CHAPTER 1

THEORETICAL AND TEXTUAL BACKGROUND

1.0 Introduction

This chapter has a twofold aim: first to provide the reader with the

theoretical background to the textual analysis proposed in Chapters 2 and 3,

and second to give some background on Exercices de style and its author. The
principle underlying all of my thinking on translation and Exercices de style is

Jakobson's "equivalence in difference." I discuss this concept, then respond to

a critique of the essay which introduced it and other theoretical notions

pertaining to translation. The remainder of this theoretical section concerns

itself with the fundamental notions of Peircean semiotics and their relevance

to translation. The second part of the chapter provides background on the

author and the text. The genesis of the text, the significance of the number of

exercices, and the literature on Exercices de style all come under discussion in

this portion of the chapter. I end with a consideration of the role played by

critical reception in analyzing the relationship between the source text and its

translation(s).

1.1 Equivalence in Difference

The concept of equivalence in difference rightfully belongs in any

discussion of Queneau's Exercices de style, as it is the structuring principle of

the entire work. Queneau tells a brief story of an encounter on a bus, but he

tells it ninety-nine times in ninety-nine different ways. Here is how he tells it

the first time:

10
 11

Notations [Queneau, 1947]

Dans l'S, à une heure d'affluence. Un type dans les vingt-six ans,
chapeau mou avec cordon remplaçant le ruban, cou trop long
comme si on lui avait tiré dessus. Les gens descendent. Le type
en question s'irrite contre un voisin. Il lui reproche de le
bousculer chaque fois qu'il passe quelqu'un, ton pleurnichard
qui se veut méchant. Comme il voit une place libre se précipite
dessus. Deux heures plus tard je le rencontre dans la cour de
Rome, devant la gare Saint-Lazare. Il est avec un camarade qui
lui dit: «Tu devrais faire mettre un bouton supplémentaire à ton
pardessus». Il lui montre où (à l'échancrure) et pourquoi.
And here is the second text:

En partie double [Queneau, 1947]

Vers le milieu de la journée et à midi, je me trouvai et montai sur

la plate-forme et la terrasse arrière d’un autobus et d’un véhicule
des transports en commun bondé et quasiment complet de la
liagne S et qui va de la Contrescarpe à Champerret. Je vis et
remarquai un jeune homme et un vieil adolescent assez ridicule
et pas mal grotesque: cou maigre et tuyau décharné, ficelle et
cordelière autour du chapeau et couvre-chef. Après une
bousculade et confusion, il dit et profère d’une voix et d’un ton
larmoyants et pleurnichards que son voisin et covoyageur fait
exprès et s’efforce de le pousser et de l’importuner chaque fois
qu’on descend et sort. Cela déclaré et après avoir ouvert la
bouche, il se précipite et se dirige vers une place et un siège
vides et libres.

Deux heures après et cent vingt minutes plus tard, je le rencontre

et le revois Cour de Rome et devant la gare Saint-Lazare. Il est et
se trouve avec un ami et copain que lui conseille de et l’incite à
faire ajouter et coudre un bouton et un rond de corozo à son
pardessus et manteau.
 12

A list of the titles of the exercices, which immediately precedes the Appendices,

will give the reader an idea of how Queneau adapted this text ninety-seven

more times.

Thus, while every one of the exercices recounts more or less the same

event, each one remains distinct from the other ninety-eight. And just as the

Bach listener perceives a constant, recurring theme in each variation, so the

reader of Queneau's book understands that there is something undeniably the

same (invariant) about each of the different exercices (variants).

Given that we are discussing not only Exercices de style but also its

translations into English and Italian, the idea of equivalence in difference

becomes even more germane, for what is the relation between a source text

and its translations if not one of equivalence in difference? Indeed, as J.C.

Catford, one of the pioneers of linguistic theories of translation, wrote: "The

central problem of translation-practice is that of finding target-language

translation equivalents. A central task of translation-theory is that of defining

the nature of conditions of translation equivalence" (Catford, 1965:21).

Subsequent theorists have added the terms functional equivalence, stylistic

equivalence, formal equivalence, textual equivalence, communicative

equivalence, linguistic equivalence, pragmatic equivalence, semantic

equivalence, dynamic equivalence, ontological equivalence, and others

(Gorlée, 1994:170).

