Investigating homicide investigation in France

The aim of this article is to explore how criminal justice agency personnel in France respond to homicide. The French system remains, despite shifts towards adversarial elements and recurring attacks on the pivotal role of the eponymous Juge d'instruction, one of Europe's inquisitorial judicial systems, and its structure of judicial supervision of police enquiries and epistemological ethos of truth seeking has periodically been advocated to improve the probity of cases in countries like England with adversarial systems pitting partisan cases supported by discrete pieces of evidence. This article explores how the supervisory system works in practice, the methods and thought processes of the French investigators in all their forms, and the societal ethos within which they operate.

The aim of this article is to explore how criminal justice agency personnel in France respond to homicide. The French system remains, despite shifts towards adversarial elements 1 and recurring attacks on the pivotal role of the eponymous Juge d'instruction, 2 one of Europe's inquisitorial judicial systems. Its structure of judicial supervision of police enquiries and epistemological ethos of truth seeking has periodically been advocated to improve the probity of cases in countries like England and Wales with adversarial systems pitting partisan cases supported by discrete pieces of evidence. This article explores both how the inquisitorial legal system frames police procedure and how French culture shapes police practice and how the two combined affect the nature of murder investigation in France. The inquisitorial legal system will be examined through one of its distinct constituents Á the reconstitution or reconstruction Á which occurs towards the end of the investigation stage of the process. Aspects of French culture Á such as the workÁlife balance and attitudes to women Á will be addressed in a second section. The conclusion will discuss the inevitable contradictions between the inquisitorial ideal of the system, the 'humanity' of the detectives' world view and the actuality observed.

Methods and data
Empirical data reported in this article are drawn from a four-year ethnographic study of French and English 3 'murder squad' detectives. Because of this source of data, although the focus of this article is on the French police their work will be implicitly and sometimes explicitly compared with the working methods of their counterparts in England and detectives working in a more adversarial system. As well as 12 months of overt observation of detectives at work (covering aspects of 35 different cases in France), 20 formal and numerous informal interviews were undertaken with criminal justice personnel (specialised squad detectives, local detectives, prosecutors and judges) in the two countries and case and policy files were examined, as well as cautious comparison of available statistics. Much of the French research was done in a city setting with a specialised murder squad. This 'standing army' option was chosen for practical reasons in order to maximise the potential for relevant cases and overcome one major difficulty in researching murder investigation Á that of ensuring research presence at the occurrence of an unpredictable event. The extent to which practice observed in specialised squads was representative of practice in other areas was assessed by short-observation periods in other sections of the police in both countries Á both a serious crime unit in a rural area and generalist detectives in urban stations. There were very few overt limitations put on data collection in France. The French police were not used to being researched by outsiders and for this reason access agreements are perhaps less nuanced than might be the case in the UK or the USA. They were, however, used to being observed by other police officers Á a key component of French training is following a period of time (a 'stage') with a unit following what they do Á and therefore officials do not find being observed off-putting or necessitating a front. In an alien culture, it was often possible to draw false conclusions about sensitivities. Initially, for example, I was not allowed into post-mortems but this turned out to be only because I did not have the right form stamped by the prosecutor's office. Murder investigation tasks happen in parallel undertaken by different members of the teams Á it is sometimes difficult to tell if one is not invited to observe one particular task because the investigators do not want you there or because they think you will be more entertained elsewhere. In France, I was allowed into interviews with families and offenders (neither of which was allowed in England) but permission was not granted for me to speak to a psychologist who did offender profiling.

Criminal justice actors
In France, suspicious death is investigated by Officiers de la Police Judiciaire (OPJ), an accreditation giving holders specific rights in relation to criminal investigation such as arrest and search under supervision initially of the Procureur and later of a Juge d'instruction. The French Procureur, though usually translated as 'prosecutor', is a 'different creature' (Hodgson 2000, p. 143) to the Crown Prosecutor in England. The Procureur in the French system becomes palpably involved in criminal cases at an earlier stage than within the adversarial system, attending the crime scene, for example. Procureurs are officers of the state sharing the same magistrat training and viewpoint as the Juge d'instruction and the trial judges. They have a wider remit to ensure that justice is done and certainly do not have the same attacking function vis-à-vis the defendant in the French trial that occurs in its English equivalent; some of this function is performed by the lawyer representing the victim (see below).
As a role foreign to Anglo-Saxon culture, the Juge d'instruction is a difficult concept to translate. Many translations only partially capture all the connotations of the role often over-emphasising one at the expense of another. For example, 'investigating judge' conveys well the concept of the Juge being involved in the investigative stage of the enquiry Á the instruction Á but suggests a rather more active role than is actually the case while also suggesting a judging function which is an anathema. 'Examining magistrate' is better in depicting the role of the Juge in overseeing the work of the police, particularly by inspecting files presented to them, and also re-interviewing witnesses, but it still fails to convey all the status of the role including the power over the OPJ of the two police forces in France Á the civil police the Police Nationale, who mainly operate in the cities, and the military police, La Gendarmerie, who work in the rural areas.
In large cities, a special squad of OPJ deals with 'crimes' Á which in French has a narrower meaning than in English, meaning a very serious offence and used quite often synonymously with the word 'murder' Á a Brigade Criminelle. Officers tend to work in small groups (of six or seven) led by the 'chef de groupe', the team leader. A 'Commissaire' (roughly superintendent) oversees these groups and liaises with the prosecutors and the Juges. These teams do not in general investigate 'simple' homicides Á those for which a suspect is immediately evident Á unless the case is for some reason sensitive, which in France would often mean political. 'Simple' homicides are assigned by the prosecutor to be investigated by detectives locally unless it later transpired that there was a problem in which case the prosecutor could reassign the case to another team. In smaller cities, which have no Brigade, all homicides would be investigated by these sorts of teams.
Within the Gendarmerie murders would also be investigated by OPJ. Complex cases would be given to officers from a Section de Recherche (SR). A SR covers a much wider geographical area Á two 'départments' or counties Á than a Brigade though not the cities in these areas. The remit of SRs varies a little regionally but would probably have financial, drugs, personal violence and serious theft teams. SR officers live in barrack accommodation and when active on a case are often displaced for weeks or months nearer the scene. Local officers are used to form the bulk of the murder investigation team as is the case in many forces in England and Wales. Again the SR does not investigate 'simple' cases, such as violence between partners, which are processed locally.
Allowing for differences in legal categorisation, there are roughly the same number of homicides per capita in France as in Great Britain (Eurostat) with the same kind of distribution patterns Á with more homicides generally in city areas (with the exception in France of Corsica historically because of its unique culture and more recently because of links with organised crime (Mucchielli 2002a). Because of more relaxed gun laws allowing for greater legal possession of firearms in France, there are slightly more firearm homicides in France than in England. There are also a high percentage of cases where the victim is known in some way to the offender (80% according to Mucchielli 2002b).

Section one Á inquisitorial legal structure
There are four key stages to an investigation into suspicious death in France: (1) en flagrant délit 4 under the direction of the prosecutor, immediately after the crime has occurred; (2) the enquête préliminaire 5 for crimes older than eight days 6 ; (3) the investigation under Commission Rogatoire (CR) 7 once the prosecutor has nominated a Juge; and (4) the investigation by the Juge once someone is Mis en examen. 8 It is these last two stages and particularly stage 4 which distinguishes the French system as the Juge pursues investigations into the 'why' of the occurrence of the crime Á to ensure not just that the legal qualifications of the offence are met but to uncover 'the maximum information' (F412). 9 Towards the end of this fourth stage, once all the evidence and statements are known the Juge's office usually organise a reconstitution Á translated into English as 'reconstruction', 'to form again'. In French criminal law, the reconstitution is a reconstruction of all or part of a crime involving the accused and/or witnesses and takes place at the scene of the crime. The reconstitution can be seen physically and symbolically to epitomise the culture and ways of thinking lying behind the Gallic criminal justice process presenting visibly the legally prescribed roles of juge, police, suspect, the victim's family, the media and the relationship between them resulting in the much vaunted manifestation de la vérité Á uncovering of the truth. The extent to which that culture and schema is reflected in empirical practice will be explored in the following sections.

The reconstitution Á judicial supervision
The reconstitution is directed by the myth-shrouded figure of the Juge d'instruction who, in the inquisitorial ideal, impartially gathers information from all parties to the event Á the police, the suspect, the victims or their representatives, and the witnesses representing society as a whole Á to uncover the truth during the reconstruction, the police play the minor role supposed by many who eulogise the inquisitorial system, they are subservient to the Juge's command.
Under the CR, the Juge legislatively directs the police enquiries Á on paper the magistrate is described as the 'directeur d'enquête' (F98). The extent to which the Juge does this can vary from individual to individual. On the first day of research, a police officer said: 'Police officers have the power to do what they want, the Juges are limited in the extent of their intervention, even in serious offences like murder' (F1). This assertion seems to be backed up even in the wording of some CRs where busy Juges write that they are delegating 'investigations' to the police 'in the impossibility' of doing tasks themselves. Observed contact was sporadic and seemed mainly on the initiative of the police keeping the Juge informed and asking for telephone and scientific requisitions that only the Juge could commission, with the Juge only asking police officers for specific defence and victim requests to be carried out. While more detailed CRs appeared to be given to local forces particularly where two separate teams were operating on the same enquiry and there could be communication difficulties, police and magistrat within the closed community of the elite murder squad agreed that supervision was often merely contact. One Juge said: 'They know what to do. That's why they are there. We can discuss the cases as intelligent people can discuss anything but we don't have to tell them how to do their investigation' (F434). Nevertheless, the hierarchy is preserved Á 'The police decide what to do, they take the initiative, we are not told 'do this' 'do that' but if the Juge decides they want something done it is done' (F28), a police officer said. To the Anglo-Saxon eye this supervision limits the opportunity that the police would have to 'construct a case' and then turn a blind eye to other possibilities but this way of thinking is not in any case in the French police mindset. One chef de groupe said: I think a great difference between here and England is that you in England present one case and the mention of other possibilities is seen as a weakness whereas here when I talk about all the facts, and facets and false leads, it demonstrates the depth and complexity of the investigation. (F754) Detectives interviewed relatives and associates of the suspect in order to understand the scenario of how the crime occurred Á how the victim and suspect, and others, had interacted to produce the crime event. While such interviews occur in English investigations too, in France the focus is wide and historical Á eliciting information on the childhood and life views of the suspect and at what point their lives took a wrong turn.

