[CAFR-L] Some more laughs from the new reappointment process

Kent Peacock kent.peacock at uleth.ca
Sun Jan 15 03:16:00 MST 2012

Colleagues --

I've been trying to grasp the new rule (that Dan quotes below) from an 
algebraic point of view.

This is difficult, because there are at least two key ambiguities in the 
rule given.  First, it does not state whether "membership" means the 
number of persons who are supposed to be on the committee, or the number 
who actually happen to be present when the vote is taken.  Second, it 
does not state what the threshold is for---is it the threshold for 
reappointment, or the threshold for failure of reappointment?  That is, 
does the reappointment fail if the incumbent gets a number of no votes 
equal to the threshold, or one more than the threshold?

As Dan suggests, on the face of it the rule does not make mathematical 
sense.  I'm going to suggest a charitable interpretation here, and I 
want a consulting fee for it.

Forging ahead, for simplicity, let's take it that "membership" means the 
number of people who are supposed to be on the committee.  Define some 

M = membership

Y = number of yes votes (i.e., votes in favour of reappointment)

Now if for further simplicity we assume there are no abstentions, then M 
- Y = the number of negative votes.  But as with the old system, on my 
charitable reading the number of negative votes and/or abstentions 
doesn't matter.

Let T(M,Y) be the threshold function.

Now, the quoted rule seems to state the following:

T(M,Y) = M - Y + 1.

The explanation of how this is to be applied is garbled and ambiguous in 
the ways noted above.  However, let us for the sake of argument take it 
that the authors of the rule /meant/ to say that the number of yes votes 
must exceed the threshold. Read this way, it at least makes some sense.  
Perhaps part of the the intent was to have a nifty formula that would 
apply to all appointment committees, regardless of their size.

Okay, on my charitable interpretation, we have

Y > M - Y + 1.

That is,

Y > (M + 1) /2.

Say M = 10.  Then the threshold for reappointment is 11/2 = 5.5.  That 
is, 6 yes votes will yield a pass on a 10-member committee.  Perhaps, if 
my charitable interpretation of the GFC rule is correct, this is an 
attempt to rewrite history (or, at least, a certain piece of recent 

