• 2 Rousseaux (1997).

1In a recent survey Xavier Rousseaux points to several ways forward in historical research on criminal justice2. Although it deals with medieval and early modern times, his essay provides an entry into current theoretical debate about modern criminal justice. It is more or less conventional wisdom that changes to criminal law and policing, greater state involvement in prosecution and adjudication and a revolution in state punishment allowed for an unprecedented degree of interference in people's everyday lives. A transformation of the socio-economic order which intensified with industrialisation and rapid urbanisation is inextricably linked with these reforms. However, Rousseaux’ essay stresses that they are also connected with a set of other factors.

• 3 Knafla, Binnie (1995, pp.8-13).

2One of these is the process of state formation. State involvement in criminal law and criminal justice reached new heights for many continental European countries by the close of the French Revolution and is recognised as a central feature of the rise of the modern state. Another factor is the development of « rational-bureaucratic » forms of administrative control. Judicial administrations were structured according to specific models, notably of a legal (e.g., inquisitorial vs. accusatorial regimes) and political variety (e. g., a centralised or decentralised model of government). The cultural representation of criminal justice is a third factor. The notion that law and justice convey powerful messages which may serve ideological functions has been carried further by neo-marxist historians, cultural theorists and legal anthropologists who have developed theories about its role in constructing collective identities3.

• 4 The expression « imagined community », originally reserved to refer to forms of solidarity based on (...)
• 5 Emsley (1996) ; Levy (1996) ; the history of modern criminal justice in the Netherlands remains und (...)

3This article seeks to show how theorising about criminal justice in a comparative framework can help in asking questions about the nature of the modern liberal state, the modernity of administration and the creation of « imagined communities » in the late 19^th century4. Our study builds on empirical research on the prosecution and court system at base and medium level. These were aspects of law enforcement that, like policing, were revolutionised during the 19^th century. Research in this domain, let alone comparative research, remains to be further developed, especially for the latter decades which heralded a new phase in criminal justice reform in France, England and the Netherlands5. Our investigation is largely based on three case studies of criminal justice in a provincial urban setting. Although historians tend to insist on the individuality of each town, the comparative approach serves a useful purpose in that allowing us to more clearly define those peculiarities that say something significant about criminal justice in France, England or the Netherlands as a whole. Ipswich, the setting of the English case study, was a middle-sized provincial town (50 to 69000 inhabitants), situated in an agricultural county. The town had a sizeable manufacturing industry but large numbers remained employed in the traditional trades. Caen (40 to 45000 inhabitants), also the centre of a rich agricultural region, was very much a town of commerce and artisanal enterprise. Finally, Maastricht (30 to 36000 inhabitants) was situated in a predominantly agrarian province (Limburg) in the Catholic South of the Netherlands but was dominated by manufacturing industry.

• 6 A notable example is King (1984).

5The extension of the scope of criminal law was conditional on the co-operation of public institutions that were not state bureaucracies (e. g., a municipal police force), as well as that of important sections of the population. The participatory nature of criminal law has been a classic focus in much writing on early modern England6. The literature on the English tradition of decentralised legal administration and the continental pattern of centralised, bureaucratic law enforcement is vast and well known. The latter part of the equation will therefore take first place. The considerable influence which members of the public continued to have on the successive stages of law enforcement in modern times is particularly under-researched in French and Dutch historiography. We will limit ourselves to two major types of middle-ranking offences.

• 7 Davis (1985, ch.5).
• 8 A prosecuted case has been taken to mean every single prosecuted charge of larceny, embezzlement or (...)
• 9 This study is based on witnesses’ depositions in cases tried by the magistrates’ court (under the S (...)
• 10 Our material is drawn from dossiers de procédure for the years 1879, 894, 902 and 1910, checked aga (...)
• 11 By larceny we have understood both the misdemeanour (délit) of common larceny (vol simple) and the (...)
• 12 Included are cases of common larceny (eenvoudige diefstal), larceny as an employee (diefstal in die (...)
• 13 None of the surviving archives of prosecutorial offices in the Netherlands seem to have preserved p (...)

6Research by Jennifer Davis has found that in later 19^th-century London private individuals instead of the police were responsible for reporting the great majority of minor property crime, as well as for detecting most offenders7. Our findings generally confirm this picture. They are based on an analysis of cases of larceny, embezzlement, and receiving stolen goods, committed in the towns of Ipswich, Caen and Maastricht and prosecuted in the local courts between 1880 and the mid-1900s8. As for Ipswich Table 1 gives a breakdown of all cases for four years, in total 297. Under legislation passed between 1847 and 1879, an average of 85 % of those prosecuted for one of these offences was tried at the lowest level, in the local magistrates’ court9. The case study of Caen is based on a compilation of samples over four years which cover three-quarters of all prosecuted cases10. Included in this total are 351 cases of larceny (vol), embezzlement (abus de confiance par salarié) and receiving stolen goods (recel) processed by the Correctional Court of Caen11. Similarly, the Dutch case study is based on five samples from the Arrondissemental Court of Maastricht12. Although our samples contain over 80 % of all prosecuted offences, their size is comparatively small (131)13.

• 14 See categories 7 and 8 of Table 1.
• 15 See categories 16 to 19 of Table 1 ; on police skills and novel detection techniques see Emsley (19 (...)

8Further in line with Davis’ work our evidence also highlights the key role of private individuals in detecting thieves and embezzlers. Listed in the table are various strategies most likely to lead to arrest and prosecution. Regarding Ipswich this table reveals that in approximately 54 % of cases studied the perpetrator's identity was either known to the victim from the outset or elicited following suspicions raised by the victim or following a description from the latter14. The sample from Caen shows an average of 70 %. Again, Maastricht's score is the lowest with an average of 44 %. Only a small number of the cases in Ipswich and Caen (15 and 12 % respectively) was successfully solved by the police through some sort of special expertise15. The percentage for Maastricht turns out to be relatively high (27 %). The peculiarity of the Dutch case throws up a series of questions addressed further on. For our purposes it is important to observe that the « typical » theft or embezzlement that reached the authorities in Ipswich or Caen was a « victim-reported » crime committed by someone who was either known to or suspected by the victim from the outset.

• 16 The English case study is based on the summary offence of common assault ; the French sample includ (...)
• 17 The analysis is based on compilations of greater numbers of cases than those analysed for Table 1 ; (...)

9Perceived as a lesser threat to social order, minor violent crime was even more likely to be policed by members of the public. Generally speaking, our analysis of assault cases heard by the Ipswich magistrates’ court, the Correctional Court of Caen and the Arrondissemental Court of Maastricht corroborates existing work on London. Similarly, our findings confirm the suggestion that the victim behind most prosecutions was a working person16. This stands in direct contrast with common property crime : in approximately eight out of every ten prosecutions the court in each town acted on behalf of members of the local middle-class groups17. This last conclusion contains few surprises.

• 18 Hay, Snyder (1989a, pp. 34-39, 42-45) ; Emsley (1987, pp. 147-150) ; Taylor (1998, pp. 577-580) ; S (...)

