Review Philosophy

Identity’s Difficulties

About: Vincent Descombes, Les embarras de l’identité [Identity’s Difficulties], Gallimard.

by Stéphane Haber , 1 May 2014
translated by John Zvesper
with the support of Institut du monde contemporain

Télécharger l'article : PDF

What do we actually mean when we say we want our identity to be recognized? How much do our identities depend on the choices we have made? How are collective identities constructed? These questions are addressed in the work of Vincent Descombes, who although acknowledging the multiplicity of our affiliations tends to give priority to the national one. Stéphane Haber’s review here is followed by Decombes’ reply.

Reviewed: Vincent Descombes, Les embarras de l’identité [Identity’s Difficulties], Paris, Gallimard, 2013, 282 pages, 21€.
Stéphane Haber’s review is followed by a response by Vincent Descombes (see below)

One of the standard moves in modern social and political thought during the last two centuries has been to reveal the falseness of individualism. The argument goes: when we think that individuals owe the essence of what they are to themselves, and that they should think primarily about themselves, to look after their rights and the conditions of their flourishing, we are thereby erring both cognitively and morally. Vincent Descombes’ position is far removed from the authoritarian outlook that can be seen as an implication of this standard account of individualism’s faults. He accepts that individualism, as the desire to live ones own life and to distance oneself from the conformity demanded by institutions and social groups, is now the atmosphere in which we live and breathe.

In Descombes’ view, individualism is an essential part of our existence and our self-understanding – and it is clear that this “our” has long since ceased to refer only to the west. No alternative vision of humanity (empowering tradition, religion, nationality, gender, and so on), claiming to reactivate a past form of life, has any chance of receiving lasting attention. In fact, on the level of principles, there is nothing with which to oppose individualism morally. You cannot blame it for the countless disorders linked to selfishness, stupidity or malice. So anyone intending to question individualism (and it is here that we can begin a philosophical investigation) must simply and calmly observe those moments when individualism unknowingly and unwillingly pays homage to its others – to the forms of life that precede and support individuals, i.e. to things these individuals have not chosen.

That is what Descombes sets out to do by examining the meaning of the vocabulary of identity. He thinks that people who talk about their “identity” as something interesting and vulnerable that deserves to be protected, promoted and recognized by others, are placing themselves within the framework of modern individualism. And indeed, in prevailing linguistic usage, identity appears primarily as a good belonging to a subject. It goes with the process of an individual’s personal affirmation, the work on oneself by oneself that defines a life that is singular and is attached to remaining so. But whoever talks about identity thereby assumes that the continued process of individuation feeds on our first attachments, which we appropriate and which are reflected in our lives: “identity” has become a reference point for self-definition by individuals who situate themselves in relation to the historical communities to which they belong, and which are older than they.

This paradoxical situation, which is reflected in political movements that make claims to identities, is philosophy’s point of departure.

The Identity of Objects and Subjective Identity

Descombes pursues a philosophical investigation intended to cover all of the most significant contemporary uses of the term “identity”. He reminds us that historically ontology has used the term identity to designate identity to itself: the first characteristic of a thing is its being identical to itself and, with certain conditions, remaining so with the passage of time. Such a position naturally raises several questions: should this identity be conceived as an inherent property of the thing, or rather as a kind of convention imposed by the mind? From the outset, Descombes argues that identity, though it does not present itself as a timeless essence, does have a kind of reality; it is not a phantasm. To put it positively: in the first place, identity means giving to something a name, which is one of the most meaningful human practices.

The theme of naming leads to questioning the status of personal identity. In modernity, this identity has become a completely original phenomenon, because of the ascendancy of the vocabulary of “subject” and “subjectivity”. Talking about subjectivity essentially comes down to assuming that only I can discover or do certain things, because of the fact that I possess a mental interiority. This singularity actually defines me as a subjective I. “My” identity thus becomes something very different from the identity of any thing: it is based on the fact that I am conscious of myself. More precisely, what is then supposed to make me exist as a coherent and unique person is memory, the internal capacity of identifying oneself and establishing continuities among remembrances. Self-awareness becomes the basis of the identity (to itself) of a self that is aware of its actions. Although it is fair to take on board the originality of the phenomenon of personal identity, focussing on self-awareness and memory leads to insoluble paradoxes that philosophers have dwelt upon for some time now. For example, do forgetfulness and amnesia to a certain extent involve ceasing to be the same self? To escape such paradoxes, we have to return to the theme of subjective identity, which deserves to be questioned on a different basis.

