Muddled conspiracies: Maths, Science, and Vincentian politics, 1951-2015

by Dr Garrey Michael Dennie, St Mary’s College of Maryland

On December 9, 2015, three hundred constituents of Central Leeward awoke. Some had breakfast; some did not. They showered; some long, some short. Some were men; some were women. Some were young; some were old. They then left their homes, went to the voting booths, and they “mutilated” their ballots, or someone else did. And then contrary to Vincentian election law, the election officials counted the “mutilated ballots”.{{more}}

This story as told by the New Democratic Party (NDP) is, “The Conspiracy of the Mutilated Ballots.”

No one knows “The Three Hundred;” their ballots remain secret. But they were not alone in this. Sixteen other voters in Central Leeward did the same thing. Indeed, across the country 186 persons also “mutilated” their ballots. But this is unsurprising. Their invalidated ballots were but a decimal point – less than one third of one per cent of all votes cast in 2015. And this was fully in line with voter performance in St Vincent and the Grenadines since 1979. In other words, over the past 40 years, more than 99.5 per cent of Vincentians have cast their ballots properly. These numbers affirm that long before 2015, the Vincentian voter had mastered the art of voting.

That is, until December 9, 2015, when in the Central Leeward constituency they regressed. By definition, “the Conspiracy of the Mutilated Ballots” repudiates the idea of a Vincentian electorate superbly skilled in the art of voting. The NDP insists instead that voter incompetence destroyed 300 ballots placing them beyond the count. But the NDP may not have been aware of an absolutely astonishing fact: this error rate in one constituency exceeds the combined error rate of all the other constituencies. Indeed, it also exceeds the combined national error rate in every election since 1994.

No mathematical model exists to explain this staggering anomaly. Yet, it is upon this stony ground that the NDP seeds “the Conspiracy of the Mutilated Ballot.”

For NDP supporters, this was not the only conspiracy which robbed them of victory. In its Friday, 29 January 2016 edition, The Searchlight offered another. In a story it described as “extraordinary,” the Searchlight detailed Clive Lewis’ allegation that on the election day itself, some of the NDP’s election monitors joined in a criminal enterprise with ULP’s officials to “stuff ballot boxes.” He offered as proof a self-designed mathematical theory that allowed him to reduce the popular vote count by 20 per cent, thereby handing victory to the NDP.

This story, as told by Mr Clive Lewis, is “The Conspiracy of the Stuffed Ballots.”

Mr Lewis’ account, however, runs into a problem from which he cannot escape, “the Conspiracy of the Mutilated Ballot.” The problem is this: any claim of electoral malpractice located in a single constituency fatally undercuts Lewis’ claim of a broad conspiracy conducted across virtually all constituencies. If true, both claims would have the same effect – yield the election to the ULP, unfairly. But these claims are also mutually exclusive: both cannot be true at the same time. Out of this arises a simple question for the NDP to answer – which of these claims applies to the Central Leeward constituency: stuffed ballots or mutilated ballots?

These competing conspiracies reflect the deep disarray within the ranks of the NDP leadership tasked with making sense of a devastating defeat. Uniting around a single theory of electoral fraud, however, is virtually impossible. Two problems exist. First, everyone is well aware of both claims. The bell has been rung and cannot be un-rung. To disavow any of these claims would invite the unwelcome question, if one is false, how can we trust the other? Political expediency therefore dictates that the NDP defends both claims – even if in so doing they erode the foundations of each.

Holding to the line of a double conspiracy also spares the NDP the pain of addressing the second problem: which of these two theories should we discard, the stuffed ballots theory or the mutilated ballots theory? In truth, none is preferable to the other, because they are both verifiably false.

Sixty-five years of Vincentian electoral data provide indisputable proof that these claims of conspiracy are like foams on the ocean waves, drifting, indeterminate, and unanchored to any truth. To dig through this data is to excavate nuggets of knowledge upon which one can construct a deeply detailed account of Vincentian electoral history. For example, we know that in 17 elections held during this period, at least 115,000 Vincentians registered to vote. We know they cast 662,529 votes. And we know that in 65 years, only 9,789 ballots were declared as invalid, a tiny 1.47 per cent. Armed with this data we can reach towards a more mathematically informed historical analysis to illuminate the darkness where conspiracies reside.

We therefore return to Lewis’ “stuffed ballot theory.” He postulates a mathematical relationship between the voters’ registration and the age eligibility lists. If the number of voters on the voter registration list exceeds the number on the age eligibility list, then the voter registration list is inflated. Therefore, to get to the “true”