10The following stages of law enforcement remain to be analysed in much greater detail, partly because the policing of public order offences has been central to English historiography of this period. The point here is to scrutinise prosecution arrangements with respect to medium-range crime in Ipswich. Only the generalities of prosecution are well known. There was no formal system of state prosecution in late 19^th century England, so far as the vast majority of crimes was concerned. Though the expense of prosecuting felonies (e.g., larceny) and certain misdemeanours was largely paid from public funds, the victim or someone acting on his/her behalf was still legally responsible for initiating (i.e. lodging a complaint with a magistrate and possibly applying for a summons) and forwarding most prosecutions. Research suggests that without acquiring new legal powers either to prosecute or to compel a party to do so, the local police were increasingly managing prosecutions for felonies, largely due to their extensive powers of arrest. In brief, a policeman was obliged by law to arrest anyone whom he saw in the act of committing a felony –in which case the police themselves were authorised to prosecute–, or any person positively charged with a felony by a victim or a third party. The latter scenario appears to have been much more common : the victim or another person called an officer to have the suspect arrested and the officer brought the former before a magistrate to lodge a complaint in order to start proceedings. Thus, what was legally a private prosecution amounted in practice to a semi-public prosecution18.

• 19 Virtually all criminal proceedings for indictable offences began in the magistrates’ court ; follow (...)

11Criminal records reveal little about prosecution practices. Nevertheless, the notion that victims of property crime tended to rely on the police lending a hand in prosecution, is confirmed by two pieces of evidence. One is that on average only 8 % of minor property offenders tried by Ipswich magistrates were discharged. This is surprisingly low given prosecution was rarely preceded by a distinct, formal preliminary hearing (designed to consider prima facie evidence and possibly a referral to a higher trial court). We will see just how stark the contrast is with prosecutions for assault19. The low discharge rate as well as the high proportion of prosecutions for minor property crime brought by or on behalf of middle-class people (in contrast with assault prosecutions) can only be fully explained if one takes on board the new function of the police as « clearing-house » for most felony prosecutions. Basically, the police tended to apply a great measure of discretion before backing a private individual to take the matter further.

• 20 It is estimated that on average only five percent of all prosecutions for minor property crime were (...)
• 21 SRO, IBR, C8/8 Papers, Box 47 [Jan. 882] and S[uffolk] C[hronicle], 19 Nov. 881, Suppl. p. 2 (Th. W (...)
• 22 SC, 18 Sept. 1880, Suppl. p. 2 (H. Sergeant) ; it was common for the police to inform complainants (...)
• 23 The ease with which a suspected thief could be arrested and taken into custody on flimsy evidence i (...)

12Any suggestion that victims were per definition bent on outright prosecution, let alone conviction, has been dispelled by research that has emphasised the diversity of public attitudes towards crime and justice. The important question is to what degree victims were effectively in charge of initiating prosecution given the involvement of the police. Instances where the aggrieved party's formal consent was required for prosecution (ie. the procedure of lodging a complaint with a magistrate) constitute the overwhelming majority of cases and form the focus of our study20. One conclusion is that when called to arrest a suspect, the police tended to exercise pressure on the victim depending on the suspect's identity (and especially on this person's criminal record) as well as the specific nature of the crime. In one case, a chemist had a 13 year-old paper-boy arrested for shoplifting. After a night in police custody the chemist let the youth off, provided he never enter his shop again. However, proceedings followed because the police, the prosecuting shopkeeper claimed during cross-examination, wanted him to. It is likely they had been motivated by the fact that the offender already had several convictions to his name21. In a similar case, the landlord of a public house who had caught a 12 year old attempting to steal money, handed him over to a constable « with the intention of frightening him ». Again, he ended up prosecuting the boy, in this instance for attempted larceny. It is likely that the police encouraged him to do so on account of the youth's record as « one of the worst young criminals they had in Ipswich »22. Clearly, the other side of the coin is that it was possible to have someone arrested, cautioned and even undergo a spell of detention in a police cell, without the matter going any further23.

• 24 SC, 20 March 1897, p. 6 (E.A. Keys) ; for an example of the first category see SRO, IBR, C8/8 Paper (...)

13The Ipswich police also seem to have applied considerable pressure if they had been involved in tracing the suspect or had been near the scene of the crime. In a case of the second type, a builder found himself prosecuting a slater, employed by another builder, for stealing 24 slates. At the time of the incident, the builder explained to the court, he did not think that the person charged had any felonious intent by taking some slates. He claimed that he « would not have instituted proceedings had not the constable been with him »24.

• 25 As explained in SC, 9 March 1889, Suppl. p. 3 (E. Daniels).
• 26 See SC, 14 April 1883, Suppl. p. 3 (F. Allen), SC, 14 April 1883, p. 6 (J.T. Clements), SC, 3 Nov. (...)
• 27 SRO, BB 15/2/1/5&6 ; the absence of relevant depositions makes any interpretation a hazardous enter (...)

14Once a complaint for a felonious offence had been lodged with a magistrate, there was less room for manipulating the criminal process. Legally, magistrates could not consent to a withdrawal from prosecution. In practice, however, a prosecution would drop if the private prosecutor offered no evidence, provided the police did not apply pressure or prosecution was not pursued under a different guise (e. g., if the police decided to bring another victim in the case before the court)25. The overall impression is that compromise was more of an option if the police were not involved in prosecution ; if the offence was relatively minor and the accused without criminal record ; and in the case of a dispute between a parent and child26. We have also systematically analysed all charges of larceny and embezzlement recorded in the court registers of 1886 –our analysis shows that a mere 6 % were withdrawn27. One might speculate that the finding that so few withdrew from prosecution at this point supports the notion that a majority of proceedings were initiated with the real consent of the victim. Clearly, we need further proof to substantiate this assessment. We believe that analyses of trial proceedings can also provide some clues. First, however, let us briefly consider prosecution practices in the field of minor violent crime.

• 28 The police were only entitled to arrest an offender if they had witnessed the act or had seen obvio (...)
• 29 My estimate is based on prosecutions for assault, as well as aggravated assault on women and childr (...)
• 30 For an explicit reference to magisterial custom see SC, 5 April 1884, Suppl. p.2 (F. Playford and C (...)
• 31 King (1996, pp. 46-47, 72) does agree that much more research is required ; see also Landau (1999).

15There is nothing very new about the finding that most prosecutions for common assault were « private »in the true sense of the word. Because police powers of arrest were limited in the case of misdemeanours, most proceedings tended to be initiated without police interference (i. e. in addition to lodging a complaint with the court, applying for a summons which was within the reach of nearly all)28. Excluded from our analysis are assault-related disputes that ended in a prosecution for a public order offence. The notion that prosecution was relatively easy and accessible is corroborated by evidence suggesting that rates of cases thrown out of court were considerably higher than in the case of small property crime29. Further evidence lies in the fact that there was relatively more room for manipulating criminal procedure in its further stages. Various private prosecutors were only interested in intimidating the accused with a court appearance or having him or her cautioned by the court. What is novel about our analysis is that evidence suggests that, even during this period, the local magistrates were prepared to allow prosecutors to compromise in nearly all common assault cases. It was customary for the court to comply with such a request, provided both parties agreed to a withdrawal and the accused paid the cost of the summons. In jurisprudence this remained a matter of dispute30. I would take issue with research that suggests that the judicial tendency to treat assault cases as a private concern was radically transformed after the late 18th century. Unfortunately, the vast majority of cases, dealt with by the local JP, has not yet been subjected to detailed analysis31.

• 32 On official thinking on the symbolical qualities of the jury system see, e. g., Emsley (1987, p. 16 (...)
• 33 See Steven's Directory of Ipswich, 1894 and various other local directories ; from 1893 the Ipswich (...)
• 34 Radzinowicz and Hood (1986, pp. 618-624) ; Emsley (1987, pp. 154-162).