Twentieth-century existential philosophies are the only ones that have tried to evade the expensive privilege of self-awareness, and for a time Descombes follows their basic inspiration. In these philosophies, what I am is not independent of what I want to be, in the sense that a series of my existential choices outlines the approximately unitary course of my life. Thus, what I am is not independent of everything that manifests something of me in the world. Being a “self” or a “subject” – being “identical to itself” in the sense of ipseity – is all related to the substance – the concrete directions – which I commit myself to and maintain. The continuity of a human existence is an affirmed, “willed” continuity (or at least one that is accepted and gone along with), in that it is a lateral element inherent in the normal psychological continuity that means that what I am and what I do are not constantly questioned and interrupted: character traits crystallize, habits form, personal styles stabilize. Descombes writes: “What is subjective is what comes from an individual and says something about this particular subject because it expresses him or her: not in the sense that it reveals or perhaps betrays them, but in the sense that it is this subject who by his or her act is expressing himself or herself, as if he or she were talking in the first person” (pp. 118-119).

But doesn’t saying that the singular individual that I am defines itself in its relative permanence (its “identity”) by something coming from the will (a view that some existential philosophies have radicalized during the past century) perpetuate the myth of radical liberty or even self-creation? Is it not in any case very problematic to base subjective identity on the “choice of oneself”?

Perhaps; but at the same time, “choosing oneself” does constitute a specific kind of experience that is not at all mysterious. In Descombes’ view, it is specifically a historical experience. We must even say moreover that we moderns are encouraged to choose ourselves after we have questioned (in identity crises) facile conformism and definitions of the self lacking any element of reflexivity. “Only a modern individual can live through an ‘identity crisis’ and come out of it by ‘constructing his or her identity’. Thus, Taylor talks about our ‘modern identity’. To use the expression ‘modern identity’ is to ask that the idea of identity (in terms of moral psychology) be explained in a comparative perspective: we have to show how modern men respond in their own particular way to a question always confronting all men” (p. 136).

It is characteristic of modern society that it furnishes individuals with a list of typical experiences to go through, and of words to use when expressing an identity crisis, for example during adolescence or early youth. We are even encouraged to question our social affiliations. But of course that encouragement is part of us being socialized individuals, who have to learn certain language games, and to internalize a number of what are presented as typical experiences. The modern person can be defined as one who is socially called upon to pretend, if only for a specific period, not really to belong to a society. But this desocialization is nevertheless secondary. No matter how profound our identity crises, we at no point cease to remain social beings, i.e. individuals who have had to learn and to reproduce language games and social habits. “Subjects are surprised to discover that they can find the reasons why they chose to be modern only by having without reasons already chosen this modern identity” (p. 165). Once we have reached this crucial point, by pursuing Descombes’ approach, we can proceed very quickly, because we have now recovered the territory of social habits. By generalizing the scope of the “sociological” conclusion to which we have been led, we arrive at the idea that individuals do not choose themselves; we have to say, less dramatically, that their only choice is to endorse such and such specific ends. So really, to speak absolutely, individuals start nothing; they adhere to ways of doing things that are already given. Only within society does one individuate oneself and acquire one’s own identity.

The Primacy of the Collective

In this sense, there are many collective identities that are not merely linguistic conventions. These identities form the natural correlate of the process of individuation by which each individual acquires an identity in a society, by linking up with ends and habits that are already there. Of course, these collective identities neither constitute nor reveal any particular substance. A collective identity is primarily what must be minimally assumed when people wonder who they are and understand that they are acting in a way that has been known before them. For this implies that they accept being the same as they or others were in the past. Descombes thus clearly takes a “modern” point of view, in which identities (whether individual or collective), rather than prevailing on account of their intrinsic stability, actually appear only in the light of questioning and problematizing (in short: of identity crises): who am I? ; who are we? To get our bearings when responding to this questioning, it is enough for others to recognize us as being in a certain respect the same as before, and that we endorse this relative permanence, whatever conclusions we decide to draw from it. Saying “we” is always a matter of addressing someone whom we do not include in this “we” and whom we place ourselves opposite from.

