13 The Law of Genre*

JACQUES DERRIDA

As a focal point of structuralist endeavours to create a 'science of

literature', the concept of genre was an obvious target for decon-
struction. In this intricately argued essay, the French philosopher
Jacques Derrida duly performs such an analysis, dramatising - in
order to subvert - the authoritarian undertones that attach to
traditional critical discourse about genre, whilst simultaneously
parodying modem structuralism's quest for the 'laws' of literature.
Originally delivered as a lecture at an international symposium
on genre at the University of Strasbourg in 1979 (contributions to
which were published, together with translations, in a special
issue of the journal Glyph), the discussion falls into two parts.
The first examines the logical status of genre classifications, and
the nature of 'genericity': that is, the means by which texts be-
long to, and inscribe themselves within, a genre or genres. Genre
logic, Derrida argues, is contradictory, and the 'law of genre'
self-defeating, not only because individual texts frequently elude
classification, but also because the textual signals which indicate
membership of a genre cannot themselves be part of that genre.
To illustrate these claims, the second part of the essay (not
included here) offers a detailed reading of Maurice Blanchot's
La Folie du jour (The Madness of the Day), a text which self-
consciously enacts the problematic which Derrida's essay ad-
dresses, and in so doing invalidates, according to Derrida, the
assumptions on which even the most advanced genre theory rests.
As his references indicate, Derrida's discussion is in part a
response to Genette's essay on genres and modes (see above,
Chapter 12), and to Philippe Lacoue-Labarthe and Jean-Luc
Nancy's book The Literary Absolute (1988), a post-structuralist study
of the literary theory of German Romanticism. Whether Derrida's
essay is simply 'a modest annotation on the margins' of these

* Reprinted from Glyph, 7 (1980), 202-13. Translated by Avital Ronell.

219
Modern Genre Theory
predecessors' work (as he at one points suggests), or a more far-
reaching intervention in modem genre theory, is a matter on
which opinion is divided. For a positive assessment, see Leitch,
'(De)Coding (Generic) Discourse' (1991). The gender issues raised
in the second part of the essay are discussed by Mary Jacobus,
'The Law Of! And Gender: Genre Theory and The Prelude', Dia-
critics, 14 (1984), 47-57. The full text of Derrida's essay, together
with some of his other contributions to literary theory, can be
found in Derrida, Acts of Literature, ed. Derek Attridge (London:
Routledge, 1992). For a selection of critical responses, see Derrida:
A Critical Reader, ed. David Wood (Oxford: Blackwell, 1992).

Genres are not to be mixed.

I will not mix genres.
I repeat: genres are not to be mixed. I will not mix them.
Now suppose I let these utterances resonate all by themselves.
Suppose: I abandon them to their fate, I set free their random
virtualities and tum them over to my audience - or rather, to your
audience, to your auditory grasp, to whatever mobility they retain
and you bestow upon them to engender effects of all kinds without
my having to stand behind them.
I merely said, and then repeated: genres are not to be mixed; I will
not mix them.
As long as I release these utterances (which others might call
speech acts) in a form yet scarcely determined, given the open
context out of which I have just let them be grasped from 'my'
language - as long as I do this, you may find it difficult to choose
among several interpretative options. They are legion, as I could
demonstrate. They form an open and essentially unpredictable
series. But you may be tempted by at least two types of audience,
two modes of interpretation, or, if you prefer to give these words
more of a chance, then you may be tempted by two different genres
of hypothesis. Which ones?
On the one hand, it could be a matter of a fragmentary discourse
whose propositions would be of the descriptive, constative, and
neutral genre. In such a case, I would have named the operation
which consists of 'genres are not to be mixed'. I would have
designated this operation in a neutral fashion without evaluating it,
without recommending or advizing against it, certainly without
binding anyone to it. Without claiming to lay down the law or to
make this an act of law, I merely would have summoned up, in a
fragmentary utterance, the sense of a practice, an act or event, as you
wish: which is what sometimes happens when we revert to 'genres
are not to be mixed'. With reference to the same case, and to a
220
 The Law of Genre
hypothesis of the same type, same mode, same genre - or same
order: when I said, '1 will not mix genres', you may have discerned
a foreshadowing description - I am not saying a prescription -
the descriptive designation telling in advance what will transpire,
predicting it in the constative mode or genre, i.e. it will happen thus,
I will not mix genres. The future tense describes, then, what will
surely take place, as you yourselves can judge; but for my part it
does not constitute a commitment. I am not making you a promise
here, nor am I issuing myself an order or invoking the authority of
some law to which I am resolved to submit myself. In this case, the
future tense does not set the time of a performative speech act of a
promising or ordering type.
