This thesis is a study of how war comes to be known in French literary texts of the two World Wars.
My examination of the relationship between "war," "language" and "literature" shows how our understanding of each is determined by its interaction with the others. In the vast corpus of primary and secondary French writing concerning the two wars, one looks in vain for any account of the literary function of war, or indeed the function of literary war. Yet in discussions of literary "realism," it Is always likely that certain seminal war texts will be cited; conversely, in discussions of war texts, the gauge of •·authenticity" almost always comes to be used. This can be taken as indicating that there are literary strategies common to war texts, and perhaps peculiar to them, and that some version of "realism" Is the required outcome of such strategies.
In the preface to the study. I suggest that current preoccupations with literature which displays or questions the preconditions and conditions of its own construction have called for both a fresh look at a whole range of texts which have otherwise not excited interest, and a better definition of what it means to be "literary." These two concerns of what is broadly known as "metafiction" are in turn shown to be of possible major interest to a reading of the literature of war. Since in both the experiential and the literary sphere the First World War is thought of as effecting a radical disjuncture with what had gone before, war writing must at one and the same time reflect the new reality of war and deflect the tradition it opposes.
In introducing the texts for study, I first briefly consider some psychological, sociological and anthropological accounts of war, and suggest that their incompleteness points not so much to the inadequacies of those sciences as to the language in which they are couched. War itself is shown to be language-bound, not in a restrictive sense, but because it does not flourish in social systems where the language for war is not in place. That there is therefore a war in words seems to justify an examination of the language of war in fiction and the claims to realism it makes, that is, an examination of the referential function of war literature.
Further to this, I propose that referring to war in "literature" might be understood in two ways: firstly, through the breakdown of norms and the absence of a war tradition; secondly, by a strident rejection of literary antecedents to the text in hand, which in turn comes to question and possibly deny its own status as written. So in the decrying of older warfare and depictions of war, this war might assert itself textually in a way which is essentially non-literary.
The texts selected for exegesis are all French; most of them are written by ex-combatants. They are studied chronologically, because it is hoped that they illustrate a development, a sustained thematics of war writing. The texts are: Henri Barbusse, Le Feu (1916); Roland Dorgeles. Les Croix de bois (1919); Drieu La Rochelle, La Comedie de Charleroi (1934); Jules Romains, Prelude a Verdun and Verdun (1938); Antoine de Saint-Exupery, Pilote de guerre (1942); Jean-Paul Sartre, La Mort dans l'ame (1949); Claude Simon, La Route des Flandres (1960).
Using the working hypothesis that if war is fought in words there must be a war in the words of fiction, I identify constant and variable structures through the texts. This leads (after an examination of texts of the Second War) to a reformulation of what writing war entails. and some applications of that new formulation are suggested in the conclusion. My exegesis focuses on specific internal disclaimers, indications of earlier, false models of war (both historically and, more importantly: in writing). These are understood as markers of the integrity of the new, non-written war text. They therefore include the (non-) play of antecedents and the (denial of the) possibility of intertextuality, both structures suggested for readings by that interest in metafiction outlined in my preface. They also include generic ploys generated by titles of works, and their narrative "framing," both apparently "literary" strategies which nonetheless confound the "written" status of the text in question.
All of these strategies point to a tension between opposing gestures: the movement of accumulation and continuity which an awareness of a text's autoreflexivity might suggest interacts with a counter-move of abnegation and discontinuity, at a historical and textual level. Implicit in the opposition between "older war (writing)" and "this war (text)" is the assertion that post-1914 war texts occupy a space which is non-literary. This looks like the beginning of a powerful revocation of the dichotomy between "literature" and "war." In moving to texts of the Second War, we must ask what happens to that process now that a second radically new, fundamentally different war is in play, one which, for the French at least, is identified not as "guerre" but as a "drôle de guerre." Texts of that strange. new First War arrive at a kind of "writing" (henceforth to be understood as enclosed by inverted commas) which promotes a rejection of the tradition of earlier wars, wars which generated strictly literary texts which could not then be referred to. If this is the new "written" war, we need then to ask whether that way of writing war can be sustained, and used to inform texts of a second war with the same kind of attendant claims to realism, or whether those texts too effect a discarding of antecedents (now understood as including those First War texts).
In the Second War texts, the tradition of concern as to the difficulty of writing war is both sustained and advanced. In fact, disclaiming the merely literary writing of war is brought to the fore in the final text for consideration, La Route des Flandres. It still shows a systematic concern over what it means to be "about" war: it refers to a body of older literature in terms which equivocate as to its validity; it exhibits a high degree of self-consciousness as to its own status as "writing"; it claims not to represent but to re-iterate war, by reflecting the rampant disorientation of the drôle de guerre in writing.
Since that final text belongs to a the nouveau roman "school" of writing, in which many of those strategies are thought to be central, I propose in my conclusion that the renewal in La Route des Flandres of concerns already adumbrated in Le Feu suggests two things: one, that in noticing textual "narcissism" in the nouveau roman, we should be aware that it is a change of degree rather than kind. and that therefore perhaps that school could be resituated in a much broader tradition; and two, that the role of the two wars in foregrounding changes in writing strategies has not been given full weight in accounts of what is, after all, Post-War writing.