ask yourself: what test can we make that would disprove the theory?
maybe i’m just not smart enough to come up with one, but i can’t conceive of one. an untestable, that is, an undisprovable hypothesis, is an empty tautology. or, at least modern epistemologists and critical rationalists have treated them this way.
maybe disprovability isn’t a necessary facet of sound scientific theories. i tend to agree with popper, though.
Okay, the test would be that we have first past the post (single winner elections, like for president, or local electorates with single candidates elected, not proportional voting, which is better), produce elections with a spread of votes across many candidates, and don’t consistently trend towards two.
This is definitely testable and disprovable, it’s just that the outcome is overwhelmingly the case I have described, the spoiler effect leading to two dominant parties. There may be outliers and times where a third candidate does win, but these are the overwhelmingly rare exceptions.
we need to define terms like “consistently” and “trend”. but even once you do that we still have the problem that you’re already explaining away exceptions. this theory is not disprovable because there is no outcome that you would say actually disproves it. you would say we just need more data.
I’m not explaining away exceptions, they’re called outliers. In any set of data there will be deviations. When I want to plot some viscosity data and get a few random points on my chart that don’t line up with the rest of the curve, I’m still very confident that my curve is close to being accurate, as long as I have enough data points.
We have enough data points on first past the post elections.
For it to be disproven you would show first past the post elections don’t have to two party systems in the vast majority of cases (which isn’t the reality).
Now, you can try and handwave this away by saying, “oh but that’s what people were TOLD TO BELIEVE, so you can’t prove it”. That’s why we have not just the correlation to rely on, we have maths.
And you can’t (I hope you don’t) really disagree that you either have many candidates, who then win with less than a majority, or two parties, which then necessarily means the third smaller candidates can’t win, and so people then vote for one of the larger parties so their vote counts. That’s the binary state of affairs, there are no other options, the reality of maths doesn’t allow for anything else, the votes add up to 100% ¯_(ツ)_/¯
I’m not saying we need more data though, we have the data, plurality voting overwhelming results in two party systems. This is disprovable and I’m totally happy to change my mind based on the evidence and data.
I’m not straw-manning, you said before with regards to looking up the spoiler effect “I have. it’s not a natural phenomenon, it’s a story that the media tells.”
Apologies if I misunderstood what you were saying there.
ask yourself: what test can we make that would disprove the theory?
maybe i’m just not smart enough to come up with one, but i can’t conceive of one. an untestable, that is, an undisprovable hypothesis, is an empty tautology. or, at least modern epistemologists and critical rationalists have treated them this way.
maybe disprovability isn’t a necessary facet of sound scientific theories. i tend to agree with popper, though.
Okay, the test would be that we have first past the post (single winner elections, like for president, or local electorates with single candidates elected, not proportional voting, which is better), produce elections with a spread of votes across many candidates, and don’t consistently trend towards two.
This is definitely testable and disprovable, it’s just that the outcome is overwhelmingly the case I have described, the spoiler effect leading to two dominant parties. There may be outliers and times where a third candidate does win, but these are the overwhelmingly rare exceptions.
we need to define terms like “consistently” and “trend”. but even once you do that we still have the problem that you’re already explaining away exceptions. this theory is not disprovable because there is no outcome that you would say actually disproves it. you would say we just need more data.
I’m not explaining away exceptions, they’re called outliers. In any set of data there will be deviations. When I want to plot some viscosity data and get a few random points on my chart that don’t line up with the rest of the curve, I’m still very confident that my curve is close to being accurate, as long as I have enough data points.
We have enough data points on first past the post elections.
For it to be disproven you would show first past the post elections don’t have to two party systems in the vast majority of cases (which isn’t the reality).
Now, you can try and handwave this away by saying, “oh but that’s what people were TOLD TO BELIEVE, so you can’t prove it”. That’s why we have not just the correlation to rely on, we have maths.
And you can’t (I hope you don’t) really disagree that you either have many candidates, who then win with less than a majority, or two parties, which then necessarily means the third smaller candidates can’t win, and so people then vote for one of the larger parties so their vote counts. That’s the binary state of affairs, there are no other options, the reality of maths doesn’t allow for anything else, the votes add up to 100% ¯_(ツ)_/¯
this is a strawman. you’re not dealing with what I actually said.
I’m not saying we need more data though, we have the data, plurality voting overwhelming results in two party systems. This is disprovable and I’m totally happy to change my mind based on the evidence and data.
I’m not straw-manning, you said before with regards to looking up the spoiler effect “I have. it’s not a natural phenomenon, it’s a story that the media tells.”
Apologies if I misunderstood what you were saying there.