Under this heading come the papers by Professors Frankena, Hedenius, and Kuhn; Mr. Hare's paper; and one section of Professor Blanshard's.

(A) PROFESSOR FRANKENA'S QUESTIONS.

      In order to formulate the questions which Professor Frankena puts to me, I will begin by introducing the phrase "moral sentence in the indicative." This is to denote a sentence in the indicative mood, in which the grammatical subject is a name or a description of a person, an action, an experience, or a disposition (or of a class of such), and the grammatical predicate is some word like "ought" or "ought not," "right" or "wrong," "good" or "evil," used in its specifically moral sense. It would not be difficult to show by instances and counter-instances what I have in mind

      In terms of this phraseology, I think that what Professor Frankena asks me may be summarised as follows: -- Have I any decided opinion and, if so, why do I hold it, on the following interconnected questions?

1. Do moral sentences in the indicative express judgments or not?
2. If not, what does the utterance of such a sentence express?
3. If so do words such as "ought" and "ought not," etc., when used in their specifically moral sense, stand for predicates of a certain peculiar kind which has been described as "non-natural?"
4. If such words stand for predicates which are "natural," what account should be given of the "natural" characteristics for which typical words of this kind stand?

      Now a short answer, and a true one so far as it goes, to Professor Frankena's questions would be: No! I have no decided opinion on any of these points. But I could say the same about almost any philosophical question. The reasons which incline one to or against a certain opinion on any one philosophical question are always highly complex, and they are always bound up with the reasons which incline one to or against certain opinions on many other philosophical questions. Here, as elsewhere in philosophy, I have tried to clear up the questions and to indicate logical connexions between certain answers to some of them and certain answers to others. These are necessary preliminaries to any attempt to come to a reasoned decision about them. But it does not follow that it is sufficient to enable a person to do this. So far as I am concerned, I find myself now inclined to favour one kind of alternative and now another, but never to come down decisively in favour of any. At most I feel fairly confident that some proposed answers to some of the questions are inadequate by themselves.

      I will now try to be a little more concrete. Let us give the name "predicative" to all theories which hold that moral sentences in the indicative express judgments, in which a moral attribute is ascribed to a person or action or experience or disposition. I will begin by mentioning and dismissing one general argument against all predicative theories, which has been thought by many intelligent contemporaries to be conclusive.

      It is alleged that a sentence can express a synthetic judgment, if and only if one can conceive and describe some kind of possible perceptual situation or introspectable situation which, if realised, would tend to confirm it or to invalidate it. Now consider such a sentence as, e.g., "Acts of promise-breaking tend as such to be morally wrong." If this expresses a judgment at all, the judgment is certainly not analytic. But, it is said, one cannot suggest any possible perceptual or introspectable situation which, if realised, would tend to confirm or to invalidate what it expresses. So it is concluded that it cannot express a judgment. And a similar argument is applied to all moral sentences in the indicative.

      This argument leaves me wholly unmoved. The account of synthetic judgments, which is its main premiss, is obviously a generalisation based exclusively on a review of non-moral indicatives, and in particular of statements about physical and psychological phenomena. Now there are admittedly whole classes of sentences in the indicative which seem prima facie to express synthetic judgments, and which are plainly not of that kind. Moral indicatives are important instances of them. If you first exclude all such sentences from your purview, in making your generalization about the conditions under which alone a sentence can express a synthetic judgment, and then use that generalisation to show that such sentences cannot express synthetic judgments, you are simply begging the question. For the only legitimate ground for excluding these from your purview, and nevertheless holding that your generalisation covers all sentences which express synthetic judgments, would be a prior conviction that these sentences do not express synthetic judgments.

      Dismissing this kind of argument as circular, I would next remark that there are two general principles to which I should appeal in preferring one type of theory to another. They sound rather platitudinous when stated baldly; but, in default of anything better, they are not to be despised.

1. Other things being equal, a theory is to be preferred if it does not have to postulate anything of a kind which is not already admitted as a fact and found to be readily intelligible.
2. Other things being equal, a theory is to be preferred if it does not have to suppose that all men are fundamentally mistaken on certain matters with which the whole race is and has always been constantly concerned.

