A while ago I considered the story of referenda under the Labour Party, when they were used, and when they were not used.
Now, it's been suggested that I was overly harsh, but it transpires that the Labour Party line (say that no one wants independence, but don't ask the question, just in case they do) has been around for 30 years. One of the documents released under the Thirty Year Rule (why have the SNP not asked for this under FOI?) is a memo by Sir John Garlick, then head of the Government's constitution unit,and is quoted in today's Herald:
"The underlying idea of a referendum on independence for Scotland is to secure a firm No and thus weaken the Nationalists by being able to dismiss their policy ... at best, the move works only if there is a clear no with a high poll.
"Many factors might mitigate against this. The Nationalists might go for abstention, people might not turn out, there might be tactical voting, eg a significant yes element designed simply to convey a demand for more devolution. Or perhaps worst of all, there might be a significantly large and genuine yes vote."
There you have it: clear evidence that opposition to a referendum on independence is not based on a belief that people don't want it, so there's no point in asking the question. Rather, that opposition is based on a fear that people might want independence, so Unionist politicians dare not ask the question. But does it apply today?
From the perspective of either camp, the best outcome would, as Garlick tells us, be an overwhelming victory on a high turnout. A low turnout and/or a close result would allow the losing side to question the legitimacy of the outcome (the Welsh Assembly, for instance, started out from a weak position on account of a wafer-thin majority for the Yes camp, and a ~50% turnout), and would add extra acrimony to the events that followed.
But as for Garlick's concerns: I don't see either side boycotting a vote. The SNP are more likely to see any referendum as an opportunity to secure independence (that being the purpose of the Party, after all). Unionist parties will need a victory, both to maintain the constitutional status quo and enhance or at least maintain their reputation across the UK: none of them want to go into a future English election as 'the Party that lost Scotland'.
Turnout, which has been consistently low over the last decade is a real worry for both sides though, and the manner in which the campaign will be fought will probably make that more of a problem than less.
The idea of the 'tactical Yes vote' isn't one I accept though: the Tories tried in 1979 to offer people a 'tactical No vote', suggesting that rejecting Labour's proposals would allow the Conservatives to put forward something better. That was a lie, of course, but I just can't see people who want extra powers for Holyrood (but not independence) voting Yes just to scare politicians into offering them. If people don't want independence they are not going to vote for it in a referendum.
Then there is the final possibility, the one Garlick sees as 'worst of all': the suggestion that there might actually be a large, genuine Yes vote. Remember that there was a YouGov poll that showed a plurality in favour of independence, a poll for the Scotsman that gave independence 51% support, and one for the Telegraph that put it at 52%. That means that whether or not Garlick was right to fear the result in 1976, his successors have every right to fear it 30 years on.
And that, more than anything else, is why the Labour Party's position on a referendum is to stop it even from being held.