Philip Cowley is astounded and embarrassed by what the results of a survey of politics academics reveal. "That's ludicrous," spluttered the man sat next to me as the results were announced. "Absolutely ridiculous. Who are these people?"
"These people" were the distinguished academic members of the UK Political Studies Association, 300 of whom had taken part in a poll asking who was the greatest Prime Minister we'd never had. Denis Healey came top (18 per cent of members), followed by Roy Jenkins (12 per cent), Kenneth Clarke (10 per cent) and Neil Kinnock (8 per cent).
The list showed a depressing lack of historical knowledge or appreciation on the part of most political scientists. Some once great figures received only negligible support. Compared with Neil Kinnock's 8 per cent, did Rab Butler really deserve just 3 per cent? And Butler did well compared with Iain Macleod and Herbert Morrison, both of whom gained support from just 1 per cent of members each.
It wasn't just old-timers who got overlooked. John Smith didn't make the official PSA list; some 3 per cent wrote in his name as an alternative suggestion. But compare that with the 4 per cent who wrote in the name of Tony Benn or the 3 per cent who suggested Charles Kennedy. Think about that last one for a second: 3 per cent of professional political scientists think that someone with a drink problem would have made the best Prime Minister we'd never had.
The exercise was, of course, largely meaningless. There's often a reason why these individuals didn't make it - usually because they had a weakness that would have prevented them being very good at the job had they got there. The presentation of the findings included a picture of Healey sticking two fingers up. Ho, ho. Yet his behaviour, what Kenneth Morgan once described as "an aggressive, even bullying, personal manner", was part of the reason Healey didn't make it to No 10 - and why he would probably have been a disaster as Prime Minister had he got there.
It wasn't, however, the list of could-have-beens that gave my lunch companion the heebie-jeebies. It was the second part of the survey, which asked members which characteristics they thought should matter in determining prime-ministerial success - and then which characteristics did matter in reality.
The most striking feature of the responses was the difference between the two lists. Top of the "should" list came "direct experience of real life" (whatever that means), which 78 per cent thought should matter. But just 24 per cent thought that it did matter. Bottom of the list of things people thought should matter came the ability to perform well in Parliament (9 per cent), psychological strength and physical stamina (9 per cent) and the ability to perform well in television interviews (3 per cent). Yet two of these came top of list of things that political scientists thought did matter: the ability to perform well in Parliament was mentioned by a staggering 92 per cent, followed by the ability to perform on TV (72 per cent).
We don't have a comparable list of the variables that political practitioners themselves would see as crucial, but I suspect that the reaction of my lunch companion - and he had more than 40 years of experience advising politicians - would be pretty typical. Most of the judgments were ludicrous.
The idea that performance in Parliament is more important than someone's ability to perform well on TV is ridiculous. A satisfactory Parliamentary manner might be a necessary condition for political success these days, but only in the way that adequate toilet training is.
Equally ridiculous is the view that physical stamina doesn't matter (just 16 per cent said it did). Everyone who works in politics testifies to its breakneck pace. To make it requires reservoirs of energy that would shame most academics.
The other day, the last voice I heard before I fell asleep was that of Geoff Hoon, the Chief Whip, on some late-night radio programme, and I woke up listening to him the next morning, too. No one - other than Mrs Hoon, perhaps - should have to endure that, but the point is that he was up, briefed and ready to speak to the nation before most of us had drunk our first cuppa.
There are two possible reasons why the professionals could take a different view from the academics. The first would be because academics, with their more highly developed analytical edge and advanced sophisticated reasoning (not to mention their much broader historical time-frame) are able to cut through the smoke and identify the most salient issues. The second - and I'm afraid in this case much more likely explanation - is that most politics academics haven't got a clue about what really goes on in the political world.
The Political Studies Association should stop carrying out such surveys. They generate useful cheap publicity, but they also demonstrate embarrassing levels of ignorance among its membership. If this really is what PSA members think, then we'd be better off not publicising it.
Philip Cowley is professor of parliamentary government at Nottingham University.