My first book, Club Government (available here from Amazon), contains a number of findings based on my exhaustive research around the club memberships of over 2,500 British Members of Parliament between the first two Reform Acts, drawing on over 50 print and archival sources.
As a firm believer in open data, I have decided to share my dataset, and to make it publicly available. Anyone is welcome to make non-commercial use of it - provided you give an attribution for any use of it, or any work derived from it - under CC BY-NC-SA 4.0 Creative Commons licence:
Seth Alexander Thévoz, 'Database of MPs' club memberships, 1832-68'.
The default tab lists the aggregate of all MPs who sat between the first two Reform Acts. Further tabs give a parliament-by-parliament breakdown, using the date of the first sitting, i.e. "33" for 1833, since Parliament did not sit after the 1832 general election until 1833. Party labels in red recognise a measure of personal interpretation that has been necessary from the (sometimes ambiguous) party labels for that MP listed in Dod's Parliamentary Companion. (Labels used here were the first ones given for an MP in each Parliament. This was usually in a New Parliament Edition of Dod's released after a general election - unless they had been returned mid-Parliament in a by-election.) If an MP's year of joining a club is known, it is given - otherwise, their membership is simply listed as "X".
Such is the inherent complexity of the minefield of party labels in the mid-nineteenth century, that I enclose below an appendix from my PhD thesis, explaining something of the methodology that was used in arriving at these party labels, based on the source material.
A note regarding party labels in the 1832-1868 period
The use of party labels among MPs – and the appropriateness thereof – remains a heavily contested notion in mid-nineteenth century British politics, worthy of a full-length study in its own right. Indeed, Joseph Coohill's recent study Ideas of the Liberal Party deconstructed many of the ideological and practical implications of party labels in Liberal politics for the 1832-52 period – although the Conservatives have not so far enjoyed any comparably detailed dissection across the 1832-68 period. Accordingly, this appendix contains an explanation of how and why the figures in the database have been reached, and so a rationale behind the categorisation system used in this thesis. The database can be viewed in the attached CD-ROM.
MPs’ party labels in this thesis have been sourced from the labels allocated to MPs in Dod's Parliamentary Companion (originally published as the Parliamentary Pocket Companion), which in turn were sourced from a combination of MPs’ self-descriptions in the questionnaires Charles Dod sent them, and the Dod's editors’ own observations of MPs’ positions at elections and in crucial Commons divisions.
Stenton and Lees’s approach, and the difference from my own approach in the attached database
Previously, Michael Stenton's reference work Who’s Who of British Members of Parliament, Volume I: 1832-1885 (and its immediate post-1885 sequel volume co-edited with Stephen Lees, which included some biographies of long-standing MPs who had served prior to 1868) applied one single party label to each MP, usually (but not always) based upon the MP's final Dod's entry.
What has been retained from Stenton and Lees's work has been their use of six party categories: Conservative, Liberal Conservative, Liberal, Radical Reformer (sometimes abbreviated to ‘Radical’), Reformer, and Whig; plus a seventh category, if one includes a handful of Independents in the 1830s, although for most of these, Stenton usually gave no party. However, the use of the ‘New Parliament Editions’ of Dod’s has meant the judgments about MPs’ labels at different times have varied considerably from the Stenton and Lees judgments, particularly in tracking the changes of MPs’ views over time. The categories employed by Stenton and Lees themselves differ considerably from a number of other categories historians have used, but the following notes seek to set out how and why these categories are thought to encompass a fair representation of the range of views represented in Parliament in 1832-68; and to show how a very different set of assumptions have been used in reaching these labels, rather than just emphasising the views at the end of an MP’s often-lengthy career, and instead stressing the views they espoused each time they had sought a fresh electoral mandate at a general election (or in a handful of cases, between the last election, and the first post-election edition of Dod’s being published).
Labels marked in red in the database
Given the considerable ambiguities over labelling, the attached database marks many MPs’ party label in red. This denotes that no explicit label was given in their corresponding Dod’s entry for a particular year, but that sufficient information was provided to be able to provide an approximation. Since such entries required more speculation than others, they are not necessarily deemed to be of equal value to those where a party label was explicitly stated in Dod’s, and so they have been identified as such through this colour-coding. The number of these MPs each Parliament is outlined in Fig. A.1. More on the nuances of these MPs is outlined below.
