Julius Kovesi
However much the preoccupations and problems of moral philosophy have changed in the last decade or so, we retain, with a ritual observance, a basic conceptual framework. Apart from a few bold spirits who disregard the ritual, most moral philosophers, before they can say anything, have to re-enact the moves of trying to justify how they dare to move from description to evaluation, while others, opposing them, claim that they have disregarded sacred texts and violated the most sacred of ritual moves.
Some, and I would like to count myself among these, would like to argue that the whole ritual is unnecessary, misleading, confused and confusing and even detrimental to moral philosophy. It is most difficult to argue for this, because, as I said in a brief postscript to Moral Notions, in our arguments we have to make use of a terminology which is not neutral but embodies the very theory one is arguing against. But added to this conceptual handicap is the probability that the effort of replacing this terminology with a more profitable one will be misunderstood as a subtle way of moving from ‘is’ to ‘ought.’ Thus the effort of replacing a framework is thought to be a move within the framework.
In my Moral Notions I was using, as I said, the terminology of contemporary moral philosophy as Wittgensteinian ladders that I was trying to throw away surreptitiously on the way, but some of them I had to use right to the end. Only when I finished that study did I realize that the whole of it should be thrown away in order to start doing moral philosophy properly. I still think it is good as a moral philosophy game but otherwise I think of it as Marx and Engels thought of theirGerman Ideology: the work in which they settled their accounts with the Young Hegelians and with their own former philosophical consciousness, but otherwise did not want to publish any more. (Only, without immodesty, I do think that my book is not so thoroughly bad as theirs.)
John Searle left the ladder–the throwing away of which was the main point of one of his articles–ostentatiously in his title when he called it ‘How to Derive an “Ought” from an “Is”.’[1] This is some excuse for reading his article as an attempt at ‘a feat which many before have thought to perform,’ that is, deriving a moral judgment from statements of empirical fact.[2] I am not taking such chances with the title of my paper and I want to make it clear that this is not another attempt to derive an ‘ought’ from an ‘is.’
There are many reasons for wanting to do away with the whole ritual. In this paper I shall try to give three such reasons. First, I want to indicate that the dichotomy is a thoroughly amorphous family of problems. Secondly, I shall indicate that the dichotomy is not the result of a neutral philosophical analysis but is an ideologically motivated doctrine. My main concern will be the third reason, which is that our preoccupation with this ritual completely distorts what moral philosophy is about and misdirects our attention as to what constitutes moral life.
THE AMORPHOUS NATURE OF THE DICHOTOMY
Hume and G.E. Moore are thought to be the two who pointed to the dichotomy most succinctly, though others before them were thought to have seen ‘it’ before. The fact that one talked about deducing an ‘ought’–in some sense of ‘deduce’–and the other about defining ‘good’–in some sense of ‘define’–does not seem to worry many people who invoke them indiscriminately in support of a set of modern theories that neither of them had in mind. The phrases ‘as Hume has pointed out’ and ‘as G.E. Moore has pointed out’ are used almost interchangeably. In fact the only similarity between Hume and G.E. Moore is that both of them thought that certain value terms are like color words, but there the similarity ends, because they thought they were like color words for diametrically opposed reasons. Hume thought that virtue and vice are like colours, sounds, heat and cold, which according to ‘modern philosophy’ are not part of the furniture of the world but are our contributions to the description of what the world is like, and G.E. Moore thought that his little indefinable good is like yellow because, like other innumerable indefinable simple objects, it is part of the basic furniture of the world. If these views are put back into their context then they are even further away from anything said by contemporary moral philosophers whose views acquired initial prestige by invoking the names of Hume and Moore. One cannot see how either of them are even dimly related to views such as: only a command can be an answer to the question ‘What shall I do?’ or that our moral life should be regulated by nothing else but universal imperatives which by definition are made to have the force of a command addressed to oneself. True, these views are not presented as something that either Hume or Moore have actually said, only as the only solution to the otherwise inescapable problem they presented to moral philosophy. But what solution is offered indicates what the problem was thought to be. Most of us can offer half a dozen different versions of what the naturalistic fallacy or an evaluative/descriptive dichotomy is supposed to be, and these are not merely different formulations, but drastically different versions.
What is detrimental in all this to philosophical reasoning is that when one argument for the supposed dichotomy is discredited or disputed, then it can be countered that the argument was directed only against a misunderstanding of the dichotomy, and so on over and over again about each and every argument. And when all the reformulations of the dichotomy have been discredited or disputed, it can still be assumed that there is an unassailable dichotomy, and only inadequate formulations have been picked on by philosophers who can even be blamed for misunderstanding the dichotomy. In this way the dichotomy can take on an independent existence, not depending on any one particular argument. I am not saying, of course, that it is supported by the cumulative force of all the arguments rather than by any single one of them, rather that it is sustained by the assumption, each time an argument fails, that there is, indeed there must be, another argument for it. This is largely responsible not only for the continued deferential treatment accorded to the dichotomy, but also for the assumption that its origin is to be found in Hume and Moore. This move has been applied to them in the first place. ‘There is indeed’ wrote Hare in his Language of Morals ‘something about the way in which, and the purposes for which, we use the word ‘good’ which makes it impossible to hold the sort of position which Moore was attacking, although Moore did not see clearly what this something was.’[3] I am suggesting that ‘this something’ has a mystique like the emperor’s new clothes.
THE IDEOLOGICAL NATURE OF THE DICHOTOMY
That the problem is so amorphous is partly due to the ideological nature of the dichotomy. It is characteristic of ideological beliefs that their truth is upheld independent of the arguments for them, and the arguments are looked for and produced in support of beliefs already held independently of arguments, and for other reasons. When it was found, for instance, that the proletariat did not bring about the renewal of creation, this was not regarded as confounding predictions of the imminent renewal of our condition, but another agency was looked for to bring it about. Though the original claim was made because of the nature of the proletariat, the claim itself acquired such an independent existence that by now some can allege that those who made the original claim did not quite understand what would bring about that transformation when they said that it would be the proletariat. There are many other examples in other religious or political theories of a claim becoming detached from the origin which gave it its first impetus and acquiring a life and force of its own.
Another feature of ideological thinking is that the believer creates a position for his ideological opponents within his conceptual framework. Often his opponents, by trying to argue against him, take up that prearranged position, thereby accepting his system by occupying a place within it. It is sad to think how many moral philosophers took up the role of being naturalists because they were trying to argue against some theory of an anti-naturalist; how many of them tried to derive value judgments from brute empirical facts because they wanted to argue against the theory, say, that we make our values by our decisions.
But I want to say more than that the arguments about the dichotomy exhibit some of the features of an ideological belief. The dichotomy is animated by ideological beliefs.
Professor Hare at the end of his article in reply to Searle’s, offers his own answer to the question why one ought to keep promises, and ends the article by saying that his answer ‘needs no “is”-“ought” derivations to support it–derivations whose validity will be believed in only by those who have ruled out a priori any questioning of the existing institutions on whose rules they are based.’[4]
One would fall into the ideological trap either if one took up the role assigned by Hare for those who wanted to derive an ‘ought’ from an ‘is,’ or if one tried desperately to protest that one did not wish to be such an authority-oriented personality. But I cannot forbear adding to these reasonings an observation which may, perhaps, be found of some importance, which is that the author proceeds for some time in the ordinary way of reasoning, quotes the O.E.D. and makes use of some morally neutral rules of logic, when all of a sudden one is surprised to be saddled with a particular moral and political position.
