Jump to ContentJump to Main Navigation
The Quality of Life$

Martha Nussbaum and Amartya Sen

Print publication date: 1993

Print ISBN-13: 9780198287971

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0198287976.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 05 August 2021

Objectivity and Social Meaning

Objectivity and Social Meaning

(p.165) Objectivity and Social Meaning
The Quality of Life

Michael Walzer

Oxford University Press

Abstract and Keywords

Walzer defends a qualified form of relativism in which much attention is paid to the various forms of criticism and discontent that may arise in society. However inarticulate these dissenting voices may be, argues Walzer, they are to be taken as important data in understanding what a good life would be for society as a whole. Walzer stresses that even if no particular meaning was objectively true or right or necessary, it would still be the case that the construction of meaning is a real process, and ‘objectivity’ is tenable only to the extent that it hangs on the accommodation of the object or social meaning by a knowing, inquiring subject.

Keywords:   construction of meaning, discontent, objectivity, perspective, relativism, social meaning, subject


I probably do not have an objective view of objectivity. Having been accused so often of disdaining it, I come to it now with some trepidation; I want to make a cautious approach, repressing for a while the uneasy sense that the conjunction in my title misrepresents the likely outcome of my argument. Let me begin with a strong, simplistic, and usefully wrong definition of objectivity: a given perception, recognition, or understanding can be called ‘objective’ if its content is wholly or largely determined by its object—so that a range of human subjects, differently placed, with different personalities and different, even conflicting, interests, would agree on the same content so long as they attended to the same object. The table determines the objective perception of the table. What makes for objectivity is simply this: the object imposes itself. The subject is passive and undiscriminating, a promiscuous consumer of available ‘data’.

For reasons philosophers have long understood, that cannot be right. Human beings are active subjects. Our faculties of perception and cognition help to determine whatever it is that we finally see or recognize or understand. But we are still inclined to call the perception ‘objective’ so long as these faculties are so widely shared as to constitute what we might call a normal subject. Then perception is objective when it is jointly determined by the object and the normal subject. If someone without depth perception reports on the existence of a table different from the one the rest of us see, his is the subjective report. The table and the normal person looking at the table (who represents ‘the rest of us’) together determine what the table objectively is (looks like). The object still imposes itself, but perception is conditioned by the character of the receptive organism, and the idea of ‘objectivity’ incorporates the results of this conditioning.

But that cannot be right either, and this time for reasons that have generated a long series of complex, difficult, and sometimes high‐flying philosophical arguments. We do not come to the object with faculties alone but with interests and ideas too. And what we see, recognize, and understand depends (with a strong but not absolute dependency) on what we are looking for, our cognitive concerns, and the ways we have of describing what we find, our conceptual schemes. Given our concerns and our schemes, what opportunity is there now for the object to impose itself? We seem armoured against imposition, shaping the world to our own purposes.

But I do not mean to surrender objectivity so quickly—indeed, it is the scientific perception of the world, driven by a strong purposefulness and (p.166) structured by elaborate and highly speculative schemes, that makes the most insistent claim these days to be called an objective perception. The claim takes many different forms, but in all its forms it must hold that if the object does not impose itself it is still recalcitrant to conceptual and purposive impositions. Scientific concepts must accommodate the object—not as the object appears, perhaps, but as it really is. I am not going to comment on that last assertion, except to say that for most of us, at least, appearance is an important aspect of reality. But I want to accept the claim that objectivity hangs (somehow) on the accommodation of the object by a knowing, inquiring subject. The knowing subject shapes the object, but he cannot shape it however he likes; he cannot just decide that a table, say, has a circular or a square shape without reference to the table. Similarly, someone self‐confidently applying a conceptual scheme that divided the world into friends, enemies, reading matter, and edible plants would get the table wrong (objectively wrong), or he would miss the table entirely, and deny its reality, and that would be a merely idiosyncratic (subjective) denial.