The majority of MPs, listed in the database in black, had relatively easily-identifiable party labels. Yet as the length of this appendix testifies, even categorising these was not free from methodological difficulties.
Changing labels
A layer of complexity has been added by the way that an MP’s party label could change over time. There were numerous causes of this, the most common being:
MPs genuinely changing their party affiliation, such as Lord Arthur Lennox, who after no formal category being assigned in 1833 was deemed to be a Whig in 1835, 1838 and 1841, yet by 1847 had been classified as a Liberal Conservative.
Certain party labels falling into disuse, most notably ‘Reformer’.
Dod's editors reappraising and correcting their earlier choices of label, often in light of more information, such as Martin Joseph Blake’s entire original 1833 entry, which simply gave his constituency, and the label ‘A Reformer’ – no other information was provided. By 1835, his Dod's entry was much more extensive, and Dod had reassessed his label as ‘A Repealer.’
MPs declaring themselves to be of one label, but Dod's noting that their voting record put them partly or substantially closer to the position of another party, for instance, in 1857 the Marquess of Stafford was listed as ‘A Liberal’, but it was noted that he differed with many Liberals on several policies: ‘Voted for inquiry into Maynooth, and against the ballot 1853, and against church-rates 1855.’
In addition, as is recognised throughout this thesis, there has also been substantial debate among historians about when the notions of ‘party’ and ‘parties’ became meaningful terms.
Consequently, according to Dod's some longer-standing MPs could change labels more than once, as with Robert Ingham, who is listed as a Reformer in the Parliaments of 1832, 1835 and 1837, then a Whig when he resumed his Commons career in 1852, and then a Liberal in later Parliaments.
The use of Dod’s “New Parliament Edition” labels
Given such changing labels, and the need to assess the whole of the period 1832-68, it has thus been deemed misleading to choose just one arbitrary label for the whole of an MP’s tenure. This has directly affected the way some figures have been counted – for instance, in Figs. 2.11 and 2.22, there is no party breakdown for all MPs who belonged to a particular club, or no club, as an aggregate for the whole of the 1832-68 period. Instead, the figures take a sample from the beginning of each Parliament, using each ‘New Parliament Edition’ of Dod's – the first edition to appear after every general election – to lend uniformity and consistency to my data set. This is, of course, not entirely free from problems, but as I seek to outline here, this approach is freer of problems than any alternative approach.
The main problem which remains from using ‘New Parliament Editions’ is that it does not take into account changes of label arising from point (c) above, when Dod's subsequently revised their earlier judgment of an MP's label. However, there are several reasons why this has been deemed acceptable:
The actual number of MPs this applied to, where there was an error in an early volume of Dod's, as opposed to an actual change in party by the MP, barely amounted to a handful; and was certainly not enough to be statistically significant for the 2,588 MPs who sat across this period.
Parliaments in the period 1832-68 varied in length from two to six years. If a later Dod's edition had been used, this would have raised serious methodological issues from a statistical perspective. Using the final edition of a Parliament would have meant that in some Parliaments, Dod's had six editions over six years to correct their earlier work, whilst in others, they would have only had one or two. (For instance, the first edition of the 1832-5 Parliament appeared in 1833. A revised, second edition appeared in 1834, but no further editions were produced in the Parliament, since the 1835 edition covered the 1835-7 Parliament. By contrast, the 1841-7 Parliament saw an initial Dod's edition published in 1841, and subsequently corrected in editions of 1842, 1843, 1844, 1845, 1846 and 1847.) Similarly, selecting an edition from the middle of a Parliament would have suffered from similar inconsistencies.
