I did say provinces/territories. Northwest Territories, Yukon Territories and Nunavut all have 30-40k inhabitants (I'm not sure where you think I got that number, but you must've been giving me either too much or too little credit - I remembered it because I'd just been reading about the Canadian provinces and territories on Wikipedia a few days earlier), and they all have a seat. It's not as if you could afford to *not* give them a seat, that's what I was saying - but still they have a seat for a far smaller population than the other provinces have per seat.

That seems like a no-brainer.

And again with the no-brainer. But perhaps I was a bit vague with "favours regions with low population"; I meant no more than that the regions with low population were artificially favoured in the voting system, precisely because a strictly population-based seat division might result in untenable and unacceptable results such as entirely disenfranchising Nunavut. I hardly intended to say that Nunavut held any real power in Canadian politics - I'm not aware of any country that gives its thinly populated regions as disproportionate an amount of power as the United States, though no doubt there are a few.

I never intended to say (nor said) the regions had "a lot" of power compared to the cities.

But your explanation about the lack of stability is interesting, alright. Canada does conform to the general rules in its amount of parties, and in the BQ's being a regional party, which is the most obvious way of having more parties than one would expect (it's the same way in the UK), but evidently not in terms of stability.

It'd be interesting to learn what makes Canada such a special case... I'm inclined to say part of it must simply be the power of precedent and expectation - if you've seen a landslide as spectacular as that 1993 election you mention, you know it can happen again, and you're not afraid to cast your vote in such a way as to make a new landslide possible. Third parties in the US and UK lose tons of votes every election because voters do *not* believe such a thing is possible.

But there has to be more to it than that. Do Canadian MPs have much of a bond with their district - I don't necessarily mean in the pork sense, though that can be part of it?

I have to admit, looking at that election, one is almost inclined to reject the "single-seat constituencies promote stability" theory altogether. Going from 43% of the popular vote to 16% would've been a catastrophe for any party in any country, but in a multi-seat constituency system it wouldn't have been a killing blow like it was there, at least not in terms of number of seats. I guess it's more of a "it's less likely to happen, but when it does happen it's going to be an earthquake" deal, with Canada somehow having managed to get into a rhythm where it does happen on a regular basis.

Out of interest, what do you mean by "the Queen might see worse soon"?

The Dutch precedent isn't a pretty one. But then, that really was a one-man party, so when someone murdered their one man it was fairly obvious the party was going to disintegrate very fast. Just not fast enough for them not to be voted into the government, sadly.

Still, one does expect things to go back to normal at some point - not necessarily with the same two big parties you used to have, but still with two big parties. No?

Yeah. Political volatility has a way of getting worse and worse when it starts - I think it's grown worse in a lot of countries, even in the US though it's still rather limited there.