16This brings us to the question of private individuals’ input (victims and third parties) in trials. Magistrates’ courts occupied a central position among trial courts in late 19th-century English society. The large shift of medium-range criminal business from higher courts to the lowest court was a shift away from trial by jury – still advocated as a cornerstone of the constitution which supposedly inculcated a belief in « British justice » – as well as, in most cases, from decision-making by a legal professional.32 As in most other English towns, the lowest criminal court in Ipswich was in the hands of unpaid, untrained and locally selected JPs. They were men of local standing, Liberals as well as Conservatives33. Proposals to introduce a professional local magistracy, appointed, salaried and promoted by central government, were partly defused due to a persistent belief in a tradition of involving local elites in law enforcement34.

• 35 Some reference to solicitations of private prosecutors and third parties is made by Radzinowicz and (...)
• 36 Despite the growing complexity of the law of evidence, magistrates’ courts were known to adopt a co (...)

17Unfortunately, there is insufficient evidence to establish with any certainty whether trial proceedings for minor property crime were actually conducted by the private prosecutor or by the police acting on the latter's behalf. Was there anything of real substance to the formal part of the private prosecutor in trial procedure ? The real significance of this procedural feature can be read, in as far as magistrates cared to motivate their judgement and the local press recorded the hearing, from those instances where a prosecutor interceded on the accused's behalf and the sentence was tempered in view of the prosecutor's action. Although this phenomenon has been noted by others, it has scarcely been seriously analysed for this period35. The same goes for the way in which trial procedure allowed third parties to intervene. Magistrates were generally known for their flexibility in applying the rules of evidence36. We find that both a leniency plea and a submission of evidence testifying to the accused's good character were devices which, depending on the social status of the person intervening (e. g. an employer, a clergyman, a landlord) could substantially affect sentencing. In the case of property crime the most likely beneficiaries were minor, first offenders.

• 37 SC, 31 Aug. 1895, p. 6 (F. Krauter).
• 38 SC, 20 April 1889, Suppl. p. 2 (Ch. Read).
• 39 SC, 18 Nov. 1893, p. 6 (H. Stacey) ; SC, 21 Oct. 1893, p.7 (R. Gair) ; SC, 21 Sept. 1895, p.3 (J. . (...)

18In a typical case, magistrates decided to fine a waiter convicted of embezzling a small sum instead of jailing him in view of the defendant being a first offender, the prosecutor's desire not to press the charge unduly, in addition to character statements received « from several places »37. Even if the bench favoured an exemplary punishment, an intercession by the private prosecutor on the accused's behalf might still be a weighty consideration. To give one example, sentencing a van foreman to one month imprisonment for a minor theft from his employers, the court stated that « the sentence would have been more severe but for the company's suggestion of a light sentence »38. On various occasions a person who enjoyed some social status, particularly an employer, came forward to testify to the accused's good character or statements or letters to that effect were submitted to the bench39. The suggestion that victims, particularly in the case of minor, casual offences committed by first offenders, were given some leeway can also be used to improve our understanding of prosecution practices.

• 40 SC, 28 Jan. 1888, p. 6 (R. Snell) ; for other examples see SC, 15 May 1886, Suppl. p. 2 (A. Lockwoo (...)
• 41 Based on an extensive reading of newspaper coverage of domestic violence cases brought before the I (...)

19Finally, the evidence regarding minor violent crime shows unsurprisingly that the say of private prosecutors was more far-reaching. Usually the court was prepared to meet a prosecutor's request of leniency by recording a fine. In a few instances victims gave to understand that they preferred the assailant to be bound over to keep the peace – a well-established sanction in English law – rather than seeing him/her punished. What transpires is that this strategy was most likely to be allowed in assault cases related to a family dispute. In one such instance, the father of a labourer prosecuting his son intimated that he did not want to press the charge, but « all he wanted was to get him out of his house, for when he was at home he was always upsetting it ». The court decided to bind over the defendant on the understanding that he left the house40. Evidence further suggests that this sort of arrangement was always allowed in spouse abuse cases, provided they did not constitute an aggravated assault41. For a fuller understanding it is extremely useful to place these findings in a comparative perspective.

• 42 Martin (1990, pp. 137-142) ; Schnapper (1991, pp. 377-378).
• 43 The exception are a series of offences prosecuted by other administrative bodies.
• 44 Half of all private prosecutions were for slander, mostly relating to press offences see Schnapper (...)

20Following the radical departure from a long-standing tradition under the Constituent Assembly, the Napoleonic régime returned to a strong system of State prosecution42. The state prosecutor's office was responsible for initiating and forwarding most publicly funded criminal prosecutions43. Although private citizens were legally entitled to bring private proceedings in the Correctional Court under the supervision of the Public Prosecutor (Procureur de la République), provided they bore the cost, in practice people generally relied on state prosecutors taking action.44 That being the case, short of lodging a complaint with the police or directly with the prosecutor's office (parquet), the victim bore no formal responsibility in the criminal process, except when it came to certain singular crimes such as adultery. In other words, the decision to bring a case into a trial court, which in this period was more often made by the Public Prosecutor than by judges sitting in private, was not conditional on the victim registering a formal complaint with either of these agencies.

• 45 It has been observed that when drawing up an official report (procès-verbal) Commissioners and Insp (...)

21However, prosecution practices on the ground in Caen were much more complex. First of all, a word of caution about the relationship between police and parquet is important. Our evidence has confirmed the observation made by several historians that in spite of the stipulations of French criminal procedure, Police Commissioners and Inspectors exercised a measure of discretion in referring reported misdemeanours and felonies to the parquet45.

• 46 The police were the main agency in urban areas responsible for reporting offences to the prosecutor (...)
• 47 ADC, U 2593, 6 Sept. 1902 (F. Vallet) ; the parquet archive provides our primary source material.
• 48 Dossiers of prosecuted cases (dossiers de procedure) contain a statement concerning or reference to (...)
• 49 See, e. g., ADC, U 3228, 16 Aug. 1879 (A. Coupey) ; ADC, U 3083, 25 Jan. 1894 (V. Lemonnier) ; ADC, (...)

22Again, for reasons outlined before, our focus is on small property cases where it was the victim who informed the authorities46. Although legally immaterial, the victim's consent seems to have been, under certain circumstances, a significant consideration even beyond this stage. For example, in one instance the Public Prosecutor decided not to initiate proceedings against a maid charged with stealing 40 fr. from her employer, considering, according to the primary sources, the complaint had been retracted and financial restitution had been offered47. Something more about this practice can be inferred from evidence on correctional sentencing in Caen. The reason is that there appears to have been an inverse correlation, at least in the 1900s, which was a period when a suspended sentence (sursis) was an appreciable penalty in dealing with (deserving) first offenders, between its application and the parquet's willingness to take on board the complainant's opinion about prosecution48. Therefore, judging from the evidence on suspended sentencing, we would suggest that the weight of an intervention by the owner hinged, among other things, on the suspect being a first offender and having offered restitution or compensation. We have also analysed a mixed bag of cases prosecuted despite the retraction of the complaint. This tends to show, rather unsurprisingly, that another major precondition was the small value of the goods appropriated49.

• 50 See, e. g., ADC, U 2538, 31 May 1883 (D. Blanchard) ; ADC, U 2541, 16 June 1884 (P.Doux) ; ADC, U 2 (...)
• 51 See, e. g., ADC, U 2540, 7 Jan. 1885 (H. Saudin) ; ADC, U 2593, 6 Sept. 1902 (F. Vallet) ; ADC, U 4 (...)
• 52 Sometimes in the form of an IOU ; see, e. g., ADC, U 3267, 11 Dec. 1902 (Lebourgeois) ; ADC, 2U/119 (...)

