1. Today we conclude the examination of fundamental modes of human experience which we undertook with a view to elucidating the general nature of human selfhood. We are to study in this lecture experience in its moral mode.
As is well known, there are two sharply contrasting schools of thought, the Naturalistic and the Non-naturalistic, concerning the proper analysis of ‘moral experience’, or ‘the moral consciousness’. According to the Naturalistic view (characterising it very broadly, but sufficiently for present purposes), there is nothing unique or ultimate about moral experience. Consciousness of the moral ought resolves itself on analysis into a complex whose components are all of the nature of the ideas, desires, hopes, fears, etc., found in the context of non-moral experience. According to the Non-naturalistic view, the moral ought is either ultimate and unanalysable, or, if analysable, can be analysed only in terms which include some other ethical concept, such as ‘good’, in a sense in which that is ultimate and unanalysable.
I think it is not easy to exaggerate the extent to which one's conception of the character and status of human selfhood is affected by the decision one comes to upon the relative merits of these two rival types of analysis. If the Non-naturalistic analysis be sound, there is, I believe (and shall later argue), a legitimate inference from the fact of moral experience to the objective reality of a moral order, and to a relationship of the human self to that order which confers upon the life of man a dignity and a cosmic significance of which the alternative, Naturalistic analysis affords no hint.
Now where the outcome of an investigation is so patently fraught with momentous consequences for the value and destiny of the human individual, there is a danger that either wishful thinking, or the anxiety not to be a wishful thinker, may exert a distorting influence. Of the former danger philosophers of our present disillusioned age may be said to be constantly and acutely aware. But in a good deal of the modern literature of ethics there are at least undertones which suggest that the latter danger is not to be discounted either. Indeed I should myself doubt whether any greater bias has been imparted to the writings of moral philosophers by the desire of some to reach agreeable results, than by the ruthless determination of others to be what they call ‘toughminded’ at all costs. Assuredly there is no duty laid upon the philosopher to ‘think nobly of the soul’. But he has no duty to think ignobly about it either. By all means let us be tough-minded. But it is well to remember that if one makes a fetish of toughmindedness the result is apt to be not too easily distinguishable from thick-headedness!
Of the two concepts commonly regarded as the fundamental concepts of morals, the ‘ought’ and the ‘good’, it is the former which will be our concern in the present investigation. For whatever else the moral consciousness may be, it is at least a consciousness of oughtness or obligation. No analysis of the moral consciousness can possibly be adequate that does not include an analysis of the consciousness of obligation. In making the ‘ought’ the central thing in our study, however, I am by no means wishing to foreclose any of the issues about the relationship of the ought to the good, or of the right to the good, which have been the subject of so much intensive and profitable research in recent decades. It may be that our consciousness of moral obligation is always of an obligation to ensue certain ends because, and only because, they are independently approved as good, or as means to good; and that the concept of good has thus a logical priority. It may be, on the other hand, that our consciousness of moral obligation is, at least sometimes, of an obligation to ensue certain ends just because they are obligatory, independently of any thought of a good which may attach to, or result from, their achievement. For our present purpose all such questions are irrelevant. On any theory, I feel free to assume, the moral consciousness is at least a consciousness of obligation. In order that a good may approve itself to our moral consciousness, and thus be esteemed morally good, it must be apprehensible (as, of course, many ‘goods’ are not) as a good which ought to be ensued by someone.
This lecture, then, will be concerned to examine the nature and implications of moral experience, understanding by that the experience of moral obligation. I propose to lead up to my own (as usual, old-fashioned) view by way of a critical survey of the main types of analyses of that experience which have been competing for the philosopher's favour in recent ethical literature.
2. These analyses are numerous, and it may seem as though we were embarking upon an impossibly large undertaking for a single lecture. Nevertheless, though a trifle more space would certainly not have come amiss, I am not so sure that any very remarkable feat of condensation is called for. Taking my courage in both hands, I venture the opinion that a good many of the ethical novelties that have been proposed for our instruction in recent years have been accorded much too solemn a reception. I shall submit to you at the outset three critical canons which, it seems to me, should guide us in appraising presented analyses of the moral ought. If they are accepted (and the first two at any rate seem to me to need no defence), we shall find, I believe, that their application to contemporary analyses causes a surprising number of these to collapse completely.
1. The meaning of the moral ought is intrinsically the same in all moral judgments, whether they be in the ist Person, 2nd Person, or 3rd Person; whether they be in the Singular or the Plural Number; whatever their tense; and whatever their mood. Any analysis must be rejected which fails to apply satisfactorily to even one acknowledged case of genuine moral judgment.
2. The meaning of the moral ought is not the same thing as the criterion of the moral ought. This platitude is perhaps hardly worthy to rank as a canon of criticism. But analyses are not seldom suggested which are plausible enough as accounts of the latter but are so implausible as accounts of the former that one is bound to suspect some confusion of meaning with criterion.
3. The meaning of the moral ought is its meaning in moral experience, which need not, and frequently does not, find overt expression in speech or writing. Let us (conveniently, if a little arbitrarily) use the term ‘moral sentences’ to denote the overt expression of moral experiences. We may then say that any analysis fails if, though apparently appropriate to all moral sentences, it will not apply to the ‘silent’, uncommunicated, moral judgment.
This does not imply, I hasten to add, that analysis should rely wholly upon introspection, and take no account at all of the ‘public’ evidence afforded by moral sentences. Introspection must be safeguarded against personal idiosyncrasies by constant reference to the moral experience of others as this is disclosed (however imperfectly) in the overt expression of moral sentences both written and uttered. What is implied is that the analysis of moral sentences can be of only ancillary value, and that in the last resort the test of validity for any analysis of the moral ought must be ‘Is this really what I mean by “ought” when I judge that I (or he, or they, or you, etc.) morally “ought” to do something?’ The study of overt moral expressions is not merely unhelpful, but positively harmful, unless it is steadily borne in mind that what we are ultimately concerned with is the moral experience that underlies the expression.
As the implications of our third canon are in manifest conflict with the linguistic method of ethical research in such high favour today, some elaboration and defence of it will rightly be expected. I welcome the opportunity, for it seems to me that ethical studies at the present time are suffering severely from failure to appreciate the palpable defects of moral sentences as an index to the nature of actual moral experience. Let me instance a few of these defects.
In the first place, moral words—‘right,’ ‘wrong’, ‘ought’ and the rest—are very far from being univocal. In many sentences they are used with no moral import at all, and we should go wildly astray if we sought to elicit from the sentences as such truths about moral experience. If we did not go behind the sentences to the thought that informs them, we should not know whether what we were doing had any relevance to ethics at all.
