Mister Deputy Vice-Chancellor,
I shall begin by welcoming the existence of this Report on voting arrangements, although I shall have rather more to say about its content later. There are a good few examples of voting in the University that has progressed in an unexpected way, and improving confidence in the decision-making process is something that I welcome. It is also notable that none of the speakers at the last Discussion on this matter was completely in favour of the status quo.
On 11th October last year, Professor Edwards perspicaciously opined on the matter of governance that “the right way forward is first to persuade the so-called conservatives that change is needed (which is not difficult if you have a good case, because they are not stupid), and then to get them to oversee the changes because they are the only people who understand enough to do it properly”1. To me, these Reports on voting give a tantalising view of what could be achieved if his words were taken to heart. I have already remarked that there seems at least to be consensus that voting arrangements can be improved – the “so-called conservatives” agreed that change is needed. They were even consulted on how to go about effecting a useful change. Unfortunately, the Review Committee then reverted to the form seen more than once in the administration before, and what we have instead is the constitutional equivalent of a Poldovian trying to sell me a Mars bar for twenty pounds sterling because he thinks I shall not notice the rip off in a foreign currency.
There is much contained within this Report with which I take issue, and it is difficult to know where to start, or whether my attempts at timeliness will result in the omission of something important. It occurs to me that I should quickly put on record my view that the proposals to change the Regulations are such that essentially all of my last speech on this matter2 remains valid. I shall not bore people by reading it again.
I am concerned that the content of the Report seems to suggest that there are still people involved in the consideration of voting arrangements who believe that it is necessarily possible to come up with a system that will avoid ambiguity in a given case. The whole point of Arrow's Impossibility Theorem is that this is just not true. Giving power to the Vice-Chancellor to decide what happens in a particular case does not solve the problem: it merely moves it somewhere else. I encourage those who are still not convinced of this to read the first chapter of Saari's Decisions and Elections: Explaining the Unexpected. I particularly note his semi-joke that “For a price, I will come to your organization to design your election procedure. You tell me who you want to win. After talking with the members of your organization to ascertain their preferences, I will construct a ‘democratic voting procedure’ which will ensure the victory of your candidate.” He goes on to illustrate with an example of how it is possible to ask the questions in such a way that a candidate can be rigged to win despite everybody preferring somebody else to that candidate.
This sort of thing is not just a curiosity that will never occur. Another good read for the layman, Archimedes' Revenge by Paul Hoffman, quotes an example analysed by William Riker, where in the United States legislature a bill failed following an amendment, where if the amendment had not been introduced the bill would likely have passed. In anticipation of an Old Schools claim that the circumstances in Cambridge are entirely different, I challenge the Council to allow the analysis of papers from real Cambridge ballots. If they are worried about people objecting to this use of used anonymous papers, they should put forward a Grace to test the opinion of the Regent House on the matter.
Returning to the proposal to give the Vice-Chancellor powers to decide how questions are presented to the Regent House, I am sure that whoever is drafting the Council's response to the remarks made in this Discussion will by the time of reading this far in the transcript of my remarks be thinking about writing something along the lines of how the personal integrity of the Vice-Chancellor means that there would never be a problem of the type I am suggesting. I regret that it seems necessary to make this observation, but I shall observe that a conflict of interests cannot be resolved by personal integrity. A man cannot serve two masters, no matter how good a man he is. The Vice-Chancellor is simply too close to the administration, and in particular to the Council, to be able to make such decisions purely in service of the Regent House. This is precisely the reason also why the Vice-Chancellor should not be the presiding officer for such ballots. The argument applies just as well to the Administrative Secretary. The Proctors are elected as the servants of the Regent House, and are therefore the correct people to serve the Regent House in such matters.
The only way to ensure that the manner in which proposals are presented to the Regent House is both as clear as possible and as fair as possible is to legislate before the event to that effect. Giving somebody power to change the rules at the last minute is neither clear nor fair. I should note that the Council attempt to justify their bland claim that “[no] dangers attach to such provision, and indeed … that significant advantage would arise if such arrangements were in place” by claiming that “Statute A, VIII, 8 already envisages that such authority can be given to the Vice-Chancellor.” Statute A, VIII, 8 says “In respect of Graces and amendments of Graces initiated under section 7 [this is the section that allows fifty-member Graces and the submission of amendments], the Vice-Chancellor shall have the power to rule inadmissible any Grace or amendment which directly concerns a particular person, and shall have such further powers as may be specified by Ordinance.” The Report seems correct therefore in claiming that the Statutes would not prohibit the proposed legislation; however, the implication that the Statutes expect or justify such legislation is such a good example of subtle spin that I suggest that whoever drafted it considers his talents wasted here and instead applies this evening for a post working for a Government minister.
