Methodological notes regarding sources
Records of the justices in the district of Montreal
For the proceedings of the justices outside of the formal courts, both in their ministerial function and in their summary jurisdiction, my main serial source was the sessions papers preserved by the clerk of the peace. As far as criminal matters are concerned, these contained two types of documents. First were those produced by justices throughout the district concerning complaints that were destined for trial in the Quarter Sessions, especially depositions and recognizances for the appearance of the parties. And second there were at least part of the documents produced by urban justices in cases brought before them for summary resolution, especially those where the prosecutor asked for the defendant to be bound to keep the peace but also including some depositions from cases involving prostitution, vagrancy, and breach of service. Distinguishing between the two sorts of documents was not always easy, since only a portion of cases destined for Quarter Sessions actually appeared in the registers of the court. Of the others, some were clearly either Quarter Sessions or summary cases from the nature of the offence (as with master/servant cases, always tried summarily, or brothel cases, always tried in sessions), or from accompanying documents, where indictments or recognizances to appear at the next sessions indicated Quarter Sessions cases and recognizances to keep the peace indicated summary cases. This left a small group mainly composed of cases involving the various shades of assault. Among these, I assumed that any cases in which the prosecutor simply asked for security to keep the peace were summary cases; all other assault and battery cases were destined in theory for the Quarter Sessions; and all lesser assault and threats for summary resolution, since this was the general practice followed throughout the period.
The sessions papers are an especially rich source, since most of the case-files include the original deposition describing the complaint, the social status of the parties, and much other precious information. However, they suffer from a number of lacunae. In the first place, there are almost no documents from the period before 1785 for the district of Montreal, so that there is less information on the operation of the justice system outside of the formal courts in the eighteenth century. As well, I could not be certain that all documents for all sample years had been preserved, especially before 1810. As a result, in calculating overall numbers, I often had to make estimates by extrapolating backwards from the number of cases in the registers.
Because of the state of preservation and accessibility of the sessions papers, for the district of Montreal, I used all documents that I found up to and including 1800, and systematically sampled all documents that I found for every fifth year from 1800 to 1835. Originally, the documents preserved by the clerk of the peace were organized in bundles covering business done in the three months prior to each Quarter Sessions. However, many of these bundles have since been broken apart and the documents scattered, and some bundles appear to have been opened and then repacked with documents from more than just that sessions, which made identifying which sessions a particular document belonged to very difficult. As a result, I chose to sample by year rather than by sessions, with the year of the initial deposition being determinate. To get all documents for a particular year, I had to sample the bundles for all sessions in that year plus the January sessions of the following year (which contained documents produced since the preceding October sessions).
Records of the proceedings of the Montreal justices in Weekly Sessions are scanty, since most of the registers and records of the court were burnt in a courthouse fire in 1844 (as described in a brief notation by the clerk of the peace inside the cover of the 1829 register of the court). The only records of cases that remain are a few entries for the Weekly Sessions in the Quarter Sessions register from 1779 to 1783; registers of the court at the end of the period, for 1829 and again for 1832-1837; and a few scattered documents kept in error, no doubt, among the Sessions papers. The other main information on the operation of the Weekly Sessions concerns convictions only: a detailed list of all defendants fined in the court between 1779 and 1787; the estreats of fines sent in to the colony's receiver general by the clerk of the peace from 1787onwards, which cover only convictions where the fine was actually paid, from 1796 do not cover fines imposed for infractions of the rules and regulations of police in Montreal (which went directly to the city's road treasurer) and are largely missing for the period between 1796 and 1814; scattered accounts of fines received in the papers of the road treasurer; and scattered accounts of convictions in the newspapers.
Finally, for cases that actually made it to the Montreal Quarter Sessions, the records are far better, since an almost complete set of registers survives, covering from 1764 to 1773 and from 1779 to 1837 and beyond. For the eighteenth century, I recorded the full details of every case, but because of the volume of cases in the nineteenth, I followed a similar proceeding to that for the case-files: for every case, I consistently recorded the charge, the names of the parties, and the outcome; but I only recorded the full details of every case, including the presence of attorneys, the appearances of the defendant, and so on, for every fifth year plus the January Quarter Sessions following, matching my sample of Quarter Sessions complaints.