23There is also plenty of evidence from the parquet records suggesting that in minor embezzlement cases proceedings were only initiated with the consent of the aggrieved party. Occasionally an employer put a case of embezzlement in the hands of the Public Prosecutor on the express understanding that no proceedings be started if the money embezzled was returned within a set number of days50. The records also include various examples of theft committed by a domestic servant or charwoman where proceedings were stopped due to a retraction of the complaint and restitution51. The notion that embezzlement and employee theft constituted a distinct category is corroborated by evidence suggesting that this type of consideration was acted upon by the police. The overall impression is that rather than having to register a complaint with a senior police official and seeing it referred to the parquet, if the matter was to be taken any further, employers brought employees to the police station only for the police to supervise and sanction some kind of arrangement between the parties52.

• 53 (My emphasis) ; ADC, U 2541, 3 June 1884 (A. Manchon and V. Tourmente).
• 54 Cases of wounding involving the use of a weapon appear to have been prosecuted irrespective of a vi (...)

24It would be even more misleading to assert that complainants in cases of assault had no say once the parquet had been informed. In one instance, describing their visit to the police station, the two parties involved reportedly said that « [il] nous a été répondu que c'était notre affaire, Mr. le commissaire ajoutant qu'il fallait que nous le prévenions si nous ne voulions pas donner suite au procès »53. Based on the available evidence, it is our contention that the public prosecution was prepared to halt the criminal process if the relevant complaint was withdrawn, provided the violence inflicted was « sans gravité ». This term seems to have covered all but serious wounding54.

• 55 Levy, Rousseaux (1991, pp. 117-122) ; Martin (1990, pp.157-162) ; Badinter (1989).
• 56 That is, cases where the victim brought a civil suit for damages concurrently with the criminal pro (...)
• 57 [Original spelling] ADC, U 3435, 19 June 1902 (Lance) ; for other such instances see ADC, U 3299, 3 (...)
• 58 The input of complainants in this matter is suggested particularly by ADC, U 2636, 30 Nov. 1889 (Ca (...)
• 59 See Hélie (1920, pp. 548-549) ; Cambuzat (1879, p. 502) ; some examples from Caen are ADC, U 2542, (...)

25Moving on from prosecution arrangements to court hearings, the most striking difference with the English case is that private citizens, including men of local standing, were formally virtually excluded. It is a commonplace that the French Revolution, with the exception of a period of radical experimentation, and particularly the Napoleonic régime furthered the subordination of the judiciary to the power of the state.55 More important for the purposes of analysis is the fact that correctional cases were conducted by the public prosecution. Victims could only formally participate as anything other than a witness in a capacity as civil litigant (partie civile) bringing simultaneously an action for monetary damages against the accused. Extremely few were able to do so and the only known instances were assault proceedings56. To be sure, this does not mean that the complainant's opinion made no impact at all on sentencing. This was marked in assault cases related to a family dispute. In one such instance, having reported her 21 year-old son, a woman day labourer called on the Public Prosecutor « de lui donai la loi Béranger quil ne recommencera pas ». In the event, the court granted her wish for a suspended sentence (loi-Bérenger)57. Furthermore, there is evidence suggesting that the victim's views could affect the reduction of a charge of assault to a minor violation and could thus indirectly influence the severity of the sentence58. What is implied is that for an accurate comparison with the English case, we need to include the offence of violences légères (a scuffle) which was tried by the tribunal de simple police (Police Court), the lowest criminal court. Contrary to the letter of the law, this minor violation covered a huge range of acts of aggression. For instance, it was common practice for the public prosecution to redefine an assault case as one of violences légères59. In the final analysis, however, the qualitative difference with the English case lies in the fact that the victim's opinion was not formally, nor systematically, nor publicly articulated, nor were trial judges allowed as much room to accommodate private interests. To further our analysis we will now present the Dutch case which displayed many similarities with the French case.

• 60 Note that the region of Limburg, including Maastricht, was a special case in that it had been integ (...)
• 61 van Boven (1990) ; Bosch (1992, pp. 19-23) ; on the prosecution of tax offences see Wortel (1991, p (...)

26One year after the country's annexation by the French Empire in 1810, the Napoleonic legal codes and administration were introduced in the Netherlands60. The new state prosecution bureaucracy replaced a polycentric, republican system of public prosecutors who represented the interests of a local or provincial authority. Despite alterations in 1838 and 1886, the essentials of the French system were kept in place. A significant change was that private citizens relied by law exclusively on public prosecutors, that is, with the exception of certain crimes (e. g., tax offences) prosecuted by other government departments. The impact of annexation on law, the court system and the judiciary was equally revolutionary and long-lasting. An indication of the growing distance between the two legal systems was that following legislation in 1877 regarding Cantonal Judges, those at the bottom of the judicial hierarchy, all Dutch judges were legally educated61.

• 62 See, e. g., Faber (1993, pp. 13, 16, 9) in Maastricht a proportion of criminal complaints were refe (...)
• 63 In 1890 16 prosecutions were instituted out of a case-load of 22 complaints of theft, embezzlement, (...)
• 64 Police case books have been preserved for the early 1880s and the 1900s for an example see GAM, Arc (...)

27In view of the legal and institutional similarities in policing, prosecution and trial, we have decided to concentrate on two features distinguishing the Dutch from the French case. Firstly, as with French historiography, Dutch writing has now incorporated the idea that the police which was the main agency reporting offences to the Public Prosecutor (Officier van Justitie), displayed a degree of discretion in dealing with crimes that reached them62. What we would add is that its scale was much larger in Maastricht than in Caen. One of the pieces of evidence is that more than eight out of every ten cases of minor property crime sent on by the municipal police to the prosecutor's office ended in a prosecution, which calls for an explanation along the lines suggested63. Further proof can be deduced from a preserved set of police case books that include reported thefts that were never recorded on an official form (proces-verbaal), let alone sent on to the prosecutor's office64. On a day-to-day basis, substantial powers were delegated from the judicial authorities to the municipal police.

• 65 The analysis is based on court registers and covers all minor property offences committed in the to (...)
• 66 Patterns of judicial repression of petty public order offences are discussed in the second part of (...)
• 67 For a personal view of the « true » amount of larcenies and burglaries in Maastricht see Notulen va (...)
• 68 In the French case, by comparison, on average only 27 % of prosecutions fall into these categories  (...)
• 69 Both French and Dutch state prosecutors of this period were repeatedly urged by their superiors to (...)

28Second, it is crucial to analyse major variations in prosecution rates for minor property crime. During this period rates were on average three times lower in Maastricht than in Caen : 8.7 per 10, 000 inhabitants, which is broadly in line with a national trend, compared with Caen's 23.765. This leads us into an apparent paradox : how can we explain the coexistence in the Dutch case of comparatively low prosecution rates for minor property crime with a model of local policing (e. g., powers of arrest, professionalism, patrolling and surveillance) and prosecution levels for violent crime and for petty public order offences that were similar to Caen's66 ? A plausible, partial explanation may lie in differences in the « true » incidence of certain kinds of property crime, yet we would like to develop a supplementary hypothesis, building on figures presented earlier67. Judging from Table 1, an above average proportion of prosecutions for minor property crime in the Dutch case were for offences discovered by the police. In addition, a study of detection techniques reveals that a peculiarly high proportion of prosecutions (53 %) concerned cases where the suspect had been traced either by the police or using a technique that implied their involvement68. In short, in dealing with a « typical » theft or embezzlement (a « victim-reported » crime committed by someone who was either known to or suspected by the (middle-class) victim from the outset) the police and judicial authorities in Maastricht appear to have been more inclined to consider prosecution « inexpedient »69. The question that arises next is why this should have been so. At the same time, our findings will be placed in a wider, interpretative framework.

