is religious (or thinking of being religious) or to regard such a quest as misguided; to be “pluralistic” in one’s approach to religion, or to regard one religion as “truer” than all the rest – these are deeply personal choices, choices of who to be, not just what to do or what to believe. I do not believe that philosophical or scientific discussion can provide compelling reasons for making them one way rather than another, although it can help us make whichever choices we make more reflectively. (Avi Sagi once told me that, in a still unpublished fragment of - I think it was a diary – of Kierkegaard’s, he found the words “Leap of faith – yes, but only after reflection.”)

I did say, however, that there are aspects of these issues that a philosophically sophisticated discussion (as well as a scientifically sophisticated discussion) can illuminate. In today’s lecture, I shall say something about one such aspect, namely the notion of experience .

Both in life and in philosophical reflection, experience is sometimes seen as intrinsically shallow, as mere surface, and sometimes as deep. I want to investigate the origins of our notion of experience in Cartesian and post-Cartesian philosophy, and explore with you the relevance of the long-standing philosophical disputes about experience for our broad themes of “science, religion, and human experience”.

The depths and shallows

When I speak of “religious experience” in what follows, I will not mean experience that purports to be of supernatural beings, or of “revelation” conceived of on the model of having words dictated to one by a divine being. (One can find a very different model – the model of revelation as the ongoing connection between the individual and God – in the writing of Franz Rosenzweig. [1] ) Rather, I will have in mind the way in which a religious person may at any time experience something or some event, whether it be an obviously significant one, say the birth of a child or the sort of deep crisis in one’s life that James describes in The Varieties of Religious Experience , or whether it be a superficially “ordinary” one, as full of religious significance. Speaking for myself, I cannot imagine being religious in any sense, theistic or not-theistic, unless one has had and cherished moments of religious experience in this latter sense. Yet the concept of “experience” that we have operated with from Descartes and Hume to today’s cognitive scientists has a troubled history, and it will repay us, I believe, to reflect on that history.

What I shall be talking about for the most part will not be what I just called “religious experience”. Rather, I am going to spend a few minutes trying to explain why so many people have (and from where they got) a concept of experience which leaves literally no room for depth , a conception of experience as, so to speak, all psychological surface , one traditionally summed up in the conception of experiences as “sensations”, and after that I shall try to explain why that conception is wrong, drawing especially on Kant’s profound analysis of experience.

We all know that the philosophers of the seventeenth and eighteenth centuries are classified by the standard texts as “empiricists” and “rationalists”. While the classification is in many ways a Procrustean Bed, it certainly captures a broad divide between, say, the British philosophers Locke, Berkeley and Hume, and the continental philosophers Descartes, Spinoza and Leibniz, and while the pattern of disagreements is by no means as tidy as the labels “empiricism” and “rationalism” suggest, it is certainly true that we find very different conceptions of experience in the two groups, and especially in Hume and Leibniz. (What is not often remarked is that Hume, the empiricist who makes experience - under the name “impressions and ideas” - the be-all and end-all of his philosophy, and who prides himself on being a sort of Newton of psychology [2] , is, in fact, far less subtle in his description of experience than Leibniz. [3] ) Be that as it may, the line that came to be recognized is between conceptions of experience that go back to Hume, and conceptions that go back to Kant (who hoped, of course, to sublate the categories “empiricism” and “rationalism”). [4] I shall briefly sketch these two conceptions, because they epitomize the idea of experience as shallow and the idea of experience as deep.

Hume and the shallow conception

For Hume, the very paradigm of an “impression” (and the other sort of experience, “ideas”, was identified by him with “faint copies” of impressions) is a visual image. Descartes and Berkeley had both tried to read the nature of visual impressions directly from the newly investigated nature of retinal images . [5] The result of this approach was a tendency to think of all “impressions” on the model of pictures (not necessarily visual, of course – there were also tactile, olfactory, etc., representations; but like pictures, these, and the “ideas” or faint copies that corresponded to them, were thought by Hume to refer only to what they resembled ). Content, on this resemblance-semantics, is a rather primitive affair. [6] The very idea of a fact that cannot be sensorily pictured was rejected by Hume (as it was later by the early Logical Positivists. [7] The only other sort of “content” arises from “association” – especially the association of “passions” (feelings and emotions) with images.