But to my mind one of the most revealing and instructive articles that has appeared about the evaluative dichotomy is Montefiore’s reply to Philippa Foot’s paper on ‘Goodness and Choice.’[5] Montefiore presents there the doctrine of what he called an ‘individualist,’ whose freedom and very existence as an individualist is threatened by the fact that our language provides us in most cases with criteria of evaluation. What his individualist has to do in order to have his ideal existence is to eliminate all ‘criteria-setting’ terms from our language, such as ‘knife’ or ‘pen,’ and operate only with the category of ‘object.’ Montefiore considers the possibility that even to categorize something as an object already carries with it ‘some peculiar even if as yet unknown function.’ Though he rejects this possibility, he thinks that if it had been so ‘we should clearly have had the greatest difficulty in ever conceiving the individualistic project of stating all the facts about pens in some overtly non-functional, non-evaluative way.’[6]
The individualist however has another problem as well. ‘The category of “object” cannot be pressed into service for the defunctionalization of all functional or similar terms.’ ‘Farmer,’ ‘trap,’ ‘writing,’ ‘puzzle,’ ‘appeal’ are his examples that cannot easily be categorized as objects. In such cases the individualist’s solution is that of Hare’s horse, which, I think, should be as well known by moral philosophers as G.E. Moore’s horse. Hare’s horse refuses the description of himself as a ‘charger’ in case he may fail to be a good charger, and redescribes himself as ‘a solid-hoofed perissodactyl quadruped, having a flowing mane and tail.’[7] As long as the individualist can redescribe himself, his action, or the objects he wants to choose, he can retain his freedom. While this escape route is open his obligations are only hypothetical. But if he cannot redescribe his objects or actions his obligation is categorical. The strangeness of this view will become clearer when, towards the end of this paper, I shall outline how moral obligations are different from evaluations. Philosophers who would like to derive our obligations from the type of considerations which show that terms like ‘pen’ or ‘farmer’ provide us with criteria of good pens and good farmers, would probably very much like to know how Montefiore thinks that categorical obligations follow from having such words in our language as ‘pen’ and ‘knife.’ The third section of my paper will argue that the problems of moral life are quote unrelated to the humdrum affair of evaluations.
But more is at stake for the individualist than his freedom. The very possibility of evaluation seems to be at stake if we have ‘criteria-setting’ terms, for Montefiore claims that ‘he alone evaluates who creates or chooses his values for himself.’[8] It is very difficult to make sense of this claim, but I think I see why he says something like this; he tries to make the individualist’s position secure by this attempted definition. In the paragraph to which this is a conclusion, he asks in what sense of ‘must’ one must not try to force the individual to accept values that he does not freely accept for himself. The reason for it cannot be that it would be wrong to do so because this would refer to ‘standards of extra-individual value.’ The only answer is that it is impossible to do so and thus he tried to make his individualist’s position follow from the very definition of what it is ‘to evaluate.’ I should quote at length what Montefiore has to say a few pages earlier:
To face him [the individualist] with values that were given to him as facts would be to restrict his freedom on the issues that were most important to him; it would be self-defeatingly to concede that what is of supreme importance is in some instances at least after all above and beyond the creative control of individuals as such. It is for this reason that a fully rigorous version of the doctrine of no ‘ought’ from an ‘is’ becomes an essential feature of any such thorough-going individualism. Its acceptance is the only guarantee that there can be no observations of fact by which the individual valuer might be committed one way rather than another…. When the individualist incorporates it as a principle of the very logic of his language he rules out any would-be non-individualist system of values as strictly unintelligible as such.[9]
If the ‘descriptive words’ that an anti-naturalist claims to be irrelevant to evaluations were the ‘descriptive words’ of our ordinary language, then the individualist would not need to worry about our description of the world because it would be irrelevant for his evaluations. Montefiore does not only agree that such words as ‘pen’ and ‘farmer’ are ‘criteria-setting’ terms without any decisions of ours; he fundamentally assumes that the description of the world is of the most vital relevance for our moral life, and this is why his individualist desperately wants to change the description of the world. He does not want to leave the world as it is: he is trying to derive an ‘is’ from an ‘ought.’ He does not choose or create values irrespective of the facts, he eliminates the facts. Montefiore even conflates facts and values when what he is concerned about is that his individualist will be faced with ‘values that were given to him as facts.’
Now there are many things wrong with Montefiore’s individualist, but again one should not, by way of criticism, try to defend an ‘authoritarian’ or an ‘other-directed personality,’ nor should one anxiously try to show that one is just as good an individualist as he is. It would be easy to show that an individualist could not exist in a world where we operated only with the category of ‘object’ (nor could anyone else for that matter), not only because he has to eat before he could be an individualist, a point to which I shall return later, but because an individualist can exist only in societies that have attained a certain level of conceptual development. Indeed, I am not sure that Montefiore could have made his point without describing the person he is talking about as an individualist. A role, however, that he cannot avoid is that of a chooser, and an analysis of what would make him a good chooser would be very instructive. In a criticism one should also show that the way of life the individualist is recommending is incoherent, or that if it came true, his aim would be defeated because in a world where everyone operated with the category of object, he could not show how different he is from the common run of people who have pens and knives.
The nature of man that is presupposed by the various anti-naturalist theories deserves a special study of its own. Here I shall say a few words, not about Montefiore’s individualist, to whom I shall return briefly in the next section, but about Hare’s individual whose life is governed by universal principles that have the force of commands.
