1. Philosophy as Scientific

In an interview Quine says: ‘I think of philosophy as concerned with our knowledge of the world and the nature of the world.’ He goes on: ‘There have been philosophers who thought of philosophy as somehow separate from science, and as providing a firm basis on which to build science, but this I consider an empty dream. … I think of philosophy as continuous with science, perhaps even as a part of science.’ The difference between philosophy and science is only one of generality. ‘Philosophy lies at the abstract and theoretical end of science.’ ‘Philosophy is abstract through being very general.’ ‘The philosopher wants to know, in more general terms, what sort of things there are altogether. Philosophy seeks the broad outlines of the whole system of the world.’Footnote ¹ In another interview Quine repeats: ‘I think of philosophy as continuous with science … [it] undertakes to analyze the general basic concepts of science … such basic notions as truth, existence, and necessity.’ And the interviewer asks: ‘That view seems to reduce the work of philosophy to simply tying up the loose ends of science. Is that accurate?’ And Quine answers: ‘Yes, I think “tying up loose ends” is a good way of condensing philosophy's purpose.’Footnote ² And again, ‘I think of philosophy as very much a handmaiden of science and something that should go along with it rather than being a separate pursuit.’Footnote ³ If philosophy is understood in this way, then it should be concerned with constructing a synoptic view of the world that combines the partial views of the various specialized sciences. This is why philosophy differs from the sciences only in generality and why it is their handmaiden.

Quine, then, thinks of philosophy as attempting to answer the time-honored philosophical question: what is the nature of the world? It is continuous with science because science is the best guide to it. But there is another time-honored philosophical question: how should we live? This is a predominantly evaluative question and the answers to it are aesthetic, moral, personal, political, or religious. If philosophy were scientific, then what would happen to all those concerns that are encompassed by evaluative questions and answers about how should we live?

Quine says very little about evaluative questions, but little is not nothing. He writes: ‘one regrets the methodological infirmity of ethics as compared with science … we can test a prediction against the independent course of observable nature, we can judge the morality of an act only by our moral standards themselves.’Footnote ⁴ But what about the moral standards themselves? He answers: ‘we do what we can with our ultimate values, but we have to deplore the irreparable lack of the empirical checkpoints that are the solace of the scientist. Loose ends are untidy at best, and disturbingly so when the ultimate good is at stake.’Footnote ⁵ He thinks that the ‘the ground of moral values’ is ‘the welfare of a society’ and ‘conflict between societies is outside of society and is thus morally neutral.’ Because of the untidy loose ends, ‘moral dilemma is rife.’Footnote ⁶ And when we face a moral dilemma we can do no better than ‘deplore the irreparable lack of the empirical checkpoints’.

This is, for Quine, an unusually mealy-mouthed way of saying what would be, if true, disastrous for us. For then all answers to the time-honored philosophical questions about how we should live would be untidy loose ends left dangling without reasonable answers after the scientifically ascertained facts are in. It follows from Quine's view that ideological and religious conflicts between societies, the distinction between just and unjust wars, and humanitarian intervention by one society to prevent the mass murders of innocent people in another are all morally neutral loose ends. If philosophy rightly done were concerned exclusively with formulating a synoptic view of the nature of the world, then the distinctions between meaningful and meaningless lives, noble and craven existence, reasonable and unreasonable hopes, great works of art and kitsch, self-control and self-indulgence, and so forth, would also be, one and all, untidy loose ends, since they would affect only individuals and have at most only statistically negligible bearing on the welfare of a society.

Other untidy loose ends are our attempts to cope with conflicts between morality and unconstrained scientific research, public and private life, seeking depth and breadth, friendship and loyalty, hope and fear, pride and shame, happiness and duty, long and short-term satisfactions, work and leisure, and dozens of other conflicts that all of us face in daily life and manage to cope with more or less reasonably. If philosophy were exclusively scientific, then much of the evaluative dimension of life concerned with the vitally important question of how we should live would be excluded from it. And, of course, so would be many great philosophical classics whose moral importance Quine must regard as only edifying at best. Quine's attitude to the evaluative dimension of life leaves him with an amateurishly simple-minded and indefensible version of utilitarianism. Something has gone very wrong.

My diagnosis of it is that Quine's reasonable claim that philosophy should be concerned with forming a synoptic scientific view of the nature of the world has turned into the unreasonable claim that philosophy should be concerned only with that. But philosophy always has and should continue to be concerned with questions about how we should live, given our understanding of what there is. This is the concern of humanistic philosophy that aims to give reasonable answers to questions about the evaluation of the human significance of what we take to be the nature of the world. The humanistic evaluative view coexists with the scientific synoptic view. Neither needs to be unreasonable. What is unreasonable is to criticize one on the grounds that it proceeds differently from the other.

Both the synoptic scientific and the evaluative humanistic view of philosophy should aim to conform to elementary requirements of reason, such as logical consistency, the explanation of relevant facts, responsiveness to criticism, justification of its own assumption, methods, and aims, and criticism of incompatible ones. Beyond these elementary requirements, however, there are likely to be many other respects in which they overlap and others in which they differ. Science is the best guide to the nature of the world. But it is no more the best guide to how we should live than aesthetic, moral, personal, political, or religious evaluations are the best guides to what there is. If pursued reasonably, scientific and humanistic philosophy are compatible. They have different aims and pursue them in different ways, sometimes reasonably, sometimes not, but always fallibly.

