Julius Kovesi
It might seem unfair to take as my point of departure an isolated argument from Professor Perry’s Moral Reasoning and Truth, but although I make some critical comments on that passage I make them in order to continue the struggle with the difficult problems Professor Perry there presents rather than to show imperfections in it. In fact Professor Perry does mention in a footnote that my comparison of judging an object to be a table and an action to be a murder is similar to the one he presents there, so my critical comments are intended more as a further contribution to the clarification of a problem that interests us both.[1]
Professor Perry points out, I think rightly, that the appropriate analogy of a moral judgement in the world of objects is not with contingent generalisations, explanations or predictions, but with empirical classifications. ‘More concretely, statements like “That action of Jones (of which we are otherwise fully informed) is wrong” should be compared with statements like “That other object before us is a table” or “That object before us is a person”, and not with statements like “That table will still be in the room tomorrow” or “That person will never marry”…’ He wants to use this analogy to cast light on the question whether a particular action, say a particular sadistic killing, would still be what we now call wrong even if practically everyone who considered it from a moral point of view would say it is not wrong. And to do this, he first considers this imaginary case: suppose that practically everyone who is in a position to observe and use English normally, no longer calls even paradigmatic tables, ‘tables’. Would they still be tables?
If the object we pointed to had the same uses as what we now call table, then whatever other word we used in place of our present word, it would have the same meaning as our present word ‘table’, ‘for it is the uses of a thing which determine the meaning of the words we use to refer to it’. But if the object did not have the same uses then it would be a ‘useless object, or perhaps a device for warding off evil spirits, or just some curiously arranged wood we can build a fire with’.
It is in the application of this theory to moral judgements that I find some ambiguity in Perry’s argument. Coming to the question whether the sadistic killing of a human would still be wrong even if practically everyone who might consider it from the moral point of view would say that it is not wrong, this is how Perry formulates his answer: ‘Yes, it would still be what we now call wrong if the people who take the moral point of view towards things continued to have the same “uses” for it: that is, if they continue to have the same evaluative response to the sadistic killing of humans’. Although the ‘it’ refers to the sadistic killing of a human, we do not hear of its ‘uses’ at all but instead we are asked to envisage a different moral judgement of it. This is far from what Perry intends to say, rather it is almost the opposite of what the whole machinery of the comparison was meant to show; nevertheless something went wrong, for this is what he goes on to say: ‘But if they would not so characterise it with a synonym of “wrong”, but would use some value term of an entirely different meaning–one expressing mild approval, for example–then under this more radical supposition it wouldn’t still be wrong, just as this table I am now writing on wouldn’t still be a table under the corresponding radical supposition.’
I cannot blame Perry for some ambiguity here, for the problem he has attacked is very complex and subtle, and furthermore, our philosophical habits of thinking have hardly prepared us for handling it. I am not sure that I can clarify the problem, but I would like to attempt it for I think it is a very important one.
The main problem is, of course, what is parallel to what; what should be compared to what in judging physical objects and judging moral acts. But before we can map that, we should clarify an important point. The operative concept is not that of using something, but is a higher or more embracing concept, of which using is an instance in the case of tables and bicycles. We have to ask something like ‘what is the place of that something-or-other in our lives’ or ‘what is the point of having something-or-other in our lives’, or even ‘what was the point of forming the idea of and then making and maintaining certain objects’. It just so happens that the answer to such questions in the case of tables is in terms of their use.
It is strange to ask for the uses, even in a very broad sense of ‘uses’, of a sadistic killing of a human. We should look for the appropriate comparison in terms of the point of selecting certain features of human actions for special recognition, which in the case of actions is not their ‘use’.