      On the second principle, taken by itself, I should be strongly inclined prima facie to prefer an ethical theory of the predicative kind to one of the non-predicative kind. The normal use of uttering a sentence in the indicative is undoubtedly to convey information (true or false). The fact that our moral utterances are commonly couched in the indicative mood strongly suggests that most men at most times take for granted that they are making and expressing and conveying to others moral judgments on such occasions. If they are in fact doing nothing of the kind, but are only e.g., evincing or evoking certain emotions, issuing certain admonitions or commands, etc., their mode of expression seems to betray a fundamental misapprehension of their situation.

      On the first principle, taken by itself, I should be inclined prima facie to favour an ethical theory which holds that moral concepts are empirical, in the sense that they are derived from data presented in sense-perception or introspection, in the familiar ways in which, e.g., the concepts red or angry are derived, and the concepts mermaid or hot-tempered are derived. On the same principle I should be inclined prima facie to favour a theory which makes universal propositions of the form:

Anything that had the non-moral character N would have the moral character M

      Now, in formulating the two principles I have prefixed to each the conditional clause "other things being equal." The basic requirement of a philosophic theory is that it shall do justice to all the facts characteristic of the region with which it deals (including, of course, "higher-order" facts about the inter-relations of the "lower-order" facts), and that it shall neither ignore nor distort any of them. When this fundamental condition of inclusiveness and non-distortion is taken into account, I think that the two principles point in opposite directions.

      I have tried to show, in various papers quoted by Professor Frankena, that it is doubtful whether any predicative theory can do justice to the facts unless it admits

1. that the concepts of moral attributes are non-empirical, and
2. that there are universal propositions, connecting certain non-moral attributes with certain moral ones, which are synthetic and yet necessary.

      It is plain that philosophers of two different kinds, who might agree in accepting my argument up to this point, would here diverge from each other.

1. Some are quite convinced that there can be no non-empirical concepts and no synthetic a priori judgments. They will have to accept some form of non-predicative theory, and make the best of it.
2. Others (including myself) have no such convictions. They will be in a freer position. They are not obliged at the next move to accept any form of non-predicative theory, but they are equally not obliged at this stage to reject all forms of it. They can view that type of theory sympathetically as a praiseworthy attempt to do without non-empirical concepts and synthetic a priori judgments in an important region of human experience. They may even offer a helping hand, as I have tried to do in certain of the writings quoted by Professor Frankena.

      Those who feel obliged to accept some form of non-predicative theory will be most usefully occupied in the following tasks.

1. In trying to account plausibly, in terms of their theory, for the main outstanding facts which seem prima facie to demand a theory of the predicative type.
2. In trying to adduce facts which seem to fit better into a non-predicative type of theory than into any of the predicative type. One such fact, e.g., is that the state of mind (whatever it may be) which is expressed by uttering sincerely and wittingly such a sentence as "That act would be wrong," always tends to evoke a reaction against doing the act in question. It might be alleged that this seems to be a necessary proposition, and not a mere empirical generalisation about human nature. Now it might be argued that, if what such a sentence expresses is a judgment, one will have to hold either
   1. that the psychological proposition in question is merely an empirical generalisation, or
   2. that it is a necessary synthetic proposition known a priori.

      Whether the non-predicativists have succeeded in these tasks or not, I think that there is no doubt that, in the course of their very strenuous efforts to perform them, they have made some valuable contributions to moral philosophy. At the time when I wrote FTET moral philosophy in England and the U.S.A. might fairly be described as dormant and apparently moribund. Since then, partly owing to the writings of certain predicativists (like Prichard and Ross) and partly owing to those of certain non-predicativists (like Professor Stevenson and Mr. Hare), it has become one of the liveliest branches of philosophy. Plurimi pertransibunt et multiplex erit scientia.

      There is one other topic, closely connected with those which I have discussed above, on which I will briefly comment. That is the phrase "non-natural characteristic." As a student at Cambridge I was brought up to believe that it is a fundamentally important proposition of ethics that moral attributes belong to a peculiar category called "non-natural," and that there is something called "the naturalistic fallacy," which most moralists had committed who had written before the light dawned in 1903. When I became Professor of Moral Philosophy, and had to write a course of lectures on ethics, I was unable to discover any intelligible and tenable account of the meaning of this distinction between "natural" and "non-natural" attributes. It also seemed to me that, unless "fallacy" be used in the improper and question-begging sense of "mistaken opinion," instead of in its proper sense of "invalid bit of reasoning," there was nothing which can be described as "the naturalistic fallacy."