In the second place, even when the words are being used with a moral intent, the sentences that embody them are often the vehicle of what can only by courtesy be called ‘moral’ judgments. Many, perhaps most, of the ‘moral’ utterances of everyday life are no more than the casual, uncritical application of the conventional moral rubrics of one's society. But it is surely clear that the analysis of moral sentences can help us to understand moral experience only if what these express really is a moral experience. Analysis of the moral sentences that embody the rough and ready, ill-considered, almost ‘slap-happy’ judgments that so often pass for ‘moral’ in everyday intercourse may have a certain interest of its own; but it is bound to be an exceedingly unreliable guide to the deliverances of the moral consciousness.
In the third place—and this I take to be the most serious, because the most frequently overlooked, defect in the approach to ethical research by way of moral sentences—as soon as moral judgment, even well-considered moral judgment, is given overt expression in a sentence, new and extraneous elements enter into the situation. Overt expression—save in abnormal cases like ‘talking aloud to one's self-differs from the silent judgment in having an ulterior purpose, viz. communication to others. And we make communication to others only because we wish to affect in some way their future states or actions. This is true even of plain indicative sentences, which are normally uttered only because we want the persons addressed to be in possession of certain information. It is conspicuously true of moral sentences, and above all of 2nd person imperatives relating directly to future action. The analysis of moral sentences, therefore, is virtually bound to issue in the discovery that an important function of many of them is to influence future action in one way or another. But what has this got to do with the question of the import of moral judgment? The ‘social purpose’ of moral sentences is clearly extraneous to the moral judgment. For nothing seems plainer than that there are ‘silent’ moral judgments, judgments which we make mentally but to which we give no overt expression, and that these have no purpose beyond themselves—social or otherwise.
3. The point we have just been making seems very obvious. Yet it is precisely the failure to distinguish clearly between the import of the moral judgment on the one hand, and the function of the moral sentence through which the moral judgment may or may not find overt expression on the other hand, that has led to the modish view that moral judgments are wholly or partly of the nature of recommendation, or advice, or persuasion, or advocacy, or command. If our ‘third canon of criticism’ be sound, and the moral ought of whose import we are in search is the ought of moral experience, then we can reject almost out of hand and en masse those theories which hold that the import of moral judgment lies wholly or partly in the desire of the judging subject to affect the mind, and thereby the future behaviour, of some person or persons. Let us call all such theories, for short, ‘Hortatory’ theories—a more convenient term, I think, than ‘Command’ theories, since very different types of theory from that which now engages us (including our own) recognise something at least like a command in the moral situation. These Hortatory theories have a slightly greater plausibility in the case of 2nd person imperatives than elsewhere, so let us look for a moment at the moral judgment ‘You ought to repay that money’. If I say (aloud) ‘You ought to repay that money’, it is reasonably certain that there is a hortatory element present. Normally, at any rate, I speak in the hope that my words will increase the likelihood of your discharging your debt. But this hortatory element cannot possibly be part of the meaning that the moral ‘ought’ has for me, for the simple reason that I can refrain from uttering the sentence ‘you ought to repay that money’ while still thinking ‘you ought to repay that money’, and it would be nonsensical to look for a hortatory element in my ‘silent’ judgment. I am certainly not trying to persuade you, yet it is surely beyond question that I am making a moral judgment, I am apprehending moral obligation or oughtness, when I merely ‘think’ that you ought to repay the money.
There is, indeed, formally, a last refuge open to the Hortatory theory in face of this objection. The only person I can possibly be exhorting in the silent moral judgment is myself—since I alone am aware of what is being exhorted—and it might be suggested that the import of the moral judgment lies in such exhortation addressed to one's self. When I merely think, and do not communicate the thought, that you ought to have repaid that money, what I am really doing (according to this view) is telling myself to be sure that whenever I am in a like position I pay back what I have borrowed. Whether any theorist has in fact ever sponsored so fantastic a view, I do not know. I can only insist that there is a limit, even in moral philosophy, to the hypotheses that deserve discussion, and give my opinion that this one passes well beyond that limit.
If it were not so clear that the application of our third critical canon is in itself fatal to the Hortatory theory in all its forms, it would be worth while to examine the theory in the light of the first canon also, and to show how extremely implausible it is even for many moral judgments that do find overt expression—for all, indeed, save those that happen to be in the second person, and for some even of these. But to myself this seems a work of supererogation. I shall confine myself, therefore, to offering a single illustration of the sort of difficulty which the Hortatory theory has to try to meet. Suppose A is on what he knows to be his death-bed, and is sorrowfully reflecting upon some unworthy past action. Obviously in mental distress, he is asked by his Father Confessor if there is anything in particular that is troubling him, and he replies ‘Yes, I ought not to have behaved in that shabby way to B’. To whom is the supposed ‘exhortation’ addressed? Is A exhorting himself not to act in that sort of way in future? No, for he is well aware that for him there is no future. Is he then exhorting his audience? No, for his sole audience is his Father Confessor; and it would stretch credulity too far to be asked to believe that the death-bed penitent is in effect saying to his Father Confessor ‘Don't you ever behave in that sort of way!’ Whom, then, is he exhorting? I leave it to the friends of the Hortatory theory to find an answer. I can supply none that has the faintest plausibility.
4. I turn now to another type of theory that is perhaps still more popular today. The advocates of this type of theory (at any rate in its pure or unmixed form) are not misled by the social purpose that is inherent in the overt expression of moral experience, and in fact proclaim that the essential import of moral ‘judgment’ lies not in communicating anything to others, but in expressing a particular emotion of one's own. When I say ‘You ought (or ought not) to do X’, the significance of the ‘ought’, it is contended, is to express a particular pro-emotion (and of the ‘ought not’ to express a particular anti-emotion) which I feel towards your doing X—or perhaps towards the kind of action designated by X.
This type of theory (and so far it resembles the pure Hortatory type) denies that the so-called moral ‘judgment’ is really a judgment. Its ethics is, in the expression made familiar by the title of an extremely stimulating paper1 by Professor Barnes, an ‘ethics without propositions’. There may, indeed, be now-ethical propositions involved. Thus if I say ‘You ought not to be doing X’, I may reasonably be taken to be implicitly affirming that you are doing X—a non-ethical proposition. But the whole significance of the ‘moral’ word—‘ought’—lies in its expressing a particular emotion which I feel towards the conduct in question.
The Emotive theory—or the Emotive-Expressive theory, as we may better call it to distinguish it from the Emotive-Assertive theory which we shall later notice—has assumed a wide variety of forms in contemporary ethics; but it will not be necessary for our purpose to give to each of these separate consideration. For it seems to me that there is one conclusive criticism to which the theory lies open in every form it can assume; viz. that when one tries to give an account of the distinctive character of the emotion that is expressed in moral sentences, one finds that this can only be done in terms which imply that the ‘emotion’ in question is also a judgment. Let me explain.
It is evident that there are any number of experiences in which we feel pro-or anti-emotions towards ways of behaving and which we yet do not identify as ‘moral’ experiences. An Emotive theory is manifestly defective if it tells us nothing about the distinctive character of the emotion typical of moral experience. What then are we to say of it? It does not seem difficult to show that any combination of ordinary ‘non-moral’ emotions that may be offered fails in fact to yield the particular emotion we feel in any serious moral ‘judgment’. Professor Barnes is right, as far as he goes, in insisting that the moral emotion is unique—a ‘peculiarly moral feeling’. But it seems to me clear that he does not go far enough. That there always is an emotion present in moral experience, and that this emotion is unique, I believe to be true. But we cannot stop there. Persons who experience the moral emotion do think that they can say something in description of its nature. Moreover (and this is important) they are in substantial agreement as to what it is that can be said. If the ordinary man is asked how he would distinguish his pro-emotion towards keeping promises from his pro-emotion towards playing golf, he would not surely, unless abnormally inarticulate, be wholly gravelled for a reply. He would not be reduced to saying that ‘they just feel different’. He would probably tell us something like this: that the former emotion, in contrast with the latter, is felt not as a mere personal liking for the action in question, but as carrying with it a claim upon him, and a rightful claim, to perform the action. Indeed, the ordinary man's most natural way of putting it would be to say that what distinguished his moral pro-emotion towards an action is the feeling that he ought to act in this way.
Now that this is as a matter of fact what we feel when we experience the moral pro-emotion, it seems to me extremely difficult to dispute. It is simply not true that we find ourselves unable to describe the moral emotion in any way; and for myself I cannot conceive of any essentially different way of describing it appropriately. The ordinary man's description can, indeed, be challenged on the ground of defective phrasing. Thus it can be objected that ‘a feeling that…’ is not a mere feeling at all, but implies a judgment. The objection is technically correct. But it brings out the very point I am trying to make; viz. that moral experience is intrinsically incapable of being described adequately in terms of mere feeling. The specific feeling which characterises it is a feeling which is in essential integration with a judgment; a feeling which can be itself—can be a ‘moral’ feeling—only through its self-completion in a judgment. On this ground alone, the strictly ‘emotive’ theory of the ought breaks down. The ‘peculiarly moral feeling’ is there; but only in indiscerptible union with a ‘peculiarly moral judgment’ in a complex experience which requires both the feeling and the judgment for its proper characterisation.
On a previous occasion I had an opportunity of discussing this important point in greater detail. Perhaps I may be permitted, in amplification of the present brief treatment, to quote from the article in question a passage on the manner of integration of feeling and judgment in moral experience:
‘My contention is that the peculiarly moral feeling towards keeping promises involves the judgment that promises ought to be kept. The moral feeling, I am suggesting, is not itself without the judgment—it requires the judgment for the completion of its own nature. Two opposite, but equally ruinous, errors of analysis must be carefully avoided. We must not say that our moral feeling towards promise-keeping generates the judgment that promises ought to be kept: for (we have urged) it is not the distinctively moral feeling at all unless the judgment of oughtness is already present. But neither must we say that the moral feeling is generated by the judgment of oughtness: for if this judgment is a moral judgment proper, and not the mere mechanical repetition of a conventional formula (if, in other words, the ‘ought’ of the judgment is before our minds in its intrinsically moral meaning), then the distinctively moral feeling towards promise-keeping is already present. Most of the moralists who have held that the moral feeling is prior, and the moral judgment sequent upon it, have held this, I think, largely because it appeared to them that the only alternative was to suppose that the moral judgment precedes the moral feeling; an alternative which they rightly reject. Similarly, most of the moralists who have held that the moral judgment is prior, and the moral feeling sequent, have held this largely because it appeared to them that the only alternative was to suppose that the moral feeling precedes the moral judgment; an alternative which they, too, are right to reject. The possibility which is missed by both parties is that feeling and judgment are twin aspects of a single experience, neither of which has precedence over the other. This I believe to represent the true state of the case.’2
5. For the most part, philosophers who favour an ‘ethics without propositions’ have interpreted the ‘ought’ of moral judgment as expressing primarily either an exhortation or an emotion; though the total experience expressed would commonly be regarded by them as an ‘attitude’ in which both elements are present, and in which there may be present also something to which the names ‘resolve’ or ‘intention’ may be assigned. Only very recently, so far as I am aware, has the view been seriously put forward by any moralist that it is the last named of these elements that is the primary one. According to Professor Braithwaite, however, when a man says ‘I ought to do X’, what is primary in his ‘pro-attitude’ towards X is ‘his intention to perform the action when the occasion for it arises.’3 It is not suggested that all expressions of intention are moral assertions. ‘For the notion of morality to be applicable it is necessary that the policy of action intended by the asserter should be a general policy (e.g. the policy of utilitarianism) or that it should be subsumable under a general policy which the asserter intends to follow and which he would give as the reason for his more specific intention.’4 But with that qualification Braithwaite is prepared to hold that ‘when a man asserts that he ought to do so-and-so, he is using the assertion to declare that he resolves, to the best of his ability, to do so-and-so.’5
We shall, I hope, have from Professor Braithwaite before long a much fuller and more systematic exposition of his theory than was possible in his Eddington Memorial Lecture or in his earlier Critical Notice of R. M. Hare's The Language of Morals6—our only sources of information so far. As it stands, however, this ‘conative’ version (as Braithwaite calls it) of ‘ethics without propositions’ seems to me, I must confess, exceedingly unplausible. I shall briefly develop two major difficulties which are likely to give pause to most of Braithwaite's readers.
The type of case that is most favourable to Braithwaite's theory is, of course, the 1st person moral judgment relating to the future, ‘I ought to do X’. But even here the difficulties seem to me fatal.
