‘The art of the legislator’

‘Ends that ought to be kept in view in a code of regulations’ introduced Bentham’s discussion of code-writing projects.¹⁰ Bentham instructed legisla­tors to employ the necessary ‘art’ when they wrote their parliamentary pro­cedures.

The tactics of deliberative assemblies, as well as every other branch of the science of government, ought to have reference to the greatest hap­piness of society: this is the general end.¹²

The guiding preference of the times divided, headed, numbered should- statements - written in semi-regimented language - and presented them in codified format. The headings supplied condensed references to the discrete logics that code-writers employed and also drew attention to the functions that these logics served. Bentham took these should-statements, numbered them, dressed them with minimal historical support on the page - or ignored history entirely - and delivered them to the reader. Bentham preferred that should-statements appear in a ‘code of regulations’. This preference over­matched the print culture of the 1790s.

Bentham’s favoured posture mimicked a parliamentarian (or programmer) writing instructions for those charged with the task of writing the if-then- else statements that resemble, to a modern eye, pseudo-code. I underline the one-step-removed posture that framed Bentham’s Essay. There is at least one similar example of an investigator striking such a pose. In the Federalist Essay no. 34 (1788), Alexander Hamilton declared:

Constitutions of civil government are not to be framed upon a calcula­tion of existing exigencies; but upon a combination of these, with the probable exigencies of ages, according to the natural and tried course of human affairs.¹³

The one-step-removed guidance that Hamilton and Bentham offered must be placed alongside James Harrington’s A System of Politics (1700), to be discussed in Chapter 7, this volume.

Bentham, however, insisted that his Essay could be taken as an instance of code-writing, without referring his project to Harrington’s.

I have chosen all along, as far as the nature of the case would give leave, to exhibit the proposed regulations in the very words in which

they might be couched.... By specification, description is saved, atten­tion arrested, and expectation satisfied: description, however well per­formed, leaves the main work still undone.¹⁵

Contrary to this pronouncement, Bentham did leave his ‘main work undone’. Henry Scobell produced a serviceable mini-code of procedure in the course of listing 169 prescriptions guiding and governing collaborative behaviour in the House of Commons. Scobell’s presentation of algorithms appeared in his Memorials (1670) under the heading ‘Publique Bills’. Bentham’s Essay acknowledged that Honore Riqueti, cmte de Mirabeau, commissioned Samuel Romilly’s Reglemens Observes dans la Chambre des Communes, an 88-page summary of House of Commons procedure. Mirabeau wrote the preface.

Bentham was familiar with this work. Romilly’s summary of parlia­mentary practice in the House of Commons ran 9,659 words, omitting footnotes.¹⁶ Romilly’s Reglemens condensed the essential points of a parlia­mentary code with commentary and algorithms interwoven in a single text. That concise treatment was, however, outdone by the ‘Chapitre Premier - De L’assemblee Nationale Legislative Section V - Reunion Des Represent­ants En Assemblee Nationale Legislative’ of the first French Constitution (3 September 1791).¹⁷ This mini-code contained the essentials of internal law-making procedures in less than 350 words, done inventory-style, as one would expect. Romilly’s well-presented reduction of parliamentary proce­dures in his Reglemens offered Bentham a guide by which to measure his own ambitions.

Bentham, however, had other things on his mind. He had recently learned of the ‘approaching meeting of the French States-General, since termed the National Assembly’.¹⁹ This development, as significant as it was, distracted Bentham from consideration of other procedural projects. Americans were hard at work crafting codes of parliamentary procedure; these were as comprehensive in scope as the codes produced by French republican assem­blies. Bentham could have drawn on mini-codes of procedures written by the newly organised House of Representatives and Senate: these legislatures opened for business in April 1789. The House of Representatives divided its first code into 52 rules. The result consumed a mere 1,985 words. Unlike the Standing Orders of the House of Commons in the eighteenth century, these rules supplied a comprehensive, if thin, treatment of public business. The House of Representatives’ procedures did not address internal improvement projects.²⁰ Bentham had no business ignoring these rules. Nevertheless, he promised his readers ‘the essential articles’ of ‘a system of well-digested rules, such as the English practice, with little improvement... will be found comprised within the compass of a page’.²¹

Despite this sloppiness, Bentham also presented a useful point of depar­ture by drawing attention to the code-writer’s encounter with the blank page. This visualisation, in turn, supported his broader claim that best prac­tices in procedure could be learned, more or less, on the job. On the subject of his ‘system of rules’, Bentham referred the reader to the skill acquisition path.

In a system of well-digested rules, such as the English practice, with little improvement, would supply, will be found the only buckler of defence that reflection can have against precipitancy, moderation against vio­lence, modesty against arrogance, veracity against falsehood, simplicity against deception and intrigue.²²

Bentham pursued one aspect of coding procedural rules, by calendaring motions and their operative effect:

§12-1: Amendments are thus reduced to six kinds, and are capable of receiving clear and precise denominations: -

Amendment { Suppressive.

{ Additive.

{ Substitutive.

Amendment { Divisive.

{ Unitive.

{ Transpositive.²³

Under the first heading Bentham gathered three types of motions to amend; under the second class he addressed these motions taken as functions of word-management. ‘The second class will include all those which directly or indirectly tend to cause the original motion to be rejected’, Bentham explained, ‘as by demanding priority in favour of some other motion, or by proposing an adjournment of the question for an indefinite time, &c’.²⁴

Bentham’s analysis lines up with that of Hatsell and Jefferson. Balloting marked both the conclusion of a discrete event state in which debate had taken place and the beginning of a new event state, in which motions to revise the text were in order. In this way, Bentham made clear how ‘com­position’ of a complete text may be broken down into stages so that ‘all the parts are only considered successively’.²⁵ Bentham’s exposition must be taken as independent confirmation of the Hatsell-Jefferson analytic. Debate- and-ballot, as a driver, enabled members to move from one discrete event state to the following stage.

These three investigators assumed that the assembly’s procedures were designed to deliver quantified scores of members’ preferences - gradatim, Jefferson’s phrase - during ordered discourse. Each score was posted at the watershed moment when one debate terminated and the next began. Even if a casual observer, following ballot after ballot, might not grasp the sig­nificance of any given ballot, to a member or an astute observer, the succes­sive quantifications yielded meaningful patterns of collaborative behaviour. Members wanted to see where they were. Taken functionally, members did what they did to keep track of what they were doing. Score-keeping bol­stered and supported the reflective or contemplative enterprise to which I now turn.