Warren F. Motte Jr. ed., (1998) OuLiPo: a Primer of Potential Literature, trans. Warren F. Motte, Illinois: Dalkey Archive Press.


Ouvroir de Littˇrature Potentielle can be defined more closely by drawing attention to the title itself. Firstly, the multiple meanings of 'ouvroir,' in French, include a place where people work together on a difficult task deriving new techniques, and more precisely a sewing circle that operates as a rich metaphor. Workroom is inadequate to describe this, from 'ouvrer' to work, as it does not amply describe the work, or indeed play, of literature as praxis. The word 'potential' too is suitably descriptive. Raymond Queneau describes this as: 
'less a question of literature strictly speaking than of supplying forms for the good use one can make of literature. We call potential literature the search for new forms and structures which may be used by writers in any way they see fit.' (Motte, 1998: 38) 

The concerns over form and structure were shared by structurAlism, although a clear distinction suggested by Francois Le Lionnais in the term 'structurElism' to emphasise the Oulipo's concerns over structures and the inevitable levels of constraints (over form and constraints of literary genre, etc). The concerns are syntactic rather than semantic in this sense: 'Indeed the creative effort in these works is principally brought to bear on the formal aspects of literature: alphabetical, consonantal, vocalic, syllabic, phonetic, graphic, prosodic, rhymic, rhythmic, and numerical constraints, structures, or programs.' (Le Lionnais, in Motte, 1998, 29) The 'virtualities of language are revealed by constraint' and the use of rules 'so cherished by the classics, were principally used as a means of channelling eventual overflowings of a poorly controlled verbal flood' (Marcel Bˇnabou, in Motte, 1998: 43). Through structure, linguistic objects might be organised into new arrangements. An example of automatic transformation of text, alluding to cryptography, is Jean Lescure's S+7 method in which a text is taken and each word (s for substantive) is replaced by the seventh following it in a dictionary. Raymond Queneau, who proposed the 's-additive' series, points out that if a 2000 word dictionary is used, the S+2000 method would produce an exact copy of the original (Motte, 1998: 61). In this sense, and according to Perec, 'the Book is a cryptogram whose code is the Alphabet' (in Motte, 1998: 96), 

The legacy of semiotics, structuralism and formalism is clear - in Vladimir Propp's analysis of combinatory and permutational forms not least, or J. R. Pierce's semiotics. To the Oulipo, two general principles apply: the analysis and appropriation of historical experiments in form; and, the synthesis or elaboration of new forms. Le Lionnais summarises this as 'Anoulipism' (analytic mode) devoted to discovery and 'synthoulipism' (the synthetic mode) to invention (Motte, 1998: 2). Here we have the quarrel of the ancients and the moderns comparable to the 'laboratory synthesis of living matter' (Le Lionnais, in Motte, 1998: 30). When something thought new is unwittingly discovered to be invented in the past, as is often the case, Le Lionnais refers to this as 'plagiarisms by anticipation' (Motte, 1998: 31; referring to Lautrˇamont's 'plagiarism is necessary'). In this way, Georges Perec's La Disparation (1969), in which a 300 page novel is written without using the letter E, can be seen as less a work of avant gardist experimentation and more a contemporary example of the literary tradition of the 'lipogram' traced back to the sixth century BC. Perec himself studied the 'History of the Lipogram,' (in Motte, 1998: 97-108) so was aware of the long tradition that includes, for instance, Lope de Vega who wrote five stories without each vowel in turn; and most tellingly Ernest Vincent Wright, who in 1939 wrote the novel Gadsby: A Story of Over 50,000 Words without Using the Letter E. (1998: 106)
(Note: The reverse constraint of La Disparation is Les Revenentes (translated as The Exeter Text) that only uses the vowel E.)   

Queneau's Cent Mille Milliards de Poemes (one hundred thousand billion poems) (1961) is a classic example: ten sonnets that can be arranged according to formal rules. To each of the ten first lines, the reader can add any of ten different second lines, and so on. The sonnet has fourteen lines so the possibilities are of the order of 10 to the power of 14, or one hundred trillion sonnets. Le Lionnais, in the postface to the work, describes the 'technical superiority, the work you are holding in your hands represents, itself alone, a quantity of text far greater than everything man has written since the invention of writing' (Motte, 1998: 3). The potential writing in this sense implies its potential reading. It is exponential. This exemplifies the Oulipean project in both taking an analytic and synthetic approach: the traditional constraining form of the sonnet and imposing a multiplicity of constraints upon it 'arbitrary at the outset but [that] become highly codified through use (and it is precisely this "use" that separates the normative text from the experimental). (Motte, 1998: 4) The full potential of this work lies unrealised for practical reasons: literally existing in a potential state. 