In any case, even if mid-Parliament or end-of-Parliament editions were used, they were not necessarily free of such errors or misjudgements. For instance, Sir Francis Burdett is generally accepted to have been a Conservative in the late 1830s; certainly by the May 1837 by-election he triggered over his change of allegiance. Yet even the 1838 ‘New Parliament Edition’ of Dod's still gave a highly misleading description of him, little changed from the 1835 edition. No explicit present-day party label was given, but it was noted ‘Sir Francis’s early advocate of Reform…[is] well known. He was a strong supporter of the Whig government, and declared himself without confidence in that of Sir Robert Peel; but refused to declare that he would vote against it. In favour of the ballot.’ No mention was made of his having crossed the floor; the impression given was that he was still a Whig. This listing was repeated in 1839 and 1840. Only in 1841 was it noted, after he had switched from representing Westminster to a county seat in Wiltshire, ‘since his return for the county [he] has generally voted with the conservative party.’ (Original capitalisation retained.)
Use of later editions mid-Parliament or towards the end of a Parliament would also have been prone to distortions caused by party splits which occurred mid-Parliament. In particular, Dod's sent out new questionnaires after every general election. Although individual MPs' entries were constantly revised and updated, wholesale reappraisals of all MPs and their party labels were not always undertaken mid-Parliament. Indeed, it is an admitted weakness of the Dod’s source material that as the above case of Sir Francis Burdett shows, even after a general election, not all existing MPs had their party labels reappraised; this was even more so in editions appearing mid-Parliament.
As noted below, the Dod’s party labels were informed by ‘Political Principles and Pledges’; namely, the pledges given by a successful candidate at the preceding election. Since the Dod’s labels reflect an MP’s pledges when they were returned, it is then most accurate to use as data points these labels when they would have been most recent, in the New Parliament Edition. Note that not all Dod’s entries reflected such pledges, but that a significant number made reference to them and were clearly informed by them. Thus to use Dod’s data from later in a Parliament would risk interpreting out-of-date pledges as a fresh mandate.
Inconsistent labelling
By the last third of the nineteenth century, Dod's had settled on clear, categorical labels for all MPs. This was not the case in earlier volumes.
The most obvious instance was with inconsistent capitalisation; and on closer scrutiny, it is difficult to see that this denoted substantial rather than typographical differences. Early volumes of Dod's from the 1830s to the mid 1850s frequently referred to:
‘A Whig.’
‘A whig.’
‘Of Whig principles.’
‘Of whig principles.’
‘Of whig opinions.’
‘Is of decidedly whig opinions.’
This variation was not a peculiarity exclusive to the Whigs. The labels ‘Conservative’ and ‘Liberal’ were also highly prone to such variations, and further discussion can be found below on how the term ‘Conservative’ could embrace a broad variety of sub-categories. However, it should be recognised that different historians have been able to take different – and equally valid – interpretations of such capitalisation. For instance, in his analysis of Liberal Party identity, Joseph Coohill does not count MPs listed by Dod’s as ‘liberal’ with no capitalised ‘L’ as Liberals, and differentiates the capitalisation (certainly from 1834/5) of liberal/Liberal as representing liberal ideas compared to a Liberal Party. This is, of course, a perfectly valid linguistic interpretation, especially for the purposes of exploring the evolution of liberal thought, and it is undoubtedly grammatically correct. However, such labels are open to multiple interpretation, and if the aim of the database is to attribute a party affiliation to as many MPs as possible, where appropriate supporting evidence exists – which is, of course, a very different objective to Coohill’s in breaking down the make-up of the early Liberal Party – then different totals can be extrapolated from the data. A strong case can be made that an identification with liberal politics, with a small l, in the absence of any other stated party affiliation, is indeed arguably an identification with Liberal party politics with a capital ‘L’. Furthermore, in sifting through the sometimes arbitrary capitalisation and punctuation in Dod’s (as with the above-noted instance of Burdett’s support for the non-capitalised ‘conservative party’ from 1841), one is struck by inconsistencies in capitalisation between editions, and even in the same volume; and so it seems quite likely that at least some of this was due to the peculiarities of individual sub-editors, or inconsistent capitalisation from different MPs’ self-descriptions, which did not necessarily carry any greater partisan significance. Naturally, given the non-existence of an ‘official’, formally-defined, structured, nationwide Liberal party in the country until the creation of the National Liberal Federation in 1877, all of these assumptions are open to question. However, such arbitrary rules have had to be laid down in order to make the statistical analysis possible, and no alternative rules would be wholly free from such questions. It is hoped that the enclosure of the full database, and its comparison to the original Dod’s volumes, will allow others to note such assumptions that have been made, whether they share them or not.
