Julius Kovesi
In a lecture-seminar at Oxford a few months ago I heard the following turn of argument. After some discussion of the similarities and differences between the good or virtuous man and the good carpenter or good musician, it was suggested that perhaps similar comparisons could be made to elucidate our social virtues. For these similarities and admitted differences we were asked to turn to the world of bees. It was felt that a musician can exhibit excellences that concern only his own abilities and potentialities, and so this can illustrate only those virtues that concern our individual development into excellent persons. Morality however concerns our dealings with other people. So we had to turn to the bees.–Of course we could have turned to a concert pianist instead, but even this would not have made me happy. I would have been happy only if we had abandoned the whole pattern of argument by saying that the concert is over, the pianist has pocketed the whole takings of the charity performance and gone on holiday on the proceeds, and now let us talk about moral problems. But then we would have been at a loss because we would not have known what our pianist is good at or bad at; we could not have said that he is bad as a this or a that. So we turned to the bees because there we were able to say that the bee which goes off with the honey to Majorca on his own, instead of bringing it back to the hive is bad as a bee. He is not functioning properly. Actually, the example was not even this, but, as far as I remember, a bee which is not interested in honey.
After a lecture-seminar it is not easy to question in a few words the whole procedure of the enterprise, so I tried to make only one small distinction. I tried to distinguish between a bad bee, who is bad as a bee, and a bad little bee. For this I was greeted with perhaps well-deserved laughter. The distinction I was trying to make is that between a bee which is indeed insensitive to honey and is just buzzing around aimlessly and a bee which performs his beeish functions properly, but cleverly puts aside a bit of honey in a corner, wakes up at night and eats it up secretly by himself. He is not bad as a bee, he is a bad little bee.
Incidentally, I am not sure whether we would really say of the first one that it is a bad bee. We are very egocentric in our use of ‘good’ and ‘bad’. We do not say that the weather is good when it–whatever ‘it’ is–functions properly as ‘weather’, but we say that the weather is good when we can go for a picnic. (Of course this is why we talk about ‘weather’ at all, not because there is such a thing but because we want to have picnics and get in the harvest.) We say that the baby is good when she sleeps through the night and we can get to sleep too. So a bad bee would be one that a child would bring to a zoology class and the teacher would find that its wings are crushed or that it is too small to exhibit. It is a bad specimen.
To have made the distinction between a bad bee and a bad little bee was a rather hopeless small protest. What I would like to do now is to question the whole procedure that made us go to the bees in search of some light on our moral life. I want to begin with our familiar picture that in our activities and endeavours we can roughly distinguish between our dealings with factual matters and with matters of values. What I want to attempt is to make a further distinction in the second field, in that field of our lives where we are grappling with values. In doing this I shall have to end up with some criticism of the way in which the distinction has been drawn between these two large fields. But to begin with I would like to say that its recognition has been very important and quite understandably it impressed us so much that we tend to regard other types of activities as only subdivisions or variations of these two main activities, of making factual statements and of making evaluative judgments. It is this further assumption, namely that all, or most, human activities are subdivisions or variations of these two, that has got moral philosophy entangled in the second of our great divisions. Quite understandably if we can choose only between these two fields, the subject-matter of moral philosophy is somewhere in the second field. Then usually this is what happens. We regard those features of these two fields to be their general and typical characteristics that distinguish them from each other, and so various philosophers arrive at various sets of characteristics as the general and typical characteristics of evaluative judgments. Then, since moral judgments are thought to be within this field, they too are first burdened with these characteristics and then, since in fact they are not at all like evaluative judgments, they are somehow distinguished from evaluative judgments.
Might it not be the case, however, that the field of moral philosophy differs from the field of evaluation just as fundamentally and radically as the field of evaluation differs from that of description? Perhaps we place our moral judgments in quite unsuitable company and spend most of our energies on how to extricate them, when they should not have been there to begin with.
