OPINION
A bundle of conjectures is not a habitable world4
Facts and Values
In the preceding chapter we looked at the simplest points about the structure of our knowledge. The knowledge I was primarily thinking of was that contained in the various sciences. But people are inclined to think that the sciences are somewhat specialized; most of our active knowledge relates to the more diffuse, less specialized context of our ordinary life. The sciences are, however, convenient for my purposes because the claims we make there are usually meant to be simply factual. They are attempts to say how things are, how they work, without any other ideas or attitudes impinging on them. But much of our less specialized thinking is shot through with such attitudes or opinions. We can sometimes describe the world about us in a disinterested way - we can report that the wall has been painted pale green - but most of the time our reports reflect somewhat more involvement or engagement with the issues - we might say that the wall has been painted a pleasant shade of green, or we might indicate by some other feature, intonation or gesture perhaps, how we felt about the matter.
The distinction I am alluding to is usually discussed as a distinction between facts on the one hand and values on the other. It is not very easy to find an uncontentious way of making this distinction, partly because in our linguistic practice there is no distinction: factual and evaluative matters are tightly intertwined. Another reason is that there is serious disagreement among philosophers and other people about whether there really is an important difference. There is some difference, however, and perhaps the simplest way to see what it might be is to say that factual claims try to say how things are while evaluative claims add something about how people feel about things, about how they think we should act or choose, about which directions we should go, or about how we should judge. If you think these extra things are equally factual, then you will not think that there is an important difference between saying 'The wall is pale green' and saying 'The wall is a beautiful shade of pale green'; but even so you can admit that the second remark adds something to the first, viz. a suggestion about how people should respond to the colour.
Talking on the one hand of pale green and on the other of a beautiful shade of green is, as I have said already, somewhat untypical. A great deal of our ordinary language is not so explicitly factual or evaluative. It incorporates both aspects. If I say that someone told a lie I am saying something descriptive about what he said, that it is not true and was known or believed by him not to be true, but I am not making a neutral report of the fact that these things were the case; I am condemning the action. The word 'lie' incorporates the two things; it has descriptive truth-conditions (if we discovered that the person honestly believed what he said, we should withdraw the claim that he lied) but it also points us towards a particular judgment, it tells us to condemn rather than acquit or applaud. And the word 'lie' is by no means unusual; rather it is typical of the vocabulary of ordinary life.
The variety of notions which are involved here is worth some brief mention. We have just been looking at concepts like lie here, and in a previous chapter we noticed a similar complexity in the notion of courage. There are other cases in which the factual element is somewhat closer to the surface. One important category consists of grading terms. Eggs and vegetables, as well as examination candidates, are often graded in quite detailed ways. Typically what is involved in such grading is the comparison of individuals with a standard, either another existing individual or an ideal set of features. Such comparisons are as straightforwardly factual as you can get (although it may not always be easy to make the standards explicit, and no doubt there are cases which pretend to an accuracy and factual basis which they lack) though ascribing the grade to the individual is usually taken to be also endorsing this kind of grading scheme: choose eggs by size, rather than putative taste, for instance.
A somewhat similar comparison of what exists with a standard is involved in many judgments of fairness or justice. It is a factual matter whether someone has broken the rules of a game, though whether anything should be done about it is not. Similarly, it is a factual matter whether the accepted procedures have been followed in a trial or in appointing someone to a job, though again the decision that these procedures should be followed is an evaluation. But given agreement on such matters, these questions of fairness or justice are, from within the institutions involved, purely factual issues.
In saying that factual and evaluative matters are intertwined in much of our language I am trying my best not to commit myself to any particular theory of how this is done. One popular but crude picture involves a factual core of meaning with a free-floating evaluative appendage. You can then predict that the same factual core could be joined to quite different evaluations. There are plenty of examples where such would seem to be the case (as suggested by Hobbes:
there be other names of Government… as Tyranny and Oligarchy: But they are not the names of other Formes of Government, but of the same Formes misliked. For they that are discontented under Monarchy, call it Tyranny and they that are displeased with Aristocracy called it Oligarchy: So also, they which find themselves grieved under a Democracy, call it Anarchy ([1651] 1929, Part 2, ch. 19; italics in original)
and see also Flew, 1975, ch. 5, for pertinent discussion) But we need not assume that it is universally true. When people do want to evaluate the same factual situation in radically different ways, then we can expect them to create the words to do so. But when such diversity is not so pressing, there need be no parallel but evaluatively different concept. The main point, however, is that even here reflective analysis can uncover the two aspects that have been welded into one. We have no excuse for avoiding the philosophical examination of the factual status of evaluations simply because of the unavailability of concepts or vocabulary.
