Select Committee on Procedure Minutes of Evidence


Memorandum by the Clerk of the House

INTRODUCTION

  1.  This paper is intended to assist the Committee in its inquiry into the Rules governing the Election of a Speaker. It falls into three Parts. The first is a survey of the occasions since 1817 when elections to the Chair were contested. The unusually long sweep of this part of the evidence is deliberate. The House's understanding and expectations of the Speakership are deeply engrained, and it is important, before coming to conclusions on what change may now be necessary, to see how and in what circumstances these expectations came into being. For example, the reasons advanced by Peel as long ago as 1841 for not claiming the Speakership for his victorious party conditioned the nineteenth and to some extent twentieth century view of the relationship between elections to the Chair and party politics.

  2.  The second Part deals briefly with the change made in the method of electing a Speaker following the Procedure Committee study of the matter in 1972, also with the further thought given to the topic in 1996.

  3.  Finally, I have included in the third Part some observations which I hope the Committee find relevant to their substantive decisions, not of course suggesting what these decisions should be but simply drawing attention to the problems which will demand solution.

PART I

CONTESTED ELECTIONS TO THE SPEAKERSHIP

  4.  There were a group of contested elections in the early part of the nineteenth century, just as modern political parties and the modern Speakership took shape. They occurred in 1817, 1833, 1835 and 1839: and all but one of them involved Mr Speaker Manners-Sutton.

  5.  Manners-Sutton was a Tory. His first election in 1817 was opposed by the Whigs[1]. He was re-elected unopposed in the new Parliaments of 1819, 1820, 1826, 1830 and 1831[2]. In 1833 his re-election to the Chair was unsuccessfully opposed, (from the back-benches, by Radical and Irish Members)[3]. In 1835, however, the Whigs blamed Manners-Sutton for intriguing to bring about their dismissal from office. He failed to secure re-election to the Chair by ten votes, 306 to 316, being succeeded by Mr Speaker Abercromby. It is significant however that not all the Whigs voted against Manners-Sutton's continuing in the Chair: some at least were opposed to rejecting a Speaker who had presided over the House in the previous Parliament[4]. This is the last time on which a Speaker was returned at a General Election and not re-elected to the Chair.

  6.  Mr Speaker Abercromby was re-elected in 1837: he resigned in 1839. The Whigs then carried by 18 votes the election of their candidate, who became Mr Speaker Shaw Lefevre[5].

  7.  When in 1841 the Tories returned to power, they re-elected Shaw Lefevre. Sir Robert Peel said that he would vote for the returning Speaker because it was "most in conformity with precedents in the best periods of the history of the country". Only in 1780 and 1835, he argued, had that course not been followed. Peel went on:

    I did not think it necessary [in 1835 and 1837] that the person elected to the Chair, who had conscientiously and ably performed his duties, should be displaced because his political opinions were not consonant to those of the majority of the House . . . I do not mean to say that the principle ought to be held as an inflexible and invariable rule. It may be perfectly competent to any man to object to a re-election on the ground of neglect of duty or conduct which may show an individual to be incompetent to the office . . . [but Shaw Lefevre had] by his ability, impartiality and integrity secured the confidence of the House[6].

  8.  Between 1841 and 1895, three Speakers[7] were elected in 13 elections. None of these elections was contested.

  9.  In 1895, as the Parliament elected in 1892 was coming to an end, the Speaker in office resigned. In contrast to the many decades of uncontested elections, this election was hotly disputed for a number of reasons. In the first place, there had been five consecutive Liberal Speakers, and the Conservative thought it was their turn to have the prize. Secondly, the Conservative Opposition expected to win the forthcoming election, and resented what they saw as a Liberal attempt to use their majority to elect a Speaker in effect for the next Parliament. They did not guarantee to re-elect Gully, the Liberal candidate, if they became the government[8] and once he had been elected to the Chair they opposed him at the General Election[9]. Finally, there were evidently doubts about the candidate's personal suitability. Gully was nevertheless elected to the Chair by eleven votes[10]. In the event, the Conservatives did not carry out their threat to replace the Speaker in the new Parliament. Indeed, Gully was re-elected on a Conservative motion.

  10.  There was no contested election for the Chair between 1895 and 1951—a period almost identical in length with that separating the previous two contested elections, 1841 and 1895. In those years, the House elected four Speakers[11] in 18 elections. In that period, it is noticeable that in 1906, a previously Conservative Speaker was re-elected without opposition by a Parliament in which the Liberals had a very large majority; and in 1945, the newly-returned Labour House re-elected a Speaker who had been first elected to the House as a Conservative[12].

