The existence of natural abstractions is entirely compatible with the existence of language games. There are correct and incorrect ways to play language games.
Dialogue trees are the substrate of language games, and broader reality is the substrate of dialogue trees. Dialogue trees afford taking dialogical moves that are more or less arbitrary. A guy who goes around saying “claiming land for yourself and enforcing your claim is justice; Nozick is intelligent and his entitlement theory of justice vindicates my claim” will leave exact impressions on exact types of people, who will in turn respond in ways that are characteristic of themselves. Every branch of the dialogue tree will leave an audience with an impression of who is right, and some audiences have measurably better calibration.
Just because no one can draw perfect triangles doesn’t mean it’s nonsense to talk about such things.
There are correct and incorrect ways to play language games.
That’s the crux. Wittgenstein himself believed otherwise and spent the most part of the book arguing against it. I think he makes good points.
At one point, he argues that there’s no single correct interpretation for “What comes next in the sequence: ‘2, 4, 6, 8, 10, 12, …?’”
Maybe this goes a bit too far. :) I think he’s right in some nitpicky sense, but for practical purposes, sane people will say “14″ every time and that works well for us.
We can see this as version of realism vs anti-realism debates: realism vs anti-realism about natural abstractions. As I argue in the linked post, anti-realism is probably the right way of looking at most or even all of these, but that doesn’t mean “anything goes.” Sometimes there’s ambiguity about our interpretations of things, but reality does have structure, and “ambiguity” isn’t the same as “you can just make random stuff up and expect it to be useful.”
That’s the crux. Wittgenstein himself believed otherwise and spent the most part of the book arguing against it.
I could be wrong, but my understanding was that Wittgenstein did think there were correct and incorrect ways of playing language games, but that this was context-dependent, and of course, someone could always choose to play another language game instead.
According to this article, the point being made with the sequences is that the correct completion is subject to interpretation and even though I could try to explain how the sequence should be interpreted, this explanation would itself be subject to interpretation, leading to an infinite regress. Wittgenstein ends up arguing in the end that we learn things through training rather than explanation.
and some audiences have measurably better calibration.
It’s not straightforward in all contexts to establish what counts as good calibration. It’s straightforward for empirical forecasting, but if we were to come up with a notion like “good calibration for ethical judgments,” we’d have to make some pretty subjective judgment calls. Similarly, something like “good calibration for coming up with helpful abstractions for language games” (which we might call “doing philosophy” or a subskill of it) also seems (at least somewhat) subjective.
That doesn’t mean “anything goes,” but I don’t yet see how your point about dialogue trees applies to “maybe a society of AIs would build abstractions we don’t yet understand, so there’d be a translation problem between their language games and ours.”
Sorry, I can’t quite follow why you are saying that dialogue trees are the substrate of language games or how this ties into the arguments. Any chance you could clarify?
The existence of natural abstractions is entirely compatible with the existence of language games. There are correct and incorrect ways to play language games.
Dialogue trees are the substrate of language games, and broader reality is the substrate of dialogue trees. Dialogue trees afford taking dialogical moves that are more or less arbitrary. A guy who goes around saying “claiming land for yourself and enforcing your claim is justice; Nozick is intelligent and his entitlement theory of justice vindicates my claim” will leave exact impressions on exact types of people, who will in turn respond in ways that are characteristic of themselves. Every branch of the dialogue tree will leave an audience with an impression of who is right, and some audiences have measurably better calibration.
Just because no one can draw perfect triangles doesn’t mean it’s nonsense to talk about such things.
That’s the crux. Wittgenstein himself believed otherwise and spent the most part of the book arguing against it. I think he makes good points.
At one point, he argues that there’s no single correct interpretation for “What comes next in the sequence: ‘2, 4, 6, 8, 10, 12, …?’”
Maybe this goes a bit too far. :) I think he’s right in some nitpicky sense, but for practical purposes, sane people will say “14″ every time and that works well for us.
We can see this as version of realism vs anti-realism debates: realism vs anti-realism about natural abstractions. As I argue in the linked post, anti-realism is probably the right way of looking at most or even all of these, but that doesn’t mean “anything goes.” Sometimes there’s ambiguity about our interpretations of things, but reality does have structure, and “ambiguity” isn’t the same as “you can just make random stuff up and expect it to be useful.”
I could be wrong, but my understanding was that Wittgenstein did think there were correct and incorrect ways of playing language games, but that this was context-dependent, and of course, someone could always choose to play another language game instead.
According to this article, the point being made with the sequences is that the correct completion is subject to interpretation and even though I could try to explain how the sequence should be interpreted, this explanation would itself be subject to interpretation, leading to an infinite regress. Wittgenstein ends up arguing in the end that we learn things through training rather than explanation.
Yeah, what I meant was the belief that there’s no incorrect way to set up a language game.
14 is certainly the most likely continuation but it could also be
16 if it’s a list of numbers k where k^2 + 7 is prime
18 if it’s a list of numbers of the form 3^i +/- 3^j
These continuations are unlikely in general but are the kind of thing that might show up in an academic mathematics paper.
It’s not straightforward in all contexts to establish what counts as good calibration. It’s straightforward for empirical forecasting, but if we were to come up with a notion like “good calibration for ethical judgments,” we’d have to make some pretty subjective judgment calls. Similarly, something like “good calibration for coming up with helpful abstractions for language games” (which we might call “doing philosophy” or a subskill of it) also seems (at least somewhat) subjective.
That doesn’t mean “anything goes,” but I don’t yet see how your point about dialogue trees applies to “maybe a society of AIs would build abstractions we don’t yet understand, so there’d be a translation problem between their language games and ours.”
Sorry, I can’t quite follow why you are saying that dialogue trees are the substrate of language games or how this ties into the arguments. Any chance you could clarify?