Skip to main content Accessibility help




For some years now, the repression of the Algerian Front de Libération Nationale (FLN) by the Paris police on 17 October 1961 has been a subject of controversy. This is despite the fact that those who have written about this incident are in agreement about its magnitude (more than 14,000 people arrested in two days) and its senseless violence. Yet there remains disagreement over the emotive question of the number of people killed. The recent publication of a study by Jim House and Neil MacMaster provides an opportunity to return to this issue, and to raise important questions about the methodologies and standards of research that have been used.


Corresponding author

8 rue Pierre Langlade, 92340 Bourg-la-Reine,


Hide All

1 Jean-Paul Brunet, Police contre FLN. La drame d'octobre 1961 (Paris, 1999).

2 ‘Et cela dans un but politique qui était, autant que d'impressionner le gouvernement français, d'affirmer son poids vis-à-vis du Gouvernement provisoire algérien de Tunis'. François Maspero, Postface to Paulette Péju, Ratonnades à Paris (Paris, 2000), p. 199.

3 See Jean-Luc Einaudi, Octobre 1961: un massacre à Paris (Paris, 2001). The two first chapters of my Charonne: lumières sur une tragédie (Paris, 2003) are a closely argued reply to this work.

4 Jim House and Neil MacMaster, Paris 1961: Algerians, state terror, and memory (Oxford, 2006).

5 Brunet, Police contre FLN, pp. 119–20, 125, 129ff, 154ff.

6 ‘General call to all PM [police municipale] services in the Paris and suburban Paris area. Make known before 0400 to all the EM-PM the number of FMAs who have been sent to the various identification centres and those who might still be there. By the same general call, please indicate the number of FMAs killed or injured in this demonstration.’ Police contre FLN, p. 205. EM refers to état-major de la police municipale (police high command); FMA refers to Français musulmans d'Algérie (French Muslims of Algerian descent).

7 In other words, 2 deaths (a third was to be documented shortly), 1,200 injured, and 11,538 detained.

8 Jacques Delarue was at the time a commissaire for the judicial police. He has since become a well-known historian and witness. Paul Roux was at the time a commissaire at the Renseignements Généraux (RG) of the prefecture of police (in charge of the Maghreb and overseas territories). He became director-general of the RG in 1981 and then prefect. He subsequently volunteered his services for the Ligue des Droits de l'Homme. In my book Police contre FLN, I have cited a number of other witnesses that confirm my conclusions.

9 Brunet, Charonne, p. 26.

10 Saïd Djamouny was found dead on the pavement in the Rue aux Ours, 3rd arrondissement, on 19 Oct. 1961 at 20:25.

11 Brunet, Charonne, pp. 18–21.

12 See Jean-Luc Einaudi, La bataille de Paris: 17 octobre 1961 (Paris, 1991), p. 191, and Octobre 1961: un massacre à Paris (Paris, 2001), p. 83.

13 Brunet, Police contre FLN, p. 244 (see also p. 246).

14 ‘He is drunk and forced to demonstrate. In order to hide, as much from the police as from FLN activists, he walks down to the banks of the Seine. It is dark, he is drunk, he is scared and falls into the river by accident’. Ibid., pp. 209–10 and 177.

15 An extreme example is the file on Mouloud Bouarif, who died on 13 October at the hospital in Saint-Denis. It contains no less than 179 pages of close-typed text. Its conclusion is unclear: the death is said to have been the result of an ‘affaire politique compliquée d'un règlement [de comptes] entre souteneurs nord-africains'. It is in any case clear that Bouarif was some kind of ‘souteneur’ (pimp) who, in the preceding few days, had been terrified because of his inability to pay the significant amounts requested by the FLN.

16 Without attempting to be exhaustive, here are two errors made by the authors. First, the interior minister Roger Frey and the prefect of police Maurice Papon are described as ‘ultras’ (p. 5), when, on the contrary, they always claimed to be loyal Gaullists. The only supporting evidence for this assertion is Maurice Faivres, Conflits d'autorité durant la guerre d'Algérie (Paris, 2004), pp. 50–60. On closer examination, there is little in Faivres's book which would support such a conclusion. Second, House and MacMaster claim that ‘from the 5 September onwards over 500 men were arrested each day, controlled at the CIV and, in some cases, interned in camps or deported to Algeria’ (p. 97). The evidence they use is a report dated 4 October, written by the director of the SCAA and addressed to the prefect of police force. It shows that, during the month of September, 15,414 identity checks were carried out. House and MacMaster have simply confused ‘identity checks’ with ‘arrests’.

17 I consulted the APP following the ‘mission Mandelkern’. I had at my permanent disposal two large files containing a number of uncut documents. There were similarly ideal conditions at the archives of the judicial police at the Quai des Orfèvres. They allowed me to work quickly: consulting these archives took three days a week over the course of fifteen months.

18 Another repeated error on the part of House and MacMaster (pp. 9 and 13) is that I am supposed to have obtained authorization for archival work at the APP on 29 March 1999. Since my book appeared in September that year, it is said that I ‘rushed’ it into print. The implication is that my work is not scholarly. Had they paid attention, they would have seen that I received authorization on 26 May 1998 (see p. 17 of Police contre FLN). Moreover, the two authors claim that, in my ‘rush to get into print’, I ignored a number of FLN documents in the APP. They are referring here to eighty ‘rapports de manifestation’ by Algerian demonstrators seized by the police on 9 November 1961. Again, had they paid sufficient attention to my work, they would know that I devoted two pages to these reports (pp. 319–20) and that I both discuss their reliability and quote from them at length.

19 Brunet, Police contre FLN, pp. 325–31.

20 See in particular Brunet, Charonne, pp. 30–1.


Full text views

Total number of HTML views: 0
Total number of PDF views: 0 *
Loading metrics...

Abstract views

Total abstract views: 0 *
Loading metrics...

* Views captured on Cambridge Core between <date>. This data will be updated every 24 hours.

Usage data cannot currently be displayed