I conclude that Quine is partly right and partly wrong. He is right because philosophy is ‘concerned with our knowledge of the world and the nature of the world’, the best guide we have to that is science, and so philosophy should be scientific. But he is wrong in implying that philosophy should be concerned only with our knowledge of the world. It should also be concerned with asking and reasonably answering the many questions about how we should live in the light of what we suppose ourselves to know about the world. The context of those questions and answers is the evaluative dimension of life, whose concern is with the aesthetic, moral, personal, political, and religious implications of our knowledge of the world. Such questions and answers are of vital human importance. If they are reasonably asked and answered, they accept the findings of science, but go beyond them to evaluate how we should live, given what science tells us about the nature of the world. Humanistic philosophy coexists, and should be recognized to coexist, with scientific philosophy.

The synoptic scientific view gives us knowledge of the world that is independent of how it affects us. It is neither humanistic nor evaluative. The evaluative humanistic view, by contrast, is concerned with the evaluation of how the knowledge we have of the world affects us. Both have been abiding and reasonable concerns of philosophers. Quine is wrong to treat the concerns of humanistic philosophy and the evaluative dimension of life as untidy loose ends that reasonable people may perhaps acknowledge but with regret or disdain relegate to the margin of their attention as the sort of edification that is best left to preachers.

2. Philosophy as Humanistic

Our knowledge of the world is what it is independently of our attitudes toward it. The human significance of that knowledge, however, depends on our attitudes, on whether we regard the world or segments of it as, in some sense, good, bad, mixed, or indifferent. We evaluate the significance of what we know from the human point of view, informed by our modes of evaluations and the many conventions and values implicit in them that jointly form the evaluative dimension of life. And that is the subject matter of humanistic philosophy. Its components continually change in response to the changing conditions of the society that forms their context. The evaluative dimension thus varies with societies and periods. I leave ‘periods’ intentionally vague. I take them to last perhaps decades; neither centuries, nor only a few years. Regardless of such variations, they are essential parts of the life we share with others who inhabit our society during the same period.

A shared life is a bond, not bondage. Its many components are individually weak. Yet they are jointly strong, because we derive from them what we see as the possibilities of life open to us and the limits to which we are subject. They are part of the context in which we do what we can to make what we want of our life. This remains so even if we are dissatisfied with some of the prevailing modes of evaluation, conventions, and values because we think that they impoverish life or set arbitrary limits to how we might live. They may be partly or wholly faulty, and we may be partly or wholly mistaken both in our satisfactions and dissatisfactions with them.

In fortunate circumstances when the ambient social life is in tolerably good order, we typically do not pay much attention to it. We are immersed in it, much as we are in our native language. Most of the time we do not have to think about it. We just say what we want to say. And so with the prevailing modes of evaluation, conventions, and values. They are something like a language of conduct. We usually know what we can and cannot, should and should not do, how to do what we want to do, do not have to think about it, and we just do it. There are difficult situations of course, and then we have to stop and think. But if the evaluative dimension of our life is in good enough order that does not happen often.

If we routinely encounter difficulties about our aesthetic, moral, personal, political, religious, or other possibilities and limits, if we are frequently uncertain about what is and what is not in some sense good or bad, right or wrong, permissible or forbidden, then our life itself has become uncertain. The bond that we used to share with our fellow participants has been weakened. If uncertainties spread from the periphery to the center, then the serious danger looms that we will not know what many of our possibilities and limits are, how we can or should live and act, what we owe to others and they to us, and how we should try to cope with the inevitable problems we encounter.

The evaluative dimension is the key to civilized life – a life in which we have choices, know how to make them, and how to draw many of the distinctions we need to make. If it is threatened, the conditions of life as we know it are threatened. Uncertainty and insecurity loom because no possibilities are safe and no limits are generally accepted. Participating in our society's evaluative dimension civilizes how we live, we derive from it many of the possibilities we value and many of the limits we accept. It guides how we can or should live and forms the bulwark between civilized and barbaric life. That is not an untidy loose end, but an essential condition of the life whose protection we reasonably want.

The evaluative dimension is a human construct, but human constructs are parts of the world and subject to its laws and conditions. It and all that forms it are facts, just like all that is outside of it. These facts exist in time, endure for a moment or for many years, come into being, change, function as causes, and are the effects of causes. Causal chains go on with no beginning or end. As far as we know, that is how the world is. We may be wrong, of course, but there seems to be no reason to think that now. It makes no difference to this understanding of the world whether God is part of it, for that would also be part of the world. What makes an enormous difference to us – only to us – is how the facts affect us. Some facts have great significance for us, some very little, and some none. We judge the significance of what we take to be facts by relying on the evaluative dimension of life of our society and period. Embedded in it are the prevailing modes of evaluation, conventions, and values. Taken together, they are the means by which we evaluate the human significance of our knowledge of the world, determine what possibilities we have, and to what limits we are subject.

Humanistic philosophy and the evaluative dimension, therefore, are essentially concerned with human interests, whereas scientific philosophy and the synoptic view of the world are concerned with the nature of the world, regardless of how it affects our interests. Our evaluations, however, depend on whether our view of the world is true. This is not just an epistemological problem, but one of vital human interest, since how well or badly life goes for us depends on the verisimilitude of our view of the facts and on our evaluations of their bearing on our interests. The closer we come to an accurate view of the facts, the more reliable our evaluations become, and the more control we have over how we live. But our evaluations always and unavoidably go beyond the facts because they concern what we take to be their significance for us. All of this will come as stale news to those who think about such matters. But if we do think about them, we need to avoid some familiar and endlessly troublesome misunderstandings of the distinction between scientific and humanistic philosophy.

One way of misunderstanding the distinction is to confuse it with the distinction between nature and convention. There is a sense in which the synoptic view that emerges from the combination of the knowledge derived from various sciences is natural, whereas the evaluative dimension of life is formed of conventional components. But in the sense I intend, the distinction is quite different. The synoptic scientific view is meant to include everything, and a fortiori to include both what is natural and conventional. Furthermore, it is a mistake to suppose that facts are, in some sense, natural while evaluations are conventional. It is natural for us to have and live by conventions. And it is natural that, while many of our conventions are contingent products of contingent circumstances, we could not exist without conventions that guide how we should treat each other, satisfy our basic needs, procreate, distinguish between friends and enemies, make rules, raise children, and so on. The enormous variety of conventions does not alter the fact that a recognizably human life could not be lived without conventions that enable us to distinguish between possibilities and limits that are in some sense good, bad, mixed, or indifferent. Both synoptic scientific philosophy and evaluative humanistic philosophy include natural and conventional elements.

Yet another possible misunderstanding of the distinction I am drawing is to suppose that it is between facts and values. Just as it is a fact that human lives are impossible without conventions, so it is a fact that human lives are impossible without values. We must value some possibilities more than others. Were we to fail, our species would be soon extinct. It is also a fact that the satisfaction of our basic needs must be universally valued as a necessary condition of sustaining human life. The fact that we sometimes override the claims of a particular universal value because we regard some aesthetic, moral, personal, political, or religious non-universal value as more important than the satisfaction of a basic need does not cast doubt on the fact that valuing the satisfaction of our basic needs in general is a necessary condition of our continued existence. It is true that facts are one thing, and how we evaluate them is another. But it is also true that how we evaluate the facts is a further fact. It is a fact that we are conventional, value-holding, and significance-seeking beings. Perhaps there are other kinds of beings like us, but that does not change the facts about us.

Another misunderstanding I want to avoid is that scientific philosophy aims at a realist view of the facts provided by the synoptic view, while humanistic philosophy aims at a relativist view that recognizes the great diversity of aesthetic, moral, personal, political, religious, and other evaluations. All our views – scientific and humanistic – are human views. They have different aims, of course, but they are both human aims. The scientific view aims at knowledge of facts; aesthetic, moral, personal, political, and religious views aim at evaluating the significance of the facts. Both kinds of views are fallible and corrigible, but I do not think that the evaluative ones are more fallible and less corrigible than the scientific. It is just as much a matter of fact whether our many particular aesthetic, moral, personal, political, and religious modes of evaluation, conventions, and values are more reasonable than alternatives to them as whether our particular scientific theories are more reasonable than alternatives to them. The two views certainly differ about what makes them reasonable and how we might decide which are more and which are less reasonable. But that decision can and should be made within a realist, not a relativist, view. Just as scientific theories aim to be true to the facts, so our modes of evaluation aim to be true to our interests.

Lastly, the distinction between scientific and humanistic philosophy is not between what Thomas Nagel has called the objective and subjective points of view.Footnote ⁷ The objective point of view, in Nagel's understanding, is the non-perspectival view of how the world is, the view from nowhere, in his felicitous phrase. It is impersonal, impartial, and the closest human approximation of it is the synoptic scientific view. The subjective view is personal and partial, a view we each individually form from our own perspective. I agree that there are two such views, but the distinction between the scientific and humanistic views is different.

Both aim at objectivity. The synoptic scientific view aims to form an objective view of the facts. The evaluative humanistic view aims to evaluate objectively the human significance of the facts. Both may be wholly or partly correct or mistaken. It is true that the scientific view aims to be as purely cognitive as possible, whereas the humanistic view includes, in addition to cognitive considerations, also emotive and volitional ones. It is natural for us to have attitudes that include not just beliefs but also emotions, desires, and motives elicited by our evaluations of the facts. Such attitudes are essential parts of how we live. But they do not exclude objective answers to the question of whether our emotions are directed toward the right objects in the right way, whether the satisfaction of our desires would enhance or diminish our chances of living as we want, and whether our motives are consistent, realistic, or misguided.

The objectivity of such attitudes does not consist in all of us having the same attitudes. Our attitudes certainly vary widely, since both our contexts and our beliefs, emotions, desires, and motives vary. Their objectivity consists in the fact that we all have such attitudes toward the significance of the facts, even if we have different ones. It is an objective fact that part of our psychological constitution is that, barring dehumanizing extreme situations, we evaluate the significance of facts, our evaluations guide how we live, and that is not a purely cognitive matter.

The fact is, however, that our evaluations are often mistaken. It is not unusual for us to be wrong about what possibilities are open to us and to what limits we are subject. The possibilities and limits we actually have are matters of fact, what we suppose we have partly depend on our attitudes. There is a gap between the facts and our attitudes toward them, and that gap is the context in which we make mistakes in our evaluations of the possibilities and limits we think we have. Such mistakes are not mere logical possibilities, but frequent occurrences. They explain why few of us can honestly say to ourselves in the dark hours of a sleepless night that we are satisfied with how we live, that we would not like to change it in various ways, that we do not regret the major decisions we have made, and that if we could do it all over again, we would not want to do it differently. Since few of us can say this, it is of vital concern to many of us to understand what causes our mistakes.