Combinations of labels
Sometimes, the compilers of Dod's would hedge their bets by attributing more than one label to the same MP. For instance, the 1841 edition of Dod's refers to Charles Hindley as being ‘Of Whig principles, inclining to Radicalism.’ If one is to transcribe party labels in a limited number of set categories for that data to be machine-readable – as has been done here – this presents a dilemma over which label to select. In such cases, I have had to use my discretion, but as a general rule of thumb, the first label named is taken as the principal one. With Hindley, for instance, this was evidently the case: his Dod’s profile went on to describe how he had previously stood in opposition to a Radical candidate, indicating that he should not be primarily seen as a Radical.
Some of these combinations could be quite baffling, naming factions that might be thought to be mutually exclusive: In 1847, Sir Robert Alexander Ferguson was believed to be ‘Of Whig principles, inclining to Conservatism.’ (As per my default preference for using the first label named, he has been categorised as a Whig). Ferguson’s case is of course symptomatic of the considerable confusion to be found when labelling Conservatives of any denomination in the late 1840s and throughout much of the 1850s.
The default choice of the first label as the principal one can lead to some aberrations: numerous Repealers did not have their label capitalised, yet a lengthy description of their views, without a label, included a mention of their favouring repeal of the Act of Union, without any other obvious views to link them to other factions. In the absence of a label, this has usually been enough to classify them as Repealers (albeit cautiously, as denoted by a label appearing in the database in red). Yet in the second 1847 edition of Dod's, John Thomas Devereux was classified as a ‘A Liberal, and in favour of a repeal of the Union’; there is thus a reasonable claim to classify him as a Repealer, but he has been classified as a Liberal, as the capitalisation and ordering of the sentence suggested that he was to be seen as a Liberal first, and a Repealer second. Similarly, in the same edition, James Fagan was given as ‘a Liberal and a Repealer’ - and so is counted as a Liberal, since it implies he was a Liberal first, and a Repealer second.
Such combinations of labels were particularly baffling in the handful of instances where three separate party labels were given as a description, as with William Pinney in 1833, ‘a Reformer, entertaining Whig principles, inclining to Radicalism’; again, the standard assumption has been that the first label given is the principal one, with subsequent labels being descriptive, and so Pinney is categorised as a Reformer.
(Note that the label ‘Liberal Conservative’ is not considered to be such a combination of separate labels, but a recognised, distinctive contemporary identity, most commonly used as a synonym for ‘Peelite’ – although the use of the term in describing a handful of MPs prior to 1846 confirms the complexity of the Liberal Conservative identity, and attests to it being open to interpretations other than ‘Peelite’, pre-dating Conservative schism. It is for this reason that I have not sought to automatically describe all Liberal Conservatives as Peelites, and have sought to preserve the contemporary idiom.)
Such instances of multiple labels naturally mean that the same data is quite naturally open to different historians making different tallies – even producing different labels – simply by using slightly different counting methodologies, and placing different emphases on the inconsistent categorisation employed by Dod's. Numerous examples of this are discussed below, under ‘Categorisation’.
No labels
A further problem of Dod's Parliamentary Companion is that although each entry contains something about the ‘Political Principles and Pledges’ of every MP included, this did not necessarily add up to a convenient party label.
Particularly in the first two decades of Dod's, a number of MPs had no party label explicitly stated. Instead, some statement of their principles was provided, often based on the questionnaire filled out by those MPs.
Perhaps more surprising is the category of current or recent Government ministers: while their history in office was faithfully recorded, they would often have no explicit party label recorded. This was, of course, fully in line with the long-standing parliamentary tradition of regarding office-holders, or ‘Ministerialists’, as a party in their own right. Consequently, throughout the ‘New Parliament Editions’ of Dod's in the 1832-68 period, some of the most well-known names of the age, such as Lord John Russell and Viscount Palmerston, were never allocated an explicit party label during their Commons careers. This perhaps requires some explanation. As Joseph Coohill outlines in his study of parliamentary guides, the purpose of Dod's and the other parliamentary companions which proliferated after the Great Reform Act was to respond to a new breed of candidate returned after 1832, who would not necessarily be established in London society and already known to other parliamentarians. As such, parliamentary companions served an important role in embracing an increasing non-aristocratic element in the House of Commons. However, for well-known national figures that had recently served or were serving in cabinet, it would have been wholly unnecessary to allocate labels.
