To this, I think, the answer is reasonably clear. It is, quite simply, so that we have an adequate and consciously-held view about what we are trying to do, about the nature of the enterprise in which we are engaging. Now of course actual people in actual jobs - teachers, civil servants, and so on — will be engaged from time to time in many different enterprises. A teacher does not only teach: he may keep the register, referee football matches, attend union meetings, and so on — and in time of war, or some other crisis, he may find himself having to keep his pupils safe from bombs or plague. Similarly a doctor does not only cure people: he may also have to fill in forms, keep accounts, tidy his consulting-room, and all sorts of other things. But we (rightly) have the feeling that there is some enterprise with which these people are, or ought to be, specially connected: something which is central to what they do. Just as doctors are centrally concerned with medicine and promoting health, so — we may feel — teachers and others are, or ought to be, centrally engaged in the enterprise of educating. Education is, so to speak, what teachers are for (cf. Hirst and Peters, 1970, pp. 17-19). 19).

This feeling, so far vaguely expressed, does not allow us to conclude that the other enterprises are unimportant, or ought somehow to be got rid of. Much will plainly depend on circumstances. If we are attacked by barbarians or have not enough to eat, it will no doubt be sensible for teachers (and perhaps even doctors) to stop teaching (curing) and turn their attention to getting food for themselves and other people, or to fending off enemy attacks. The feeling is rather that there are in principle — ‘in theory’ if you like, though we hope also in practice — enterprises whose nature just is different. Educating people is one thing, curing them is another, keeping them properly fed is yet another, and so on. We have different words which may fairly mark these enterprises — ‘education’, ‘medicine’, ‘economics’; but the words in themselves may not give us a sufficient grasp of what the enterprises are and how they differ from each other. For the enterprises exist in their own right (‘in principle’ or ‘in theory’), whether or not people mark them clearly with certain words. Even if people did not mark them at all, or practise them at all, they would still be important: the enterprises we mark by ‘science’, ‘medicine’, ‘democracy’ and so on are important in themselves, even though many societies may have had no understanding of them, and lived by superstition, witch-doctoring, and tyranny.

Nevertheless one has to begin by using some word or words with which to conduct the discussion; and I shall start with a brief criticism of what one or two philosophers have said about ‘education’ — not simply for the sake of criticizing, but because it is very important to be properly aware of a number of temptations which beset us here. The word seems to open the door for almost any kind of partisan prejudice or fantasy; and not a few writers seem to want to keep the door open in this way. Thus when Max Black says that ‘ “Education” is, and should be, a term as comprehensive as “life” or “experience” ’ (in Lucas, 1969, p. 283), and tells us on the next page that ‘In practice, philosophy of education becomes nothing less than philosophy, without qualification or restriction’ (p. 284), we recognize that things have got out of control. We know perfectly well that ‘education’ is not as ‘comprehensive’ a term as that; and we know that, if somebody claimed to be doing ‘philosophy of education’ when conducting a detailed examination of (say) Berkeley’s theory of perception or Aristotle’s categories, we should not readily understand him.

There are two, perhaps three, general temptations which need especial notice. The first is evident in Frankena’s remark that ‘Education is the process by which society makes of its members what it is desirable that they should become, either in general or in so far as this may be carried on by what are called “schools” ’ (p. 288), or in O’Connor’s ‘The word refers to the sort of training that goes on in schools and universities and so on’ (O’Connor, 1957, p. 5). Here ‘education' is identified (roughly) with what we may call the educational system or ‘what society does’. One obvious objection is that ‘society’ can make its members richer, or better-fed, or plenty of other ‘desirable’ things without doing anything to them which we should dream of describing as educating them; but the real danger lies in identifying an enterprise which exists in its own right with particular social practices or institutions.

A parallel may help here. Human beings may engage in an activity or enterprise which we may want to call ‘religion’ — though no doubt we are not entirely clear just what this enterprise is. It would obviously be wrong to think that this was the same as saying that there were certain people and social practices — parsons, funerals, churches, grand inquisitors, and so on — actually in existence: or even the same as saying that there were certain sets of beliefs and doctrines which were called ‘religions'. For we could always ask ‘Are these people (institutions, beliefs, etc.), whatever they may be called, actually concerned with religion?' and we may often find that they were not. If we identify religion with social practices, we make the same mistake as the man who is humorously quoted as saying ‘When I say religion of course I mean the Christian religion, and when I say the Christian religion I mean the Church of England.’ The point is not just that the man is prejudiced: it is that he has no idea of religion as an enterprise in its own right (Wilson, 1971, Part 1).