      I do not propose to traverse again now this much trodden ground, but I will state briefly and dogmatically the conclusion which it seems fair to draw. If words like "morally good (or evil)," "morally right (or wrong)," etc., stand for characteristics, then the characteristics for which they stand differ from non-moral ones in being dependent on the latter in a way in which no non-moral characteristic appears to be dependent on others. No doubt some non-moral characteristics are necessarily dependent on others, e.g., to have a shape entails having a size. But none of these cases of necessary connexion between non-moral characteristics seems to be at all like the connexion between being a breach of promise and being morally wrong, which we express by saying that being a breach of promise necessarily contributes towards making an act morally wrong.

      Now a non-predicativist might accept all this, and simply use it as water for his own mill. He might proceed to argue, in accordance with my first Principle, that any ethical theory which can avoid postulating characteristics of such an odd kind as moral ones would have to be, if there were such, is to be preferred (other things being equal) to one which has to postulate them. Suppose he could then explain in detail, in terms of a certain form of non-predicative theory, how it comes about that moral adjectives seem to stand for characteristics of this peculiar kind. Then I think that there would be a fairly strong prima facie case for preferring his form of the non-predicative theory to any form of predicative theory known to me.

      Now non-predicativists have attempted such detailed explanations. I am impressed, if not completely convinced, by their efforts up to date and I am inclined to think, at the moment of writing, that it is likely that the truth lies somewhere in that direction rather than on predicative lines. I could not be more definite if Professor Frankena (that kindest of men) were to hold a pistol to my head, which I cannot imagine him doing.

(B) MORAL PHILOSOPHY AND MORAL PRACTICE.

      The main topic which Mr. Hare discusses is the bearing or lack of bearing of moral philosophy on moral practice. As regards the historical part of his essay I would make the following comments.

      Mr. Hare rightly mentions Moore and Prichard as the two most influential English moral philosophers at the time when I was young and for many years afterwards. Each held that the moral concepts which he took as fundamental are not only unanalysable, but also of a unique and peculiar kind. Now anyone who takes such a view must, if he would be consistent, hold that any proposition, in which the subject is described in purely non-moral terms and the predicate is or involves one of these moral notions, must be synthetic. Mr. Hare thinks that this commits such a philosopher to the particular epistemological view, called by Sidgwick "aesthetic intuitionism." This view he ascribes to Moore and to Prichard, and he thinks that for those who hold it moral philosophy can give no guidance to those who seek to know what they ought to do in various types of situation.

      Now I do not think that a person who holds the Moore-Prichard type of theory as to the nature of moral concepts is necessarily committed to aesthetic intuitionism. The latter view may be stated roughly as follows. The only way to discover what is morally good or morally obligatory (as the case may be) in a particular situation is to put oneself actually or imaginatively into that situation, and to note what kind of value-judgment or deontic judgment one then makes. Now I do not doubt that it would be a necessary preliminary to giving practical guidance to others that one should oneself often have done what the aesthetic intuitionist has in mind. It would also be a necessary preliminary that other men should have done the like, and should have recorded the moral judgments which they then made. But at that stage there are the following two conceivable developments.

      (1) Suppose a person admits (as Sidgwick certainly did, and as I imagine both Moore and Prichard would do) the possibility of necessary synthetic universal propositions, which can be seen to be true ex vi terminorum. Then it is conceivable that one might arrive by "intuitive induction" at a number of synthetic a priori axioms, stating necessary connexions between certain non-moral and certain specifically moral characteristics. This alternative would no doubt be rejected unhesitatingly by Mr. Hare and by most of his English and American contemporaries. But in a historical account it must be remembered that it has been held by many eminent and influential moral philosophers.

      (2) Even if this alternative be rejected, there remains the theoretical possibility of inductive generalisations, of a high order of generality and reliability, similar in content to the alleged synthetic a priori axioms of the rejected view.

      Now such a set of moral axioms, or of well established moral inductive generalisations, might be capable of elaborate deductive development, and might be found to entail consequences which no one could have foreseen. These consequences, together with factual information about the situation in which a particular person is placed, and about the probable consequences of this, that, or the other alternative action might enable a moral philosopher to provide him with valuable (though never infallible) guidance as to how he morally ought to act.

      The legitimate source of scepticism here is of course the very general conviction that none of these "mights" is in fact realised. The first alternative would involve admitting that there are synthetic necessary propositions knowable a priori, and this is very commonly held to be an exploded superstition. The second of them, though it might be admitted to be theoretically possible, seems not in fact to be true. Either

1. there are no well established inductive generalisations in morals; or
2. if there are, they do not (like, e.g., the laws of motion and the law of gravitation) form a system capable of elaborate deductive development and detailed application.