Though the Quarter Sessions registers was the only "complete" source that I was able to find, even it had its problems. Most importantly, unlike the registers of the higher criminal courts, which almost exclusively contained cases where an indictment had been laid before the grand jury, the Quarter Sessions registers often contained references to cases that had reached a variety of stages in the formal process, from those where the only mention was the defendant being called on his or her recognizance, through cases determined summarily in sessions, to cases where an indictment was laid and a formal trial had. Some historians who have used Quarter Sessions records have considered only the second two categories, where there was a formal "case" before the court, in other words either an indictment laid or a summary trial had, and excluded other cases as simply procedural matters. However, there are a number of objections to applying this approach to the Quarter Sessions in Montreal. In the first place, not all cases in which indictments were laid made it into the registers of the court, though those that did not were relatively rare. Even when cases were entered in the registers, for many up to about 1790 that were resolved before going to formal trial, the clerk of the peace did not record the grand jury's verdict, perhaps at the behest of parties who wanted to save fees. Further, from the perspective of the defendant and the prosecutor, simply appearing in court gave the experience of coming in contact with the criminal justice system an entirely different meaning from appearing before a justice out of sessions, whether or not the formal proceedings went any further, though again the registers do not always indicate with certainty whether the parties actually appeared. For instance, when a prosecutor or a defendant was called on his or her recognizance and simply marked as not appearing, there was often no indication whether the other party had appeared. Finally, since only about a third of cases that were in theory supposed to come before the court were ever recorded in the registers, the sheer fact of their being recorded meant that there was something particular about them, and that either a private party had paid the clerk of the peace the fees he was allowed for entering each case, or the justices themselves had ordered that the entry be made. As a result, I have included all cases that officially came before the notice of the court, in the sense that they appeared in the registers. The only cases that I excluded consistently were those where the court imposed fines on jury-members or constables for non-attendance and then remitted them later; however, where there was no indication that the fine had been remitted, I considered these to be cases (with the prosecutor being the court itself) since the court had exercised its summary jurisdiction to impose a punishment.
A further problem with the Quarter Sessions registers was that they provided only partial information on the parties involved. Entries in the registers of course almost always gave the names of the defendants, except in the very rare instances where these were left out by an error of the clerk or otherwise, and up to 1793 they also usually gave the names of the prosecutors. The only exception in this regard was petty larceny cases, which about half of the time did not give the name of the prosecutor; however, these constituted only a very small part of the business of the court, and accounted for no more than thirty unknown prosecutors altogether. However, the registers almost never gave any information about the parties beyond their names, allowing me to determine only their ethnicity and gender, and then only in those cases with names that were unambiguous; as a further complication, they prevented me from excluding cases from areas outside the post-1790 district of Montreal, though these probably represented only a very small number. Further, from 1794 the clerk of the peace increasingly began to leave out the names of the prosecutors in cases where the crown was technically a party, which included most cases brought before the court, and after 1800 almost never included them; this meant that I could not rely on the representativity of prosecutors listed in the registers after the mid-1790s. I was able to address both of these lacunae in part by cross-referencing all cases in the Quarter Sessions registers (based on the names of the defendants) with complaints in the case-files, which gave far greater information on the parties; but this was only possible where the case files existed, in other words mainly from the 1790s on, and from 1800 only applied to years where I sampled the case-files, in other words every fifth year. Because my sample was based on the year of the document rather than the Quarter Sessions it was based on, in all proportional calculations I have included prosecutors from the January Quarter Sessions of the year following the sample year.
Records of the justices in the district of Quebec
My analyses of everyday justice in the district of Quebec was based on several different sources, but the most important was BAnQ-Q TL31 S1 SS1, which contains the sessions papers for the district of Quebec from 1802 onwards. Unlike the equivalent series in Montreal, this series includes both Quarter and Weekly Sessions records, as well as records of summary commitments. It is also especially important since unlike in the district of Montreal, there is no surviving run of Quarter Sessions registers for the district of Quebec before 1823, and also very few surviving weekly sessions registers.
A complete analysis of this series, using the same systematic date-based sampling techniques as for the district of Montreal, would have been impossible in any reasonable time, since unlike the Montreal sessions papers, those in Quebec are not in date order, with groups of documents from throughout the nineteenth and early-twentieth centuries scattered through the 230-odd boxes of the series. Luckily, the entire series has been described essentially document by document in the Thémis 2 database by Archiv-Histo, covering some 40 000 individual documents or case files (and perhaps 55 000 individual documents in all) up to 1837, with each document assigned a serial number. The database records the date, the type of document, the names, occupation and place of residence of the parties and the charge, and for case files (generally groups of two or three documents, such as a deposition with its accompanying recognizance) also gives an indication of the documents. It is especially useful since it provides an efficient way to constitute a systematic date-based sample is to use the database. Given the level of detail, I was able to use the database records as the basis for my quantitative analyses of prosecution rates, places and parties. By transforming and systematizing the information presented in the database (which had no distinct fields for any information apart from the date and the document number), I used four sample years at ten-year intervals, namely 1805, 1815, 1825 and 1835. I also made estimates of the total number of cases and documents in the case files.
The accuracy of any such database is of course key, but checks on a large number of these documents in the context of another research project have led me to the conclusion that the data entry was relatively accurate, with the occasional errors coming mainly in the misspelling of English names and words. The one significant difficulty with the Thémis 2 database concerns individuals described as "habitants" in the documents. The database creators decided that this was too uncertain a term (since in French it can mean either "inhabitant of" or the more usual "farmer") and thus omitted it altogether in their document descriptions. In order to deal with this problem, I entered the limited number of rural inhabitants without stated occupations as farmers, but even if some of these are in error, it would not much change the overall picture.