RFAS 2024-4/ CFC Perspectives on the evolution of social protection in France since the end of the 19th century

Multidisciplinary call for papers on:

Perspectives on the evolution of social protection in France

since the end of the 19th century

for the fourth issue of RFAS of 2024.

The report will be coordinated by Paul V. Dutton (Northern Arizona University, USA), Yannick Marec (Université de Rouen Normandie), Bruno Valat (Institut national universitaire Champollion, Université de Toulouse), and Vincent Viet (Cermes3).

This call for papers is addressed to researchers in history, sociology, political science, demography, economics and law, as well as actors in the field of social protection.

Articles must be submitted by Tuesday, 9 April 2024.

This call for papers aims to take advantage of the abundance of multidisciplinary work on social protection since the 1970s to review the construction of the social state in France, with a view to providing a historical overview and in light of the forthcoming eightieth anniversary of social security. Unlike analyses that postulate that the social state emerged from the most underprivileged margins of French society[1] or traditionally limited approaches[2], the aim is to question the evolution of social protection as understood both in its broadest sense and in its multiple sectoral interactions[3]. By “social protection”, we mean all the measures, systems and benefits designed to reduce the uncertainties of life for a human community, by protecting its beneficiaries against socially identified and individually perceived risks (disability, accident, unemployment, illness, old age and, even partially, dependency or loss of autonomy).

The proposed approach is both cultural and structural. We wish to understand how the French have adapted to the various forms of protection to which they are now subject or stakeholders, and what effects these have had on their mentality and relationship to risk. It is a daunting task when you consider that French social protection, whose gendered[4] and family-oriented beginnings date back to the late 19th and early 20th centuries, is a multifaceted system. It is quite clear that “protecting” one’s own or “protecting oneself” (voluntary welfare cover: private insurance, savings, membership of mutual funds) is not of the same nature, does not have the same meaning, does not involve the same institutions as “being protected” (assistance, compulsory welfare cover via social insurance), even if these three configurations or attitudes towards preparing for the future can indeed be found in the same household or be adopted by the same individual. However, there are many signs to suggest that this combination of reflexive, transitive and indirect forms of protection stems from the same “polycentric environment[5]” which, since the end of the 19th century, but at different stages, has structured social protection around three main areas of focus: assistance (public assistance and private charity)[6], voluntary welfare cover (savings, membership of mutual funds, private and even public insurance) and compulsory welfare cover (social insurance, public or private insurance[7]). Yet these broad categories, clearly established by Pierre Laroque in the mid-1930s[8] and still used as a basis for understanding France’s system of social protection, were never really watertight – far from it. They have conflicted with[9], intersected and mutually sustained each other, with each being aligned with radically different, even antagonistic, philosophical ideas or ends of the political spectrum, but also breaking with these through numerous compromises and departures from the principles that are supposed to set them apart[10]. Unaffected by partisan divisions, the astonishing plasticity or fungibility of the field of social protection stems both from its very early claim (from before 1914) to combine[11] protections without favouring one in particular[12] – in a country which, like Germany or England, did not experience an economic take-off and thus an early process of people becoming salaried employees – and from its later propensity to hybridise forms of protection[13]. In fact, most forms of social protection we have today were already known about and had already been trialed before 1914, whether in the form of assistance (public or private), cooperation, public or private insurance, membership of mutual funds, individual or collective savings, or corporate social protection. The same could be said of protection techniques and the chosen compensation methods, which were already considered non-standardised: actuarial calculations based on mortality and morbidity tables, annuity calculations based on scales, funded or pay-as-you-go schemes, etc. Finally, all the classes of protection, i.e. the ways of protecting oneself or protecting others, were already in place: social protection could be public or private, individual or collective, optional or compulsory, legal or contractual, local or national, professional or inter-professional, territorial or sectoral; guaranteed or by the state or not. The originally exclusive “or” can, depending on needs and circumstances, be transformed into an inclusive “or” by means of cross-cutting measures or institutional and legal arrangements[14]. The fact remains that, apart from voluntary welfare cover, which is based mainly on annuities, savings or property capital, no single form or class of protection really predominated over the others.