For Descombes it is not a matter of defending on this basis a substantialist concept of the “we” and of “historical communities” as examples of that concept. His semantic approach lets him say that it is primarily a matter of subjects, in the grammatical sense, but also in the practical sense of activity. As factually given sets, these communities act, and we have to understand them in this perspective. They do not, then, enclose individuals in some conformist destiny; so it would be inept to understand them as constraints that one could and should free oneself from in order to become oneself. Yet at the same time, social belonging is deeply resonant. For example, it is a psychological fact that I feel concern if one of mine is attacked – i.e. one of those whom I often include when I use the word “we”. This fact and many similar ones suggest that the assertion of a “we” is neither superficial nor contingent. It seems as significant as the capacity to say “I”. In other words, communities exist, and they are primary.

It is clear how distant Descombes’ account is from the perspectives of Will Kymlicka [1], who asserts the existence of an individual’s right to flourish in conditions of life that respect the historical, ethnic, and cultural attachments that determine each and every one of us; and of Axel Honneth [2], for whom the struggle for recognition is the main life force in collective existence, as if existing communities, merely by existing and thus creating limits and routines, signified nothing but inertia and alienation. From Descombes’ point of view, the apriority of human sociality and historicity is not really appreciated in those perspectives.

But what are these historical communities? Some contemporary political issues point us in the right direction. For example, it is easy to see the relevance that “identity politics” can have here. There is only one meaning that legitimately explains its substance: as members of a minority and/or ostracized group, by committing to making a claim, we are saying that our ways of living deserve to be recognized and respected. But by whom? The only possible response is: by our co-citizens. Rightly understood, the act of making this claim consists of demanding forcefully that reasonable arrangements be adopted so that the way of life and the habitual practices that we represent can be given new attention, in order to make changes in existing habits and legal norms. This act presupposes the existence of a community – meaning both having things in common and the society as a whole – between those who are demanding and those to whom the demand is being addressed. So this claim is addressed by one “we” to another, more encompassing “we” in which the former is still included; otherwise the process would be reduced to adopting a position of unproductive exteriority, similar to the one taken by people who talk about us elsewhere. Therefore, logically there is a priority of belonging, of homogeneity within a historical-political group, and of the “shared meanings” (in Charles Taylor’s words) that flow from this. Diversity comes afterwards.

Descombes is not here taking a position in favour of a “strong” concept of republicanism along the lines of the French tradition of the state, treating as invisible – and even pursuing a policy of actively reducing – the cultural and ethnic differences that make up the community of citizens. While not emphasizing the limits and difficulties of that approach, he clearly perceives them. In fact for him, in contrast to some advocates of republicanism, it is not abnormal or alarming sometimes to claim one’s identity and difference. His analysis takes place at a different level, more analytical and conceptual. However, he does advocate a distinct principle that is not unrelated to the republican vision. In societies like ours, he says, fundamental historico-political affiliations, with their communitarian dimension, are not like other affiliations. They provide the background of all the various affiliations that we experience in our lives and make claims about. From the philosophical point of view, “identity” politics and ideas are therefore less interesting in themselves than because they let us rediscover the primacy of co-citizenship and of related collective forms.