  11.  In 1951, with a Conservative government having recently taken office, there was a contest for the Speakership between Mr W S Morrison and Major Milner, who had been a Deputy Speaker in the previous Parliament[13]. There was a misunderstanding between the Conservative and Labour parties, the former believing that the Labour party would second Mr Morrison, while the latter were insistent that the claims of Major Milner should be considered[14]. Mr Speaker Morrison was not opposed when he sought re-election to the Chair in 1955. The misunderstanding of 1951 shows how hard it will probably always be to elect a new Speaker, when there is also a new Parliament and a new government.

  12.  Again, there followed a number of years when elections were not contested, in this case from 1951 until 1992, leaving aside the 1971 case when there was a division but no real contest (see paragraph 14). In that period there were five Speakers[15] and 13 elections to the Chair.

  13.  In the years in question, however, the House was not wholly at ease with the procedures for election. The principal reason for concern was the perceived lack of consultation by the government, which was felt to exclude the Opposition or Members on the back-benches. Although there was no election when Sir Harry Hylton-Foster became Speaker in 1959 , the then Leader of the Opposition (Mr Gaitskell) expressed "strong dissatisfaction with and disapproval of the methods by which in practice this election has been conducted"[16].

  14.  In 1971, this dissatisfaction took the form of the proposal of a second candidate who, when called on to submit himself to the House, asked that his name should not go forward. However, since a motion to elect him had been made and seconded, a division took place on the Question that the first Member proposed, Mr Selwyn Lloyd, take the Chair as Speaker. (It should be recalled that under the old procedure, there was no division if only one name was proposed). Comments by Members in the debate included the proposal that there should be "a system of nominations followed by a secret ballot"[17]. The outcome was an inquiry by the Procedure Committee, which led to a change in the method of election (see paragraph 20).

  15.  The elections of new Speakers in 1976 and 1983 were unopposed. The new procedures were not therefore put into operation until the nest contested election, in 1992. The amendment proposing Miss Boothroyd in place of Mr Peter Brooke was agreed to by 372 votes to 238, and the Main question was agreed without division[18]. Madam Speaker Boothroyd was re-elected in 1997[19].

  16.  In 2000, Madam Speaker Boothroyd resigned. As the Committee will recall, there were 12 candidates in the ensuing election—itself an unusual event: there had never been more than two names formally before the House in any contested election (though other candidates were in the background at preceding elections). Each of the 12 nominations was divided on, and the Speaker was not elected until nearly seven hours after the beginning of the sitting[20].

PART II

THE PROCEDURE COMMITTEE REPORTS

  17.  In 1972, the Procedure Committee made a report on The Election of a Speaker[21]. The immediate occasion of undertaking the inquiry was the events of 1971, and the nomination of the Member who was unwilling to stand, but whose name had been put forward simply in order to permit a division to take place (see paragraph 14).

  18.  The Committee observed that there had been two contested elections in the preceding 20 years (1951 and 1971) though the previous occasion was as far away as 1895. They recorded the recent increase in the number of complaints made in the course of the election of a Speaker, even uncontested elections. These centred on perceived lack of consultation, deficiencies in the procedure, and qualifications of the candidates (see paragraphs 13-14).

  19.  The procedure which was the subject of criticism was very ancient. The Clerk presided, standing and pointing to Members in place of calling their names aloud as a Speaker would do. He directed Members to the lobbies and appointed tellers if the House divided. He could not however, respond if a point of order were raised, or accept a dilatory motion[22]. If only one Member's name were proposed, it was not the practice of the House to divide against the nominee: the new Speaker was elected by acclamation[23].

  20.  The Procedure Committee recommended that if the retiring Speaker did not preside over the election of his or her successor, the Member with the longest unbroken period of service present in the House on the back-benches should do so. He or she should have "powers to enable him to exercise his functions in the House with proper authority"[24]. A question should be put on the election of a sole candidate. If there were more than one candidate, the old procedure of successive substantive motions, each in favour of a particular candidate, should be abandoned. Its place should be taken by the proposal of a question for the election of the first nominee, followed by amendments to the original motion bringing forward the names of alternatives, and finally—if no amendment was carried—by the putting of the question on the original motion. The argument was that by first voting on the second candidate (as an amendment to the original question) the House would gauge the degree of support he or she had attracted. If that proposition were defeated by a large majority, "the unanimous election of the first or of another candidate might thereby be secured, a result which your committee believe to be highly desirable"[25].