Though one usually associates Hare’s ethical theory with the closing paragraph of Part I, section I of Book III of Hume’s Treatise, far more helpful passages for understanding why Hare’s man would respond only to commands are some of those where Hume distinguishes between passions and reason, and claims that ‘reason is perfectly inert, and can never either prevent or produce any action or affection.’ Because Hare replaced Hume’s passions with imperatives and put them not into one’s breast but into major premises, we tend to mistake a psychological theory for a logical one. Of course Hare would be the first to deny that his is a psychological theory; this is indeed why he replaced Hume’s passions with imperatives. But what I am saying is that it is Hume’s passions that he has replaced with his imperatives, assigning to them the role of moving men to act, while leaving reason ‘perfectly inert.’
The imperative is then made to move us by definition. ‘I propose to say that the test, whether someone is using the judgement “I ought to do X” as a value-judgement or not is, “Does he or does he not recognize that if he assents to the judgement, he must also assent to the command ‘Let me do X’?” Thus I am not here claiming to prove anything substantial about the way in which we use language….’[10] But now one cannot give reasons for acting on one imperative rather than on another, or for not acting on some of them, because reason is inert. An ‘active principle’ cannot be opposed by an ‘inactive principle.’ But if we universalize whatever we want to act on then it will entail not only ‘let me do X’ but also ‘let X be done to me,’ and this is something I might not want. There could indeed be no reason why I should not prefer the destruction of the whole universe to the scratching of my little finger, except that the destruction of the whole universe would entail the destruction of my little finger, which I very much want to scratch. Not even this is a reason for not wanting the destruction of the universe but here an active principle opposes another active principle.
It is strange to think that views about the questioning of institutions, about the individualist’s way of life, about the role of reason in human life, are fought over in terms of good strawberries, good knives and good farmers. Neither the anti-naturalists, nor the anti-anti-naturalists (who are not necessarily naturalists) leave alone the arguments about good pens and knives as irrelevant to morals. The anti-anti-naturalists too, know that there is more at stake in these arguments than the fear that someone might decide that a blunt knife is good for cutting. It is the desire to show that it is possible to give reasons for value judgments, and that values are not as random as they apparently fear that others might think, that keeps some of them arguing about good knives however otherwise unrelated such arguments are to the problems of morals. The arguments about evaluating artefacts are like elaborate tournaments with set rules fought in place of and away from the real battles.
THE DIFFERENCE BETWEEN EVALUATING AND MORAL LIFE
My main grudge about the descriptive-evaluative dichotomy is that it takes our attention away from the subject matter of moral philosophy to areas which might not even be connected with it, or if they are connected, it is only by convention that they are. To argue against a well established convention is hard, but nevertheless I shall try to do this.
The convention is that moral life falls on the evaluative side of our dichotomy, and this looks plausible enough if we place brute facts on one side and values on the other, and if there is nowhere else to put them. But I want to argue that there might be an even greater difference between evaluation and the problems of morals–not just a gap to be bridged but a complete difference–than there is usually thought to be between description and evaluation. The problems of moral life do not constitute a subsection of the evaluative side of the descriptive-evaluative dichotomy. I want to come back here to a theme that I dimly came to see towards the end of my Moral Notions, which is that no light is thrown on the problems of values, nor, which is yet a further problem, on the problem of obligations, by considering what we do when we evaluate.
I want to emphasize that moral judgments are not a subsection of evaluations because the assumption that they are has led philosophers to look in the wrong place to find what makes a judgment a moral judgment, and to attribute as a chief characteristic of moral judgments that which is not their characteristic at all. Of course we all know that moral judgments are different from evaluations of artefacts and skills. But if we assume that they are alike, only different in some vital respect, then we will think that we have found in that vital respect a defining characteristic of moral judgments.
