Bernard Williams, Truth and Truthfulness Princeton: Princeton University Press, 336pp. £19.95


Bernard Williams enjoys such preeminence as a moral philosopher that it is easy to overlook his interests and achievements in other philosophical areas, including metaphysics, epistemology, and history. These other interests are splendidly on display here, although there is still a sense in which the book is centrally a contribution to moral philosophy. It shows all Williams’s characteristic virtues. He is always a pleasure to read, and as it has often done before, his deft, sparkling, intelligence newly illuminates an old philosophical subject, scattering light into many surprising corners as it does so.

            In general analytical philosophy has had rather an unsatisfactory relationship with the topic of truth. On the one hand there is considerable hostility to those descendants of the sophists who pronounce themselves suspicious of the very idea. Williams calls these the deniers, recognizing that there is often some indeterminacy in just what they are denying. At any rate, they include the familiar relativists, postmodernists, ironists, and debunkers who are supposed to skulk in other corners of the academy, such as departments of literature, history, languages, and cultural studies. Williams shares this hostility, and his book is in part a rebuttal of the deniers. Yet, in spite of this overt hostility, it is important to realize that for the last half-century analytical philosophy has tended to offer nothing but comfort and ammunition to these subversives. Sellars attacked the idea of anything being simply given in experience, while Goodman and Kuhn attacked the objectivity of any scientific theorizing we might do on the basis of experience. Quine attacked the idea of determinate meanings and determinate interpretations of other people, while Wittgenstein attacked, or was taken to attack, the existence of determinate rules for the application of terms and also gave grateful postmodernists the playful idea of a language game. Austin sensitized people to the variety of activities involved in the use of language, making contact with the subversive idea that all speech is political speech. Williams does not take us through this piece of history, but to many it must surely look as though denial is the order of the day, and Protagoras has finally got his revenge on Plato.

            Nevertheless in the analytical tradition, perhaps surprisingly, truth survives intact. It does so mainly by keeping its head down. This posture is known as deflationism, and was fathered by the venerated founder of analytical philosophy, Gottlob Frege. The basic deflationist point that you get nowhere by adding the terms of truth. It makes no difference whether you say that fish swim, or it is true that fish swim, or really true or a fact that fish swim. So, it seems, there is nothing in talk of truth of which we need to be suspicious. Truth is itself too little to be a subject of doubt and denial. The right response to Protagoras was not some ambitious piece of Platonic mythology, but a homely reminder that it is true that fish swim if, and only if, fish swim. So there is no doubting or denying anything about the truth of the claim which goes beyond doubting or denying what fish do, and to settle that we need not metaphysics but ichthyology.

 So far so good, but it hardly suffices to tell the deniers that there is nothing of which to be suspicious, since they are apt to reply that this is precisely their own point. They can welcome deflationism, simply adding the warning that people falsely think they are adding something important when they lay claims to truth, particularly something in the nature of dignities or titles to authority, and it is the hollow nature of those titles that needs exposing. The high-priest of the deniers, Nietzsche, said that there are no facts, only interpretations. But he could have made the same point in deflationist times by allowing that there are plenty of facts, but then insisting that to identify anything as a fact is itself to make an interpretation, and indeed this is a good way to think of Nietzsche’s mature position. And then it is the situatedness or contingency or fluidity of those interpretations that is taught by Sellars and the others, at least as effectively as it is by any of Nietzsche’s more flamboyant descendants.

            Williams’s strategy is not to tackle these motivations head-on. Rather, his route into the minefield is via an attempt to sketch a genealogy of truth. The notion of a genealogy is itself associated with Nietzsche, but Williams’s paradigm example of its successful application is Hume’s account of the way the exigencies of human life make it natural or inevitable for us to form some concept of justice. A genealogy, in this sense, is something like a “just so” story: given some fairly basic facts about the human condition, perhaps thought of as a kind of state of nature, we can see how natural it would be for some practice to emerge. Williams argues that in order to do this work the state of nature need not be thought of historically, and he sharply distinguishes the method he is pursuing from any kind of evolutionary psychology. The story is intended make us more comfortable with our concept, and this is precisely its potential for discomforting the deniers. By analogy, before Hume a denier might have contended that justice is nothing but a mystery or a fetish, but afterwards that is not so easy to do, since Hume explains how it has its feet firmly planted in the human soil. So an explanation of this kind has the potential to vindicate the practice it explains, although, as Nietzsche’s own story about the emergence of religious and moral thought illustrates, it also has the potential to debunk it. The question will be whether we are proud or ashamed of the features of humanity that are displayed in the unfolding of our practice.

            The basic element in Williams’s vindicatory story is our need to act successfully in an environment, about which we must therefore gather information. Yet a solitary individual can only gather so much information by himself. It is much better to be able to use other peoples’ signals as an extension of one’s unaided observations and inferences. Pooling information requires a division of labour, and we then need something from other people. If we are to find out how things stand, for instance at a place where they have been and we have not, we need them to possess the two virtues of accuracy and sincerity. Accuracy is the virtue of reporting how things are only when the report is the outcome of sufficient investigation to make it likely to be true, and sincerity is the virtue of communicating only how you take things to be. In this manner the virtues of acquiring and disseminating information seem inevitable in any human culture, however much other functions of speech crowd in. Furthermore the real exigencies of not getting eaten, or finding water or food or shelter scarcely encourage the idea that these reporting uses of language are self-contained, or illustrate the free play of signifiers, or only make up an activity that ought to be called a game. 