  21.  The Committee's recommendations were agreed to by the House in August 1972[26], in what is now Standing Order No 1.

  22.  Elections to the Chair were uncontested for 20 years. In 1992, there was a contested election, and Madam Speaker Boothroyd was elected when the amendment nominating her was carried, and the original question as amended was agreed to. The Procedure Committee looked into the matter once more in 1996, as part of a report on Proceedings at the start of a Parliament[27].

  23.  It had been argued that in 1992 "discord" on the government side of the House had prevented the emergence of a candidate who might have enjoyed greater support. The Committee had little sympathy with that view. They concluded that "the onus is plainly on the parties concerned to agree on their favoured candidate and failure to do so cannot be attributed to procedural obstacles"[28]. More generally, while the Committee admitted that the burden laid on the Father of the House to decide who is to catch his eye to move the first candidate was something of a weakness, the 1972 solution should not be changed. "There is in our view no better system and many worse"[29].

PART III

CHANGING THE METHOD OF ELECTION: SOME PRACTICAL CONSIDERATIONS

  24.  Before coming to a conclusion on whether any change in the method of electing a Speaker is required, the Committee may wish to look at the situation broadly and analyse the circumstances in which the demand for change has arisen. Paragraphs 25 to 28 attempt to do this: paragraphs 29 to 40 consider three particular issues to be addressed—canvassing, a ballot, and re-elections.

(i)  General analysis

  25.  Contested elections are relatively rare and the circumstances of each unique, so that the possibility of reviewing the matter in its broadest context has arisen infrequently. It seems clear however from the historical background in paragraphs 4 to 16 that concern about the procedure began seriously somewhere around the 1950's. It has been gathering speed over the last half-century, and was not halted by the reforms of 1971-72. If there is a secular trend, it seems to point to the timeliness of the present review.

  26.  The events of 23 October 2000 were not fundamentally at odds with the history of the previous half-century of argument: indeed, they can be seen as the logical extension of it. Yet as the same time, the background to that election was not that the House was still struggling to free the election of the Speaker from the constraints of government or party direction. The problem was that it seems to have won these battles but did not have a procedure, which was either adequate for that new situation or capable of dealing with numerous candidacies. The opportunity, which faces the Committee, is to devise a system of election, which both entrenches the freedom enjoyed by the House in the election of 2000, and is devised to cope with a broader range of possible circumstances.

  27.  At the same time, there are many features of the present arrangements which the Committee will wish to carry over into any new regime, such as the complete political impartiality of the Chair. This may need to be manifest in procedural terms, perhaps by codifying the need for proposers and seconders to be from different parties in the House (for which see paragraph 33). Whatever new procedure may be devised, there will be a decision to be taken on how far it will incorporate elements of the old. I have in mind, for example, ensuring that a newly-elected Speaker is supported, at the final stage of decision-taking, by as high a proportion of the membership of the House as possible.

  28.  Finally, it is axiomatic that any solution put forward by the Committee will have to be robust enough to take account of all situations—appointment of a new Speaker at the beginning of a Parliament; appointment of a new Speaker in the place of a Speaker who has died or who resigns in the course of a Parliament; and election in a new Parliament of a Member who had been Speaker in the previous Parliament.

  29.  The following paragraphs set out some detailed practical matters which the Committee may wish to bear in mind in coming to its substantive recommendations.

(ii)  Canvassing

  30.  The practice of the House, universally observed so far as I know until the last election to the Chair, was that the candidates did not openly seek support on the basis of their attitudes or policies in areas of concern to the Chair. It is not for me to say whether or not the old practice should be re-affirmed: it may in any case prove difficult to get the genie back into the bottle. There are however some considerations which the Committee may wish to give thought to.

  31.  Every Speakership is different, and the problems which confront each Speaker are not identical to those faced by his or her predecessor. Nevertheless, it is truism that the Speaker is the servant of the House and there are thus limits to what he or she can do. Canvassing surely ought not to lead to undertakings being given which, if implemented, would take the role of the Speaker in a direction or into an area not intended by the House.

  32.  These considerations seem to point to recommendations restricting in some detail the activities of candidates for the Speakership in presenting themselves to their "electorate", or (if the Committee prefer to return to the no-canvassing traditions) a clear indication that canvassing should not be the practice of the House.