But another hypothesis, another type of audience, and another
interpretation would have been no less legitimate. 'Genres are not to
be mixed' could strike you as a sharp order. You might have heard
it resound the elliptical but all the more authoritarian summons to
a law of 'do' or 'do not' which, as everyone knows, occupies the
concept or constitutes the value of genre. As soon as the word 'genre'
is sounded, as soon as it is heard, as soon as one attempts to conceive
it, a limit is drawn. And when a limit is established, norms and
interdictions are not far behind: 'Do', 'Do not' says 'genre', the word
'genre', the figure, the voice, or the law of genre. And this can be
said of genre in all genres, be it a question of a generic or a general
determination of what one calls 'nature' or physis (for example, a
biological genre in the sense of gender, or the human genre, a genre of
all that is in general), or be it a question of a typology designated as
non-natural and depending on laws or orders which were once held
to be opposed to physis according to those values associated with
techne, thesis, nomos (for example, an artistic, poetic or literary genre).
But the whole enigma of genre springs perhaps most closely from
within this limit between the two genres of genre which, neither
separable nor inseparable, form an odd couple of one without the
other in which each evenly serves the other a citation to appear in
the figure of the other, simultaneously and indiscernibly saying T
and 'we', me the genre, we genres, without it being possible to think
that the T is a species of the genre 'we'. For who would have us
believe that we, we two for example, would form a genre or belong
to one? Thus, as soon as genre announces itself, one must respect a
norm, one must not cross a line of demarcation, one must not risk
impurity, anomaly or monstrosity. And so it goes in all cases,
whether or not this law of genre be interpreted as a determination or
perhaps even as a destination of physis, and regardless of the weight
or range imputed to physis. If a genre is what it is, or if it is supposed
to be what it is destined to be by virtue of its telos, then 'genres are
221
Modern Genre Theory
not to be mixed'; one should not mix genres, one owes it to oneself
not to get mixed up in mixing genres. Or, more rigorously, genres
should not intermix. And if it should happen that they do intermix,
by accident or through transgression, by mistake or through a lapse,
then this should confirm, since, after all, we are speaking of 'mixing',
the essential purity of their identity. This purity belongs to the typical
axiom: it is a law of the law of genre, whether or not the law is, as
one feels justified in saying, 'natural'. This normative position and
this evaluation are inscribed and prescribed even at the threshold of
the 'thing itself', if something of the genre 'genre' can be so named.
And so it follows that you might have taken the second sentence in
the first person, 'I will not mix genres', as a vow of obedience, as a
docile response to the injunction emanating from the law of genre.
In place of a constative description, you would then hear a promise,
an oath; you would grasp the following respectful commitment: I
promise you that I will not mix genres, and, through this act of
pledging utter faithfulness to my commitment, I will be faithful
to the law of genre, since, by its very nature, the law invites
and commits me in advance not to mix genres. By publishing my
response to the imperious call of the law, I would correspondingly
commit myself to be responsible.
Unless, of course, I were actually implicated in a wager, a
challenge, an impossible bet - in short, a situation that would exceed
the matter of merely engaging a commitment from me. And suppose
for a moment that it were impossible not to mix genres. What if there
were, lodged within the heart of the law itself, a law of impurity or
a principle of contamination? And suppose the condition for the
possibility of the law were the a priori of a counter-law, an axiom
of impossibility that would confound its sense, order and reason?