Their causes may be psychological or social. Some psychological ones are irrationality, stupidity, self-deception, sentimentalism, and various vices. I will say nothing further about them because my present concern is with the social causes of our mistaken evaluations. Their causes are embedded in the conventions and values of the various modes of evaluation that jointly form the evaluative dimension of life in our society. Some of the prevailing conventions and values may rest on false beliefs (e.g. about races or ethnicity). We can rely on the best available scientific knowledge to correct such mistakes. But many conventions and values do not rest on false beliefs and are nevertheless mistaken because they lead us to misjudge what the possibilities and limits are in our context. They may lead us to condemn reasonable possibilities (e.g. requiring literacy for holding public office); value possibilities that may be quixotic (e.g. legislating against extra-marital sex); impose arbitrary limits (e.g. on voluntary euthanasia); and fail to set reasonable limits (e.g. on the commercialization of politics).

Many of the social causes of our mistaken evaluations are faulty modes of evaluation, conventions, and values that permit possibilities or limits that should be prohibited and prohibit those that should be permitted. If there are many such mistakes, then the evaluative dimension of life in a society is seriously defective. Dissatisfaction with it is reasonable, reforming it is desirable, and abandoning it altogether may be warranted. The problem is to find reasonable ways of identifying faulty modes of evaluation, conventions, and values. That problem is what Quine thinks of as an untidy loose end and to which Bernard Williams offers a solution.Footnote ⁸

3. The Genealogical Approach

Williams' defense of humanistic philosophy relies on the genealogical approach. Early on in ‘Philosophy as a Humanistic Discipline’ he writes that ‘one definite contrast to a humanistic conception of philosophy is scientism … which tends to assimilate philosophy to the aims, or at least the manners, of the sciences’ (182). This is what Quine does. In Truth and Truthfulness Williams explains that ‘I use a method which I call “genealogy”. It is a descendant of one of Nietzsche's own methods’ (18). And he goes on: ‘a genealogy is a narrative that tries to explain a cultural phenomenon by describing a way in which it came about, or could have come about, or might be imagined to have come about’ (20). But ‘genealogy is not simply a matter of what I have called real history. There is also a role for a fictional narrative, an imagined developmental story, which helps to explain a concept or a value or institution by showing ways in which it could have come about in a simplified environment containing certain kinds of human interests or capacities, which, relative to the story, are taken as given’ (21). Such a fictional account is ‘an imaginary genealogy’. It ‘proceeds by way of abstract argument from some very general and, I take it, indisputable assumptions about human powers and limitations’ (39).

Here are a few examples (some my own, some Williams') of genealogies: Hobbes in the Leviathan on the state as the means of escaping from the barbaric state of nature; Hume on justice in the Treatise; Hegel on history as the self-realization of Geist; Marx on history as the dialectic of class conflict; more recently Nozick on the minimum stateFootnote ⁹; Craig on knowledgeFootnote ¹⁰; and also Williams on truth and truthfulness as requirements of civilized life. Williams calls such genealogies vindicatory. My understanding of Williams' view is that a vindicatory genealogy shows that the possibilities and limits that are embedded in the evaluative dimension of life serve a function necessary for a civilized existence. This is the sense in which we cannot live in security unless we have a political organization we may call, very broadly, a state; we cannot do without rules that define and protect rights, so we need justice; and, according to Craig and Williams, we cannot do without knowledge and truth.

Genealogies may be subversive, not vindicatory. Some examples are Thrasymachus in The Republic on justice as the rule of the strong over the weak; Hume in The Natural History of Religion on religion as the product of primitive fear; Rousseau in the second Discourse on the Origin of Inequality on civilization as the corruption of innocence; and Nietzsche in The Genealogy of Morals on Christianity as the product of ressentiment, and on morality as the revenge of the weak on the strong. Subversive genealogies unmask the mistaken possibilities or limits of the evaluative dimension of life in a society by showing that their real functions are very different from what their defenders falsely claim for them. They are not conditions of civilized life, but ways of protecting the self-serving interests of their defenders.

A vindicatory genealogy shows that a rightly valued possibility or limit is a necessary condition of civilized life. This is the sense in which we cannot get along without valuing security, justice, knowledge, and truth, and condemning what violates them. Particular interpretations of the forms these valued possibilities and limits may take vary with contexts and periods, but in some form or another they are either valued or barbarism follows. Williams thinks that vindicatory genealogies show that beings like us must regard vindicated possibilities or limits as intrinsically valuable, otherwise we are doomed to barbaric life that no one can reasonably want. ‘No society can get by, (I shall argue) with a purely instrumental conception of the values of truth’ (59). ‘The concept of truth itself – that is to say, the quite basic role that truth plays in relation to language, meaning, and belief – is not culturally various, but always everywhere the same’ (61). A subversive genealogy, by contrast, shows that a possibility or a limit is valued for the wrong reason. Its real function is to serve the interest of its defenders, not to protect a necessary condition of civilized life. It has become an instrumental value and it is being misused.