So with the notion of educating people. We may call certain things ‘schools’ and certain people ‘teachers’, and we may say that what we are doing is to educate children. But we have to be able to show that this is, in fact, what we are doing. The mere existence of social practices with the word ‘education’ attached to them shows nothing; any more than, in the police state of Orwell’s 1984, the existence of an institution called ‘The Ministry of Truth’ showed that the institution was, in fact, concerned with truth (rather than, as Orwell represents it, with propaganda). It is, in fact, a very open question how much that is actually education does go on in those institutions which we currently classify under that heading. Clearly we cannot answer the question until we know, or decide, what ‘education’ is to mark: but equally we cannot assume that the answer is given by existing institutions.

The second temptation is to use (or abuse) ‘education’ to endorse, not a particular set of social practices, but some particular ideal or set of values which we happen to favour. Most writers on the subject have some general ideology, or ‘doctrine of man’, or political or moral theory which they want to sell: and their ‘educational theory’ (together with what they want ‘education’ to mean) exists chiefly as a kind of spin-off, so to speak, from this general ideal. Thus Plato (Laws, 643—4, Saunders, 1970, p. 73):

It is worth noticing here that it is not just the English word ‘education’ which tempts us in one or the other of these two directions. Other words in other languages suffer the same fate: thus Plato, in the passage above, monopolizes the Greek word paideia (here translated ‘education’) for his particular purposes, and the same can be, and has been, done with the French I’éducation, the German Bildung, and so forth. The same temptations beset words (in any language) which may mark enterprises of a fairly general nature, about which we have not taken the trouble to get clear: ‘religion’, ‘politics’, ‘morals’ and many others. They lure us to endorse either some existing social practices, or else our own partisan views (and, indeed, these two may obviously be more connected with each other than I have here made them appear).

A third approach is to use ‘linguistic’ or ‘conceptual analysis’; and I hope to make clear how far this can be construed as a ‘temptation’, as against a perfectly proper procedure. One obvious difficulty is that it seems somewhat arbitrary to select a particular word in a particular language, used by a certain group of speakers at a certain time. Why should ‘education’, as used by certain twentieth-century English-speakers, be of more interest than, say, paideia as used by certain Greek-speakers in Athens of the classical period? The difficulty can partly be met by what we said earlier about distinguishing different enterprises, rather than being concerned just with the words that may mark them. As Hirst and Peters put it (1970, p. 8): ‘the point of doing conceptual analysis, which is to get clearer about the types of distinction that words have been developed to designate. The point is to see through the words, to get a better grasp of the similarities and differences that it is possible to pick out.’ But we have to assume that the words, which are our inevitable starting-points, have in fact developed in such a way as to designate or pick out the really important distinctions. Often this assumption is justified: there are many concepts which are, so to speak, both transcultural and inevitable — that is, which men in almost any culture are virtually bound to possess, and hence to mark by some words. But it remains an assumption. What the philosopher has to do is to elicit the important distinctions: how far these are to be found in actual linguistic usage cannot be determined in advance, and is ultimately a separate question.

In any case, if we start by engaging in this sort of analysis we have to beware that we do not fall victim to either of the two temptations mentioned earlier. If the procedure has merit, it must be partly by way of preventing ourselves from smuggling in our own particular values or doctrines. Thus in Peters’s earlier work (1966, p. 30) he claims that ‘we would not say that the ancient Spartans had received a military or moral education’ and similarly, having said some fairly stringent things about the necessity for ‘commitment’, ‘being on the inside of' various intellectual activities, and ‘a man’s outlook’ being ‘transformed by what he knows’, he adds (p. 31): ‘I do not think that we would call a person “educated” whose knowledge was purely external and inert in this way [i.e. who did not possess these attributes].’ These are overt linguistic claims, and very dubious ones. Similarly (in Tibbie, 1966, p. 71): ‘ “Education” . . . encapsulates criteria to which a family of processes must conform. The first is that something valuable should be transmitted in a morally unobjectionable manner’; but one might have strong moral objections to, say, very severe methods of teaching Latin grammar and plenty of other things, and yet think that one had (pro tanto) been educated or even well educated by so learning them.

In his later writings Peters makes it clear that his concern is with a ‘more recent and specific concept’ of education, a concept which implies ‘the development of states of a person that involve knowledge and understanding in depth and breadth’, and which also implies that these states are desirable (Hirst and Peters, 1970, p. 25). This is, by and large, the concept explicated by Hirst and Peters conjointly, and they recognize that it is only one concept that ‘education’ might mark: ‘in so far as we are concerned about education in what we called its specific sense, we are committed to processes which assist the development of desirable states in a person involving knowledge and understanding’ (p. 40). Other writers (Downie, 1974, p. 173) similarly stipulate a particular meaning for ‘education’, derived from some (very sophisticated) notion of ‘the educated man’ (ibid., pp. 11ff.).