However, to claim that the French population’s attachment to their social protection system is due to its traditionally composite nature would be highly simplistic. This is because it relates first and foremost to the most restrictive form of protection of all, compulsory welfare cover, which involve employer contributions and charges that are deducted from wages. Their conversion to this form of protection, which was far from a given at the start of the 1930s, appears above all to be correlated with their collective (and not individual, as voluntary welfare cover would imply) and cumulative (the example of pay-as-you-go pensions is a good illustration of this) experience of a social security system that assumes “bad risks” (long-term illnesses since 1945) without resisting, and ensures compensation and care far more than it does prevention. The predominant role of social security, which obliterates optional forms of protection in public opinion, and its ability to nurture solidarity between those of the same generation or different generations, are of course key factors in this, as is its “universal” character, which developed around social insurance by increasing the number of beneficiaries from the “economically weak”, many of whom were working class (1930), to include the general population (1946). Indeed, it was the institutional and fiscal extension of these types of insurance, which already formed a system within the broader framework of social security,[15] which now includes compensation for occupational risks (1898) and family allowances (1932), that gave the French welfare state its main characteristics. These include the strong involvement of relatively autonomous operators (the social security funds) and ad hoc administrations, complemented or rivalled by private actors or bodies working in the public interest (e.g. mutual funds); a degree of homogeneity in compulsory social protection, embodied by a general scheme (not exclusive of any special schemes), most of the funding for which comes from paid employment; redistribution ensured by national solidarity in a system endowed with a high degree of unity; a capacity to magnetise, relay, amplify and disseminate social policies implemented by private and public actors (such as local authorities)[16]; and a (fluctuating) system of social relations capable of assuming the governance of protection (with or without trade unions). Added to this are the impositions placed on providers of voluntary welfare cover as a result of the emergence and affirmation of a system that forces them to contend with its generosity. Either to fill the inevitable gaps; or to create new needs that justify their very often complementary work; or to help reduce, often with the benevolence of the public authorities, social expenditure that has historically soared since the 1930s.

The question, which covers the entire period from the end of the 19th century to the present day, is how, from a given state of protection – rich in potential before 1914, but unequal, fragmented and devoid of preference for any particular form of protection (apart from voluntary welfare cover) – we have moved on to a cohesive protective state, which in practice favours compulsory welfare cover. This is not to disqualify public assistance or especially voluntary welfare cover, which were both, albeit for very different reasons, at the very heart of the Third Republic’s first republican social model. It is precisely this “historical selection”, which is not exclusive to any particular form of protection and is punctuated by “social models of protection[17]” that are very different from one to another (which puts the notion of a French “model” into perspective), that this call for contributions aims to question, problematise and inform. The general hypothesis is that the French have adapted to what they consider to be their social protection[18] (or “social property[19]”), despite strong opposition, from the top (from the first workers’ laws to social security), from the bottom (through compensation for certain social risks[20] or by comparing one’s own social situation with that of others), and through the public and private provision of an abundance of voluntary welfare cover (often discriminating on the basis of income).

To this end, a number of complementary themes are proposed to researchers from an interdisciplinary perspective, even if the historical perspective logically holds a central place. 


1st line of focus: The first theme invites us to take a socio-cultural approach to the history of social protection. Its goal is threefold.

Firstly, we need to identify and characterise protection cultures based on political or economic visions and discourse, which often give rise to controversy (the disincentivising effects of social benefits, etc.). Then we need to examine these cultures over time, comparing them with the reality of protection to assess how they are used, and any changes they may undergo.

Secondly, to examine the various “social models of protection” that have succeeded one another from the Belle Époque to the present day, taking into account the relative impact of the forms of protection that were put into practice, their distribution within society and their reciprocal organisation (internal economy). This means paying close attention to the socio-demographic and socio-professional segmentations of French society, which have evolved considerably over time. What heuristic models can be identified during the period under consideration? How well did they fit the social structure? What was their capacity for redistribution and what inequalities did they generate, particularly through the degressive mechanism of the contribution cap (created in 1930)? What were the effects of the registration cap (social insurance in the 1930s) on the development of conventional sector or company protection in socio-professional categories that were not subject to insurance (engineers, supervisors, technicians, foremen, etc.)? What influence did the development of a system of protection that was primarily based on salaried employment have on the protection of the self-employed, who are traditionally tied to a security system based on the ownership of their working tools? How were social risks shared between the various operators? What were the interactions, hybridisations, arrangements[21] or, on the contrary, the compartmentalisations or juxtapositions that emerged between the main forms of protection? What extensions, by means of assimilation or analogy[22], to the old model of legal protection for workers (from children to adult men via female minors and adult women), did social protection undergo?

Lastly, to assess, date and document the factors behind French people’s participation in a differentiated system of social protection, which does not compensate for all social risks in the same way (disability annuities and pensions are not of the same nature as sickness or unemployment benefits), and which can be perceived in different ways (benefits considered “undeserved” or, conversely, stigmatising for those who receive them).