The Logical Structure of Evaluation
Evaluative claims are, then, made by sentences that usually make straightforward descriptive claims as well. The logical relations of such sentences are identical with those we have already examined. The singular claim 'It was wrong for John to say what he believed false' is covered by but does not entail, even in company with many analogous claims, the open universal generalization 'It is always wrong for people to say what they believe to be false.' But given that, for whatever reasons, we have accepted some such generalization, it is built into the terminology we use so that we would now more naturally say 'It is always wrong to tell a lie.'
Just as evaluative theory (formulated as principles rather than hypotheses) is built into reports of what is going on, so the generalizations we use are equally provisional, though people are often unwilling to recognize the fact. But as in the case of our scientific knowledge, reflection on history can bring home to one just how provisional our evaluative beliefs actually are. We often pride ourselves in fact on the amount of moral progress that has been achieved; but that is to admit that what were once thought to be acceptable principles are now seen to have been inadequate. In saying that there have been many changes, I am not of course saying that every evaluative belief has changed.
Similarly our evaluative beliefs are not overwhelmingly weighted in favour of either side of the logical fence: singular judgments or general principles. Rather, in sophisticated thought at least, there is what Rawls (1972) has called a 'reflective equilibrium': judgments of individual cases qualify principles, but principles and reflection on their applications can lead us to revise judgments on particular cases. Of course, some people, blinded by false views of the status of evaluations, can think that some particular set of principles is unrevisable; but so can they think about empirical knowledge. It is, however, often more dangerous for other people when they have such views about their own evaluations.
The Factual Status of Evaluations
While I suspect that a very great deal of what we have said about the structure of knowledge can be carried over virtually unchanged into the topic of evaluations, the most important philosophical question for us to deal with is whether evaluations are themselves true or false; whether there can be evaluative knowledge. The question is, then, whether evaluations are objectively given or whether they are only subjective.
It is worth noticing that this is not the question whether human beings agree on certain evaluations. Clearly they do, and it is even more clear that they do when we restrict ourselves to particular social groups. Equally clearly there are other evaluations on which there is not universal agreement, but the mere fact of different opinions tells us nothing. There are different opinions about a lot of straightforwardly factual matters too; people can be wrong, and everybody could be wrong about something. So agreement or lack of it tells us nothing on its own. Again we are not dealing with the question whether it is good for us to think that some values at least are objectively given. It is simply the question whether that belief is true.
I have suggested that what is often known as the argument from the relativity of values is not persuasive, at least as it is usually put. People notice quite correctly that we do things one way, the Romans do them another, and conclude that therefore there is no right way to do them. But that conclusion doesn't follow. All that follows is that at least all but one of us has got it wrong. (It is logically possible that we are all right, if the objective facts are simply that we should do things our way and they should do them theirs; but most people who want objective truths here want something less arbitrary than that - they usually think that moral principles at least, apply to people qua people and not to pretty arbitrary groups of people.) But notice that this common argument proceeds in the way I suggested we re-orient. It moves from evidence or data - the diversity of values - to a general conclusion about the status of values. But we have seen that no such argument is deductively valid, and that it is often better to view the matter the other way round. What do we need to postulate to explain the data, in a way that fits into our other explanations of other data?
If we approach the diversity of values in this way it immediately seems that the subjectivist line about their status is in a very strong position. It says people differ because there is no fact of the matter - differences are to be expected in fact. Since moral values are part of the machinery for getting people to do or accept things they might not otherwise want to do or accept, these values portray themselves as objectively given, but this is an ideological disguise. If one takes other areas of purely factual concern where there is also considerable diversity of opinion, it is usually the case that one's explanation of that diversity will itself involve some account of how the facts are: people have differed over the shape of the earth, but you will tend to explain this by reference to what visual evidence was available to them, and this explanation will involve the way light behaves and the actual shape of the earth and the relative positions of the sun and the earth, etc. You will have to say that the earth has a certain shape in explaining why people have differed over it; but in the case of values, we do not need to postulate any such correct values. Objective values are yet another hypothesis we can do well without.