1. May not a man ‘resolve, to the best of his ability’ to pursue a policy of action X even though he believes that he morally ought not to follow it? Granted that a man can, in a particular situation, choose to do what he believes to be morally wrong, it is surely possible in principle for him to resolve to pursue a policy, even a policy of the highest generality, where he believes that to be morally wrong? Indeed it seems likely that such resolves are not only possible in principle but have sometimes, if rarely, been made in fact. Instances at least seem to have occurred in which, for example, a man with a passion for power has resolved to devote all his talents and energies, and to subordinate every other claim, including the claim of morality, to the achievement of this objective. It is enough, however, that such a resolve should be possible in principle. If it is, then ‘I ought to do so-and-so’ cannot mean the same as ‘I resolve, to the best of my ability to do so-and-so’.
If at first sight there is a faintly unrealistic look about this objection, that is only because ordinary linguistic usage has accustomed us to give to the word ‘resolve’ a certain moral, or quasi-moral significance. We tend to think of ‘resolve’ as the kind of thing which, if not always directed towards what the agent believes he ought to do, is at least never directed towards what he believes he ought not to do. We talk freely of ‘good resolutions’; seldom if ever of ‘bad resolutions’. And ‘bad resolutions’ are implied in the objection I have been raising. But of course Braithwaite is not entitled, and would not claim to be, to a usage of the word ‘resolve’ which contained a covert reference to ‘oughtness’, since his whole object is to explain ‘oughtness’ in terms of ‘resolve’. In some contexts, as we have seen, he substitutes for ‘resolve’ the normally neutral word ‘intention’. When his view is formulated in terms of ‘intention’, our criticism of it is, I think, relieved of any appearance of ‘unrealism’. For it does not conflict with ordinary usage to say that one intends to follow a policy of action even though one believes it to be morally wrong.
2. The second fundamental objection to this ‘intentional’ theory of oughtness may be summarily stated as follows. Whereas moral oughtness is in certain cases accepted as self-justifying, in no case is an intention accepted as self-justifying.
A moral principle is acknowledged as a valid moral principle either because it is believed to be valid in its own right or because it is supposed capable of derivation from some more general principle (or principles) believed to be valid in its (or their) own right. In the case of the second of these alternatives, the question ‘Why ought I to observe the principle?’ is obviously meaningful. The answer lies in exhibiting its relationship to the more general principle or principles from which it is derived. In the case of the first alternative, a ‘Why?’ is just as obviously not meaningful. If the principle is acknowledged as valid in its own right, its oughtness is accepted as self-justifying, and any question of ‘Why?’ is recognised to be inept.
So much must, I think, be agreed. Ultimate moral principles, ultimate ‘oughts’, are their own justification. In these cases of oughtness the question ‘Why?’ has no meaning. But must we not also agree that in no case whatsoever of an intention is the question ‘Why?’ meaningless—not even where it is an ‘ultimate’ intention (or ‘resolve’, or ‘personal decision’) to adopt some total plan of life?
This seems to me so evident, and yet to be so manifestly fatal to the intentional theory of the ought, that I find the absence from Braithwaite's exposition (brief as it is) of any serious attempt to grapple with the difficulty a little surprising. It would appear, however, from a late reference in his lecture,7 that he would try to meet it along lines suggested by Mr. Hare. Mr. Hare argues that an ‘ultimate personal decision’ can be self-justifying, admitting of no question ‘why?’ that is not able to be answered within the ambit of the decision itself. According to Hare, it will be ‘a complete justification of a decision’ if we give ‘a complete specification of the way of life of which it is a part’—‘a complete account of its effects, together with a complete account of the principles which it observed, and the effects of observing these principles’. ‘If the inquirer still goes on asking “But why should I live like that?” then there is no further answer to give him, because we have already, ex hypothesis said everything that could be included in this further answer.’8
This is ingenious, but I cannot think that it will really do. Has everything already been said that is relevant to a further answer? To suppose that it has seems to me to involve a petitio principii.
Consider how the ordinary moral man would react to Hare's claim. ‘You have given me’ he might say ‘a complete specification of a way of life to which you are evidently prepared to commit yourself. But one thing that specification does not include which is extremely relevant, indeed fundamental, to the question why you or I or anybody else should adopt such a way of life, viz. any reference to its obligatoriness. To me at any rate this omission renders the ‘answer’ essentially incomplete.’
The retort, no doubt, would be ‘Of course I make no separate reference to obligatoriness or oughtness. For the essence of my view is that the “oughtness” of an ultimate principle is nothing other than its acceptance as a total way of life. To ask me to justify that way of life in terms of an “ought” that means anything else is just to betray that you do not understand my theory.’
But is there not here, as I have suggested, a petitio principii? The contention that no question can arise concerning the oughtness of the said way of life which is not already covered by the specification of it, is defended on the ground that its oughtness just means its ultimate adoption as a way of life. But the very point at issue is whether the meaning of the moral ought is thus reducible to a resolve, or personal decision. To the man who does not already so regard the ought, the argument can have no force whatever. For him the simple objection stands, that the specification of the way of life lacks one thing relevant and fundamental to the question why anyone should adopt that way of life, viz. any reason to suppose it morally obligatory. To allege, as Hare expresses it in one passage, that the decision to adopt such a fully specified way of life ‘would be the most well-founded of decisions, because it would be based upon a consideration of everything upon which it could possibly be founded’9 is merely to assume that a decision could not possibly be founded upon ‘moral obligation’ in its ordinary sense.
6. Let us pass on now to theories of moral experience which do admit that judgment is involved—‘assertive’ theories, as they are often called, by contrast with ‘expressive’ theories. Characteristic of the forms of assertive theory that have had a vogue in recent times is the contention that the ‘ought’ in moral judgment is not ultimate and unique. As to what precisely it is that we affirm when we judge that X ‘ought’ to be done, there is much diversity of opinion, but it is a matter of common agreement that at least this ‘ought’ is something capable of analysis without remainder in terms that do not include or imply itself. It is these forms of assertive theory that I wish now to consider.
I do not propose to consider them, however, at any very great length. For two reasons. Firstly, because it is the Hortatory and Expressive types of analysis upon which by far the greater part of contemporary interest has centred. Secondly, because the standard objections—to which I have little to add—against the Assertive type of theory are both well known and, I think, widely accepted as cogent. I shall do little more here than remind you, in regard to each of the Assertive theory's main varieties, of certain criticisms which to myself seem fatal. If they should appear to you equally conclusive, I venture to hope that in their light, and in the light of what has earlier been said in criticism of the Hortatory and Expressive types of theory, you may be sympathetically disposed to reconsider the claims of the once common view that the moral ought is ultimate and unanalysable.