The reconstitution Á the victim's family
The legally constituted role for the victim's family as partie civile is distinctive in the French system. Their right of access to information regarding the enquiry is visually symbolised by their presence, or at least representation, at the reconstruction and their right to have questions asked. Once an instruction is begun, the family of a victim of murder, in common with victims of other serious crime, can constitute themselves as parties civiles. This means they are able, through a lawyer, to have access to the full dossier, have opportunities to speak to the juge in the case directly, are kept informed of developments and can petition the juge for ancillary investigations (demandes d'acte) to be carried out. They can also be represented in the court hearings, both in relation to the criminal charges and the civil hearing regarding damages that follow immediately after a guilty verdict. 10 In France, too interest groups and associations can also constitute themselves parties civiles in trials relating to their area of concern. All this has the effect of widening the remit of the trial and the issues that might be covered and sounds as though the victim has a more precise and valued place in the French criminal justice system than in England and Wales.
However, in a number of reconstructions witnessed during the research, the family of the victim, though present, stood aloof from the proceedings largely ignored in one case even by their lawyer and at least one prosecutor told me that in fact the treatment of victims in France was at best average. Most crimes are not looked at by a Juge at all and the treatment of the victims of such crime was often 'bad' and even victims of serious crime found it difficult to find a lawyer. 'With murder, it's ok, the police treat the families of victims very well' one Juge said (F439), though also citing a case where the brother of the victim had been shooed away unceremoniously from a crime scene Á 'the needs of the investigation are paramount'. Police officers told me that although it was mainly the chef de groupe who spoke to the family, 'we always have time for family' (F32). This time is, however, limited. In one case, I heard an officer say to the dead man's daughter that she was welcome to ring him but 'not too often' (F182). Certainly, the French police structure does not provide for the systematic use of dedicated Family Liaison Officers in murder cases introduced following the recommendations of the 1999 Macpherson Report in England and Wales. Saleem (2011 p. 15) highlighted the contrast in this regard in connection with the killing of two Frenchmen Laurent Bonomo and Gabriel Ferez in London in 2008 Á 'where one informed the victims' family regarding the death via a note stuck on the front door, the other designated full-time French speaking FLOs to the families of each victim'. However, in other cases, detectives were seen showing much care and consideration for victims. In a crowded working environment where privacy was rare, officers tried to find empty offices to interview family members but did not always succeed. 11 In many cases, the contact between detectives and family was mainly by telephone and families did not fully partake of the opportunities offered to them though whether this was because of lack of confidence and knowledge of the system, or because they were content with what had already happened is uncertain. Jouve (2004), in his journalistic study of the Parisian murder squad, suggests a closer, more emotional relationship, between police and family in some cases, notably those involving the killing of children. Nevertheless, whatever the level of attention given to the victim's family during the police investigation, the Juge's work in trying to make the victim 'a person rather than a name' (F728), including information about them in the dossier, realigns the focus of the eventual trial.
The reconstitution Á the treatment of the suspect During the reconstitution, the suspect is treated with consideration and is invited to take part in the dialogue. This might appear at odds with the infamous reputation of the French system of custody (garde à vue), suspect interview (evocatively called interrogatoire) and the frequency of subsequent confession, not unassociated with the violence reported Á the most notorious case concerned Ahmed Selmouni, who took France to the Court of European Human Rights alleging he was both physically and seriously sexually assaulted while in police custody. 12 Many professionals working in this field Á both police and magistrat Á are recorded as accepting even an 'interrogatoire musclé' (Le Taillanter 2001, p. 196) as part of the pressure of the police station where 'there are no innocents' and a slap across the face was not 'shocking' (Hodgson 2001, p. 354). The character of the French interviews observed was certainly different in terms of atmosphere, location, personnel and tempo to that in England.