• 70 Lévy, Rousseaux (1991, pp. 114-124).
• 71 Rosanvallon (1990, pp. 97-109) ; Schafer (1997, pp. 1-13).
• 72 On political conformity within the judiciary and judicial handling of political crime see, e. g., R (...)

29Over the past decade the integrative role of law and justice has become a major research topic. It is clear that, historically, the rise of the nation-state was accompanied by the expansion of « state law » and royal justice. Criminal justice has increasingly been interpreted as a mechanism that served to reinforce the power of the sovereign in the eyes of the population70. The picture that emerges is that partly due to the country's major political crises and early démocratisation, governments particularly in Third Republic France used law and judicial action as critical instruments to constrain regionalisms and the influence of traditional local élites and assert the moral authority of the nation-state71. It is obvious that judicial centralisation in France and the limited discretionary powers of judges were about much more than the set of factors usually mentioned (e. g., usefulness in dealing with civil disorder ; a concerted process of unifying law and legal administration ; ideologies of popular sovereignty and parliamentary supremacy)72.

• 73 Lévy, Rousseaux (1991, p. 124).

30Historians are now much more aware that centralisation was a drawn-out process, extending well after the 19^th century73. Drawing on our own case study, we believe due attention should be given to two points. First, the municipal police, which was in the first place a local government institution, was arguably one of the major constituent agencies of criminal justice. Secondly, to dispel generalisations about French criminal justice more emphasis also needs to be placed on the major role of private citizens in law enforcement, instead of concentrating on the legal powers of public officials. In particular, our study proposes that individuals rather than the police were responsible for channelling the majority of minor property and violent crime to the authorities, as well as for detecting most suspects in these areas. Although victims rarely bore legal responsibility further in the criminal process, our contention is that in practice there was a degree of flexibility which allowed complainants to express their opinion as to whether proceedings should be initiated and to substantially influence the decision, provided the offence in question was relatively minor and the guilty party a first offender. We have suggested that in the case of assault there is no need for such a qualification. As far as trial proceedings are concerned, although complainants were formally virtually excluded and the court was given very little leeway to take on board their opinion, to claim that the former had no say at all would be incorrect.

• 74 See, e. g., de Bruin (s. d.) ; te Velde (1994, pp. 49-52) ; Dudink (1994, pp. 132-135) ; for a comp (...)
• 75 Faber, van Ruller (1997).

31To assess the Dutch case let us start by identifying the process of state formation. The accepted view is that the period 1798-1830 was a decisive accelerating phase (e. g., major internal transformations resulting from the Batavian Revolution (1795) and French intervention). However, partly because major political divisions only emerged after 1870 and the franchise was (only) extended in phases after 1887, a major impetus for more intensive state intervention was largely absent for much of the 19^th century74. Historians have started to substantiate the notion that judicial centralisation lagged behind the French pattern75. Our own observation is twofold. First, we need to theorise the extent to which even in this period legal powers regarding medium-range crime were delegated, on a day-to-day basis, from judicial officials to police officers. Secondly and more interestingly, we would suggest that the nature of the Dutch state carries a considerable burden of explanation for variations in prosecution levels for common property crime that existed between Maastricht and Caen. Faced with a « typical » theft or embezzlement (reported by or on behalf of a middle-class person) the judicial authorities and senior police officials in Maastricht were, it is argued, less inclined to detracting from existing informal sanctioning or regulation procedures, thereby affirming the power of state sanctions and the state at large, so long as these informal techniques reflected the local community's hierarchical character of (e. g., restitution or compensation, whether or not sanctioned by the police).

• 76 Thane (1990, pp. 1-2, 35-36, 47).

32The English political classes, for their part, were extremely apprehensive about greater state intervention and about extending the « bureaucratic system »76. With our case study of small property and violent crime in turn-of-the-century Ipswich we have tried to stress two points. We have mainly emphasised that private individuals continued to play a considerable formal and practical role in running the system. Notably, in the majority of cases the victim was still formally responsible for initiating proceedings ; furthermore, the impression is that police pressure was scarcely brought to bear on potential prosecutors in the case of minor, casual property crime committed by first offenders and in most cases of assault. In addition, the significance, in comparative terms, of the formal participation of the private prosecutor in a trial lies in the fact that he/she was able to substantially influence sentencing decisions, even if proceedings for property crime were often effectively conducted by the police. Interestingly, trial procedure even allowed third parties to voice their opinion about the preferred judicial outcome.

• 77 Conley (1991, pp. 6, 16, 41-42, 203-204).
• 78 Harris (1993, pp. 10-11, 213-214) ; McKibbin (1990, pp. 24, 38) ; Johnson (1993, p. 166) ; an exemp (...)

33The political argument against centralisation was reinforced by a certain ideology of justice, recently again articulated by Carolyn Conley. She argued that criminal courts, using their discretionary powers, tended to administer justice in a way that was in keeping with the local community's values and priorities (e. g., social reputation, common interests) and suggested that officials took care that they were being seen to do so77. Unfortunately, what is not stressed is that the image of a « community-based » system of criminal justice is at odds with that commonly described by historians of Victorian Britain. In line with British sociological thinking about the criminal law and E. . Thompson's and Douglas Hay's analyses of 18th-century English criminal law, historians have tended to argue in favour of an ideology of fair and correct « British justice » (adherence to abstract, legal principles observance of complex rules of procedure), which is seen as a major factor in generating popular support for « law » as a basic form78.

• 79 Davis (1984) ; Emsley (1987, pp. 154-162) ; Conley (1991, pp. 12, 19) ; Radzinowicz, Hood (1986, p. (...)
• 80 In contrast with the rules of ceremony and degree of solemnity of English higher courts and the Cae (...)
• 81 Report R. . Divorce, vol., q. 7643 (testimony of Theodore W. Fry, stipendiary magistrate of Middles (...)
• 82 The Ipswich magistrates described justice as « the quintessence of common sense » (Ipswich Corporat (...)

34However, the judicial style of the Ipswich magistrates’ court was significantly different from that in higher criminal courts, as well as from that in the Caen and Maastricht courts discussed so far. Again, the comparative perspective provides some useful insights. Historians would agree that magistrates’ or police courts which were largely unsupervised, were peculiar institutions in that they applied unwritten principles external to the case on a wide scale and dispensed summary justice with little concern for procedural niceties79. Arguably, magistrates were not overly concerned about presenting « justice » as being associated with the application of the law (the substantive and formal criminal law), which is also suggested by the court's notorious lack of judicial ritual80. Contemporaries noted that ordinary people « [did] not look upon [the police court] as a legal place so much »81. Our proposition is that a weightier ideology was articulated around the notion of justice as being based on a consensus within the local community. Two analytically different aspects of this ideological agenda have to be separated out. One, well-known side of the argument refers to the accommodation of popular or commonsensical ideas of crime and justice (vide Conley)82. This exercise could be genuine but could just as well be designed merely to give the impression that justice was done in accordance with popular notions. The other side of the argument, to do with the role of private individuals at successive stages of law enforcement, has remained under-developed. A sound analysis of this theme depends on integrating our findings with a study of the lowest tier of criminal justice.

• 83 The 17^th and 18^th centuries saw the first moves towards increased control over « consensual » offen (...)
• 84 See, e. g., Emsley (1991, p. 70) ; Bernard (1987, pp. 128, 132) ; Manneke (1993, pp. 7, 92).

