An indyref is much more likely to take place in 2019 or 2020 than next year

First things first: a Survation-Watch update.  Stuart Dickson has spotted an article in the Sunday Post about a Scottish poll conducted by Survation between the 27th and 30th of November.  It was commissioned by an organisation called 38 Degrees and asks whether EU powers over devolved areas should be transferred direct to Holyrood after Brexit (as would happen automatically if it wasn't for the UK government's power grab in the Great Repeal Bill).  The results are predictable, albeit devastating for the UK government - almost two-thirds of the Scottish public want the powers to go to Edinburgh, not London.

The poll looks for all the world like a bolt-on question added to a full-scale poll commissioned by a different client.  So it looks like Survation have carried out the Scottish poll they promised - but so far the voting intention results are nowhere to be seen.  Maybe those will still appear over the coming hours or days - and if they don't, there'll be the tantalising possibility that maybe, just maybe, the client is holding them back due to disappointment with the numbers.  Something like that happened a couple of months ago, as you may recall.

What did turn up last night was a UK-wide Survation voting intention poll, which caused a sensation because it puts Labour in overall majority territory for the first time in years.  The Scottish subsample shows the following: SNP 34%, Labour 29%, Conservatives 23%, UKIP 8%, Liberal Democrats 7%.  That's not to be sniffed at - any sort of SNP lead in a poll that puts Labour on 45% at UK level is pretty good going.  Across all firms, twenty-five of the last twenty-seven subsamples have shown the SNP in first place.

To turn to a different subject, this month's issue of iScot magazine features Peter A Bell and myself making entirely opposite points about the timing of the second independence referendum, and doing so with equal confidence.  For reasons that I find hard to pin down, Peter is certain that the referendum will be held in September 2018, whereas I think a 2018 referendum is close enough to being impossible as makes no difference - although of course I do firmly believe that it should (and probably will) be held before the current mandate expires in May 2021.  Yesterday, Peter claimed that Nicola Sturgeon had dropped a heavy hint that 2018 was going to be The Year in remarks to the SNP National Council.  Others disputed that she had done any such thing, and I'm not surprised, because I'm baffled as to how Peter thinks a 2018 referendum is even feasible in practical terms.

If the UK government were prepared to pass a Section 30 order without fuss upon request, then of course holding a snap referendum at almost any time would be a trivial matter.  But all the indications are that they intend to persevere with the "now is not the time" tactic for a few years, which means any vote in 2018 would have to be of the consultative variety, held without Westminster's permission.  That makes it harder to do on a tight timetable, because the following steps would have to be followed -

1) Nicola Sturgeon would need to allow ample time for a renewed Section 30 request to be considered, in order to demonstrate that she isn't just going through the motions in making it.  She'll want to establish in the public mind that she bent over backwards to reach an agreement, and wasn't hellbent on going it alone.

2) She'll then need time to explain to the public why an "unauthorised" consultative referendum has become necessary - not least because the media will point out she's changed her own mind on that subject.

3) The Scottish Parliament will then need to go through the process of legislating for a referendum, which from a legal perspective will not necessarily be that easy.  The Presiding Officer's legal advisers will have to be satisfied that the proposal is within the parliament's current powers, which will require very careful wording.  If that is achieved, it'll still be important that the SNP are not seen to railroad such controversial legislation through parliament - there'll have to be proper time for reflection and debate.

4) There may then be legal challenges to overcome.

5) Last but not least, Ms Sturgeon will need to allow an appropriate length of time for the campaign proper.  I don't think anyone would want a campaign anything like as long as the one that preceded the 2014 vote - but any attempt to cram it into a few short weeks would look like a cynical tactic and might backfire badly.

I would suggest the last realistic date for a 2018 referendum is late October - anything beyond that would lead to concerns about the weather.  (Alex Salmond initially proposed that the first indyref should be held on St Andrew's Day 2010, but fate proved that wasn't such a great idea - there was heavy snow and traffic chaos on that day.)  So basically we're talking about a little under eleven months from now.  If Ms Sturgeon got the ball rolling right now or very soon, there might be just about enough time.  But that clearly isn't going to happen.  There'll be no announcement this side of Christmas, and probably not until well into the New Year.  That means the 2018 option will be effectively timed out, and we'll be looking at a probable date of 2019 or 2020.  I'm not at all downhearted by that, because it wasn't very long ago that siren voices within the SNP were seemingly trying to use the general election result as an excuse to "park" any talk of a referendum until beyond the 2021 election.  They now appear to have comprehensively lost the internal debate.

Does the near-impossibility of a 2018 vote mean that the referendum will be held after Britain officially leaves the European Union?  Quite possibly, yes.  Is that sub-optimal?  In my opinion, absolutely.  But if we wanted a September 2018 referendum, the time for making that case was during the summer.  It seems to me we lost that particular battle, but won the wider war to keep a pre-2021 vote firmly on course.  Speaking personally, I'm more than satisfied with that outcome.