All the best,


On 14/01/2012 10:29 PM, O'Donnell, Dan wrote:
> Postings to this list are *Publicly Archived.* This is an unmoderated list and posters are solely responsible for the content of their messages.
> --------------------
> Hi all,
> Time for some more laughs from the new reappointment procedure from 
> the GFC bylaws... you know the one our well-paid senior admin has been 
> working on as a priority for the last year.
> You've probably forgotten, but the prompt for all this was the 
> President's perception that the old bylaws were unclear about how to 
> tell if somebody was reappointed or not. The old method of determining 
> if somebody was reappointed was found, in the case of deans, in 
> article IV M 4 (each office had a similar article indicating the 
> relevant test):
> "A motion to recommend the reappointment of the incumbent or the 
> appointment of a candidate shall require seven (7) affirmative votes 
> from among Committee members."
> Obviously we can't have that. So here's what they came up with instead 
> (IV D 3 in the new--this applies to all senior admin):
> "The Committee shall consider the information available to it, and 
> shall make either a
> recommendation that the incumbent be re-appointed or that the 
> incumbent not be re-
> appointed. For reappointments, the total number of negative votes 
> against the incumbent
> are determined. The threshold here is total membership minus votes for 
> appointment,
> plus one. If there are sufficient negative votes, then a 
> recommendation is put forward to
> not reappoint. Otherwise, a recommendation is put forward to reappoint."
> There are a couple of interesting bits to this. The first is that the 
> onus is no longer on administrators to prove that they have sufficient 
> positive votes to win reappointment, but rather on the committee to 
> show that they have enough negative votes to unseat the incumbent. And 
> since it involves only negative votes, you can no longer abstain in 
> any meaningful way: an abstention is not a vote against, so it doesn't 
> show an absence of support when we are trying to see if the 
> administrator has to go. In fact if you read "membership" to mean "the 
> total seats on the committee whether they are occupied or not," it 
> might even be impossible to resign in disgust without having this 
> count in favour of an incumbent.
> This is a pretty strong reversal of the normal burden of proof, and 
> I'm not sure it is necessarily a good idea to appoint as our chief 
> executive and strategic officers people about whom the best you can 
> say is there wasn't a clear majority against. But if you were an 
> administrator and saw your primary job as being to ensure the 
> reappointment of yourself and your colleagues, I can see how you might 
> not want to run the risk that you could end up with a majority against 
> you even under such tough circumstances. After all, if it turned out 
> you'd done a lousy job, you might be held accountable and then forced, 
> after a year or more of administrative leave, to go back in front of 
> the classroom.
> So what to do?
> Well, one solution might be to make the threshold for negative votes 
> mathematically impossible to meet. Then it wouldn't matter how bad a 
> job you'd done, you would never have to be held accountable. And 
> /that/, as Black Adder might say, would be a cunning plan.
> Our guys have done Black Adder proud.
> The key to all this is that line about the threshold: "The threshold 
> here is total membership minus votes for appointment, plus one." 
> Remember the threshold determines the number of /negative/ votes you 
> need to be removed from office. The formula "Membership - Positive 
> Votes + 1" creates a threshold you can never meet: it is 
> mathematically impossible to acquire a sufficient number of negative 
> votes to unseat a candidate under these conditions.
> In case you are finding this hard to get your head around, let's run 
> some examples. Say you have a committee of 10 people and 9 members of 
> that committee vote for the incumbent. To unseat this candidate, the 
> malcontents would need 2 negative votes (10 members - 9 positive votes 
> + 1 = threshold of 2). This might seem like a low threshold, but since 
> it is impossible to meet and the incumbent is popular anyway, we 
> probably shouldn't sweat the details.
> But let's say we have a more unpopular administrator whom only 4 
> people on our committee are prepared to support. The threshold for 
> unseating this administrator is now 7 negative votes (10 - 4 + 1 = 7). 
> This is also an impossibility, but it is also a bit of a more serious 
> problem, because we are about to reappoint a senior leader who can't 
> even command the support of a reappointment committee.
> But what about the absolute worse case: an incumbent whom nobody is 
> willing to support. What happens then? Well in that case our 10 person 
> committee needs to find eleven negative votes to put the University 
> out of its misery: 10 - 0 + 1 = 11. In other words, if we want to 
> remove an administrator that is so unpopular he or she cannot get even 
> a single positive vote from his or her reappointment committee, we 
> need to move our deliberations into another dimension.
> (By the way, this is where the problem with abstentions come in. Since 
> abstentions are not negative votes, they don't count against the 
> number needed for the threshold. But since they are not positive votes 
> either, they don't do anything to lower the threshold either. So a 
> committee with a lot of abstentions will actually have a higher 
> threshold than a committee with a lot of positive votes. Cf. the 
> following scenarios on a 10 member committee with 6 votes against: if 
> the remaining 4 votes are all positive, the threshold is 7 negative 
> votes to unseat (10 - 6 + 1 = 7); if 2 of those 4 non-negative votes 
> are abstentions, then the threshold rises to 9 negative votes to 
> unseat: 10 - 2 + 1 = 9. Of course since it is impossible to meet the 
> threshold under any circumstance, all this is moot.)
> This is a lower standard than any any of us have ever faced in any 
> context, except perhaps in winning our mothers' affections (my mother 
> was actually a little tougher than this: she loved me unconditionally, 
> but not under compulsion).
> I have my suspicions this may not be the route to ensuring excellence 
> in governance.
> -dan
> P.S. I really hope I'm wrong about this. But in case you are 
> wondering, I'm not sure I see any obvious typo or drafting mistake 
> here. A threshold of "positive votes + 1" (i.e. not subtracted from 
> the total votes) would have become progressively easier to meet as 
> popularity fell: if nobody voted in favour of a candidate, you'd only 
> need one negative vote to unseat him or her), but I can't see that 
> being the intention of the current administration. And "Total 
> Membership - Vote for [or against] the incumbent + 1" would be even 
> less characteristic of them, as it would usually give a veto to 
> anybody who voted negatively. Can anybody else see a way of reading 
> this that doesn't guarantee reappointment? Maybe we should just be 
> honest and decide we'll offer a 10 year first term.
> _______________________________________________
> cafr-l mailing list
> cafr-l at uleth.ca
> http://listserv.uleth.ca/mailman/listinfo/cafr-l

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://listserv.uleth.ca/pipermail/cafr-l/attachments/20120115/874c36b6/attachment.html

More information about the cafr-l mailing list