2nd line of focus: The second line of focus takes a more dynamic, less endogenous view of the social and health consequences of wars and economic crises which, despite their largely exogenous nature, seem to have had a decisive impact on the extension of compulsory welfare cover in France, if only by upsetting the social and socio-professional structure of the population. In short, they have made it possible (but in what way, and at the cost of what social reclassifications and societal renunciations?) to move from one social protection model to another, without abandoning the intrinsic polymorphism of social protection. By way of illustration, the First World War marked the twilight of the “republican social model”, which was based on a high degree of local autonomy for social and health policies, total confidence in the virtues of savings (now wiped out by inflation), a carefully contained welfare sector, a certain contempt for salaried workers, and self-protection that was indexed to one’s capacity to save, which varied greatly from one individual to another and from one social group to another. At the same time, it would have brought about the introduction of another model, the equally composite social insurance model, characterised by the deterritorialization and professionalisation of social policies, with local elected officials and social reformers overwhelmed by the consequences of the conflict. At the same time, new categories of beneficiaries were created by assimilating war with a new social risk: war wounded (Lugol Law of March 31, 1919), who were covered by insurance-like assistance (presumption of origin and reparatory compensation based on the idea that a social debt was owed to them); war widows and wards of the nation; war victims; and veterans, whose status was created in 1926 (veteran’s card) and which, in 1930, gave them access to various measures of national recognition (creation of a mutual annuity funded by the state, awarding of an annual indemnity, tax benefits).


3rd line of focus: The third line of focus is the relationships of competition or complementarity and, more generally, the interactions and interferences between systemic forms of social protection, embodied social policies and private and public actors (mixed economy of welfare). The assumption here is that the landscape and practicalities of social policy have been radically altered by France’s decision to introduce compulsory welfare cover. The latter became the pivotal axis of these policies, notably through its financing by the social security funds, whose health-related and social action (which is yet to be extensively explored by historiography) made it possible to renew and coordinate policies that had been passed down but hampered by the original polycentrism of French social protection. The political, economic and social stakes involved in the historical linking of pensions and supplementary insurance with systemic social protection (social insurance in the 1930s and social security) could also be considered.


4th line of focus: The last line of focus has two objectives. Firstly, to introduce a comparative dimension that puts into perspective or better defines the specificity of the polymorphous French case. It should also be pointed out that comparing systems of protection is not a recent academic phenomenon. From the end of the 19th, it was developed within the framework of international congresses on welfare and social insurance, which enabled the circulation of theoretical reflections and technical know-how relating to protection, while pioneering examples played out in France[23].

Secondly, and from a more epistemic perspective, to compare the genealogical approach of the polymorphous French “case” with that of the transnational construction of the main social states. In this respect, there would appear to be an imbalance or a strong dissonance between the former, which is based in particular on the archives and writings of the main actors in the field of social protection, whose references and frames of thought are primarily national, and the growing number of works that postulate a circulation of ideas and reforms between comparable countries. To the point, moreover, of “petrifying the very idea of circulation” or inducing the “lazy concept of influence” that “makes us stop and merely consider the idea of exchange without actually doing it”. How can this discrepancy be explained, and how can this “cross-fertilisation”[24] be given its rightful scope?


The Revue française des affaires sociales will be particularly attentive to proposals for review articles focusing on the extensive and cohesive dynamics of social protection. Coordinated reports made up of articles on specific topics will be particularly appreciated. In view of the vast scope of this call for papers and the abundance of related works in all disciplines, and in order to preserve the authors’ ability to make proposals, no bibliographical references are provided. Teams or authors are instead asked to refer to clearly identifiable works in the humanities and social sciences, the references to which will appear in the appendix to the proposals.


Further information on the content of this call for papers can be obtained from the journal team:


Authors wishing to submit an article or report to the journal on this topic should submit

the final version of the text together with an abstract and a presentation of each author (cf. RFAS “advice to authors” [online https://drees.solidarites-sante.gouv.fr/sites/default/files/2021-02/Charte%20deontologique%20et%20conseils%20aux%20auteurs.pdf),

bearing in mind the multidisciplinary nature of the journal and its need to be accessible to a non-expert readership,

before Tuesday, 9 April 2024. 

RFAS is committed to its publication schedule. The dates indicated are therefore firm, except in cases of force majeure. Thank you for your understanding.