Perry suggested that to continue to have the same ‘uses’ for it is to continue to have the same evaluative response to the sadistic killing of humans, and to change its use is to have a different response to it, perhaps one expressing mild approval. My further objection now is not only that this should not be discussed in terms of ‘uses’, nor that its ‘use ‘ in a broad sense of it is perhaps different from what Perry suggests. The whole logic of the problem and the whole point of drawing the parallel are distorted here. As Perry presents the case it looks as if the very same action, the sadistic killing of a human, could be right or wrong, everything else remaining the same. I am sure this is not what he would have liked to say, but by presenting the judging of the action to be right or wrong as the parallel to the use of the table, he has destroyed his parallel, with drastic consequences. To reverse the parallel: it is as if the use of tables consisted of judging them to be tables.
Now in the case of actions (though less so than in the case of making and doing) one must bring about some observable movements or interfere with the world in some way, but those movements or interferences are not the equivalents of observable objects that we have in the case of tables. In the case of actions we have only descriptions, or if this offends anyone, we have only specifications. So the parallel to the use of tables should be something like the reasons for selecting certain features of our lives, actions and situations for special recognition.
The elusive nature of our problem is such that it is very easy, or natural, to make two mistaken and diametrically opposite assumptions at the same time. This can happen because the comparable levels or elements of the parallel are shifted but at the same time there is a parallel, so the two mistakes are due to a sort of distorted double vision. Incidentally, I believe that it is at the heart of so much of the cross-purpose arguments about ‘description’ and ‘evaluation’, and this is why I said that our philosophical habits of thinking have hardly prepared us for handling this problem.
Since in the case of actions there is no observable object such as we have in the case of tables, we easily fall into the assumption that a wrong act continuing to play the role that wrong acts play is to be continued to be judged wrong, and at the same time to assume that since in the case of objects there is something observable even when their use and their role in our life have changed, so in the case of actions there must be as it were some residue of the action which remains the same. In the process the most important element in the comparison, the very element which the comparison was meant to bring out and spotlight, was conjured away: namely that there is something in our moral life comparable to the roles which objects play in our lives. What the parallel intended to show was that in the same way as we cannot succeed in judging a table other than a table as long as it continues to play the role tables play in our lives, so we cannot succeed in judging a wrong act other than wrong as long as that act continues to play the role wrong acts play in our lives.
The philosopher who came nearest to seeing what is happening is Locke in his analysis of mixed Modes. Moral notions for Locke are mixed Modes, and mixed Modes are made by the understanding. They are not copies of real existing things in the world but ‘are the creatures of the understanding, where they have a being as subservient to all the ends of real truth and knowledge as when they really exist’.[2] The names of mixed Modes always signify the real essences of their species. ‘For, these abstract Ideas being the Workmanship of the Mind and not referred to the real Existence of Things, there is no supposition of anything more signified by that Name, but barely that complex Idea, the Mind itself has formed…’[3] In mixed Modes the real and nominal essence is the same. The examples of mixed Modes are moral notions, and he does call them notions. There is a reason for forming these notions. ‘But though these complex Ideas, or Essences of mixed Modes, depend on the Mind, or are made by it with great liberty; yet they are not made at random.’[4] Now of course the reason why we select certain features of our lives, actions and situations into one notion and describe them by one term is because certain such configurations have moral significance. Taking away that reason is like taking away a kingpin, and the whole configuration will fall apart and the bits and pieces will take their place in other, not necessarily moral, configurations, or just remain as scattered pointless pieces. And those features of our lives that are features because of their moral significance will cease to be features at all. Here is the equivalent to tables turning into useless objects or perhaps devices for warding off evil spirits.
We see now another reason why Perry made that slight mistake in his analogy, but we can also see now a further consequence of that mistake. If their wrongness is the point of bringing together certain configurations into one notion, then it is easy to mistake this role of these notions in our lives for judging them wrong. But by conflating the two–their roles and judging them–Perry has let slip from our attention the point that the reason for judging a notion wrong is the role it plays in our lives.