This plethora of different types of equivalence would seem to suggest

that translation theorists have come to the sensible, though usually tacit,

conclusion that a translation may be equivalent to the source text in one way

and not equivalent in another. In other words, a source text and its

translation(s) are inevitably both equivalent and different. As Jakobson wrote

in his famous essay "On linguistic aspects of translation": "...translation

 13

involves two equivalent messages in two different codes. Equivalence in

difference is the cardinal problem of language and the pivotal concern of

linguistics" (Jakobson, 1959:233). It is also the pivotal concern of translation in

all senses of the word. Equivalence in difference, or invariance in variation

was, according to Jakobson, the common thread in his work from 1911

onward (Jakobson, 1990:61). Although Jakobson was primarily concerned

with interlingual translation in his essay, he also characterized two other types

of translation: intralingual and intersemiotic. Interestingly, Queneau's

Exercices de style provides an almost prototypical example of the former, and

hence its translations are (interlingual) translations of (intralingual)

translations and involve one level of equivalence in difference, i.e. the

equivalence between the various exercices, within a larger system of

equivalence in difference, that of the original and its translation.

1.1.1 On Sturrock on Jakobson on Translation

Jakobson's essay is often mentioned in the field of translation studies

but rarely examined in depth. The only full-length scholarly treatment that I

know of is J. Sturrock's 1991 contribution. Since Jakobson's essay and the

Peircean concepts it refers to provide the basis for my approach to Exercices de

style, I will spend some time discussing the essay and a few of Sturrock's most

relevant objections. These objections, I believe, are quite simply the result of

Sturrock's misreading of Jakobson.

In the opening paragraph of his essay, "On Jakobson on translation",

Sturrock correctly observes that "...where there is meaning there is also

translatability, or the possibility of interpretation, the act by which new signs

are asked to do duty for old" (Sturrock, 1991:307). One might add that these

new signs relate semiotically to the old signs as interpretants (in the Peircean
 14

sense) of the latter. The relationship between a sign and its interpretant sign is

one of equivalence in difference. Different signs never mean the same thing,

and even identical signs can never mean the same thing, as a sign signifies

within a spatio-temporal, socio-cultural context which is by definition in a

state of flux.

Sturrock at first agrees with Jakobson's characterization of translation

as interpretation or semiosis, but then qualifies this by adding that "...in

contemporary English usage, a distinction would normally be made between

the translation and an interpretation of an original text--translation being

looked upon as a more rigorous semantic exercise than interpretation, whose

practitioners enjoy a greater license in determining the meaning of the signs

they are working on..." (Sturrock, 1991:309). Jakobson chose the word

"interpretation" in reference to Peircean semiotics, with which he had been

acquainted since the early 1950's, and in this context, Jakobson's use of the

term is entirely correct (see Jakobson, 1990:19).

Jakobson's distinction between intralingual, interlingual, and

intersemiotic translation provides an excellent framework for the various

editions and versions of Exercices de style All three types of translation involve

an interpretation of signs by other signs. In the case of intralingual

translation, verbal signs interpret verbal signs within a single language. In

interlingual translation the verbal signs come from two different systems.

Finally, intersemiotic translation involves verbal and non-verbal sign systems.

Each edition or version of Exercices de style represents an example of

intralingual translation, or more precisely, ninety-eight examples. The English

and Italian versions are interlingual translations of, respectively, the original

and modern French editions. Finally, there exist intersemiotic translations of

Exercices de style into both musical and pictorial forms (Bens, 1962:240-249).
 15

Sturrock (1991:312) reproaches Jakobson for blurring the distinction

between synonymy and equivalence. I believe Jakobson was quite clear and

shall attempt to explain my reading in what follows. Jakobson did not believe

that translations are synonymous with the original, but rather interpretations

of it which might be perceived as more or less equivalent depending on the

translation. The disjunction he creates between synonymy and circumlocution

(1959:233), or more precisely, between synonymous word (synonym) and

circumlocution, seems clear enough: "grub" is a synonym of "food", whereas

"animal or vegetable matter consumed for sustenance" is a circumlocution.

Synonymy has to do with two code units of the highest level (a word or

idiomatic phrase), while circumlocution has to do with a code unit and a

sequence of code units. In his writings Jakobson uses synonym/synonymy

/synonymous to refer to the relationship between code units within the same

code. He states this explicitly in "Shifters and Verbal Categories", in which he

writes: "Any elucidating interpretation of words and sentences--whether

intralingual (circumlocutions, synonyms) or interlingual (translation)--is a

message referring to the code" (Jakobson, 1990:388).

