The voting system used by the Citizens’ Assembly may distort or misrepresent the views of its members, according to an independent critique of the advisory body.
The central claim is that while offering members a broad range of options on which to cast their vote, the structure of the assembly is such that these options must then be reduced to a binary “yes or no” outcome.
"In a nutshell, some voting procedures are not just inaccurate, they actually fail to achieve the desired outcome – to measure the collective will," writes Peter Emerson, director of the de Borda Institute which produced the analysis.
In one example highlighted by the institute, assembly members were asked to vote on the acceptability of abortion in different circumstances. A choice of five ways to vote was then sub-sectioned into three outcomes: in favour, against or abstain – and it is here Mr Emerson finds fault.
For instance, in a ballot on cases where a woman is raped, nine assembly members voted “never for this reason”; 23 for “up to 12 weeks gestation only”; 25 for “up to 22 weeks”; 25 for “no restriction as to gestational age”; and four for “prefer not to state an opinion”.
“At this point the chair gave her casting vote so that 22 weeks was the winner,” writes Mr Emerson. “They are, therefore, pretending that 25 votes plus that of the chair’s is a majority which it is not.”
The system suggests one of the outcomes is a majority when it is not, he continues, arguing the assembly could just as easily create a different set of binary divisions and get different results.
In this case, the assembly found 89 per cent in favour of permitting abortion in the case of rape, and 11 per cent against.
"They are trying to use a good method...different voting methodologies, but then they are stuck because the terms of reference say you have to have a majority [position]," Mr Emerson told The Irish Times.
He said while the majority voted in favour of terminations under differing gestational periods, the result here was presented as a binary 89 per cent “for” and 11 per cent “against”, and simplifies the complexity of the vote.
In response, the assembly explained three varying options in favour of abortion were grouped together in order to “facilitate reporting of the results” of the vote.
While the headline figure was used in reporting, the assembly also broke down the figures “to show how the citizens had voted in relation to gestational limits, or none”.
Mr Emerson, a long time campaigner on deliberative democracy, stressed that he supported the work of the assembly and his critique applied only to what he saw as an imperfect system of decision-making.
He advocates the use of a voting system known as the modified Borda count (MBC), which is designed to find the option that attracts the greatest approval, not necessarily a clear cut majority. He argues this is best suited to complex policy matters where there is not a clear yes/no decision to be made.
The institute contacted the assembly’s steering group with its analysis late last year, but Mr Emerson said the body appeared to change its mind on hearing a presentation.
The assembly said it had the guidance of former Dublin county returning officer John Fitzpatrick on the structure of its voting system.
"How the votes would be counted and how the majority would be determined" on each ballot was explained to the members by assembly chair Ms Justice Mary Laffoy.
“For the purposes of clarity and maximum transparency, the assembly voting process not only provided the percentage results of each vote, but also the absolute results in each case,” it said.
According to the assembly, the de Borda Institute had offered to present its steering group not an analysis of completed votes but views on “a broad range of topics” including the benefits of the MBC for future votes.
“There was broad consensus amongst the [steering group] about the value of consistency, and therefore it was agreed that we would not deviate from the mechanism we have had in previous meetings,” it said.