// Is there ever any point in holding a referendum at all, then? //
It depends, is my unhelpful short answer.
In a post that it seems I was drafting but never finished, I also mentioned the forgotten referendum of the last decade, the one on whether we should switch to the Alternative Vote or stick with FPTP, held in 2011. In any case, I have no wish to revisit the question posed in 2011, but my point would be that it was a curious exercise from start to finish. It posed a very narrow question -- a choice between two electoral systems out of dozens -- but seems to have been used to address public sentiment on any kind of electoral reform. It also posed a question, for that matter, that even its proponents weren't that interested it. The LibDems will have wanted a proportional voting system, which AV is not, but apparently settled on that as a compromise. It also didn't really get that many people engaged -- turnout was 40%, or thereabouts, as compared to the whopping 85% who voted in the 2014 IndyRef, or the 72% who voted in 2016.
My real point is that the 2011 Referendum was held, not to resolve the issue, but to kill the issue. Most politicians were actively against it, and I'm fairly sure this also explains partially the choice of alternative offered -- different enough that it can be used to represent all electoral reform, whilst also not really addressing any of the concerns of reformers; thus, it was likely to be rejected, and it serves to preserve the status quo for a generation or so, that of course suits in particular the Tories and Labour to the favour of anybody else.
This is also true of the initial Brexit vote. The referendum in 2016 did, in fact, follow a few minor tweaks to our relationship with the EU -- it hardly matters what these were, now, and in any case they clearly didn't come close to addressing the concerns of EUsceptics -- and was also offered in part during the election campaign of 2015 to try and head off UKIP's challenge. It was offered not to resolve the issue of our EU Membership, then, but to kill it. In 2011 that succeeded; in 2016 it failed spectacularly.
The same, I would argue, is also true of the 2014 Referendum. Pressure was building in Scotland, partly due to two majority victories on the bounce for the SNP, and it seemed politically expedient to test the waters and undercut their drive for Independence by giving the SNP what they wanted -- it is obvious, however, that there was no desire for the Referendum to be successful on the part of the people who needed to allow it.
The same could also be held to be true of the only other Referendum in our history, the one in 1975, that Labour offered in order to strengthen their campaign in the previous election. I'm not as familiar with the history of that one, but it seems likely that it was also proposed in part to address divides within Labour itself, just as the 2016 one addressed Tory divides.
All of this leads to the conclusion that Referendums in this country are, and always have been, conducted dishonestly, and never with the intention of deciding the issue in question. As long as that remains true, then the only real "point" to Referendums, in the UK at least, is to try and hold one Party or the other, or a Coalition, together.
That shouldn't be the end of the story, of course (TBC...)