Equivalence is a more general concept which can apply not only to the

intralinguistic phenomena of synonymy and circumlocution, but also to code

units (or sequences of them) compared interlinguistically. In any case, he

never meant to imply that synonymy was absolute, when he wrote "... more or

less synonymous.." and "...synonymy, as a rule, is not complete equivalence"

(Jakobson, 1959:233). Concerning equivalence, Jakobson was equally clear,

stating: "Likewise on the level of interlingual translation, there is ordinarily no

full equivalence between code units..." (Jakobson, 1959:233).

I hope to have convincingly shown that not only does Jakobson clearly

differentiate between synonymy and equivalence, but also that he does not
 16

conceive of either one in terms of absolutes. In the discussion following a

lecture he gave in 1972 he says of himself: "...I am a relativist to the end and

would never say deep but deeper or less deep." (Jakobson, 1990:322)

Sturrock objects to Jakobson's definition of interlingual translation as

"...two equivalent messages in two different codes" (Jakobson, 1959:233), in the

course of his critique of the "semantic Platonism" inherent in the "intuitive

notion of synonymy with which we habitually operate" and the "Saussurean

analysis of the linguistic sign, as that has generally been understood"

(Sturrock, 1991:313). Sturrock appears to not realize that these

characterizations apply neither to Jakobson's notion of synonymy nor to

Jakobson’s reading of Saussure. Sturrock correctly argues against the

misreading of Saussure which gives rise to such spurious concepts as the

"transcendental signified", and convincingly demonstrates how a strictly

Saussurean conception of the linguistic sign is incompatible with the notion of

its arbitrariness or non-arbitrariness (Sturrock, 1991:314-318). However, he

fails to recognize that Jakobson neither misread Saussure in this way, nor

strictly adhered to Saussure. Jakobson's conception of the linguistic sign,

which draws upon both Saussurean and Peircean semiotics, is perfectly

consonant with the notions of arbitrariness and iconicity.

Due to this confusion, when Sturrock comments on Jakobson's

statement, he is convinced he sees in it a misreading of Saussure. He deems

that:

The notion of `codes' and `recoding' is not a happy one, for a

reason which by now I need hardly spell out. It is Platonist
through and through. For what is `coded' in language if not
`meanings', those supra- linguistic entities the translator has
been commissioned to effect? In Jakobson's model the translator
 17

can but `decode' the `message' of the original text, before

`recoding' it into what he believes to be an equivalent form in a
second language. [...] And then, if the communicational model of
translation involves a decoding of the original followed by a
recoding, what is the original decoded into? The model invites
us to posit an intermediary `language', or Code of Codes,
between source and target languages, as an ineffable repository
where the `message' may be stored before being recoded
(Sturrock, 1991:319).

Sturrock has just oversimplified Jakobson's model here. He speaks of code

and message as if Jakobson intended for them to be considered as entirely

independent of one another. Instead, it is the subtle interplay of code and

message, invariance (context-independent meaning) and variation (context-

dependent meaning) that characterizes Jakobson's model. Hence, meaning

does not reside in the code as Sturrock would have it, but rather in both the

code and the message. The former contains invariant meaning, while the

latter engenders contextual meaning. Even this is simplifying things to a great

extent as some contextual meaning is also part of the code (e.g. basic

meaning), while variation provides the basis of subcodes and the invariant

resides in each contextual variant in each message. Sturrock wonders what a

message is decoded into, claiming that the terminology implies a "Code of

Codes". This does not reflect Jakobson's ideas. For him, as for Peirce,

understanding is the interpretation of signs by other signs (interpretants),

hence all decoding is recoding.

Sturrock concludes that Jakobson's essay "does little to support the idea

that a semiotic terminology is helpful when trying to understand the nature of

translation" (Sturrock, 1991:320). I claim instead that when reading the essay

in the context of Jakobson's other writings the relevance of semiotics to

 18

translation theory is obvious, particularly with regard to the concept of

equivalence in difference.