Paul V. Dutton is a health and social policy historian at Northern Arizona University, USA. He is the author of Beyond Medicine: Why European Social Democracies Enjoy Better Health Outcomes than the United States (Cornell University Press, 2021), Differential Diagnoses: A Comparative History of Health Care Problems and Solutions in the United States and France (Cornell University Press, 2007), and Origines de l’État-providence français : la lutte pour la réforme sociale en France, 1914-1947 (Cambridge University Press, 2002). His current research focuses on the relationship between health and mobility.

Yannick Marec is Professor Emeritus of Contemporary History at the University of Rouen Normandie. Chairman of the Scientific Advisory Board of the French Society for Hospital History. He is also Vice-Chairman of the Scientific Advisory Board of the Social Security History Committee. A laureate of the Institut de France (Académie des Sciences morales et politiques) and the SFHH, he is the editor of Accueillir ou soigner ? L’Hôpital et ses alternatives du Moyen Âge à nos jours (PURH, 2007); co-edited with Daniel Réguer, De l’hospice au domicile collectif. La vieillesse et ses prises en charge de la fin du XVIIIe siècles à nos jours (PURH, 2013) ; and, in collaboration with Jacques Poisat, Hôpitaux et médecine en guerre. De la création du service de santé militaire aux conflits contemporains (PURH, 2018) and Hôpital, ville et citoyenneté, (PURH, 2021). Together with Antony Kitts and Olivier Vernier, he also co-authored La pauvreté et sa prise en charge en France 1848-1988 (Éditions Atlande, “Clefs Concours, Histoire”” collection, 2022).

Bruno Valat is a history professor at the Institut National Universitaire Champollion, University of Toulouse. A specialist in social protection, he has published several works on the history of social security, social policies and health, including Histoire de la Sécurité sociale. L’État, l’Institution et la Santé (1946-1967), Economica, 2001. He recently published Les Marchés de la Santé en France et en Europe, au XXe siècle (ed.), PUM, 2021. He is editorial secretary of the Revue d’Histoire de la Protection Sociale.

Vincent Viet is a research historian at Cermes3 (CNRS, INSERM, EHESS and Université Paris-Cité). He is a member of the scientific commissions of the Comité d’histoire des administrations chargées du travail, de l’emploi et de la formation professionnelle (CHATEFP) and the Comité d’histoire de la Sécurité sociale (CHSS), and is co-author of the books: Se protéger, être protégé : une histoire des assurances sociales en France, Rennes, PUR, 2006; and Pour une histoire plurielle du paritarisme. Fondements, formes et usages (XIXe -XXIe siècles), La Documentation française, 2021. His two most recent publications are Votre travail nous regarde ! Enquête associative et institutionnelle sur l’Inspection du travail (1980-2020), Nancy, Arbre bleu Editions, “Le corps social” collection, 2022, and La Santé en guerre, 1914-1918. Une politique pionnière en univers incertain, Paris, Presses de SciencesPo, Collection académique, 2015.


[1] As Henri Hatzfeld did in his pioneering work, Du paupérisme à la Sécurité sociale 1850-1940, Paris, Armand Colin, 1971. We believe it possible to adopt a different point of view, starting from the heart of French society during La Belle Époque (the non-salaried middle classes before 1914, the electoral base of the Radical and Radical Socialist parties and the privileged clientele of voluntary welfare cover), and show that, with the slow decline of these strata in favour of the salaried middle classes (via the “economically weak” of the 1930s), coverage of the middle classes (salaried or not) was in reality the historical concern – albeit hidden and deferred by voluntary welfare cover – of the French welfare state.

[2] In other words, focused on a particular form of protection.

[3] For example, the creation of social insurance schemes in 1930 had a significant impact on spending and the scope of assistance, which is quantified annually by the Ministry for Labour.

[4] Little has been done thus far to place gender-focused studies of social protection (or the gendered nature of social protection) (the family-oriented nature of which, as evidenced by the coverage of spouses and children in the event of affiliation, is better known), in their historical context. In particular, the data collected since the 1930s by the statistical services of the Ministry for Labour could be used in ministerial reports on the application of social insurance legislation and on the application of social security legislation from 1930 to 1962 (https://sante.gouv.fr/ministere/documentation-et-publications-officielles/ressources-documentaires/article/rapports-sur-l-application-de-la-loi-des-assurances-sociales-et-sur-l). See also the surveys published in the Revue française du Travail, the predecessor of the RFAS, since its inception in 1946.