This is still a very simple account of objectivity, a rough, commonsensical approach to philosophical difficulties that lie beyond my immediate congnitive concerns and possibly also beyond the conceptual schemes at my disposal. But the account works, more or less well, for simple objects‐in‐the‐world. The question that I want to ask now is whether it works at all for objects to which we assign use and value, objects that carry ‘social meanings’.1 This term, borrowed from anthropology, seems to cast a cloud over all claims to objective knowledge. Social meanings are constructions of objects by sets of subjects, and once such constructions are, so to speak, in place, the understanding of the object has been and will continue to be determined by the subjects. New sets of subjects learn the construction and then respect or revise it with only a minimal accommodation of the object. The object may or may not limit the constructive work in which they are engaged. Obviously, the table cannot be constructed as an intercontinental ballistics missile. But it can become a desk, a workbench, a butcher's block, or an altar, and each of these can take on meanings to which the ‘mere’ table gives us no positive clue. Can perceptions of objects like these, objects‐with‐meanings, ever be called objective? It is easy enough to imagine situations in which one person's altar is another person's butcher's block. But we do accept reports on social constructions. Now objectivity (in the reports) hangs on an acknowledgement of the construction. Our shared understanding of what an altar is, what we have made it for, determines our perception of the table‐that‐is‐an‐altar. The holiness of the altar is similarly (p.167) objective, since it is part of the same construction. All normal persons living within the system of social meanings would deliver similar reports on the objective reality of tables‐that‐are‐altars‐that‐are‐holy.

But this may go too far. Suppose that there are dissenting voices within the society where some tables are holy altars, people who deny the construction, who announce, ‘There's nothing there but an old table.’ That is also an objective report of a kind. Can we say that it is an incomplete report, that it misses something of real importance? Imagine a fuller report: ‘Some people claim that it's an altar and treat it as if it were holy, but there's nothing there but an old table.’ No incompleteness now, and now only the disagreement can be objectively reported: ‘I think there's nothing there but an old table.’ Nothing in the nature of the table will lead us to say that it is or is not a holy altar. The altar is objectively there only for those who understand it to be objectively there. It is holy only for those who acknowledge its holiness. And what they will have to say, if they are to report objectively, is that it is holy for them.

Believers will want to say more than that. They will want to say that God has sanctified the table and made it into a holy altar; hence everyone who knows how things really are in the world will acknowledge its holiness. But I shall take it as given that altars and holiness are alike human creations. The believers are wrong, then to take the holiness of their altar as a universally recognizable (objective) fact. The altar is holy only because and only in so far as they have made it so. With regard to such creations, the rest of us are not bound by majority rule; only the voice of the people as a whole resembles the voice of God. Social constructions must reflect a general agreement—or, better, since no vote is ever taken, there must be a consensus—if there is ever to be an unqualified objectivity, an objectivity without pronouns, in our reports about them. (Reports from outside observers will always need pronouns: ‘their altars’ or, in more extended form, ‘these tables, which they use as altars’.) The more complex and specific the construction the more surprising it is when a consensus is actually reached. The social processes that make this possible are mixed processes, involving force and fraud, debate and consent, long periods of habituation; overall, they remain mysterious.

Compared to ‘altar’, ‘table’ is both uncomplicated and indeterminate; hence its meanings will rarely provoke significant or stirring dissent. Someone who says, ‘That's not a table’, while pointing to a flat piece of wood with a supporting structure of appropriate height, will probably lead us to talk about mistakes, not disagreements. (I will not stop here to imagine bizarre cases of table‐like objects that are rarely not tables.) We would suspect some failure of normal understanding. A table is indeed a social construction, as well as a physical construction, but the socially constructive work is so rudimentary that we are unlikely to recognize much more in it than the assignment of a general name to the object. And then we expect people of normal understanding to remember the name. Nothing much follows from remembering it; the construction of the flat piece of wood, etc., as a table does not require (p.168) us to use or to value the table in a certain way. More specific constructions, by contrast, have normative consequences.

