Sir David Butler 1924-2022

[swanarcadian]

There have been comments and tributes made on the Rest in Peace thread, but it really does seem appropriate that this great man, the father of psephology, should have a thread specially dedicated to him.  I hope the forum will agree?

Sir David was very helpful to me on the one occasion I corresponded with him, when he was already in his nineties.  He was able to direct me straight to the source of the BBC's notional 1970 general election results by constituency, which were used to calculate the swings on for the BBC's February 1974 election night programme (the ITN worked out separate figures).  I'd been chasing this important data for some time. Had he not lived into his nineties, I might still be chasing it today.

His name was given to the concept of swing, the Butler swing, which is the increase in the Conservative vote plus the decrease in the Labour vote divided by two. This was what we saw in Times Guides to the House of Commons up to 1979. The custom was that positive swings were swings to the Conservatives, and negative ones to Labour. I don't think anything cynical should be read into why that was - it seems logical if you think of the two parties next to each other on a chart, with Labour on the left, and the Conservatives on the right.  Having said that, there is naturally some speculation as to his own loyalties.  He seemed saddened when George Brown lost Belper, but read into that what you will.

His time appearing on BBC's election night specials, from the age of just 25 - 1950 to 1979 - seemed to go hand in hand with the age of consensus - Butskellism.  In fact he was a cousin of Rab Butler.  I think it is a pity that the swingometer was abandoned in 1983 and 1987; in fact, it was unforgiveable.  The fact that it was the SDP-Liberal Alliance, not Labour, who were in second place in about two thirds of Tory seats at that time, is by the by - I can only assume that was the reason it was abandoned - although the sad death of Robert Mackenzie, another enthusiast of the swingometer, in 1981, might have been an influencing factor.  

He lived to a great age, but it is still sad that he is gone.  Rest in peace.

[xenon]

A rather nice obituary piece for him was the last item on tonight's late-night Radio 4 news.

[The Bishop]

Actually his political leanings were fairly well known, ie lifelong moderate Labour - the "mask slipping" just for once when George Brown lost in 1970 was no accident. It was to his credit that he generally managed to keep them so well hidden, though.

[nodealbrexiteer]

Nov 10, 2022 11:22:36 GMT The Bishop said:

Actually his political leanings were fairly well known, ie lifelong moderate Labour - the "mask slipping" just for once when George Brown lost in 1970 was no accident. It was to his credit that he generally managed to keep them so well hidden, though.

this is a great thread.I think in a Sunday times interview during the 1992 campaign he revealed a sympathy for the Aliance in the 1980s and Michael Crick's biography does say he voted Labour in 1950

[grahammurray]

The most remarakable story I ever heard about him was where he picked up his skill for numbers and working out swings on the spot.

He said that his real boyhood passion was for cricket and just after the war there wasn't any, but there was a General Election and so he transfrred his abilities to voting habits. At cricket matches he could recalculate batting and bowling averages after every ball without the need for a reference book or writing materials.

[nodealbrexiteer]

Lile swanarcadian I did email him when trying to get to the bottom of the mystery of the mess the BBC made of their October 1974 exit poll,he did kindly reply but apologised that his memories of the 1970s were fading, nice to get a reply though

[Sibboleth]

Nov 10, 2022 11:22:36 GMT The Bishop said:

Actually his political leanings were fairly well known, ie lifelong moderate Labour - the "mask slipping" just for once when George Brown lost in 1970 was no accident. It was to his credit that he generally managed to keep them so well hidden, though.

I think the mask slipping just once in literally days of footage, for a few seconds, for George Brown, is, to use a bit of Internet English, undeniably and utterly Based. There's something very old-school BBC, very Reithian, about the whole thing though: his views were not a guarded secret, but viewers would never have known what they were. Even the incident with Brown's defeat would not have been picked up by more than a literal handful of people out of the millions that watched at the time.

[Sibboleth]

Crossposted from the RIP thread:

Nov 9, 2022 0:23:39 GMT Sibboleth said:

Elections before Butler were subject to serious analysis, but most of what was published about them tended to aim more for colourful narrative than sober analysis, and the basically Marxist framework* used by Butler and others associated with the 'Nuffield School' was genuinely groundbreaking at the time.

_{*In an analytical sense rather than an ideological one.}

The approach of the Nuffield School is very unfashionable in Political Studies/Science these days, but it did at least explain the politics of the post-war decades pretty well even if not periods before or after, which is more than can be said for the record of 'New Political Science' that has dominated since the 1980s for its own period.