1.1.2 Translation and the Peircean Sign

Assuming that language is a system of signs, and that the study of

linguistic signs (linguistics) is hierarchically subordinate to the study of signs

in general (semiotics), it is interesting to note that the notion of equivalence

plays an equally vital role in semiotics. In Saussurean semiology equivalence

is the organizing principle of the paradigmatic axis, and in Peircean semiotics

it is the means by which signs signify. Semiosis depends crucially on

equivalence. As Peirce states in his oft quoted definition, "A sign is something

which stands to somebody for something in some respect or capacity. It

addresses somebody, that is, creates in the mind of that person an equivalent sign

(italics mine), or perhaps a more developed sign" (Peirce, 1932:135). This

equivalent sign or interpretant, to use Peirce's terminology, is, as Savan notes,

"...a translation of an earlier sign" (Savan, 1987-88:17). He explains that for

Peirce "...interpretation is translation" (ibid). Jakobson appears to concur in

"On linguistic aspects of translation", and Waugh writes "...interpretability is

translatability, the potentiality in the signatum of being translated into another

signum (sign)..." (1984:412). In its most general sense then, translation is the

interpretation of a sign by another, more or less equivalent sign. From this

perspective, (mis)translation underlies all (mis)communication. Referring to

Jakobson's tripartite typology of translation Waugh notes that translation can

involve a more explicit or less explicit sign, "...a translation may reduce or

enhance redundancy, reduce or enhance ambiguity" (1984:413).

In Peircean theory the sign is a process powered by the act of

interpretation. The sign consists of a representamen, an object, and an

 19

interpretant. The interpretant is in turn another representamen which refers

to another object and generates another interpretant ad infinitum. In fact the

picture is significantly more complex than I portray it here. For one, the

concepts of firstness, secondness, and thirdness lead to the hypothesizing of

an immediate object (firstness), and the dynamic or signified object

(secondness), as well as an immediate interpretant (akin to Jakobson's

invariant), a dynamical interpretant (akin to Jakobson's contextual variant),

and a final interpretant (a potential rather than a reality unless semiosis comes
to an end). The signified of the Saussurean sign is part object and part

interpretant in the Peircean system.

As Eco explains (1976:68), "The interpretant is that which guarantees

the validity of the sign...it is that which the sign produces in the quasi-mind

which is the interpreter; but it can also be conceived as the definition of the

representamen..." In Peirce's model the interpretant is itself a representamen

which refers to another object, hopefully a more or less equivalent one. Thus

the exercices are both representamina and interpretants. The concept of

equivalence in difference applies to both signifier and signified which in the

Peircean system includes representamen, object, and interpretant. Eco further

notes that, "The idea of the interpretant makes a theory of signification a

rigorous science of cultural phenomena, while detaching it from the

metaphysics of the referent" (1976:70). Indeed, while in the next chapter I shall

use the concept of referential information, it should be clear that this is not

meant to refer to a real-world referent but instead to its signification.

1.2 Background on Queneau

Raymond Queneau was born on February 21, 1903 in Le Havre where

he completed his elementary and secondary education. He received his

 20

baccalauréat with honors in philosophy from the Lycée du Havre in 1920, then

moved to Paris to study philosophy and mathematics, and eventually

obtained a licence in philosophy. Although he never earned a university

degree in the mathematics, he spent a few hours every day on math (Bergens,

1963:10), was a member of both La Société Mathématique de France and the

American Mathematical Society, and published his findings about s-additive

sequences in the Journal of Combinatorial Theory in 1972.

He became part of the surrealist movement in 1924 while studying in

Paris, and contributed to the movement's journal, La révolution surréaliste, but

eventually broke ties with the group definitively following a disagreement

with André Breton in 1929. As Hale notes though, "...it is clear from

Queneau's later literary and theoretical writings that his attitude toward

literary creation as the product of conscious, voluntary, disciplined formal

research would have sooner or later precipitated a break with this group over

doctrinal differences" (1989:10-11).

In 1934 he began translating (with his wife Janine Kahn, Breton's sister-

in-law!) Edgar Wallace's Kate Plus Ten. In the following years Queneau

continued his English to French translations of a wide assortment of writers

including Hart Crane, Sinclair Lewis, Vachel Lindsay, Cotton Mather, George

du Maurier, Henry Miller, Marianne Moore, Maurice O'Sullivan, Wallace

Stevens, Amos Tutuola, Walt Whitman, and William Carlos Williams.

Queneau joined the College of 'Pataphysics (an absurdist "anti-

academy" founded in post-war Paris) in 1950 and held the titles of

"Transcendant Satrape" and "Grand Conservateur de l'Ordre de la Grande

Gidouille." An interesting note with respect to the concept of equivalence is

Hale's remark, "The College was founded upon the theory of equivalences,
 21

i.e., that any one thing is worth any other, or that everything is the same thing"

(1989:27) .

In 1960 he co-founded with François LeLionnais the Ouvroir de

Littérature Potentielle (Oulipo) which began as a Subcommission of the

College of 'Pataphysics. The activities of the group centered around

anoulipisme, the analysis of existing literary works in search of forms,

patterns, or structures, and synthoulipism, the synthesis of new forms, patterns,

or structures. Queneau's essay in Bâtons, chiffres et lettres provides an excellent

introduction to the activities of the Oulipo (Queneau, 1965:319-345).