Although there were sometimes bespoke interview rooms for prisoners in the police building, the suspects were often interviewed in the offices of the detectives, to enable the interviewing officers to manipulate the atmosphere. The doors to the office in which the interview was conducted were unusually shut to the casual passerby coming in to say 'Bonjour', but officers besides the interviewing officers did go into the room during the interview precluding the sterile seclusion of much of the English procedure. The defendant was interviewed alone against two detectives. At the time of the research suspects in France had fairly recently won the right to see a lawyer at the beginning and at the 20th hour of custody. The visit was, however, limited in its scope to English eyes Á the lawyer had no access to the dossier at this stage and was expected merely to check that the defendant was being treated satisfactorily and make a note to that effect. The French detectives said they would not welcome the introduction of a lawyer into the room. One chef de groupe said: 'In England you give too many rights to the suspect and then you have to spend more on policing to make up for it' (F302). The whole group concurred: 'We rely on creating the right conditions to make it easier for him to talk, to tell the truth. As soon as you have a defendant with a lawyer, you have a strategy, and not the truth' (F804). 'There's too much about proof for us in the English system, it industrialises the process. We rest human. English police are only interested in covering themselves not the passion to find the murderer' (F304).
One needs to engage with murderers. Often they are not the easiest of people or in the best frame of mind, and this is not going to be improved with a tape recorder, an empty grey room and a solicitor advising them to stay stum. (F604).
Before interviewing the accused, the officers discussed not only their prisoner's character and how best to treat him 13 to encourage him to talk to them, but also how best to match him with one of the team, recognizing the forthcoming encounter as 'a complex confrontation between two personalities' (Diaz et al. 1994, p. 174). While this could be viewed as a strategy in itself, the officers treated the suspect as 'a fellow human being', often explaining what they were doing and chatting. Officers spoke on one case about the need not to make a particular suspect feel in any way slighted or inferior as they felt his violence had stemmed from his not believing others took him seriously enough. Other writers have spoken about officers talking to prisoners freely between interviews, even sharing sandwiches if sufficient rapport had been built up (Diaz et al. 1994, p. 174). There was shouting and cajoling. One interviewer described his interviewee's demeanour shortly before he confessed: 'He's pacing the room like a wild animal'.
This confession remains pivotal, at least in the culture of the French investigation. Described variously as the 'queen of proofs' (Dawant and Huercano-Hidalgo 2005, p. 63) and 'the obsession of police officers' (Van Geirt 1995, p. 51), the 'religion of the confession' (Bell 2001, p. 113) is presented as a battle between investigator and suspect 'a fight without mercy' (Van Geirt 1995, p. 51). Officers in the research presented more of a rounded picture.
Confession is important but not obligatory. Usually at the Brigade you have been looking at a suspect for several months. You have several material proofs. It would be a mistake not to have these; otherwise if he goes back on his confession the case could be confounded. One interviewee (F809) 'The best moment for us is before the arrest, when you know everything and you know who the killer is. The confession too, but that comes almost as a relief', said another (F807). Legal theorists have suggested that the confession in France has less of a symbolic significance and is merely a practical equivalent of the guilty plea in Anglo-American systems (see Van den Haag 1984). In France, there was traditionally no plea-taking ceremony with the defendant (Hodgson 2012) and this is still the case with crimes. The logic of the French system is that a criminal offence has to be fully investigated so that the truth can be manifested regardless of any admissions. In reality, however, both for volume crime and for the more serious interpersonal crime with which we are concerned here, a confession is a major element of a solid dossier. A culture of confession looms large in totalitarian societies, such as China, and shame-based communities, such as Japan. However, the confession is France is also seen as psychologically beneficial to the accused (if guilty) and to the wider society, in terms of reintegration and explanation. 'He's crying now, talking about his father, his family. This happens a lot. In the end they are relieved to have told the truth' (F819), said one police officer of a suspect.
In 2011, France introduced further rights to the suspect including being told of his right to silence and to have a lawyer present throughout the interviews. It remains to be seen how this will affect the psychological strategies and atmosphere manipulation of the French police but one told me that it might encourage quicker confession if the lawyer was 'watching his watch'. Because of the low hourly fees paid by the state (fees so low that it resulted in strike action), the detective felt that it was unlikely lawyers would be inspired to do a very detailed or effective job.
The defendant can spend two years 14 between being Mis en examen and his trial on remand. Leigh and Zedner (1992) stated that France had one of the highest rates of pre-trial detention in Western Europe when their report was written in the 1990s and this remains the case despite the introduction of the le juge des libertés et de la détention in 2000. On several occasions the defendant is transferred from his prison to the offices of the Juge d'instruction to be interviewed. The Juge confronts the defendant with inconsistencies between his account and other witnesses' putting each point to him and noting his response. The defendant spends time with professional mental health experts, whose reports on his personnalité Á the French word encompassing both the English homonym and also the accused's psychological state at the time of the crime and life events and circumstances in order to assign legal culpability but also any potential for recovery or rehabilitation in the future Á he was asked to comment upon. Each interview takes place in the presence of the defendant's lawyer who is also allowed access to the entire dossier.