Today there are very few if any old fashioned empiricists in philosophy. But what survives of the older view is the very influential idea that experience (still identified by empiricists with sensory inputs) is “non-conceptual”. Quine’s idea that, for philosophical purposes, experience-talk could simply be replaced with talk of “surface irritations” (stimulation of the nerves on or near the surface of the body) in many ways foreshadowed this influential idea.

Kant and the deep conception

In Kant’s writings one can find a response to the empiricist view of experience as consisting of sensory images, a response so deep that even today few philosophers who are not primarily Kant specialists have fully appreciated it (Strawson, Sellars, and more recently John McDowell and James Conant being among the happy exceptions). In the few minutes I can afford to devote to it this evening, I cannot, of course, do justice to it, but I hope to point out at least some of the leading ideas of the Kantian conception. It is important, however, to realize that no one book of Kant contains all of it. From The Critique of Pure Reason to Religion Within the Bounds of Mere Reason [8] , Kant constantly broadens and deepens the presentation of his view, if not the view itself. The account in The Critique of Pure Reason is, nonetheless, the basis on which the deeper and broader reflections in Kant’s subsequent writings depend.

Hume, as we just saw, conceives of experiences on the model of pictures, and their cognitive content as contained and communicated via (sensory) resemblance . Only sensory qualities are, thus, properly cognizable at all. If one accepts this, then many of Hume’s other famous doctrines readily follow: for example, Hume’s claim that we don’t "observe" causal connection depends both on Hume’s limitation of what we can observe to sensory qualities and on his very narrow inventory of sensory qualities: since causal connection is not a sensory quality for Hume, it is evident to him that casual connection is never observed . On the other hand, although objective time hardly consists of sensory qualities either, Hume never worries about the question, “How and why are we able to think of impressions and ‘ideas’ as succeeding one another in an objective time?”

Kant did, however, worry about this question, and he concluded that our notions of objective time, causality, and lawful connection are interdependent. For example, our awareness of a boat’s sailing down a river (coming, let us say, to a certain bridge) as earlier than the boat’s sailing beyond the bridge even though we think of a building’s back as existing at the same time as the front even if we look at the front before we look at the back, are internally related to our beliefs that we could have chosen to experience the front before the back, but we could not, conditions being as they were, have chosen to experience the boat’s sailing beyond the bridge before we experienced it approaching the bridge [9] , and these beliefs are in turn related to the system of causal connections we accept. The notion of time is inextricably connected with the notions of space, time, and causality. This is not just a fact about Newton’s physics, or Einstein’s, but about our ordinary conceptual scheme as well. Imagine, just as a thought-experiment, that there is a (more or less instantaneous) world-state, call it “A”, consisting of a sense impression as of a cat chasing a mouse, a world-state, call it “B”, consisting of a sense impression as of a twelve-foot cat singing “Yankee Doodle”, and a world-state, call it “C”, consisting of a sense impression as of a purple tidal wave sweeping over a field of flowers with heads like Charlie Chaplin. What sense would it have to say that these are states of one and the same world , let alone to speak of them as temporally ordered, if there are no causal connections of any kind between them? [10] Hume’s argument depends upon our thinking of the concepts “experiences A and B [think of experiences at different times here] lie in one and the same phenomenal world” and “A is earlier than B” as presuppositionless .

Our question, however, concerned how we experience things, and not how we conceive them. But – long before modern psychology – Kant questioned the coherence of such a dichotomy. We do not experience familiar objects and events – a cat’s drinking milk, a tree swaying in the wind, someone’s hammering a nail into a wall – as collections of “color-points” on a spatial grid. As William James put it, [in the case of a “presented and recognized material object”] “sensations and apperceptive idea fuse … so intimately that you can no more tell where one begins and the other ends, than you can tell, in those cunning circular panoramas that have lately been exhibited, where the real foreground and the painted canvas join together.” [11] In Kant’s language, in the sort of perception James described (or – an example Kant himself uses – in the case of experiencing something as a boat’s sailing down a river), we have not mere unconceptualized sensations, whatever those might be, but a synthesis of experiences and conceptual ideas, the ideas of space, time and causation. This is something that the phenomenogical school, beginning with Husserl, likewise emphasized: I see a building as something which has a back, Husserl pointed out, even when I don’t see the back. Such perception is fallible, to be sure; but so is the perception that something is red or circular. And the retreat to “sense data” in the hope that there we can find something “incorrigible” has long been recognized to be a “loser”.