Locke’s views on mixed Modes are hardly made use of in moral philosophy. Such views as these are put down to his rationalist side and his analogy with ‘Mathematicks’ makes him a bit of a curiosity in the history of moral philosophy. It even seems incredible that anyone could compare the practical affair of morals with such theoretical study.
Before we go further I just want to mention one of Locke’s examples, an example I look on as one looks on past battles or skirmishes, the different outcome of which could have changed our whole subsequent history. Here is one of Locke’s examples with which he illustrates how mixed Modes have no natural articulations corresponding to them in nature from which they could be derived. ‘What Union is there in Nature,’ asks Locke, ‘between the Idea of the Relation of a Father, with Killing, than that of a Son, or Neighbour; that those are combined into one complex Idea and thereby made the Essence of the distinct Species Parricide, whilst the other make no distinct Species at all?’[5] This is indeed the question Hume asked later, and as we know, Hume’s answer was not that it is our mind and understanding that creates such mixed Modes and creates them for good reasons. He looked for the answer in his breast and we know what consequences it had for moral philosophy, while Locke himself went on to make the unfortunate analogy with mathematics. A few chapters later he says ‘…I am bold to think that Morality is capable of Demonstration, as well as Mathematicks: Since the precise real Essence of the Things moral Words stand for may be perfectly known, and so the Congruity or Incongruity of the Things themselves be certainly discovered, in which consists perfect Knowledge’.[6] No wonder Hume treated Locke’s question on parricide in the same chapter in which he argued against Wollaston.
There is an interesting philosophical tradition where we find the constant recurrence of the partly helpful but also rather misleading analogy of our knowledge of what we have made with our knowledge of mathematics and geometry. The analogy is usually invoked to show the superiority and completeness of the maker’s knowledge. Vico, for instance, claims that God knows the world as we know mathematics, not because the world is mathematical, but because God made the world as we made mathematics, and concluded that since we made history, we should have a divine knowledge of history and adds that the historian should have divine pleasure in contemplating history. Hobbes also draws the same analogy: ‘Geometry therefore is demonstrable, for the lines and figures from which we reason are drawn and described by ourselves; and civil philosophy is demonstrable, because we make the commonwealth ourselves’.[7] Earlier still, Maimonides made use of the idea in order to show the difference between God’s complete knowledge of the universe and our incomplete understanding of it. He makes use of the example of a water clock which is known by its producer according to the principles by which he made it but which others know only by observing it as an object. This is how he contrasts God’s perfect knowledge of the world with our poor understanding of it.[8]
Interestingly we find the same idea in Lukacs’ interpretation of Marx: throughout history we make the object of our knowledge, nature and society, more and more intelligible by transforming them. Finally in the proletariat this social reality becomes completely rational and knowable, and at the same time this subject matter is the knower, so in the proletariat the knower and the known become one, as God is described in some theories.
Locke, in his theory of mixed Modes, gives, I think, without mystification and without megalomania, a very helpful account of what is involved here. In our knowledge of substances like gold or cabbages, the nominal essences never really exhaust, do not really coincide with, the real essences. However much we know them there is always a residue beyond what we know or what is knowable. But, as I said, in mixed Modes the real and the nominal essences coincide.
The claim that maker’s knowledge is divine knowledge is of course an admission that we haven’t got that knowledge; and the mathematical analogy is a very misleading attempt to indicate something which is true about our moral and social life, namely that in studying it we have to chart out and explore intricate structures of conceptual relationships. But knowing this is not like knowledge of our intentions and in fact it is even more difficult to know than the physical world. The embodiments of our intentional endeavours in our language and culture are not the making of an individual agent, and yet only individuals can know, so however much that world is in a sense our creation, the maker and the knower are not the same. This is what is partly behind the prescriptivist’s objections that on the descriptivists’ account our moral decisions are, at least in part, made for us by others. But of course it is not our decisions which are made for us by others, but what we make our decisions about, the world in which we make our decisions, the world which includes the reasons why we have to make decisions at all. As others cannot know for us, our decisions cannot be made by others, only what we know about or make decisions about can be made by others. But as some Husserlian phenomenologists fear that our knowledge is falsified as soon as an intentional object is described, some prescriptivists have similar fears about the embodiments and expressions of our decisions in moral terms. Such terms can only be a descriptive corpse without a vital element. We can see also in the views of the prescriptivists another way in which the tradition of maker’s knowledge went wrong. We meet here again the promethean temptation of a world where I make my own values, and values not made by me confront me like objects made by others: they appear to me as descriptions.
