Confiscated Histories

Access to ‘Sensitive’ Government Records and Archives in France


In France, the culture of secrecy continues to dominate access policies. The acceptance of or resistance to this culture by various social actors, including government officials, civil servants such as archivists, historians, independent scholars, and journalists, partly explains the historical tension between advocates of a more restrictive or liberal policy of access to government records deemed ‘sensitive’. Unlike the American case with its long-established right to access, in France, access to information is just starting to be considered a citizen’s right.1 Initial reactions to the first version of my book (1994) sparked a rather violent debate. In the controversy, most of the archivists and some influential historians either denied or justified the difficulty of accessing so-called ‘sensitive archives’. Indeed, thanks to the ‘invisibility’ of this question until then, a book dedicated to the ‘Vichy Syndrome’, which had been published some years before, did not even mention this problem as evidence of France’s difficulties in facing the past.2

Since the American Revolution, the press has often played the role of the citizen’s watchdog and has earned the status as the unofficial fourth branch of government. In part, this tradition helps explain the consensus in favor of government disclosure in the United States and the Freedom of Information Act (F.O.I.A.), which was enacted in 1966. F.O.I.A. did two things: it marked a victory for the social actors who claimed open access as their civil right, and it set an example of transparency for other countries to emulate. France still lags behind many similarly democratic countries, and it has not opened its ‘sensitive’ archives related to the Second World War as well as to the Algerian War of Independence, which are still classified and require a ‘special permission’ (dérogation).

One might argue that France’s history has been less consensual than that of the USA. On the one hand, national scandals and state secrets in France, such as the nature of collaboration between Vichy France and Nazi Germany or France’s role in the wars of national independence, have conspired against liberalizing access to government records, especially by social actors who have a vested interest in keeping their past secret. On the other hand, the revolutionary tradition of protecting civil and human rights has militated in favor of expanding access to government records. After a brief presentation of the French legal framework governing records, i.e., the three laws of 1794, 1979, and 2008, I will give an overview of the so-called crise des archives that is impacting the Archives nationales, provide some examples of ‘sensitive’ archives, and analyze the current issues in public discourse and academic debates.


In France, with each piece of major legislation on the issue (1979 and 2008), the notion of ‘one step forward, two steps back’ characterizes what appears to be a paradox: the expansion of access to certain categories of government records along with the introduction of new restrictions. This paper argues that despite what appears to be an expansion of rights, the culture of secrecy and restricted transparency still dominates access to information in France and, moreover, a variety of social actors (archivists, historians, independent scholars, and journalists) end up going along with the system. Willingly or not, they contribute to the persistence of this culture of secrecy that pits privilege against civil rights and, in many ways, serves their professional interests.

The French Revolution granted a series of rights to the country’s citizens. Among them, the so-called law of Messidor An 2 (according to the revolutionary calendar, 1794) granted free access to archives for all citizens, thereby abolishing the prevailing notion that state documents are the property of the monarchy. This liberal law enabled scholarly research on the Ancien régime. Until then, only the monarch’s historians were given access to government records for the purpose of legitimizing royal power. By providing this access, the French Revolution contributed to the training of professional historians and freed them from their dependency on the monarch or government. After the Revolution, access to public records stopped being a privilege and became a right. However, and this is an important caveat, the law did not apply to recent government records, i.e., those documents produced after the Revolution. There were good technical reasons for this: archivists needed to collect and then process the documents. Nevertheless, during the course of the nineteenth century, several decrees were implemented that significantly restricted the 1794 liberal right-of-access legislation by extending the waiting period for access to government records and arbitrarily authorizing access – if at all.

The first attempt to broaden access to records came as a result of the First World War. The Ligue des droits de l’homme (League of Human Rights, created during the Dreyfus Affair) tried in vain to gain access to the (military or court martial) files of the soldiers who had been convicted of ‘mutiny’ and executed in 1917. Most of them (les mutins) did not even get a trial. In the hope of rehabilitating the condemned soldiers, the Ligue tried to obtain documents that they were told had been destroyed or classified.


After France’s liberation in 1944, in contrast, the postwar purges and trials of collaborators (Epuration) was made possible by the availability of documents that provided evidence of collaboration with the Germans or of the repression of French resistance fighters. However, a few years later, in 1951 and 1953, two laws were passed that granted amnesty to all government officials of the Vichy period. For all intents and purposes, this legislation led to the complete sealing of Vichy government records.