1.3 Genesis of the Work

We know, according to Michel Leiris, how and at approximately what

point in time Queneau got the inspiration for Exercices de style. Leiris states in

an interview with Jacques Bens:

Dans le courant des années 30, nous avons entendu ensemble à

la Salle Pleyel un concert où l'on donnait L'Art de la fugue. Je me
rapelle que nous avions suivi cela très passionnément et que
nous nous sommes dit, en sortant, qu'il serait bien intéressant de
faire quelque chose de ce genre sur le plan littéraire (en
considérant l'oeuvre de Bach, non pas sous l'angle contrepoint et
fugue, mais édification d'une oeuvre au moyen de variations
proliférant presque à l'infini autour d'un thème assez mince). A
mon avis Queneau y est parvenu avec les Exercices de style, qu'il
serait erroné de regarder comme relevant du seul humour (Bens,
1962:12).

Queneau gives essentially the same version in the preface to the 1963

edition of Exercices de style, although he wonders if the concert he and Leiris

attended really took place before the second World War. In a radio interview
 22

with Georges Ribemont-Dessaignes the author recounts: "Les Exercices de style:

je suis parti d'un incident réel, et je l'ai raconté d'abord douze fois de façon

différente, puis un an plus tard j'en ai refait douze autres, et finalement il y en

a eu quatre-vingt-dix-neuf." (Queneau, 1965:43)

1.3.1 Editions and Versions

In this dissertation I use the word "edition" only in reference to the

numerous manifestations of the work in French. I reserve the word "version"

for referring to foreign language translations of the work. For the sake of

brevity let it be simply said that Exercices de style has been reissued on multiple

occasions since its original printing in 1947. Until 1973 the editions differ only

very slightly. A certain number of revisions of what appear to have been

typographical errors can be seen by comparing the original edition with the

revised and corrected edition from 1964. Nonetheless, a number of "errors" or

"oversights" continue to appear in the text even in the 1964 edition. These

"errors", which often involve only a letter or two in an entire exercice, are not

easy to identify and may have been overlooked. Examples of these "errors"

may be found in Appendix C.

While I believe the idea of "error" or "lapsus" to be a fruitful one for the

analysis of the work (I am thinking in particular of the Oulipian interest in the

clinamen), that is not the approach taken here. To begin with, it would be

nearly impossible to obtain an accurate picture of what is "error" and what is

not without consulting Queneau's manuscripts for Exercices de style, and also

studying the many different editions of the work. The preeminent Queneau

scholar Emmanuel Souchier did just that in his doctoral thesis, which is

unfortunately not available in this country.

 23

From what I can gather, reading the abstract of his dissertation and

selected articles drawn from it, Souchier concerns himself primarily with

literary history. As Souchier explains it, his approach aims to open "...les voies

d'une lecture susceptible de livrer un certain nombre d'informations sur les

procédés d'associations et de raisonnements, bref, sur ce qu'on a appelé les

«réseaux cognitifs» de l'auteur." (Souchier, 1985:183) Souchier's analysis of a

particular exercice, Translation, will come up in Chapter 3.

Not all of the editions are mere reprintings. In many cases the text has
been augmented with prints, lithographs, and even typographical exercices de

style. The text changes radically however in the 1973 edition, and these

changes remain in all subsequent editions of the work. In what I shall

henceforth call the modern edition, i.e. any edition published since 1973, six

exercices have been eliminated (Permutations de 2 à 5 lettres, Permutations de 9 à

12 lettres, Réactionnaire, Haï Kaï, Féminin) and replaced with new ones

(Ensembliste, Définitionnel, Tanka, Translation, Lipogramme). Eight other exercices

appear with different titles (Homéoptotes > Homéotéleutes, Prétérit > Passé simple,

Noble > Ampoulé, Permutations de 5 à 8 lettres > Permutations par groupes

croissants de lettres, Permutations de 1 à 4 mots > Permutations par groupes

croissants de mots, Contre-vérités > Antonymique, Latin de cuisine > Macaronique, A

peu près > Homophonique). The necessity of choosing one original and

one modern edition of Exercices de style to work with had an impact on some of

the data presented in Appendix C. In particular, many of the "errors"

perceived in the exercices based on permutation were rectified in the 1964

corrected original edition which preceded the modern edition. What I am

calling the original edition is any edition published prior to 1964.