We are further handicapped in our knowledge by the moral world’s differing from that other branch of the subject matter of maker’s knowledge, from the world of artefacts. As Locke so clearly argues the mixed Modes of our moral notions, unlike complex ideas of substances, do not have references to substances existing in the world; they come to exist only when by our actions we instantiate them.
We just have not got the models and the vocabulary to chart properly this world of conceptual structures, their multiple interrelationships and the patterns of structural and other changes within it. I am not referring to problems of methodology developed and argued over by philosophers and social scientists. Our handicap is inherited from our pre-academic development and perhaps from our pre-intellectual development, from the time when we developed primarily as visual observers.
Even Wollaston who came very near to saying that by our actions we instantiate concepts and propositions which are appropriate or inappropriate in relation to other instantiated concepts, retained a correspondence theory of truth, appropriate to propositions we use in talking about the physical world. For this Hume rightly criticised him by arguing that in all his arguments ‘there is an evident reasoning in circle’. To make a false statement by my action presupposes a moral world about which my action makes the false statement. It is very difficult to express the idea that my action is not a statement about a moral world but in a sense constitutes it.
Others who borrow from the models of the visual world to express the fact that they are not talking about the visual world postulate non-natural qualities or other invisible entities. What should strike us about these is not the implausibility of such qualities or entities but that in trying to talk about our moral life they only extend the furniture of the physical world.
But the most prevalent way in which we borrow models from our dealings with the physical world is when, in order to draw the distinction between talking about the physical world and talking about our moral life we first draw it within the ways in which we talk about the physical world and then transfer that model to our moral life, without the objects in connection with which we made the distinction. We are all familiar with those introductions to moral reasoning which begin by saying that there are two sorts of things we can say about strawberries or tables: one sort is to say that they are sweet or round, the other sort is that they are good. Then, in order to have a reason to choose we have to add that I like strawberries. I have argued elsewhere that the difference between these two is not as great as that between both of these types of statements about objects, and our statements about our moral life.[9] We must not assume that out of the two ways–among many others–in which we can talk about objects one of them provides the pattern for talking about objects and the other provides the pattern for talking about our moral life.
I want to bring out by the help of an absurd sounding example the difference between the two worlds. The difference is not that in one case we say that this is a table, or that it is round, and in the other we say that it is a rather good table. To envisage our moral life we have to envisage ourselves as instantiating a table. If I want myself to be described as a table I have to be careful not to wobble, and whenever I see that the occasion demands it I would have to bend down sufficiently low so that the appropriate people could write on me or place objects on me.
Let us observe the completely different relationship between description and action on the pattern I am suggesting. On the model of choosing objects we are choosing from already existing objects and we choose the one which comes up in a sufficient degree to what the object is supposed to be under that particular description. On my pattern I am not choosing the best instance of different particulars for there are no particulars there to choose from. The connection between description and action is that I am instantiating by my behaviour the description. I choose, if I can, what description I want to instantiate, or choose actions falling under different descriptions. I choose an act not because it is the best instance of something under a certain description but because I want to instantiate that description–or I refrain from doing something because I do not want to instantiate that description.