A second issue which plays a large role in The Critique of Pure Reason , and one which figures in contemporary attacks on what Postmodernists consider to be the metaphysical illusion of the “ego” is the issue of right and wrong ways to think about what it means to be or have a self . (As Nicholas Boyle has observed, Postmodernist doubts about whether there is such a thing as a self, or an “author”, never stop the Postmodernist from cashing a royalty check.) Here again, paying more attention to Kant would help to clear our heads.

For Kant, rational thought itself depends on the fact that I regard my thoughts, experiences, memories, etc. as mine . To illustrate Kant’s point, imagine yourself going through a very simple form of reasoning, say “Boiling water hurts if you stick your finger in it; this is boiling water; so it will hurt if I stick my finger in this.” If the “time-slice” of me that thought “Boiling water hurts if you stick your finger in it” was one person, person A, and the “time-slice” that thought the minor premise, “This is boiling water” was a different person, person B, and the person that thought the “conclusion”, “It will hurt if I stick my finger in this” was yet a third person, person C, then that conclusion was not warranted, indeed, the sequence of thoughts was not an argument at all, since the thoughts were thoughts of different thinkers, none of whom had any reason to be bound by what the others thought or had thought. We are responsible for what we have thought and done in the past, responsible now , intellectually and practically, and that is what makes us thinkers , rational agents in a world, at all. Kant, like Locke before him, can be seen as making the point that the thinking of my thoughts and actions at different times as mine does not depend on a metaphysical premise about “self-identical substances”, and is none the less a form of conceptualization that we cannot opt out of when we are engaged in judgement in action.

As before, to say that Kant’s point is valid for conceptualization but not for experience would be to miss the way in which experiences and concepts interpenetrate, the way in which they are “synthesized”. When I reason (say, about the boiling water), I experience my successive thoughts as “mine”. Hume is right in holding that this is not a sensory quality; there is no “impression” of “my-ownness”; and Kant would emphasize this just as much as Hume. But whereas Hume concludes that the self is an illusion , Kant sees that experience transcends Humean “impressions”. Whereas for Hume, experiences are sheer psychological surface, for Kant even the simplest perception links us to and interanimates such deep ideas as the ideas of time, space, causality and the self. And this is something that Kant does not just claim, but that he argues in detail, and with incomparable brilliance. That experience is intrinsically deep is the heart of the Kantian conception. It is not something that was overthrown by the collapse of Kant’s “synthetic apriori” and the metaphysics Kant tried to base upon it.

Kant on aesthetic experience

I said above that Kant deepens the presentation of his views (and perhaps the views themselves) in successive books, and, I should add, not only in books. For example, a wonderful (and sadly neglected) discussion of what is right and wrong in mysticism may be found scattered in Kant’s writing. [12] But no where is this more true than in The Critique of Judgment .

I cannot, of course, even sketch the complex and rewarding aesthetic theory of that Critique. Fortunately, that is not my goal. What I want to do is extract one item from that complex discussion, although to do that I will have to say a little about the ideas that surround it. [13] The “item” in question is the fascinating notion of an indeterminate concept. When we experience a work of art, Kant tells us, we experience it as escaping capture by “determinate concepts”, but we do perceive it as – not being captured by, but evoking – a kind of concept, an indeterminate concept, one which is deeply connected with what Kant calls “the free play of the faculties” [imagination and reason – under the guidance of the former].

Here I do have interpret the aesthetic theory I said I wouldn’t discuss to the extent of warning my readers against two common misinterpretations. The first, which I am indebted to Paul Guyer for pointing out, is the assumption that when Kant speaks of “pure” aesthetic experience he is using “pure” as a value term. The reverse is the case; the art that Kant values and thinks we should all value, Guyer has conclusively shown, is mixed, impure. “Pure aesthetic experience” in Kant’s sense is concerned only with form; but to value, say, a painting which moves us both on account of its subject matter and its formal properties, or a novel or a poem, is to respond not only to the “purely aesthetic” features in this technical sense, but to the interplay of description, valuation, and purely formal experience. [14] The second misunderstanding is that it is only “the concept of beauty” that Kant has in mind by the term “indeterminate concept.