Williams, then, uses the genealogical approach to distinguish between justified and unjustified evaluations of the possibilities and limits embedded in a society's evaluative dimensions of life. Evaluations are justified if they can be vindicated by a genealogical account showing that the function they serve is necessary; and they are unjustified if a subversive genealogical account shows that their function is to protect the interest of their defenders, rather than a condition of civilized life.

Thus genealogy is intended to provide an approach to the justification of evaluations that relies on history rather than on science. If successful, it shows how what Quine regards as loose ends can be brought within reason. This is why Williams says in ‘Philosophy as a Humanistic Discipline’ that ‘history is central to my argument not just because history is central among humanistic disciplines, but because … philosophy has some very special relations to it’ (180). That special relation is that just as scientific philosophy rests on the scientific method, so humanistic philosophy rests on the historical genealogical approach.

Williams, then, defends humanistic philosophy by relying on the genealogical approach to distinguish between justified and unjustified evaluations. If he is right, then the evaluative dimension of life is not the repository of loose ends that are left dangling as the serious work of scientific philosophy is being done. It is an essential part of civilized life concerned with how we should live, given what we think we know about what there is. Scientific philosophy is concerned with what there is. Humanistic philosophy is concerned with how we should live. Both are indispensable components of civilized human existence. Or, so Williams might be interpreted as claiming.

I accept that we can construct both vindicatory and subversive genealogies, but I do not accept that they would enable us to avoid unjustified evaluations of our possibilities and limits. A genealogy may reveal what we genuinely believe are our possibilities and limits, but not whether our beliefs are true. It needs to be shown that the genealogy is in fact vindicatory rather than subversive. Self-deception, wishful thinking, fantasy, stupidity, bad faith, and other psychological causes may subvert the evaluations we reach by way of a genealogy. What we identify as a necessary condition of civilized life may involve life-diminishing commitments to unreasonably permissive possibilities or restrictive limits, and adherence to mistaken modes of evaluation, conventions and values. Our commitments may be to vicious ideologies or religious practices, to irrational beliefs in magic, astrology, or witchcraft, or to being God's chosen scourge of humanity. The genealogical approach would enable us to distinguish between justified and unjustified evaluations of our possibilities and limits only if the genealogy on which we rely is shown to be either vindicatory or subversive, and, if vindicated, justified. I doubt that this can be done.

Before turning to those doubts, I want to get a pseudo-problem out of the way. The genealogical approach is not a new version of the genetic fallacy. It remains a fallacy to confuse the reconstruction of the history of anything with a justification of it. The genealogical approach is concerned with tracing the historical development, for instance, of security, justice, knowledge, or truth, but its concern is a special one, namely, to identify the social function of its subject and ask whether it is an indispensable condition of civilized life. Only if it is indispensable could its genealogical account be vindicatory; otherwise, it is not.

4. A Problem with Subversive Genealogies

The problem concerns the identification of subversive genealogies. Consider the not entirely imaginary examples of Catholic orthodoxy, utilitarian morality, and socialist ideology. Suppose that a genealogy traces the historical development of each and shows that its real function is not what its defenders claim it is. Catholic orthodoxy is intended to coerce others to accept certain dogmas as the sole means of salvation. Utilitarian morality expropriates exclusive moral authority to decide what the common good is and how it ought to be pursued. And socialist ideology is a means of seizing power and dictating policies that lead to the ideal of egalitarian justice that takes precedence over all considerations that may conflict with it. These subversive genealogies unmask the unfounded dogmatic certainties of their defenders, show that again and again they have led to persecuting dissenters, and imposing on others The Way they should live their lives.

If this were the end of the matter, these versions of Catholicism, utilitarianism, and socialism should be consigned to the dustbin of history to join countless other human follies. But, of course, the last word is far from having been spoken. Catholics, utilitarians, and socialists can acknowledge with various degrees of chagrin the regrettable excesses of the past and explain that they were caused by misguided dogmatists who should have known better. Catholics can leave the complexities of theology to theologians and endeavor to live a Christian life modeled on Christ. Utilitarians can accept a pluralistic view of the common good and acknowledge that there are deontological constraints on how the common good ought to be pursued. And socialists can resist dictatorial tendencies and opt for socialism with a human face. By means of these corrective strategies – I do not say stratagems – they defuse the supposedly subversive genealogy of what they are doing and transform it into a vindicatory genealogy that has been strengthened by correcting the mistakes that have been made in the past.

Moreover, Catholics, utilitarians, and socialists can rightly claim that they are not guilty of special pleading. The conventions and values of all modes of evaluations always have and always will be in need of post hoc corrections and improvements, since we are all fallible, and liable to make mistakes. It is unreasonable to abandon important components of the prevailing evaluative dimension of life if they can be corrected in appropriate ways. When that can and cannot, should and should not be done is usually a hard question. Answering it requires making the difficult judgment of whether the status quo can be reformed by small, incremental changes, or whether it is beyond reform and should be abandoned. Cherished modes of evaluations, conventions, and values rarely disappear suddenly. They usually sink, if they do, slowly under the weight of hundreds of small corrections.

