4 ms·

Political view aside, if you repeatedly let the people vote to be independent, then eventually they are going to be independent. It is similar to the principle

Political view aside, if you repeatedly let the people vote to be independent, then eventually they are going to be independent. It is similar to the principle of p-hacking. A fundamental problem here is that getting independent is easier then uniting. If you want to make it a habit to be able to vote for independence, then you also need to make it a habit to vote for uniting countries. Only then the statistical equilibrium can be reached where the outcome represents what the people want, on average.

voakbasda 4y ago
What if people want smaller countries in general? Maybe we should stop pretending that everyone can get along.
- KolenCh 4y ago
  There's nothing wrong with that want. But you're not following the logic—the current asymmetry favors flowing from one direction than the other. Imagine 4 years from now the people want to get back. It is incredibly hard to do that. And also you're not understanding the p-hacking analogy here. Imagine there's a truth, objective fraction of people wanting independence (which is not true, that's why above we talk about that may change at later time), assuming that p = 0.49. If a vote can probe this true value with no error, there's no problem in making it a habit to vote even annually. But since voting itself has error-bar (we still assume the true p is time-independent so far, but the errors come from randomness such as if you can make it to the voting center), if you keep voting, once upon a time you're going to get 51%. So P(leave | keep voting N times until leave) -> 1 almost surely. That's logically why you shouldn't allow such kind of vote appears too often. And whenever it happens, make sure the campaigns are really educating people to do the best decision, not spreading misinformation. The point about the asymmetry is another matter, imagine after you leave, you can keep voting too, P(join | keep voting N times until join) -> 1 almost surely again. But P(leave) average over time (say hundreds of years) will give you back 49%. (Now we have another matter, say if 30% of the population want to leave, do we really want 30% of the time the countries are split? Instability is not good here.)
mike_hearn 4y ago
The interesting thing here is that official SNP policy is to try and join the EU if they get independence, which is a form of uniting with other countries (or more accurately, uniting with a single semi-country). It seems however very unlikely that in the event of independence the SNP would grant the Scottish people a referendum on joining the EU.

voakbasda 4y ago

What if people want smaller countries in general? Maybe we should stop pretending that everyone can get along.

KolenCh 4y ago
There's nothing wrong with that want. But you're not following the logic—the current asymmetry favors flowing from one direction than the other. Imagine 4 years from now the people want to get back. It is incredibly hard to do that. And also you're not understanding the p-hacking analogy here. Imagine there's a truth, objective fraction of people wanting independence (which is not true, that's why above we talk about that may change at later time), assuming that p = 0.49. If a vote can probe this true value with no error, there's no problem in making it a habit to vote even annually. But since voting itself has error-bar (we still assume the true p is time-independent so far, but the errors come from randomness such as if you can make it to the voting center), if you keep voting, once upon a time you're going to get 51%. So P(leave | keep voting N times until leave) -> 1 almost surely. That's logically why you shouldn't allow such kind of vote appears too often. And whenever it happens, make sure the campaigns are really educating people to do the best decision, not spreading misinformation. The point about the asymmetry is another matter, imagine after you leave, you can keep voting too, P(join | keep voting N times until join) -> 1 almost surely again. But P(leave) average over time (say hundreds of years) will give you back 49%. (Now we have another matter, say if 30% of the population want to leave, do we really want 30% of the time the countries are split? Instability is not good here.)

mike_hearn 4y ago

The interesting thing here is that official SNP policy is to try and join the EU if they get independence, which is a form of uniting with other countries (or more accurately, uniting with a single semi-country). It seems however very unlikely that in the event of independence the SNP would grant the Scottish people a referendum on joining the EU.