The reconstitution Á the role of the media in French murder investigation The media are excluded from the reconstitution but later cobble together a report based on comments from whichever of the parties can be persuaded to talk to them. During the investigation, the media are kept at a distance and spoken of with contempt by the police. They are not considered part of the truth-finding team; they report their own version of events. 'We don't trust journalists', said one officer (F206). 'We've not gone to the press, it's not the French way', confirmed another (F454). The Juge agreed: 'Culturally we don't do it in France. In Anglo-Saxon countries I know they have posters asking for witnesses. There is a different civic sense here and I feel if you were to ask the press for help they would want things in return and they would deform everything'. Legislatively all legal actors within the investigation are bound to keep the details of the case secret Á le secret de l'instruction 15 Á but as one officer commented 'it is broken everyday', particularly by the defence team if they feel they have something to gain by this. In one of the cases observed, the prosecutor, who does have rights to speak to the press 'in the interests of justice', released the details of a named suspect but in general there is little thought given to using or working with the media. In one case observed almost in its entirely during the research, officers spent six months attempting to identify a dead body when a placed article in a local newspaper would have led to a much earlier identification from one of the victim's large network of friends and acquaintances.

The reconstitution Á witness testimony
Finally, the reconstitution brings together all the witnesses to the crime event that the investigation has unearthed. Of all the sources of information that Osterburg and Ward (1997) identify to assist in 'Reconstructing the Past', French justice relies heavily on the 'témoignage', or witness statements, of human subjects, with forensic criminalistics playing a secondary role. 16 Interviews with witnesses are referred to as 'auditions' Á it might appear that a witness is given the opportunity to be 'heard' ('entendu') rather than restricted to the basic details of what will be affirmed in a 'statement'. As state subjects witnesses, in rhetoric at least, are given a role in the truth revelation process which will heal society.

The reconstitution, confrontation and truth
The reconstitution is the last and largest of a series of 'confrontations' between witness' accounts which take place in the Juge's office throughout the instruction and occasionally in the police offices before Á I witnessed a particularly harrowing confrontation between a prostitute and a man accused of rape which took place at the dead of night in the suburban offices of a city police force. While the atmosphere in the Juge's office is far less intimidating than the busy police office, the brutal element of confrontation, having one's account put to the test, is a constant feature as the following excerpt from a reconstitution shows. The Witness (W) is trying to excuse his departure following a fight in a restaurant in which a man died: W: 'But my wife was pregnant'. J: 'There had been a fight and you just leave?' W: 'I didn't know he was dead'. J: 'You knew he had been badly attacked, hit with a piece of wood and stabbed. Why did you leave?' It is far more difficult to remain silent, or repeat no comment in such a direct attack and indeed this witness eventually admitted he had been told to leave by his boss. It is this process of direct confrontation between accounts and the reasoned questioning of the Juge which is deemed to result in the unveiling of the truth. There is much talk of the 'truth', 'la manifestation de la vérité' in France during formal interview but also in observed conversation. Although some of this use, particularly on paper, is rhetorical and formulaic, nevertheless, many detectives spoke to me of la vérité, and with prosecutors and Juges this tendency was even more apparent. This is in stark contrast to English detectives who almost as often spoke of lying Á 'everybody lies Á the suspect, the witnesses' (E149). This is not to suggest that English witnesses and suspects necessarily do lie more than their French counterparts or that the epistemological basis is sounder in France but that for confident French investigators 'the truth' is something findable.

The contents of the dossier
The end of stage four (the investigation by the Juge) marks the end of the pre-trial phase of French criminal justice. This ends the major involvement of the police and Juge in the case Á it is not even certain that the police will be called on to give evidence at the eventual hearing. Again this is a stark contract to the situation in England where a large proportion of a murder team's work is spent in preparing the case for court and in attending lengthy court sittings. A number of researchers have remarked on the importance of the pre-trial phase compared to the trial itself in the French process. Leigh and Zedner (1992, p. 16), for example, argue that, due to the multipartite input into the investigation Á involving defence and victim Á the issues are all forecast pre-trial (see also Greilsamer and Schneidermann (2002, p. 16) and Hodgson (2005, p. 116): the reverse of the situation in adversarial systems such as England's. Along with this goes a parallel weight attached to the dossier, the written record of the pre-trial phase (McKillop 1997, p. 71). This written record of the investigation controls the subsequent phases of the process. A large proportion of the written record in cases would be the file produced by the police (of around 500 pages in many of researched murder cases). According to the myth of the primacy of the written word in the French criminal procedure, every aspect of the investigation would appear in the dossier, in order for the truth to be revealed to the reader. 'French law is a written law, only the written is valued. In investigation everything thought is done, everything done is written, you write up everything', 17 Balland quotes an investigator as saying (2003, p. 107). Another insists: 'A policeman's job is primarily a writer's'. 18 It is at the time of arrest when the file is being put together which sees the officers completing almost the longest shifts they spend on cases, working throughout the night.