Further, my objection to last-minute legislation extends to the suggestion that because the rules for voting are enshrined in Ordinance rather than in Statute, “if at some future time it seemed desirable, change could relatively quickly be made”. I do not disagree with this phrase solely because the rate of progress of the current debate makes a mockery of it: it is worrying that anybody should consider that such a fundamental part of the University's constitution as a central part of the decision-making process should be changed at short notice to avoid some unspecified inconvenience. (Incidentally, I consider the sentence elsewhere in the Report “This is a constitutional matter rather than one related to the voting system as such” to be meaningless sophistry: one cannot separate consideration of voting from consideration of a sizeable part of the University's constitution.) If the Council truly consider that a defining characteristic of an Ordinance is that it is acceptable for the content to be changed at short notice, then I believe they should propose enshrining the procedures in Statute. Moreover, the phrase about future change could be read as a suggestion that, because it could be considered at another time, the Council need not consider a rather central aspect of voting arrangements in a Report on voting arrangements.
Given what I have said, those present may be forgiven for wondering how I do wish this matter to proceed. I regret that I am going to repeat my call for the voting system to be considered in the context of a statement of Arrow's Theorem, and a voting system selected on the basis of deciding which premise or premises of the theorem we should be most content to see dropped. The fundamental importance of the voting system being used in the decision-making process is such that no consideration of voting arrangements can be complete without a rigorous consideration of voting systems. Whilst I note that the Review Committee did consider alternative voting systems, it is not clear how this consideration was effected, and given that a major problem we are facing is that any non-trivial decision making mechanism can lead to anomalous results, my concern about the manner of the consideration undertaken should be understandable.
I believe it may also be worth reconsidering the practical arrangements for the enfranchisement of the Regent House. The current balloting system seems to be gaining a reputation as a “nuclear option”, because of the amount of delay and disruption caused by a ballot. Indeed, in the recent case of pay and grading the sole reason the Council gave for not using their powers to call a ballot was the existence of “significant practical disadvantages” and “uncertainty” because of the intended timescale of moving to a new system3. Arguments about whether matters should be presented to the Regent House at the eleventh hour aside, it seems that if a faster workable alternative could be instituted, this could be of benefit both to those who do not want proposals that will eventually be approved to be delayed, and to those who would like to see a lower threshold of uncertainty on a matter at which it became reasonable to test the opinion of the Regent House. It seems to me that a system of Regents voting in person on a given day, perhaps with the ability to appoint a proxy, may be one that should be given consideration. It would at least remove the need to distribute ballot papers and flysheets around the University, with the concomitant time delay and administrative burden, if everybody were to vote in one place.
Following on from this is the question of voting in the Senate. If postal voting in the Senate is to be considered, then I believe the issues of establishing voter identity and of establishing entitlement to vote become more complicated. I suspect one of the reasons that the list of members of the Senate is not fully up to date is that such a list is extremely difficult to maintain (although this is not to say that I should not find an effort to improve the current situation gratifying). I can see that voting in person at least makes it easier to deal with cases arising out of an imperfect list of voters. Perhaps a possibility that may be worth exploring is that of allowing members of the Senate a postal vote on a per-ballot basis, subject to confirmation of entitlement to vote and of address.
I have one comment that I should like to make regarding electronic voting, and that is to observe that the question occurs to me of whether the Proctors would necessarily have sufficient technical expertise to be able to scrutinise adequately on behalf of the Regent House a vote conducted electronically.
Finally, I wish to address the question of the Vice-Chancellor's ability to withdraw a Grace submitted to the Regent House. I fail to see any reason for the Vice-Chancellor to be able to withdraw a Grace except that there is something so substantially wrong with the Grace that continuing the process of its submission to the Regent House is patently ludicrous. In particular, I am saying that once the Council has decided to submit to the Regent House a particular course of action reflecting its view on a matter, it cannot officially (at least) change its view until this part of the process of consideration by the Regent House is complete, so that there is no reason to allow the Vice-Chancellor to withdraw a Grace particularly because a ballot has been requested or an amendment proposed. (Indeed, it may be inferred from this statement that I am uncomfortable with the Council being able to propose amendments to its own Graces when they go to ballot.) Further, the restriction on the Vice-Chancellor being able to withdraw a Grace once a ballot has been requested is a constitutional necessity to prevent the obvious abuse of the system of withdrawing a Grace on which a ballot has been requested with the intention of resubmitting the Grace in the same form in order to avoid a ballot. I am sure I am now risking the usual “only the paranoid would envisage the commission of that abuse of the system” response from the Council, but my point is that the only way to avoid any possibility of impropriety in this matter is to leave things as they are. I cannot really comment on the minutiae of the practicalities of this proposal, because there seems to be some ambiguity in the wording: I wonder whether the first “and” in the amended section is supposed to read “or”, and I wonder to what clause the last clause (beginning “within”) is supposed to apply?
Mister Deputy Vice-Chancellor, in conclusion I support the continued consideration of voting arrangements, but believe there is much more consideration and review necessary.
1. Reporter 2005–6, p.85
2. Reporter 2004–5, p.140
3. Reporter 2004–5 pp.1002–3