Barbara Wright's British English version of Exercices de style, which first

appeared in 1958, is based on the original edition of the work. Although this
 24

version was reprinted once in England, and released in the United States later

on, the translation was never updated to reflect Queneau's changes in the

modern edition. It must also be noted that the British English translation

remains somewhat inaccessible, or at the very least odd-sounding, to many

American readers. Eco's Italian version was published for the first time in

1983. He bases his translation on both the original and modern French

editions.

1.3.2 How Many Exercices?

In all editions Queneau includes the same number of exercices;

whenever he deletes he adds anew. Queneau certainly did not exhaust his

stock of ideas, in fact, in a special 1963 edition of the work (Queneau,

Carelman, Massin, 1963:97-98) he includes an annex of titles of "exercices de

style possibles" containing not less than 124 new ideas. One must then

wonder if the number 99 holds some special significance for the author.

Indeed, Queneau himself states (1965:29) that he never could let chance dictate

the number of chapters in his novels. He also writes of his particular fondness

for certain numbers: "...quant à 7, je le prenais, et puis je le prends encore

comme image numérique de moi-même, puisque mon nom et mes deux

prénoms se composent chacun de sept lettres et je suis né un 21 [3 x 7]" (ibid).

Considering that many of the important secondary sources on

Queneau make note of this (Bens, 1962:56; Bergens, 1963:95; Guicharnaud,

1965:11; Shorley, 1985:62; Jouet, 1988:63; Hale, 1989:118; Renard, 1995:189), it

seems odd that none but Renard has noticed the connection to the number of

exercices. By his account, "Si l'on isole le texte central de l'ouvrage

(«Maladroit», le cinquantième), on obtient: 49 (7 x 7) + 1 + 49 (7 x 7). Ce

quadruple 7 peut apparaître comme une signature de l'écrivain..." (Renard,

 25

1995:189). Another possibility worth considering involves only the three

sevens representing Queneau's three names, i.e. a first and middle name plus

a family name. An even stronger argument can be made for the number

ninety-nine being a numerical signature when only three sevens are involved.

We know from Queneau (1963:9) that Notations, the first exercice in Exercices de

style, was also the starting point in terms of Queneau's writing of the work. If

we exclude this initial exercice, ninety-eight variations remain. Does it not

seem plausible that Queneau arrived at this number by means of his cherished
seven? A first and middle name of seven letters each, or (7+7), and a last name

of seven letters (7). The product of the sum of his two first names and his last

name, (7 + 7) x 7, gives the number of exercices written after his "notes"

(Notations).

Queneau addressed the significance of the number ninety-nine, but was

not consistent. In 1953 he said that originally the project was to include more

exercices but that he stopped out of laziness and for fear of boring the reader.

Ten years later he states that ninety-nine is a satisfying quantity, neither too

few nor too many, and invokes the Greek ideal (Renard, 1995:171-172).

Whatever explanation one prefers, it remains true that the number ninety-nine

connotes abundance (ninety-nine is many) yet incompleteness (it is not quite a

hundred).

1.4 Queneau on Exercices de style

Queneau occasionally discussed his Exercices de style. In the previously

cited conversation with Georges Ribemont-Dessaignes (Queneau, 1965:43-44)

he states:
 26

Les Exercices de style... On a voulu voir là une tentative de

démolition de la littérature, ce n'était pas du tout dans mes
intentions, en tout cas mon intention n'était vraiment que de
faire des exercices, le résultat, c'est peut-être de décaper la
littérature de ses rouilles diverses, de ses croûtes. Si j'avais pu
contribuer un peu à cela, j'en serais bien fier, surtout si je l'ai fait
sans ennuyer trop le lecteur.

Queneau agrees with André Gillois who senses in Exercices de style

"...une satisfaction d'ordre sensuel à manier les mots et à s'en servir de toutes

les manières possibles" (Bens, 1962:216), but to Gillois' statement that the

essential meaning of the work is "the end of literature" he replies: "Non, c'est

décanter, décaper plutôt, la littérature de choses futiles que souvent on

n'aperçoit pas; et là peut-être ces choses futiles sont-elles mises en valeur pour

précisément atteindre une moindre futilité." (Bens, 1962:218)

1.5 Secondary Sources on Exercices de style

Most of the secondary sources available for consultation have little to

say about Exercices de style. The only sources I found dealing exclusively with

this work, apart from Souchier's previously discussed dissertation, are: 1)

