In La Vie des Idées, Jean-Yves Le Naour offers an astonishing reading of the controversy that has in recent years embroiled the field of First World War studies. Astonishing because, rather than reconstructing the intellectual issues at stake in the discussion, he prefers, as have others before him, to summarize them in terms of a simple binary construction (consent versus constraint) which, it is said, merely masks the “squabbles particular to university circles”, with each of the two camps defending its own little patch in an “artificial” confrontation. Describing the debate as “bogged down”, it is true, has the advantage of enhancing the standing of positions that appear to be above the fray. Concluding his discussion, Le Naour thus expresses his desire that a “compromise peace” be found. Given that the stakes involved are barely identified, it is obviously impossible to say upon what basis such a compromise might be reached.
As members of the Great War International Research and Debate Collective (Crid 14-18: Collectif de recherche international et de débat sur la Grande Guerre) and as such directly concerned by the debate , we would like to reexamine several aspects of this presentation of the controversy in order to restate its terms in a way that takes account of what we see as two central dimensions: the practical conditions of possibility for scholarly debate, on the one hand – what might be called the “supply train of ideas” – and a thorough reconstruction of the controversy, on the other. It is thus far from our intention to deny that this discussion contains extra-intellectual dimensions: quite the contrary, we think that the practical conditions presiding over the elaboration and circulation of ideas are inseparable from their “content”. Reconstructing a controversy, in our view, requires that the conditions under which research is conducted as well as the intellectual issues involved in it be taken into account with the same rigor.
The “Supply train of Ideas”
Let’s be up front about it: it is rather pointless to protest against the ways in which scholarly debate is caricatured as soon as it leaves the academic realm. As it happens, it is tempting in discussions of the Great War to present two opposing “schools” by hardening up their positions and employing battlefield metaphors. Yet certain limits should be respected in this regard. Indeed, the question of what metaphors are employed to describe the activity of research (“war machine”, “fortress”) is far from being insignificant: what is at issue here is a general manner of disparaging as “vulgar” the practical dimensions of research, with the result that these are addressed only in order to be dismissed as impure. This is an essential point for Crid 14-18: far from seeking to create a stronghold for a certain way of writing about the Great War, our project consists in practicing a form of “open research”.
To clarify what this project involves, we should begin by recalling several facts. Initially, there were two collective undertakings – a book and a colloquium  – which permitted a team of researchers to be gradually formed. Coming together in November 2005, the team consisted of university professors, CNRS scholars, doctorate-holding teachers, non-professional historians and doctoral candidates. They quickly reached agreement on the idea of practicing “open research”, which aims at taking into account and profiting from the public’s intense interest and involvement in the Great War as an historical object. Indeed, in what concerns the Great War – in fact, not all subjects lend themselves to this approach to the same degree – it is in our view essential to supply the means for reconciling the demands of scientifically rigorous research with that of establishing genuine dialogue with the large number of enthusiastic amateurs (who it would be unthinkable to do without) and teachers (who must not be considered as merely the passive recipients of academic discourse). In the most concrete terms, this involved creating and expanding an internet site that had been conceived as a tool for diffusion and exchange.
But “open research” as we understand it also requires that the instruments of research properly so-called be made available. More particularly, it means jointly developing documentary collections for the purposes of verification and authentication and as material open to all forms of reutilization. The “Chemin des Dames” database is as an example of this. Originally created to analyze how the received account of the 1917 battle came to be constructed – what did it include, what did it leave aside?  – the database can be of use to both the curious amateur who wishes, for example, to put the wartime career of an ancestor into context and the teacher who wants to carry out an educational project with his students. But, in keeping with its initial scholarly vocation, it has also allowed a precise map of the 1917 mutinies to be drawn up and, in doing so, has renewed the manner in which the “mutiny phenomenon” is measured.
So, yes, rather than clutching to the shopworn opposition between “history” and “memory”, we are resolutely searching for ways to take advantage of the intense interest generated by the Great War. And, yes, we reject the stance that consists in identifying with the superb isolation of a “purely” academic history while at the same time targeting the broadest possible audience through major editorial coups and by seizing upon even the smallest commemorations as an occasion to advance one’s agenda.
Finally, to read the accounts of those who have been frightened away by the critical debate that has opened up concerning the Great War, one comes away with the disagreeable impression that the principal criterion for reconstructing and evaluating the controversy should be that of good taste. But if decorum is the only legitimate argument for overcoming disagreement, it is because the life of ideas in the historical domain has fallen into a pitiful state. Against those who would advance the “coteries” argument (once again a question of good taste), it is to be recalled that Crid is, for us, a way of insisting on the necessity in scholarly controversies of giving serious attention, as is standard practice when examining the controversies of the past, to their most concrete dimension. This in no way prevents genuine engagement with the disagreements themselves, which are real, nor does it require them to be reduced to the kind of inanity that a worthy “compromise” or “consensual” concepts would be sufficient to overcome.
On a theory and its criticism
Let’s thus try to specify the properly scientific dimensions of the controversy. And, since it has lately been summarized in terms of the consent/constraint dichotomy, let’s tackle this head on. We should begin by recognizing that this dichotomy has the virtue of raising crucial questions that significantly exceed the domain of the Great War concerning the place that ideals, beliefs and representations of the world are to be granted in explaining behavior . That said, one must immediately note that it does a very poor job of answering them, advancing an alternative that has been artificially tailored to the needs of the cause and can thus be dispensed with by cleverly mixing the ingredients (here a pinch of consent, there an ounce of constraint). In other words – this time those of the Great War – the problem goes well beyond the rather perfunctory one of knowing whether the soldiers “held out” because they were patriots or whether they did so because they were threatened with the officer’s pistol.What then is the crux of the disagreement? To answer this question, let’s take another look at the theory of “consent”. For this, it is best to begin with the argumentative framework at work in 14-18, retrouver la guerre , where it is presented with the greatest coherence, even if the same line of reasoning, among others, is to be found in Stéphane Audoin-Rouzeau’s last book, Combattre. The argument consists of three steps:
1. It begins with a dual observation, at once empirical and historiographic: the interpersonal violence and cruelty of practices of combat are said to have been systematically sanitized and under-estimated, when not simply covered up, by both historians and the combatants themselves.
2. However, understanding the War requires one to fully and completely take account of “paroxysms of violence” [“paroxysmes guerriers”] among the soldiers. In support of this claim, two extremely weak arguments are advanced. On the one hand, it is argued that the essential characteristic of these acts of cruelty is that they were voluntary for we are not here dealing with the anonymous, industrial death produced by bombardments (upon which, it is said, too much emphasis has in any case been placed). On the other hand, it is claimed that the violence of war was overwhelmingly accepted by the soldiers since refusals were rare, the soldiers held out almost without exception for four years and the violence never ceased. The perpetuation of interpersonal violence over the long term thus supplies the proof that the combatants consented to the War.
3. What remains to be explained is why the soldiers not only held out but, more than that, actually desired, trivialized and accepted violence. Here, the operation consists in specifying what is concealed by this desire. The observation of warrior practices in this way acquires by inference the power to reveal the poilus’ beliefs, perceptions of the enemy and other representations. These, however, constitute the superstructure of a manifestly shared and lasting culture essentially defined by “the investment of men in their nation”. The circle is closed: it is this “culture of war” that, by a complete explanatory reversal, allows the outburst and perpetuation of violence to be explained, the very violence that served to define and describe the cultural framework in the first place.
The only aim of this excessively schematic reconstruction of the argument is to throw light upon what we believe to be the social philosophy at work in what might be called the “consent paradigm”. One will no doubt here recognize the comprehensive stance often associated in the social sciences, rightly or wrongly, with the work of Max Weber: understanding what someone does (his behavior) requires showing what his gestures mean (their meaning), that is to say, specifying what he wanted to do (his intention). In this particular case – and it is not the least of the problems raised by this approach – the operation is immediately totalized: knowledge of a collective state of mind constitutes the principal key for interpreting the conflict, its length, its violence and its consequences. The method employed in support of this interpretation calls upon cultural anthropology: it draws on traces (texts, objects) to bring out a “culture of war” shaped by the internalization of feelings of hatred for the enemy, one consequence of which is said to be a phenomenon of lasting “brutalization”. Thus conceived, the “paradigm of consent” presents a citizen who is “autonomous in his will” and makes decisions fully aware of the consequences of his acts (themselves analyzed as simple manifestations of thought) as the actor of its social philosophy. By pushing this logic to its extreme, one might describe the resulting model by analogy with the figure of Rodin’s thinker – in other words, a man who reflects on his own, examines his ideals (patriotism, pacifism, the republican model and its values) and, having carefully weighed them all, chooses to make war or not to make it, to “hold out” or to revolt, depending on the supposed terms of the “choice”.
Against this interpretation, which one may describe as “intellectualist” and which has even found its way into school curricula , there is obviously no question of substituting the converse, that is to say, the figure of an entirely constrained man, a robot who responds in a virtually mechanical way to orders. It is rather a matter of resolutely abandoning the alternative between consent and constraint in order to bring out other attitudes composed of indifference, resignation, self-sacrifice and, more generally, conformism.
It should be stated that the notion of “war culture” as it has been developed over the past fifteen years is not in itself problematic: it is entirely legitimate to seek to describe sensibilities and representations, their differential appropriations and their transformations over time. But it becomes problematic as soon as one makes it play a role in explaining individual and collective practices and behavior. Describing the patriotism of French society in 1914-1918 is one thing; saying that, if the soldiers held out, it’s because they were patriots is something else altogether. The explanation thus falls into culturalism. As the philosophy of practical reasoning has shown, this constitutes a logical error to the degree that it systematically relates observed behavior to an unobservable culture, which in turn is always postulated on the basis of observed behavior. Culturalism thereby explains the way people act by… the way they act . Explaining a soldier’s endurance by reference to a patriotic way of enduring is to claim that the soldier endures the war in order to behave as a patriot, and this without knowing anything else about him apart from the fact of his endurance and his nationality. Yet the authors of Retrouver la guerre engage in precisely this type of purely speculative argument (that is to say, one that has no quantitative dimension, no examination of appropriations and includes no effort to define the body of evidence under consideration) when they claim that the discourse of hatred could not have been so broadly diffused without expressing a very widespread state of mind.
Faced with this type of apparently obvious deduction, it is worth recalling a point that is too often forgotten despite being the result of several decades of work in the social sciences: beliefs do not observe themselves. They do not randomly collect themselves from the pages of the documents being consulted in order to come together in a high level “discourse” or “national text”, as if the historian were an old-fashioned botanist gathering plants in order to stick them in a herbarium. It is true that, given the mass of documents and objects that it offers for investigation, the Great War inevitably gives rise to the temptation to create collections (texts, images and objects). The fact that such collections allow the omniscient scholar to himself reconstitute the great cultural text of history only encourages the tendency to neglect their original contexts of use. Yet before we decree a particular state of mind – a collective one, at that – and immediately enshrine it in a museum of Great War beliefs, it would be useful, to say the least, to ensure that we are not disregarding those who are doing the believing. Rather than being surprised when confronted with the most incredible and unlikely representations – those that strike us at first glance precisely because we do not understand them – it is worth attending to more ordinary beliefs, those that organized everyday experience in 1914 and continue to do so today. For this, we must first give ourselves the means to do so.
Restoring men and their situations to the question of belief
How can one restore men and their situation to the question of belief? One may, it seems to us, apply the most classic methods and instruments of the social sciences by counting, comparing and affirming the primacy of the context.
Let us begin with the issue of counting, which is essential. The interest of counting, measuring and supplying orders of magnitude does not amount to waving the aged banner of a quantitative history that has been roughly handled by the cultural turn. First and foremost, it has to do with the fact that soldier’s “choices” can only rarely be interpreted as the result of pure internal arbitration. The history of each combatant is not simply a personal and singular history; it is in fact connected to and comparable with those of other soldiers and is, as such, significantly determined by its social context. Appropriately conceived and supervised, the creation of databases thus allows individual and family trajectories to be compared. Rather than treating these from the point of view of their singularity, this comparative approach allows one to go beyond psychological judgments concerning the choices made by individuals. The latter are no longer reduced to moral decisions supposedly made in full knowledge of the facts nor are they evaluated through the prism of the investigator or reader’s categories of judgment. Rather, they are related to the military, familial, associational or economic environment in which they take place.
It is from this perspective that the joint elaboration of documentary collections such as the “Chemin des Dames” collection mentioned above or the Dictionary of Témoins are decisively important for research. First of all, these databases allow one to form a better idea of what one is talking about and to clearly identify matters of which one is ignorant. Contrary to what the over-abundant bibliography on the Great War would seem to suggest, information is sometimes still cruelly lacking: is there not something astonishing about the fact that we debate one another over general interpretations of the Great War even though we have only a vague idea of how many men were actually sent to the trenches or who they were?  Documentary collections can thus serve to support research, either as simple bases for comparison and empirical validation or in order to raise new questions. The Dictionary of Témoins lends itself to such scholarly uses. In addition to supplying a minimum of biographical information about each author, it should in the future permit the development of a genuine sociology – or prosopography, if one prefers – of this population.
The idea is not simply to achieve a better knowledge of this group of “authors” but above all to supply oneself with the means for better monitoring what one does when parsing a quotation from this or that testimony (how many authors are quoted in our works about whom we know nothing or next to nothing?). The received idea according to which “there are neither good nor bad témoins” tends to reduce what is nevertheless a decisive question to the idea that a historian should not bring moral judgment to bear on his sources. This is of course true. But does it follow that all testimony is not only equal but indeed interchangeable from the point of view of the possible uses to which it is put? To take just one example: the testimony of the Durkheimian sociologist Robert Hertz  is a precious source for analyzing the discovery of the lower classes by a socialist intellectual and man of private means  but much less useful for the purposes of deducing the contents of what is taken to be a shared culture. The central issue here is to be capable of measuring the diversity of relations to culture – a culture of war no less than any other – in sociological terms, particularly given that the privileged access of elites to the printed word represents a major bias of our sources. Indeed, it will be agreed that if the “culture of war” is only vouched for in the writings of writer-intellectuals , one must call into question the generality of the phenomenon.
From the point of view of this complex relationship to the social status of témoins, one can only be astonished by the intense criticism to which the work of Jean Norton Cru has been subjected.
Jean Norton Cru and war testimony
Jean Norton Cru, a language and literature professor in the United States, was mobilized in 1914 and spent three years at the front before becoming an interpreter for American troops. Starting in 1916, he began to read war testimony, an interest that he would transform into a gigantic undertaking of critical analysis. His inventory, which was composed of 250 authors and 300 volumes published between 1914 and 1928, was based on precise principles: the author had to have been actually present at the front and not have held a rank above captain, his personal and military biography was verified, topographical details, statistical observations and extensive quotations were supplied and sources were crosschecked. Norton Cru’s aim was to supply a judgment concerning the documentary value of the works under consideration and he made a particular effort to hunt down what he called “war legends” – the collection of exaggerations, excesses and literary artifices which in his view too often filled the accounts. In the end, he delivered a prize list of authors according to six categories. As Frédéric Rousseau usefully brings to light in Le Procès des témoins de la Grande Guerre, the study provoked violent debate in the two years following its publication in 1929 . To put it briefly, Norton Cru believed that his concern for exactitude and objectivity (as clear-cut or arguable as it may sometimes have been) was at odds with the notion that factual errors were somehow excused by the account’s literary value (certain detractors described him as a “simpleton of the card catalogue and the marginal note”). The controversy was revived along very similar lines after the work was republished in 1993 by the Presses universitaires de Nancy. Annette Becker and Stéphane Audoin-Rouzeau condemned Norton Cru for complacency and as a “perfectly representative [example] of the desire to sanitize” the violence of war that had prevailed “first among témoins and then among historians” .
While certain of the stances taken by Norton Cru and his taste for “best of” lists deserve to be debated, the case is altogether different concerning the strict rules he gave himself for selecting témoins (date of birth, family origin, studies, profession, military career). These rules allowed him to ensure the uniform application of the procedure by which he judged texts. In our view, moreover, they are simply in keeping with the principles of historical criticism of sources: cross-checking of information, analysis of the conditions in which testimonies and documents are produced, the study of relations between observed practices, discursive stances and social positions. Indeed, once historians start to unearth shared representations, analyzing the collection of testimonies mobilized to this end constitutes a sine qua non condition of interpretive credibility. Representations may indeed be shared but by whom exactly?
Restoring men and their situations to the question of belief also involves offering other forms of comparison and deflation so that the general applicability of the premise of the autonomous will may be examined. There is an obvious argument here for consulting other fields and areas of social science: thus armed with comparisons drawn from classic issues in social history and political sociology (the strike, the labor union movement, the vote), we have been able to discuss, assess and give context to the idea of a “rate of mutiny”, both from the point of view of its numerator (from 2 to 8% of the soldiers actually present at the front, a considerable number given the proportion of those who, under much less trying circumstances, participate in a union, a strike, a demonstration) and that of its denominator (thinking in such statistical terms relegates to the background the sociologically unrealistic idea that 100% of soldiers could have mutinied) . What’s more, the implicit politics of the notion of a culture of war, which deduces generalized consent to the war from the rarity of “refusal”, must be underscored. In the absence of a perfectly general culture of war, guided by the wisdom of enlightened reason concerning the wrongs of the conflict and perhaps endowed – why not? – with an intimation of the horror of the next four years, would it have been possible in 1914 for all of the soldiers to rise up in mass to proclaim their unanimous opposition to violence? It is clear that, far from representing a break with the more traditional “public opinion” approach to such questions – that practiced, for example, by Jacques Becker  – recourse to the notion of culture has only made it more inflexible: where the former could still describe certain behaviors as those of a minority (refusal, avoidance, etc.), the latter renders them inconceivable except as marginal and thus unimportant phenomena. This all or nothing raises the question of the legitimacy of the nation state’s power. The thesis of consent to the war very clearly goes hand in glove with the idea that the legitimacy of a political regime depends on the acquiescence or collective support of its citizens (one here recognizes the deliberative and contractual model according to which society is nothing more than a sum of individual consciousnesses). Contrary to those who continue to believe that a functioning society, if only in ideal-typical form, requires what Renan called a “daily plebiscite”, one must insist on the crucial fact that those who do not profess their attachment to a regime or its values do not for all that lose their status as “nationals” or, for that matter, the rights and obligations attaching to this status. The reason for this is simple: like those who promote and police national sentiment (as well as nearly all citizens), the latter were never required to claim or justify their membership in the national community by reference to high level principles. We do not deny, in other words, that some of the soldiers of 1914 fought with their heads full of ideals and values by means of which they were able to justify their involvement to themselves or other people. We simply posit that these ideals and values were differentiated according to social group, that they were not to be found among all those whose relation to the state was chiefly a matter of indifference or withdrawal and, last but not least, that the absence of such ideals and values in no way altered the “duties” prescribed by the state – in particular those that took force once the state imposed via the organization of a levy en masse.
Rather than affirm the thesis of national unanimity which underlies the notion of a culture of war, we thus prefer to explore the complex phenomenon of social conformism by considering it from the perspective of socially differentiated relations to obedience (consider Richard Hoggart’s examination of the British working class’ relationship to the army ) and by restoring the social thickness and mobilizing power of the institutions that were brought to bear on these relations. These included the republican school (through scholarly training of group obedience) the conscription army (and its slow educational work of “nationalizing” the masses), political and labor union organizations, the Catholic Church and the powerful bureaucratic and coercive apparatus with which the nation state endows itself in wartime .
By putting the issue of social constraint at the center of discussions of processes of societal “nationalization”, the perspective proposed here allows one to fully restore the force of the collective behaviors promoted by these institutions. This, in turn, allows us to see the soldiers as acting without necessarily improvising, reflecting or justifying to themselves what they are in the process of doing. For the soldiers of 1914, this labor-saving reflexivity, which all of us have experienced in everyday life – we do not reinvent the world each morning! – took place on at least three levels: that of their gestures (charisma is in the movement, not person who performs it), their interpretation of others (the forms through which homage or obedience is expressed are immediately recognizable) and, last but not least, their judgment of their own actions – the reason for which one obviously cannot maintain an equivalence between endurance and acquiescence. In this respect, the present perspective further suggests that the study of beliefs and of representations – “what did the soldiers of 1914-1918 think of the War?” – should never be detached from that of the situations and frameworks in which individuals develop. As with all other major social ordeals (wars, epidemics, terrors, economic crises), the First World War was first and foremost a self-evident event which, by very significantly reducing the possibilities for choice, imposed itself on one and all alike. Understanding these choices is not simply a matter of reconstructing a culture, shared or otherwise; one must also fully reconstruct a complex network of representations, institutional mechanisms and social relations.
What we are positing here obviously can and should be discussed. These points have given rise to many debates within the social sciences and will long continue to do so. But denying that we here find ourselves at the crux of a genuine subject of disagreement and debate is difficult, to say the least. We hope that we have shown how badly such debate is needed and, to that extent, deserves to be taken seriously.
Translated from french by Ethan Rundell.