In cases of government ministers without a formal label, the default has been to duplicate the pre-existing label from their earlier entries before they entered government, or failing that, a later entry from soon after their departure.
However, as noted, a handful of ministerialists never had a party label recorded by Dod’s during their time in the House of Commons. In the complete absence of one, the political composition of the government has had to be considered, as a "tie-breaker". Thus, although Lord John Russell never had a party label in his years as an MP, Dod’s later allocated him one when he moved to the House of Lords (‘Liberal’). However, although this may be an accurate reflection of his views by the 1860s, it would be a perplexing description of his views from the 1830s to the 1850s, and so for all but his last Parliament in the Commons, he is listed as a Whig, since this is a widely-accepted description of the ministries he served in, and Dod’s draws attention to his ministerial service as an indication of his views.
Figure A.1 – Breakdown of the party tallies from Dod’s Parliamentary Companion New Parliament Editions, using the methodology outlined
(TABLE)
Note: Upper figures in black denote totals. Lower figures in red denote how many MPs of that label, in that year, have required a significant amount of interpretation in the absence of an explicit label provided in Dod’s Parliamentary Companion. Red figures include MPs for whom multiple labels were given, and a choice had to be made.
Source: Seth Alexander Thévoz, ‘Database of MPs’ Club Memberships, 1832-68’ with party labels derived from the 1833-65 New Parliament Editions of Dod’s Parliamentary Companion, using the counting methodology outlined in this appendix.
The figures in red in Fig. A.1 show the number of MPs in each Parliament for whom no explicit label was given by Dod's; by this, it is meant that key party label words - Conservative, Liberal, Liberal Conservative, Radical Reformer, Reformer, Whig - did not appear in the corresponding Dod's entry; or else, more rarely several conflicting labels were given. (Note that in the database on the attached CD, these are the MPs whose labels are written in red.) However, in the vast majority of cases, enough was said of their principles/election pledges/voting record to enable a relatively straightforward judgment as to enable an educated guess, and so the main method of deducing labels in these instances has been to simply look at what Dod's says about their views, and how it tallies with positions associated with the different parties. Many such decisions are not difficult: attachment to the ballot and universal manhood suffrage in the 1830s tallies well with ‘Radical Reformers’ (although such view were arguably no longer ‘Radical’ by the 1860s), while a cautious supporter of modest reform whose aristocratic lineage was emphasised has often been categorised as a ‘Whig’. In some cases, when next to nothing was said about the MP in their original entry, their Dod's entry from the preceding or following Parliament (if they had one) was used as a ‘tie-breaker’ – but only in instances when there was a lack of other available information. Many of the conclusions reached on individual MPs have tallied with the Stenton and Lees labels, though this has by no means always been the case, particularly with MPs who changed labels. For instance, Charles Edmund Rumbold was classified by Stenton and Lees as a Whig. This does indeed reflect his affiliation in the Parliament of 1852. However, in the Parliament of 1832 Rumbold had been a Radical, and throughout 1837-47 he is listed in the database as a Liberal – although this latter construal in turn conflicts with Coohill’s interpretation, for Dod’s stated Rumbold ‘holds liberal opinions’, and so the lack of capitalisation in the label meant that he failed Coohill’s definition of a Liberal.
As noted, in taking this approach the emphasis has been on allocating party labels to as many MPs as possible, where appropriate, as opposed to not allocating labels at all as a default whenever there has been some ambiguity. Consequently, very few MPs emerged from this as true ‘Independents’, because the Dod's data usually stressed key stances and principles in the absence of a clear-cut label. An alternative, ‘purist’ methodology in the absence of a specific Dod's label would have been to count all such MPs as Independents. However, given the number who clearly still declared themselves supporters of particular causes, and given that this says as much about the shortcomings of Dod's information-gathering and lack of systematised formatting as about the MPs themselves, this would risk becoming a misleading misconstrual of the original source material. It would also be completely at odds with most other labelling systems which also seek to extend labels as widely as possible (see section below, on ‘other party label tallies not used’).