Of course, in the slave market I might be chosen because I am a good instance among other slaves of a sturdy physical specimen. But I am being chosen as an object; the choice is not a moral choice. The moral choice is choosing to choose you under that description, and what is more, it involves treating you under that description. We can see that there is a connection not only between the description I instantiate and my behaviour but also between that and the behaviour of others: both I and the slave owner enter into a world of conceptual relationships. And now we can see that the absurd sounding example was not quite so absurd after all. The absurdity is only in our repulsion at thinking of a man as a table, to think of him and to treat him as what he is not. As Wollaston said: ‘To talk to a post, or otherwise treat it as if it was a man, would surely be reckoned an absurdity, if not distraction. Why? because this is to treat it as being what it is not. And why should not the converse be reckoned as bad; that is, to treat a man as a post…’.[10]
We can work out, for instance, as a theoretical exercise, what is implied in the notion of a friend. In analysing the concept we work out what is entailed in being a friend, what behaviour we expect from friends, with what intentions they do what they do, and so on. And when I am a friend, I have to act out in my life the implications of the concept. When someone reproaches me by saying ‘I do not mind you doing that but do not call yourself a friend’ he is pointing to the logical incompatibility of my action and my describing myself as a friend. Of course if I have reason to take on a different description I can cease to regard myself as a friend. But my description of myself needs justification just as much as the description of anything else. Moreover, changing it is itself an action, just as much as resigning, marrying or divorcing or joining the resistance movement are actions.
I am not giving in this paper an account of moral obligation, nevertheless I want to mention some reasons why we cannot lightly take on or abandon descriptions, reasons that are connected with my limited objective in this paper.
One is the interrelatedness of whole webs of connection in such a way that I have to consider a whole set of consequences–and I do not mean utilitarian but logical consequences–in renouncing or taking on a description. During the height of student revolutions some years ago a student handed in to me a sheet of paper at the time when essays were due. On the paper was a drawing of a steam engine with a flower in its funnel and with big bold letters the inscription: ‘I expressed my freedom but you academics do not care about it.’ With my limited artistic ability I drew a little pussycat in the corner of that paper, the sort children draw, with whiskers, and handed it back to the student next day. He came to me fuming that I had penalised him, that he was right to say that I did not care about his freedom to express himself. I did my best to look puzzled and asked him in what way he was penalised. ‘You did not give me a mark’. — ‘I cannot mark your expression of freedom. If you expected a mark then perhaps what you submitted was an essay’. ‘You just want to treat me as a student, you don’t treat me as a human being’. ‘In that case my giving you a mark is even more incomprehensible. I cannot mark you as a human being. But if I treated you are a human being I would certainly ask you not to intrude on me and not to take up my time. For as a human being you are disagreeable, and I don’t want to talk to you. But in so far as you are a student I would have to give you a distinction if you produced good work. In this case it is not because I do not like you that I did not give you a mark but because I took your paper as an expression of freedom and not as an essay. So instead of a mark I just expressed my freedom and do not know why you are so upset about my expression of my freedom.’
This was regarded as bourgeois formalism. I don’t think that young man had worked out the implications of opting out of certain descriptions. Notions like relevance, fairness, corruption and a whole host of vital moral notions come to life and into prominence as relevant concepts logically connected even with such a simple case as this.
Another type of consideration that one might have for not wanting to get rid of a description is the recognition that the aspects of our life that it carves out and brings together into what Locke calls a mixed Mode, and other aspects that it leaves out and sharply distinguishes itself from are important and worthwhile distinctions. To use again my example of a student, we have to realise what conceptual transformations were needed to separate out and to create the notion of student in distinction to the notion of disciple. In marking the essay of a disciple it would be relevant to consider whether he is reproducing my thought and whether he is furthering the cause I am dedicated to. In considering the value of having such a description as being a student again we have to trace and chart a whole set of relationships with other concepts with which our concept is connected and of which it forms a part. But without such a description one might not be able to be a student at all.