(iii)  A ballot

  33.  In the course of the election of 2000, most of the critics of the current practice favoured a ballot. Many decisions would be implied in such a change on which it would not be proper for me to advise. However, the Committee may be assisted by a check-list of some of the choices facing them should it be agreed to recommend a balloting procedure. These include:

    —  arrangements for proposing and seconding candidates, including any proposals for a minimum number of signatures to make a nomination effective, and perhaps also demanding cross-party nominations;

    —  time-limits for nominations and casting of votes (and on this subject see also paragraphs 34 and 35);

    —  setting a threshold of support below which a candidacy would not be permitted to be included in the ballot proper;

    —  whether the ballot should be open or secret;

    —  what the voting system should be;

    —  whether an absolute majority of the House should be required for final election;

    —  determining the balance between the balloting process and any proceedings on the floor of the House, and (if there is to be a second stage on the floor) deciding what provision should be made for eliminating or run-off ballots to establish those who proceed to the second stage;

    —  establishing the procedure (if any) on the floor of the House, including whether motions should continue to be seconded, and whether the candidates should address the House;

    —  arrangements for counting votes; and

    —  the procedure to be used when there is only one nominee.

  34.  The timing consequences of a ballot to elect a Speaker (whether at the beginning of a new Parliament or in mid-session) are of crucial importance. In the case of a new Parliament, the present arrangements permit the Speaker to be elected as the first business on the first day. Thereafter the taking of the oath by the newly-elected Speaker allows him or her to preside over the swearing of the House at large. Under the present arrangements, selecting a candidate or candidates for the Speakership will usually have taken place informally on days before the new Parliament meets. A ballot, however, demands specific and reasonable periods of notice—of nomination and of the ballot process itself—and it may be considered that these should not begin until Members have gathered at Westminster[30].

  35.  The problems are less acute in the case of the sudden resignation or death of a Speaker in mid-session, perhaps in the middle of a sitting week, or on a time-scale similar to that of 2000. Nevertheless, a ballot ought to be capable in these circumstances of permitting the election of a Speaker without loss of much more time than the present arrangements.

  36.  Though I do not wish to express any view on whether or not a ballot arrangement would be appropriate, it may be useful to summarise the arguments deployed by the Committee's predecessors in coming to the conclusion that it would not. This, after all, is the case, which any favourable recommendation will have to meet. The Procedure Committee in 1972[31] noted arguments that a secret ballot reduced the influence of the government and party leaders. There was no reason (it was said) for the Speaker to be elected by a process different from that used by the major parties in electing their leaders. Opposite arguments were also put. A ballot might weaken the Speaker's authority and impartiality, and diminish continuity. It would accentuate divisions and not compose them. Balloted votes might still break down on party lines. The Committee's conclusions were grounded on the need for Members to be publicly accountable for their votes, on the likelihood that a ballot would lead to canvassing and lobbying which would be undesirable, and on the argument that a ballot would encourage opposition to a candidate who, with proper consultation, might have commended himself or herself to the House.

  37.  The Committee in 1996 identified some of the subsidiary decisions involved in adopting a ballot[32]. There would have to be advance declaration of candidates, including perhaps rules about notice and number of supporters. The choice of electoral system might also involve a decision about the majority required and rules for the elimination of the less successful candidates[33]. The Committee recommended no change in the then current procedure.

(iv)  Elections and re-elections

  38.  If any new procedure is to respect the long-standing tradition of the House however (and for other reasons which are touched on below) the Committee may think it wise to recommend a procedure in the last of the cases in paragraph 28, the re-election of a "sitting" Speaker, which differs from that of the other two.

  39.  Paragraph 5 shows that, for as long as the Speakership has been fully non-partisan—that is, going back to Shaw Lefevre in 1841—the House has not refused to re-elect to the Chair a Member who had been Speaker in the previous Parliament. Even in 1835, the last occasion on which this happened, there were those in the majority party who were unhappy at what was done. Any other arrangement would effectively put a newly elected Speaker "on probation"; and nothing could be more calculated than that to bring party politics into elections to the Chair.

  40.  The Committee may also wish to reflect on the wider consequences of throwing doubt on the existing presumption that a "sitting" Speaker will be re-elected to the Chair. When first elected, a Speaker resigns from his or her political party. A Speaker in office when a General Election is called fights that election as "The Speaker seeking re-election". If the House did not normally return the "sitting" Speaker to the Chair, and the defeated candidate chose not to leave the House (itself a major break with tradition over centuries) the outcome would be anomalous indeed.

  41.  At the same time, when such occasions arise, it perhaps should remain possible for the House (as under the "ancient" procedure before 1971 it was not) for a vote to take place on a re-election. The easiest way to provide for that would be to except such occasions from whatever new procedure is recommended.