It seems to me that this is one of the reasonable conclusions we can form if we reflect on genealogies that reconstruct historical processes. They can help us understand how and perhaps why what we now have has become what it is. But genealogies will not subvert them because they can be interpreted as spurs to correction, rather than as reasons for condemnation. Nor will they often enable us to say whether a particular genealogy on offer is vindicatory or subversive when we live with slowly changing values, conventions, and modes of evaluation. Revolutionary changes of the prevailing evaluative dimension of life are rare. What happened during the emergence of philosophy in 5^th Century B.C. Athens, Christianity, the Renaissance, the scientific revolution, and the Enlightenment are momentous but exeptional. They have all been preceded and followed by centuries of gradual changes of sensibility. In the light of the uncertainties of those who live through them about where such historical changes might lead, I do not see how most genealogies could be reasonably identified as subversive rather than corrective. Apparently subversive genealogies may become vindicatory by prompting small changes that correct whatever the defects are that all status quos always have.

The claims made on behalf of the genealogical approach, however, are far more ambitious. They aim at revolutionary changes that subvert the prevailing evaluative dimension of life, not at making small corrections to it. Genealogists interpret Thrasymachus in Republic I as aiming to subvert dikaiosyne; Hume in Natural History as wanting to do away with religion; Nietzsche in Genealogy, as bent on destroying conventional morality; and they also interpret the various genealogies of Pyrrhonian skeptics, Rousseau, Marx, psychoanalysts, Foucault, and other would-be subverters as aiming to bring about evaluative revolutions. All these interpretations, however, are inconclusive because their targets can reasonably interpret the genealogy as warranting only the correction, not the abandonment, of whatever they aim to subvert. Of course, there have been and are obviously vile and incorrigible conventions, values, and modes of evaluation that should be unmasked and abandoned. As we learn from history, however, that is not what happens most of the time.

5. A Problem with Vindicatory Genealogies

The problem is caused by conflicts between vindicatory genealogies. Let us grant that Hobbes' genealogy of the state, Hume's of justice, Craig's of knowledge, and Williams' of truth are vindicatory. Their defenders correctly identify each as serving a function that is an indispensable condition of civilized life. The state provides security, justice defines and protects rights, knowledge and truth are means of coping with the exigencies of life. Having granted all this, have we granted that the prevailing evaluations of security, justice, knowledge, and truth are justified? Genealogists can give a clear answer to this question only if they specify what exactly the justification of security, justice, knowledge, and truth involves.

If their justification amounts to no more than showing that they are indispensable conditions of civilized life, then no one can reasonably disagree with the claim that they are justified. The problem is that vindicatory genealogies can be constructed of many other indispensable conditions of civilized life, such as, for instance, of the possibility of enjoyment, freedom, health, hope, love, meaning, order, responsibility, and so forth. It is a fact of life that these and other indispensable conditions of civilized life often conflict. The claim that they are all justified because vindicatory genealogies can be constructed of them amounts to very little unless we are told how to resolve conflicts between vindicatory genealogies. Without providing an account of how such conflicts can be reasonably resolved all genealogists have done is to express the no doubt correct wish that it would be good to have security, justice, knowledge, and truth. We can accept that platitude, however, without relying on a historical reconstruction of how the present conditions of security, justice, knowledge, truth, and all the other things we value have been arrived at. If we already value them, we do not their genealogies. And if we doubt their value, then our doubt will be carried over to the genealogies that are intended to vindicate them. Who in the civilized world would nowadays take seriously a vindicatory genealogy of monarchy, primogeniture, or compulsory church attendance?

Genealogists may say that we need genealogies not to understand the indispensable function of something we value, but to understand what its function is. This, however, is not a persuasive claim. We reasonably value protection against murder, assault, theft, and fraud, and rightly prefer knowledge to ignorance and truth to falsehood without having any information about the past social conditions and the many changes of them that have led to the present conditions of security, justice, knowledge, and truth.

Suppose, then, that genealogists go beyond platitudes, acknowledge that vindicatory genealogies often conflict, and attempt to provide some reasonable approach to how we should resolve their conflicts. Whatever that approach may be, it would have to provide some reason for valuing one of the conflicting values, conventions, or modes of evaluation more than the other. But if both of them have been justified by vindicatory genealogies, then the justification of preferring one of the conflicting evaluations to the other cannot be merely genealogical because both evaluations have already been genealogically vindicated. The consequence is that as a result of conflicting modes of evaluation, conventions, and values vindicatory genealogies unavoidably fall short of the justification of whatever they have vindicated. I ask: what then is the point of vindicatory genealogies beyond the intrinsic interest of the historical account implicit in them? If the genetic fallacy is indeed avoided by the genealogical approach, as it should certainly be, then vindicatory genealogies must always fall short of justification.

6. A Problem with Justification

This problem emerges if we consider further the frequent conflicts between modes of evaluation, conventions, and values. As we derive what we can reasonably claim now to know from the best synoptic scientific view we currently have of the nature of the world, so we derive the evaluations of what we know from the modes of evaluation available in our context. Since there are many modes of evaluation, with many conventions and values, there will often follow from them quite different ways in which we may reasonably evaluate whatever concerns us. Our moral, personal, political, and religious evaluations, for instance, of security, justice, knowledge, and truth will be quite different. We cannot do without valuing each, but our modes of evaluation prompt us to evaluate them very differently, for very different reasons, and to resolve the conflicts between them in very different ways.

