Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[infangthief]

So, say someone listen's to Beethoven's ninth symphony. I can see how you look at that and say "the physical aspect (how the particles bounce around) is not everything that happened". I can also see how someone looks at that and says "the physical aspect is everything that happened". This is why I'm still unvonvinced that there is a real difference here.

That is exactly the difference. A reductionist says that is everything that happened; a non-reductionist says that is not everything that happened.

Also, this is quite the necro for this thread in recent times 20 hours!

[infangthief]

The claim I would want to make is "the non-physical aspects are all perfectly reconstructible from the physical aspect". Tell me the exact position of every particle in the room, and I could in principle tell you everything about what happened, about every aspect, physical, sensory, biotic, whatever.

I agree that a reductionist wants to make that claim. And in principle they can accomplish it, to their satisfaction as a reductionist. To them there was no thing and no meaning in that room except as could be explained by the particles and their physical properties. If biotic things and meanings can be reduced to physical then it is simply a matter of reassembling - maybe the biotic meaning is like a number and the physical properties are like its prime factors.

I think a non-reductionist cannot describe everything to their satisfaction from just the physical properties of the particles. But I haven't come up with an accurate and clear way to articulate what happens here. One thing that complicates the situation is the fact that the person attempting to analyse the physical data is themselves a human (perhaps with the help of human-designed machines) and thus capable of all the different aspects. Indeed, any description of an aspect (say, the social aspect) is only possible because the describer is capable of that aspect themselves. I will continue to ponder.

[silverspawn]

A useful question could be "is this view falsifiable?". Could you, in principle, run an experiment that has different results if this kind of reductionism is true vs false?

[infangthief]

A useful question could be "is this view falsifiable?". Could you, in principle, run an experiment that has different results if this kind of reductionism is true vs false?

Probably not. It is a philosophy, or worldview kind of thing.

[silverspawn]

Noticing how people who aren't especially amazing writers use "very" and "actually" way too much. Glad it's not just me who does, or at least used to, do this naturally.

[Awaclus]

Noticing how people who aren't especially amazing writers use "very" and "actually" way too much. Glad it's not just me who does, or at least used to, do this naturally.

Actually, very amazing writers do this.

[silverspawn]

like who?

[Awaclus]

like who?

The (originally not very good, but about to be explained and therefore now hilarious) joke was that I was pretending to imply that I'm a very amazing writer, by actually doing the very thing I claimed amazing writers do, and then there was the second layer of the joke which was doing the thing you just said people who aren't especially amazing writers do, demonstrating that I am not an especially amazing writer, despite appearing to believe otherwise.

[silverspawn]

      - - - joke -->

(°-︶) <- me

[MiX]

like who?

The (originally not very good, but about to be explained and therefore now hilarious) joke was that I was pretending to imply that I'm a very amazing writer, by actually doing the very thing I claimed amazing writers do, and then there was the second layer of the joke which was doing the thing you just said people who aren't especially amazing writers do, demonstrating that I am not an especially amazing writer, despite appearing to believe otherwise.

I'm very disappointed you didn't say "actually" twice here.

I suppose it's easier to say "very" more often though, as evidenced by the previous statement.

[faust]

like who?

The (originally not very good, but about to be explained and therefore now hilarious) joke was that I was pretending to imply that I'm a very amazing writer, by actually doing the very thing I claimed amazing writers do, and then there was the second layer of the joke which was doing the thing you just said people who aren't especially amazing writers do, demonstrating that I am not an especially amazing writer, despite appearing to believe otherwise.

I'm very disappointed you didn't say "actually" twice here.

I suppose it's easier to say "very" more often though, as evidenced by the previous statement.

Actually, I was about to make that very same comment.

[infangthief]

So I was very close to actually saying that Awaclus might be a very amazing writer, but that actually I had very little idea whether or not that was actually the case.

So, what is the other word which I think gets abused, besides "very" and "actually" that I actually notice people using very often, perhaps more in speech than in writing, but I have actually used twice in this very post?

[silverspawn]

So.

[silverspawn]

I had this transition period where I purchased grammarly premium (still pay for an annual subscription), and it mercilessly told me to cut out the actually's. This led me to the style I've mentioned before, where I just focused on cutting out every unnecessary word and make my writing as concise as possible. I think this generally has the effect of sounding more confident and forceful.

But while I think the step is important, you can take it too far. In the extreme, you sound confident but not cold and not entertaining, and also it may be too dense. It's extremely difficult to spell out when to be more lengthy, which is why writing is an optimization problem in a high dimensional space and as such art rather than science.

I definitely sometimes add "definitely" and "so" on purpose. "Actually" really is rare though.

[silverspawn]

Below is a tiny LW post I've written that at the time felt really strong, but which I would now condemn for being too dry. Yes I think the point is valid and in principle self-evident, but just take the extra step and spell it out a bit more!