The importance of the file in the psyche of the French detective is suggested in some of the words they use. The English 'case' is referred to as an 'affaire' in France, especially when it is unsolved. Although 'affaire' has little of the frivolous overtones of the English homonym, its use connotes a wider view of the circumstance and context of the event of murder compared to the legalistic 'case' or colloquial 'job' used in England. Once a suspect is found, the affaire is then referred to by the word 'procédure' or 'procès' referring to the file and process Á and eventual trial Á which the suspect undergoes; the officer who puts the paperwork together is a senior figure 'la procédurier'. Witness statements are written down in a 'procès-verbal' Á literally 'verbal item towards the process'. 'Exhibits' are not referred to by their function within the eventual trial as in the English system, but by what happens to them within the process Á they are 'sealed' Á 'scellé'. The statements and reports in the file are prefaced by lines of archaic legal language in the first person plural (Diaz et al. 1994, p. 167). 19 This presents the document in an acceptable form for the procès. One of the main reasons for evidence not to be accepted in court in the French system is if they are in an incorrect form. The former gendarme Sesmat (2006) describes how failure to amend a requisition to a handwriting expert in l'affaire Grégory 20 who could link the original accused to the anonymous letters that had been circulating among the victim's family prior to the child's murder meant that her subsequent report was not included in the dossier.
But Sesmat (2006, p. 200) adds in his recapitulation of the Grégory case, that no dossier Á 'a fixed and formal image' 21 could capture the full scale and movement of an investigation. He criticises the police officers who later took over his case for not coming to speak to the original investigators; instead relying on gleaning all they needed to know from the dossier. For Hodgson (2005, p . 161n), the dossier is a record of events that simplifies reality, eliminates doubt and avoids future complications and Mouhanna (2001, p. 122) makes the same point in his study, saying there are many things that are difficult to insert formally into a dossier, not least because witnesses are often aware that an eventual suspect will have access to all documents through their lawyer. Officers told me that they sometimes delayed documents going into the dossier knowing that the defendant would have access to it. In many of the researched cases, there was no record made in the dossier of activities I had seen; especially meetings with informal contacts.
The printed statements (of witnesses or suspects) also do not reflect the interview process in its entirety. Unlike an English interview, which is tape-recorded and can be transcribed word-for-word later, the French interview process proceeds in two-time (Levy 1987, p. 81). The question is asked, a reply received, and then the two are repeated aloud Á inevitably in paraphrase Á before being typed. 22 This procedure was speeded up to some extent during an interrogation when two officers would conduct the interview; one asking the questions, the other typing. Once the suspect started to confess, the defendant's words were written up as a monologue. This is in part procedural because legally the French police are not allowed to continue to question someone who has confessed. It also gives a more voluntary feel to the confession.
Nevertheless, the French dossier is a more complete and tidy record of the progress of the case than is available in most English cases. The use of the work procès throughout the investigation and trial stages reflects that the work of the police is seen as part of a whole which includes the work of the justice department.

Section 2 Á French culture
French working life Á the team As other writers, such as Van Geirt (1995) have commented, a French detective officer's life is centred around the seven-member groupe he or she is part of Á 'it's like a little family, but more than family because you get all sorts as family, it is working with really close friends', an officer said to me (F805). The teams observed were even known by the surname of the chef de groupe Á les Vignerons 23 Á like blood relatives. Each member of the team observed, when speaking about what was good about their work, mentioned their group and the easy relationships between them. They bonded in team activity outside work Á often taking their long lunch breaks 24 together Á having a meal, playing sports or going to the cinema. They said their chef de groupe fostered a 'festive spirit' within the group, arranged social outings for his group after particularly long stretches of work; they spent a week together on holiday at his cottage in the country 25 and supported his quite frequent appearances on the stage as a member of a band made up of other police detectives Á whose posters appeared on the wall of the office. The group were often teasing each other, shouting comments from room to office Á the dossier was thin because the victim was a woman, the chef de groupe did not do much interviewing because he was lazy. In another case, the chef de groupe dressed up in a wig borrowed from the victim's wardrobe at the scene, telling me part of his role was to maintain the ambience. Although the group did have times of jollity, they were also often to be found having serious discussions and showing genuine interest in each other's lives.
While group family was important, an officer's real family life was also considered. The long lunch break in France allows for workers to return home for lunch, as one of the officers in the group observed regularly did. The school summer holidays in July and August Á les grandes vacances Á are reflected to a greater extent in work patterns in France than in England. In common with many French workers, half of the Brigade officers took July off work, the other half August. When I asked what would happen if a number of murders occurred with limited staff, an officer quipped: 'There won't be Á everybody else is on holiday too'. 26 French detectives tended to minimise the deleterious effect of the work on their lives. While they did talk about the long hours causing some friction at home, they also spoke about the positive consequences of undertaking murder enquiries. 'It allows me to put life in general into perspective. I have friends who think they have problems, but which on a grand scale are not that important and I can see that', said one (F805). Even mundane tasks that Wilson's (1978) detectives might have called 'garbage work' like house-to-house, the enquête de voisinage, is somehow translated by the French into an art, requiring the personality and verve of 'seduction' (Diaz et al. 1994, p. 161).