      Passing from the historical to the other parts of Mr. Hare's essay, I agree that many young persons take up the study of philosophy because they are morally perplexed and hope that moral philosophy will give them practical guidance. But I think that this attitude covers a number of different troubles and demands, and I propose to distinguish some of them.

      (1) A person may have been brought up to accept as unconditional a number of general moral principles, as to how one ought or ought not to act in any instance of certain frequently recurring types of situation. It may be that each of these maxims, considered in isolation on its merits, still seems to him on reflexion to be obviously true. But he may become aware, either in his own life or in the lives of others, of situations in which several of these principles are relevant and it is impossible to act in accordance with one without acting against another

      Moral philosophy could help here, if it could carry out the following programme.

1. Indicate a certain more general principle, which seems on its merits to be at least as obviously true as any of the more special ones.
2. Show that, in acting on each of the more concrete principles in the relevant kinds of situation, one will generally (though not invariably) be acting in accordance with this more general one.
3. Show that, in the exceptional situations, where several of the more concrete principles are relevant but it is impossible to act in accordance with all of them, this more general principle provides a satisfactory answer to the question how one ought to act.
4. Suggest the causes which may have made the more concrete maxims seem to be true in their unconditional form, when really they are true only in the majority of situations in which they are relevant.

      (2) A great many conscientious plain men and several very eminent moral philosophers, e.g., Plato, Butler, and Sidgwick, seem to hold the following conviction. All moral maxims are subject to a certain implicit condition. When this is made explicit, any acceptable moral maxim would take the form:

"In situations of the kind S one ought always to behave in the way W, if and only if such behaviour would not be in the end and on the whole detrimental to one's own interests."

      If such a person appealed to moral philosophy in his perplexity, its first move should be to clear up the many ambiguities in the phrase "one's own interest." Is this supposed to be confined to one's own happiness or unhappiness; or is it to be extended to cover the improvement or worsening of one's own character, intellect, and personality? If the latter, is it to be confined to improvement or deterioration in non-moral respects, or is it to be extended to cover specifically moral improvement or deterioration also?

      So much might fairly be regarded as within the range of moral philosophy. But what might be demanded is an assurance that behaviour, which we all agree to be morally obligatory, but which often seems to be to all appearances detrimental to the agent's long-term "interest" (however that may be interpreted), can never really be so, and therefore is no exception to the general principle in question. Now it seems to me that any attempt to show this would fall outside the realm of specifically moral philosophy, since it would turn on the nature and destiny of the human individual and the organisation of the rest of the universe. Philosophy has traditionally been held to be closely concerned with such questions, but the prevalent view among professional philosophers in England and America at the present time is that that is an elementary mistake.

      (3) What troubles many intelligent and conscientious persons nowadays is something still more fundamental. There is a certain view of the nature and destiny of man, which seems to have the whole weight of biology and experimental psychology behind it, viz., a "behaviourist" or "epiphenomenalist" view, which I will call for short "scientific materialism." To many people it seems that, if this view be true, the notion of moral obligation must be a mere figment, which arose somehow in the days of men's ignorance of their nature and destiny, and now survives precariously like a vestigial organ. When they contemplate the scientific evidence they cannot help accepting the materialist account of human nature. When they are engaged in co-operating or competing with their fellow-men they cannot help thinking that they have moral obligations. When they try to bring together these two convictions into one focus it seems impossible to reconcile them. They naturally, and I think quite legitimately, appeal to professional philosophers to help them.

      Now philosophers might seek, and in fact have sought, to do this in various ways. One is to try to show that, when the scientific materialist view of human nature and the notion of moral obligation are both properly understood, there is no incompatibility between accepting the former and continuing to hold that men are subject to moral obligations. This type of solution will be helpful, only if it can succeed without having to give such an account of moral obligation as seems to the intelligent and conscientious non-philosopher to distort it or eviscerate it or altogether to dissolve it. Another way would be to admit the conflict, but to deny the adequacy and the ultimate coherency of the scientific materialist account of human nature, whilst granting its plausibility and usefulness in the limited context in which it has arisen. I think that the first type of answer might fairly be said to fall within moral philosophy, and the second only within philosophy in a wider sense.