35There is now an extensive body of research describing the way the criminal law in the mid-19^th to the turn of the century was extensively moving beyond protecting property and person and suppressing long-standing public order offences such as vagrancy into areas of life where there was least value consensus. Activities such as drinking, street trading, gambling and prostitution became the objects of a disciplinary enterprise which bore to a very large degree on working-class culture83. This aspect of criminal justice squares with revisionist historiography in so far as the law was no longer primarily enforced by the courts but by, amongst others, the new, reorganised police, meting out their own « justice »84. The argument generally goes that they were instrumental in law enforcement given the extremely wide scope for legitimate arrest, their use of techniques of cautioning and move-on as well as physical force, and the deployment of preventative strategies. Clearly, this thesis applies most to one range of « consensual » offences, namely those against public order and public morality (e.g., public drunkenness).

• 85 Radzinowicz, Hood (1986, p. 119).
• 86 See, e. g., Wiener (1990, pp. 257, 264).
• 87 See, e. g., Schnapper (1991, pp. 379-380) ; Wortel (1991, p. 30) ; Bosch (1992, pp. 25-26) ; Arthur (...)

36The danger is that one fails to analyse and theorise about the part played by criminal courts in enforcing what French and Dutch law defined as minor violations (contraventions ; overtredingen) and English legal historians as summary offences of a « non-criminal nature »85. There are historians who have noted a process of « administratisation » of the lowest tier of criminal justice, starting in the mid- to late 19^th century86. Implied is that the low scale of punishment, combined with a range of other factors, contributed to the rise of a criminal justice machine that was routine, impersonal and bureaucratic. This approach is part of the debate about the « modernity » of judicial administration, particularly about the rationalisation of judicial procedure. The best-known aspect of this process is the generalisation, in France and the Low Countries, of an inquisitorial criminal procedure, whilst England maintained its long-standing tradition of accusatorial criminal justice. The growing delegation of law-enforcement functions from law courts to the police and new administrative bodies is another well-known facet of procedural rationalisation87.

• 88 For instance, an outfitter was tried and convicted of public drunkenness only one day after his arr (...)
• 89 The latter draws upon observations about metropolitan police courts, see Gamon (1907, p. 97).
• 90 In France proceedings were generally initiated with a simple warning notice instead of a summons wh (...)
• 91 According to A.J. Rethaan Macaré, Public Prosecutor of Haarlem, in H[andelingen der] N[ederlandse] (...)
• 92 No Police Court registers have been preserved of Caen, but those of Elbeuf, a Norman textile town o (...)

38Assessing and comparing the pace of justice is a difficult task. However, starting with a study of criminal procedure, there is no doubt that the pace of the English variant was extremely rapid. Under summary jurisdiction procedure cases could be directly brought into court to be tried. Those arrested were either released at the police station on their own recognizances to appear in the court or kept in custody –in case of a disturbance of public order- provided they were brought before the court within 24 hours88. If no arrest had been made, the police would lay an information before a magistrate and the accused would either come into court voluntarily or be summoned to appear, generally about a week later89. In Maastricht and Caen minor violations were always deferred to the public prosecution which was responsible for summoning the accused90. However, to understand why the time span between the reception of a charge by the judicial authorities and court proceedings was comparatively greater in these two places, two other factors have to be considered. One was that the relevant courts sat only once a week, the other refers to a tendency of prosecutor's offices to « hoard up » cases that were similar in nature. It is estimated that in the Netherlands the average length of time was one month91. The French evidence is as yet too impressionistic to make any strong assertions92.

• 93 With regard to the former see, e. g., Wiener (1990, pp. 263-264) ; Vidal (1903, p. 318) ; van Hamel (...)
• 94 Hélie (1920, p. 203) ; van der Feltz (1896, p. 437).
• 95 van der Feltz (1896, p. 437) ; Cnopius (1896, p. 270) ; A.J. Rethaan Macaré in HNJV, 1895, 26, pp. (...)

39Secondly, moving on to trial procedure, rapidity was a common feature. In a wide range of « consensual » offences, proof of criminal intent, a basic legal principle, was less important, occasionally even immaterial and police evidence alone constituted sufficient proof93. Significant for our further argument is the idea that procedure was generally simplified, be it to varying degrees. Magistrates dispensed with much of the procedural complexities of higher courts, whereas trial procedure in the Cantonal Court and in the Police Court was, by comparison, nominally fairly sophisticated. One of these rules, that of rendering motivated decisions, could force judges to reserve judgement until the following hearing94. This is known to have been common in the Netherlands95. Although this requirement tended to lengthen the criminal process, the pre-judgement stages of Dutch and French petty trials were surprisingly swift regardless of their relatively elaborate rules. We will develop this point in the next section.

• 96 On France see Perrot, Robert (1989, p. clii) ; report presented by M. Du Mouceau (Public Prosecutor (...)
• 97 We have analysed all convictions for petty public order offences (public drunkenness, drunk and dis (...)
• 98 Radzinowicz, Hood (1986, pp. 649-651) ; see also Howard Association Report, 1896, pp. 10-11 ; Howar (...)
• 99 By law the distress procedure which was compulsory in case of a summary offence of a non-criminal n (...)
• 100 van der Does de Willebois (1891, pp. 90-91) ; Cnopius (1896, p. 269) ; van der Feltz (1896, p. 440) (...)
• 101 Reports presented by R. Demogue, M. Dubois, M. Du Mouceau, and J.-A. Roux to the Internat. Penitent (...)

40As for the final stage of the criminal process, the enforcement of a sentence, our emphasis lies on enforcement procedures for fines. French and Dutch figures suggest that on average nine in ten of those convicted of a minor violation were fined96. For the English case we cannot rely on aggregate figures but research points in the same direction97. Historians have observed that English magistrates tended to be unwilling to allow those fined either an extension of payment or payment by instalments98. Legally, they were bound to first levy a distress on the goods of a fine defaulter in the case of a petty public order offence before resorting to a custodial alternative. A legal loophole, however, allowed them to send fine defaulters to prison immediately after a judgement, a practice that appears to have been routine in Ipswich99. The French and Dutch enforcement procedures were, by comparison, time-consuming. Dutch law allowed all defendants two months’ time to pay. Given few paid and most were as a result ordered to be imprisoned, court hearing and punishment were usually separated by a period of at least three months100. French defendants, for their part, with the exception of juveniles, were legally not entitled to an extension of payment, nor to payment by instalments. However, due to the amount of red tape to do with enforcing sentences, which involved the Public Prosecutor's and the tax collector's offices, fine defaulters could be committed to prison usually only several weeks after the trial. Again, most of those convicted by Police Courts ended up as fine defaulters101. To sum up this first section, differentials in terms of criminal procedure, the pronouncement of a judgement, and the enforcement of a sentence combined to make « low justice » decidedly fast-track in Ipswich when compared to the two continental cases.

• 102 On popular attendance and press coverage see, e. g., Davis (1984, pp. 316-317) ; Behlmer (1994, pp. (...)
• 103 On the presentation of evidence in the Dutch and French cases see Hélie (1920, pp. 201, 204, 212-21 (...)

41In direct connection with these differences it is essential to scrutinise the public character of court proceedings. To start with, magistrates court hearings of cases involving « consensual » offences were per definition more likely to be subject to a fair amount of public and press attention given the institution's central role in processing medium-range crime102. A second factor that affected the public character of court proceedings refers to the balance between its oral and written components. Let us merely point out here that evidence was given orally in Ipswich, whereas in Caen and Maastricht this was also submitted in written form, the official report (procès-verbal ; procès-verbaal), the significance of which becomes clear in the light of a short survey of the use of trial by default103.

