Almost invariably, where elections are referred to in the Constitution, it is emphasised that "voting shall be by secret ballot". Clearly, the intention is to avoid the contamination which might result from public voting. Obviously, an election in which each voter could look over the shoulders of others and citizens shouted out their preferences as they filled in their ballot papers, would not satisfy this constitutional requirement, writes John Waters.
But is this not, in a way, how modern elections are conducted? At least as important as the issues of this past month's referendum debate has been the emerging graph of public opinion sketched by the various polls published throughout the campaign. This is, in a certain sense, analogous to the kind of open-voting melee hypothesised above, though perhaps less worryingly with regard to voting intentions as in respect of decisions about whether to vote at all.
I am not arguing that the Yes outcome in the Nice referendum was inappropriately influenced by opinion polls. More interesting are the effects, visible from the comparative turnouts and results of the two Nice campaigns, of the workings of electoral complacency arising from impressions created by successive polls during a campaign.
Three weeks ago in this space, I confidently predicted a Yes vote at a time when many commentators were talking about the outcome being "in the balance". I made this prediction not because of confidence concerning the Government's persuasiveness, but on the basis of a common-sense awareness that there remains among the Irish electorate a high degree of support for the European project.
We do not have here the kind of split that occurs on these issues in, for example, Denmark. At most, our 2001 No vote was an opportunistic kick in the backside to the Government and/or complacent Yes voters by the slowly growing minority opposed to further European integration.
Put another way, last year's outcome was due to the complacency of large sections of the Yes majority who abstained because the referendum seemed unlosable. Yesterday's count showed a No vote relatively unchanged since last year, with the Yes vote massively increased due to the greatly-enhanced turnout.
This raises again questions about the advisability of running opinion polls for the duration of a campaign. Advocates of unrestricted polling invariably talk somewhat disingenuously about polls providing the electorate with "information". But this information is quite unlike the normal facts and background to the issues which emerge in a campaign, being in a sense absolute with regard to the actual outcome.
Although the pollsters emphasise that polls are not predictive, they are perceived to be so by a significant element of the electorate. It appears that some voters, reading the results of a final poll or observing the trajectories of support over the course of a campaign, become disinclined to vote by virtue of believing that " their" side is assured of success.
IN HIS book In Your Opinion, MRBI chairman Jack Jones provides an insight into the workings of such complacency in his account of the final MRBI poll of the 1997 general election. Commissioned by The Irish Times, it was conducted between 2 p.m. and 8 p.m. on polling day, in 1,400 homes throughout the State.
The published poll overstated support for Fianna Fáil by five points, understating that for other parties correspondingly. Mr Jones identified low voter turnout as the main factor in this divergence. Because of the time factor, some interviewees had voted before the MRBI representative caught up with them, while others had not. This necessitated a variation on the normal questioning procedure: respondents were first asked if they had voted and, if so, how; those who had not, having confirmed their intention to vote, were asked how they intended to vote.
Some days later, when the small print of the survey was studied, significant differences were identified between those who had already voted at the time of being interviewed and those who had not. The results among those who had already voted were virtually identical to the election results, whereas those who had not "yet" voted had given responses which resulted in the significant overstatement of the likely support for Fianna Fáil.
Since Fianna Fáil had been the consistent front-runners in polls conducted during the campaign, with the FF-PD coalition option showing a 10- point lead on the rainbow alternative just a week before polling, it is reasonable to assume that a significant element of intending supporters of the FF- led option finally neglected to vote because they "knew" how others would vote. This is the kind of thing the constitutional provision for secret ballots seeks to prevent.
Without opinion polls, this complacency could not develop. Although the polls indicated a gradual drift towards the No side during last year's Nice campaign, nobody could have predicted the defeat of the Nice Treaty from the final opinion poll, which suggested that 45 per cent intended to vote Yes and 28 per cent No. In the event, the referendum was defeated by 54 per cent to 46 per cent.
Had no polls been published during the last week of the campaign, it is probable that many more Yes voters would have come out to vote last year, as the fear of an irreversible No vote motivated them to do last Saturday. This strongly suggests that the opinion polls' technical elimination of the secret ballot is having a distorting effect on our electoral processes.