I have just proposed an alternative between two interpretations. I
did not do so, as you can imagine, in order to check myself. The line
or trait that seemed to separate the two bodies of interpretation is
affected straight away by an essential disruption that, for the time
being, I shall let you name or qualify in any way you care to: as
internal division of the trait, impurity, corruption, contamination,
decomposition, perversion, deformation, even cancerization, generous
proliferation or degenerescence. All these disruptive 'anomalies' are
engendered - and this is their common law, the lot or site they share
- by repetition. One might even say by citation or re-citation (rc-cit),
provided that the restricted use of these two words is not a call to
strict generic order. A citation in the strict sense implies all sorts
of contextual conventions, precautions and protocols in the mode
of reiteration, of coded signs such as quotation marks or other
typographical devices used for writing a citation. The same holds no
222
 The Law of Genre
doubt for the r€cit as a form, mode, or genre of discourse, even - and
I shall return to this - as a literary type. And yet the law that protects
the usage, in stricto sensu, of the words citation and recit, is threatened
intimately and in advance by a counter-law that constitutes this very
law, renders it possible, conditions it and thereby renders it impossible
- for reasons of edges on which we shall run aground in just a
moment - to edge through" to edge away from or to hedge around
the counter-law itself. The law and the counter-law serve each other,
citations summoning each other to appear, and each re-cites the other
in this proceeding (proces). There would be no cause for concern if
one were rigorously assured of being able to distinguish with rigor
between a citation and a non-citation, a fecit and a non-recit or a
repetition within the form of one or the other.
I shall not undertake to demonstrate, assuming it is still possible,
why you were unable to decide whether the sentences with which I
opened this presentation and marked this context were or were not
repetitions of a citational type; or whether they were or were not of
the performative type; or certainly whether they were, both of them,
together - and each time together - the one or the other. For perhaps
someone has noticed that, from one repetition to the next, a change
had insinuated itself into the relationship between the two initial
utterances. The punctuation had been slightly modified, as had the
content of the second independent clause. This barely noticeable shift
could theoretically have created a mutual independency between the
interpretative alternatives that might have tempted you to opt for
one or the other, or for one and the other of these two sentences.
A particularly rich combinatory of possibilities would thus ensue,
which, in order not to exceed my time-limit and out of respect for
the law of genre and of the audience, I shall abstain from recounting.
I am simply going to assume a certain relationship between what
has just now happened and the origin of literature, as well as its
aborigine or its abortion, to quote Philippe Lacoue-Labarthe. 1
Provisionally claiming for myself the authority of such an
assumption, I shall let our field of vision contract as I limit myself
to a sort of species of the genre 'genre'. I shall focus on this genre of
genre which is generally supposed, and always a bit too rashly, not to
be part of nature, of physis, but rather of techne, of the arts, still more
narrowly of poetry, and most particularly of literature. But at the
same time, I take the liberty to think that, while limiting myself thus,
I exclude nothing, at least in principle and de jure - the relationships
here no longer being those of extension, from exemplary individual
to species, from species to genre as genus or from the genre of genre
to genre in general; rather as we shall see, these relationships are a
whole order apart. What is at stake, in effect, is exemplarity and its
223
Modern Genre Theory
whole enigma - in other words, as the word enigma indicates, exemplarity
and the recit which works through the logic of the example.
Before going about putting a certain example to the test, I shall
attempt to formulate, in a manner as elliptical, economical, and
formal as possible, what I shall call the law of the law of genre.
It is precisely a principle of contamination, a law of impurity, a
parasitical economy. In the code of set theories, if I may use it at
least figuratively, I would speak of a sort of participation without
belonging - a taking part in without being part of, without having
membership in a set. The trait that marks membership inevitably
divides, the boundary of the set comes to form, by invagination,
an internal pocket larger than the whole; and the outcome of this
division and of this abounding remains as singular as it is limitless.
To demonstrate this, I shall hold to the leanest generalities. But I
should like to justify this initial indigence or asceticism as well as
possible. For example, I shall not enter into the passionate debate
brought forth by poetics on the theory and the history of genre-
theory, on the critical history of the concept of genre from Plato to
the present. My stance is motivated by these considerations: in the
first place, we now have at our disposal some remarkable, and, of
late, handsomely enriched works dealing either with primary texts or
critical analyses. I am thinking especially of the journal Poetique, of its
issue entitled 'Genres' (32) and of Genette's opening essay, 'Genres,
"Types", Modes'. From yet another point of view, L'Absolu litteraire
(The Literary Absolute) has already created quite a stir in this context,
and everything that I shall risk here should perhaps resolve itself in
a modest annotation on the margins of this magistral work which
I assume some of you have already read. I could further justify
my abstention or my abstinence here simply by acknowledging the
terminological luxury or rapture as well as the taxonomic exuberance
which debates of this kind, in a manner by no means fortuitous,
have sparked: I feel completely powerless to contain this fertile
proliferation - and not only because of time-constraints. I shall
put forth, instead, two principal motives, hoping thereby to justify
my keeping to scant preliminary generalities at the edge of this
problematic.