The consequences of these restrictions are well known. The limited sources that were made available to historians facilitated an interpretation of Vichy (les années noires) that posited that Vichy officials had been strategically playing a double game with the German occupiers and choosing ‘the lesser of two evils’ (e.g., favoring collaboration over war with Germany and deporting foreign Jews presumably to save French Jews), while French society, it was claimed, almost uniformly resisted. This mythe résistancialiste was discredited decades later by the scholarship of foreign historians Eberhard Jäckel and Robert O. Paxton.3 Vichy archives remained closed, but in order to overcome the restrictions imposed on historical research, Jäckel and Paxton consulted the German archives that the Allies had captured, brought to Washington, and then returned to Germany in the 1950s and 1960s on the condition that they be made available for historical research. By contesting the prevailing ‘patriotic version’ of France’s role during the Second World War, Jäckel and Paxton sparked much controversy in the academic discourse.

In 1978, a law promoting the idea of government transparency facilitated the passing of a law that aimed to liberalize access policy and chip away at the ‘culture of secrecy’. As a consequence, new legislation on Access to Archives had to be passed. The French legislation of 1979 was also a delayed response to the American Freedom of Information Act of 1966, which allows for any person to consult government records after 30 years. Following the American example, many Western democracies reduced the waiting period to this time span. With the law of 1979, France was one of the last democracies to do so.


However, the same law that established access also created five new categories of government activity that restricted access by increasing the waiting periods. For example, the waiting period to consult administration (government) records was increased to 60 years, and to consult the personnel dossiers of civil servants and officials to 120 years. These kinds of delays made it impossible for anyone to examine the dossier of any government official in the course of his or her lifetime. Beneath the appearance of liberalization, the new law in fact increased the waiting period for access to the government records concerning Vichy and decolonization. How is it possible to research the role of civil servants if their files are inaccessible, the personnel dossiers of Vichy officials remaining sealed for 120 years? Arguments such as the protection of ‘personal information’, ‘privacy’, ‘national security’, and ‘national reconciliation’ could be seen as an attempt to protect state secrets or the misdeeds of public servants from disclosure and to thwart the efforts of scholars and journalists to investigate government activities.

The purpose of the 1979 law was twofold. A procedure was established to encourage and regulate the transfer of government papers (presidential papers, papers of the Ministry of the Interior, of the Economy, and so forth) to the National Archives. A procedure to request ‘special permission’ was instituted to allow researchers to access restricted government papers. The problem, however, was that this dérogation was not granted to all and that it was dispensed arbitrarily. Depending on whether one worked on a sensitive subject or one considered sensitive (which was often a question of the individual archivist’s mood at the lower echelon, or the liberalism of the institution which created the document at the higher echelon), the response might be ‘everything is open’ or ‘everything is classified’. One might be disappointed or grateful. Some historians received the response immediately; others had to wait for months before receiving an answer – which in some cases was negative. Sometimes people never received an answer at all. Moreover, most researchers were unaware of the special permission clause. It became a sort of secret shared among a privileged few working in the National Archives.

More than a decade later, in the 1990s, two scandals shook the historical profession as well as the archival world, setting the stage for the so-called crise des archives. The first scandal occurred in November 1991, when the French lawyer Serge Klarsfeld, best known for his work on the French persecution of Jews during the Holocaust, discovered the Jewish registration file (fichier juif) that had been processed by the French police (Préfecture de police de la Région parisienne) at the instigation of the German occupying forces in October 1940. This document had proven to be a decisive tool when the French police and the Gestapo rounded up and arrested Jews. Believed to have been destroyed after liberation, Klarsfeld found a reference to it in an internal document – an archival finding aid – that scholars and other patrons were not allowed to see. Actually, the fichier juif had been used after the war for processing survivors’ claims. Klarsfeld’s discovery dredged up the memory of the role of the police in occupied France. It also provoked fear among some of the survivors who demanded that their files be returned to them; and it demonstrated that the sheer existence of ‘sensitive documents’ like these remained clandestine and undisclosed. (In a similar fashion, in the earlier historical controversy over the Jewish captain Alfred Dreyfus in 1894, his personnel dossier was declared lost. Even after the dossier suspiciously reappeared, it remained inaccessible until in compliance with the 100-year-rule.)


The second scandal occurred in the late 1990s, and it concerned the killing of Algerian activists in Paris in 1961. A researcher wrote about the brutal acts of the police ‘under the orders of police chief Maurice Papon’. Witnesses testified that hundreds of people had been thrown into the Seine and were never found again. By this time, in the 1990s, Papon was already in retirement and had just been sentenced for his collaboration in the 1942 deportation of Jews in Bordeaux, where he had been a Vichy official. Due to the classified nature of Vichy government records, Papon had escaped investigation at the end of the war. He even claimed that he had been a resister. From his jail cell, Papon sued the researcher who had accused him of having given the order to throw demonstrators into the Seine: how could the researcher prove his claim? Papon was right – the researcher could not prove it, although it was obvious that the police had been carrying out the orders of their superiors. The only possibility to provide evidence was to gain access to the sealed police documents, which was refused to the researcher. Fortunately, two archivists decided to ignore the ‘duty of confidentiality’ and agreed to testify in court on behalf of the researcher. They said that the documents to which they could not give access, but the contents of which they knew, confirmed Papon’s role in the repression and killing of the Algerian activists. Papon’s case was dismissed, the researcher acquitted, and the two archivists ostracized (put, as we say in France, au placard): they were forbidden further contact with patrons of the archives and shunned at their workplace. Had this problem occurred some years earlier, the two archivists surely would have been fired or otherwise sanctioned.

In the meantime, it became increasingly apparent to historians, journalists, and archivists that the current special permission system (dérogation) to consult government records was arbitrary and showed that France was lagging behind other democracies in government transparency. The case of the mutiny of soldiers during the First World War resurfaced in the public debate regarding access to judicial archives (100 years). By the late 1990s, the government promised a new law. But it took another 10 years before the new legislation was passed.

The enactment of the new legislation in July 2008 provoked a national debate over the arbitrary definitions and subjective interpretations of ‘privacy’, ‘national security’, and ‘national defense’ that enabled an abuse of the notion of transparency. Some journalists and a few historians mobilized to condemn the practice of restricting access and even concealing ‘sensitive files’. This new legislation in many ways recalled the experience of 1979: On the one hand, the timespan of confidentiality was reduced to 25, 50, and 100 years for some records, but on the other hand, it increased for certain others. While the legislation of 1979 had increased the waiting period primarily for Vichy records, the legislation of 2008 created a catch-all category of government records related to the security of civil servants and to national security (documents relatifs à la sécurité des personnes et à la défense nationale), whose confidentiality period is 100 years now! And, last but not least, the law also created a totally new and unique category of documents concerning the ‘manufacturing process of weapons of mass destruction’, whose access was forbidden for all eternity. To be honest, some progress was made concerning access to documents related to property (minutiers des notaires), for example, and others related to routine government matters (e.g., drivers licenses). But this article focuses on contemporary political records, not on the archives of the nineteenth century, for example.


De facto we find almost the same barriers as before: ‘sensitive’ archives (police records, diplomatic negotiations, intelligence reports, political surveillance, and the like) continue to evade the transparency rule. In other words, national security (the raison d’état) triumphs again and again over civil rights. As long as notions such as ‘privacy’ and ‘national security’ have no clear definition, they may be also used as alibis to protect the behavior of high-ranking government officials and civil servants.4

I would now like to briefly elucidate the reasons why legislation in France remains less liberal than in other countries. Even if we acknowledge that no government is eager to communicate its records and that high-ranking officials are inclined to hide the deeds of their predecessors and protect them from embarrassment, how do we interpret what appears to be a combination of inappropriate solidarity and raison d’état?

This issue of information policy concerns and affects all citizens. Still, the principal social actors are the historians. Although they are the main users of the ‘nation’s memory’, until recently historians have paid little attention to, or have remained virtually silent on the question of access to contemporary French archives. The ways in which historians have conformed to the law and obeyed rather than questioned its restrictions expose questionable intellectual practices and raise questions about their social role and political responsibility. The same could be said of the archivists, who play an important (though less visible) role in implementing the law governing access. After all, it is they who perform the first series of triage after a historian requests a dérogation.


The most recent law (2008) has improved the system. Since then, researchers are better informed about their right to request special permission (but not necessarily to gain special access) and to expect a timely response from the archivist. But in order to be eligible for a dérogation, the applicant is required to explain her/his research, indicate her/his own titles or the names of her/his PhD advisers, and other information in addition to signing a confidentiality agreement acknowledging not to disclose what the law considers private. The application process itself as well as the amount of time it takes to secure access to classified materials in many cases discourages ‘simple’ citizens, foreign scholars, or journalists who require a quick response due to the timeliness of the news. Nevertheless, the ‘special permission’ provision has made academic research easier than before. But the question is: at what cost?