Tables‐that‐are‐altars‐that‐are‐holy must be treated in accordance with certain principles and rules. I cannot use the altar, for example, as a desk on which to write profane essays on social meaning—not because the altar will resist the use or God strike me dead, but because it would be wrong to do that given what an altar is in my society (for me and my fellow members). Nor can I chop it up for firewood, even if the church is very cold; or trade it for personal profit—a new suit, say, or a season ticket to the opera, or a place on the Stock Exchange: it would be the wrong thing to do. But would it be objectively wrong? It seems to me that it is possible to ask questions like that too soon. Clearly the rules of use and value are not determined by the ‘mere’ table; nor are they jointly determined by the table and a normal person looking at the table; nor do they represent an accommodation of the table by a knowing subject or a scientific observer. The rules follow from the social construction of the table as a holy altar and they would seem to be objective rules only for those men and women who join in the construction or acknowledge its results. The other might be bound by some notion of ‘decent respect’ for the opinions of their fellows, but not by the idea of holiness.

But perhaps we can go a little further than this. If we think of the holiness of the altar not as an isolated construction but as one feature of a more complex whole, a cultural system or a way of life, then the force of the rules is considerably enhanced. Imagine the table‐that‐is‐an‐altar‐that‐is‐holy within a set of connected constructions: socially meaningful occasions (holy days), spaces (churches), officials (priests and bishops), performances (religious services), texts (scriptures, prayers, homilies, catechisms), and beliefs (theologies or cosmologies), and the result is something from which individuals cannot so easily opt out. Some day there will be alternative occasions, spaces, officials, performances, texts, and beliefs, arising out of a long process of social change (‘secularization’, say); and then people will be able to explain to their fellows why the altar is not (really) holy. But now the refusal of some dissident or rebel to treat the altar in accordance with its rules of use and value is probably not a straightforward denial of its holiness but a specific act of desecration—literally, an effort to reverse the process through which this particular altar has been consecrated. And the religious rebel committed to desecration is likely to appeal, in much the same way as the early Protestants appealed, to other features of the existing cultural system or way of life, features that give him reasons, so he says, for what he does. The system as a whole still has objective value for him; he lives within the set of social constructions. Where else can he live?


We can still ask whether this is objectively the best place to live. But I want, again, to postpone that question in order to explore more fully the crucial (p.169) implication of my argument thus far: that social construction is also moral legislation. The meanings with which we invest objects have normative consequences. I have been calling these norms ‘rules of use and value’; they are also rules of distribution, that is, they regulate our relations not only with things but also with other people. Any number of philosophers have argued that morality is a human invention, writing mostly as if what we invent are the rules that govern a moral life. We leap to principles like equality before God or personal autonomy or the greatest happiness; and then we make lists, like the Decalogue. Maybe we do that, sometimes; but the thickness of the moral world and the density of our relationships suggest a radically different kind of invention. One of the ways we reach that thickness and density is through the social construction of objects (of all kinds). Social construction makes for a complex and rich world, many features of which will seem so obvious to us that we will not be prompted to ask whether they are, of all possible features of all possible worlds, objectively best. They will have a more immediate objectivity. So we will use and value objects in accordance with the meaning they have in our world, and we will exchange, share, and distribute them in accordance with their use and value. We will know what objects we owe to other people as soon as we understand what those objects (really) are and what they are for. And a great part of our conduct towards other people will be governed by these distributive entailments of social meanings.

At this point, it will be useful to take up another example, even though the difficulties of the table‐that‐is‐an‐altar‐that‐is‐holy have by no means been exhausted. I want to consider the construction of a human life—not a biological but a social life, not a life span but a life course in a particular society, namely, our own. What we have constructed is a life‐that‐is‐a‐career‐that‐is‐open‐to‐talents. Obviously, there is nothing in the nature of a human life that determines its construction as a career. Any given version of the life course is conditioned by the life span, so that youth, maturity, and age give rise to a pattern like training, professional practice, and retirement; but these latter three do not by themselves constitute a career. A career is an individual achievement; it is constituted by choice and qualification. Though career patterns may be collectively established and repetitively enacted, a career is none the less a projection of the self into a chosen and uncertain future. What makes this projection possible is the opening up of certain sorts of places and positions (professional or bureaucratic), which I will call ‘offices’. Offices are the objects of careers. The social construction of the two goes hand in hand, like altars and offerings. If careers are open to talents, then offices must be distributed on meritocratic principles to qualified persons. If we imagine individual men and women planning careers‐open‐to‐talents, we must also imagine competitions for office. If there are competitions, there must be rules protecting the competitors, not only against violence but also against discrimination, that is, against any refusal to attend honestly to their qualifications.