The activity of evaluation very often needs tremendous skill and knowledge, sensitivity and discernment, but its logic is simple. The whole secret of evaluation is the basic fact about our world and life that whatever we find interesting or important enough to bring under a term or description is exemplified by many instances, none of them fully embodying or exemplifying that for the sake of which we took an interest in bringing them under a term or description. There are not many, if any, objects in the world like the standard metre in Paris. The reason for this is not, as the Neo-Platonists thought, that there is some recalcitrance in matter which resists the reception or embodiment of forms. It is not that the potentialities of nature are aiming at the fulfilment of various forms without ever quite achieving those aims. It is we, human beings, who have aims, intentions, needs and purposes. Our aims and purposeful activities are prior to evaluation, and the activity of evaluation could be said to be prior to our language, except for the fact that we could have only very limited aims and purposeful activities without language. When we want something for some purpose we look around to see what would do the job. So we start looking for things that would be good for our purpose. There is nothing mysterious in the fact that the objects the caveman found for the purpose of cutting or for the purpose of piercing did these jobs only more or less efficiently. Ever since, even the specially-made objects only more or less fulfil their functions.
Montefiore’s individualist would have to eliminate not merely the terms of our language; he would have to cease to perform any activities that we have to choose or to perform. If he wanted to catch animals for his food, he would have to start looking around for implements that would help him in this. Though he would have as yet no words for knives, projectiles, or spears, nor would there be any ready made, he would nevertheless have to choose objects that would do the jobs better than others. And when he lost or broke one of the implements he used for cutting, in looking around for another one, he would not look at the broken one as an exemplar, but, like Plato’s carpenter, he would think of that which does the cutting and would look for whatever would exemplify it or embody it. He would think, without the word ‘cutter,’ that one object was better than another for …, and then sooner or later he would have the terms ‘cutting’ and ‘cutter.’ It is in this sense that evaluation is prior to our language.
Even when we have the terms that incorporate our aims and purposes in performing an activity, the term we use to identify something is only one piece in the jigsaw puzzle of an activity or situation, all of which we have to understand in order to be able to evaluate. Furthermore, the term itself is not always fully developed from the point of view of the function of the activity, or purpose, or aim for which we engage in the activity, or use the object which is identified by the term. Nor can we change the term too easily to include our aims and purposes fully, for there are many other demands on our terms at the same time. Answers to tax declarations or statistics, as well as past history and technology, will influence the development of such terms as ‘professional,’ ‘housekeeper’ or ‘coin.’ Very often there is not even one single term that we can use to refer to an object or activity that we want to evaluate. These are the occasions on which we have to use the various good phrases such as ‘good at …,’ ‘good for,’ or ‘good as…’ followed by an explanation in place of one term. We also use these phrases when we want to evaluate something under a description other than the one by the help of which we identified what we want to evaluate. I say this not so much to plug the loopholes where someone might claim that a term by itself does not provide us with the criteria of evaluation, but to indicate that our activities are prior to evaluation and, indeed, to language.
The individualist may say however that all he wants is not to be compelled to do the things we normally do, he just wants to be free to do his own thing. But he will still be confronted by the fact that whatever he wants to choose will be exemplified by a number of objects or in a number of ways and, whatever he wants to do, he will be able to do in a number of ways. He will have to cease aiming at all, wanting to do anything at all, in order to escape standards or criteria of evaluation. To fully explore this problem we should investigate the nature of man presupposed by Montefiore’s individualist, a task at which I hinted earlier. My aim is far more limited here. Let us just remember that the individualist was worried that ‘to categorize something as an object carries with it the assumption that it must have some peculiar even as yet unknown function.’[11] The assumption here seems to be that by contemplating the notion of ‘object’ one might or might not excogitate a function. It is at times like this that one can see the poverty of the type of linguistic philosophy which ignores the fact that language is part of human activity. It is not the case that if the O.E.D. tells us what an object is for then we just quote what others decided to use it for, and if it does not tell us what an object is for then we are free to create our own criteria. Even if I know the meaning of a clearly functional term like ‘fishing rod,’ I do not know how to evaluate fishing rods if I never went fishing or learnt about that activity, and even if as yet we do not have a term like ‘fishing rod’ in our language, we will still have to choose one rod rather than another as better for the activity of fishing. As I said earlier, though the logic of evaluation is a simple affair, the activity of evaluation often needs great skill and knowledge, sensitivity and discernment.