• 104 Even in very petty cases magistrates were entitled to issue an arrest warrant ; in one such instanc (...)
• 105 To be allowed time to pay an accused had to be present and put his/her case – often without success (...)
• 106 Reference could be made to instances of juveniles prosecuted for throwing stones in which parents s (...)

42The majority of those prosecuted in Ipswich for a « consensual » offence were present at the hearing. As has been mentioned before, some were brought to court in custody. Provided the offence was punishable only by a fine, magistrates could try the remainder by default, unless they issued a new summons or a warrant104. However, we would argue that those summoned or released on their own recognizances had an interest in attending. For one thing, the procedures for enforcing fines and for arranging an alternative method of payment acted as incentives105. Secondly, from a comparative perspective one might speculate that there was a greater sense among the accused that what they would say in their defence, would receive the court's attention106.

• 107 The published judicial statistics aggregate this type of information the estimates are drawn from t (...)
• 108 Pujo (1894, p. 142) ; [Kahn] (1921, p. 239).
• 109 van der Does de Willebois (1891, p. 97) Centraal Bureau voor de Statistiek (1986, p. 21).
• 110 See, e. g., Faure (1882, p. 321).
• 111 van der Does de Willebois (1891, pp. 82, 89) ; van der Feltz (1896, p. 436).
• 112 Briet (1915, p. 51) ; on popular attendance at Cantonal Court proceedings in the 1890s see van der (...)
• 113 Pujo (1894, p. 141-142) Dalloz (1894, v^eprocedure criminelle, p. 467).

43Levels of trial by default were much higher in Caen and Maastricht. Judging from the available evidence, which is restricted to the latter years of this period, the average proportion of Police Court cases heard by default within the arrondissement of Caen as a whole was one in three – in the specific cases of public drunkenness and disorderly behaviour more than two in five107. And attendance levels at urban Police Courts are known to have been generally lower than at rural Police Courts108. The Dutch case, for its part, is more conclusive and more marked. Nationally, trial by default rates rose consistently from 60 to over 85 % over this period109. The fact that none of the accused were brought to court in custody and the nature of fine enforcement procedures are part of the explanation. In addition, we would propose that in the eyes of the accused it was very difficult, if not impossible, to set up a defence110. The Dutch evidence is unequivocal about the impact of such levels of trial by default on the conduct and public nature of court hearings. They generally consisted of little more than the calling of the accused's name and the application of a prescribed penalty. The official report was often not even read over, which merely served to convey the futility of putting one's case as a defendant111. This subversion of trial procedure inevitably contributed to the lack of public attention which Cantonal Court hearings attracted. Public attendance was virtually nil by the start of the First World War112. The impression is that French Police Courts, for their part, were just as swift in hearing default cases, the effect of which on trial procedure, its oral character and, ultimately, on popular attendance and public attention is likely to have followed a similar pattern113.

• 114 Table 2.
• 115 Judicial Statistics, 1880-1905 (PP 1881-PP 1907) ; after 1898 these public order offences were clas (...)

45In view of the extensive jurisdiction of English magistrates published statistical data on overall prosecution rates cannot be used for comparative purposes. However, it is noteworthy that in this period on an annual basis on average one person for every 90 of the population of Ipswich was prosecuted in the magistrates’ court114. A reliable basis of comparison is provided by the combined rates of prosecution for public drunkenness, drunk and disorderly behaviour and breach of the peace. Annually on average one inhabitant of Ipswich for every 380 was prosecuted for one of these offences which were among the most significant petty public order offences115.

• 116 Minor violations represented 70 % of all court cases ; based on national figures (covering the quin (...)
• 117 Santucci (1986) ; Chatelard (1981) ; Désert (1981, pp. 237-238, 308-309, 311) relies on published c (...)
• 118 Table 3 ; the ressort of the Tribunal de Simple Police of Caen, the cantons Caen-Ouest and Caen-Est (...)
• 119 The correspondence between these two charges and the English offences of drunkenness, drunk and dis (...)
• 120 The ratio of prosecutions to population (minor violations) in the city of Rouen was 1 to 35 in 1893 (...)
• 121 Based on unpublished aggregate statistics covering the arrondissement of Caen the supposition is th (...)

46By 1880 most criminal cases processed by French courts, were dealt with by Juges de paix116. Nevertheless, French case studies usually fail to properly scrutinise the latter's vast criminal business or, if they do, rely exclusively on aggregate data117. Within the jurisdiction of the Caen Police Court each year saw on average one prosecution for a minor violation for every 31 inhabitants. Measured in terms of persons prosecuted, the ratio to population is likely to be a few points lower118. A valid comparison with the English case can be made on the basis of prosecution levels for public drunkenness (ivresse publique) and for disorderly behaviour (bruit et tapages injurieux ou nocturnes)119. Again, we have to rely on limited archival documentation. Evidence from Rouen, where overall prosecution rates for minor violations in proportion to the population were roughly similar to Caen's, shows an average ratio of 1 :62 for the mid– 1890s120. Another, less reliable, piece of evidence, from Caen this time, suggests a ratio of about 1 :110 for the mid– to late 1900s121. In sum, prosecution for these types of offence appears to have been up to four times more likely in Caen than in Ipswich.

• 122 Following the introduction of the new Penal Code prosecutions for minor violations represented over (...)
• 123 Table 4 the jurisdiction of the Cantonal Court of Maastricht, the kanton of Maastricht, included th (...)
• 124 The charges of public drunkenness (dronkenschap op de openbare weg) and disorderly behaviour (nacht (...)

47The Dutch case, finally, is in line with the French case122. By 1880 the Maastricht Cantonal Court tried on average one minor violation for every 123 inhabitants within its jurisdiction, which would rise to one for every 65 between 1882 and the mid-1890s and one for every 30 from 1895, largely due to a sustained effort to implement the 1881 law on public drunkenness. Again, rates of persons prosecuted are likely to be a few points lower123. Between 1882 and the mid-1890s on average one case of public drunkenness or disorderly behaviour was heard for every 71 inhabitants. From 1895 the ratio rose to 1 :53124. Prosecution for either of these offences was, we suggest, about six times more likely in Maastricht during the period studied than in Ipswich.

• 125 In addition, more research is needed to determine the comparative costs of prosecution and the fina (...)
• 126 Petro (1994, . 217).
• 127 Ipswich Corporation, Proceedings of the Town Council, 1902-1903, pp. 69-70 ; SRO, DB 32/12, Watch C (...)

48When trying to explain these variations, it is right to insist on the impact of certain procedural differences125. To start with, the burden carried by the Ipswich police of having to go to court to give evidence in the case of a wide range of « consensual » offences acted as a disincentive to report and prosecute such offences126. Secondly, whereas minor violations could be prosecuted on police evidence alone, in Ipswich the burden of proof was greater for certain categories of « consensual » offences (e. g., the offence of obstruction of footpaths and highways)127. At the same time, procedural variations reflected differing cultural assumptions about the precise function of law courts in this area of criminal justice. This is the subject of the final section which draws together the three strands of our discussion.

• 128 Gamon (1907, pp. vii-viii) ; see also Behlmer (1994, p. 232).
• 129 Gamon (1907, p. xiii) ; some thirty years later another observer of magistrates’ courts again empha (...)