Consider some familiar conflicts. We value security, but if maintaining it is incompatible with protecting our freedom and way of life, then we have to decide which we value more. Same with justice. If it requires being untrue to those we love or to endanger our life, then we have to make a difficult choice. Knowledge and truth are undoubtedly good in general, but they need not be good in every particular. Is it good to know how to insult, humiliate, or torture others? or to be a pedant whose mind is cluttered with knowledge of trivial truths? Is it good to know the truth about other people's sadistic fantasies, recurring nightmares, and embarrassing social gaffe? And the unlimited pursuit of knowledge of even important truths may impoverish our emotional life, lead us on quixotic quests, and ruin our marriage, parenthood, and friendships. Seeking knowledge and truth has obvious moral, personal, political, and religious limits. Remember the parable of the Garden of Eden where the forbidden fruit came from the Tree of Knowledge of good and evil. Knowing the truth about the good is good, but knowing the truth about evil may not be good. Of course knowledge and truth are good, but there are also other no less important goods, and they conflict. When the conventions and values that follow from different modes of evaluation conflict, we have to decide what to do.

This familiar fact leads to a question that defenders of the genealogical approach can neither avoid nor reasonably answer: can ubiquitous conflicts among the values of a society's evaluative dimension of life be reasonably resolved? One possible answer is that all responses to such conflicts ultimately rest on some arbitrary preference for one of the conflicting evaluations. As Wittgenstein put it: ‘What has to be accepted, the given, is – so one could say – forms of life.’Footnote ¹¹ To which may be added his claim that ‘If I have exhausted the justifications I have reached bedrock, and my spade is turned. Then I am inclined to say: “This is simply what I do”.’Footnote ¹² Several defenders of humanistic philosophy echo this answer.

Strawson writes that ‘inside the general structure or web of human attitudes and feelings … there is endless room for modification, redirection, criticism, and justification. But questions of justification are internal to the structure or relate to modifications internal to it. The existence of the general framework of attitudes itself is something we are given with the fact of human society. As a whole, it neither calls for, nor permits, an external “rational” justification.’Footnote ¹³ Nagel says in response to the felt need to justify the evaluative dimension of life that ‘one cannot just exit from the domain of moral reflection: It is simply there. All one can do is to proceed with it in the light of whatever new historical or psychological evidence may be offered. It is the same everywhere. Challenges to the objectivity of science can be met only by further scientific history, challenges to the objectivity of history by history, and so forth.’Footnote ¹⁴ And, in remarkably similar words, Williams says of our evaluative dimension of life that ‘there is nothing more basic in terms of which to justify it. We know that most people in the past have not shared it; we know that there are others in the world who do not share it now. But for us, it is simply there.’Footnote ¹⁵

This answer – minus the rhetoric – is an admission that there is no justification of our way of life, which is ‘is simply there’. Quine and those who agree with him that philosophy should be scientific are much too polite in regarding this merely as an untidy loose end. The ‘simply there’ defense of our most basic evaluations dooms them to arbitrariness. If so, the consequences for civilized life are alarming, because basic aesthetic, moral, personal, political, religious, and other modes of evaluations, conventions, and values often conflict and their conflicts have led and are leading to the all too familiar horrors of our past and present. I note but will not dwell on this dismal fact.

What is of immediate relevance is that it follows from this answer – which is no answer – that the genealogical approach cannot be relied on to provide a reasonable justification of our evaluations. Subversive genealogies are not intended to justify anything but to warn against relying on evaluations that have been subverted by showing that they serve the interests of their defenders, rather than how we should live. And vindicatory genealogies that do aim at justification cannot by themselves be sufficient for deciding between vindicated genealogies that conflict. If one of the conflicting vindicated genealogies can be justified rather than arbitrarily accepted, then its justification must be found in an approach that goes beyond the genealogical.

The three problems I have discussed with Williams' defense of humanistic philosophy by relying on the genealogical approach, lead to a predicament that friends of philosophy will deplore and its enemies welcome. Scientific philosophy that rests on a synoptic view derived from the knowledge provided by particular sciences cannot provide reasonable answers to the central philosophical questions involved in how we should live. Humanistic philosophy that relies on the genealogical approach cannot reasonably distinguish between justified and unjustified evaluations, but only appeal to what in our context is simply there, instead of justifying or criticizing what is simply there. Is there a way out of this predicament?

7. Philosophy: Scientific and Humanistic

If the arguments I have given throughout this paper are correct, then the way out of the predicament is obvious and anti-climactic. Philosophy should be both scientific and humanistic. Of course, not at once and not in trying to answer the same questions, but in recognizing that understanding the nature of the world and evaluating how we should live are different kinds of questions and reasonable answers to them are also likely to be of different kinds. Defenders of scientific philosophy are right: science is the best way of acquiring knowledge about the nature of the world. Defenders of humanistic philosophy are also right: how we should live depends on the evaluative dimension of life that goes beyond facts and evaluates their significance from the point of view of human interests. But both are also wrong. Scientific philosophers should not exclude from philosophy evaluations that concern our vital interests and by their very nature go beyond scientific knowledge. And humanistic philosophers should not rely on the genealogical approach that cannot distinguish between justified and unjustified evaluations. The remedy is also obvious. Scientific philosophers should recognize that the justification of evaluations will not be scientific. And humanistic philosophers should recognize that the historical reconstruction of the function of evaluations is not enough to justify them.