Categorisation
Because of the Dod’s labels’ inconsistencies in style, punctuation, capitalisation and formatting, literal use of the raw labels from Dod's would be utterly impractical - unless one were to construct bar charts with several dozen different polysyllabic variant categories with mildly differing punctuation, many of which were synonyms for one another. Instead, some interpretation is required to put them into categories. This had already been done by Michael Stenton and Stephen Lees; but as mentioned, almost entirely using party labels from the end of an MP’s careers, rather than throughout their careers.
As noted, even basic conclusions about an MP’s label cannot be reached without some interpretation. For instance, in describing the ‘parliament of such uncertainty regarding its party composition’ that first met in 1847, Angus Hawkins notes that on the Conservative benches:
The difficulty of assessing party numbers accurately became apparent when Conservative MPs submitted their affiliation to the new Dod's Parliamentary Companion after the election. The labels Free Trade Conservative, Liberal Conservative, Moderate Conservative, Conservative, Protectionist, and Tory were variously used.
while also summarising the then-Lord Stanley's view about the ‘mixed nature of the Liberal majority...composed of discordant elements.’ This is, of course, a judicious and nuanced recognition of the subtleties of party labels at a time when they were prone to considerable upheaval. Yet even reaching such a balanced conclusion requires some interpretation with the original raw material from Dod’s. Thus a precise tally of the exact breakdown of different Conservative factions listed in the second 1847 (New Parliament) edition of Dod’s shows the following permutations of party labels, and shows how just as Hawkins has taken the same data set to conclude that there were at least six variant Conservative labels, so the data transcribed into this thesis’ database has come up with a differing tally, using two variant Conservative labels, Conservative and Liberal Conservative – and as can be seen, other valid permutations and interpretations are possible:
Figure A.2 – Tally of exact Conservative descriptions in Dod’s, 1847 (second edition), showing precisely how they have been aggregated together as ‘Conservatives’ and ‘Liberal Conservatives’, and preserving original spelling, punctuation and capitalisation
(TABLE)
Note: Labels marked in red denote entries in the database in red, which in turn denote that the precise phrase ‘Conservative’ or ‘Liberal Conservative’ is not used in the description, but that the corresponding Dod’s entry gives some reasonable grounds to group these MPs in this category. It is hoped that by tabulating such detailed information for the Conservatives in 1847, some insight is given into the extent of interpretation required (and deemed acceptable by the author) in deciding how to categorise MPs for whom an imprecise description was given, but sufficient information still provided for a judgment to have been made rather than misleadingly listing such MPs as having no party or no principles.
Even this detailed breakdown risks presenting something of an oversimplification of MPs’ views, for in seeking to isolate their labels, and given that different statements on MPs were separated in Dod’s by a semicolon, I have made a rule of thumb to record everything in Dod’s pertaining to the party label, up to the first semicolon. (Exceptions to this rule have been noted in the footnotes.) Yet some descriptions of an MP’s views could give several sentences of text, describing unorthodox combinations of stances, separated by numerous semicolons, and so even the above table does not fully capture all pertinent information. Clearly, such a listing would be wholly impracticable for a machine-readable format.
In presenting this breakdown, it is hoped to convey not only the diverse range of MPs’ labels and the inherent difficulty in grouping MPs together, but also the particular difficulty in agreeing on categories that are feasible for a machine-readable format. The above chart gives thirty-four categories of Conservative, and twelve categories of Liberal Conservative; but thirty-seven of these forty-six categories contain just one MP, and most are separated by minor nuances of capitalisation and ordering. It is conceded that the labels marked in red have involved more interpretation than most. Nonetheless, they are included to illustrate how sufficient information was still provided to give a reliable party label; or at least a party label that is more plausible than the alternatives, and that would be far more accurate than recording such MPs as ‘Independent’ or as having had no party political affiliation.