One detective said: 'I like my work enormously. Of course some tasks are more interesting than others 27 but in general it works well' (F811). 'I'm paid for satisfying my own curiosity'. Most spoke of the inherent interest of the work, and used words like 'enjoyment' and even 'laugh' when describing how they viewed their jobs. Part of this satisfaction could be connected to the high regard in which the Brigade Criminelles are held by professional colleagues. One young officer said that the Brigade officers were known as 'les seigneurs' ('stars', 'heavyweights') Á 'some of my colleagues who work in other units are very jealous of me, they say I don't work in the real world like them'. (F807). Magistrats told me they thought the Brigade was 'performant', a word which can be translated as 'efficient' but which I had only heard previously in the context of high performance cars. Paradoxically, in some ways detectives in France enjoy less esteem from the general public Á perhaps because of the low profile they keep, they are not routinely sought as experts by the media for example, and often appear as shady figures in drama (the television series La Crim or the film 36 for example). 28 The basic contentment, the officers felt could also be linked to the manner in which the detectives appraised their own work, less as a comparison against an ideal image of investigation and more compared to other public service appointments. When talking about long hours, one officer said: 'We expect to work different hours from other civil servants' and when I asked others if they liked their work they said 'Of course, otherwise we would do something else'. 'If you don't like being on call, you could go and join another unit'. Although criticism was made of some of the physical deficiencies in their work life Á in terms of equipment, cars, mobile telephones and computer accessories 29 Á the attitude of the officers was that they still managed well Á 'On arrive bien quand meme'.

Working life Á women
The impression from reading the literature was that the world of police in France was very much a male world and that what women there were regarded as in some ways second rate for aspects of the detective trade, treated in a subordinate fashion. While it is true that women are not well represented within the French Police Nationale as a whole Á figures from the Ministère de l'Intérieur put the proportion at just over 13% (in 2005), 30 the dual entry appointment system operating in the French police means the level of women in management positions is actually higher Á at around 16%. I was proudly informed that there was at least one woman in most teams at the Brigade researched. While it initially seemed to me that the woman officer in the team was assigned some of the more routine tasks, she was also the most junior and inexperienced in terms of service history. However, there were no female Chefs de groupe at the Brigade at the time of my research and at least one chef had refused to have women in his team saying it would be distracting for the men. 31 In interview, male officers stated baldly that women could not intervene successfully in arrest situations and would be unable to exert the necessary man-to-man pressure to confess in the interrogation as if these were indisputable facts and not opinion. One of the women officers said to me: I have never had any problems but there are those who think women have nothing to do with the police. It is true that there are things I can't do, I have less power than my male colleagues but there is little need for that in the murder team, and there are some things women are better at. We are different, we work from different sides of our brains, but it is ridiculous to suggest we can't do interviews. If a man MEX 32 doesn't like women, it could be an advantage or a disadvantage to have a woman there.

Working life Á coffee
The work ambience of the French detective can be seen to be symbolised by the taking of the 'petit café du matin' Á a ritual whose communicative potential was demonstrated by Zauberman's (1997) work on Gendarme detectives when it was banned by new management. French detective groups observed Á whether in the Brigades, the Police Nationale in local police stations or the Gendarmes in the SRs Á had some provision for coffee making and the first thing an officer would do on arrival at work was put on the coffee or be offered coffee Á even in the very early days of new cases. There would be some kind of space, even in small offices, for taking coffee. Within the Brigade, one group had a sofa, two breakfast bars, another a corner table, all had a coffee jug or machine. Conversation at coffee could be about domestic issues Á taking a daughter to school before work, traffic; it could be about resources, seizing the opportunity of the head of the Brigade dropping by to talk about needing more mobile telephones. It could be about ongoing or past cases. There could be shouting, teasing, laughter, quiet talk. On occasions, an officer would bring in croissants, biscuits or fruit. The ability to take coffee was part of the culture of the work. One chef de groupe was felt to be 'a bit lacklustre' as a manager Á he often took a back seat, going off on his own to pursue searches leaving his second-incommand to organise the rest of the team. He was also said to 'only go through the motions of coffee'Á it was a quick and quiet affair with not much talk or spirit Á a 'dry ritual' (F274). He found it difficult to involve others in the decision-making process and did not encourage discussion. By contrast, the team mainly observed 'discuss all day, at le petit café du matin and after'. Their coffee break was a popular affair often attended by many others, quickly creating a crowd and standing room only. The taking of 'le petit café du matin' represents the joie de vivre of the French officer in his life and work and the place that work takes in his life. By contrast, the nearest equivalent to social drinking that occurred in my English fieldwork was the traditional after trial celebration in the pub. Conversation here though was very little about the intricacies of a case or philosophical discussion, especially as the night wore on, and the event often excluded those who did not drink alcohol or had family commitments.

Conclusion
The image of the French criminal justice system is one of an inquisitorial ideal with state employees seeking and finding truth. In interview, French detectives and other justice personnel believed in the possibilities of finding that truth through a detailed and complex investigation and detectives' commonality with suspects and witnesses who want to be persuaded to tell the truth to them and be heard. The detectives were supported in their work ethic by the team spirit engendered in the engineered small teams and convivial atmosphere in which the importance of a life away from work is not forgotten. French detectives saw the Anglo-Saxon system as mechanistic and bureaucratic; their investigations as limited and curtailed in a trial-ready vacuum; their detectives defensive. The French people would be protected from some of the excesses of overseas investigators by the overseeing role of the indomitable Juge d'instruction.