However much interest there may be in outlining — one might fairly say, creating or fashioning — these ‘specific’ concepts of education, or in simply stipulating what ‘education’ is to mean, it is entirely plain that we are no longer engaged in conducting a non-partisan analysis of how words are used. We are being asked to direct our attention to a particular concept which the (various) authors think to be very important; as Hirst and Peters (1970, p. 41) put it, they ‘are conscious that a definite moral point of view is implicit in their approach’, and disappoint us by adding ‘but it is not part of the intention of this book to attempt any explicit justification of it’. But our attention will only be engaged if this particular concept is shown to be important. We want to know whether it is, and why it is, and what sort of importance it has (‘moral’ importance is only one kind of importance). If this is not done, all that we have is just another attempt to promote a particular idea or set of values, this time put together in a rather confusing way with the quite different business of ‘conceptual analysis’.

So far as I can see, these and other similar manoeuvres are not only mistaken but largely unnecessary. As I have suggested elsewhere (Wilson, 1975, chapter 1), there is a concept which, when properly explicated, makes tolerably clear the kind of enterprise we need to distinguish, and — though this is, in one way, a secondary consideration — best fits the term ‘educate’ as now normally used by English-speakers (and, I would add, other parallel terms that exist in other languages). To state this as briefly as possible: ‘education’ is the marker for a particular enterprise or activity, which has as its aim or ‘good’ the sustained and serious learning of rational creatures, planned in some coherent or overall way. We educate people (rather than treating them in other ways) when we are engaged in bringing such learning about; and people become educated — using that term as the ordinary past participle passive of‘educate’, and not with any ideological accretions — when or in so far as they have done some learning of this kind. (A good parallel can be found in the Latin ‘doctus’, which can either mean just ‘having been taught’ [‘having learned’], or something much more specific like ‘cultured’, ‘well-read’, ‘learned’.)

Compared with the particular pictures presented by most authors, this is a fairly broad concept; but it is, I think, the concept that most contemporary English-speakers do actually mark by such terms. There is some delimitation on what learning will count as education: we do not use the term of trivial or fragmentary bits of learning, nor of the learning of animals or infants. But we do use it where what is learned may be undesirable (bad habits, or to hate Jews, or plenty of other things): and where the amount of knowledge or understanding is very small (one can learn, in a serious and sustained way, to acquire certain habits or skills or attitudes). We speak of bad — that is, not just incompetent but evil — education (just as we can speak of bad religion, bad moral principles, bad political ideals, and so on: we have to distinguish these from cases which are not cases of religion, or morality, or politics at all); and also of education that does not involve much knowledge or understanding.

This (very brief) sketch needs to be supplemented by two elaborations. First, some delimitation, not so much of content as of general intention, is placed on the concept of education by virtue of the fact that education is a rather general or comprehensive kind of enterprise. Thus the Oxford English Dictionary speaks of ‘systematic instruction, schooling or training’, and for ‘educate’ gives ‘to bring up (young persons) from childhood so as to form (their) habits, manners, intellectual and physical aptitudes’. Not just any learning counts as education; the learning has to be seen as part of a systematic and coherent enterprise. Hence the grammar of ‘educate’ is different from the grammar of (for instance) ‘train’; we can train people in particular skills, or for particular tasks, or as fillers of particular roles. But we can only educate people as such: that is, if we claim to educate people, we claim to be viewing their learning from some general, overall or comprehensive point of view, not just with an eye on certain jobs or skills.

Of course since people have minds, and education consists of learning, it is likely that a large part of this enterprise will be seen as developing knowledge and understanding in people; indeed, an educational ideal which involved no such development would be hard to conceive. Yet one might easily think that the really important things for people to learn — still in a comprehensive and coherent sort of way — did not involve much intellectual or ‘cognitive’ sophistication, but were more in the area marked by ‘character’, ‘habits’, ‘attitudes’ and so forth; and one might believe that these things were best learned by imitation, or practice, or exhortation, or playing games, or other methods of that kind. Again, one might think it rash to lay down any particular content as being ‘really important’ for all pupils — such content might reasonably vary according to the pupil’s particular needs, abilities, or station in life. But the notion of education is neutral with regard to any questions of content, so long as there is an enterprise of this general kind, the title of education cannot be denied.

Second, some delimitation arises from the fact that ‘education’ normally marks a fairly formal, structured or institutionalized enterprise: something designed to raise people above the level of what they would naturally learn for themselves in the ordinary course of events. We do not speak of parents and other language-users educating their children, or even teaching them to talk, if the children just pick up the use of language from the adults — even though this learning might be thought crucially important for any mental development. We may, indeed, loosely say that certain people, or experiences, exercise an ‘educational’ effect; but ‘education’ and ‘educate’ is a much narrower term than ‘bring up’, ‘rear’, or ‘nurture’.

These, I think, are at least some of the features which would emerge from a thorough and systematic study of how the words are actually used. Much more work, in my judgment, needs to be done in this field, both on English words and those terms which are, at least prima facie, parallel in other natural languages. But whatever may or may not be true o...