Claude LeRoy's "Etude sur la perte d'information et la variation de sens dans

les «Exercices de style» de Raymond Queneau" (Queneau, Carelman, & Massin,

1963:99-114) which appears to be based largely on psychoanalytic theory; 2)

Paul Miclau's "Structure et information dans «Exercices de style» de Raymond

Queneau" (1967:345-358) which combines statistics and stylistics; and 3)

Francesco Ragusa's 1988 Raymond Queneau attraverso gli Esercizi di stile, an

essentially biographical and historical piece. None of these proved especially

relevant to my work. This lack of commentary may reflect the fact that the

opus does not represent any single genre. It is neither a novel, nor a play, nor
 27

an essay, nor prose, nor poetry. Though it tells a story it is certainly more than

a short story, but also less than one since it takes at most two pages to recount.

And it could hardly be called a collection of short stories as each story consists

of more or less the same events.

Some critics (e.g. Shorley, 1985:15) refer to it as a literary experiment,

while no less an authority than François Le Lionnais considers it akin to the

work Queneau did with Oulipo (Bens, 1962:22), although its original

publication predates the formation of the group. Chronology aside, I would

maintain that Exercices de style quintessentially exemplifies potential literature;

like Queneau's Cent mille milliards de poèmes its structure contains the

possibility of infinite production. There is literally no end to the potential

number of exercices that might be written following Queneau's ninety-ninth.

The direct influence of Queneau's work with the Oulipo cannot be overlooked

in the modern edition. Définitionnel, Tanka, Translation, and Lipogramme clearly

reflect Oulipian experimentation. As Souchier (1985:187) points out, all of

these techniques are "...répertoriés et développés dans les annales

oulipiennes."

Some sources also recognize the impact of Exercices de style on the work

of Oulipo. Jouet (1988:33) for example, notes that "[t]out le travail oulipien de

manipulation-traduction de textes sources procède des Exercices de style

(variations de Perec sur «Longtemps je me suis couché de bonne heure...»,

microtraductions du poème de Verlaine «Gaspard Hauser chante», chapitre

«Trans» de l'Atlas de littérature potentielle) et lorsque Raymond Queneau

s'exerce à la toute neuve méthode S + 7, il choisit de traiter...l'un des Exercices

de style."

Most secondary sources on Queneau remark on his use of a spoken

style of French; quantifying and tracking such usage is the topic of a section in
 28

the following chapter. Queneau traces his familiar style to several sources

including popular writing, the military barracks, Vendryes' Le Langage, and his

acquaintance with the two varieties of Modern Greek (cathaverousa and

demotic) (Bens, 1962:50). Hale (1989:93) comments that "Queneau is

extremely sensitive to, and interested in , the variety of linguistic registers,

levels, or tones of language, and his oeuvre is replete with examples of every

imaginable style of French--written and oral, archaic and ultra-modern, formal

and casual, learned and illiterate, abstract and concrete, original and banal.
When, as is often the case, the tone adapted is ill-fitted to the occasion, the

result is surprising, delightful and amusing." Nowhere could these

observations be more relevant than in the case of Exercices de style. Jouet also

seizes upon Queneau's fascination with the complexity of language noting

that for the writer "...la langue n'est pas un absolu. Elle n'est pas donnée à son

utilisateur comme un bloc achevé, inerte, incontestable ou uniforme. La

langue est un réseau complexe de sous-langues spécifiques" (1988:27). I

cannot help but be struck by the similarity of this conception of language to

that of Jakobson and Waugh. As the latter writes in "Some Remarks on the

Nature of the Linguistic Sign": "It should not be assumed that the code

described here is a monolithic, undifferentiated code: it is rather a code

consisting of subcodes, a `system of systems', a multiform code which

commands a variety of styles of speech..." (1984:393).

Many scholars other than Souchier have remarked on the significance

of error in Queneau's work. Jouet, for example, mentions Queneau's fondness

for "la discordance des temps des verbes," and "les fautes d'accord en genre"

(1988:27). While he does not give any examples from Exercices de style, a few

grammatical mistakes can be found. For example, the exercices Ignorance and

Alors both show instances of anomalous auxiliation (i.e. Il s’aurait querellé, j’ai
 29

monté), Vulgaire is missing an indefinite article (avec autre zozo), and

Exclamations has a finite verb form where one would expect an infinitive (Ils

vont se fiche des gifles). Jouet further remarks that in such cases "...corriger

consiste à fabriquer de l'erreur" (1988:28). He also notes the implications of

error within the context of Oulipo, writing: "La déviance, l'erreur de calcul, le

lapsus, le clinamen (qui, pour l'Oulipo, sera une déviance programmée)

peuvent jouer un rôle fécond dans la recherche scientifique, comme dans la

pratique littéraire. Et bannie par les clercs, Queneau l'accueillera toujours, la

bonne erreur..." (Jouet, 1988:21).