The Assertive types of theory can be most conveniently classified, I think, under three heads; the Emotion-assertive, the Command-assertive, and the Production-assertive. We shall begin with the Emotion-assertive; and we shall look first at that form of it which holds that what is asserted when anyone makes the moral judgment ‘You ought to do X’ is the presence in the person judging of a pro-emotion towards your doing X.
The most obvious objection to which this theory lies open is that it becomes impossible in terms of it to allow that there can ever be real disagreement between two or more moral judgments. When I assert ‘You ought to do X’, and you reply ‘No, I ought to do Y’, what, on this theory, can you possibly mean by your prefatory ‘No’? For all I am doing, if the theory is correct, is to assert the presence of a certain pro-emotion in my mind, and all you are doing is to assert the presence of a differently directed pro-emotion in your mind; and there is nothing incompatible in these two assertions. Both may be, and probably are, correct. Yet we do all in fact believe (as the prefatory ‘No’ indicates) that the two moral judgments do contradict one another. Apparently, then, we cannot really be meaning in our moral judgment what this theory declares that we mean.
We might also, of course, apply to this theory what we said earlier about the peculiar character of the moral ‘emotion’. When we try to distinguish the kind of pro-or anti-emotion whose presence in ourselves we are supposed by this theory to be asserting in our moral judgment, we find that it is not merely an emotion. We find that it is the specifically moral emotion only if it is also a judgment. Hence this form of the assertive theory reduces to the absurd contention that moral judgment asserts the presence in the person judging of the moral judgment itself.
A second, perhaps commoner, form of the Emotion-Assertive theory is that what is asserted when I say ‘I ought to do X’ is that my doing of X evokes, or tends to evoke, a certain pro-emotion in the minds of some other person or persons or groups of persons; e.g. of most members of one's society, or of the majority of mankind.
This form of the theory is superior to the first form in not making disagreement between different moral judgments a logical impossibility. But it is difficult to see what other virtues it possesses. The simplest way of appreciating its inadequacy is to consider whether it is, or is not, a meaningful question to ask ‘Ought I to do that which evokes, or tends to evoke, a certain pro-emotion in P?’—where ‘I’ denotes the particular person or persons or group of persons named in the theory. Everyone would agree that the question is meaningful.10 But it could not be meaningful if ‘I ought to do X’ meant ‘My doing of X evokes, or tends to evoke, a certain pro-emotion in P’. For in that case what I should be asking would be ‘Ought I to do what I ought to do?’
The same general line of criticism—which has, of course, been fully exploited by Ross and (in relation to analysis of the ‘good’ rather than of the ‘ought’) by Moore—applies with equal force to Command-Assertive theories of the ought. According to this type of theory, the moral judgment ‘I ought to do X’ means ‘My doing X is commanded by (or is in accordance with the will of) P’, where P denotes whatever person or group of persons the theory accepts as having authority. This theory of the ought (so far at any rate as secular ‘authorities’ are concerned) is perhaps the implicit theory of certain ‘plain men’ rather than an explicit theory of philosophers. One might suppose it to have been the implicit theory of most citizens of, e.g. Nazi Germany and pre-War Japan. But again it does make sense, and on the theory it should not make sense, to ask ‘Ought I to do what Hitler (or the Emperor) commands?’ If ‘I ought to do X’ simply means ‘My doing X is commanded by Hitler (or the Emperor),’ the question reduces here also to the tautologous ‘Ought I to do what I ought to do?’
The main responsibility for this theory, as for several other false theories, lies in a failure to distinguish clearly between the meaning of ought and the criterion of ought. Many people have come, at all times in history, to accept as a belief not open to question that what they morally ought to do is whatever some specific ‘authority’ commands or wills; and this is apt to be taken as implying that for these people the moral ought has no meaning independently of the command of their authority. In fact it only implies that the command of their authority is for them the criterion or norm by which morally obligatory acts are determinable: not that even for them, this is the meaning of the ought. No doubt such persons often enough talk as though for them moral oughtness and the command of their favoured authority were identical. But this way of talking probably means no more than that the distinction between them is not one of which they have become explicitly conscious. For wherever there is unquestioning belief in the validity of a given criterion of the ought, wherever people have therefore ceased to ask (because the answer seems to them self-evident) ‘What ought I to do?’ there is little or no occasion for the explicit emergence in their minds of the notion of a criterion of the ought as something distinct from its meaning. Yet the distinction is surely latent in their minds all the same? Even the most fanatical Hitlerite would have deemed it significant—even if he regarded it as merely stating the obvious—to say, ‘I ought to do whatever the Führer wills’, He would not think he was merely saying ‘I ought to do what I ought to do’. But this is how the sentence would have to be translated if he in fact takes what the Führer wills to be the meaning, and not merely the criterion, of the moral ought.
The same criticism naturally applies, mutatis mutandis, to the Command-assertive theory where it takes a religious form, in which case the assertion is (roughly) that ‘ought’ means ‘commanded by God’: and I mention it separately only because it makes so much wider an appeal than any secular variety of the theory. To some religious men it may, indeed, seem a rather silly question to ask ‘Ought I to do what God commands?’ It may seem to them that the reasons for an affirmative answer are so compelling that no one ought to be in any doubt about it. But that they do not regard it as a meaningless question is obvious from the very fact that they are ready with their reasons to show that the affirmative answer is the right one. Yet if ‘I ought to do X’ means ‘I am commanded by God to do X’, the question ‘Ought I to do what I am commanded by God to do?’ becomes the purely tautologous ‘Ought I to do what I ought to do?’ ‘Divine Command’ may or may not be the criterion of the moral ought, but it certainly is not its meaning.
At the same time it must be conceded that there is, formally, a line of defence open to the last-ditch champion of these Assertive theories of the meaning of ‘ought’ against the objections we have been raising in terms of our second critical canon. He may deny that there is any confusion in his mind between ‘meaning’ and ‘criterion’. If he favours the Emotion-assertive theory, he may boldly avow that the question ‘Ought I to do what evokes or tends to evoke a pro-feeling in P?’ is for him a tautology, and that he finds no difference between the statement ‘This morally ought to be done’ and the statement ‘This evokes or tends to evoke a pro-feeling in P.’ A corresponding line of defence (or act of defiance?) is of course open to the Command-assertive theorist.