The current system of ‘special permission’ highlights the predicament of historians dealing with recent events. In this regard, they differ from their colleagues researching earlier periods, whose work may have no immediate social or political implications. Historians who are put in the position of having to request ‘favors’ from a government office are subjected to a policy verging on illegality in the sense that the granting of special privileges, in an open society, might signal an abuse of power. Instead of being upset that access to archives is difficult to obtain, some historians may take satisfaction in being ‘in on’ the secret, real or fancied. By pledging non-disclosure to the government office granting special permission, historians do gain access to ‘sensitive’ documents, but risk losing their independence. In many cases, they are legally forbidden from disclosing the details of what they uncover, such as the names of officials. Knowing if this is decisive or not, and whether they respect their commitment or not, does not alter the question of the principle involved. Bound by their commitments, historians find themselves in a situation of dependence and of self-censorship vis-à-vis the state.

The notion of ‘privacy’ is highly questionable. Respect for privacy, to which every citizen is entitled, denies in practical terms access to public documents detailing the activities of officials, who enjoy a form of immunity. Seeking ‘special permission’ to access documents raises moral and methodological questions. Some books become reference works by default because they are the only ones available by virtue of their authors’ privileged access to holdings that remain inaccessible to others. What about the rule of scholarship according to which every peer should have the opportunity to examine or review a quoted source? And how can we explain the fact that there are two categories of citizens – private citizens (e.g., genealogists, journalists, and independent scholars) who encounter difficulties in gaining access to the archives, and scholars doing academic research who enjoy privileged access?


If the debates surrounding the 2008 legislation changed ever so slightly the thinking of professional historians, they did not change the historians’ reluctance to admit their own privileges. Most historians continue to claim that nothing would be gained by opening the archives to all and that, on the contrary, the officials at the National Archives, irritated by all the fuss, would only set up further roadblocks. This attitude reflects an insular political culture. Such prudence is part and parcel of the advantages of the special permission system, first on the symbolic level (recognition of the scholar as standing above the crowd), and second on the material level. Still, the debates provoked by the 2008 law give us reason to hope.

We have seen how various public scandals involving the nation’s past and present gave historians and journalists the opportunity to protest against new restrictive categories such as ‘eternally inaccessible documents’. Strangely, some scholars do not take it seriously: who really believes it is possible to find in the public archives information on how to produce weapons of mass destruction? The question is not whether such data are to be found. The problem is that such a category exists in the first place. Do we need to be a lawyer or philosopher to understand the repercussions of such categories?

At least it has become possible to question the state’s control of the archives. Recently, a journalist denounced in Le Monde not only the impossibility of accessing the files of the domestic French intelligence services (Direction centrale du renseignement intérieur – DCRI), but also the fact there is little or no form of social control or public monitoring of the intelligence services – not to mention access to their records and even archives (older records).5 Here, it is important to point out the paradox that French historians – who believed that the former Communist countries should open the files of their secret political police – have so far made no efforts to achieve this in their own country!


It is a risky business to conduct research in contemporary history. The need of modern societies to access information about the recent past and about the events citizens have witnessed and endured is a relatively new claim. It demands an open information policy under the effective control of society, without arbitrary restrictions or privileges.


1 Sonia Combe, Archives interdites. L’histoire confisquée, Paris 1994 (third revised edition, Paris 2010). I am deeply grateful to Harriet Jackson and Eva Schissler who revised the entire English version of this article.

2 Henry Rousso, Le syndrome de Vichy, Paris 1987.

3 Eberhard Jäckel, Frankreich in Hitlers Europa. Die deutsche Frankreichpolitik im Zweiten Weltkrieg, Stuttgart 1966; Robert O. Paxton, Vichy France. Old Guard and New Order, 1940–1944, New York 1972. See also Claudia Prinz, Vor und nach Paxton. Der Paradigmenwechsel in der Deutung des Vichy-Regimes, in: Zeithistorische Forschungen/Studies in Contemporary History 4 (2007), S. 263-268.

4 Gilles Morin, Archives: secret et patrimoine, in: Histoire[at]Politique 5 (2008), URL:

5 David Dufresne, Le renseignement français doit pouvoir être contrôlé, in: Le Monde, 31 May 2012.


Copyright © Clio-online – Historisches Fachinformationssystem e.V. und Autor/in, alle Rechte vorbehalten. Dieses Werk ist zum Download und zur Vervielfältigung für nicht-kommerzielle Zwecke freigegeben. Es darf jedoch nur erneut veröffentlicht werden, sofern die Einwilligung der o.g. Rechteinhaber vorliegt. Dies betrifft auch die Übersetzungsrechte. Bitte kontaktieren Sie: <>.

Für die Neuveröffentlichung von Bild-, Ton- und Filmmaterial, das in den Beiträgen enthalten ist, sind die dort jeweils genannten Lizenzbedingungen bzw. Rechteinhaber zu beachten.