      It would take me too far afield to attempt to discuss adequately the "test" for rightness or wrongness, which Mr. Hare very tentatively puts forward at the end of his essay. I will consider only the following point. Mr. Hare says that A will be inclined to judge it to be wrong for him to treat B in a certain way, if, on imagining himself to be in a similar situation as patient instead of agent, he finds that he would dislike to be treated in that way. What is not clear to me is what Mr. Hare takes to be the relevance of this "dislike" on A's part.

      It seems to me that all that is logically relevant is that A should judge that it would be wrong for another to treat him as he is proposing to treat B. Whether he would dislike or like being treated in that way seems logically irrelevant.

      Perhaps Mr. Hare wishes to assert only the psychological proposition that A will be inclined to judge that it would be wrong for another to treat him as he is proposing to treat B, if and only if he would dislike to be treated in that way. If so, I think it is a very doubtful generalization. Perhaps, then, what Mr. Hare wishes to assert is only the following. A needs to be convinced that he would dislike to be treated in the way in question, not in order to judge that such action by another towards him would be wrong, not in order to judge (in accordance with Mr. Hare's principle) that such action by him towards B would be wrong, but in order that the latter conviction should have any practical effect on his conduct towards B. If that is what Mr. Hare means, I think it is a rash generalisation about human motivation.

      I am inclined to think that the only relevance of A's disliking the experience which he would have if he were to be treated as he is thinking of treating B is this.

1. An important, though neither a necessary nor a sufficient reason for thinking that it would be wrong to treat B in a certain way, is that B would dislike to be so treated.
2. An important, and perhaps indispensable, way for A to gain a vivid and practically effective belief that B would dislike a certain experience is that A should imagine himself to be having a similar experience in similar circumstances, and should find the idea strongly distasteful.

(C) "OUGHT" AND "CAN."

      The relations between the former and the latter of these notions form the main topic of Professor Hedenius's paper. I would like at the outset to make the following general remark. The treatment of the whole subject in my lecture "Determinism, Indeterminism, and Libertarianism" is extremely condensed and somewhat dogmatic. It omits much that should be included in any adequate discussion; the points raised are not sufficiently developed; and objections and counter-arguments are not considered. Such defects are inevitable when a vast and intricate subject has to be handled in the course of an hour's lecture.

      Professor Hedenius draws a distinction between acts which are morally obligatory and acts which are morally imputable to the agent. He argues that a conceivable act, which it is impossible or inevitable for an agent to do, may nevertheless be morally obligatory. But he holds that, for an act to be morally imputable, it must be at any rate what I have called "conditionally substitutable." I am inclined to think that any difference between us on this matter depends mainly on different usages of certain terms, which undoubtedly are used sometimes in a wider and sometimes in a narrower sense. I will now proceed to develop this suggestion.

      Consider the statement that A is under an obligation to do X at t. Does this entail

(1, 1) that it is not impossible for him to do X at t?

(1, 2) that it is not inevitable for him to do X at t?

(2, 1) that it is not inevitable for him to do Y at t?

(2, 2) that it is not impossible for him to do Y at t?

      I think it is easy to show that (2, 1) can be reduced to the form of (1, 1), and (2, 2) to the form of (1, 2). In order to do this one need only note that an obligation not to do Y is equivalent (subject to two conditions which I will state in a moment) to an obligation to do something-other-than-Y. The two conditions are these.

1. It is to be understood that "to do something other than Y" includes, as one alternative, refraining from all positive relevant action, e.g., just not answering a question.
2. It is also to be remembered that to be under an obligation to behave in one-or-another of several alternative ways does not entail being under an obligation to behave in any particular one of those ways.

      Professor Hedenius is undoubtedly right in saying that we often use expressions which seem to imply that the alleged entailment in (1, 1) does not hold. Here are some examples.

• "He ought to have lectured from 9 to 10 A.M. yesterday; but it was impossible, since he was then undergoing an operation."
• "He ought to be lecturing now; but it is impossible since he is now stricken with aphasia."
• "He ought to begin to lecture at 6 P.M. in London this evening; but that will be impossible, since it is now 5 P.M. and the train in which he is travelling from Cambridge is held up by a derailment at Bishop's Stortford."