To what do these two motives essentially relate? In its most recent
phase - and this much is certainly clear in Genette's propositions -
the most advanced critical axis has led to a rereading of the entire
history of genre-theory. This rereading has been inspired by the
perception - and it must be said, despite the initial denial, by the
correction - of two types of misconstruing or confusion. On the one
hand, and this will be the first motive or ground for my abstention,
Plato and Aristotle have been subjected to considerable deformation,
224
 The Law of Genre
. as Genette reminds us, insofar as they have been viewed in terms
alien to their thinking, and even in terms that they themselves would
have rejected; but this deformation has usually taken on the form of
naturalization. Following a classical precedent, one has deemed
natural structures or typical forms whose history is hardly natural,
but rather, quite to the contrary, complex and heterogeneous. These
forms have been treated as natural - and let us bear in mind the
entire semantic scale of this difficult word whose span is so far-
ranging and open-ended that it extends as far as the expression
'natural language', by which term everyone agrees tacitly to oppose
natural language only to a formal or artificial language without
thereby implying that this natural language is a simple physical or
biological production. Genette insists at length on this naturalization
of genres: 'The history of genre-theory is strewn with these fascinating
outlines that inform and deform reality, a reality often heterogenous
to the literary field, and that claim to discover a natural"system"
wherein they construct a factitious symmetry heavily reinforced by
fake windows' (italics added, p. 408)? In its most efficacious and
legitimate aspect, this critical reading of the history (and) of genre-
theory is based on an opposition between nature and history, and,
more generally - as the allusion to an artificial construct indicates
(' ... wherein they construct a factitious symmetry ... ') - on an
opposition between nature and what can be called the series of all its
others. Such an opposition seems to go without saying; placed within
this critical perspective, it is never questioned. Even if it has been
tucked away discreetly in some passage that has escaped my attention,
this barely visible suspicion clearly had no effect on the general
organization of the problematic. This does not diminish the relevance
or fecundity of a reading such as Genette's. But a place remains open
for some preliminary questions concerning his presuppositions, for
some questions concerning the boundaries where it begins to take
hold or take place. The form of these boundaries will contain me, and
rein me in. These general propositions whose number is always open
and indeterminable for whatever critical interpretation will not be
dealt with here. What however seems to me to require more urgent
attention is the relationship of nature to history, of nature to its
others, precisely when genre is on the line.
Let us consider the most general concept of genre, from the minimal
trait or predicate delineating it permanently through the modulations
of its types and the regimens of its history: it rends and defends itself
by mustering all its energy against a simple opposition that arises
from nature and from history, as from nature and the vast lineage
of its others (techne, nomos, thesis, then spirit, society, freedom, history,
etc.). Between physis and its others, genos certainly situates one of the
225
Modern Genre Theory
privileged scenes of the process, and, no doubt, sheds the greatest
obscurity on it. One need not mobilize etymology to this end and
could just as well equate genos with birth, and birth in tum with
the generous force of engenderment or generation - physis, in fact-
as with race, familial membership, classificatory genealogy or class,
age class (generation) or social class; it comes as no surprise that, in
nature and art, genre, a concept that is essentially classificatory and
genealogico-taxonomic, itself engenders so many classificatory vertigines
when it goes about classifying itself and situating the classificatory
principle or instrument within a set. As with the class itself, the
principle of genre is unclassifiable, it tolls the knell of the knell (glas),
in other words of classicum, of what permits one to call out (calare)
orders and to order the manifold within a nomenclature. Genos thus
indicates the place, the now or never of the most necessary meditation
on the 'fold' which is no more historical than natural in the classical
sense of these two words, and which turns phyein over to itself across
others that perhaps no longer relate to it according to that epoch-
making logic which was decisory, critical, oppositional, even dialectical,
but rather according to the trait of a contract entirely other. De jure,
this meditation acts as an absolute prerequisite without which any
historical perspectivizing will always be difficult to legitimate. For
example, the romantic era - this powerful figure indicted by Genette
(since it attempted to reinterpret the system of modes as a system of
genres) - is no longer a simple era and can no longer be inscribed as
a moment or a stage placeable within the trajectory of a 'history'
whose concept we could be certain of. Romanticism, if something of
the sort can be thus identified, is also the general repetition of all the
folds that in themselves gather, couple, divide physis as well as genos
through the genre, and through all the genres of genre, through the
mixing of genre that is 'more than a genre', through the excess of
genre in relation to itself, as to its abounding movement and its
general assemblage which coincides, too, with its dissolution. 3 Such
a 'moment' is no longer a simple moment in the history and theory
of literary genres. To treat it thus would in effect implicate one as
tributary - whence the strange logic - of something that has in itself
constituted a certain romantic motif, namely, the teleological ordering
of history. Romanticism simultaneously obeys naturalizing and
historicizing logic, and it can be shown easily enough that we have
not yet been delivered from the romantic heritage - even though we
might wish it so and assuming that such a deliverance would be of
compelling interest to us - as long as we persist in drawing attention
to historical concerns and the truth of historical production in order
to militate against abuses or confusions of naturalization. The debate,
it could be argued, remains itself a part or effect of romanticism.
226
 The Law of Genre
A second motive detains me at the threshold or on the edge of a
possible problematic of genre (as) history and theory of history and
of genre-theory - another genre, in fact. For the moment, I find it
impossible to decide - impossible for reasons that I do not take to
be accidental, and this, precisely, is what matters to me - I find it
impossible to decide whether the possibly exemplary text which I
intend to put to the test does or does not lend itself to the distinction
drawn between mode and genre. Now, as you may recall, Genette
demonstrates the stringent necessity of this distinction; and he rests
his case on 'the confusion of modes and genres' (p. 417). This implies
a serious charge against romanticism, even though 'the romantic
reinterpretation of the system of modes as a system of genres is
neither de facto nor de jure the epilogue to this long history' (p. 415).
This confusion, according to Genette, has aided and abetted the
naturalization of genres by projecting onto them the 'privilege of
naturalness, which was legitimately . .. that of three modes .. .' (p.421).
Suddenly, this naturalization 'makes these arch-genres into ideal or
natural types which they neither are nor can be: there are no arch-
genres that can totally escape historicity while preserving a generic
definition. There are modes, for example: the recit. There are genres,
for example: the novel; the relation of genres to modes is complex
and, perhaps not, as Aristotle suggests, one of simple inclusion.'
. If I am inclined to poise myself on this side of Genette's argument,
it is not only because of his ready acceptance of the distinction
between nature and history, but also because of its implications with
regard to mode and to the distinction between mode and genre.
Genette's definition of mode contains this singular and interesting
characteristic: it remains, in contradistinction to genre, purely formal.