49This comparative survey demonstrates that prosecution for « consensual » crime was much less regular in Ipswich than in either Caen or Maastricht. In addition, we have tried to show in the case of Ipswich that « low justice », from its first until its last stages, was fast-track and that court hearings at this level were public events. Against this background and given the concentration of this amount of judicial business in one institution, it is easier to grasp why, in the words of one observer, Canon Barnett, a social reformer, the magistrates’ court functioned as « the visible representative of ‘justice’ » and « the very centre of observation and information »128. He was convinced that « [p]eople [...] largely owe to the police-court their conception of justice »129. What is clear is that the judicial processing of « consensual » crime only served to strengthen the view that criminal justice had little to do with the application of abstract, legal principles and sophisticated rules of procedure.

• 130 van Hamel (1884, p. 271) ; see also Bool (1887, p. 121) ; Cnopius (1896, p. 269).
• 131 ADC, U 2542, 2 Sept. 1886 (M. Chaignard).
• 132 Taylor (1998, pp. 588-589) emphasises how prosecution levels and prison sentences with respect to « (...)

50The notion of « administratisation » has proved to be especially useful in describing the Dutch and French cases. Striking is that probably due to the slower pace of the criminal process and the waning of the public character of court hearings, « the deterrent and cautionary effect [of the court] ha[d] been lost entirely », to use the words of a Dutch observer who went on to assert that « people know a neighbour has been charged, but perhaps they will never learn about the outcome »130. In a typical case from Caen, having lodged a complaint against a neighbour, a working-class woman inquired with the parquet how her case stood, only to discover that her case had already been followed up and that the offender had been convicted by the Police Court131. The limited moral educative role of this court and the Maastricht Cantonal Court is likely to have served to reinforce existing prosecution levels which were strikingly high in comparative terms. It adds a critical comparative note to the thesis put forward, amongst others, by Howard Taylor that the English criminal justice machine was increasingly guided by routine and by impersonal, bureaucratic considerations132.

• 133 Trial proceedings begin with the hearing of testimony and presentation of evidence and conclude wit (...)

51We would also like to put forward the hypothesis that « administratisation » of « low justice » was particularly advanced in the Dutch case, which is consistent with two important episodes in Dutch legal history. One was the rejection by the Dutch political class of a jury system, which was perceived as a symbol of distrust of the magistrature. The other was the late introduction of fully public trial proceedings in 1848. Both innovations had been introduced in the Netherlands by the French in 1811 but were discontinued in 1813133.

• 134 van der Does de Willebois (1891, p. 82) ; van der Feltz (1896, p. 436).
• 135 For such proposals see van Hamel (1884, pp. 298-299) ; Briët (1915, p. 51) ; [Kahn] (1921, pp. 241- (...)
• 136 These points draw upon McBarnett (1981, pp. 143-153, 166-167) ; this leads to a refinement of E. . (...)

52A further reason why « administratisation » was more in evidence in Maastricht and Caen refers to the subversion of dominant notions of procedural fairness. According to Dutch legal voices Cantonal Court hearings left a distinct impression on the public of a sham or a public sale, which made it virtually impossible to keep up the appearance of « justice »134. Proposals were debated in both countries at the end of this period either to simplify trial procedure or to direct more criminal business to administrative bodies ; one impetus behind such suggestions was a desire to better uphold prevailing ideals of formal legal justice135. It is at least arguable that further « administratisation » of « low justice » may have been seen as useful in that it allowed higher criminal courts alone to feed into the public image of criminal justice. By contrast, given the openness of English summary procedure and the flexibility of English criminal law, perceptions of English magisterial justice were never quite so directly at odds with the image of justice conveyed by the English higher criminal courts136. At this point we need to integrate these insights with our findings regarding the middle tier of criminal justice.

• 137 Their pre-eminence is assumed by writers such as McKibbin (1990, pp. 24, 38).
• 138 E. g., SC, 17 Oct. 1891, p. 6 ; EADT, 14 June 1904, p. 5.
• 139 See, e. g., Smith (1982, pp. 225-247) ; Morris (1990, pp. 318-331) ; Hills (1988).

53In the end, a few preliminary conclusions can be drawn from our comparative analysis of the English case study. We suggest it has put in proper perspective the notion of « community-based » justice which seems to have represented a more significant ideal than the presumedly pre-eminent ideology of fairness and adherence to the rules137. They embodied two competing tendencies in official thinking about justice. It was not only, as has so often been claimed, a matter of accommodating the values of the « respectable » sectors of the local community. We have argued that it was just as much about the ways in which private individuals participated or were seen to be participating in law enforcement. As far as « middle justice » is concerned, further research is needed to corroborate our hypothesis that for the majority of routine cases of property and violent crime known to the Ipswich police and magistrates around the turn of the century private individuals, both victims and third parties, played a considerable role in policing, prosecution and adjudication. There is a rather misleading overemphasis in current research on the role of police and magistrates. Equally important, in the case of « low justice », a neglected area of research, we have stressed how in comparative terms the court was much more subject to public scrutiny, which fits in with the English tradition of an accusatorial judicial procedure for deciding medium-range crime. Crucially, our argument has been that law-enforcement agents in Ipswich attempted to realise the ideal of justice as being administered with broad approval from the local community, in order for it be convincing. In this light, our study proposes that more attention should be devoted to the significance of local criminal justice institutions as a source of integration among a town's « decent and respectable citizens », which included many better-off working men138. Until now social historians have tended to concentrate on various other public bodies, cultural enterprises and associational life as foci of social cohesion139.

• 140 On the cultural meaning of the spectacle of courtroom proceedings in mid-19^th century France see Fi (...)
• 141 See, e. g. Tombs (1996, pp. 99-101, 04) ; Burdeau (1989, pp. 188, 98, 33-334) ; Rosanvallon (1990, (...)

54Our case study of turn-of-the-century Caen clearly demonstrates the limitations of the common view of the French criminal justice machinery as being « typical »(in a Weberian sense) of a highly autonomous state bureaucracy. To start with, we have re-emphasised the role of the municipal police in the early stages of law enforcement but in particular we have reviewed evidence substantiating the notion that private citizens were actively involved in policing much medium-range crime and in initiating its prosecution. The second part of the article, then, explored the theme of rationalisation in judicial administration. Given the fundamental difference between inquisitorial and accusatorial regimes, perhaps it is not surprising that we have observed that a process of « administratisation » was at work in the area of « low justice » both in Caen and in Maastricht. Interestingly, however, the French Police Court appears to have been more open to outside scrutiny than its Dutch equivalent. This also opens a perspective from which to examine the nature of correctional and Assizes trials140. Although several general studies have taken on board this alternative view of French state institutions, particularly in the context of the Third Republic, research is still too narrowly focused on the influence exerted by local elites rather than on, what our case study has tried to analyse, various, more complex ways in which groups further down the social scale affected the day-to-day working of administrative bodies141.

• 142 Unfortunately, Faber (1983, pp. 270-279) does not raise any comparative questions.

55At the same time, we have further developed the notion that the criminal justice system served a much more significant ideological function for the state in Caen than in Maastricht. In particular, the hypothesis has been put forward that this can help explain variations in prosecution levels for minor property crime between the two towns which were very similar in terms of its system of policing and judicial disapproval of violent crime and public disorder. Clearly, more detailed local research is required. Another fascinating feature of the Dutch case is that in comparative terms judicial agents seem to have been less concerned about symbolic displays. A major factor is that modern Dutch society never experienced a massive crisis of public confidence in the judiciary or other sections of state officialdom. The political climate was such that instead of public scrutiny the procedures and rules alone were seen as what guided and legitimated the actions of those running the system. More comparative research will have to establish the strength of the early modern roots of judicial rationalisation142. Our approach may perhaps be fruitful in investigating the procedures and administrative practices developed by other Dutch bureaucracies of this period.