Particularly problematic is the lop-sidedness of the above chart in categorising far more MPs as Conservatives (283) than Liberal Conservatives (fourteen), and it is at odds with J. B. Conacher’s tally of up to 115 MPs who were sitting in 1847 voting with the Peelites at least some of the time. This is especially notable amongst the seventy-three MPs listed as ‘Conservative’, but also noted to have expressed some support for free trade. One could argue that this was the very definition of a Liberal Conservative, and that such MPs should be listed amongst this group rather than the Conservatives. However, I have listed them among Conservatives for three reasons. Firstly, the great strength (and weakness) of the Dod's labelling system was that it accommodated how MPs viewed themselves. Thus it is revealing that early in the 1847 Parliament, the great mass of so-called Liberal Conservatives and/or Peelites still viewed themselves primarily as Conservatives. Secondly, if they were listed as Conservatives in Dod's, then retrospectively re-terming them Liberal Conservatives is arguably a reinterpretation too far, and would change the original meaning Dod’s intended. And thirdly, as noted in the thesis, historians have fiercely debated whether repeal of the Corn Laws was the primary cause of Conservative schism in the 1840s, with religious schism frequently suggested as a rival factor. Dod's additional observations on MPs who were ‘Conservative, but in favour of free trade’ further supports the notion that the Conservative schism carried a religious as well as economic dimension. For instance, on the issue of free trade, City of London MP John Masterman was at one with most of the Liberal Conservatives; but he is listed as a Conservative, and a clue to his attachment to the mainstream Conservative cause is offered by his 1847 Dod's description as being ‘prepared to resist any further concessions to Popery.’
Even among the sixty-six MPs explicitly listed in 1847 as ‘A Conservative, but in favour of free trade’, no clear trend was apparent. While one might expect to find such MPs were listed as Conservatives in the 1841-7 Parliament, and as Liberal Conservatives in the 1850s, the picture is confused by MPs like Earl Jermyn, who was given just such a free trade label in 1847 yet remained consistently classified as a Conservative throughout his Commons career, from 1830 until 1859. Equally baffling was the case of the notoriously inconsistent William Keogh. He occupied a category of his own, as ‘A Conservative; in favour of free trade, and a supporter generally of Sir R. Peel's policy’. The part of the description following the semicolon would have made him a textbook definition of a Peelite. Indeed, Keogh was classified by Conacher as a Peelite in the 1847-52 Parliament. He then played a role in the foundation of the Irish Independents, and by the time of the 1852 Dod’s New Parliament Edition, he was categorised as a Liberal, ultimately holding office under both Aberdeen and Palmerston. Nonetheless, I have striven for consistency in preserving the spirit of Dod’s in labelling him as a Conservative when he was a newly-elected MP in 1847.
In the instances when it was used, ‘Tory’ has been accepted as a synonym for ‘Conservative’; not only is this in line with the contemporary vernacular label ‘Tory’ as a shorthand for ‘Conservative’, but it is in line with most other categorisation systems of the period, including Dod's Electoral Facts, Stenton and Lees, F. W. S. Craig, and the History of Parliament labels. It would thus be inappropriate to arbitrarily begin using the term ‘Tory’ as a separate label, without undertaking significant additional research, which more properly belongs in a study of mid-nineteenth century conservatism and party labels, rather than one on clubs and politics. Furthermore, as the word was only used in 1847 by four Conservatives and one Liberal Conservative (plus one Liberal Conservative who used it denote what he wished to distance himself from), it really was too rare a label to consider as a separate party grouping for the purposes of statistical analysis of club membership trends.