[5] The term refers to the institutions, laws and agreements governing each of the main categories of protection (assistance, voluntary welfare cover and compulsory welfare cover) and, more specifically, their breakdown by type of protection (private charity, public assistance, mutual funds, cooperation, workers’ and farmers’ pensions, compensation for occupational accidents, savings, etc.).

[6] Historians speak of “modern” assistance to describe the overhaul of this protection at the end of the 19th and distinguish it from the primitive version that had been in place since the French Revolution,

[7] like car insurance (1958) or home insurance for tenants.

[8] Laroque, P. (1935), “Protection de la faiblesse physique et sociale”, Encyclopédie française, section A – Activités et fonctions de l’État, T. X : L’état moderne. Aménagement et crise.

[9] For a long time, those who promoted and supported social insurance (including the socialists) believed, or pretended to believe, that it would eventually eradicate public assistance, which was rightly or wrongly perceived as the work of the Prince (state, local authorities), whose moods were deemed fickle.

[10] Perhaps the most striking feature is the coexistence of the principle of obligation underlying social insurance – albeit, as in the 1930s, within the limits of a registration cap – with the principle of voluntary welfare cover based on individual initiative, the two of which can theoretically be combined.

[11] This trend continued with the creation of social insurance by the laws of 1928-1930 which, through the establishment of a single institution (subject to payment of a one-off double contribution), covered most of the social risks that had been identified at that time (sickness, maternity, disability, old age, death), with the exception of unemployment and occupational risks. This was confirmed in 1945, when social security took over as a single system covering the former forms of social insurance, family allowances and compensation for occupational risks.

[12] It seems that doing away with the liberal rejection of obligation resulted in greater political encouragement of voluntary welfare cover, the state supervision of which was paradoxically a factor in bureaucratisation, much more so than in Germany or in those states that had adopted the principle of obligation at an earlier stage (Viet V. (2018), “Comment suivre l’exemple de l’Allemagne sans perdre son âme ? Le pari de Fuster sur la Mutualité française dans la lutte contre la tuberculose”, Edouard Fuster (1869-1935) et la construction de l’état social, Les Études sociales, n° 167-168, 1st and 2nd semesters 2018, p. 157-179).

[13] Particularly assistance and insurance, as in the 1930s or, more recently, through CMU, now PUMA.

[14] One example among many: the principle of free membership inherent in mutuality has recently been undermined by company mutuals, which are based on the principle of obligation.

[15] Hence the interest in revisiting the historical and functional continuity between social security and the forms of social insurance of the 1930s, which is often underestimated, if not overlooked entirely. This extensive continuity has been little explored by historians, political scientists, economists, legal experts and sociologists, no doubt because social security was perceived and presented as a sui generis or ex nihilo institution (even by its promoters), or as the fruit of workers’ struggles.

[16] This is the very focus of studies based on the mixed economy of welfare. See the special report in issue 15 of Revue d’histoire de la Protection sociale, 2022: Nouveaux terrains de l’économie mixte du Welfare (XIXe-XXe siècles).

[17] By “social protection model” we mean a configuration of mutually arranged forms of protection, in relation (which does not mean in adequacy) to the social structure of society at a given moment. For example, the pre-1914 social model was organised around a welfare system aimed at the needy and the destitute, a predominantly voluntary system of welfare cover aimed at the bourgeoisie and the unsalaried middle classes, based on savings, capital and property rent, and an embryonic protection system for the “economically weak” (salaried workers and peasants).

[18] We could speak of the “socialisation” of compulsory welfare cover, insofar as it is so popular with the French.

[19] What Robert Castel has historically described as “the property of non-property owners” (labour law and social protection).

[20] As demonstrated by the success of maternity and sickness insurance in the 1930s, despite the fact that contributions were deducted from very modest salaries.

[21] These arrangements can also be of an institutional, financial or legal nature: between social security and mutual funds, which lost their role as manager of social insurance in 1945 while retaining its complementary function; between the public authorities and private insurance companies, which lost their (lucrative) leadership role in occupational risk compensation in 1945 (nationalisation of the eleven largest companies), but assumed additional home and vehicle risks, etc.

[22] For example, the gradual assimilation of disabled workers with disabled war veterans.

[23] For example, the contribution of Paul Pic (1913), Les assurances sociales en France et à l’étranger, Paris, Librairie Félix Alcan.

[24] Quotations are taken from Pierre-Yves Saunier (2003), “Administrer le monde ? Les fondations philanthropiques et la public administration aux états-Unis (1930-1960)”, Revue française de science politique, n° 2, April, pp. 237-255.