Reference to a content has no pertinence. This is not the case with
genre. The generic criterion and the modal criterion, Genette says,
are absolutely heterogenous: 'each genre defined itself essentially by
a specification of content which was not prescribed by the definition
of mode .. .' (p. 417). I do not believe that this recourse to the
opposition of form and content, this distinction between mode
and genre, need be contested, and my purpose is not to challenge
isolated aspects of Genette's argument. One might just question the
presuppositions for the legitimacy of such an argument. One might
also question the extent to which his argument can help us read a
given text when it behaves in a given way with regard to mode and
genre, especially when the text does not seem to be written sensibly
within their limits, but rather about the very subject of those limits,
and with the aim of disrupting their order. The limits, for instance,
of that mode which would be, according to Genette, the recit ('there
are modes, for example: the recit'). Of the (possibly) exemplary text
227
Modern Genre Theory
which I shall address shortly, I shall not hasten to add that it is a
'reeit', and you will soon understand why. In this text, the recit is not
only a mode, and a mode put into practice or put to the test because
it is deemed impossible; it is also the name of a theme. It is the
nonthematizable thematic content of something of a textual form that
assumes a point of view with respect to the genre, even though it
perhaps does not come under the heading of any genre - and perhaps
no longer even under the heading of literature, if it indeed wears
itself out around genre-less modalizations, it would confirm one of
Genette's propositions: 'Genres are, properly speaking, literary [or
aesthetic] categories; modes are categories that pertain to linguistics
or, more precisely, to an anthropology of verbal expression' (p. 418).4
In a very singular manner, the very short text which I will discuss
presently makes the recit and the impossibility of the recit its theme,
its impossible theme or content at once inaccessible, indeterminable,
interminable and inexhaustible; and it makes the word 'recit', under
the aegis of a certain form, its titleless title, the mentionless mention
of its genre. This text, as I shall try to demonstrate, seems to be
made, among other things, to make light of all the tranquil categories
of genre-theory and history in order to upset their taxonomic
certainties, the distribution of their classes, and the presumed stability
of their classical nomenclatures. It is a text destined, at the same
time, to summon up these classes by conducting their proceeding,
by proceeding from the proceeding to the law of genre. For if the
juridical code has frequently thrust itself upon me in order to hear
this case, it has done so to call as witness a (possibly) exemplary text,
and because I am convinced copious rights are bound up in all of
this: the law itself is at stake.
These are the two principal reasons why I shall keep to the liminal
edge of (the) history (and) of genre-theory. Here now, very quickly,
is the law of abounding, of excess, the law of participation without
membership, of contamination, etc., which I mentioned earlier. It will
seem meager to you, and even of staggering abstractness. It does not
particularly concern either genres, or types, or modes or any form in
the strict sense of its concept. I therefore do not know under what
title the field or object submitted to this law should be placed. It is
perhaps the limitless field of general textuality. I can take each word
of the series (genre, type, mode, form) and decide that it will hold
for all the others (all genres of genres, types, modes, forms; all types
of types, genres, modes, forms; all forms of forms, etc.). The trait
common to these classes of classes is precisely the identifiable
recurrence of a common trait by which one recognises, or should
recognise, a membership in a class. There should be a trait upon
which one could rely in order to decide that a given textual event, a
228
 The Law of Genre
given 'work', corresponds to a given class (genre, type, mode, form,
etc.). And there should be a code enabling one to decide questions of
class-membership on the basis of this trait. For example - a very
humble axiom, but, by the same token, hardly contestable - if a genre
exists (let us say the novel, since no one seems to contest its generic
quality), then a code should provide an identifiable trait and one
which is identical to itself, authorizing us to determine, to adjudicate
whether a given text belongs to this genre or perhaps to that genre.
Likewise, outside of literature or art, if one is bent on classifying, one
should consult a set of identifiable and codifiable traits to determine
whether this or that, such a thing or such an event belongs to this set
or that class. This may seem trivial. Such a distinctive trait qua mark
is however always a priori remarkable. It is always possible that a set
- I have compelling reasons for calling this a text, whether it be
written or oral- re-marks on this distinctive trait within itself. This
can occur in texts that do not, at a given moment, assert themselves
to be literary or poetic. A defense speech or newspaper editorial can
indicate by means of a mark, even if it is not explicitly designated as
such, 'Voila! I belong, as anyone may remark, to the type of text
called a defense speech or an article of the genre newspaper-editorial.'