Once careers and offices are in place, nepotism becomes a wrongful practice. It would be wrong for me, the member of some search committee, say, to (p.170) favour my brother over a more qualified candidate. It does not matter that I have a very strong, and to my mind overriding, belief in family loyalty; I am caught up in a complex set of social constructions that has normative entailments. Someone who fails to respect the table‐that‐is‐an‐altar‐that‐is‐holy does no injury to the table, and in so far as the failure is private he does no injury (causes no offence) to other men and women either. His is one of the minor sins. But once constructions determine distributions, private refusals make for a more serious wrong. And when behaviour is in question, general agreement is no longer a necessary condition of rightness or wrongness; the rule against nepotism, for example, is binding even on individuals who argue that offices are family holdings and not objects of careers. There will not be many people, however, who will actually argue that offices are family holdings—except in the unlikely case that a strong familial idealism is part of the same set of social constructions as the career‐open‐to‐talents, and then we might well recognize the public nepotist as a conscientious objector. Acts of refusal and opposition commonly have a basis of this sort, in the coexistence of contradictory constructions. Then people have to choose, guided only by their best understanding of the complex social world they inhabit.

I want to stress that it would not be objectively wrong to adopt the argument for offices‐as‐family‐holdings. Majority rule does not govern arguments about social meaning; it only governs behaviour. The rules of behaviour, then, are objectively right relative to the prevailing meanings, but the prevailing meanings are not objectively right (or wrong). They are only objectively there, the objects, that is, of more or less accurate reports. The life‐that‐is‐a‐career could, over time, be constructed in an entirely different way, and offices could be reconstructed to match the difference, and no wrong would have been done. It would not be the case that lives or offices had somehow been misunderstood; nor would the men and women leading lives and holding (or not holding) offices under the new dispensation have been treated unjustly.

It is not my claim that the whole of morality is objectively relative (relatively objective?) in this way, only whatever part of it is entailed by the social construction of objects. Even here we might plausibly ask whether there are cases where construction is jointly determined by its objects and its human agents in such a way that the same normative entailments appear again and again, in all or almost all human societies. Then the same behaviour would be wrongful for the same reasons in all human societies; morality would lose its particularist character without ceasing to be relative to social construction. The easiest case has to do with the things we call ‘food’: given the human body, the construction of edible objects is not an entirely free construction—though people in different cultures do choose different things to eat and not eat, edibility itself is (in part) socially determined. In any case, the experience or expectation of hunger and the possibility of eating certain things work together to turn some of those things into human provisions, and it would seem to follow from this that provisions should be provided for those in need; (p.171) food belongs to the hungry. (Who should do the providing and at whose expense are questions not so easily answered.) More complex and specific constructions will still be culturally relative: we save certain food for festive occasions or we burn it before the gods or we waste it at extravagant banquets. But the original construction of things‐that‐are‐food‐for‐the‐hungry entails certain distributive rules that have, I suspect (this could be checked), always been recognized. Hoarders in time of famine act wrongly, for example, given what food is for.

