I've written occasionally here that I sometimes find political scientists better at working with political statistics than at original thinking about political questions. Emmett Macfarlane is one I exempt from that criticism. His books and his SubStack are frequently interesting and original takes on Canadian political and constitutional questions.
His most recent SubStack posting is about political "conventions." Not conventions meaning the big meetings that political parties hold, but conventions meaning the unlegislated but generally accepted rules that govern aspects of political life.
Here Macfarlane is mostly at pains to reject recent claims that there is or should be a convention to the effect that "the party that 'wins the most seats' in an election automatically get to form government."
The alleged convention at issue is, I think, an argument being put forward by Conservatives who fear that they will face an endless string of Liberal minority governments supported by the NDP unless they can create a "convention" that would require a governor-general to invite them to form government if they win more Commons seats than the second-place party -- even if the second-place party is in government, has support from other parties in the Commons, and wishes to test its support in the new House of Commons.
I don't disagree at all with Macfarlane on all that. The alleged "convention" makes no sense, and he explains why.
What does puzzle me is what existing conventions he would defend under his definition that "Conventions are binding political rules of behaviour rooted in reason. It is important to distinguish these from mere practice, precedent, or custom."
Binding? Beyond "reason," who or what binds us to them? It is sometimes suggested that responsible government, meaning the accountability of the executive ("the Crown, if you like) to the legislature, is itself a convention, and a binding one. But it is not. In Canada at least, it has since 1867 been guaranteed by bedrock text in the constitution itself.
What "conventions" are there in Ottawa that have the force of law without actually being in the law? I might suggest that a convention is a practice that tends to makes the running of a political system more efficient and practical, which does not contravene existing law, and which Parliament allows to persist.
Do we need the convention that compels the governor-general to allow the governing party to face Parliament after its slate of candidates has placed second in a general election to be binding? Hardly. The second and third parties, if they in combination hold a majority of seats in the new Parliament, will have no trouble overcoming the will of the G-G who defies the convention. They would toss out the new government that lacked majority support in Parliament as soon as Parliament met. Indeed, that's why the "convention" exists and is followed, whether "binding" or not. It would be foolish and impractical to impose on Parliament a doomed new government to which the new Parliament would not give majority support.
I may indeed be wrong about the limited need for binding conventions. Enlighten me.
P.S. I cannot help pointing out the title of this post is a riff on political scientist John C. Courtney's book Do Conventions Matter? But that book is indeed about the conventions held by political parties in hockey arenas and cow palaces, not about the rules of procedure about which Macfarlane is writing. As it happens, the need for them to exist is not binding either.