The preceding argument is very crude. Its claims about the social role of morality are highly oversimplified and incomplete, but it should indicate the way a more adequate argument would be mounted. It should also show the fruitfulness of the re-orientation I argued for in the previous chapter.
But if its conclusion is accepted, we must say that a very large amount of what is taught, especially in the human and social fields, is only partly knowledge, at best. A lot of history or civics or the kinds of diluted psychology that gets into the schools is not simply factual data and theorizing, but is much more a matter of transmitting socially held values. There is now an extensive industry committed to uncovering the value commitments of large chunks of the official curriculum of schools (and not just in the human and social fields), and of virtually the whole of the so-called 'hidden' curriculum. None of this aspect of schooling can be said to be the transmission of knowledge. It is merely socially entrenched opinion. The doubts that usually fester into relativistic claims about truth are to that extent fully supported. A lot of what people are brought up to think truth or knowledge is a fraud. It is not true for some but not for others; it is rather not true at all for anyone. As I claimed in the discussion of truth, simple truth gives one a sharp cutting edge, which is blunted by facile talk of relative truths.
Disenchantment
In these last remarks we are focussing upon a very important aspect of the 'modern' world. In most societies, and indeed in parts of our own, social arrangements are underwritten by beliefs about the nature of things. In some form or another, societies have what anthropologists call 'charter myths'. The structures of human life, and much of its petty detail, connect with grand theorizing about the cosmos. In saying that evaluative pushes and pulls are not part of the natural world at all, but only something that we misrecognize as being there, I am endorsing that disenchantment of the world which Weber saw as central to secular modernity. A central part of education, as I conceive it, is this debunking, or at least uprooting, of many of our values.
As Bourdieu and Passeron note in passing (1977, p. 12), this creates a pedagogical problem: you cannot start by teaching the cultural relativity of cultural values, although you can hope to end up by passing on this bit of anthropological wisdom. To take a related case, you must learn some language or other in order to formulate the true belief that it doesn't matter which language you learn. Similarly, you must be given some food or other while your teacher is bringing you to an awareness of the arbitrariness of food taboos. While this is so, it should not blind us to the facts that it is possible to end up with the beliefs we are discussing and that societies vary considerably in their self-awareness of these matters.
Truths of the Heart
In this brief examination of what properly should be considered merely matters of opinion rather than knowledge, I have focussed on the evaluative aspects of much ordinary thinking. But there are other aspects, of considerable importance to us as people, if not central to schooling, in which it is by no means clear that we have much that can be called 'knowledge'. Much of our thinking about ourselves is, for various reasons, in this predicament.
These questions deserve extended treatment, but perhaps we can look at one or two points now. Part of our problem comes, I think, from a misapprehension of what meaning is like. Earlier I noted that we may think the meaning of what we say is a lot more precise and definite than it actually is. We are inclined, for instance, to think that talk of emotions or motives must refer to actual determinate psychological states when it is possible that we really use such language not on the basis of internal happenings, but rather as a way of unifying complex patterns of behaviour. This possibility may seem somewhat more plausible if you consider how prone to alternative interpretations are our judgments of such matters. I may judge that A loves B while you think A is indifferent to B; both of us are tying together disparate bits of behaviour; and the question may not be resolvable by reference to what A 'feels' - she may well be deceiving herself. What then is the truth of this matter?
I do not have an answer to this question; but I raise it now to suggest that one possible answer might be that really there is no more truth here than in the evaluative case we have examined. The terms we use about each other are designed for our purposes, but as we noted long ago, telling the sober truth is not a very high priority, so they may not be contributing very much to such an aim. The truth here might not be so anthropomorphic as we would like to believe.
In general I would urge you not to make too much of a common contrast between matters of fact and matters of interpretation, since what are called 'interpretations' are often only factual accounts at a higher level of abstraction or explanatory power. But in some areas of inquiry the best we may be able to achieve is to pit one interpretation against another, without hope of any rationally persuasive resolution becoming available. This situation might arise because our concepts are not really suitable for the facts in question, or are simply too indeterminate to permit more definite progress. The concepts we use to characterize each other may well exemplify these possibilities.
Another example of misrecognition of meaning occurs in the case of abilities and aptitudes. What we have in these parts of 'folk psychology' are terms that we can use to make true or false claims on the basis of observed behaviour but which carry with them ideas that seem to refer to properties hidden from view. We use them but without having any idea of what we are really talking about. Consider different cars with different gearing: one may be ideal for driving up moderate hills, another may be very awkward. We could say the first was apt for such roads, and we would know what it is about the car that makes it so. But when we talk of a person's aptitude for foreign languages we usually have no idea whatever of what it is about that person that gives him or her that aptitude. We can also engage in debates about whether one aptitude is or is not identical with another, but since the licence to talk of aptitude carries with it no knowledge of the psychological reality, these disputes are usually insoluble. We make claims that are to some extent testable and which may be useful for various purposes, but their greatest utility for us is that we can remain in ignorance of what is going on. Again, I think people tend to suppose that they must be meaning something rather more solid than this airy emptiness, but that is virtually all analysis can uncover (cf. Brandon, 1985a).
Subjectivity and Arbitrariness
What I have been talking about is the simple truth of our evaluations, or more precisely of the evaluative aspects of the evaluative claims we make. I have argued briefly that there are no such objective evaluations or prescriptions, so that our ordinary thinking is pervaded by a kind of error here. But while I think this is a significant claim to insist upon, it must not be confused with other claims that people are often inclined to make.
As I said in the first chapter, the normal conception of what is rational or sensible or appropriate for human beings is a fairly complex one. It certainly isn't exhausted by the notion of truth. So while I have argued that our evaluations are not true or false, that claim alone should not be read as saying that our evaluations are irrational or non-rational, arbitrary, pointless or gratuitous. Some of them might be all or any of these things, but the argument so far has not impugned evaluations generally in these ways. All I have said is that, in Williams' words, 'there is no moral order "out there"' (1973, p. 29) to back up our evaluations in the way that our ordinary empirical claims to knowledge are backed up by the way the world is.
At a superficial level we can see that these other charges need not be justifiable by returning to the fact of the intertwining of factual and evaluative matters in so much of our language. Since the use of this language is partially governed by truth-conditions, we can and do carry on lengthy and perfectly rational disputes about whether or not what someone said was a lie, or whether an action was courageous or a person saintly. If we use ordinary language, we cannot escape using such vocabulary and the descriptive claims it brings with it; and there is no other language for us to use. As we have seen, our ordinary terms for grading or ranking things or people or their performances, our ordinary talk of justice and fairness in relation to rules, as well as these less specific descriptions of character or action, all crucially and centrally involve factual matters. It is partly for this reason that some apparently 'verbal' disputes are not so silly, and why it is worth protesting at some cases of the appropriation of ordinary words for technical use when evaluative aspects are treated as easily dispensable (thus one might well think Bourdieu somewhat disingenuous in claiming that his talk of 'symbolic violence' is purely descriptive (Bourdieu & Passeron, 1977, p. x) or applaud Kleinig's (1982, p. 232) complaint that behaviourists have misused the word 'punishment' in talking of unpleasant stimuli).
But while it is important to see that we cannot use all this vocabulary any way we happen to want, it is clearly not the fundamental issue. Using the vocabulary is normally to endorse the evaluations built into it, but the deeper question is whether we should endorse those values, and here again some people want to claim that they are arbitrary. But if 'arbitrary' includes in its meaning some idea that we could equally well do things differently, then it is very difficult in many cases to see how our in-built evaluations could be radically changed. We have an extensive vocabulary that incorporates a rejection of gratuitous violence between members of a social group. There is no objective moral law that tells us that such violence is wrong and to be abhorred, but it is difficult to see how social groups could survive or reproduce themselves if such a value were not fairly generally endorsed. (Notice that I am not supposing, falsely, that we totally reject violence, but that we reject what we label 'murder', if not capital punishment or the killing involved in warfare.) Again, to take a much more rarefied example, earlier on we noted the distinction between valid and invalid deductive arguments. I briefly explained the factual basis for this contrast, but it is obvious that an evaluative assessment is tied to the normal use of the terms as well. We want our deductive arguments to be valid, not invalid. But given what the factual basis is, and what our aims in rational argument and criticism are, it is not feasible for us to switch these evaluations around. Of course, if persuasion were our aim, the distinction and its associated values might well not matter; but given a particular 'form of life', the values have precious little freedom for variation.
