The Omnibus Homo Sacer

by Giorgio Agamben

  given of the enunciated, before or beyond their semantic content. The speaking

  being or enunciator has thus been substituted for Kant’s transcendental subject,

  and language has taken the place of being as historical a priori.

  This linguistic declension of ontology seems today to have reached its com-

  pletion. Certainly language has never been so omnipresent and pervasive, su-

  perimposing itself in every sphere—not only in politics and communication

  but also and above all in the sciences of nature—over being, apparently without

  leaving any remainder. What has changed, however, is that language no longer

  functions as a historical a priori, which while remaining unthought, determines

  and conditions the historical possibilities of speaking human beings. In being

  totally identified with being, it is now put forward as a neutral ahistorical or

  post-historical effectuality, which no longer conditions any recognizable sense of

  historical becoming or any epochal articulation of time. This means that we live

  in a time that is not—or at least pretends not to be—determined by any histor-

  ical a priori, which is to say, a post-historical time (or rather, a time determined by the absence or impossibility of such an a priori).

  It is from this perspective that we are seeking to trace out—even if purely in

  the form of a summary sketch—an archeology of ontology, or more precisely, a

  genealogy of the ontological apparatus that has functioned for two millennia as

  a historical a priori of the West. If ontology is first of all a hodology, which is to say, the way that being always historically opens toward itself, it is the existence

  today of something like a hodos or a way that we seek to interrogate, by asking

  ourselves whether the track that seems to have been interrupted or lost can be

  taken up again or instead must be definitively abandoned.

  This page intentionally left blank

  1

  Ontological Apparatus

  1.1. An archeology of first philosophy must begin from the apparatus of be-

  ing’s division that defines Aristotelian ontology. This apparatus—which

  divides and at the same time articulates being and is, in the last instance, at the

  origin of every ontological difference—has its locus in the Categories. Here Aristotle distinguishes an ousia, an entity or essence, “which is said most strictly, primarily, and first of all” ( kyriotata te kai protos kai malista legomene) from the secondary essences ( ousiai dueterai). The former is defined as “that which is not said of a subject

  [ hypokeimenon, that which lies under, sub-iectum] nor in a subject” and is exemplified by the singularity, the proper name, and deixis (“this certain man, Socrates; this

  certain horse”); the latter are “those in whose species the essences called primary are

  present, as are the genera of these species—for example, ‘this certain man’ belongs

  to the species ‘man’ and the genus of this species is ‘animal’” ( Categories, 2a 10–15).

  Whatever may be the terms in which the division is articulated in the course

  of its history (primary essence/secondary essence, existence/essence, quod est/

  quid est, anitas/ quidditas, common nature/supposition, Dass sein/ Was sein, being/

  beings), what is decisive is that in the tradition of Western philosophy, being,

  like life, is always interrogated beginning with the division that traverses it.

  א We translate hypokeimenon with “subject” ( sub-iectum). Etymologically the term means “that which lies under or at the base.” This is not the place to show the turns

  and vicissitudes by means of which the Aristotelian hypokeimenon became the subject of modern philosophy. It is certain in any case that, by means of Latin translations,

  this passage of the Categories has decisively determined the vocabulary of Western philosophy. In the terminology of Aquinas, the Aristotelian articulation of being thus

  appears in this way:

  According to the Philosopher, substance is said in two ways. In a primary sense it

  means the quidditas of the thing, signified by its definition, and thus we say that the definition means the substance of a thing, which the Greeks call ousia and we

  call essentia. In a second sense, it is the subject [ subiectum] or suppositum [“what is put under”] which subsists [ subsistit] in the genus of substance. This can be

  1131

  1132

  HOMO SACER IV, 2

  expressed with a term that means intention and this is called suppositum. It is also called by three names signifying the thing: that is, a thing of a nature [ res naturae], subsistence [ subsistentia], and hypostasis [ hypostasis]. For insofar as it exists in itself and not in another, it is called subsistentia, and insofar as it underlies some common nature, it is called a thing of a nature; as, for instance, “this man” is a thing

  of human nature. Finally, insofar as it is pre-supposed by the accidents [ supponitur

  accidentibus], it is called hypostasis or substance. (Aquinas 2, I, q. 29, art. 2, Resp.) Whatever may be the terms in which it is expressed at various times, this division of

  being is at the base of the “ontological difference” that, according to Heidegger, defines Western metaphysics.

