Philosophy Now Forum

Obvious Leo wrote:... Even a simple overview of the UTM would be beyond the scope of a forum such as this but in a nutshell the UTM is a computer which programmes its own input. ...

But it's not? Turing's UTM is a theoretical computer that can simulate any other TM and the input required from it's own tape, it's not simulating itself?

The ET is a moral, mythical issue. It has fuck all to do with maths.

Arising_uk wrote:

Melchior wrote:Why does anyone take this thread seriously?

Have you bothered to read Nietzsche?

Yes, and I have translated four of his books into English. Why do you ask?

Arising_uk wrote:Yes but in his sense exactly the same thing is going to be repeated, not some eternity for all possible outcomes to be computed.

The only reason (in his thinking) that anything is going to be "repeated" or "recur" is because of the eternity factor. He thinks the "eternal" part of "eternal return" or "eternal recurrence" is important to making his case. As you can see, though, he's mathematically wrong.

How he thought it I thought, so instead of choosing a moral action with the idea that it would become a universal law for all if acted upon, i.e. pretty much a restatement of the 'do unto ...' of Christianity, one chooses as though one will be doing that action in the light of having to do it endlessly along with all it's consequences.

I see what you're saying.

You might find it interesting that Alan Wood has done the best recent work on Kant. You'll find a rather different view of Kant's metaethics there. But the traditional (perhaps now debatable) view of Kant is that he uses the principle of universalizability to ground his argument, not the "do unto" principle. And in neither view, Wood's or the traditional one, does Kant suggest that one is actually going to be "doing that action...endlessly." That's merely his heuristic device, not a real-world postulate for Kant.

Additionally, the traditional view of Kant is that he rejects all thought of "consequences" in framing his ethics. That's a point traditional Kant scholars make very forcefully, and with some textual evidence from Kant -- and if their view is right, then no "consequences" could be included in any ethical heuristic.

So the parallel you suggest would be questioned very hard by both by more modern Kant scholars and the traditionalists as well. But thank you for sharing that insight. It may turn out to be something you can eventually show, but it will likely be a bit of a hard sell in view of the general expert consensus on Kant.

Arising_uk wrote:

Obvious Leo wrote:... Even a simple overview of the UTM would be beyond the scope of a forum such as this but in a nutshell the UTM is a computer which programmes its own input. ...

But it's not? Turing's UTM is a theoretical computer that can simulate any other TM and the input required from it's own tape, it's not simulating itself?

No. Definitely not. In modern information theory the UTM does not simulate itself but quite literally creates itself by generating new information. It emphatically does NOT merely process an existing dataset in a Newtonian and mechanistic way. It is in this respect that the analogy between the cosmos and the UTM is applicable and also the analogy between mind and the UTM is of great significance. As it happens it is also the hottest theoretical field in the science of computation because neurally networked evolutionary algorithms will comprise the software infrastructure for the next generation of computers. Obviously it is also crucial to the further development of AI because a computer which can merely execute a programme can never be definable as intelligent.

Obvious Leo wrote:No. Definitely not. In modern information theory the UTM does not simulate itself but quite literally creates itself by generating new information. ...

Could you point me towards some links as I've been out of the field for quite a while now and this doesn't sound like the Turing UTM to me.

It emphatically does NOT merely process an existing dataset in a Newtonian and mechanistic way. ...

Not sure about the 'Newtonian' nor the 'mechanistic'(?) but a Turing UTM does not program itself, what makes it universal is that it can simulate any TM by holding the inputs and the machine description on its own infinite tape but I can't see how it can program itself?

It is in this respect that the analogy between the cosmos and the UTM is applicable and also the analogy between mind and the UTM is of great significance. ...

I can't understand how, are you saying that the cosmos is a TM running on a UTM?

As it happens it is also the hottest theoretical field in the science of computation because neurally networked evolutionary algorithms will comprise the software infrastructure for the next generation of computers. ...

Well if they are digital then I can see that a UTM could simulate them.

Obviously it is also crucial to the further development of AI because a computer which can merely execute a programme can never be definable as intelligent.

Then a UTM won't be the solution but I stand to be pointed to the new ideas about them.

Immanuel Can wrote:The only reason (in his thinking) that anything is going to be "repeated" or "recur" is because of the eternity factor. He thinks the "eternal" part of "eternal return" or "eternal recurrence" is important to making his case. As you can see, though, he's mathematically wrong.

Can't understand how as the 'eternal' part applies to the process eternally repeating? As it doesn't depend upon 'eternal' as a factor in producing the conditions just that they will repeat eternally.

I see what you're saying.

You might find it interesting that Alan Wood has done the best recent work on Kant. You'll find a rather different view of Kant's metaethics there. But the traditional (perhaps now debatable) view of Kant is that he uses the principle of universality to ground his argument, not the "do unto" principle. And in neither view, Wood's or the traditional one, does Kant suggest that one is actually going to be "doing that action...endlessly." That's merely his heuristic device, not a real-world postulate for Kant.

Additionally, the traditional view of Kant is that he rejects all thought of "consequences" in framing his ethics. That's a point traditional Kant scholars make very forcefully, and with some textual evidence from Kant -- and if their view is right, then no "consequences" could be included in any ethical heuristic.

So the parallel you suggest would be questioned very hard by both by more modern Kant scholars and the traditionalists as well. But thank you for sharing that insight. It may turn out to be something you can eventually show, but it will likely be a bit of a hard sell in view of the general expert consensus on Kant.

I take your points and agree that consequences were not in Kant's E&M and understand that he wished Reason to rule E&M but for myself I think the Universalizability Principle was obviously based upon the idea of 'Others doing it as well'(if you like) which for me is really the 'Do unto ...' inverted. Now Nietzsche's ER was, for me, the opposite of such a thought as it was not based upon 'Other's doing it at all', just oneself deciding whether one could live with having to do it forever and ever and ever... . Although I think both wished to make Reason the factor to behave.

Cheers for the thoughts.

Arising_uk wrote:Cheers for the thoughts.

My pleasure. Thanks for sharing.

Immanuel Can wrote:
You might find it interesting that Alan Wood has done the best recent work on Kant. You'll find a rather different view of Kant's metaethics there. But the traditional (perhaps now debatable) view of Kant is that he uses the principle of universalizability to ground his argument, not the "do unto" principle. And in neither view, Wood's or the traditional one, does Kant suggest that one is actually going to be "doing that action...endlessly." That's merely his heuristic device, not a real-world postulate for Kant. .

Are we conflating Kant with Nietzsche?

Where is Kant saying this ; this that Nietzsche is so obviously saying with the ET.

Melchior wrote:Yes, and I have translated four of his books into English. Why do you ask?

My apologies. As I was surprised that you thought Nietzsche should not be taken seriously but you did say 'this thread' so my mistake, although we weren't exactly taking it too seriously.

Hobbes' Choice wrote:

Immanuel Can wrote:
You might find it interesting that Alan Wood has done the best recent work on Kant. You'll find a rather different view of Kant's metaethics there. But the traditional (perhaps now debatable) view of Kant is that he uses the principle of universalizability to ground his argument, not the "do unto" principle. And in neither view, Wood's or the traditional one, does Kant suggest that one is actually going to be "doing that action...endlessly." That's merely his heuristic device, not a real-world postulate for Kant. .

Are we conflating Kant with Nietzsche?

Where is Kant saying this ; this that Nietzsche is so obviously saying with the ET.

Five days later and he can't find the quote.
Tut tut.