To illustrate what I believe Kant actually had in mind, think of a painting by Vermeer (pick your favorite!). It is not “indescribable”; a great deal about it can be described. The notorious Vermeer-forger, Van Meegeren, could undoubtedly have given a precise (determinate) description of a great many features of this or of Vermeer paintings in general. But the description, although it might teach us a lot, and even add to our appreciation of such a painting, would not answer the question: “Why is this painting so beautiful?”. Indeed, as Van Meegeren’s rather unpleasant forgeries testify, a painting could satisfy this “determinate” description and not be beautiful . What Kant, interestingly, says about the discussion of works of art is not that it is impossible to describe what is it that strikes us as beautiful (which it would be, if the only alternatives were to apply to them determinate concepts of the kind a Van Meegeren – or an art historian – might offer, or to apply the single indeterminate concept “beautiful”). What he says is that the aesthetic ideas which are the content of works of artistic genius evoke so much thought that language cannot fully attain them, or make them intelligible. [15] {He also says that that we add to a [determinate] concept “a representation of the imagination that belongs to its presentation, but which….aesthetically enlarges the concept itself in an unbounded way. [16] ) In short, certain concepts seek – and manage – less to finish a discussion, or answer a determinate question, as to further provoke both thought and imagination, and to raise an “unbounded” number of further questions. And these are the concepts we need and have to use to talk meaningfully about art.

What connects the notion of an indeterminate concept with my topic of experience is that it is precisely in the context of discussing how we perceive works of art that Kant invokes it. Indeterminate concepts are not purely intellectual concepts; they require both a sensible subject matter and an active imagination to apply. That perception is fused with conceptual content is something we learned from The Critique of Pure Reason ; that some of the perceptions we value most are fused with indeterminate , with open-ended, conceptual content, content in which imagination and reason cooperate under the leadership of imagination, is something we learn from The Critique of Judgment .

The notion of an “indeterminate concept”, understood in this way, naturally extends to moral notions. If Kant does not use it in the area of morals, it is because, I think, of a desire to keep morality rigorous and transparent. But morality, good morality, cannot always be rigorous and transparent, and a thinker who has seen that something like the notion of an “indeterminate concept” I just described applies also to the highest type of moral awareness is Iris Murdoch, even if she does not cite Kant or use his terminology. (Thus, in her philosophical masterpiece, The Sovereignity of Good , she writes, “Moral tasks are characteristically endless not only because ‘within’, as it were, a given concept our efforts are imperfect, but also because as we move and as we look our concepts themselves are changing…We do not simply, through being rational and knowing ordinary language words, ‘know’ the meaning of all necessary moral words. We may have to learn the meaning; and since we are human historical individuals the movement of understanding is onward into increasing privacy, in the direction of the ideal limit, and not backwards towards a genesis in the ruling of an impersonal public language.” [17] )

Beyond both aesthetics, in the sense of the open-ended appreciation and discussion of works of art, and morality, in the sense of Murdoch’s “loving attention” to the whole complexity of human beings and human moral life, it should be obvious, I think, that religious experiences are both guided by and spontaneously give rise to indeterminate concepts in a way analogous to the ways in which aesthetic and moral experiences do. And if we see religious, aesthetic and moral experiences in this way, as I have been urging we should, we will avoid Hume’s mistake of trying to analyze them as a chemist analyzes a compound, analyze them into so much of this factor (“ideas and impressions”), so much of that factor (“passions”), and so much of this other factor (“beliefs”). In the deepest human experiences, ways of perceiving things that are inseparable from those experiences but nonetheless conceptual, at least in the way indeterminate concepts are conceptual, fuse so intimately that you cannot tell where one begins and the other ends, to mimic William James words quoted earlier.

Although the phenomenological school of philosophy which began with Husserl inherited and extended the Kantian insights I have been describing, the most influential twentieth century phenomenologist, Heidegger, had a contemptuous attitude to science, which, for him, was merely an aspect of technological civilization (which he regarded as intrinsically evil). In Heidegger’s writing, everything I have been saying about the depth of religious experience (including the experiences of “Being” and of being “thrown” into the world and of finding a destiny which is one’s “ownmost”, which are Heidegger’s versions of or substitutes for religious experience), as well as of artistic experience (especially experience of poetry), and even of our everyday experiences with artifacts is recognized and phenomenologically interpreted; but science is denigrated.