I must confess, however, that (as in the case of the last-ditch defence of the Hortatory type of theory against the objection from the side of the ‘silent’ moral judgment) this seems to me the sort of desperate refuge to which no moralist would have resort unless to save a theory at all costs. I think it reasonable to believe that most sponsors of the Emotion-assertive or of the Command-Assertive theory, once it is appreciated by them that their account of the meaning of ‘ought’ has the consequence of rendering tautologous the question ‘Ought I to do…?’, will recognise that their account is mistaken, and that they have in fact only been led to adopt it by permitting themselves to slip into the fallacy of identifying meaning with criterion. Still, it is logically open to them to continue to hold their theory if they are prepared to accept the consequence of the tautologous character of the above question. If they are so prepared, if they insist that the meaning conveyed by the words ‘this morally ought to be done’ is for them precisely identical with that conveyed by the words ‘This evokes or tends to evoke a pro-feeling in P’, or ‘This is what P commands’, there is not much one can do about it; though one may speculate with justifiable curiosity about the genesis in their minds of so eccentric a linguistic usage. One can be confident, however, that they will make few converts. For most people it will be more than sufficient to discredit the alleged identity of meaning that in their experience the word ‘ought’ has always connoted essentially a relation of acts to possible doers11—a relationship about which the equivalents suggested by the Emotive-assertive and the Command-assertive theories have nothing to say.
It is worth noting, however, that if the problem before us concerned the definition of ‘good’ rather than of ‘ought’, we could not quite so easily rest content with the ‘argument from tautology’. In his extremely able little book Logic and the Basis of Ethics Professor A. N. Prior suggests that it is not beyond the power of the Naturalistic moralist to explain, consistently with there being no non-naturalistic quality of goodness, how it might come about that, whatever naturalistic meaning X is offered in definition, it is felt to be still significant to say ‘X is good’. The pages he devotes to the topic deserve to be read in their entirety, but briefly his argument is this. As Mill has pointed out, ‘A name not infrequently passes by successive links of resemblance from one object to another, until it becomes applied to things having nothing in common with the first things to which the name was given; which, however, do not for that reason drop the name; so that it at last denotes a confused huddle of objects, having nothing whatever in common; and connotes nothing, not even a vague and general resemblance’. This, the naturalist may argue, has happened to the word ‘good’. Professor Prior states their case as follows:
‘At present, when we call a thing good we may mean that it is pleasant, or that it is commanded by someone, or that it is customary, or that it promotes survival, or any one of a number of things; and because we use the same term to connote all these characteristics, we think there must be some other single characteristic which they all entail; but in fact there is not. When it is said that being good means promoting survival, we are dissatisfied; we feel that it is still significant to say that promoting survival is good; and the same thing happens with every identification that is suggested; but this is just because, in each case, the other meanings are still hovering in our minds—to say that promoting survival is good is significant because it means that to promote survival is what we desire; to say that what we desire is good is significant because it means that what we desire promotes survival; and so on.’12
It is quite clear, of course, that this naturalistic ‘explanation’ can support no one specific naturalistic definition of ‘good’. On the contrary, the explanation rests upon the plurivocal character of the word ‘good’ in current usage, and the naturalist can do no more than recommend adherence to a specific usage as a useful linguistic convention. Hence even if a parallel ‘explanation’ were possible in the case of the word ‘ought’, the Emotion-Assertive and the Command-Assertive theories of the meaning of ‘ought’ could find therein no succour. It is much more important to notice, however, that a parallel explanation in the case of the word ‘ought’ has little or no plausibility. ‘Good’ is admittedly a word with a wide variety of current usages; but there is no comparable collection of naturalistic meanings of ‘ought’ in current usage by ‘ringing the changes’ upon which we might (analogously to the case of ‘good’) still find it significant to say ‘X ought to be done’ where ‘X’ stands for some one naturalistic meaning. One could perhaps allow that ‘ought’ is sometimes taken in current usage as equivalent to ‘required, under penalties, by law’, or to ‘decreed by social custom’, or to ‘commanded by God’. But reflective persons can usually be quite easily brought to see that none of these meanings is really the meaning he attaches to the moral ought. The fact that a small number of exceedingly reflective persons—i.e. certain philosophers—have lately offered us some new meanings of oughtness besides these (and of moral oughtness at that) is beside the point. For these new meanings have certainly not acquired currency in ordinary linguistic usage, and cannot, therefore, function in the way that Prior suggests that the different meanings of ‘good’ ‘hovering in our minds’ might function.
The third variety we distinguished in the Assertive type of theory—the Productive-Assertive—holds that ‘I ought to do X’ means, broadly speaking, ‘My doing of X will be productive of maximum good’. Its forms differ, often sharply, with the very different interpretations that can be given to the expression ‘maximum good’. If this view were offered as a theory of the criterion, and not of the meaning, of the moral ought, we should obviously require to treat it, in some of its forms, seriously and at length. As it is, it can readily be seen to collapse on the application to it of our old canon of criticism, and I do not propose to weary you with yet another variation on the old theme. Suffice it to say that if the objections advanced against the two previous varieties of Assertive theory are accepted as valid, there seems no ground for denying the validity of similar objections to the present variety.
7. Now we have not, of course, attempted in the foregoing to examine all the permutations and combinations that occur in contemporary analyses of the moral ought in terms other than itself. That would be impracticable. But we have, to the best of our belief, considered the leading types; and our conclusion is that there are insurmountable objections to each of them. I want to put forward now, as a ‘substantive motion’, that those philosophers are right who maintain that the moral ought is ultimate and unanalysable, incapable of definition save in terms that imply the definiendum itself. According to this view, we can know what the ought means in, but only in, actual experience of it. That meaning is as incommunicable to a person who cannot enjoy the requisite experience (if such there be) as the sensory quality ‘redness’ is to a colour-blind person. We can, indeed, put a man on the road to grasping the meaning of the moral ought by drawing his attention to the kind of situation in which it is typically experienced, and to some extent by the use of analogy; but these are never more than, at best, devices for aiding him to enjoy at first hand the direct experience which will alone reveal its nature. It is always, in a manner, unsatisfactory to have to say that one knows what a thing is, yet cannot by any description of it communicate that insight to others. But this is of course a misfortune by no means peculiar to the moral ought. It pertains to every indefinable notion; and some indefinable notions there must be. Besides, as in the case of other indefinables, we can (as I have just remarked) say something by way of a ‘pointer’ to its nature. For instance, it is not without reason that the moral ought is commonly spoken of as a ‘command’. It is certainly like a command in certain respects; though not, even in 2nd person moral judgments, a command issued by the person judging—that was the mistake of the Hortatory theory—but rather a command apprehended by the person judging. But it is also unlike a command in certain important respects. It differs from ordinary, non-moral commands, in the first place, in that it is not, for the experiencing subject, intrinsically connected with any specific imponent. We may indeed come, for extrinsic reasons, to associate it with God or with some other being as imponent. But a relationship to God or to some other being as imponent is not something intrinsic to the ‘ought’ as experienced. The moral ought can be experienced as the moral ought with no reference whatever to a Divine or any other imponent. And it differs from ordinary commands in this second respect, that it is experienced as having absolute authority for us. Of all ordinary, non-moral commands we may legitimately ask ‘What justification is there, if any, for its claim to our obedience?’ But in the case of the moral ought, as experienced, such a question has no point. ‘Why ought I to do what I morally ought to do?’ is a question that, for the moral consciousness, does not make sense. The moral ought, in other words, is experienced as an unconditional, or categorical, imperative.