Ultimately, it is quite feasible to arrive at differing figures from other historians using exactly the same Dod’s data. For instance, the notion of MPs without party has been noted – termed ‘Independents’ above. In his own database, Coohill used a very different burden of proof in identifying these MPs of no party: ‘Not all Liberal Party MPs have an exact and literal party term in Dod, even though their entries may go on to list obviously liberal (usually radical) political statements. If this was the case ‘None listed’ was used.’ Yet in the database compiled for this thesis, because of the slightly different objective in assigning party labels as widely as possible, party labels have ended up being assigned differently, particularly to those with ambiguous party affiliations. As Coohill himself acknowledges, his own figures for Liberal MPs holding different party labels vary from those compiled by T. A. Jenkins, Jonathan Parry and Miles Taylor, due to the different summary labels each has used. For instance, Jenkins's tally of 163 MPs using the label ‘Liberal’ in the 1847 New Parliament Edition of Dod's is quite close to Coohill's tally of 166, but still differs noticeably from Parry's tally of 175, and my own tally of 182. Similarly, taking a different basis from Dod's (or a different source altogether) could also produce different results. Based on press reports throughout Parliaments (and not Dod’s), Miles Taylor places the number of Reformers (of which he has his own definition, based partly on their independence and autonomy as a group) at sixty in the Parliament of 1847-52. This figure differs significantly from the equivalent tallies of thirty-seven (Coohill), thirty (Parry), or forty (me), all of which are based upon Dod’s; while Taylor's tally of forty Reformers in the 1852-7 Parliament still differs from my figure of twenty-seven. Such differences are natural, as the Dod's New Parliament edition labels reflected both an MP's election pledges, and their voting record. Consequently, when there was some variance between the two, distortions to party labels could and did occur. Each tally, naturally, is the product of slightly different methodologies. Crucially, it is not contended here that the labels and tallying system I have used are inherently any more or less accurate than those compiled by Coohill, Jenkins, Parry, Taylor, et al. – merely that the counting methodology I have used is more suited to the aims and requirements of this particular study into clubs and politics.
Other party label tallies not used
In addition to Stenton and Lees, alternative party labels for this period have been attributed by others, but it was decided they were less suitable than Dod's labels for a variety of reasons.
Journalist James Grant provided party labels in his 1836 book Random Recollections of the House of Commons. However, the labels do not comprehensively cover the 1832-68 period, and so would be inappropriate to use here. As Joseph Coohill has noted, of particular interest is the way in which Grant used unique labels not found anywhere else, although this was abandoned by his 1838 volume The British Senate: Tory Party - Late Members, Tory Party - Present Members, Neutral Party, Members of the Government, Metropolitan Members, Country Liberal Party, Irish Liberal Members, Literary Members, Religious Members and New Members.
Charles Dod provided a set of labels for the period 1832-1852 in Dod's Electoral Facts (1852), which the History of Parliament's ongoing research indicates contains some of the most accurate pre-existing labelling of the post-Reform period. However, these labels were not comprehensive in covering the period up until 1868; and serious methodological problems would arise in seeking to combine these 1832-52 figures with another source for 1852-68.
In addition to Dod’s, numerous lesser-known parliamentary companions were produced from the 1830s; however each of these have their own problems with party labels. Not only were many non-comprehensive in their coverage across the 1832-68 period, but party labels could be hopelessly incomplete (as with The Parliamentary Test Book for 1835, which classified MPs as either Reformers or Conservatives), or even non-existent ( as with Vacher’s, which did not include party identities).
F. W. S. Craig's reference work British Parliamentary Elections 1832-1885 used a format and labelling system originally developed for twentieth-century British politics. Consequently, Craig's focus on two-party politics led him to conflate the vast majority of candidates as either ‘Conservative’ or ‘Liberal’, even when these concepts were ahistorical. Thus Craig's labels are of limited use in the pre-1868 period.
The History of Parliament Trust is engaged in comprehensively writing biographies of all MPs who sat in the 1832-68 period, and as part of this, each MP's party affiliation is being reappraised. However, this work is not expected to be completed for another decade or so, and so this data is not presently comprehensive. When it is complete, it will of course provide arguably the most reliable and in-depth guide to party labels in this period, compensating for inaccuracies in the original Dod’s volumes.
Accordingly, even though Dod’s is by no means perfect or unambiguous, and even though it is far from the only usable data set relating to MPs’ party affiliations in this period, it is less flawed and more comprehensive than any other available alternative; and the counting methodology used has sought to acknowledge expressed party affiliations, even when this has necessitated some flexibility in the interpretation of the admittedly-sometimes-ambiguous original source material.
(Seth Alexander Thévoz, January 2014.)