We do not enter the world at the level of brute objects. We enter the world as intending people and make use of brute objects. It is for this reason that we do not ‘move’ from ‘facts’ to ‘values’ or ‘derive’ them from ‘facts.’ By the time we have described the world it is not ‘brute.’ For whatever was good for cutting before we had the term ‘cutter’ can now be said to be ‘a good cutter’ and whatever is not good for cutting cannot even be described as a cutter. I shall come back to this briefly when, discussing our moral life, I shall want to say that in the same way as I do not first find that I can shout ‘help’ and then decide to shout for help, but find that I need help and make use of shouting to get it, we do not first keep uttering sounds like ‘I promise to pay you’ and then make a decision about keeping promises, but first find the need for the institution of promising and then make use of words to make promises.
We are now in a better position to see why it is often argued that the essential feature of moral judgments is that in the case of moral judgments we decide on the criteria of goodness. It rests on the following simple blunder. It is characteristic of evaluation that we evaluate according to criteria of goodness. Moral judgments are (thought to be) evaluations, so the characteristic that moral judgments share with the genus is that they are evaluations according to criteria of goodness. It is an established fact that the family of terms that one might call functional terms provides the criteria of goodness. But moral judgments are not like judgments of functional goodness. Therefore the characteristic feature of moral judgments is that in their case we have to provide the criteria of goodness.
The idea that moral judgments are not evaluations does not even occur. When we make a moral judgment we do not evaluate according to criteria of goodness, therefore in the case of moral judgments we do not provide the criteria of goodness.
This is how the tournament about evaluation is being fought away from the battles of moral problems. It is thought that this tournament can decide the fate of the battle. The anti-naturalists think that they have to demonstrate that even in the field of evaluation the criteria are not always given, and then claim that moral judgments are like that. The anti-anti-naturalists want to show that in the field of evaluation the criteria are given, and then claim that moral judgments are like that. The battle has to be fought out in this substitute tournament because it cannot be fought over examples from moral life–because in moral life we do not evaluate.
When in the previous paragraphs I tried to indicate that the lack of criteria given by a term does not mean that there are no criteria, only that they are not given by a term, I did not do this in order to show that in our moral life the criteria of evaluation are given because the criteria are always given. I did it partly to indicate what sort of activity evaluation is, in order to contrast it briefly with moral judgements and partly to show that the anti-naturalists would lose even in the tournament.
Just one more example of evaluation might be instructive. In answer to Geach, Professor Hare gives ‘a good sunset’ as his example to show that here is a case where the ‘standard is not even hinted at in the meaning of “sunset,” let alone in that of “good”.’[12]
Now where we do not evaluate there we do not evaluate. I do not think we have occasion for waiting for the sunset, day after day, in expectation that the next will be a good one. There are, however, beautiful sunsets and Hare in fact gives us a description of a beautiful sunset: ‘it has to be bright but not dazzling, and cover a wide area of sky with varied and intense colours, etc.’
Let us observe that an interesting feature of judging something beautiful, as against judging something good, is that in order to know in what manner something could be beautiful one has to be familiar with the appearance of the actual physical object, rather than what the object is supposed to be. Take a leaf for example. Knowing what a leaf is would enable an expert to say what a healthy leaf is (the nearest to judging a leaf a good one). But knowing what a leaf is does not enable us to know what shape a beautiful leaf would have: for that one would need to know how leaves do and can look. Then with discernment we would be able to choose one particular leaf as more beautiful than another. Such are the examples that are expected to help us solve problems of moral philosophy, but this example does not even show what it was meant to show about evaluation.
But now when we are confronted with a decision to be honest and perhaps suffer for it, or to tell a lie and get someone else into trouble, we are not confronted with several acts all of which are supposed to be honest but some more so than others. The choice before us is not between several honest acts some of which are better than others insofar as they are honest. I am not suggesting that we demand that one had better be honest or dishonest and there should be no half measure of just trying to be honest to some extent. I am suggesting that we are not presented with the type of choice with which we are when we are evaluating. Although in some manuals of moral life we might find several examples of, say, brave acts, and the manual might even suggest that one was braver than the other (though probably it will turn out to be a better example of a brave act, or perhaps a more difficult brave act but not a braver one); when we have occasion to be brave, we are not choosing from possible brave acts some of which are braver than others. Nor is it the case that honesty and dishonesty are instances of something of which one is a better instance than the other. We do not choose to be honest because honesty is a better instance of that of which dishonesty is a less perfect instance.
In our moral life we are interested in the description under which our action falls and the description of the situations in which and because of which we are confronted with choices. We are not interested in the qualities that make one of several things falling under the same description better than the others, but in the relevant facts that justify us in regarding our acts as falling under one rather than another description.