      I am very doubtful, however, whether these expressions in fact show that the entailment alleged in (1, 1) does not hold. I suggest that in each of them "ought" is used in a certain conditional sense; that the condition is regarded as obvious and as nearly always fulfilled; and therefore is not explicitly stated. I would expand my first example as follows: --

"If and only if he had been able (as he normally would have been) to lecture from 9 to 10 A.M. yesterday, he would have been under an obligation to do so. But (owing to the exceptional circumstances of undergoing an operation at the time) it was then impossible for him to do so, and therefore he was not in fact under an obligation to do so."

      It should be noted that the collapse of a categorical obligation, through the impossibility of performing the relevant action, very often imposes on the agent a categorical obligation to perform a certain other action, which is in his power. The lecturer in the delayed train, e.g. ought, if he can, to send a telegram to the person in charge of the arrangements for his intended lecture in London.

      Let us now consider the alleged entailment (1, 2), i.e., that if A is under an obligation to do X at t, it follows that it is not inevitable for him to do X at t. Can we think of a relevant and obvious counter-instance?

      The first point to notice is this. An action, such as answering (truly or falsely) a question, returning or withholding a borrowed article, etc., has to be considered in two aspects, viz., in reference to the person affected by it and in reference to the person doing it. In respect of the patient the important question is: Does the action in fact treat him as he has a right to be treated in the situation? In respect of the agent the important question is: Is the action done from the intention (inter alia) of treating the patient as he has a right to be treated in the situation? An action of the former kind may be called "right-securing," and one of the latter kind "right-intending."

      Now I think it is certain that we often use "obligation" and "obligatory" in such a way that an action which the agent is under an obligation to do is one that is right-securing, whether or not it be right-intending. If we use our terms in that way, it is obvious that an action which the agent could not help doing may be obligatory upon him. (It is equally obvious that one which he could not possibly do might be obligatory on him.)

      But I think it is no less certain that we often use "obligation" and "obligatory" in such a way that an action which the agent is under an obligation to do must be right-intending. Now it seems to be that an action, which the agent could not help doing, might indeed be in accordance with an intention on his part to treat the patient as he has a right to be treated in the situation. But one could hardly say that such an action was done from that intention (inter alia). So I do not think that an action which the agent could not help doing could be called "obligatory," if that word is used (as it often is) to connote right-intending and not merely right-securing.

      Professor Hedenius says, quite correctly, that we can talk of a man being forced to do his duty in a certain manner, e.g., forced to repay money that he owes. I doubt, however, whether this is relevant to the issue. In the first place, "duty" is here used in the first of the two senses which I have just distinguished. What we mean is that A is forced to do an act which in fact treats B as he has a right to be treated. And, secondly, to say that A was forced to do X is not generally equivalent to saying that it was inevitable for him to do X. What it generally means is that A would have preferred antecedently not to do X, but that he was in a situation where it was practically certain that the consequences to him of not doing it would be extremely unpleasant. It was open to him to refrain from doing X and to put up with the unpleasant consequences. So his doing of X was not inevitable.

      Very likely I used "obligable" in my lecture in roughly the sense in which Professor Hedenius uses "morally imputable." Let us assume this for the sake of argument, and use the latter phrase in the rest of the discussion. I understand that Professor Hedenius is inclined to agree, up to a certain point, with my account of the conditions which must be fulfilled if it is to be morally imputable to A that he behaved in the way W in a certain situation S. He agrees with me up to the point that A's behaving in the way W would not be morally imputable unless it were, in a certain sense, "determined by A's ego or self." Now I offered a certain analysis of this latter condition, and said that it seemed to me self-evident that it could not be fulfilled. Professor Hedenius offers an alternative analysis, which would not be open to that objection.

      If I understand him aright, the essential features in his account are as follows. We have at the back of our minds a reference to a certain large class of persons (e.g., contemporary middle-class Englishmen above the age of puberty); and we have the thought of a certain type of personality as normal in that class in respect to the nature and strength and organisation of a number of important conative-emotional dispositions (e.g., desire for food and drink, desire for money, sexual desire, tendency to react with hostility when thwarted, and so on). The agent is assumed to be a member of such a class of persons. We regard a bit of behaviour on the part of a member of such a class as "determined by his ego or self," when and only when the following conditions are fulfilled.

1. The stimulus must be of a kind to which
   1. all members of the class are quite often subjected, and
   2. in response to which most of them on most occasions would behave in a certain way Z.
2. The individual in question A behaved, when so stimulated, in a markedly different way W.