I shall assume that reiterated social construction rather than diffusion from an authoritative centre is the preferred explanation for the appearance of identical or similar used and valued objects in different societies. There is no authoritative centre, no Jerusalem from which meanings go forth. The list of similarly constructed uses and values, then, constitutes what we might think of as a universal and objective morality—relative to social construction where construction repetitively takes the same form, relative to the prevailing argument where the same argument always prevails. We could go on to further explanations: if certain things‐in‐the‐world are constructed in the same way again and again, presumably there is something in the nature of the things and/or something in the nature of the human agents that accounts for the construction. As the example of food suggests, the account is likely to be a naturalistic one. But I doubt that the list of similar constructions would be very long; nor would it include the complex and specific constructions that make for the thickness of moral life: food for eating would get on the list, not food for offerings. This is what it means to say that complexity is free: the more complex the construction the more room there is for cultural difference. Complex constructions do not turn up again and again, and they do not have plausible or satisfying naturalistic explanations.

There is no universal model for social construction, and the range of difference among actual outcomes is very wide. It might be argued, however, that this is so only because the constructive work takes place under a great variety of adverse and advantageous (mostly adverse) conditions. Only a common necessity, like the need for nourishment, makes for sameness. But if we imagine social construction in ideal conditions (and if ideal conditions are a single set of conditions), then we will get a model outcome, that is, a free construction that is at the same time the best construction. I am afraid that this is an impossible dream. For we can replace actual with hypothetical social construction only if we know, and not hypothetically, what conditions are ideal. And if we know that, then we already know the model outcome. We simply pull into our account of the imagined (ideal, original, natural) conditions all those materials, and only those materials, out of which we want society constructed. We might as well draw a blueprint of the good society and give up the idea that construction is free.

(p.172) [3]

But if we do not have a model outcome, how can we ever criticize actual outcomes? This question is motivated, I think, by a misunderstanding of what an actual outcome is. Social construction is first of all conceptual in character. Holy altars and careers‐open‐to‐talents are ideas, and the distributive norms that follow from them are also ideas. These ideas are never more than partially instantiated in the world; holiness and openness are more often than not honoured in the breach. What social critics commonly do is to hold the idea, or some more or less elaborated interpretation of the idea, over against the instance of the idea. Or, just as commonly, they hold some other idea or complex set of ideas, also the product of social construction, over against this idea and its instances. They say, if careers are open to talents, then why are they not open to the talents of Jews, or blacks, or women? Or, if our society is a union of families or a democratic and co‐operative community of citizens, how can we tolerate the intense competitiveness generated by careers‐open‐to‐talents?

Criticism of this sort depends on objective values, where objectivity is a true report on social meaning. The criticism itself, however, is not objectively true or false, for it also depends on an interpretation of social meaning, and interpretations are (except at the margins) only more or less persuasive and illuminating. But surely there are times when we want to say something stronger than this. We want to say that though the report is objectively true, the meaning is wrong (and not just wrong for us). Or we want to say that that is not the way we ought to think about altars or careers or whatever. Or even, that is not what an altar or a career really is.

Is it possible for a whole society to get things wrong in this fundamental way? This is the question that I have been postponing, and it is time now to try to deal with it. I want to make sure, though, that we understand exactly what the question is. Clearly, it is possible for individuals within a society to get things wrong, even fundamentally wrong, and it is also possible for groups of individuals to do the same thing. We should think of the Nazi case in these terms. It would strain the imagination to describe a fully elaborated world of complex meanings of a Nazi sort; in any case, no such world has ever existed. Within German or European or Western culture, the Nazis were an aberration, and in so far as we can make out their distributive principles—air for Aryans, gas for Jews—we can readily say that these are objectively wrong, immoral, monstrous. All the resources necessary for a judgement of this sort are already available, the products of a long history of social construction. It is a great mistake to make of the Nazis a hard case. The hard case comes when we begin to think that a long history of social construction has somehow gone awry.

Consider, then, those societies where women (all women) seem to have been socially constructed as objects of exchange and where rules of exchange follow from the construction. I will not attempt an internal account of the exchanges (p.173) that actually take place or of the meanings attached to them. Perhaps our understanding of an ‘object’ and even of an ‘exchange’ is not available to the participants. All I will say about the social construction at this point is that women are transferred among households, from one patriarchal jurisdiction to another, as if they were objects of exchange. What should we think of that? Are exchanges conducted in accordance with the rules objectively just?