But by default, if we do not examine the impact of science on our ways of experiencing the world in a more sympathetic spirit than Heidegger was capable of, we are likely to fall back into the empiricist picture of science as consisting of deductive and inductive inferences from simple “sense data” (or Machian “Empfindungen”). To find a sustained critique of this way of thinking, we have to turn to the American pragmatists, and especially John Dewey. Extending the line of thought that William James had begun with his talk of apperceptive ideas and sensations as “fusing”, Dewey saw that science endlessly and inventively creates new observation-concepts, and that by so doing it institutes new kinds of data. [18] A scientist with a cloud chamber may now observe a proton colliding with a nucleus (without being able to answer the question “Exactly what visual sensations did you have when you observed it?” except by saying “It looked like a proton colliding with a nucleus”), or observe a a virus with the aid of an electron microscope, or observe a DNA sequence, and so on. And the impact of science on the conceptualization of experience is not confined to specialists; the way in which all of us experience the world was changed by Darwin, was changed by Freud (whether on thinks this or that claim of Freud’s was well- or ill-founded) as the notion of the unconscious became part of our vocabulary, and is being changed today by computer science and the concepts and metaphors it adds to the language.

On the “meta-level”, the level of the methodological appraisal of scientific theories, we also find something in science analogous to the indeterminate concepts involved in aesthetic judgment, indeterminate concepts that figure in judgments that are internal to scientific inquiry itself: judgements of coherence, simplicity, plausibility, and the like. The similarity of such judgments to aesthetic judgments has, indeed, often been pointed out. Dirac was famous for saying that certain theories should be taken seriously because they were “beautiful”, and Einstein talked of the “inner perfection” of a theory as an “indispensable criterion”. [19]

But it is time to say something of the wider relevance of this picture of experience, the picture of experience as deep, for the concerns of this series of lectures.

Conceptuality and skepticism

At first blush, recognizing that perception (and experience that purports to be perception or resembles perception) is always conceptualized may seem to make the problem of skepticism much worse, especially when religion is the issue. From Kant to John McDowell, philosophers who point out that experience is conceptualized have been told that they are problematizing our access to reality. Concepts can, after all, mislead as well as lead, conceal as well as reveal.

The fact that religious concepts are no longer intersubjectively shared within Western culture, and have not been for a long time, makes this more than a purely theoretical issue, as skepticism about the existence of houses and rocks happily has become. (For the ancient Greek skeptics, it is often pointed out, it was anything put a “purely theoretical issue”, but that is another story, and not one I need to tell today.) While no one can say that there are only so and so many possible answers a religious person can give to the atheist or to the religious skeptic, three main approaches are familiar to all of us.

The traditional approach, and the one that is still that of the Roman Catholic Church, is to continue (albeit with contemporary sophistication) the medieval attempt to “prove the existence of God” (“neo-Thomism”). This is not an approach I find possible for myself, at any rate, for the following reasons:

First, in order to understand talk about God, whether or not that talk takes the form of a "proof", one must be able to understand the concept God. But there are very different possible conceptions of what it is to understand the concept God, in a way that has no analogue in the case of, say, mathematical proof. Secondly, even if one understands the concept God, to accept any of the traditional proofs one has to find a connection between that concept and the highly theoretical philosophical principles involved in those proofs, premises about conditioned and unconditioned existence, and about what sorts of necessity there are. Some of the most profound religious thinkers of the last two hundred years (particularly the religious existentialists from Kierkegaard to Rosenzweig and Buber) have had no use at all for this sort of philosophizing; and I would be the last to say that they lacked the concept God. What the traditional proofs of the existence of God in fact do is connect the concerns of two different salvific enterprises: the enterprise of ancient and medieval philosophy [20] , which, after all, is the source of the materials for these proofs, and the enterprise of monotheistic religion. While it is certainly possible to have a deeply worthwhile religious attitude which combines these two elements -- indeed, the effort to do so has contributed profoundly to Judaism as well as to Christianity and Islam -- it is also possible to have a deeply fulfilling religious attitude while keeping far away from metaphysics. Speaking for myself, I would say that while I do conceive of God as a "transcendent Being", as a "necessary Being", as an "unconditioned ground for the existence of everything that is contingent", I feel that insofar as I have any handle on these notions, I have a handle on them as religious notions, not as notions which are supported by an independent philosophical theory . (Certainly not by the theory of Aristotle's Metaphysics .) For me the "proofs" show conceptual connections of great depth and significance, but they are not a foundation for my religious belief. (In spite of Maimonides’ prestige, they never played an important role in Judaism.) Nor are “proofs” the way in which I would try to bring someone else to religious belief of any kind. [21]