      Now I think that the distinctions which Professor Hedenius draws are important in reference to the degree of merit or demerit which we ascribe to a person in respect of a bit of intentional behaviour. We do not get morally excited when a person behaves rightly under circumstances which frequently occur in the lives of all of us, and in which most of us generally do act rightly. Nor do we get morally excited when a person behaves wrongly under circumstances which are highly exceptional, and in which one suspects that most such persons would act wrongly and is very doubtful whether one would have acted rightly oneself.

      It seems to me that all this comes fairly easily under the sense of "ought" and "ought not" which I described in paragraph (ii) of the Section entitled, 'Various Senses of "Obligable" ' in my lecture. I said of this that "a clear-headed Determinist should hold either that this is the only sense, or that, if there is another sense, in which obligability entails categorical substitutability, it has no application." But I added that I am inclined to think that we often use "ought" and "ought not" in another sense, and that in this other sense they entail categorical substitutability. I think that this is most obvious when one makes judgments about oneself, of the form "I ought to have done so-and-so" (which I did not do), or "I ought not to have done so-and-so" (which I did). I cannot help thinking that a reference to what the average middle-class Englishman above the age of puberty would or woud not generally do, when subjected to the stimulus to which I was subjected, would serve only as a rough measure of the degree of my delinquency, and not at all as an analysis of my conviction that I, under the very circumstances in which I in fact failed to do my duty, could instead have done it.

(D) THE EXISTENTIAL ACCOUNT OF HUMAN PERSONALITY.

      I understand Professor Kuhn to be using "existence" throughout nearly the whole of his essay in a certain technical sense, viz., to denote the peculiar kind of being which he holds to be characteristic of a person, and to be revealed to each of us by the reflexive awareness which is an essential factor in personality. If I understand Professor Kuhn aright, what he takes such reflexive awareness to reveal to each person may be described as follows. What a person now is is what he has made himself, through the reaction of himself as active, spontaneous, and selective, upon himself as passive and malleable. Furthermore, he, as he now is, is actively engaged in determining and generating himself as he will become, by a further process of selection and action. This interest in, and self-direction towards, the future is particularly characteristic of a person. Moreover, each person has a unique and fundamental concern for himself, and this is alleged to be an essential condition of "the absolute validity of moral obligation and moral claims in a person."

      I am willing to accept much of this, if I am allowed to interpret it as follows, and to put certain qualifications upon it. In the first place, it is certainly characteristic of human beings (as contrasted with other animals, and especially with certain insects) to be born with extremely few and comparatively unimportant first-order dispositions. They are born, instead, with what Professor Ducasse calls "aptitudes," i.e., dispositions to acquire dispositions and to organise those which they acquire. In so far as statements to the effect that a person is not "an entity fixed and bound by its own whatness" are interpreted in this way, I think that they are true and important.

      On the other hand, we must not overlook the fact that what a person can make of himself, even under the most favourable conditions, is limited by his innate endowments. It is true that no one knows even approximately what are his own or another person's ultimate limitations. It is true too that it is generally undesirable for a person to dwell on this topic in his own case, or for his neighbours to express a confident and narrow view about it. Lastly, it is true that experience shows that a person, who seems prima facie to be hopelessly handicapped, physically or intellectually or morally, sometimes does (if he seriously takes himself in hand, and if others give him understanding help) achieve a development of personality which seems well nigh miraculous. But I see no reason to believe that the possibilities are in fact unlimited in any case, or that the limits in each particular case are not fixed by the innate constitution of the individual.

      Allowing that there is an important sense in which it is true that each of us is continually making and re-making himself, we must not exaggerate the part played by the deliberate action of the individual himself in this process. In the case of most of us it is but fitfully and for short periods that one "takes oneself in hand" and sets out to make oneself a person of such and such a kind. In the main each man's personality is moulded for him in early life by the pressures of family, of school, of business, by the newspapers, the wireless, and the films. These influences are (after occasional struggles, which leave their scars in all, and mar the personalities and wreck the lives of some) generally assimilated fairly thoroughly, though of course in modo recipientis. Thereafter the reactions of most men of a given social group in normal situations are almost automatic. Doubtless the power to make a hard deliberate choice which one realises will profoundly modify one's life and personality, remains latent in everyone. If faced with a crisis, some few of us might make such a choice. But I suspect that in most men that power has become so repressed and overlaid and atrophied in middle life that the chance of its being exercised, if a crisis should face one, is negligible.