Scientific philosophy is in fine mettle because it can rely on the scientific method that has led to great achievements in understanding the world and making human lives better. Humanistic philosophy has no analogous method on which it could rely. The genealogical approach has been proposed as such, but, as we have seen, it does not provide what is needed. The crucial question humanistic philosophers need to answer is whether there is a reasonable way of justifying evaluations that guide how we should live. I think there is such a way, and in closing indicate – no more than that – what I take it to be.

A society's evaluative dimension of life rests on widely and often inarticulately shared basic evaluative assumptions. They are typically taken for granted by those whose shared identity is formed by participation in the same society during the same period. These assumptions are embedded in Wittgenstein's forms of life, they among what Strawson calls inescapable commitments, what Nagel says is simply there, and what Williams claims are unhintergehbar beliefs. What kind of assumptions are they and can they be justified?

Some of the assumptions are beliefs: about human nature, whether it is on balance good, bad, mixed, perfectible, or malleable, and whether and to what extent we can or should try to live and act contrary to our upbringing and social conditions. Other beliefs are about the world, whether there are parts of it inaccessible to science; whether its order is ultimately benign, indifferent, hostile to human interests, or even comprehensible. Some other assumptions are about the extent to which we should be guided by reason, faith, imagination, ideology, or prudence; rely primarily on past experiences or on hoped for future possibilities when we take risks, accept or try to overcome limits; or form overall attitudes toward life and its possibilities and limits that are optimistic, pessimistic, ironic, exploratory, resigned, or rebellious. Other assumptions are about whether there is some dominant overriding ideal that we should aim at, such as truth, beauty, goodness, salvation, love, understanding, and so forth. Or perhaps the assumption is that there are many ideals, they conflict, and we should live by balancing as many of them as well as we can.

Those who participate in the same evaluative dimension of life share many of these assumptions. The prevailing modes of evaluation, conventions, and values presuppose them. Different societies and different periods have different evaluative dimensions partly because they take for granted different basic assumptions and, as a result, may be committed to different evaluations of the same facts. They can agree about the facts, even about their relevance and importance to how they should live, and yet evaluate the significance of the same facts differently. Such evaluations are unavoidably context-dependent, partly because they depend on what the basic assumptions are in a particular society during a particular period, and partly because the prevailing physical, social, international, technological, economic, and cultural conditions are different.

If this is understood, it will be recognized how hopeless it is to try to justify basic aesthetic, moral, personal, political, or religious assumptions, and consequently the modes of evaluations, conventions, and values that rest on them, by saying no more than that they are parts of our form of life, or that we make inescapable commitments to them, or that we find them simply there, or that they seem to us unhintergehbar. The futility of such feeble attempts at justification – or of denying the need for justification – comes from the recognition anyone who reflects on such matters will come to that even though the basic assumptions are ours and we make and rely on them, there are alternatives to them that are known to even moderately educated people from reading history, literature, ethnography and from travel, watching television, or living in a multicultural setting. Those who recognize this, will naturally ask what reason there is for accepting their own basic assumptions rather than alternatives to them. That question needs to be answered by humanistic philosophers, if they are indeed committed to reason. I think they can answer it.

The answer is that the evaluative dimension of life – including the basic assumptions on which it rests and the modes of evaluations, conventions, and values that form its content – is justified if it stands the test of time. I take that to be an indication that those who participate in it are satisfied with it, provided the following conditions are met. First, it endures for a period longer than a few years and lasts for at least decades. Second, parts of it slowly change as the ambient conditions change, new possibilities become available, needlessly restrictive limits are removed, and new limits called for by new adversities are set. It stands the test of time because it is flexible enough to correct unavoidable and all too frequent mistakes about possibilities and limits and because it is open to prudent incremental changes. Third, participation in it is voluntary. Participants have exit rights, they can exercise them if they wish, but a large majority of them do not because they are by and large content with the prevailing evaluations. Fourth, their contentment is not the result of indoctrination, manipulation, coercion, or ignorance of alternatives. Participants are aware of having other options but they do not take advantage of them because the satisfactions they derive from the prevailing evaluative dimension of life outweigh the natural dissatisfactions that result from being constrained by limits.

This justification, of course, is only the sketch of an ideal and a possibility whose countless details would have to be filled in. Furthermore, even if that were done, no evaluative dimension of life could be completely justified. The conditions I have described are always imperfectly met. Sometimes circumstances change rapidly and the efforts to respond to them are slower than they can or should be. Participants are not motivated only by reason and good will. Crime, poverty, ignorance, prejudice, and internal and external enemies are also parts of the conditions that must be faced and possibly corrected or at least ameliorated. Standing the test of time is a desirable possibility, not the description of any society that has ever existed or is likely to exist in the future. But its possibility is also the possibility of a reasonable justification of the prevailing evaluative dimension of life that humanistic philosophy can provide.

The contents of this justification change in response to the passage of time, shifting conditions, the extent to which participants comply with the limits, and the available possibilities are rich enough to provide a sufficient variety of different forms lives can take. And, of course, the prevailing modes of evaluations, conventions, and values will conflict and the conflicts will often be basic and difficult to cope with. But there is a standard that can be appealed to in trying to cope with even particularly troublesome conflicts: the shared interests of the participants in protecting the entire evaluative dimension of their lives. Would this be a good justification if its details were filled in? It seems to me that it would be good enough to enable us to come as close to living the way we want as our conditions, fallibility, weakness, and perversity allow us to do. It would be an imperfect justification for imperfect societies and imperfect people, but it would be a justification.