This illustrates also the reverse relationship between description and action. Very often when people are told that moral distinctions are man-made they are not only shocked but think that moral distinctions cannot be very important–if they are only man-made. But this is all the more reason for their importance and we cannot take it for granted that they will exist by nature. If we value some concepts we have to exemplify them and instantiate them by our actions in order to keep them alive. I do not mean ‘exemplify them’ in the sense that by our example like a living advertisement we shall propagate them. I mean that only by us being instances of it can the concept stay alive for this is its mode of existence.
I think, incidentally, that Locke could have dealt with the connection between the conceptual relationships of moral notions and our actions in a similar fashion. ‘If it be true in Speculation, ie. in Idea, that Murther deserves Death, it will also be true in Reality of any Action that exists conformable to that Idea of Murther‘. [11]
Someone might object that this shows how wrong Locke was, and say that most people would dispute that murder deserves death. The point is, however, that we do work out ‘in Speculation’ whether murder deserves death and if we come to the conclusion that there is no such implication then ‘it will be true in Reality of any Action that exists conformable to that Idea of Murther’.
But the problem still remains, even supposing for the sake of argument that murder deserves death, why should I doanything about it. This would depend on my description, or on what idea I instantiate, and how I am related under the appropriate description to the rest of the situation which consists of a configuration of other appropriate descriptions. If it be true in speculation that judges should sentence criminals and I am the judge then I should be doing the sentencing. If I am the murderer then I should think of myself as deserving death. But however much murderers deserve death, if I am not the executioner and he has not yet been condemned, then if I killed him I would be described as someone who took the law into his own hands, and that has yet further implications. If, however, there is no appropriate legal system in the land and I am the murdered person’s brother, I could not be described as taking the law into my own hands, there being no laws that I could be supposed to take into my own hands. This last example also shows how an act of revenge could change its role, which would help to illustrate Professor Perry’s discussion of actions which could play different roles in our lives.
If, on the other hand, I am not in a situation, that is, if I am not related conceptually to the other concepts in question, all I might be expected to do is merely to make a moral judgment. Versions of the fact/value distinction which locate the distinction in the gap between describing something and doing something, suffer from a systematic confusion because we have not worked out the difference between moral judgments and moral decisions. Let me just say here that the occasions for making judgements, and along with them the reasons for making judgements are quite different from the occasions and reasons for making decisions. I am making a moral decision when I have to make one, and I have to make one when I am in a situation. I am in situation when I lock in, under a certain description, to an interlocking conceptual field, as one of the elements of that interlocking conceptual field.
So the connection between a description and an action is not like the connection between describing something and judging it a good object of that sort; nor is it like the connection between describing it and choosing it, when the connection has to be supplied by our liking it or wanting it or making a decision about it. In a situation I have to choose because I have to choose, that is what puts me in a situation. And my reasons for choosing are in the descriptions under which the alternatives present themselves.[12]
[1] Thomas D. Perry, Moral Reasoning and Truth (Oxford: Clarendon Press, 1976), 132-134.
[2] Essay, III, Ch.V., 5.
[3] Ibid, III, Ch.V., 14.
[4] Ibid, III, Ch.V., 7.
[5] Ibid, III, Ch.V., 6.
[6] Ibid, III, Ch.XI., 16.
[7] English Works (London: Bohn, 1839-45), Vol.7, 184.
[8] Maimonides, The Guide of the Perplexed (Chicago: University of Chicago Press, 1963), Part III, Ch.21, 484.
[9] ‘Against the ritual of “is” and “ought”,’ Midwest Studies in Philosophy, Vol.III, 1978, pp.10-16, above pages 30-51, and ‘Valuing and evaluating’, in Jowett Papers, ed. by B.Y. Khanbhai, Blackwell 1970, above pages 16-29.
[10] D.D. Raphael, British Moralists (Oxford: Clarendon Press, 1991), Vol.I, 245.
[11] Essay, IV, Ch.4., 8.
[12] I worked on versions of this paper while I was Visiting Fellow at the Research School of the Australian National University. I would like to thank the University for its hospitality and especially Stanley Benn for his persistent helpful criticism.