While Queneau clearly prizes form and style, it seems to me that he

nonetheless conveys information in his exercices. It was thus heartening to

read Shorley's remarks concerning Exercices de style in his recent critical work

on Queneau. Shorley confirms my contention, writing that although

"...language obviously takes precedence over content, information -- albeit of

the most banal kind -- is being communicated" (1985:57). I hasten to add that

the banal information Shorley alludes to is what linguists refer to as

referential.

Since I argue that equivalence in difference, the structuring principle

behind all types of translation, is also the structuring principle behind

Exercices de style, the following quote from the introduction of one of the most

recent critical works on Queneau seems particularly illuminating:

The concept of translation is a useful analytical tool for the study

of Queneau. The Latin root of to translate means, literally, "to
carry across or beyond," and all of Queneau's work may be read
as a crossing of established intellectual and aesthetic boundaries
in an effort to transcend the limitations they impose upon his,
and our, mind and imagination. While Queneau's translation
efforts have taken many forms, they are all results of his lifetime
 30

quest for synthesis among various domains of thought and

expression. They include: actual translations of English
literature into French; the editing of an encyclopedia that
translates the erudition of many disciplines into organized
summaries for nonspecialists; the translation of darkness and
solemnity into lightness and laughter through both Queneau's
lucid perception of the tragedy of the human condition and his
renowned gift for humor; the translation of one style or level of
the French language into another; the translation of characters,
events, languages, and images from many different places and
periods into his own twentieth-century French literary universe;
and the translation of scientific and mathematical concepts and
forms into a more widely accessible poetic language (Hale,
1989:3).

I have already discussed this broad conception of translation in the theoretical

section of this chapter, and shall return to it again in Chapter 3 when

analyzing Queneau's exercice entitled Translation.

1.6 Reception and Equivalence

Shorley (1985:205) notes that Queneau first gained recognition with

Exercices de style thanks to its musical adaptation by the Frères Jacques (what

Jakobson would have undoubtedly called an instance of intersemiotic

translation). The book did not receive a great deal of attention initially: the

reviews appearing in 1947, while recognizant of Queneau's literary talent, do

not seem to take the work very seriously. Only later as it gained enough

popularity to allow for numerous different editions and reprintings did it

begin to receive the serious attention of critics. By the time the modern edition

of the work appeared in press Queneau was widely recognized as a literary

genius and his Exercices de style was considered a masterpiece. Barbara

Wright's English version received accolades from the moment it first

 31

appeared, as did Umberto Eco's Italian version. While I do not intend to

plunge into the deep and murky waters of reception theory, it does occur to

me that critical reception represents another possible basis for comparison of

editions and versions. It offers another perspective in which equivalence in

difference provides a powerful analytical construct. In terms of reception, the

modern edition of Queneau's work and Wright's English version of the

original edition are more equivalent than the original and modern editions,

since the original edition of Exercices de style was not nearly as well received as
the modern edition, while the modern French edition and Wright's English

version were both well received. This type of approach recalls the final of Jiri

Levy's three relationships defining the problem of translation, namely, "[t]he

relationship between the ultimate value of the original work and that of the

translation (here we would apply the methods of literary criticism)" (Levy,

1976:226).

1.7 Concluding Remarks

In the theoretical section of this chapter I have shown that equivalence

in difference as conceived by Jakobson plays a central role in linguistics and

all types of translation, including the production of meaning through semiosis.

Any understanding of a sign involves its translation into another sign (the

interpretant), and so equivalence in difference is the principle that allows us to

communicate using linguistic sign systems. I conclude that Jakobson's

thinking on translation, as well as the Peircean model of the sign, provide an

excellent framework for the analysis of Exercices de style and its English and

Italian translations. As part of the background on Queneau and Exercices de

style I clarify the relationship of the two translations to their source texts. I

discuss the probable significance of the number of exercices, and explain why
 32

the Italian translation differs fundamentally from both the original and

modern French editions. Now that I have established a theoretical framework

and given some background on the text and its author, I will show how the

theory can be applied to the analysis of Exercices de style and the English and

Italian versions of the same.