[nodealbrexiteer]

Nov 10, 2022 13:45:45 GMT Sibboleth said:

Crossposted from the RIP thread:

Nov 9, 2022 0:23:39 GMT Sibboleth said:

Elections before Butler were subject to serious analysis, but most of what was published about them tended to aim more for colourful narrative than sober analysis, and the basically Marxist framework* used by Butler and others associated with the 'Nuffield School' was genuinely groundbreaking at the time.

_{*In an analytical sense rather than an ideological one.}

The approach of the Nuffield School is very unfashionable in Political Studies/Science these days, but it did at least explain the politics of the post-war decades pretty well even if not periods before or after, which is more than can be said for the record of 'New Political Science' that has dominated since the 1980s for its own period.

I also like the consistency book to book of the sections(i know there were one or 2 changes/expansion) such as the newspapers,polls, appendix-always love the section on the workings of the electoral system/bias

[Sibboleth]

Nov 10, 2022 14:23:06 GMT Andrew_S said:

Why is it unfashionable these days? I've read some of these studies and can't think what the problem with them might be.

It became unfashionable in the 1980s for the same reasons that other scholarly frameworks rooted in a Marxist understanding of class did: it could not be used to explain the politics of the period, and so the suspicion grew that perhaps it hadn't really done so during the decades when it appeared to be able to. In the humanities this was also the period of the 'linguistic turn' and other postmodern experiments, but in the social sciences the movement was more towards an obsession with quantitative data, and so the extremely empirical fundamentals of the Nuffield School (and similar classic 'materialist' approaches towards electoral studies, such as those associated with the Norwegian political scientist Stein Rokkan, or the 'geographic' approach associated with the French historical geographer André Siegfried) became unfashionable as well: the view developed that it was all a little bit crude. And after a certain point it remained unfashionable because it was an approach to the subject that people used to do, and so was seen as a little old hat. Structuralism, Materialism and Marxism were all out: Big Data was in.

Yet... there was very little that Butler could not explain in the heyday of his approach, something that can hardly be said of the leading political scientists of the past four decades, who seem to find themselves bemused and searching for explanations at every election.

[johnloony]

Nov 10, 2022 12:36:04 GMT nodealbrexiteer said:

Lile swanarcadian I did email him when trying to get to the bottom of the mystery of the mess the BBC made of their October 1974 exit poll,he did kindly reply but apologised that his memories of the 1970s were fading, nice to get a reply though

What was wrong with the exit poll in 1974?

[nodealbrexiteer]

Nov 10, 2022 21:27:43 GMT johnloony said:

Nov 10, 2022 12:36:04 GMT nodealbrexiteer said:

Lile swanarcadian I did email him when trying to get to the bottom of the mystery of the mess the BBC made of their October 1974 exit poll,he did kindly reply but apologised that his memories of the 1970s were fading, nice to get a reply though

What was wrong with the exit poll in 1974?

See my posts in interesting electoral facts and elsewhere

[nodealbrexiteer]

Nov 10, 2022 21:56:07 GMT Andrew_S said:

Nov 10, 2022 21:27:43 GMT johnloony said:

What was wrong with the exit poll in 1974?

From memory it was predicting a very large Labour majority which as you know ended up being 3 seats.

Yes Labour majority of 125. However the authors of the BBC seat forecast program say they didn't do the seat calculation describing it as a 'non computer', 'swingometer based' prediction. I was trying to find out who did it (One of the authors Dr Clive Payne said it was done by Prof Michael steed but in correspondence with him he didn't remember doing it) and to deepen the mystery (as i've wrote elsewhere on this forum) Payne and co author of the program Prof Philip Brown ran the exit poll data through their system and got a pre result forecast from it(bizarrely not broadcast by the BBC) which had a 28 seat error in the forecast of the Labour lead over the Conservatives-a forecast(along with similar ones their program produced for the Feb 1974 and 1979 elections) I tried to obtain from the BBC using a freedom of information request and a request to the BBC archives(these failed-long story)

[tonyhill]

Michael Steed was responsible for chapter on the statistical analysis of the General Election in many of the Nuffield Studies. I'd be surprised if he had made such a mistake of such magnitude - anyone with experience of looking at data is bound to ask themselves whether what the figures are implying is plausible, and given the pre-election polls these clearly weren't.

[The Bishop]

Nov 10, 2022 13:36:23 GMT Sibboleth said:

Nov 10, 2022 11:22:36 GMT The Bishop said:

Actually his political leanings were fairly well known, ie lifelong moderate Labour - the "mask slipping" just for once when George Brown lost in 1970 was no accident. It was to his credit that he generally managed to keep them so well hidden, though.