We begin our moral philosophy by describing how we choose fire-extinguishers and cricket-bats and then say that moral choices are like this except that in moral choices we do not choose between things like fire-extinguishers and cricket-bats. Or we begin by describing how we evaluate functional objects and then say that moral decisions are like this except that in moral decisions we do not, and we should not, regard people as functional objects. These are like the story of the boy who asked his father what a telephone was. He was told: ‘Imagine a long snake with its tail in Edinburgh and its head in London. If you tread on its tail its head will move in London.’ ‘I see, but what about the wireless then?’ ‘Well, it is exactly the same, only without the snake.’
Our philosophical literature has good runners-up to this story, like the story of the social contract, of natural law, non-natural qualities, or when you are asked to listen to the voice of your heart or obey commands that you addressed to yourself. So we are asked to observe the good bee and the good fire-extinguisher. They are good because they do what they are supposed to do. The good man is exactly like this, except that we, unlike these other objects, can choose to do or not to do what we are supposed to do, and also we do not know what we are supposed to do. This is exactly our problem.
There are many activities that we can properly describe as evaluation. Some of them are of quite recent origin, as when we evaluate the results of a survey or evaluate data. We must not ignore these if we want to understand ourselves. Here I want to outline what is the most ancient and widespread of these activities, when we make an evaluative judgment on this or that as a good something or other. This is not only the most ancient type of evaluative activity but the most ancient activity altogether. I would be tempted to say that it precedes the use of language as we know it except for the fact that as we know this activity now it cannot exist without language.
Our language did not begin with some cavemen, say, seeing some large floating objects on the sea and thinking: how interesting, they all look very similar, let’s abstract their essential characteristics and form a general concept of them. Then we could call them by one word, say, ‘boat’, and wouldn’t it be convenient for communication? No, when they went down to the sea there was nothing there but fish and they thought it would be nice to be able to go after them. First came the final cause of what later was to be regarded as a boat, then this brought about its formal cause, namely the idea of the sort of thing that would do the job. Then some of the cavemen became the efficient causes by providing the material causes in the shape of wood. If the first of their constructions didn’t quite do the job, they thought that this was not what they intended, this was not quite what they had in mind, this was not … well, let’s help them and say, this was not a good boat.
What enables us to evaluate things is that our descriptions of things are standards, they embody the purposes and intentions that made us form their notions to begin with and they are all capable of being exemplified by many particulars in space and time. We evaluate particulars that fall under a certain description as they more or less come up to the standard of what they are supposed to be or regarded to be. If anyone defined what a boat was purely in terms of the characteristics of any of those visible objects that we call boats, he would commit the naturalistic fallacy in the sense of identifying a standard with something which is not a standard but is made to approximate to it, and he would logically prevent us from producing better and better boats in the future, in fact he would prevent us from evaluating boats.
When something is not good as a boat it could still be good as a container on dry land or good for firewood. In each case one has to complete the phrases ‘good as’ or ‘good for’ with a description. The phrase ‘good as’ indicates most clearly that we bring the object in question under some heading, under a description. These phrases are like levers shifting the objects from under one description to under another. We can say that something is not good as this but it is good as something else or good for something else. But we do not use the phrase ‘good for’ when the object is for what we want to judge it good for. Not because it is not good for it but because this is what it is for. This further indicates how descriptions are standards. We do not say that needles are good for sewing, except when we first explain to someone what they are at all. ‘Good at’ and ‘good with’ are used to evaluate skills and here again we can observe the same pattern. If the skill in question can be described by the use of one word or one phrase we can evaluate someone’s skill by saying that he is a good such and such, e.g. that he is a good carpenter. In the absence of such a word or phrase, e.g. ‘firelighter’, we say that he is good at lighting fires. Incidentally, it is for reasons like this that I was tempted to say that this sort of activity of evaluation may precede in a sense the formation of our language because only after continued recognition of things being good for this or that do we begin to form terms for this or that. But of course we cannot for long recognise that something is good as this or that without some notion of this or that.