  1.2. The treatise on Categories or predications (though the Greek term kat-

  egoriai means “imputations, accusations” in juridical language) is traditionally

  classified among Aristotle’s logical works. However, it contains, for example, in

  the passage in question, theses of an undoubtedly ontological character. An-

  cient commentators therefore debated over what was the object ( skopos, end) of

  the treatise: words ( phonai), things ( pragmata), or concepts ( noemata). In the prologue to his commentary, Philoponus writes that according to some (among

  them Alexander of Aphrodisias) the object of the treatise is only words, accord-

  ing to others (like Eustatius) only things, and according to others, finally (like

  Porphyry), only concepts. More correct, according to Philoponus, is the thesis

  of Iamblichus (which he accepts with some clarifications) according to which

  the skopos of the treatise is words insofar as they signify things by means of concepts ( phonon semainouson pragmata dia meson noematon; Philoponus, pp. 8–9).

  Hence the impossibility, in the Categories, of distinguishing logic and on-

  tology. Aristotle treats here of things, of beings, insofar as they are signified by

  language, and of language insofar as it refers to things. His ontology presup-

  poses the fact that, as he never stops repeating, being is said ( to on legetai . . .), is always already in language. The ambiguity between logic and ontology is so

  consubstantial to the treatise that, in the history of Western philosophy, the

  categories appear both as classes of predication and as classes of being.

  1.3. At the beginning of the treatise, immediately after having defined hom-

  onyms, synonyms, and paronyms (that is, things insofar as they are named),

  Aristotle specifies this onto-logical implication between being and language in

  the form of a classification of beings according to the structure of subjectivation

  or pre-supposition.

  Of beings, some are said of a subject [ kath’ hypocheimenou, lit.: “on the presupposition of a lying under”] but are not in any subject [ en hypocheimenoi oudeni].

  THE USE OF BODIES

  1133

  For example, “human” is said on the presupposition [subjectivation] of a certain

  human being but is not in any subject. . . . Others are in a subject but are not said of any
subject. . . . For example, a certain grammatical knowing is in a subject,

  the soul, but is not said of any subject. . . . Still others are both said of a subject and in a subject. For example, knowledge is said of a subject, grammar, and is

  in a subject, the soul. Yet others are neither in a subject nor said of a subject: for example, a certain man or a certain horse. ( Categories, 1a 20–1b 5)

  The distinction between saying (to say of a subject) and being (to be in a subject) does not correspond to the opposition between language and being, linguistic

  and non-linguistic, so much as to the promiscuity between the two meanings

  of the verb “to be” ( einai), the existentive and the predicative. The structure of subjectivation/presupposition remains the same in both cases: the articulation

  worked by language always pre-sup-poses a relation of predication (general/

  particular) or of inherence (substance/accident) with respect to a subject, an

  existent that lies-under-and-at-the-base. Legein, “to say,” means in Greek “to

  gather and articulate beings by means of words”: onto-logy. But in this way, the distinction between saying and being remains uninterrogated, and it is the opacity of their relation that will be transmitted by Aristotle to Western philosophy,

  which will take it in without the benefit of an inventory.

  א It is well known that in Indo-European languages the verb “to be” generally has

  a double meaning: the first meaning corresponds to a lexical function, which expresses

  the existence and reality of something (“God is,” that is, exists), while the second—the

  copula—has a purely logico-grammatical function and expresses the identity between

  two terms (“God is good”). In many languages (as in Hebrew and Arabic), or in the same

  language in different epochs (as in Greek, in which originally the copulative function is

  expressed by a nominal phrase with no verb: ariston hydor, “the best thing is water”), the two meanings are, by contrast, lexically distinct. As Émile Benveniste writes,

  What matters is to see clearly that there is no connection, either by nature or

  by necessity, between the verbal notion of “to exist, to be really there” and the

  function of the “copula.” One need not ask how it happens that the verb “to be”

  can be lacking or omitted. This is to reason in reverse. The real question should

  be the opposite; how is it that there is a verb “to be” which gives verbal expres-

  sion and lexical consistency to a logical relationship in an assertive utterance?

  (Benveniste, p. 189/164)

  Precisely the promiscuity between the two meanings is at the base of the many aporias and

  difficulties in the history of Western ontology, which has been constituted, so to speak,

  as a double machine, set on distinguishing and, at the same time, articulating together

  the two notions into a hierarchy or into a coincidence.

  1134

  HOMO SACER IV, 2

  1.4. A little further down, apropos of the relation between secondary sub-

  stances and primary substances, Aristotle writes:

  It is clear from what has been said that if something is said of a subject [ kath’-

  hypokeimenou, “on the pre-sup-position of a lying-under”], both its name and its

  definition are necessarily predicated [ kategoreisthai] of the subject. For example,

  “man” is said on the subjectivation [on the pre-sup-position] of this certain

  man, and the name is of course predicated (since you will be predicating “man”

  of this certain man), and also the definition of “man” will be predicated of this

  certain man (since this certain man is also a man). Thus, both the name and the

  definition will be predicated of the subject. ( Categories, 2a 19–25)

  The subjectivation of being, the presupposition of the lying-under is therefore

  inseparable from linguistic predication, is part of the very structure of language

  and of the world that it articulates and interprets. Insofar as, in the Categories, being is considered from the point of view of linguistic predication, of its being

  “accused” ( kategorein first of all means “to accuse” in Greek) by language, it

  appears “most properly, in the first place, and above all” in the form of sub-

  jectivation. The accusation, the summons to trial that language directs toward

  being subjectivates it, presupposes it in the form of a hypokeimenon, of a singular existent that lies-under-and-at-the-base.