A second familiar response to religious skepticism is the dogmatists’ “my religion is true and every other belief is wicked (especially atheism), or no better than witch doctoring (other religions).” (A friend remarked, “I understand this is very popular among people philosophers don’t talk to." [22] ) Not only is this response a denial of the very raison d’etre of philosophy itself, which John Dewey so well defined as “criticism of criticisms” [23] , but, in a marvelous discussion of the psychology of “fanaticism” in The Critique of Judgment , Kant argues that this is, at bottom, not religion but a disease of religion. [24]

Part of Kant’s point is that the “fanatic” (his term for what I just called “the dogmatist”) treats religious beliefs as if they were as sure as ordinary perceptual beliefs. I remarked a few moments ago that skepticism about the existence of houses and rocks has happily become a purely theoretical issue. In practice, as Kant pointed out in The Critique of Pure Reason , perception of such objects is passive; we have no real choice about whether to believe that there is a house in front of us when we see one. Nor do we have to “take responsibility” for believing that there is a house there when we see one or walk into one. For the fanatic, it is as if he had as simply (and as unproblematically) seen God, or seen Jesus (or, in Kantian language, seen the unconditioned). Those who do not accept what is so obvious are wicked or stupid or both, or, in the best case, waiting for the fanatic to enlighten them. Such an attitude, Kant believes, misses the essence of true religious faith, which (for him) involves the recognition that what one believes is not simply forced on one passively. The uncertainty, the unprovability, of religious propositions is, Kant believed, a good thing; for if religious propositions could be proved, there would be nothing to take responsibility for. To put it in present-day language, the fanatic is unconsciously fleeing responsibility . I find that my perceptions are in accord with Kant’s here: I find that both his psychology of fanaticism and the phenomenology of faith presupposed by that psychology are very deep.

A third approach to skepticism, often associated with existentialism, is to accept responsibility for believing what cannot be proved. I already mentioned the note Avi Sagi found in an unpublished bit of Kierkegaard’s Nachlass which reads: “Leap of faith – yes, but only after reflection.” In this approach the role of religious experience is not to prove something but to confront one with an existential choice, to make “believe or don’t believe” a “live option”, in William James’ words. A fine but deeply challenging account of this third option can be found in Wittgenstein’s “Lectures on Religious Belief”. [25] (Wittgenstein described himself as not a believer, “although I cannot help seeing every question from a religious point of view.”)

Here is what Wittgenstein writes: [26]

Several paragraphs later, Wittgenstein discusses a “Father O’Hara” [27] who, he tells us “is one of those people who make it a question of science.” And he continues: [28]

The question these remarks of Wittgenstein’s invite is the obvious one: is it ever justified to believe what is not “reasonable”. This is the question that William James dealt with in his celebrated essay, “The Will to Believe” [29] (which he considered calling “The Right to Believe”, which is what the essay actually defends). That often misrepresented and misinterpreted essay, it seems to me, gives exactly the right answer to this question, but it would take a lecture at least as long as this one to interpret and discuss it. I want, however, to make just one point about it, namely that James emphasizes that saying there is a right to believe is by no means to say that there is a right to be intolerant, [30] and that too seems to me exactly right.

Why did I focus on experience, then?

In view of what I just said, it will be clear that I did not focus on experience in this lecture because I wish to argue that religious experience answers skeptical questions. But I did have a reason for focussing on it, just as Wittgenstein had a reason for focussing on the complexity of the phenomenon of religious belief. Wittgenstein began the lectures on religious belief by pointing out that believers and atheists regularly talk past each other. If you search the web under “atheism”, you will find a great deal of intelligent and painstaking proof that the Bible contains errors, that it is silly to think that every word of the Bible was literally dictated by God etc., but precious little recognition that most religious people are not fundamentalists, and many do not believe in the idea of divine dictation at all. It is as if atheists too were “fanatics” in Kant’s sense; for them too, their [negative] religious belief is, it seems, akin to a perceptual certainty, something that involves no responsibility . Wittgenstein, if I have interpreted him correctly [31] , did not want to make us believers (he was not religious himself), but he felt an enormous respect for the literature and the spirituality contained in religious traditions, and he wished to combat this sort of simplistic stereotyping. One way of overcoming the idea – and we need to overcome it! - that it is simply obvious what having a religious faith consists in, is to overcome the idea that it is simply obvious (or if not obvious, obviously irrelevant) what the words “religious experience” refer to. In this lecture, I have tried to suggest that what “experience” refers to is far more complicated a matter than we tend to think, and that understanding how deep experience can be is a necessary preliminary to any discussion of “science and religion” – and such a preliminary, or perhaps propaedeutic is what I have attempted to offer this evening.