I used to think that the phrase ‘good to…’ when followed by a verb also falls into this same pattern as when I was thinking of examples like good to eat. What is good to eat is good as food or as nourishment. But there is a sense of ‘good to…’ followed by a verb which is not used for evaluating but for what I shall soon call ‘valuing’, as when your dentist advises you that it is good to chew, or when someone questions your beating the carpet and you say ‘it is good to beat it’. We mean in these cases that it is a good thing in the 1066 and All That sense of a ‘good thing’.–The other ‘good to…’ phrase when it is not followed by an infinitive but is used meaning ‘good towards’ (as when the stepmother was not good to Cinderella) is a notable exception among the phrases using the word ‘good’. This is the only one of these phrases which is used for talking about human relationships and which could be used in a moral context.
Before I turn to the consideration of what I want to call ‘valuing’ as against ‘evaluating’, I would like to consider two connected objections to what I have said so far.
It could be argued that I have been unfair on two counts in taking functional objects as my examples. First it could be said that I made it too easy for myself to illustrate what we do when we evaluate by taking functional objects, professions and skills as my examples when there are many other objects and human performances as well. Secondly it could be said that by taking functional objects or professions and skills as my examples I was not giving examples of what could be called, strictly speaking, descriptions.
With regard to the first objection I would like to ask what other examples one could use to illustrate evaluation? These are the things we evaluate. If I were to investigate, say, the concept of ‘growth’, it should not be objected that I am making my job easy by taking as my examples organic matter and human institutions instead of other things. Similarly I am not just making my job easier by taking functional objects and human skills as my examples of evaluations. I am showing the examples where we evaluate. This is the main burden of my paper, to argue against the view that we are evaluating all over the place, including our moral life; and that in order to introduce what we do in our moral life we first take ‘easy’ examples like the evaluations of fire-extinguishers and then proceed to the more difficult ones like deciding whether to tell a lie or not. This way we make our job not only difficult but impossible or in so far as it is possible, misleading. Of course we shall find it more difficult to find out how we evaluate in our moral life if the first thing we are asked to observe is that there we are not dealing with functional objects nor with skills, nevertheless we are evaluating on the same pattern (it is the same you know, without the snake). Why not just talk about evaluation and then take functional objects as our examples, not because it is easier but because these are the sort of things we evaluate? And when we turn to moral philosophy why cannot we recognise that there we are dealing with quite different problems, and I mean quite different?
But one could return with the objection by saying that even outside the field of morals there are cases where evaluation is rather difficult, or more complicated than in the case of functional objects. How would we evaluate pebbles for instance? Well, we don’t. We could make use of them as ballast but then we evaluate them as ballast, under a different description. We employ one of these phrases that will shift them under a different description and say ‘these are good as ballast but those are not’, or ‘this is good for ballast’. But where we do not evaluate, do not ask how we evaluate there, but say that there we do not evaluate. Especially we should not make the mistake of thinking that the ‘easy’ examples are not genuine cases of evaluation, the real examples are those where it is hard to think how we could possibly evaluate something. The ‘real’ example of evaluation is when we evaluate pebbles. Here there is no given standard but we as collectors of pebbles make up some standard of our own. So the real case of evaluation occurs when the description of what we evaluate cannot give us any help but we provide the standard by our own decision. This may be the genesis of our prevailing distinction between evaluation and description into which I do not want to inquire now directly. Let me just say that this is carrying my snake story really too far by assuming that the real and genuine cases of what is illustrated by the snake are exactly those where there is no trace of a snake. Let me also repeat that the collector of pebbles evaluates them as geological specimens, as semi-precious stones or as beautiful objects, not as pebbles. In turn we can evaluate him as a collector and, if he is not good at his standards, he is not a good collector. But let’s not pursue the collector any further because he cannot serve as an introduction to moral philosophy, not even if we make him say ‘I hereby resolve to collect stripy pebbles; do so as well’. The collector of stripy pebbles is not deciding on his standards but on a description of something which he is going to collect good examples of.