In emphasising formal constraints, inspiration and other literary myths of creation and genius, are devalued. Constraint, in this sense, works on a number of general levels, suggested by Motte as: firstly, a minimal level of language itself; secondly, an intermediate level of genre and literary norms; finally and the one concerning the Oulipo, a maximal level 'of consciously preelaborated and voluntarily imposed systems of artifice' (1998: 11). This is clear when Le Lionnais states that: 'The efficacy of a structure - that is, the extent to which it helps a writer - depends primarily on the degree of difficulty imposed by rules that are more or less constraining' (1998: 11). In the Oulipean laboratory, a constraint is a formula that is 'proven' by the production of a text. In La Disparation, a simple constraint is imposed that causes immense difficulty on implementation. La Disparation is a good example of the use of constraint in a fuller reflexive sense as a novel about disappearance of the letter E, 'thus both the story of what it recounts and the story of the constraint that creates that which is recounted' (Roubaud, in Motte, 1998: 12). (Note: The idea of parody is evident here too - in 'heteroparody' in which others work is imitated or 'autoparody' where the author imitates their own work.)  

The central analogy is between mathematics and literature, itself invoking a long tradition (the influence of Nicolas Bourbaki being one notable influence - see notes elsewhere - and the game theory of von Neumann, and the literature of Lewis Carroll). The mathematician Queneau, probably the main protagonist here, has contributed significantly to this mix: for instance, in the proposition 'The Relation X Takes Y for Z,' other binary, ternary, and 'n-ary' relations are imagined (Motte, 1998: 153-5); or where algebraic hypotheses are turned into text, such as in his 1955 Meccano (Motte, 1998: 82). In 'cellular prosody,' Conway's 'Game of Life' is used as a rule base, in a similar way to Mozart's 'musical game', using a card index to compose 'recurrent, iterative or recursive' forms (Motte, 1998: 115). Creative endeavour is seen to be programmed, and is considered in terms of its execution. Thus, the potential for permutations or 'combinatorics,' what Le Lionnais calls a 'combinatory literature', is expanded greatly by the computer and its systematic compositional structure. As an aside, Permutations is a Web site by Florian Cramer, that reproduces combinatory text systems, such as those of Ramond Queneau, in digital form.  This mechanistic approach is entirely in keeping with Oulipean 'anti-aleatoric' (anti-chance) procedures: 'Make no mistake about it: potentiality is uncertain, but not a matter of chance' (Bens, in Motte, 1998: 17). Rather, Oulipean texts are generated through the use of laws, through constraints. The ideas of chance, randomness and freedom of expression are all confined to the wastebasket in this sense. 

In a somewhat contradictory way, Le Lionnais argues that Queneau's interest in prime numbers was precisely because they 'imitate chance while obeying a law' (Motee, 1998: 18). Furthermore, what if the law itself is a random one, asks Motte, such that the system of constraint itself is generated in this way? This contradiction might open up new potentialities of course. What distinguishes the human use of constraints and the machine version is the error in the system. In 'Prose and Anticombinatorics,' Italo Calvino demonstrates the potential of the computer in serving this purpose, proposing that: 'the aid of the computer, far from replacing the creative act of the artist, permits the latter rather to liberate himself [sic] from the slavery of a combinatory search, allowing him also the best chance of concentrating on this 'clinamen' which, alone, can make of the text a true work of art.' (in Motte, 1998: 152; the 'clinamen' is the swerving of atoms from Epicurean atomic theory) 

Taking its inspiration from the bifurcating structure of computer program instructions, Queneau's 'A Story As You Like It,' (1998: 156-8) provides the reader with two choices of how to proceed at each stage of the story - towards a happy or unhappy ending perhaps. In an experiment to explicitly bring computer science and literary creation together, Paul Braffort was commissioned to program some of the Oulipo works such as Queneau's 'Cent Mille Milliards de Poemes' and 'A Story As You Like It.' (Motte, 1998: 140-2) In describing this enterprise of 'aided creation' or 'algorithmic literature', Paul Fournel argues that the machine allows the author top dominate the existing relations of computer, work and reader in new ways. Far from a deferral of authorship, the computer offers new potentialities for literary praxis. 

Alluding to dialectics, Perec proposes that introducing a flaw in the system breaks the symmetry: 'because when a system of constraints is established, there must also be anticonstraint within it. The system of constraints - and this is important - must be destroyed.' (Motte, 1998: 20) Citing Queneau's 'Cent Mille Milliards de Poemes', Harry Mathews describes the potential of the algorithm to compose and or decompose texts. The dialectical nature of this composition is even more evident his statement that: 
'Beyond the words being read, other lie in wait to subvert and perhaps surpass them. Nothing any longer can be taken for granted; every word has become a banana peel. The fine surface unity that a piece of writing proposes is belied and beleaguered; behind it, in the realm of potentiality, a dialectic has emerged.' (Motte, 1998: 126) 
For Mathews, this dialectic is virtually automatic.