But we must not, of course, suppose that when we say (rightly, I think) that the moral ought is a ‘categorical imperative’, and that the moral judgment asserts that certain conduct is unconditionally binding in some or all contexts, we are giving a description which will, per se, convey to another person a positive notion of the moral ought. He will be able to attach no positive meaning to a ‘categorical imperative’ if he has no direct personal experience of the moral ought, for nowhere else is a categorical imperative to be found. The description may well be of help in preventing him from looking for the moral ought in the wrong quarter. It can be no substitute for his own direct experience.
But what if a man tells us that, search as he may, he just cannot find in himself anything corresponding to this categorical imperative, and that if this is what a ‘moral consciousness’ implies then he at least is lacking in that commodity? He may admit, with some reluctance no doubt, that there is good evidence that most other persons do have it: and he should admit this, since most other persons, when it is put to them, can be persuaded to agree that they do not mean by the ought of morality any kind of merely conditional ‘ought’—‘I ought to do X, if I want Y, to which X is a means’. But he may insist that he at any rate never thinks of the ought in terms which imply more than a conditional significance. He recognises that certain ways of behaving (e.g. keeping faith) are required for the attainment of certain ends (e.g. social stability) which he finds himself approving for one reason or another; but the reason for approving them, he avers, is never any intrinsic obligatoriness conceived as belonging to the end. To the moral ought as a ‘categorical imperative’ he frankly declares that he can attach no meaning whatsoever.
Now if this position is consistently maintained by anyone, there is very little one can do about it. On the other hand, it is questionable if it ever is consistently maintained. At least very often, professed a-moralists slip into ways of speaking which suggest that what they are really concerned to reject is the supposed absolute obligatoriness of certain norms of behaviour proclaimed by their society, or even perhaps (in the case of large scale iconoclasts) by their ‘age’. The new standards they themselves propose, or which are implicit in their criticism of the old, seem to exact the same unconditional claim upon their allegiance as the old standards do upon the more conventionally minded. As with Nietzsche, ‘a-moralism’ sometimes signifies no more than the transvaluation of moral values—not their abnegation. A disguised ‘moralism’ may even reveal itself in a fervently felt ‘duty’ to preach ‘amoralism’!—just the same old categorical imperative in a new dress.
Again, one might (though perhaps with no great confidence) appeal to the professed a-moralist's modesty. If he admits that, as seems to be the case, there is good evidence that the great bulk of mankind, not merely at the civilised level but in the most primitive communities, do possess an authentically moral consciousness,13 then he has to make his choice between two alternatives. Either he may confess that there must be some defect in his analysis of his own experience, or he may claim the distinction of being differently constituted from almost all other members of the human race. It is uncertain which choice he would make. But it does not seem uncertain which choice we ought to make for him.
8. The existence in man of a moral consciousness—a consciousness of unconditional obligation, of a categorical imperative—I propose now to take as established. Is there a further question to be faced, the question of the objective validity of the moral consciousness? May it conceivably be the case that though we do have the idea of an unconditional obligation, the presence of that idea in us is of only subjective significance, and tells us nothing about the objective nature of things?
I take the view myself that there is no such further question; or, to be more precise, that the ‘further question’ virtually answers itself. The notion that moral experience may have only subjective significance arises primarily, I think, from the erroneous belief, already criticised, that the essence of moral experience lies in feeling or emotion. When the true nature of moral experience is appreciated, the objective validity of the ought can be seen to be little more than a corollary. But to justify this pronouncement I must insert here a brief reminder of what was earlier said about the essential nature of moral experience.
The crucial point I was concerned to make was that moral experience is never a mere feeling, but always includes judgment. It is, as I then expressed it, ‘a complex experience which requires both feeling and judgment for its proper characterisation’. Each by itself is a mere abstraction from the actual experience. This is a matter which can, and in the end must, be decided for everyone by personal experiment. Let any man imagine in himself what he is prepared to call a ‘moral’ pro-feeling (as contrasted with any other type of pro-feeling) towards his doing X, and let him answer whether he does not find that the judgment ‘I ought to do X’ belongs intrinsically to the experience. And let him imagine in himself what he is prepared to call a ‘moral’ judgment ‘I ought to do X’, and answer whether he does not find that a moral pro-feeling towards X belongs intrinsically to the experience. I must assume that he will find, as I find, that an affirmative answer is inescapable in both cases.
Now if judgment belongs thus to the essence of moral experience, a subjectivist interpretation of the ought seems to me to be ruled out eo ipso. It belongs to the essence of judgment to claim truth for what it asserts; and truth is always ‘of’ objective reality. Hence my experience of X as morally obligatory involves a judgment which can be legitimately expressed in the form ‘It is true of objective reality that I ought to do X’. In other words, the consciousness of moral obligation is at the same time the consciousness of that obligation as objectively valid, as rooted in the nature of reality.
It must be admitted, however, that there are certain considerations not directly connected with a false analysis of moral experience which tend to make people sceptical about the objective validity of the moral ought. I must say a few words about two of the commoner sources of confusion.
(a) I may be genuinely doubtful about what I ought to do, not merely in some particular, and perhaps rather complex, situation, but in regard to the general conduct of life. And lacking certainty about even the broadest principles of morality, my moral consciousness may never be in a position to express itself in any definite judgment ‘I ought to do this specific thing X’. There is, that is to say, a question in my mind about the objective validity of any specific embodiment of the moral ought. Now it is extremely easy, I think, to confuse the problematic objective validity of any specific embodiment of the ought with the problematic objective validity of the ought itself. But it is, surely, a confusion. The very statement ‘I do not know what my duty is’ presupposes the recognition that there is a duty for me, if only I could discern where it lies. The moral consciousness in any form implies the belief that there is something which I ought morally to do. And no more than that is required for the objective validity of ‘oughtness’ or ‘duty’ in general.