The possibility is always there. This does not constitute a text ipso
facto as 'literature', even though such a possibility, always left open
and therefore eternally remarkable, situates perhaps in every text the
possibility of its becoming literature. But this does not interest me at
the moment. What interests me is that this re-mark - ever possible
for every text, for every corpus of traces - is absolutely necessary for
and constitutive of what we call art, poetry or literature. It underwrites
the eruption of techne, which is never long in coming. I submit this
axiomatic question for your consideration: can one identify a work of
art, of whatever sort, but especially a work of discursive art, if it does
not bear the mark of a genre, if it does not signal or mention it or
make it remarkable in any way? Let me clarify two points on this
subject. First, it is possible to have several genres, an intermixing of
genres or a total genre, the genre 'genre' or the poetic or literary
genre as genre of genres. Second, this re-mark can take on a great
number of forms and can itself pertain to highly diverse types. It
need not be a designation or 'mention' of the type found beneath the
title of certain books (novel, recit, drama). The re-mark of belonging
need not pass through the consciousness of the author or the reader,
although it often does so. It can also refute this consciousness or
render the explicit 'mention' mendacious, false, inadequate or ironic
according to all sorts of overdetermined figures. Finally, this re-
marking-trait need be neither a theme nor a thematic component of
the work - although of course this instance of belonging to one or
229
Modern Genre Theory
several genres, not to mention all the traits that mark this belonging,
often have been treated as theme, even before the advent of what we
call 'modernism'. If I am not mistaken in saying that such a trait is
remarkable, that is, noticeable, in every aesthetic, poetic or literary
corpus, then consider this paradox, consider the irony (which is
irreducible to ~ consciousness or an attitude): this supplementary and
distinctive trait, a mark of belonging or inclusion, does not properly
pertain to any genre or class. The re-mark of belonging does not
belong. It belongs without belonging, and the 'without' (or the suffix
'-less') which relates belonging to non-belonging appears only in the
timeless time of the blink of an eye (Augenblick). The eyelid closes,
but barely, an instant among instants, and what it closes is verily the
eye, the view, the light of day. But without such respite, nothing
would come to light. To formulate it in the scantiest manner - the
simplest but most apodictic - I submit for your consideration the
following hypothesis: a text cannot belong to no genre, it cannot
be without or less a genre. Every text participates in one or several
genres, there is no genreless text; there is always a genre and genres,
yet such participation never amounts to belonging. And not because
of an abundant overflowing or a free, anarchic and unclassifiable
productivity, but because of the trait of participation itself, because
of the effect of the code and of the generic mark. Making genre its
mark, a text demarcates itself. If remarks of belonging belong without
belonging, participate without belonging, then genre-designations
cannot be simply part of the corpus. Let us take the designation
'novel' as an example. This should be marked in one way or another,
even if it does not appear, as it often does in French and German
texts, in the explicit form of a subtitled designation, and even if it
proves deceptive or ironic. This designation is not novelistic; it does
not, in whole or in part, take part in the corpus whose denomination
it nonetheless imparts. Nor is it simply extraneous to the corpus. But
this singular topos places within and without the work, along its
boundary, an inclusion and exclusion with regard to genre in general,
as to an identifiable class in general. It gathers together the corpus
and, at the same time, in the same blinking of an eye, keeps it from
closing, from identifying itself with itself. This axiom of non-closure
or non-fulfillment enfolds within itself the condition for the possibility
and the impossibility of taxonomy. This inclusion and this exclusion
do not remain exterior to one another; they do not exclude each
other. But neither are they immanent or identical to each other. They
are neither one nor two. They form what I shall call the genre-clause, a
clause stating at once the juridical utterance, the precedent-making
designation and the law-text, but also the closure, the closing that
excludes itself from what it includes (one could also speak of a floodgate
230
 The Law of Genre
(,ecluse') of genre). The clause or floodgate of genre declasses what it
allows to be classed. It tolls the knell of genealogy or of genericity,
which it however also brings forth to the light of day. Putting to death
the very thing that it engenders, it cuts a strange figure; a formless
form, it remains nearly invisible, it neither sees the day nor brings
itself to light. Without it, neither genre nor literature come to light,
but as soon as there is this blinking of an eye, this clause or this
floodgate of genre, at the very moment that a genre or a literature
is broached, at that very moment, degenerescence has begun, the
end begins.

Notes
1. Philippe Lacoue-Labarthe and Jean-Luc Nancy, L'Absolu litteraire: theorie de la
litterature du Romantisme allemand (Paris, Seuil: 1978) (Eng. The Literary Absolute:
The Theory of Literature in Gf71nan Romanticism, trans. Philip Barnard and
Cheryl Laster (New York: State University of New York Press, 1988)) [Ed.].
2. Derrida's page references are to Genette's original article, 'Genres, "Types",
Mode', Poetique, 8:32 (1977),389-421, not to the later version included in
Chapter 12 above [Ed.].
3. In this respect, the second footnote in L'Absolu litteraire, p. 271, seems
to me, let us say, a bit too equitable in its rigorous and honest prudence
[Derrida's note].
4. Compare the revised version of Genette's statement on p. 213 above [Ed.].

231