Compared to the problems of moral philosophy the logic of evaluation is such a humdrum affair that it is not worth the attention of moral philosophers. As we know by now, we evaluate particular objects or performances insofar as they fall under a description, when our knowledge of what the objects or performances are supposed to be, indicates or provides us with the criteria of evaluation, a knowledge gained either by our experience or, if we are armchair evaluators, through good dictionaries. But there can be outrageous things and performances, or such that we would rather not have them, like poisons, lies, murders and tortures. And yet there might be people who aim at choosing a good poison if they want to murder well. We all know that a good poison for a good murder should not only be efficient but should not leave traces for easy detection, a quality not indicated by the meaning of the word ‘poison.’ Neither the activity nor the logic of evaluation gives us any clue or help as to whether we value or detest something, whether we ought to choose or reject a course of action. For the moral problem is whether to lie or not lie, and the philosophical problem is what constitutes a lie as against saying what is not the case, and why we should make such distinctions, and the problem is not the problem of a good liar who wants to lie well. Evaluation is quite neutral to morals. We can and do evaluate both what we value and what we detest.
Moral notions and judgments are not about objects ‘out there’ which we can both identify with a ‘descriptive’ phrase andevaluate as a good one of that sort, nor about ourselves insofar as we too can be evaluated along the lines of evaluating ‘things out there.’ Moral notions and judgments are about our life insofar as our life is constituted by these very notions, judgments, concepts and descriptions. Perhaps this is the place to say something about a paradoxical remark I made at the end of Moral Notions: ‘Moral notions do not evaluate the world of description; we evaluate that world by the help of descriptive notions. Moral notions describe the world of evaluation.’ By the ‘world of description’ I mean the world other than the world of our moral life, the world of tables, knives and farmers. It is not the case that while descriptive terms describe this world, with our moral notions we do ‘the other thing,’ we evaluate it. No, we evaluate that world with our descriptive terms because I can describe that world with such terms insofar as things and people in that world are supposed to come up to some standards and expectations. When we turn to our moral life we do not find that there is some raw material there waiting to be described and evaluated. For the sake of the paradox it was proper to say that moral notions describe the ‘world of evaluation,’ that is, the world of our moral life. But this should not suggest that there was something there waiting to be described. Our moral notions constitute that world, and without our moral life and notions there is nothing there to be described or evaluated. Montefiore’s individualist would find it indeed impossible to reduce the description of our moral life to the category of ‘object.’
To follow up this suggestion would not lead us outside the empiricist tradition of British moral philosophy. I cannot do more here than to suggest that we would be well advised to pay more attention to some of Locke’s ideas on what constitutes our moral life.[13] For after exploring our knowledge of nominal essences we find that there is a residue of substance which might correspond to Montefiore’s object. But the mixed modes of our moral notions do not have such residue and ideally they are transparent because they are of our making.
It would be tempting to think that we could reduce moral descriptions if not to objects, then to something akin to objects, to movements, sound waves and similar happenings. ‘Promising’ could be reduced to ‘saying something’ and that eventually to sound waves. But the genus of ‘promise’ is not ‘saying something.’ We do not choose or pick out promises from the family of performances of ‘saying something’ as we choose or pick out cutters from a certain family of objects, or chargers from the family of horses. This is even more obvious in the case of ‘appeal,’ an example mentioned by Montefiore. In whatever senses of the word ‘appeal,’ if I want to appeal or make an appeal I would have to do something, most often I would have to put it in certain words. But an appeal is not a sub-class of the manner in which I appeal.
But why should the individualist or anyone else accept that what he made was a promise and not just saying something? Why should he be described in certain cases as a liar and not as someone who said something? Why should he accept that he stole a watch and not merely removed a round object that made the noise ‘tick-tock’?
I would like to distinguish now four different types of redescriptions: the evaluative, the ideological, the diluting, and the substantiated redescription.
