By Michael Gallagher

1. Previous post didn’t contain a prediction, but did observe that there were many considerations that could lead electors to a No vote while the benefits of a Yes outcome were uncertain and contested. Margin of defeat for the two proposals has taken everyone by surprise.

2. Many of the Yes advocates virtually held their noses while calling for a Yes; at pains to stress that the wording on offer wasn’t really what they wanted, damned it with faint praise as ‘better than nothing’ or ‘at least a step in the right direction’. All the conviction and passion seemed to be on the No side.

3. Voting on the two proposals very similar; correlation between Yes % votes at the two was 0.96. Unlikely that many voted No to Family and Yes to Care, so approx 68% voted No to both, 26% voted Yes to both, 6% voted Yes Family and No Care. As proportions of the electorate, only one in seven backed Family proposal and one in nine backed Care amendment. No large discrepancies between the two votes; only in Dublin Central (11.5%) did Yes to Family exceed Yes to Care by more than 10 percentage points, while smallest differences were in Carlow–Kilkenny and Wexford (less than 2 points in each).

4. As expected, quite strong relationship between these votes and 2018 abortion referendum, which correlates 0.81 with vote on Family proposal and 0.71 with vote on Care proposal. And familiar liberal constituencies (Dun Laoghaire, Dublin Bay S) recorded highest Yes votes, and familiar conservative constituencies (Donegal, Roscommon) especially high No votes. But correlations well short of perfect (ie 1.0), showing that while liberal–conservative cleavage was important in structuring these votes, there was more to it than that.

5. Comparing Yes to abortion in 2018 with Yes to Family in 2024, overall drop in support was 34%. Largest drops in support, at 44%, were in predominantly working-class constituencies (Dublin MW and Dublin NW). Middle-class Dublin Bay S and Dun Laoghaire among the smallest drops (less than 30%). Evidently relationship between class and voting wasn’t quite the same as in archetypal conservative–liberal referendums.

6. Role of citizens’ assemblies open to debate. Without a CA, this or a previous govt might have proposed a simple amendment to remove reference to mothers’ “duties” in the home (Art 41.2.2), with strong prospects of success. Bringing so many other issues into it was, with benefit of hindsight, a recipe for rejection. Question arises as to whether CAs should be confined to issues that don’t involve resource allocation; a focus on almost any area of public policy is almost bound to lead to conclusion that more money should be spent on it, but in real world all such decisions involve trade-offs.

7. Also questions about role of constitution. Should it aim to recognise, affirm, pay tribute to all those who contribute to society; or confine itself to setting out framework within which political process operates?