I think the mask slipping just once in literally days of footage, for a few seconds, for George Brown, is, to use a bit of Internet English, undeniably and utterly Based. There's something very old-school BBC, very Reithian, about the whole thing though: his views were not a guarded secret, but viewers would never have known what they were. Even the incident with Brown's defeat would not have been picked up by more than a literal handful of people out of the millions that watched at the time.

Its a while since I have watched the 1964/66 GE reruns, but I think he allowed himself to editorialise a bit over the Smethwick business as well - "good riddance to bad rubbish" was what he basically said on the latter occasion IIRC

[nodealbrexiteer]

Nov 11, 2022 6:55:55 GMT tonyhill said:

Michael Steed was responsible for chapter on the statistical analysis of the General Election in many of the Nuffield Studies. I'd be surprised if he had made such a mistake of such magnitude - anyone with experience of looking at data is bound to ask themselves whether what the figures are implying is plausible, and given the pre-election polls these clearly weren't.

The parties themselves couldnt see much of a pro Lab swing in the campaign, Wilson was talking about a 12 seat maj privately, the Tories thought they would do well to hold Lab to 20-30 maj so 3 must have been pleasant surprise. One of 2 late polls did show Lab quite comfortably ahead so there was a small chance some in the BBC studio(if not Steed) thought it might be plausible

[Defenestrated Fipplebox]

Nov 11, 2022 11:32:18 GMT nodealbrexiteer said:

Nov 11, 2022 6:55:55 GMT tonyhill said:

Michael Steed was responsible for chapter on the statistical analysis of the General Election in many of the Nuffield Studies. I'd be surprised if he had made such a mistake of such magnitude - anyone with experience of looking at data is bound to ask themselves whether what the figures are implying is plausible, and given the pre-election polls these clearly weren't.

The parties themselves couldnt see much of a pro Lab swing in the campaign, Wilson was talking about a 12 seat maj privately, the Tories thought they would do well to hold Lab to 20030 maj so 3 must have been pleasant surprise. One of 2 late polls did show Lab quite comfortably ahead so there was a small chance some in the BBC studio(if not Steed) thought it might be plausible

Tories were very very pessimistic! 

[nodealbrexiteer]

Nov 11, 2022 11:37:05 GMT Defenestrated Fipplebox said:

Nov 11, 2022 11:32:18 GMT nodealbrexiteer said:

The parties themselves couldnt see much of a pro Lab swing in the campaign, Wilson was talking about a 12 seat maj privately, the Tories thought they would do well to hold Lab to 20030 maj so 3 must have been pleasant surprise. One of 2 late polls did show Lab quite comfortably ahead so there was a small chance some in the BBC studio(if not Steed) thought it might be plausible

Tories were very very pessimistic! 

lol i've reedited it to 20-30!

[nodealbrexiteer]

Nov 11, 2022 0:43:30 GMT Andrew_S said:

Nov 10, 2022 22:08:45 GMT nodealbrexiteer said:

Yes Labour majority of 125. However the authors of the BBC seat forecast program say they didn't do the seat calculation describing it as a 'non computer', 'swingometer based' prediction. I was trying to find out who did it (One of the authors Dr Clive Payne said it was done by Prof Michael steed but in correspondence with him he didn't remember doing it) and to deepen the mystery (as i've wrote elsewhere on this forum) Payne and co author of the program Prof Philip Brown ran the exit poll data through their system and got a pre result forecast from it(bizarrely not broadcast by the BBC) which had a 28 seat error in the forecast of the Labour lead over the Conservatives-a forecast(along with similar ones their program produced for the Feb 1974 and 1979 elections) I tried to obtain from the BBC using a freedom of information request and a request to the BBC archives(these failed-long story)

If I'd been responsible for it I would probably have trouble remembering it.

Prof Steed did speculate to me after watching Humphrey Taylor's interview on the morning after on the BBC parliament replay that maybe Harris themselves did the calculation(though one of Sir Robert Worcester's books says Taylor was 'categorically' reassured beforehand the Harris data wouldn't be used to make this sort of calculation). I don't know if Robert Waller might be able to say how much when he worked on exit polls the seat calculations were done by Harris and how much by separate teams at ITN(BBC's approach was have your own team it seems)

[Sibboleth]

Nov 11, 2022 10:32:31 GMT The Bishop said:

Its a while since I have watched the 1964/66 GE reruns, but I think he allowed himself to editorialise a bit over the Smethwick business as well - "good riddance to bad rubbish" was what he basically said on the latter occasion IIRC

He was also obviously a little concerned by David Pitt's defeat at Clapham in 1970. But I don't think either really count as even a minor indicator of partisanship, just of something political in a small 'p' sense if that makes sense. Very easy to imagine a Liberal or even a moderate Conservative saying similar things in both cases.