      "Existentialism," as presented by Professor Kuhn, seems to me to be an account of human nature derived from contemplating men of forceful and original character, making hard (and for themselves and those near and dear to them, at any rate) far-reaching decisions. It is certainly most important not to neglect this heroic side of human nature, and not to forget that it can and does show itself in what we might be tempted to regard as very ordinary men and women in very humdrum circumstances. But that should not make us ignore the dim and petty background (against which these cases shine forth by their rarity), summed up in the epitaph which might so fittingly commemorate most of us:--

Too bad for heaven, too good for hell;
So where he's gone I cannot tell.

      Professor Kuhn's main criticism on what l have written about human personality is that I have treated a person and his doings and sufferings as if they were exactly like a physical thing and what happens to it, and have treated voluntary action as if it were exactly like physical causation. I must admit that there is much truth in this, as regards my published works. I can, however, assure Professor Kuhn that I am not, and have never been, a "physicalist" (as I understand that word) about human nature. I regard the differences between men and any non-human animals of whom we have knowledge, as quite fundamental, however they may have arisen in the course of evolution. And I consider that causation, as it shows itself in rational cognition, deliberation, voluntary decision, and considered action, has certain unique peculiarities as contrasted with either purely physical causation or psychological causation at the non-rational level.

      The only other matter on which I will comment is this. Professor Kuhn twits me with some obiter dicta, which occur towards the end of The Mind and its Place in Nature, to the effect that the human race might possibly escape disaster by applying psychology and genetics to "deliberately altering the emotional constitution of mankind, and deliberately constructing more reasonable forms of social organisation." He asks me what I think about that now.

      My answer is as follows. It seems to me even more likely now than it did then that, unless opportunities for organised scientific research should be destroyed in the near future, the knowledge and the power will be available to determine the kind of individuals who shall be born (or incubated), and to mould their nature at will after birth. Such knowledge or power could be used on a large scale at any moment only by that person or that group who then have control in a given society. They would be used only in so far as those in control knew of them and desired to use them, and the ends for which they would in that case be used would depend on the wishes and ideals of the controllers. Given all this, the scheme would be effective only in so far as those in control could apply it on a large scale by consent or through inadvertence, or impose it by fraud or by force or by propaganda on the rest of the society.

      Plainly that would give an unprecedented power for good or for ill to those who are in a position to use it. Beyond that platitude there is little that I can say except to add the following supplementary platitudes.

1. There is little likelihood that the scientists, who had the knowledge, would be any more than the tools, or at best the willing technical advisers, of those who had the power to apply it.
2. Even if, by some strange chance, the relevant scientists should also be in effective control, that would be no guarantee that a good use would be made of the power. There is no reason to think that the ideals of psychologists and geneticists, as such, in regard to human nature and society, would be better (as distinct from more practicable) than those of trades-unionists, businessmen, lawyers, soldiers, or professional politicians. Nor is there any reason to think that psychologists and geneticists, as such, would be any less susceptible than other men to the corruptions of power.
3. I am inclined to believe that there is a rather strong negative correlation between the qualities which help a man to get and to keep power in a highly organised industrial society of the modern type (whether capitalist, socialdemocratic, or communist), and the qualities which tend to endow a man with high ideals of human personality and human society. I should therefore think it much more likely that the powers in question, if used at all, would be misused than that they would be applied to good ends.
4. On the other hand, it seems to me plainer than ever that, unless the emotional make-up of the average citizen throughout the world be profoundly modified in certain ways in the fairly near future, the chance of humanity escaping a large-scale disaster is very slender.

      Existing societies are composed of persons whose emotional reactions are largely infantile or anachronistic, i.e., adapted to situations utterly different from those with which men are now faced. They are wholly dependent for their livelihood on a complex and delicate web of economic conditions, which no individual understands. They are now brought into ever closer and more irritating contact with each other through the development of means of quick communication and the inordinate growth of population, and their emotions are continually played upon by wireless propaganda. All the conditions for an explosion are thus given. And now such persons and societies, whom a sensible parent would hesitate to trust with a popgun, are provided with atomic and hydrogen bombs, and with rockets to convey them. So there is every prospect that the explosion, when it comes, will be shatteringly destructive.

      These seem to me to be reasonably probable inferences from fairly plain empirical facts, and I do not think that their plausibility is much affected by whether one holds a "physicalist" or an "existentialist" view of the nature of human personality.