Sir

M. de L. Brooke reasons that British and US voters will elect a leader from an age group that they respect. On the basis of an analysis of the age at which British prime ministers first came to power, he concludes that “voters give the job to somebody whose attributes are honest signals of ability in the society of the day” (Nature 398, 102; 1999). Both reasoning and conclusion are flawed.

First, Brooke fails to ask whether the ages of victorious prime ministers differ significantly from those of their vanquished opponents. The two-party nature of the British and US electoral systems makes this a vital question. If they do not differ significantly, then voters cannot choose on the basis of age — the choice has already been made for them. If they do differ, then we need to know in what direction(s).

Second, it is at least questionable whether people vote for prime ministers at all. There is plenty of psephological evidence to show that it is the political party that commands people's loyalty. This can be demonstrated anecdotally. Winston Churchill was an immensely popular war leader, but he was heavily defeated in the 1945 general election. Jim Callaghan was more popular than Margaret Thatcher, but he lost in 1979. It is also significant that Brooke does not find an age trend in the US system, where there is a more direct relationship between voter and president.

It might be possible to rescue at least a version of Brooke's thesis by shifting the emphasis away from voter choice to party structure. That is, to claim that the reason that prime ministers have become progressively younger in the twentieth century is that there is a perception on the part of the major players in political parties that the voting public has less respect for older people than in the past. But there are still problems. First, the analysis would no longer be founded on a notion of ‘honest signals of quality’. And second, it would be to oversimplify the process whereby an élite rises to the top of a party. For example, Margaret Thatcher — in electoral terms, the most successful prime minister of this century — became leader of her party in 1975 in circumstances that had nothing to do with her age (and a lot to do with Sir Keith Joseph's committing political suicide by a speech in Birmingham; and Edward du Cann's withdrawing from the Conservative leadership contest at the last moment). And Tony Blair became leader of the Labour party only after the sudden death of John Smith.

As a philosopher, I frequently try to persuade colleagues of the merits of a sophisticated reductionism. Unfortunately, in this instance, the reductionism is not sophisticated and the conclusions are, therefore, likely to be erroneous. Incidentally, I wonder if your correspondent is aware that every US president elected between 1900 and 1968 was taller than his major opponent.