(b) It is a commonplace that the moral codes of different peoples, or even of the same people in different ages, show striking divergences from one another, and in some respects contradict one another. This is apt to suggest to the mind that there is not a single moral consciousness common to all men, but a number of moral consciousnesses operating according to different and to some extent mutually conflicting, principles. Now if such be the case, it is indeed very difficult to see what could be meant by the ‘objective validity’ of the moral consciousness. There will apparently be no one objective moral order. And to say that there are several different objective moral orders seems tantamount to admitting that there is no objective moral order at all.
It seems to me, however, that the inference from the plurality of conflicting moral codes to a plurality of moral consciousnesses is, to say the least of it, over-hasty. The inference would have some justification if it were clearly impossible to interpret the conflicting moral codes as compatible with an underlying identity of ultimate moral principle. But is this impossible? Let us suppose that there is a common moral consciousness, and let us further suppose, merely for the sake of the argument, that it is informed by a single ultimate principle which is the obligation to promote the well-being of one's community. It is surely obvious that such an ultimate principle will manifest itself in very different rules of conduct, or ‘moral codes’, in different societies, corresponding to their differing cultural beliefs and different material, economic and social backgrounds? It is not merely understandable, but inevitable, that differently circumstanced societies should take different views as to what modes of behaviour best promote communal well-being; not to speak of different views as to what constitutes one's community—e.g. the family, the clan, the tribe, the nation, or all mankind. Hence rules of conduct that are superficially in violent conflict may be in reality quite compatible with a common moral consciousness functioning in accordance with the same ultimate principle.
Nor is this mere idle speculation upon abstract possibilities. Modern social anthropology has, I suggest, succeeded in making intelligible, and even in a sense acceptable, to our moral consciousness what at one time seemed to be wild and mysteriously alien ‘moralities’ in primitive peoples. Even the most bizarre moral rules have turned out to make sense as ‘instruments’ for the promotion of communal well-being, when interpreted in the light of a sympathetic appreciation of the whole cultural and material milieu of the societies in question. Indeed I think it hardly too much to say that while the relatively external and superficial reports of moral customs supplied by early anthropologists provoked natural doubts about the unity, and therefore the objective validity, of the moral consciousness of man, the far more intensive studies of later anthropologists, who have taken infinite pains to get, as it were, ‘inside the skins’ of those whose customs they seek to understand, have worked in exactly the opposite direction, and have provided truly impressive evidence of a fundamental identity of moral principle throughout the whole human race.
9. A final word about the metaphysical implications of accepting the moral consciousness as having objective validity. For these seem to me to be of enormous importance. If the moral ought is objectively valid, this means nothing less than that a ‘moral order’ is somehow ingredient in the very nature of things. To develop the ‘somehow’ into a definite and intelligible concept is, of course, a task of the most formidable character, which only metaphysics is competent to discharge—if even it be competent. But in order to know that, as distinct from how, a moral order is rooted in the nature of reality, there is no need to wait for a metaphysical theory. If we are prepared to grant the objective validity of the moral ought, the fact of an objective moral order follows as a direct implication. We have need of a metaphysical theory only to know (if we can know) the manner in which ‘the nature of things’ incorporates that moral order.
Indeed, it may well be doubted whether the achievement of a sound metaphysical theory of the universe is of quite such moment to humanity as philosophers have often liked to suppose. Not quite so much would now seem to hang upon it. Certainly to know that a moral order belongs to the very nature of things, and is no mere figment of man's imagination, is not to know everything we should like to know about this universe in which we find ourselves. But it is to know (or so it seems to me) that which effectively eliminates the most haunting of those ultimate fears that chill the heart of reflective man; the fear that human existence has no meaning or purpose beyond itself. For it is just not possible to believe that the moral order is objectively real—that there really are obligations unconditionally binding upon us as human beings to behave in certain ways—and to believe also that human existence has no meaning and purpose beyond itself. Perplexity about what in particular we ought to do may, and no doubt in some degree always must, remain with us. But the dejection of spirit that darkens and corrodes the inner life of so many men today has its primary source, surely, not in mere inability to discern wherein their duty lies, but in an inability to be sure that there really is any duty to discern. To banish this doubt is, as I believe, to emancipate the human soul at a stroke from by far the most tormenting and most spiritually debilitating of all its cosmic forebodings.
10. I have left little time for a summing-up of the general argument of this course. But perhaps, after all, little time is needed. Its object has been to find whether a view of the self could be rationally defended which makes intelligible the sort of language theology is constrained to use in talking about the human soul. Such language is clearly meaningless if the term ‘self’ stands for no more than some inter-related set of particular states and events; which is all that most modern empiricists are prepared to recognise by it. But it is otherwise if, as the argument of these lectures has sought to show, human experience is found to imply a unitary and relatively enduring subject, a being not reducible to experiences, but manifesting itself in experiences; a being in which, in the manifestations by which we know it, spirit and body are intermingled, and ‘compose a certain unity’, yet a being of which the essence is spiritual, in the sense that while we can conceive its existence as possible without body, we can attach no meaning at all to its existence without spirit; a being which, though subject to manifold influences from without, is nevertheless endowed with a creative power which suffices to constitute it a free and responsible agent, master, in a real sense, of its own destiny; a being, again, which is intrinsically related through its moral consciousness and moral will to an objective moral order, whereby there is imparted to the life of man on earth a no less than cosmic significance; and, finally (for religion, perhaps, most significantly of all) a being that is for itself, in the last resort, a mystery; knowing in some measure, indeed, what it is, but knowing not at all how it is what it is.
Does the human self as so characterised lack anything that could be claimed as indispensable to the theological conception of the soul? It seems to me that it does not. The soul is always, I think, conceived in theology as a relatively abiding entity which maintains its identity throughout changes; as a spiritual entity—however close may be its affiliations to an animal body; and as an active entity, which is no mere passive plaything of external forces. These are the minimal requirements, and a philosophical theory of the human self which is incompatible with any one of them logically entails the rejection of all theological talk about the human ‘soul’. But each of these requirements, I submit, is adequately met by the theory of the self which has emerged from the analysis of human experience conducted in these lectures.
If this be so, our question whether the nature of the self as it discloses itself to philosophical investigation is such as to render nonsensical theological talk about the soul is answered in the negative. The still larger question, whether there is in reality anything corresponding to theological talk about ‘God’, and, if so, how the nature of this ‘God’ and His relationship to the human soul ought to be understood, is what we are to debate in our second course.