There are a number of possibilities here: either women have or have not played a part in the constructive work; either they agree or they do not agree to its outcomes. Or, in a language less marked by our own conceptions of moral agency and lives‐as‐careers, they acquiesce or do not acquiesce, go along or do not go along with the outcomes. If they have played no part and do not go along, then the exchanges cannot be described as just. We can only report on the disagreement. Or perhaps we can say, as I would be inclined to say, that the exchanges are unjust, because in this case the objects are also human subjects, capable (as tables and lives are not) of going along or not, and the resistance of the constructed object nullifies the construction. It does not matter if the resistance is inarticulate, passive, hidden, or private. So long as we can in one way or another discover it, so long as we have probable cause to believe in its reality, the social construction fails.

The women involved may or may not be able to describe themselves as persons‐engaged‐in‐social‐construction; the vocabulary that I have deployed here is presumably not their own vocabulary. But we can see how their resistance ‘works’ in the world and why the construction fails. The unanimity or consensus principle plays a part in explaining this failure, but something more is involved. Constructions of persons are not free—and not only in the obvious sense that we cannot make women or men into intercontinental ballistics missiles. The theory of social construction implies (some sort of) human agency and requires the recognition of women and men as agents (of some sort). We might say, looking at the idea itself as something we have made, that the construction of social‐construction‐with‐human‐agents has certain moral entailments. Among these is the right of subjective nullification, the right of the agents to refuse any given object status—as commodities, ‘hands’, slaves, or whatever.2

But what if women, for whatever reasons, actually agree that they are objects of exchange and live willingly by the rules of exchange? The phrase ‘for (p.174) whatever reasons’ conceals a problem here, which philosophers who are quick with hypothetical examples are prone to ignore. What reasons could these women possibly have? We can easily see the reasons they might have for concealing disagreement, for stifling anger, for expressing resentment only in private or only in the company of other women. But if the experience of being treated as an object of exchange is the sort of experience we think it is, and if the women being exchanged are beings like us, what reasons could they have for agreeing? If, on the other hand, the experience does not match our understanding of it, and if these hypothetical women are beings of a different sort, then what is the philosophical issue here? What can we say, why should we want to say anything at all, about experiences and beings of which we are entirely ignorant?

Still, let us accept the hypothesis in its strongest form: here is a society in which women really do agree to the construction of themselves as objects of exchange. They do not agree because they have been brainwashed, because some chemical process or some hitherto unknown social process has turned them into moral robots or made servitude a reflex—for then, whatever they did or said, it would not constitute agreeing. Nor do they agree because they have no choice or because they are physically coerced or because they find themselves in desperate difficulties from which agreement is the only escape—like the women who sells herself into slavery in order to feed her children. For their agreement in these circumstances would not count as the construction of themselves as objects of exchange; it would represent only a reluctant and resentful acceptance of a pretence, a role that they could not refuse or escape. We must imagine reasons of a different sort: that the exchange of women brings some benefits to at least some women (even if the benefits are much greater for men); that it is only one part of a larger pattern of relationship, fitted to a system of beliefs, symbolically represented, ritually enacted and confirmed, handed down from mothers to daughters over many generations. So women accept the construction, even participate in it. What normative consequences follow?