On this basis, Descombes suggests that we distance ourselves from discourse about the diversity of affiliations or belongings – discourse that has become widespread, and has been refined by such writers as Amartya Sen. This discourse is too levelling. Admittedly, we are not made from just one piece; we do not define ourselves in relation to just one large group. However, we immediately sense that the fact that X is of a given nationality and the fact that X, having a particular taste or subjective preference, is part of the group who shares that preference, are not at all on the same plane. This example is crucial. In the real world, very different in this respect from postmodern intellectual constructions, the fact of nationality (the fact of being a citizen of a certain country) does count, both symbolically and factually. In Descombes’ book, it even forms a kind of fixed point, around which orbit the other personality components and forms of life to which we attach ourselves. The nation is at the head of the list of the historical communities that we look to. Descombes emphasizes this, reminding us of the arguments of Marcel Mauss and Louis Dumont. Significantly, it is here that his approach is the closest to empirical hypotheses (in contrast to his own register, “grammatical” elucidations). At any rate it is here that the theme of the “historical community” begins to take on a more distinct shape in his account. Which is perhaps the source of some difficulties.

The Resourcefulness of the Theme of Social Construction

To explain those difficulties, we can start with the premises of the reasoning. Without identifying his targets, Descombes talks about “constructivism” as an extreme view in which identities are artifices, going back to fictions imposed by mass manipulation. This view is his favourite adversary. In fact for him it is a caricatural denial of the thickness of collective identities. His philosophical critique of “constructions” attacks the theoretical intention of methodological individualism, which would cynically deprive us of the conceptual means to conceive at their proper level the reality of the collective and the historical.

But on reflection, there is nothing very ridiculous in saying that “social identities” as they are conceived and imposed, functioning for example as justifications for discriminatory policies or simply as support for commonplace prejudices, are “constructions”: after all, people and institutions do engage in building them, disseminating them, and securing their intended practical effects. Of course, not all identities fall into this category that is so thoroughly explored in feminist and “post-colonial” theories. So we should agree with Descombes not to let our awakening critical faculties crush our historical discernment. Not all collective identities have been from the outset manifestations of our alienation or evil ruses by powerful elites. Not all traditions have been “invented”, in the conspiratorial sense of the word. Generally things are much more ambiguous than that. In formerly colonized countries, the way in which borders (and consequent identity awareness) were marked by a variety of circumstances that had remarkably little to do with tranquil continuities in the history of forms of life passed along down the generations is very instructive, but things did not always take place along those lines. Taking issue with Descombes, we can wonder whether in a philosophical approach it really makes sense to insist on isolating, and even on idealizing as it is, the fact of nationality – the manifestations of which, after all, have been and remain extremely diverse – as the thing that in the last resort illustrates and even validates the thesis of the importance of historical communities in particular, and of the primacy of the social in general. That is a doubtful approach. It is difficult to carry on as if the alternative were seriously one between a sombre warning about the “irreducibility of the fact of nationality” and a false philosophy of asocial man. It is wrong to suppose that we could get very far in an analysis based on this too crude alternative.

We must also wonder whether Descombes’ bad-temperedness about “constructivism” is sound. In contemporary psychology, discussion of “social categorization”, showing how during concrete interactions “identities” are related to assignments and to stereotypical constructions, is actually rather perceptive [3]. Likewise when Ernest Gellner, in Nations and Nationalism, shows how nations and nationalisms in modern Europe should be seen historically as correlated with a specific phase of politico-economic globalization: there has been not only a kind of organic formation affecting cultures and practices, but also a process steered by states aiming to establish some relatively stable areas in a world with increasing volumes of displacements and population flows. So identities clearly are largely “constructed”, in many very interesting senses, and without being considered as inconsistent lies. Asserting that they are given (or that they have to appear in philosophic reasoning only as if they were given) amounts to making a choice that is more than slightly dodgy, ontologically speaking.

Finally, a more general remark: human sociality – a significant dimension of which is clearly the dependence on groups, on forms of life, and on institutions – has many faces. Although we do not choose our communities, neither does the way in which we individuate ourselves by internalizing some parts of our environment correspond to an unambiguous model known in advance and involving clear hierarchies. Granted, in its manifest substance, contemporary identity discourse reflects considerable influence by historical communities (for example religious or national) on human life, even when individualism has had an impact. And one can completely agree with Descombes that this discourse, reasonably understood, shows paradoxically the persistent ascendancy of the state and the nation. But it is not clear that this identity discourse is the best thread to follow when trying to understand individuation in general. What makes me a unique person possessing a particular identity, my own character, and an original biography, is a combination of circumstances, relationships and sedimentations, which figures only partially in that discourse. Perhaps we should not ask noisy political ideologies to furnish us indirectly with the key to human existence.

