Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

> Things only exist when you can construct them.

This is exactly what I’m saying is presumptive! If constructivism is to earn the merit of being less presumptive by virtue of not assuming the existence of various things, it should also not assume the non-existence of those things.

Which, I think many visions of constructivism do earn this merit, but not your description of it.



The underlying problem is that constructivism and non-constructive reasoning are using the word "exists" (and, relatedly, the logical disjunction) to mean very different things. The constructive meaning for "exists" is certainly more intuitive, so it makes sense that constructivists would want it by 'default'; but the non-constructive operator (which a constructivist would preferably understand as "is merely allowed to exist"), while somewhat more subtle, has a usefulness of its own.


So having a different philosophy from you makes me presumptive?

What makes you presume that you have any business telling someone with different beliefs from you, what is OK to believe? You may believe in the existence of whatever you like. Whether that be numbers that cannot be specified, or invisible pink unicorns.

I'll be over in the corner saying that your belief does not compel me to agree with you on the question of what exists. Not when your belief follows from formalism, which explicitly abandons any pretense of meaningfulness to its abstract symbol manipulation.


No, that’s not what I said. Thinking you can determine a-priori that something that is logically self-consistent, cannot exist, if there is no reason that such a thing being physically instantiated would imply a logical contradiction, is the thing I think is presumptive.

Merely believing that such a thing (a halting oracle) doesn’t exist, isn’t something I meant to call presumptive, only believing that you can know a-priori (with certainty) that such things cannot exist.

I don’t claim that you are obligated to agree with me that they do exist. Someone who believes they don’t, but doesn’t believe they can know this as certain a-priori knowledge, would be no more presumptive than I am, and someone who is agnostic on the question of whether they exist would be less presumptive than I am.

Also, I disagree with your notion of “meaningfulness”. At a minimum, all statements in the arithmetic hierarchy are meaningful. The continuum hypothesis might in a certain sense not be meaningful.


> Merely believing that such a thing (a halting oracle) doesn’t exist, isn’t something I meant to call presumptive, only believing that you can know a-priori (with certainty) that such things cannot exist.

If you think that I was making that case, then you have misunderstood something important.

Constructivism is a statement about what kinds of arguments will convince me that things exist.

Could things exist that I don't believe in? Absolutely! There could well be a bank account with my name on it that I don't know about. Its existence is possible, and my lack of belief in it is no skin off of its back. But I still don't believe that it exists.

Similarly, the Platonists could be correct. There could be an omniscient God whose perfect mind gives existence to a perfect system of mathematics, beyond human comprehension. I have no way to prove that there isn't such a God, and therefore that there isn't such a perfect mathematics.

However the potential for such things to exist is a point of theology. I do not believe in their existence. Just as I do not believe in the existence of Santa. In neither case can I prove that they don't exist. And if you choose to believe in them, that's your business. Not mine.

There is nothing presumptive in my laying out the rules of reason that I will accept as convincing to me. There is a lot of presumption if anyone else comes along and tells me that I should think differently about unprovable propositions.

Now it happens to be the case that from the rules of reason that I use, I provably can't be convinced of the existence of certain things. That's a mathematical theorem. But the fact that I can't be convinced, doesn't prove that you shouldn't be convinced. You are free to be convinced of all of the unprovable assertions that you wish. And it is also true that on something like this, I have no way to convince you that it doesn't exist.

On meaningfulness, meaning is in the eye of the beholder. For example there are people who are willing to pay a million dollars for a century old stamp which was misprinted with the airplane upside-down. (See https://en.wikipedia.org/wiki/Inverted_Jenny to verify that.) They clearly find great meaning in that stamp. But I don't.

So again, you're free to find meaning in whatever you want. But you're in the wrong to object that I don't find meaning in what you consider important.


Ah, that I wouldn't call presumptuous. I just interpreted a previous statement you made as saying that one can prove that certain such things can't exist. So, I guess there was just a miscommunication (I guess I misunderstood.)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: