Overview: Second series, volume 1 (1923–1924)

2In this study, we have only taken account of the “Analyses” (see Appendix 1, Table A1). This term should be understood in light of what Durkheim envisaged when he was thinking about establishing the journal: [2]

It would be useful, very often, and especially in the first year, if the analyses gathered under a single rubric were preceded by a preamble that would set out the current state of the issue and a general conclusion; if a conclusion is justified. We would then have individual and objective analyses that would also be articles. It would be of great advantage for anyone interested in religions, for example, to find in one place everything that has been published on sacrifice—or festivals—or funeral rites. [3]

Overview: Second series, volume 2 (1924–1925)

3Aside from two original works by Marcel Mauss (“Divisions et proportions des divisions de la sociologie” [“Divisions and Proportions of the Divisions of Sociology”]) and by René Maunier (“Recherches sur les échanges rituels en Afrique du Nord” [“Studies on Ritual Exchanges in North Africa”]), this volume contains only one General Sociology analysis: “Note de méthode sur l’extension de la sociologie” (“Methodological Note on the Scope of Sociology”), signed by Henri Hubert, Marcel Mauss, Louis Gernet, Dominique Parodi, Maurice Halbwachs, Max Bonnafous, Marcel Déat, Daniel Essertier, and Célestin Bouglé. This text, unfinished and cut off, contains a sort of stock-taking of (Durkheimian) sociology, a list of its shortcomings, and a catalog of the criticisms made against it.

4Jennifer Mergy (2004) found an unpublished part in the Mauss archive held at the Collège de France. She was therefore able to establish that the published part probably constituted a quarter of the volume that had initially been intended, that a third volume was being prepared, and that a fourth volume had been announced for 1929 (Mergy 2004, 16). This project was abandoned because, aside from the cost, Mauss did not have the skills to commercially manage and lead a project of this sort, or he did not wish to do so. He was an “old bachelor,” “unreliable,” “dissipated and boastful.” [4] The unpublished texts intended for this second volume of the new series numbered almost 400, and mainly consisted of analyses by Mauss (56), Hubert (22), Halbwachs (5), and Fauconnet (2). The seventeen authors who contributed to this second volume are: Georges Davy, Philippe de Félice, Albert Demangeon, Daniel Essertier, Paul Fauconnet, Louis Gernet, Maurice Halbwachs, Henri Hubert, Henri Jeanmaire, Emmanuel Lévy, Henri Lévy-Bruhl, Charles Lalo, René Maunier, Marcel Mauss, Jean Plassard, Jean Ray, and Jules Sion. The authors for whom no texts exist in the archive but who feature on the cover of volume 2 of L’Année sociologique are: Albert Bayet, Célestin Bouglé, Georges Bourgin, Henri Bourgin, Stefan Czarnowski, Marcel Déat, Marcel Granet, Raymond Lenoir, Lucien Lévy-Bruhl, Jean Marx, Antoine Meillet, Alexandre Moret, Dominique Parodi, Charles Blondel, André Piganiol, and Pierre Roussel. [5]

The second series of the L’Année sociologique

7Mauss’s involvement in L’Année sociologique is impressive, as he published roughly four times more than the second most active contributor, Halbwachs (Table 2). He almost single-handedly sustained the rubric of Religious Sociology, and he also contributed three of the four original works of the two volumes of the second series: “Divisions et proportions des divisions de la sociologie” for the 1923–1924 volume, and the In memoriam and “L’Essai sur le don” (“Essay on the Gift”) for the 1924–1925 volume. These figures confirm his role as “leader,” which the members of the team seemed to confer on him when their journal was relaunched, and which he assumed, at least temporarily, according to the correspondence exchanged at the time. Mauss most visibly adopted this role as leader in the “Divisions,” a work that called for an increase in concrete studies and an abandonment of premature attempts at systematization. We could also say that this was the last time he took on this role, as the end of the publication of L’Année sociologique coincided with his virtual disappearance from the collective project. Halbwachs and Bouglé already demonstrated a major involvement at this stage, which will become even more pronounced when we turn our attention from L’Année sociologique to the Annales sociologiques. Henri Lévy-Bruhl entered the scene at this point, marked by a high degree of involvement and a strong, unwavering commitment to the project. It still remains to write the history of the active involvement of Jean Ray (1884–1943), a former student of the École normale supérieure (ENS), having passed the agrégation examination in philosophy. A specialist in the League of Nations, he belonged to the original team of the first series of L’Année sociologique and subsequently led series C of the Annales sociologiques. Let us note, besides the absence of Davy, the very limited involvement of Fauconnet. [6]

Annales sociologiques: The original works

8The critical work devoted to this period of the history of Durkheimian sociology emphasizes the team’s inability to renew itself and find new blood in order to continue its tradition of research (Heilbron 1985).

9In fact, the number of original works published allows us to distinguish within the first generation of Durkheimians the “historical” principal trio of Halbwachs, Mauss, and Simiand (plus Bouglé), to which we must add the contributors of the same generation who were recruited through different circles (see below). Jean Ray and Emmanuel Lévy were former contributors to the first series of L’Année sociologique (Table 3). This body of original works contains certain texts that had a lasting significance, such as Simiand’s work on money (1934) published in the first installment, or the programmatic text published by Mauss (1934). [7]

10Nonetheless, we can consider that some space was allocated in the Annales sociologiques for younger recruits, since these account for a little over a third of the original works.

11The format of the “original work” existed from the publication of the very first issue of L’Année sociologique. As its name suggests, it is supposed to be a publication of new material, reflecting its author’s research work. Mauss is present indirectly in this group of authors, since René Maunier (although he published his original work in series D) and Stefan Czarnowski belonged to his network, whereas Robert Marjolin [8] and Jean Stoetzel gravitated around Bouglé’s Centre de documentation sociale (CDS) (Center of Social Documentation) [9] (Table 4). We should note that Stoetzel’s original work (1941) on “La psychologie sociale et la théorie des attitudes” (“Social Psychology and the Theory of Attitudes”), published in installment 4 of series A, discusses the contribution of the American literature to the theory of attitudes, on which the young Stoetzel was then writing his doctoral dissertation. This example also makes the Annales sociologiques a critical space, open to a reflection that differed in many respects from certain theoretical and methodological principles laid down by the Durkheimian school of thought. In this respect, as we can also see from the “General Sociology” analyses in L’Année sociologique and in the Annales sociologiques, Bouglé contributed to instigating a renewal of theoretical reflection, without hindering the development of alternative conceptions of sociology, including among those who, like Jean Stoetzel and Raymond Aron, would later declare themselves to be strong critics of Durkheimian sociology in order to further their own causes (for more detail on this matter, see Marcel 1998; 2001).

Annales sociologiques: The Analyses

12The list of analyses confirms the high degree of involvement of Bouglé, Halbwachs, and Simiand. The death of Simiand, five days before his sixty-second birthday, would leave the other two largely responsible for continuing the work of Durkheimian sociology, which would lead them to establish several new connections (see above). We should note the presence of Halbwachs in all the series—except the one devoted to Religious Sociology—, which proves once again his commitment to the project: he took the “resurrection of L’Année sociologique” seriously, and played the role of “convener” among the Durkheimians, at the center of different networks. Besides, there were very few contributors who authored texts in several different series (Table 5). Apart from those mentioned in the table, we find Lévy-Bruhl and Maunier (series C and D), whose legal knowledge undoubtedly allowed them to contribute to several series, but also Marjolin, who had expertise in economics. We should also note the large involvement in a single series (A) by André Kaan and Raymond Aron, who were Bouglé’s assistants at the CDS, and the contribution of Charles Lalo (whose expertise undoubtedly gave him a monopoly in his field), who exclusively contributed to the “Aesthetics” rubric of series E. The work of the assistants at the CDS, who were, along with Bouglé, the main contributors of analyses in series A, shows Bouglé’s ability, as a real leader in the field of sociology, to motivate young researchers and put them to work. By comparison, Mauss, who is considered to have influenced many young authors, did not seem to have the same enthusiasm for involving contributors in this kind of collective work, in which he was barely involved.

Those who stayed on from the second series of L’Année sociologique to the Annales sociologiques

15Among these loyal collaborators, there is a first circle made up of contributors from the outset, who were present in the team of the first series of L’Année sociologique, as we have already seen: Célestin Bouglé, Georges Bourgin, Albert Demangeon, Paul Fauconnet, Maurice Halbwachs, Henri Jeanmaire, Charles Lalo, Emmanuel Lévy, Marcel Mauss, Jean Ray, Pierre Roussel, and François Simiand.

16A second circle of contributors was made up of authors born in the 1880s, about ten years younger than the “historical Durkheimians,” but who did not contribute to the first series of L’Année sociologiques. These included Henri Lévy-Bruhl, [10] son of the sociologist Lucien Lévy-Bruhl, who joined the directorial committee of L’Année sociologique in 1924, probably thanks to his father’s connections with the Durkheimians. [11]

17Albert Bayet (1880–1961) was employed at the Sorbonne where he had taught since 1932 (Heilbron 1985). He succeeded Bouglé in 1940 with the support of Halbwachs (Pinto 2004, 44). Jules Sion (1879–1940) was a geographer and, along with Demangeon, a disciple of Paul Vidal de la Blache. A friend of Lucien Febvre, Sion taught at Montpellier from 1910 to 1940 and was a member of the Institut français de sociologie (IFS) (French Institute of Sociology). [12] He was probably recruited by Halbwachs, as he only wrote for series E of the Annales sociologiques, and he was part of the team that Febvre put together to publish the Encyclopédie française, which Halbwachs was also involved in (Müller and Weber 2003).

18The others joined L’Année sociologique through Mauss. Marcel Granet (1884–1940), who took up his studies belatedly and was a student of Durkheim’s when he took the agrégation at the Sorbonne (Hirsch 2016), was probably recruited through the socialist network, but in particular the École pratique des hautes études (EPHE) (Practical School of Advanced Studies), where he met Mauss, whose lectures he also attended (Mestre 1939). [13] Marcel Cohen (1884–1974), Stefan Czarnowski (1879–1937), and René Maunier (1887–1951) all attended Mauss’s lectures at the EPHE (Mahé 1996). André Piganiol (1883–1968) was a specialist in ancient history, a friend of Mauss, and a member of the editorial board of the Revue historique.

Those who left

19If we examine those who contributed to the second series of L’Année sociologique and then “disappeared,” we observe that, besides some students of Célestin Bouglé (Marcel Déat and Max Bonnafous, who left for political careers, Daniel Essertier, who died in 1931, Jean Laubier, and François Henry), those who left were primarily the “elders” of the first series of L’Année sociologique: Hubert Bourgin, Georges Davy, Edmond Doutté, Louis Gernet, Henri Hubert, [14] Antoine Meillet, Jean Marx, Alexandre Moret, Dominique Parodi.. . making up about 21% of the total. We know that Bourgin, disillusioned with the social engagements of his early life, especially in socialism, would later join the ranks of the far right and publish a violent pamphlet (Bourgin 1938) against what Christophe Prochasson (1993) called “socialisme normalien” (socialism of the École normale supérieure). The departure of others, such as Davy, can be explained by the demands of their careers: he became rector of the académie of Rennes in 1931. We have not been able to find out anything about Jean Plassard, who was also a member of the IFS. Henri Laoust, possibly recruited by Mauss, also appears in the second series of L’Année sociologique, but not thereafter.

The new arrivals at the Annales sociologiques

20Among the new arrivals, some belonged to the “second circle” generation identified earlier: Édouard Mestre (1883–1950) emerged from Mauss’s network at the EPHE; the legal specialist Gabriel Le Bras (1891–1970) was a colleague of Halbwachs at Strasbourg, as was the economist Henry Laufenburger (1879–1964); [15] Pierre Montet (1885–1966) was an Egyptologist, also recruited by Halbwachs at Strasbourg, and Lucien Tesnière (1893–1954), a linguist, was also employed for a time at Strasbourg. René Hubert’s experience (1885–1954) strongly suggests that he was recruited by Bouglé, as he studied philosophy at the Sorbonne. He would eventually become rector of the académie of Poitiers, and in 1926 he wrote a sociology textbook for the Écoles normales. We know that Bouglé was very active in introducing sociology to these schools (Geiger 1979). Georges Gurvitch (1894–1965) was probably brought in as a contributor by Marcel Mauss, who supported him when he put himself forward as a candidate for a temporary lectureship at the EPHE in 1932 and when he applied to cover Maurice Halbwachs’s teaching at Strasbourg in 1935 (Fournier 1994, 647). Pierre Noailles (1881–1943) was a legal specialist and a friend of Henri Lévy-Bruhl. Finally, Halbwachs introduced Pierre Depoid (1909–1968) into the team, a graduate of the École polytechnique, considered at the time to be one of the future stars in the field of demography (Lenoir 2004, 207). Georges Lutfalla (1904–1964), a specialist in mathematical economics, was also recruited. These two moved in the networks of statisticians and demographers, which Halbwachs also frequented, as he was always interested in questions relating to the enumeration of human phenomena (Martin 1999). Anatole Lewitsky (1901–1942) and Michel Leiris (1901–1990) were students of Mauss (Soustelle was initially a student of Bouglé, but he pivoted towards ethnology).

21In summary, 19 people, or around 39% of the total number of contributors to the Annales sociologiques, joined it through what we could call peer recruitment (11 people, or 23%, arrived for the first time with the launch of the Annales). Those brought in by Bouglé and Halbwachs tended to be recruited through institutional networks involving relations between colleagues, whereas Mauss’s recruitment usually relied on master-student relations.

22The others can be grouped with the generation of “students” born around 1900. Thanks to the work of Antoine Savoye (2017, 134 ff., Appendix 2), we know that 14 of them (roughly half of all new arrivals) were at some point students of Bouglé (Table 6).

23Bouglé therefore almost single-handedly supplied a new generation to take up the baton, drawing from a pool that centered around the Sorbonne, where he taught, the École normale supérieure, and the CDS (Appendix 2, Table 3A), which he directed, and in which philosophers, unsurprisingly, were overrepresented.

24In total, the recruitment for the second series of L’Année sociologique demonstrates the vitality of the network around Mauss (Appendix 3, Figures 1 and 2), as the new arrivals were more likely than those in the team since the first series to be based at the EPHE (more than half of them). Recruitment for the Annales sociologiques certainly reflects both the breadth of the sociological network and the dispersion of the former Durkheimian group, as the historiography has already shown, but it also demonstrates the vitality of the network around Bouglé (which accounted for almost half the new arrivals) and, to a lesser extent, that around Halbwachs, as well as, quite blatantly, the disengagement of Mauss.

Series A, “General Sociology”

26Célestin Bouglé, who was in charge of this series, was very dedicated to maintaining a periodical devoted to sociology, as was Maurice Halbwachs. Series A, which was already Bouglé’s responsibility at the time of the launch of the first series (and which he jokingly said was the least important), would be used as the organ of the CDS, which he directed. This is why the editorial choices made in this series must be understood in the light of the activity of the CDS, which had to obey the wishes of its American backers [16] to promote “empirical studies” for understanding the contemporary world. Bouglé’s intellectual interests and the way in which he would pursue them are also factors. As a result, the generation of students that he trained in the interwar period, even if they were not strictly his “disciples,” had shared interests that can be explained by this context, and that were manifested in their choices of dissertation and postgraduate research topics: social thought (through the works of nineteenth-century authors: Proudhon, Charles Fourier, Frédéric Le Play, etc.); theoretical questions addressed from a philosophical and sociological perspective; contemporary problems (trade unionism, economic crisis, cooperation, emigration, etc.); historical sociology and the relations between sociology and history more broadly (Savoye 2017). However, we find traces of these interests both in the rubrics structuring the category of “General Sociology” and in the choice of books analyzed. The cognitive basis of these articles, which have a certain thematic continuity from the second series to the Annales sociologiques, should be understood in this context. Finally, as we have already suggested, these publications also reflected the intellectual interests of this new generation, which they expressed by adapting them for this Durkheimian editorial context.

27The analyses in the series generally begin with a “Methodology” rubric and a “Handbooks” rubric, which can be considered connected. In his “Notes sur la méthode expérimentale en sociologie” (“Notes on the Experimental Method in Sociology”) in installment 1 of the Annales sociologiques (which includes a review of Simiand’s magnum opus [1932]), Bouglé wrote a sort of methodological manifesto (Bouglé 1934a, 83):

28

Since the perspective of sociology is not yet accepted by everyone, and some dispute its ability to grasp specific truths, it is natural that those who helped it progress should reflect on the procedures that they have used.. . to demonstrate their legitimacy and usefulness.

29This text, presented as an attempt to defend and consolidate the scientific nature of sociology, should also be kept in mind when we read the original work by Mauss that opens installment 1, entitled “Fragment d’un plan de sociologie générale descriptive” (“Fragment of a Plan of General Descriptive Sociology”), [17] as well as Bouglé’s own introduction to installment 2 (1936) on “La méthodologie de François Simiand et la sociologie” (“The Methodology of François Simiand and Sociology”). This idea is presented as being based on a methodological turn that aimed to make sociology more empirical, closer to fact. The analyses in this rubric manifest a certain eclecticism in the volume of L’Année sociologique and throughout the four publications of the Annales sociologiques. From the beginning, it is mainly German and American works that are analyzed (“two or three cases studied well are, in our view, worth twenty treatises on methodology” [Bouglé 1934a, 83]). In installment 2, the works discussed are Florian Znaniecki’s methodological handbook (by Marjolin), William F. Ogburn’s conception of measurement (by Marjolin again), and Paul Descamps’s Sociologie expérimentale, reviewed by Aron: “A monographic method, broad in scope and flexible, leads to experimentation and synthesis,” he writes. This involves proposing a hypothesis, generalizing the relations it brings to light, then testing it by gathering evidence (Aron 1936, 90). In installment 4, the contribution of the Romanian school of sociology is also discussed (Mircea Vulcanesco). In summary, in the same spirit as the programmatic articles by Mauss on the issue, it is a question of trying to make sociology more operational through the analysis and collection of facts, extending it towards sociography, without sacrificing any of the rigor related to experimental analysis.

30In L’Année sociologique, Bouglé analyzed treatises by Alfred Vierkandt, Franz Oppenheimer, but also Leonard Hobhouse and Paul Bureau, [18] while Max Bonnafous discussed the works of Leopold Von Wiese and Werner Sombart. The interest that these young philosophers took in German sociology was typical of the 1920s, when a major innovation appeared: the arrival of German authors (Edmund Husserl), whose thought represented a break from the dominant idealist culture of the preceding period (Pinto 2009). Bouglé, a flexible “patron,” but not without authority, did not discourage his students from pursuing these other traditions of thought, nor did he object to the critiques of Durkheimian rationalism to which they gave rise (Marcel 2001).

31The other rubrics that made up the analyses of “General Sociology” should be understood in the same spirit: clarifying the relations between sociology and the other disciplines “that it makes use of” (Bouglé 1934a, 83), namely history and psychology.

32The first of these rubrics, “Sociology and Psychology,” [19] is the most significant, since it belongs to the context of an attempt to redefine the boundaries between the two disciplines, through debates that would mobilize both sociologists and psychologists, and that related to precise topics: the expression of emotions, pursued by Marcel Mauss and Georges Dumas; and the question of memory, which was the subject of a dialogue between Maurice Halbwachs and Charles Blondel (Mucchielli 1994; Hirsch 2016). All these debates were summarized by Déat in the first volume of L’Année sociologique (Déat 1925). This interest in questions that were also explored by psychology formed part of a project that we might call the “psychologization” of sociology (Pinto 2009). One of the main extensions of Durkheimian sociology in the interwar period actually proceeded through a transformation of the theory of collective representations and a desire to redefine the categorization of the facts to be observed, in order to make sociology more empirical, or “concrete,” to use Mauss’s term (1934). This continued to the extent that Mauss himself (1924) proposed a program of collaboration with psychology, in which he tried to list the common concepts that could be of use to both disciplines, and, together with Halbwachs, he went as far as suggesting that the term “sociology” might be abandoned in favor of “collective psychology” (Marcel 2004). We could summarize this desire for change by saying that it aimed to gain a better understanding of how the individual experiences his or her belonging to social groups, and ultimately, to give a more phenomenological dimension to reflection on the social. This position also undoubtedly justified the presence of American-style social psychology supported by Stoetzel, whose avowed aim was to study the individual in concrete social reality and in his or her individual story (Stoetzel 1941, 4). Essertier’s reflections on the relations between sociology and psychology were also an important point in the debates that took place in the journal. Essertier was the author of a vast critical bibliography on this issue (Essertier 1927), which Bouglé had asked him to pursue, and he took a very harsh line on Durkheimian sociology, which he considered to be incapable of conducting in-depth analysis, which requires an approach that positively considers man as a “whole,” and that therefore modifies the operative concepts of sociology in order to allocate more space to individual introspection (Essertier 1934). In the program of cooperation with psychology that he argued for, it was necessary to reestablish “point by point the role of the collective in mental life,” as Bouglé put it (1934b, 143), for if...:

33

“the principal role of the notion of collective consciousness is to inform us that the representations that act as the center of human groupings are syntheses and not summations,” this does not resolve the problem of the interpenetration of the individual self and the collective self.

34In short, the project for the “psychologization” of sociology was aligned with the preoccupations of those recruited by Bouglé, who showed an interest in philosophy (Stoetzel, Aron, Essertier, Bonnafous, Déat, etc.) and who, in some cases, would play a role in the sociology of the postwar period. [20]

35The rubric “History and Sociology” was part of the attempt to construct a “universalist” history of civilizations, made necessary by war. However, there was nothing entirely new in this program for collaboration, which took up the position previously adopted by Durkheimian sociologists since the debates at the start of the century and annexed historical material for the purposes of sociological explanation, but which was also aligned with the contemporary project to make sociology more descriptive. The note on “History and Sociology” of installment 1, written by Bouglé, remarks that historians (especially from the Annales school centered around Marc Bloch and Lucien Febvre, and Henri Berr’s collection “L’Évolution de l’humanité” [“The Evolution of Humanity”]) have emphasized the importance of the comparative method in order to better take account of the transformations and evolution of institutions. This allows for a greater discrimination between effects and causes. By this account, history had been led to consider the correlations between historical facts and the facts of economics, collective psychology, politics.. . and the general facts of civilization. Consequently, the conflicts, which centered around three essential points (the place of individuals in history, the role of the accident or pure contingency in the series of causes, and the real or supposed inability of sociological generalization to grasp the original character of each historical period), were henceforth considered secondary, that is, more a matter of form than of substance. Recent developments, notably following the foundation of the Annales school, were therefore interpreted as reinforcing the sociological perspective. It was therefore appropriate to establish a new division of work in which, faced with a particular case, the sociologist would establish what can be explained by the transformations of social forms and collective beliefs, while the historian would produce an account of what is attributable to the effects of a coincidence or the action of a temperament, such as the role of Philip II’s neurosis and stubbornness in the creation of the religion of the monarchy. This place allocated to history was presented as aligning with methodological reflection on the experimental method: the historian could produce monographs posing general questions based on a particular case (as demonstrated by Marc Bloch’s work, Les Rois thaumaturges [The Royal Touch]). [21] Although this allowed a role for innovations and initiatives, it did not prevent the observation of a certain number of regularities in the life of societies: a given antecedent corresponds to a particular effect. But there was no question of borrowing its concepts from historical science, or of building new concepts with its help.

36The rubric “History of Doctrines” gave a considerable place to the German and French forms of socialism. This undoubtedly reflected Bouglé’s interest in socialist thinkers and utopian socialists, which he never lost sight of, publishing on this subject throughout his life and encouraging his students to devote their own research to it (Savoye 2017).

37Finally, installment 4 in 1941 introduced a rubric entitled “The Problem of the Classes,” “since the concept derives from general sociology and the reality of the classes is a matter of interest for society as a whole” (Mougin 1941, 62). Reproducing some of the results of the “Enquête sur les classes moyennes” (“Inquiry into the Middle Classes”) carried out at the CDS by the same Henri Mougin, this rubric presented, one could say, research as it happened. Funded by the Rockefeller Foundation, this inquiry into the middle classes was launched in 1938 by the CDS in order to carry out a synthesis of the quantitative data gathered by the American method and a theoretical reflection on the concept (Tournès 2008). The first results were published in the volume Inventaires III (Aron et al. 1939), which compared the data gathered by observers who had traveled abroad, thanks to Bouglé, of whom a large number were members of the CDS, including Raymond Polin (Belgium), Robert Marjolin (United States), Dragoljub Yovanovitch (Yugoslavia).

Series E, “Social Morphology, Language, Technology, Aesthetics”

38This series was quite varied from the beginning, including remnants from other fields that were difficult to classify. Halbwachs’s role in directing this series can be explained by the importance that he attributed to “morphological facts” or “facts of population” (Lenoir 2004). Halbwachs specified that the term “social morphology” was more accurate than the term “demography,” as “it implies science (-logy) and not mere description (-graphy); above all, it clearly indicates that the emphasis should be placed on the material and spatial form of society” (Halbwachs and Sauvy 1936, 7–76.3). Halbwachs was heavily involved in the studies of “social morphology” in the interwar years. For example, Paul Rivet and Lucien Febvre would call on him to write (in collaboration with Alfred Sauvy) the section of volume 7 of L’Encyclopédie française devoted to “The Human Species” and concerning “the point of view of the number” (Halbwachs and Sauvry 1936). The original work published in installment 1 on “La nuptialité en France, pendant et depuis la guerre” (“Marriage in France, During and Since the War”) addresses a question that he developed more completely in this volume of 1936. Similarly, he summarizes his own work on the determination of sex at birth in this installment, which prefigures the long article he devoted to this question in the Encyclopédie. The importance attributed to the facts of population corresponds to Halbwachs’s conception of life in society, which involves a theory of population. In order to exist and to feel that one exists (to create a collective consciousness for oneself), a group, as well as an individual, must create an idea of its form in space, of its body (Halbwachs 1938). By “form,” we understand the gathering and division, the concentration and dispersal of its population in space. This form structures the way in which individuals enter into contact with one another, and, consequently, both the force and quality (in terms of cognitive power) of the collective representations that result from it. Contacts are ephemeral in places with a high concentration of population, such as cities, and in these cases one has a recurring experience of alterity. As a result, collective representations make themselves felt less strongly, as they lose their force and coherence. But in the countryside, contacts are more intense, take place among fewer people, and reinforce interpersonal relations between similar individuals who share simpler collective representations that make themselves more strongly felt (Halbwachs 1930). As a result, one of the fundamental conditions for a society’s ability to exist and continue to exist is the way in which it inserts itself in matter, what it devotes particular attention to (Halbwachs 1920). In modern societies, it is the workers who are made responsible for this “insertion,” since they are subjugated to matter because they work to shape it. This makes them a sacrificed class, kept apart from the spaces where more purely human contacts craft the social (Halbwachs 1955). In this sense, the facts of population, the relation of the sexes to birth, marriage, or the reduction in the average number of children per family, etc., are not merely biological or physical facts. They are social facts that show how societies organize the struggle to survive in time and space. For example, if families have fewer children, it is because in an urban social world, where life is more complicated, more importance is attached to the child as a unique individual and more efforts are devoted to preparing the child for this harder future life. In summary, the nuclear family is the institution that allows society to adapt to urban life, which is more complicated. The enumeration of human phenomena is, in this conception, a means of showing how society strives to survive across time and space by carrying out an action on its “body”: the group has an influence on the secret acts of its members and contributes to shaping their motivations and aspirations.

39Accordingly, in the original work on “La nuptialité en France” (“Marriage in France”) (Halbwachs 1935), Halbwachs observed that what he calls the “movement of marriages,” which he analyzes over the period 1913–1931, had been changed by the war. “Social habits” and the “various motivations and incentives that lead men and women to get married” (Halbwachs 1935, 37) had been disturbed as never before. As a result, marriages entered into during the war involved new age combinations that would have been considered excessive before the war, and which therefore made up a very small proportion of marriages: unions in which the woman is far older than the man and vice versa. After the war, the proportion of marriages with men who were far younger than their wives remained much higher than before the war. “As soon as they found that they were no longer held back by their elders, as the numbers of the latter had been reduced, the tendency of young people to get married, which had been suppressed until then, was able to manifest itself,” and this continued (Halbwachs 1935, 41) in the 1930s. In order to understand why this change occurred and why it had lasting effects—a change that had no purely demographic justification (why does it not return to its “former balance,” Halbwachs asks)—we must see marriage as a competition between men and between women, with differences in their ages. Young people and older people enter into a struggle on the marriage market. For the conjugal union is the act of creating a household and founding a domestic establishment that is intended to last. Today, we would say that it is a status with a social dimension that connects it to the rest of society. This explains why the number of marriages varies according to the economic conditions. However, the state of the economy is the result of the organization of society and the circumstances it finds itself in. Therefore, these changes cannot not explained by a mechanical change in the composition of the male population due to the losses of the war. Rather it was because, due to the war, the economy had slowed down and offered increased possibilities to the young because competition from older people had reduced. Conditions had conspired to leave them no longer obliged to wait as long as before to take the place of their elders. The general conditions of life in society were therefore transformed. They also consequently changed the collective representations linked to age: the very marked reduction of the male population between twenty-three and thirty-eight years because of the war changed perceptions, such as the sense of responsibility, which developed more quickly so that younger men than before might be allowed to take on the responsibility of marriage. In other words, the social milieu as a whole changed because the relations between its parts (different age groups) had been modified.

40

It is in the nature [.. .] of a social organism to consolidate the changes that have taken a truly collective form. Therefore, beyond anecdotal efforts and tendencies and individual matrimonial behaviors, there is a sort of collective nuptial march, whose sense and rhythm are governed by the evolution of society.

41The movement of marriage is a manifestation, among others, of a sort of collective survival instinct, a behavior that the group follows in order to survive across time and space, by organizing the struggle for life. Here, this struggle takes place between older people, who are inclined, by means of their experience, to prevent their juniors from advancing in their careers and replacing them, and younger people, full of spirit and energy, who are keen to see their elders removed and to immediately take their places (Halbwachs 1935).

42This explains Halbwachs’s unwavering interest in statistical methods and, for example, his collaboration in Strasbourg with the mathematicians Jean Cerf and René Maurice Fréchet, with whom he gave courses in statistics. He even published a book with Fréchet, for which he received the prize of the Académie des Sciences (Halbwachs and Fréchet 1924). He was a member of the Société de statistique de Paris (Statistical Society of Paris), and was an authority on the subject, since, for example, in February 1936 he attended the assembly of the Mixed Commission on Nutrition of the League of Nations. [22] In 1938, he sat on the board of directors of the Institut de statistique de la faculté des Lettres (Statistical Institute of the Faculty of Arts). [23] In 1935, he became a member of the International Statistical Institute and, in the same year, he participated again in the International Labour Organization, this time as a delegate at the Conference of Labour Statisticians.

43For Halbwachs, his interest in social morphology therefore intersected with moral statistics (he would also analyze the production of these in series C), [24] facts of population, and the sociology of knowledge (Marcel 2007). This explains why he contributed to several series.

44His editorial work in the series therefore consisted in creating a critical inventory of works of demography, geography, and moral statistics, written by authors as diverse as, in demography: Adolphe Landry (installment 1 of the Annales sociologiques) and René Gonnard (in L’Année sociologique); and, in statistics: Lucien March and Michel Huber (installment 1 of the Annales sociologiques). He also devoted methodological notes to the study of population, such as his “Note sur l’application de certains procédés analytiques à l’étude de la population” (“Note on the Application of Certain Analytic Procedures to the Study of Population”) (installment 2), or the text that he devoted to the “diverses espèces de déplacements de population” (“varied types of population movements”) (installment 1). Halbwachs published several methodological articles on the processes of quantification, thereby pursuing his interest in this question.

45Regarding the other contributors, it appears that a clear division of labor was established within a small team. Charles Lalo authored absolutely all of the analyses of the “Aesthetics” rubric, which provided reviews of works that we would now categorize under the sociology of art and artists. It seems that he held this monopoly in the Durkheimian team since the time of Durkheim himself. Lucien Tesnière and Pierre Montet provided material for the rubric on language and writing, while Depoid and Halbwachs dealt with everything concerning “Population.” Sion and Demangeon took care of the rubric on the “Geographical Bases of the City.” A rubric on “Population Movements” appeared only in installments 1 and 2 and contained analyses by Halbwachs and Sion.

46In all, his varied intellectual interests, his prolific research activity during the period, and his involvement in the collective publishing venture made Halbwachs (especially after the death of Simiand) a crucial figure in the Durkheimian project in the interwar period. In this sense, series E is a good example of an area of sociological research and of certain conceptual innovations that were essentially due to Halbwachs.