Evaluative redescription should be familiar to us by now. This is when we employ the various ‘good-phrases’ to shift an object from under one description to under another description. Though we do not evaluate pebbles as pebbles we might find that some of them are good for paving, that is, by the phrase ‘good for’ we shift them under the description ‘paving material’ and we say they are good paving material. Or an object which is not good as a knife might be good as a screwdriver: the phrase ‘good as’ shifts the object from under one description to under another. The individualist could use even a good knife as a decoration around his neck if he thought it might be good for his purpose which might be to épater le bourgeois. But in all these cases the individualist would not be happy yet because he would have to be observing objectively given criteria in working out whether something would be good for one thing rather than for another. Not everything around one’s neck would show that you are different from others rather than that you have bad taste or that you are ignorant. Perhaps what the individualist wants is to redescribe his activity in order to avoid appearing bad at something merely by the description attributed to his action. Thus, if someone else assumes that he is trying to cut down a tree and he is doing it rather badly, he might want to be free to claim that he is only patting it.
Ideological redescription is a very important problem for moral philosophy, but here I won’t say anything about it except to contrast it with other redescriptions. When the Nazis redescribed their murder as ‘final solution’ it was not because they were not good at murder but better at final solutions. Though the other two redescriptions also differ from the evaluative redescription, the ideological differs from them in involving a distorted conceptual framework, an ideology or a mythology, the nature of which cannot concern us here.
Examples of diluting redescriptions are redescribing ‘lying’ as ‘saying something’ or ‘murder’ as ‘killing.’ What happens in diluting redescriptions is that an attempt is made to deny or ignore a morally relevant fact which would distinguish the act from a morally neutral family of acts. Again, such attempts are not made because someone might not be good at lying but better at saying something, or not a good murderer but rather good at killing.
Both Hare and Montefiore talk about redescription in the context of evaluation, and they seem to want the freedom to redescribe a thing or an action in order to get rid of the criteria that enable us to evaluate the particulars falling under a certain description; but in fact all their examples are of the diluting type, and one doesn’t know what they would say to the ideological redescriber. But let us turn now to the substantiated redescription.
In response to a claim ‘it was a murder’ we might ask for a description of what has happened. What we are requiring when we are asking for a description of what has happened is not a different description: we are asking for the relevant facts that would substantiate the original description. This is a normal procedure in all our life, not only in our moral life. If someone claims that a certain substance is poison we might ask for a description of the substance. What we want is not a diluting description, for example that the substance is a greyish liquid, but we want a description which would establish the correctness of the original description. Similarly, in the case of murder, we are not asking for the substitution of a different description, for instance ‘he moved a piece of steel very fast and the other person remained motionless,’ but we ask for something that would substantiate the original description. Only if the claim cannot be substantiated can we demand a different description.
In moral life and appraisal we are interested in substantiated descriptions and redescriptions, and no light is shed on the problems of these by the ritual arguments about description and evaluation or about the dichotomy of ‘is’ and ‘ought.’ Even just to indicate what the problems of morally relevant descriptions are, cannot be our concern here. But I have been saying more than that we should not rule out a priori any questioning of the existing philosophical orthodoxies that are based on the assumptions of the ‘is-ought’ dichotomy. I have been saying that we must forget that dichotomy if we want to give our attention to the philosophical problems of our moral life.
[1] Philosophical Review 73 (1964) reprinted in Philippa Foot, ed., Theories of Ethics (Oxford: Oxford University Press, 1967), 101-114.
[2] R.M. Hare, ‘The Promising Game,’ Revue Internationale de Philosophie, 70 (1964), reprinted in Foot, Theories of Ethics, 115-127.
[3] The Language of Morals (Oxford: Clarendon Press, 1952), 83-84.
[4] ‘The Promising Game,’ reprinted in Foot, Theories of Ethics, 127.
[5] Aristotelian Society Supplementary Volume XXV (1961).
[6] Ibid., 74.
[7] P.T. Geach, ‘Good and Evil,’ reprinted in Foot, Theories of Ethics, 82.
[8] Aristotelian Society Supplementary Volume XXXV (1961), 77.
[9] Ibid., 69.
[10] The Language of Morals, 168-9.
[11] Montefiore, ‘Goodness and Choice,’ 74.
[12] ‘Geach: Good and Evil,’ in Foot, Theories of Ethics, 79.
[13] I am grateful to Dr John Colman for letting me read parts of the manuscript of his forthcoming book on the moral philosophy of Locke [John Locke’s Moral Philosophy (Edinburgh: Edinburgh University Press, 1983)] and for drawing my attention to the similarity of Locke’s views to what I am groping for. (I think one can gain some similar insights from Vico.)