One possible response is that no consequences follow, for agency is inalienable. This is Rousseau's argument, not applied by him to the self‐subordination of women but obviously applicable: ‘To renounce liberty is to renounce being a man, to surrender the rights of humanity and even its duties . . . Such a renunciation is incompatible with man's nature; to remove all liberty from his will is to remove all morality from his acts.’3 Since human beings are agents by nature, and necessarily responsible for the worlds they make, the surrender of agency simply does not count; it is a gesture without effect. The argument from social construction is harder than this since it cannot refer to a universal and unconditioned moral agency. Now agents are socially produced, themselves involved in the production. It is still true that we (with our perceptions, (p.175) understandings, theories) can recognize the‐woman‐who‐is‐an‐object‐of‐exchange as a social construction with moral entailments only if we also recognize the same woman as a moral agent capable of agreeing (or not) to the construction. She can only be (morally) an object if she is simultaneously a subject confirming her object status. She is constituted by a contradiction—in so far as her subordinate status depends (morally) on her own agreement or acquiescence and is therefore inconsistent with subordination itself—and therein lies her freedom. She can never become just an object of exchange; the proof of this is that if she ever repudiates her object status, she is immediately and wholly a subject; the rules of exchange instantly lose their force. But so long as she confirms them (and even if her confirmation takes, as it commonly will, some other form than explicit agreement), they retain their force: she is partly an object.4 There is nothing in the nature of a woman, or a man, that rules out contradictions of this sort. (The case is somewhat similar, I think, to Kant's means–ends polarity. We do not have to treat every person we meet, on every occasion, as an end‐in‐himself, for persons can agree to be means, like good civil servants who make themselves into the instruments of their fellow citizens, even surrendering some of their civil rights. But they can always resign from instrumentality.)

So long as the woman‐who‐is‐an‐object‐of‐exchange confirms her object status, the contradiction in her being is an objective contradiction. We can give a true account of it. Someone who claims that she is wholly an object is wrong. But so is someone who claims that she is wholly a subject (this would be roughly analogous to insisting that we must always, on every occasion, treat the civil servant as an end‐in‐himself). This last turn of the argument may well seem to many readers too relativistic, a surrender to what Marxists call ‘false consciousness’. But once we have ruled out brainwashing and coercion, I see no morally acceptable way of denying the woman‐who‐is‐an‐object‐of‐exchange her own reasons and her own place in a valued way of life. That does not mean that we cannot argue with her, offering what we take to be better reasons for the repudiation of (what we take to be) object status. It does mean that, once the argument begins, she has to choose what she thinks are the better reasons, without any certainty as to which ones are objectively best. But we can say, and this seems to me all that we should want to say, that the choice is truly hers.

Is this not a plausible account of social construction seemingly gone awry? If nature provided a blueprint for construction, the process would not go awry as often as it does (seems to do). If something like gender equality were (p.176) a simple entailment of the constructive process and every internal contradiction were ruled out a priori, then arguments for equality would be much easier than they are. When we encounter a complex set of social meanings, we enter a moral world, and it is no tribute to the creators of that world to deny its reality. Social meanings are constructed, accepted, and revised for reasons, and we have to engage those reasons. When we engage them from the outside, as in the case of women‐who‐are‐objects‐of‐exchange, we are like missionaries preaching a new way of life to the natives, and we would do best, morally and politically, to try to work out what they find valuable or satisfying in their old way of life. More often, and more importantly, criticism of the old ways comes from within, as the result of long processes of social change. For the construction of objects‐with‐meanings and thus of moral worlds goes on and on; it is a continuous process in which we are all engaged. Conservatives try to freeze the process, but that effort is only one more instance of constructive activity (it has its reasons), one more expression of human agency. Criticism is no different in form.

Consider, for example, the construction of lives‐that‐are‐careers‐that‐are‐open‐to‐talents in a society where women are still objects of exchange. Over a period of time, institutions and practices take shape that make it possible (or necessary) for some members of the society, mostly men, to plan their lives–and, over the same period, lives of this sort are discussed, argued about, rendered meaningful. In the course of this process women will find that they have a new reason to repudiate their object status, for only by doing so can they undertake careers of their own. Some of them will seize upon this reason, and then more and more of them; at some point women‐who‐are‐objects‐of‐exchange will be relics, sad memories, their agreement to subordination hard to understand. If a few people try to act out the rules of exchange, they will appear quixotic, not so much defenders of old ways as fools of time. In similar fashion, an archaeological guide might say to us: ‘These were the holy altars of Xanadu, in the days when holiness reigned in the city.’