However, the observed practice painted a rather more nuanced picture. In practice, the reasoning process at work in the French system was less inquisitorial truth seeking and more akin to one of inference, very similar to what Atkinson (1978) many years ago imputed to coroners in assigning a suicide verdict. Atkinson argued that coroners carry out what in America are dubbed 'psychological autopsies' based on the evidence of the dead person's life history, their psychiatric record and what they had said to other people. This sounds very like what goes into a French murder file which is perhaps not surprising in that it is the coroner's court which is the closest to inquisitorial in the court system in England but nevertheless suggests at the least a mixed system operating in France. There are patently aspects of the adversarial not only creeping into French procedure via legislative reform under European edict (the increasing presence of lawyers in police/suspect interviews) but forming the basis of traditional French set pieces. The confrontation of individual narratives at the centre of an adversarial trial occurs several times during the French pre-trial procedure Á from confrontations between suspect and detective and confrontations between individuals in the police station and Juge's office to the wholesale confrontation structure of the reconstruction.
There are as well contradictions between the French inclusive world view with the French treatment of two key groups in the homicide investigation process Á the surviving family of the victims and the media. While the families of victims are accorded a codified legal role they can in practice despite (but possibly because of) their legal status be treated in a rather cursory way in person. French and English police and judicial colleagues also had a markedly different relationship with the various branches of the media. French detectives did not maintain the constant contact with members of the fourth estate that their English counterparts did and 'media relations' were never discussed. While to some extent this was because of the different nature of the media in France, which does not have the large numbers of tabloid titles eager to fill their pages with crime news that Britain has, the lack of communication with the press by official sources had both positive and negative effects. It left a lacuna for the media which was sometimes filled by defence lawyers talking about their side of the investigation. This led to media coverage that was to some extent less tied to a possible sensationalist crime angle and more focused on the human issues relating to the defence such as mitigation and background material. It also meant that some investigative opportunities were lost by the French and this is indicative of the rather old-fashioned tools of investigation still relied on. CCTV, for example, is almost non-existent in France.
In turn, the English officers regarded some of the procedures of the French police, particularly as regards the interview and the type of records made of it, as akin to what they themselves used to do in the past (pre-PACE) and thus, retrograde, though not denying that their own methods sometimes left them frustrated Á 'on occasions to the point of nausea when they say nothing and are not even considering saying anything'. Again, French attitudes towards women in the police workforce also look outdated.
Finally, there did not appear to be the detailed supervision and direction of the police enquiry by either Prosecutor or Juge d'instruction supposed by many outside observers. In my observation, this was because of the supposed expertise of the specialised squad but the situation does seem to reflect the general picture presented by other empirical commentators. While the Juge did retain narrative control she/he was dependent on police for content. In fact, in some ways the English police's relationship to the Counsel who represented 'their' case in court often seemed to me to be more akin to the supervisory and directing role expected of a Juge.
Working at its best the French system of murder investigation has the potential to dig deeper and, with its dual 'authorship', to prevent some of the miscarriages of justice that have plagued British justice. It is difficult to see, for example, how those connected with forensic evidence, which is requisitioned and examined by the Juge d'instruction in France, could have occurred there. However, France has not been free from miscarriages of justice of its own, linked to its own weak links Á the primacy of confession and witness testimony and the dependence on the integrity of the Juge Á who works alone even when inexperienced. Patrick Dils spent 15 years in prison for the murder of two young boys after repeating confessions both to police and Juge when still a teenager himself. More recently l'Affaire Outreau saw new Juge Fabrice Burgaud remand 18 people in custody for up to three years falsely accused of child sexual abuse reportedly on the basis of an overreliance on the witness testimony of one woman (Dawent and Huercano-Hidalgo 2005). 25. To have a holiday home like this has not been in the past such a luxury as it would be here as prices have been quite reasonable. 26. Followed by the often repeated comment 'We'd cope'. However, during the heat wave that took place in July and August 2003. The French medical profession, depleted by numbers taking their annual holiday, struggled to cope as hundreds died in the cities. 27. Such as the hours spent examining the telephone records of victims, witnesses and suspects. 28. This state of affairs is complained of by some of the older officers who told me it used to be different, but now it is the Juge d'instruction whose names are known. 29. There is no HOLMES and few searchable databases in France, they have only recently upgraded from typewriters to computers. 30. Compared to 26.2% in England and Wales http://www.homeoffice.gov.uk/publications/ science-research-statistics/research-statistics/police-research/hosb1311/hosb1311?view0 Binary. 31. Shortly afterwards the last team also took on a woman Á 'a man disguised as a woman, in order to make herself fit in she has had to make herself as man-like as possible', one of the other women officers commented (F808). For similar adaptational strategies among English police officers, see Holdaway and Parker (1998). 32. Arrestee.