Vincent Descombes replies

I fully agree with the idea that Stéphane Haber expresses at the conclusion of his review. The meaning of a human life cannot be reduced to the matter of belonging to a given community. Not only do I agree with this idea, I think I have explained how the political form of a modern society makes this dissociation of the individual and the collective possible, through the principle of secularism, which separates concern for the public safety from concern for personal salvation: the government of the city does not have to look after my salvation as an individual.

This principle of secularism is part of the constellation of values that sociologists call individualistic, to bring out the contrast with the holistic value constellations of traditional societies. As Haber emphasizes, here the word “individualism” is to be taken in the descriptive sense. It has nothing at all to do with condemning a lack of good citizenship in our contemporaries; it aims rather to describe the concept of the human being that we fashion for ourselves, insofar as our thinking is based on our communal ideas.

What we call “identity issues” arise for us in the intellectual and moral context of these individualistic communal ideas. Only human beings who conceive of themselves as individuals in the normative sense of the term – beings ultimately responsible to themselves – can go through the “identity crises” described by social psychology.

Modern human beings define themselves as individuals (in the normative sense). However, they have not thereby stopped living in societies. So there arises an inevitable tension between the fact of their belonging to various collectives and the ideal of autonomy. On this point, Haber has clearly seen that the sociological point of view (which is mine) does not in any way imply that individual autonomy is a chimera. It would be inept, he writes, to understand social belonging as constraints on individuals, or as enclosing them “in some conformist destiny”. There are philosophers who understand the words “social” and “collective” necessarily to mean “inertia and alienation”. They fail to understand that, after Durkheim and Mauss, and also after Wittgenstein, institutions should be conceived not as limitations on individuals’ possibilities, but quite the opposite. If there were no set practices of the game of chess, with their established rules, I would not have the possibility of playing chess. So the rules of the game are not “constraints” that interfere with the freedom of the player. Likewise, the fact that languages exist does not take away from individuals their expressive possibilities, this fact is on the contrary the condition of all their verbal inventions and poetry (granted, in their particular language).

Haber observes that my investigation of identity issues – “who am I?”, “who are we?” – starts with a clarification of the very idea of identity in the logical sense, from which idea has been drawn the principle of every entity having an “identity to itself”. He writes: “the first characteristic of a thing is its being identical to itself and, with certain conditions, remaining so with the passage of time.” I would oppose this very traditional way of stating the identity of a thing – a way that wrongly suggests that identity is a characteristic that a thing possesses – and that it could perhaps lose. In this view, attributing an identity to a thing would be to describe it, to characterize it in its status (for example, in its permanence, or stability, or internal cohesion). I would say that to understand identity in this way is to head off down the wrong road. We would see this if we were to ask ourselves: alright, we are attributing a property of identity to the thing – but to what thing? But of course in asking what is the object that is the referent in our discourse, we are already applying to the object in question the concept of identity. We are identifying this object, before we have got to describing it. The identity of an individual – the fact that he or she is identifiable, and that one can refer to him or her – must not be confused with a quality of resemblance. So for example, when I recognize that I am the same sex as one of my two parents, the identity in question is that of this sex, not that of my person.

Therefore, I intentionally refrain from entering into the traditional debate that Haber touches upon, about whether identity is an “inherent property” of a thing or just a perceived property. This debate is not relevant, because identity is simply not a property or a quality. It is better to explain the logical concept of identity very differently, starting with the logic of naming. And this lets me apply to all attributions of identity the “Geach rule” (named after the philosopher who has best clarified this idea of Wittgenstein’s). In saying that a thing remains the same, we seem to be describing it, but in fact the words “is the same” or “remains the same” make no sense unless we can complete them: is the same what? remains the same what? We have to insert after the words “the same” an individuating term (Quine), i.e. a general substantive that supplies us with a counting principle, which analytical philosophy calls a criterion of identity. From birth to old age, I change a lot, but I remain nonetheless the same human being.