I come back at last to my initial reservations. The kind of objectivity that I have attached to social meanings is probably not the kind that philosophers seeking objectivity are interested in. They are in search of things as they really are or as they must be. But I know very little about things as they really are, social construction apart. It is true (we can give objective reports) that particular constructions are reiterated in one social setting after another. The extent of the reiteration and the reasons for it–these are empirical matters. Exactly what evidence would lead us to say that such and such a construction could not or should never be otherwise, I do not know; in any case, that sort of evidence will not often be available. Interesting objects, all the more complex (p.177) constructions, can always be otherwise. Tables need not be altars; lives need not be careers.

But is it not objectively true that meanings are always constructed? Men and women who claimed to have discovered meaning in nature, say, would surely be misreporting (misconstructing) their own activity–as if they were telling us that it was not Adam but God who named the animals. Even if no particular meaning were objectively true or right or necessary, it would still be the case that the construction of meaning is a real process. Men and women really have made tables into holy altars and lives into careers. This has been the presupposition of my argument, and I have even pointed to its possible moral entailments; I do not want to run away from it now. But it is a strange ‘objectivity’ that leaves us adrift in a world we can only make and remake and never finish making or make correctly.

Note: In this paper I have tried to sketch an account of ‘social meaning’ that might underpin and uphold the theory of distributive justice presented a few years ago in my book Spheres of Justice (New York: Basic Books, 1983). My views on objectivity have been guided, stimulated, and provoked by recent philosophical and anthropological work that I can only acknowledge in a general way by listing a few crucial books: Hilary Putnam, Reason, Truth and History (Cambridge: Cambridge University Press, 1981) and The Many Faces of Realism (La Salle, IU.: Open Court, 1988); Nelson Goodman, Ways of Worldmaking (Indianapolis: Hackett, 1978); Thomas Nagel, The View from Nowhere (Oxford: Oxford University Press, 1986); Clifford Geertz, Local Knowledge (New York: Basic Books, 1986); and the essays collected in Rationality and Relativism, ed. Martin Hollis and Steven Lukes (Cambridge, Mass.: MIT Press, 1982) and in Objectivity and Cultural Divergence, ed. S. C. Brown (Cambridge: Cambridge University Press, 1984). I am grateful to Ruth Anna Putnam, Alan Wertheimer, John Goldberg, and Thomas Nagel, who read the paper in an earlier version and told me what was wrong with it. Martha Nussbaum suggested that I write the paper and provided its title, but she is responsible only for its existence, not for its argument.


(1) Are there any objects without social meaning? Perhaps the phrase ‘simple objects‐in‐the‐world’ names a null set. But I am going to assume that there are such things, which we accommodate and shape directly, without any necessary reference to their sociological significance. Stones are, for my purposes, simple objects‐in‐the‐world—until they are made into cornerstones, tombstones, grindstones, milestones, stepping stones; or doorsteps (or, more dramatically, until they are used for coronations or set up as markers for a sacred history). As for tables, see the argument below.

(2) A friend writes in criticism of this ‘right’ that some participants in any social system ‘will resist or resent or reject the position or identity they are given . . . Not all resistance nullifies.’ So we have to decide in each case whether the resistance is legitimate or not, and this requires standards that are extrinsic to the business of social construction. Yes, we will have to decide whether the resistance is just a way of avoiding particular obligations incurred by particular men or women. Agency itself cannot be denied, and promises made by agents are not subject to unilateral repudiation. But the denial of agency can always be repudiated, with the consequences described further on in the text. The right of nullification is simply the agent's right to claim her agency against any social process of objectification — and it does follow, I think, from the view of objectification as the work of human agents.

(3) Jean Jacques Rousseau, The Social Contract, trans. G. D. H. Cole (London: Dent), 9.

(4) I do not mean to say more than this. I do not mean that the construction is right because the woman confirms it, only that it is effective and consequential in the moral world. Her agreement (or acquiescence) had evidentiary, not legitimizing, force. Agreement makes for rightness only within moral systems where it is understood to do that, and in such systems it is commonly hedged with qualifications as to the freedom of subjects, the knowledge available to them, and so on. Hence women‐who‐are‐objects‐of‐exchange can be exchanged justly or unjustly; but the objectification itself is not justified by their agreement.