The Assembly election was a necessary follow-up to the referendum. But not all of those who voted in the referendum saw it that way. The turnout in the election was substantially down on that in the referendum. On the other hand, the referendum turnout was remarkable by any standard and particularly by recent Northern Ireland standards.
In fact there has been a significant decline in turnout in Northern Ireland since the early 1970s: from 77 per cent in 1970 to 68 per cent in 1997 in the Westminster elections and from 72 per cent in 1973 to 65 per cent in 1996 in the Assembly elections.
This withdrawal from the political process by a significant proportion of voters was dramatically reversed in the May 22nd referendum. Thursday's election failed to maintain that remobilisation but still produced a turnout that was five percentage points better than the 1996 Forum election and almost three points up on the 1997 Westminster election.
In terms of party support, however, Westminster elections provide a very unsatisfactory basis for the comparison of electoral support for the parties in Northern Ireland. This is so because of the incumbency advantage possessed by many UUP candidates in Westminster elections and, more importantly, the extent to which the DUP has failed to contest all the Westminster constituencies.
The implication is that the figures for support for the DUP in Westminster elections understate actual North-wide DUP support, particularly in comparison with support for the UUP. Assembly elections and district council elections provide a more satisfactory basis for tracing developments in party support.
Beginning with the 1982 Assembly election, those developments can be briefly summarised, taking the parties in the order of their performance in Thursday's election. The SDLP has made modest but fairly steady progress to become, by a whisker, the party with the largest share of the first-preference votes. It has achieved that position courtesy of a substantial decline in support for the Ulster Unionist Party. However, that support has been recent rather than long-term and is a function of that party's move towards a historic compromise on the fundamental issues that beset Northern Ireland. The point is that, having made significant compromises, it has survived as the leading unionist party.
The 18 per cent for the DUP is up on the 1997 council elections but otherwise in line with the levels of support it has won over the last decade. Note, however, that this is substantially down on its heyday support of the mid-1980s.
Sinn Fein has made the biggest gains but, coming only in 1996, those gains have been almost entirely due to the party's commitment to the peace process. Finally, the Alliance has been squeezed, the consoling point being that the squeeze has come not from voters fleeing from the centre but from other parties moving towards it.
The RTE/Irish Times exit poll provides vital clues as to why voters opted for the different parties on each side of the unionist/nationalist divide. Within unionism, the best predictors of UUP/PUP/UDP support as against DUP/UKUP support are: confidence that there will be long and lasting peace in Northern Ireland, a positive attitude to the working of the Assembly and middle-class occupation.
There is no sign that attitude to the maintenance of the Union plays any role in intra-unionist party choice. Nor is there any indication of generational differences within the unionist community.
The explanation of the division in party support in the nationalist community is quite different. True, attitude to ensuring that the Assembly is a success is a predictor of SDLP support, just as it is a predictor of support for the moderate unionist parties. Social class is also a common differentiating factor, although its effect within the nationalist community is much stronger than its effect in the unionist community. But, while supporters of the two blocs of parties on the unionist side are at one in their opposition to a united Ireland, Sinn Fein supporters can be distinguished from those of the SDLP in terms of their commitment to a united Ireland.
This underlines the well-established but often forgotten fact that attitudes to a united Ireland vary substantially within the "nationalist" community. The other factor that helps to predict Sinn Fein over SDLP support is age: ominously for the future of the now triumphant SDLP, there is a remarkable age contrast between its support and that of Sinn Fein, to the advantage of the latter. Comparing the results of this exit poll with that conducted for RTE Prime Time in the recent referendum indicates a modest increase in optimism among Northern Ireland voters as a whole. Overall, for example, the proportion believing that republican paramilitary violence will resume as it was before the ceasefire has declined from 19 to 8 per cent.
But this masks a quite dramatic change in the opinion of those who had voted No in the referendum. When they were voting in the referendum a clear majority of No voters (58 per cent) believed that republican violence would resume as it was before. But just one month later the proportion of serious pessimists among these No voters had declined by 34 percentage points to just 24 per cent.
Note, however, that the No voters have not become unbridled optimists. Rather they have moved in the main from believing republican paramilitary violence will resume as it was before to the less fearful but still cautious belief that it will remain a threat.
In a situation where perceptions and expectations, hopes and fears can have decisive effects on behaviour and on whether people move towards the centre or retreat into extremism, this striking shift in the expectations of those who opposed the agreement is quite an encouraging sign.
Another encouraging sign is that two out of every five referendum No voters want those they elected to "work to ensure the success of the Assembly as an institution". A block of more or less similar size opts for the conditional view that their elected representatives should "co-operate with the Assembly only when it advances their own principles". The crucial point, however, is that only 14 per cent of No voters want their elected representatives to "work to bring the Assembly to an end".
Richard Sinnott is director of CEEPA (Centre for European Economic and Public Affairs) at UCD and author of Irish Voters Decide: Voting Behaviour in Elections and Referendums since 1918 (Manchester University Press, 1995)