  The primary ousia is what is said neither on the presupposition of a subject nor in

  a subject, because it is itself the subject that is pre-sup-posed—as purely existent—as what lies under every predication.

  1.5. The pre-supposing relation is, in this sense, the specific potential of

  human language. As soon as there is language, the thing named is presupposed as

  the non-linguistic or non-relational with which language has established its rela-

  tion. This presuppositional power is so strong that we imagine the non- linguistic

  as something unsayable and non-relational that we seek in some way to grasp

  as such, without noticing that what we seek to grasp in this way is only the

  shadow of language. The non-linguistic, the unsayable is, as should be obvious,

  a genuinely linguistic category: it is in fact the “category” par excellence—the

  accusation, the summons worked by human language, which no non-speaking

  living being could ever conceive. That is to say, the onto-logical relation runs

  between the beings presupposed by language and their being in language. What

  is non-relational is, as such, above all the linguistic relation itself.

  It is in the structure of presupposition that the interweaving of being and

  language, ontology and logic that constitutes Western metaphysics is articulated.

  Called into question from the point of view of language, being is from the very

  THE USE OF BODIES

  1135

  beginning divided into an existentive being (existence, the primary ousia) and a

  predicative being (the secondary ousia, what is said of it): the task of thought will then be that of reassembling into a unity what thought—language—has presupposed and divided. The term “presupposition” indicates, that is to say, the subject

  in its original meaning: that which, lying before and at the base, constitutes the

  “ on-which” (on the presupposition of which) one says and which cannot, in its

  turn, be said about anything. The term “presupposition” is etymologically perti-

  nent: hypo keisthai is used, in fact, as passive perfect of hypotithenai, and hypokeimenon thus means “that which, having been pre-sup-posed, lies under.” In this

  sense Plato—who is perhaps the first to thematize the presuppositional power

  of language that is expressed in language in the opposition between names ( ono-

  mata) and discourse ( logos)—can write, “The primary names, by which in some way other names are presupposed [ hypokeitai], in what way do they manifest

  beings to us?” ( Cratylus, 422d), or again: “by each of these names is presupposed

  [ hypokeitai] a particular essence [ ousia]” ( Protagoras, 349b). Being is that which is a presupposition to the language that manifests it, that on presupposition of

  which what is said is said.

  (It is this presuppositional structure of language that Hegel—hence his suc-

  cess and his limits—will seek at the same time to capture and to liquidate by

  means of the dialectic; Schelling, for his part, will instead attempt to grasp it by

  suspe
nding thought, in astonishment and stupor. But even in this case, what the

  mind as if astonished contemplates without managing to neutralize it is the very

  structure of presupposition.)

  א Aristotle frequently expresses with perfect awareness the onto-logical interweav-

  ing of being and saying: “Those things are said in their own right to be that are indicated by the figures of the categories; according to the way in which it is said, such is the meaning of being” ( kath’ autà de einai legetai osaper semanei ta schemata tes kategorias: osachos gar legetai, tosautachos to einai semanei; Metaphysics, 1017a 22ff.). The ambiguity is after all implied in the celebrated formulation of Metaphysics, 1028a 10ff.: “Being is said in many ways . . . for in one sense it means that a thing is or is a ‘this,’ and in another sense it means that a thing is of a certain quality or quantity and each of the other things that are predicated in this way.” Being is constitutively something that “is said” and “means.”

  1.6. Aristotle therefore founds the priority of the subjective determination of

  ousia in these terms:

  All the other things are either said on the pre-sup-position [ kath’ hypokeimenou,

  on subjectivation] of the primary ousiai or they are in the presupposition of

  these. . . . For example, “animal” is predicated of the human being, and therefore

  also of this certain human being; for were it predicated of none of the individual

  1136

  HOMO SACER IV, 2

  human beings, it would not be predicated of the human being at all. . . . So if the

  primary substances [ ousiai] did not exist, it would be impossible for any of the

  other things to exist. Thus, all the other things are either said on the presupposi-

  tion of their standing-under or presupposed in the latter. ( Categories 2a 34–2b6)

  This priority of the primary substance—expressed in language by a proper name

  or an ostensive pronoun—is confirmed a few lines after: “The primary ousiai,

  insofar as they are supposed [ hypokesthai] by all the other things and all the others are predicated of them and are in them, are for this reason called ousiai par

  excellence” ( Categories, 2a 15–17).

  The primary essence is “most properly, in the first place, and above all” ousia,