Of course, there are many 'forms of life' available; and to the extent that they are up for choice their values are equally to be chosen or rejected. The peculiar deep-seatedness of our moral values perhaps arises from the fact that some of them, at least, arise out of features of every 'form of life' accessible to human beings. As I argued above, the condemnation of gratuitous violence among group members is likely to be an unavoidable feature of any human existence; a passion for rational criticism is clearly pretty idiosyncratic. The values in the former case are then not arbitrary, while it must be admitted that many people have got by without endorsing the values enshrined in our conceptualization of deductive argument.
I have been trying to counter the crude argument from subjectivism to arbitrariness. In many instances there is, however, one aspect of the situation that is worth a further comment. In suggesting that certain sorts of violence would be universally condemned, I didn't state precisely which kinds, or which factors would be thought relevant to excusing such violence. To take another and somewhat clearer case, let us consider family arrangements. All human groups have institutionalized some sort of family structure in which children are expected to be brought up (although many children will in fact be brought up in different circumstances in a lot of cases). But it is equally obvious that these kinds of institution vary considerably. Various pervasive biological and social facts about human beings may provide the basis for an argument that all human groups will necessarily have some such family arrangements; but those general facts cannot guide us as to which sort of family arrangement it will be. We must have some arrangement; but we cannot say which. It is perhaps even more obvious with language. All human groups speak a language, and perhaps they wouldn't be human if they didn't; but that doesn't force any particular language upon any group. When we are faced with this sort of situation, we have an interesting sort of arbitrariness: it is arbitrary which language or family institution we have, but it is not arbitrary that we have some language or institutionalized family structure or other.
Of course, we might be able to go on to other considerations that incline us to favour some such arrangements over others, so we might want to qualify the first claim about arbitrariness; but I think the general situation is very common and gives rise to many people's belief that the way things are done is fundamentally arbitrary. They may be at least half-right in thinking so. We should also acknowledge that we tend too readily to assume that our present way of satisfying the non-arbitrary demand for some arrangement or other is clearly the best that can be done.
While I have been trying to counter a too easy move from the philosophical claim about the status of our evaluations to a premature rejection of their claims upon us, it is this last point with which I think it is salutary to conclude. Subjectivism's refusal to underwrite our evaluations allows it to recognize more easily than would other views the degree to which our values cannot be given the kinds of rational support we have been looking at. Socially entrenched evaluations are often simply a disguise for brutal and exploitative power relations; they are often based on blatant falsehoods. A view that tells us that our evaluations have no more authority than the facts of our situation and our own choices give them encourages us to hope that these sorts of atavistic prejudice and disguised oppression can be rejected and replaced. A concern for the truth about these matters of opinion might well then be a liberating force, even as it proclaims that they themselves are not matters of truth or falsehood at all.
The Teacher's Stand
While a great deal more needs to be said about these matters, we have at least the outline of a position that prompts the question: what should it mean for teachers? I have argued that a tremendous amount of what is transmitted in schools is not knowledge but socially entrenched opinion; should this fact affect the way teachers handle it?
This book takes its stand within certain assumptions that make truth, accuracy, and the strength of argument central to education. So my question becomes one concerning the educator's response to the kind of points made about values and our self-image in this chapter. People who only wish to initiate children into their pre-existing societies need make no changes.
In looking at knowledge I suggested that teachers should seek not to misrepresent it, at two levels: in terms of its repudiation of common sense, and in terms of its provisional status. Another outcome of that discussion was some support for popular pressure to stress process rather than product, the strategies of critical inquiry rather than the transient results thereof - though I insisted that one cannot have the former without assuming the latter. An important feature of the processes here is that there are not really alternatives on offer - while there may not be a simply describable 'scientific method', there is a set of overlapping critical strategies applicable to any field, there are general aims of more comprehensive understanding in terms of which inquiry can be evaluated. So while teachers here need to be careful not to mislead, they can feel reasonably confident in appealing to the 'rules of the game' as their authority for fostering the skills and understanding they teach.