It might sound from this as though we are being offered an account of the function of truth and truthfulness that is both reductive and utilitarian, two words of round condemnation in Williams’s writings. But Williams stresses that genealogical stories are not reductive. There is nothing in them to suggest that an interest in truth is “really” something else, such as an interest in utility or social solidarity, as some pragmatists would claim (Williams enjoyably retails Sidney Morgenbesser’s supposed remark: of course pragmatism is true, the trouble is that it just doesn’t work). The trouble with someone who is inaccurate or insincere is certainly not that he is out of step, or not “one of us”. In the primitive scenario, the trouble is that if you rely on things being as he says, you risk your neck.

The story does not prevent an interest in truth from developing a life of its own, quite apart from practical applications, just as sexual desire has a life of its own quite apart from any faint interest we may have in propagating our genes. But in neither case does the genealogy shed very much light on how that independent life comes about. It is easy to see why we value accuracy and sincerity when we listen to reports of where the tiger went, but harder to see why we should when the question is one of the shape of the universe, or the lineage of the King. Evolutionary psychology is not refuted by, say, sexual jealousy among the elderly, but it cannot easily be said to predict or explain it, and similarly curiosity, accuracy, and sincerity might seem to be merely quixotic when directed at things which have no practical importance. This leaves it open that a curiosity about such things, and an extension of accuracy and sincerity to our dealings with them, are themselves optional, and indeed options that were only picked up at specific historical moments.

            Before that there is, however, a different problem. A genealogy of truth faces the same problem that bedevils social contract theories, which is the difficulty of finding an uncontaminated starting point. A theory of the origins of contract must not imagine people already endowed with a respect for promises, for that is to be the outcome of the process. Similarly a theory of the origins of truth must not imagine up-and-running notions of belief, judgment, or representation, since each of these is integrally connected with the idea of getting something right. This leaves it quite hard to see just what is legitimately allowed at the beginning of the story. Some kind of distinction between being right and being wrong, whether we are making an observation or an inference, is so firmly embedded in our thinking that it is hard to imagine starting yet further back. We would be at a point in human life at which there is no distinction between something appearing to be so and it being so, which is sufficiently far from us to be unrecognizable as human life in the first place. At the very least, the story would have to be one about the emergence of self-consciousness itself, and while it is not impossible that there should be such a story, Williams veers away from the daunting task of providing it.  So in the event, his project is not so ambitious as it may sound. It is not so much the emergence of the very idea of truth that concerns him, as the emergence of signaling to each other how things stand, which requires the virtues already mentioned.

As Williams insists the genealogy does not by itself guide us very far in many practical deliberations. It does not, for instance, give us a measure of the stringency attached to duties of diligence in finding the truth, nor of sincerity in telling it. Williams explores these themes in the central part of the book. Unlike grammar, truth does not just “carry us along”: in ordinary speech we do not ordinarily have to decide whether to speak in accordance with English grammar, but we do frequently have to decide which things to say or conceal, and which truths deserve the effort of discovery. Politically as well, we have to come to turns with the extent to which the supreme aim of government, the safety of the people, subordinates its duty to inform, or to refrain from misinforming, the people. Williams pursues these themes by developing his well-known opposition to Kantian absolutism, and he is particularly interesting on the relation between sincerity, thought of as an ordinary, everyday, disposition not to say what one does not believe, and authenticity, a much more ambitious, and contentious, ideal. He tellingly illustrates its inadequacy as a measure of virtue by the sad case of the paranoid Jean-Jacques Rousseau, who seems to have thought that since he knew himself to be utterly candid and transparent, the constant misunderstandings of him that were made by other people had to be due to their willful malice.

            The final chapters of the book take up the themes of historical truth and political critique. As we have seen, historical truth, beyond very simple reports like “a tiger just went in there”, is not a given in the state of nature. The success of human action is determined by how things are now, and only indirectly by how things were some time ago, about which myth or fantasy might do just as well, as indeed they seem to across large parts of the world. So Williams describes the emergence of historical truth as itself a specific historical process, occurring in the west between the time of Herodotus and that of Thucydides. He stresses, however, that the historian’s goal of making sense of some historical development goes beyond merely setting out the truth about it. A chronicle of events may recount nothing false, but it cannot ever be complete, and it will not include the interpretative thread that would turn it into a historical narrative. Here, Williams concedes, the relativists or debunkers have a point, or perhaps two points. First, perhaps as Tristram Shandy discovered, there is indeed no such thing as ‘the’ truth, or what Rorty might call the book of the world, in which all and only historical truths are to be found. To describe at all is to select, and any ideal of completeness is bogus, since the world will always permit further description in yet other terms. And secondly, there is no such thing as making sense, tout court. A narrative makes sense to some audience at some time in terms of categories that are themselves culturally variable and mutable. This mutability raises the thorny problem of relating our own best understanding of historical agents (or indeed, any Other) to their own self-understandings, and the status of any assumption of uniformity that enables us to be optimistic about doing so—staple exhibits of the deniers and debunkers. But it does not mean, as has been said, that historical truth has had its day (itself a historical pronouncement about which Williams is extremely funny). It only implies that the intelligence involved in making sense of events goes far beyond any simple or passive fidelity in reporting them. But fidelity to how we take things to have been remains a sine qua non: these concessions to the deniers suggest no playful, or sinister, downgrading of accuracy and sincerity.

             There are many other topics Williams touches, and that flourish under his touch: aspects of Greek philology, the concept of a marketplace of ideas, the idea of trust, the relationship between liberalism and truthfulness, and others. He is consistently amusing, but at absolutely no cost to the depth of the enterprise. And what a wonderful life it would be if even a small proportion of philosophers could write so well.


Simon Blackburn


2511 words.