I have mainly answered by now the second objection as well, which is this: my examples of descriptions are not examples of genuine descriptions. Genuine examples of description are those that do not help us with evaluation, like describing some objects as pebbles. Instead of saying that certain objects are boats, which is cheating, I should have said that some pieces of wood are arranged in a certain way and nailed together. This is what they are, not boats. How could I evaluate this, or how could I move from this to an evaluative judgment? My answer is again simply that we do not evaluate this and we cannot move from this to an evaluative judgment. But if we want to illustrate how we evaluate something we have to take examples of those things that we do evaluate. We can of course say that this construction is good for floating on and to go out fishing in and then we are right back in the stone age and soon we shall be creating some such simple word as ‘boat’. The person who says that the object in front of him is not a boat but pieces of wood arranged in a certain way and nailed together does not only make evaluation impossible, he does not know how to describe the object, he does not know what the object is. To test this we should observe whether he can follow a rule in recognising other objects as being the same or not the same. In the next object on the water the pieces of wood are arranged differently and they are not joined by nails, and the one next to it is not even made of wood. If he keeps to his original statement then he cannot regard the second and third objects to be the same as the first, nor of course the first to be same as the second and third; therefore he does not know what even the first object was; he has no notion of a boat. Let us not ask therefore how someone who has no notion of what a boat is will evaluate boats in the hope that his resolution of this problem (e.g. adopting his own principles) will shed some light on the problems of moral philosophy.
There is a use of the word ‘descriptive’ according to which to describe something as a boat (or an act as stealing) is not a good example of description but to describe it as pieces of wood nailed together (or to say that he moved an object that makes the noise tick tock from one pocket to another pocket) is a good example of description. I want to say this to people who use the word ‘descriptive’ in this latter sense: this use of the word itself is not a descriptive use in their but only in my sense of ‘descriptive’ because it is used as a standard, it was formed from a certain point of view, it was formed for a purpose. Its purpose and function is simply for use in certain types of moral philosophies to illustrate terms that are of no use or little use when we evaluate. Because the term ‘descriptive statement’ is itself used as a standard and used for a purpose in philosophical system-construction, it enables those who use it in this way to make evaluations. They can say for instance that the statement ‘he has put marks on the paper’ is a good descriptive statement and the statement ‘he signed the document’ is a bad one.
But I have to turn now briefly to what I want to call ‘valuing’. Here I am handicapped by the interesting fact that the word ‘valuing’ has not one single opposite, nor is there what I would really want for my purpose, a more general term that would include both ‘valuing’ and its many opposites. Let us keep this in mind when I want to point out the distinction between evaluating and valuing.
We value a good knife more than a bad knife but I suspect we do this because we value knives as such. We may admire, or be fascinated by a good burglar more than by a bad burglar, but we do not value him more, because we do not value burglars. We can evaluate both things that we value and things that we do not value. (And by ‘do not value’ I mean the opposite of valuing and not simply the absence of valuing.) In both cases the good one is the one that comes up to the standard of what it is supposed to be according to the description. But we do not decide what things we value as against those that we do not value by regarding them as instances of a higher description, and those that come up to the standard of the description we value, and those that do not come up to that standard to the same degree we do not value. So valuing is a different type of activity from evaluating. As I was trying to say, we evaluate particulars as they fall under a certain description but valuing is what governs the formation of our descriptions.
I would like to suggest that our moral life is more akin to valuing and is not at all like evaluation.
When I have to make a decision whether to be honest and go to jail or be dishonest and get an innocent man in jail instead, I am not trying to choose between several honest acts, some of which are better than others in so far as they are honest, nor do I regard both honesty and dishonesty as instances of something of which one is a better instance than the other. (Of course we can find better examples of honesty, if we want an illustration, as some Renaissance paintings are better than others. But the honest man’s job is not to try to perform better and better examples of honest acts as the Renaissance painter’s job wasn’t to paint the best example of a Renaissance picture. As the moral agent is not conducting a class in moral philosophy, the painter is not compiling a selection of illustrations.)
When I say that when we decide whether we should be honest or not we do not choose between several possible honest acts, and when we are praised for being honest, we are not praised for having done the best honest act but simply for being honest, I am not saying that each honest act is unique as against boats of which there are many. In a sense every act is unique but in so far as we judge an act to be honest it is not unique. In our moral life we are interested in an honest act in so far as it can be described as honest and not in so far as it may approximate to honesty. It must have been very puzzling, even for Plato, why the subject-matter of moral philosophy, like the subject-matter of mathematics, should be in the upper half of the Divided Line. But we must not think that the reason for it is as mysterious as the Divided Line and all that that system would suggest.
In our moral life we are interested in the description under which our action falls and in the relevant facts that justify us in regarding our acts as falling under one or another description. The facts we are interested in are not qualities that make a particular a good instance of something but those facts of the situation and circumstances that make what we are doing one act rather than another. The very facts that are in the running for being relevant are of different types in the case of evaluation and in the case of our moral decisions. So as an introduction to moral philosophy, instead of taking clues from how we evaluate, we have to take clues from how we describe. To defend myself from the accusation that I incite you to commit the naturalistic fallacy would require another paper in order to outline what it is to describe, how we form the terms by which we describe, and why we form terms at all in order to describe our acts in certain situations. I just remind you how I laboured earlier in the paper to indicate that our intentions, purposes, and standards are part and parcel of our descriptions. This is even more so when we turn away from the inanimate world and we become not only the describers but also the subject-matter of our descriptions.
In conclusion I would like to mention three problems that complicate my thesis.
The first is the case of decisions that one makes as part of certain types of long-range attitudes or policies like wanting to be kind to someone. This is different from the choice between being honest or not, here we have a choice of different acts, all of which can be described as kind, and I have to choose between them, as in the story of the boy whose name was Jim whose friends were very good to him
they gave him tea and cakes and jam
and slices of delicious ham
and chocolates with pink inside
and little tricycles to ride
and read him stories through and through
and even took him to the zoo…
Here is an ample choice of kind acts to choose from. There are many reasons for choosing one kind act rather than another but I don’t think one could make up a case for saying that we choose the one which is better as a kind act. The kindest act is not the one which approximates in a scale of kind acts to what a kind act is supposed to be.
The next problem I want to throw up equally briefly is the problem of the good Samaritan. It looks as if this were a problem for me because here is the word ‘good’ used in a moral sense. But to indicate briefly how this judgment differs from the usual pattern of evaluation let me draw your attention to the difference between the good burglar, who is good as a burglar, does his job efficiently, and that other good burglar who is so bad as a burglar that, when he discovers the old lady resting in the house, he makes her a cup of tea. If a policeman investigating the burglary does the same for the old lady, then he is a good policeman in the sense in which the burglar was good in the first case and not in the second. The good Samaritan is like the burglar in the second case. The operative word is the ‘Samaritan’ and again the story would not have the same moral if it were the story of the good ambulanceman.
My last problem is really a case of evaluation that should be of interest to moral philosophers. The problem is really in the field of what is usually called cultural or anthropological relativism, but it could be the problem of a moral agent either during the period of social and moral change or when the agent has to choose between different societies. Let me illustrate this by a simple account of how one might justify the institution of revenge in a certain society. There is no police force or effective judiciary in that society. Now if a strong bully would want to murder someone in that society, he might run over the following points in his mind before he would attempt the murder. He might think that although he is stronger than his intended victim, his two brothers are a rather formidable team. And even if he were to join the crusades after the murder, those brothers might kill off his family. Our potential criminal today would think about the police force this way and he would even check on the laws of extradition before he would plan to fly off to modern Constantinople. Of course, if today the brothers of victims would kill criminals or their families, we would say, among other things, that they should not take the law into their own hands. But what about a society where there is no law to take into your own hands?
Now what enabled me to make the comparison between these two cases is that one does recognise something in the two cases which is the same in spite of all the differences, even though we might not be able to say before long considerations what is it that is the same. What I want to say is that if someone is looking for evaluative problems in the field of moral philosophy then this might be it. One could argue that one of my examples is a better manifestation of something than the other. This is more like the case of evaluating boats, when we compare primitive constructions with other developments that are designed to serve a purpose. As in the case of boatbuilding we have to take into account the available material and human sophistication, so in our cultural and moral development we have to look at our human resources and available institutions, but nevertheless I think we could make evaluative judgments here and say that one arrangement is better than another. But here we are dealing with institutions and arrangements and with whole moral systems and not with the problems confronting moral agents within any of these systems though perhaps today these larger problems confront us more and more even as individual moral agents.