  42.  A Member of the House (whether the Father of the House or the senior Deputy Speaker in the last Parliament who is returned to the new House is a matter for decision) would be in the Chair, armed with all the powers of the Speaker. The Question for the re-election of a "sitting" Speaker would be put. The proposer and seconder might speak, and the candidate might submit himself or herself to the House, in accordance with any recommendation of the Committee accepted by the House. The main point is that the Question would be unamendable. Thus the House would be able to reject a "sitting" Speaker—as it has since 1971—but there would be no head-to-head contest, and the long-standing tradition of the House would be respected[34].

CONCLUSION

  43.  There is wide admiration for the Speakership of the House of Commons. The Westminster model is found (to varying degrees) in the parliaments of the Commonwealth who stand in the "Anglo-Saxon" tradition, as well as in Parliaments which participate in a modified version of that tradition (such as the United States[35]) and those in Europe which have developed quite differently. That regard for the position of the Speaker in the House of Commons is, in my experience, indissolubly linked to its wholly non-partisan character, established over centuries. That seems to me to be the principle, which ought to survive any practical changes in the rules for the method of electing a Speaker.

W R McKay
19 December 2000


1   CJ (1817) 307, Manners-Sutton had a majority of 160. Back

2   CJ (1818-19) 8; ibid (1819-20) 108; ibid (1826-28) 8; ibid (1830-31) 6; and ibid (1831) 522. Back

3   (1833) 4. He had a majority of 210. Back

4   CJ (1835) 5. Back

5   CJ (1839) 244. The voting was 317 to 299. Back

6   Parl Deb third series, vol 59, c 7. Back

7   Mr Speaker Denison (1857-72): Mr Speaker Brand (1872-84): and Mr Speaker Peel-the Prime Minister's son-from 1884 to 1895. Back

8   Parl Deb fourth series vol 32, c 1389. Back

9   In fact the "Speaker seeking re-election" increased his majority, which few Liberal candidates did at that Election. Back

10   CJ (1895) 149. The voting was 285 to 274. Back

11   Mr Speaker Lowther (1905-21), Mr Speaker Whitley (1921-28), Mr Speaker Fitzroy (1928-43) and Mr Speaker Clifton-Brown (1943-51). Back

12   It is also the case however that, even though there was no contested election between these years, Members did criticise the conduct in a previous Parliament of a Speaker returned to the House and seeking re-election to the Chair. For example Parl Deb fifth series, vol 21, c 8-9; HC Deb (1921) vol 141, c 308-12; and ibid (1942-43) vol 387, c 617-21. Back

13   CJ (1951-52) 5. Mr Speaker Morrison's majority was 67-318 to 231. Back

14   HC Deb (1951-52) vol 493, c 12-15. Back

15   Speaker Sir Harry Hylton-Foster (1959-65); Mr Speaker King (1965-71); Mr Speaker Selwyn Lloyd (1971-76); Mr Speaker Thomas (1976-83); and Mr Speaker Weatherill (1983-92). Back

16   HC Deb (1959-60) vol 612, c 6. Back

17   HC Deb (1970-71) vol 809, c 9. Back

18   CJ (1992-93) 2. Back

19   CJ (1997-98) 2. Back

20   HC Deb (1999-2000) 355, c 1ff. Back

21   HC 111 (1971-72). Back

22   In defence of the ancient usage, no such circumstance had ever arisen. Back

23   This part of the procedure was not quite as old as the rest. Evidently before 1700 a question was put on the election of a sole candidate. Back

24   HC 111 (1971-72) paragraph 24. Back

25   ibid paragraph 25. Back

26   CJ (1971-72) 514-5. Back

27   HC 386 (1995-96). Back

28   ibid paragraph 19. Back

29   ibid paragraph 22. Back

30   In that context, it is relevant to note that the results of the 1997 General Election were fully available only on the Friday before the election of the Speaker on the following Wednesday. Back

31   HC 111 (1971-72) paragraphs 19-20. Back

32   HC 386 (1995-96) paragraphs 20-21. Back

33   The report also quoted a case in 1979 when the election of the Speaker of the Swedish Riksdag was declared invalid (because there had been a miscount) a fortnight after the successful candidate took the Chair. Back

34   Of course, provision would have to be made for an election by what might be regarded as the "standard" procedure-whatever model the Committee and the House favour-to follow any rejection of the "sitting" Speaker. A double procedure of this kind would need to be considered in the light of what is said about time in paragraph 33. Back

35   The American Speakership may be seen as sharing the Westminster tradition up to but not beyond the development here of a fully bi-partisan office. Back


 
previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2001
Prepared 15 February 2001