This logical point may seem far removed from issues of collective identity and from the political debates that they inspire, but we have to take it on board in order to clarify the elusive and ambiguous idea of “affiliation” or “belonging” to a social group. As Haber rightly points out, I make a clear distinction between belonging to a purely classificatory grouping – for example, the set of those who like a particular kind of music – and belonging to a real group: a country, a school, a political party – i.e. to anything that can be seen as a collective body (in the sense of something that gives birth to an “esprit de corps”, to particular ways of thinking and feeling). However, the reason for this distinction is not that these affiliations do not have the same importance in the eyes of the individual, but that the word “affiliation” is equivocal: it can mean belonging to a purely nominal group, with no identity other than that of the elements of the logical class defined by the shared attribute (for example, liking the same music); or it can mean belonging to a historical community, with its own collective identity.

This point enables me to respond to the concerns expressed by Haber towards the end of his review: doesn’t criticizing “constructivism” in sociology imply disregarding the role that “stereotypes” and “categorizations” play in human relationships? In fact, I recognize collective identity only when we can identify a real group, i.e. when we can apply to it a criterion of identity drawn from its definition: the same school, the same party, the same town, the same orchestra, the same people, etc. In contrast, the “stereotypical constructions” alluded to by Haber are not collective identities, they are merely similarities among individuals with a shared trait. They define nominal groups, not real ones. They do not inspire the question “who are we?” in the collective sense of the word “we”, the sense that introduces a relationship between the particular will and the general will, as Rousseau puts it. They correspond to no historical totality, only to collections of individuals.

With regard to those real groups, the question is raised whether they are “given” or “constructed”. It seems to me that we should reject this question, for the reasons given by Castoriadis: in the exercise of power by instituters, we must recognize a circle involving both the instituter and the instituted. As Castoriadis says, an institution is neither created ex nihilo nor passively received: “everyone is an ‘author’ of the evolution of language, family, morals, etc.” [4]

Haber wonders why we should privilege affiliations of nationality: isn’t that idealizing what is just one affiliation among others, which moreover has historically taken a variety of forms, sometimes rather unfortunate ones? He also and rightly points out that there are many possible ways for individuals to prioritize their purposes in life, and therefore also their multiple affiliations.

True, but among their affiliations, is there one that appears to them comprehensive and in this sense not optional? To be compatible with the ideal of individual autonomy, this comprehensive affiliation must be defined on the ground of human will. Therefore the comprehensive community must be defined by a political rather than by a religious attribute. The principle of secularism comes in here. Now, what form can a community of citizens take as soon as it sets out to bring together modern individuals?

From a political point of view, the question we should be raising is the shift from the “I” to the “we”. The modern individual intends to respond for herself or himself, both on the small scale of his or her own person (“I”), and on the enlarged scale (“we”). How can these modern individuals say “we” and thus submit their particular wills to their general will? They have to enter into co-citizenship with each other. Will they do that in a universal republic or within the limits of a nation state? There is no objection in principle to all human individuals being co-citizens with each other, but this human community must first be politically instituted, in order for such co-citizenship to be effective. As long as that co-citizenship is not effective, it is only within the borders of a national territory that we can conceive of the democratic exercise of human sovereignty.

Dossier's Articles

by Stéphane Haber, 1 May 2014

To quote this article :

Stéphane Haber, « Identity’s Difficulties », Books and Ideas , 1 May 2014. ISSN : 2105-3030. URL :

Nota Bene:

If you want to discuss this essay further, you can send a proposal to the editorial team (redaction at We will get back to you as soon as possible.


[1Will Kymlicka, Multinational Citizenship: A Liberal Theory of Minority Rights, Oxford University Press, 1995.

[2Axel Honneth, The Struggle for Recognition: The Moral Grammar of Social Conflicts, translated from German by Joel Anderson, Polity Press, 1995.

[3Henri Tajfel, Human Groups and Social Categories: Studies in Social Psychology, Cambridge University Press, 1981.

[4Quoted in Les embarras de l’identité, p. 247.

Our partners

© - Any replication forbidden without the explicit consent of the editors. - Mentions légales - webdesign : Abel Poucet