When we turn to evaluative matters, things are not so straightforward. Several writers correctly note the distinctiveness of, say, moral or aesthetic thought, and assume that no more is needed to justify initiation into it. If such distinctive thought also embodied simple truth, perhaps they would be right (the truths would be ones that obviously matter to us). But if it does not, we need, I think, at least to be able to establish its uniqueness or unavoidability. Just as there is really only one, albeit general, way to make cognitive progress, so we should need to be persuaded that there is only one way to play the 'moral' game.
But this is very far from being obvious. Even if we allow that some components of moral thought are required by any conceivably worthwhile social life, the recent discussion of arbitrariness illustrates how little can be sustained by such considerations. Using typically rooted moral considerations, and even more so letting one's action be influenced by such thought, would seem options, rather than necessities. And so, by what right can teachers initiate children into them?
My point is not simply that the aim should not be to inculcate particular evaluative views (which it almost always is in 'hidden' fact if not in theory) in favour of standard evaluative ways of thinking; but rather that in these areas there are alternative ways of thinking and that one of the main alternatives is not to indulge at all in the kind of thought in question - just as one important option with respect to religious thinking is to reject it as confusion.
Choosing among alternative courses of action, choosing among virtually unlimited ways of doing things (cf. Scruton, 1980) are no doubt inescapable features of human life. We should therefore give children access to possible principles to guide such choice. But just as we should not conceal unresolved disagreements about the cognitive status of such principles as components of a moral or aesthetic system (cf. Barrow, 1981, ch. 5), so we should not ignore the dispensibility of such systems themselves. We may still offer some such system because we prefer it, but let us not suppose we have any better warrant than that preference.
So whereas the teacher can make a stand on the methods of cognitive inquiry on the grounds that they are the only way to participate and contribute to the growth of knowledge, no such basis exists for insisting on any particular version of moral or aesthetic thinking. Our cultural imperialism may well be such that teachers will continue so insisting, but they are going far beyond anything an appeal to simple truth could justify.
Conclusion
So do teachers care about truth? Do you? We saw that the problem is not so much with truth in the abstract but with whether the things we teach are simply true. We also saw then, and have been considering again in this chapter, the way that what we mean can in fact be a complex mixture of different elements, some perhaps simply true, others perhaps false, or perhaps not even trying to be true. So simple answers to our question are not likely to be available.
Once we distinguish truth from what our fellows accept and press the notion of comparison built into it, it is not so easy to agree that what goes on in schools has much to do with passing on the truth. We have traditions of inquiry that have met with some success, but their tentative findings are often in conflict with common sense. This conflict is not in general stressed in schooling; it is perhaps something many of us would wish to repress. Similarly, the logical status of knowledge is often misrepresented. So schools hardly succeed in general in transmitting a defensible view of the world or of our cognitive dealings with it.
Again, despite some awareness of the distinctiveness of some evaluations, the extent of their entanglement with our ordinary concepts for characterizing ourselves is not recognized. Much is taught that represents value orientations rather than simple facts. And even when the cognitive problems of evaluations are recognized, many writers assume too readily that such ways of thinking are somehow inescapable. Once again the desirable masquerades as the inevitable.
I have suggested that educators should at least present a defensible view of the status of our knowledge and values. The views I have presented here are of course not the only ones defended by philosophers, but they indicate limits to what can be transmitted as settled. Thus one cannot correctly teach that our scientific knowledge is fixed, or our moral beliefs straightforwardly true. A concern for truth should not misdescribe the state of debate about itself.
I have indicated some possible re-orientations of our thinking and teaching about evidence or testimony and what we normally think of as the conclusions based thereon. Perhaps students can be encouraged to seek more actively for plausible alternative explanations and to take a more probing attitude to data. One can see these suggestions as yet another plea for giving a greater role to disciplined imagination in our educating. They may not be any more successful than earlier calls, but they do at least have respectable philosophical support.
I have acknowledged that a concern for truth may not be altogether comfortable; it will clash with received opinion and the guardians of the status quo; it will force choices into the open. But it may thereby promote liberation as much as it spreads misgivings. Education, as its prophets understand it, has never been a particularly comfortable or comforting process. No doubt there is much else that schools should reasonably be doing, but I think one lesson of our discussions is that when they turn to education they should be prepared for conflict.
4. Gellner, 1985, p. 126.
Originally published by Allen & Unwin, 1987, this version last revised June 17th, 2000.
URL http://www.uwichill.edu.bb/bnccde/epb/opinion.html