A lot of people speak in terms of "existential risk from artificial intelligence" or "existential risk from nuclear war." While this is fine to some approximation, I rarely see it pointed out that this is not how risk works. Existential risk refers to the probability of a set of outcomes, and those outcomes are not defined in terms of their cause.

To illustrate why this is a problem, observe that there are numerous ways for two or more things-we-call-existential-risks to contribute equally to a bad outcome. Imagine nuclear weapons leading to a partial collapse of civilization, leading to an extremist group ending the world with an engineered virus. Do we attribute this to existential risk from nuclear weapons or from Bio-Terrorism? That question is neither well-defined, nor does it matter. All that matters is how much each factor contributes to [existential risk of any form].

Thus, ask not "is climate change an existential risk," but "does climate change contribute to existential risk?" Everything we care about is contained in the second question.

which btw is a real reason to still worry about climate change.

[silverspawn]

Damn, I make a google form question, it starts as a multiple choice, I type in the desciprtion "On a scale from 0 {{description}} to 5", and google automatically replaces the multiple choice with a scale from 0 to 5. The future is here.

[silverspawn]

So I reread the sequences with the goal of understanding exactly what they say about consciousness and morality. Putting morality aside for now, the view seems to be

Consciousness is what an algorithm feels like from the inside. This means it happens *within physics*; it's NOT epiphenomenalism. This s confusing because problems we don't understand are confusing; consciousness is like addition to people who don't understand addition. It also happens at the algorithmic level, which means consciousness is substrate-independent.

Out of all theories, this is the hardest one for me to grasp, and it's the only one that I can't find a name for; I don't think there exists one that people agree on. Eliezer called it "not-quite-faith-based Reductionism"; in the survey I just made, I called it "Functionalist Reductionism".

definitely it says that it's bits rather than atoms, a process rather than a thing, and it doesn't contradict physicalism (that the laws of physics are causally closed).

[silverspawn]

The thing that annoys me about it is that it tries to sidestep the problem of how to get from matter to consciousness. The way Eliezer frames it is that other theories invoke magic to explain consciousness but this one doesn't. That makes it sound so good. But you do in fact need to get to consciousness somehow!

Put differently, how is it different from epiphenomenalism (= consciousness appears when atoms do specific things)? Eliezer has called epiphenomenalism "the most deranged idea in all of philosophy", so it better be different.

When in doubt, draw causal diagrams! I think it comes down to this: both agree that the following two things are important...



... but they disagree on how they are related. Epiphenomenalism says that the relationship is causal; do one and the other appears, i.e.,



whereas functoinalist reductionism says that they are inherently the same, i.e.,



but I'm still annoyed by framing this as less magical than panpsychism, which after all just says that



I don't think there is any sense in which functionalist reductionism happens "within physics" but panpsychism doesn't. If you get to say "consciousness is what an algorithm feels like from the inside", then I get to say "consciousness is what atom feels like from the inside". And then we can talk about how this fits the data, at which point            Moat                     

[silverspawn]

Eliezer's big argument for why epiphenomenalism is so deranged is the so-called meta problem. Why do we talk about consciousness? Epiphenomenalism asserts that consciousness has no causal effect whatsoever, so you have to say "well I know that I'm conscious because I have experience, but this is not why I report that I'm concious since consciousness has no effect; rather there is an entirely separate mechanism that leads me to say that I'm conscious, which miraculously does exactly the same thing as this non-causal subjective awareness thing". Or as EY put it, there are epicycles but the epicycles don't do anything, and then there is this separate thing that does the same as epicycles".

Making Fun of this is a lot of fun

... and I totally agree with this reasoning. But it's just an argument against epiphenomenalism; panpyschism and functionalist reductionism both look equally good here.

[silverspawn]

I'm not sure if this is intended as a reuductio ad absurdum of Functionalism (or consciousness-as-Bits), but it works that way.

[Awaclus]

I use "actually" (as opposed to supposedly) a lot. It's a great way of phrasing positions where you are simultaneously arguing that a proposed solution doesn't work and proposing a different solution that does, because you can simply phrase your position as "we should do something that actually works, like my proposal" and omit a lot of information that is now redundant with the obvious implication. You could just say "we should do something that works, like my proposal", but that fails to address the fact that some people are claiming that other proposals work.

[silverspawn]

I think this is the one use case, at least the only one that I can think of, where I still use the word on reflection. But it's outnumbered by the cases where I used to add it for really no good reason.

[silverspawn]

So this diagram thing is pretty interesting because you seem to get most theories by writing "atoms" into a box to the left, "consciousness" into a box to the right, and then doing all combinations of

- add -> or <- or <-> or ≙ in the middle
- add "moving" or nothing after "atoms" in the left box
- remove or don't remove [one of the boxes and the middle]

I.e.,

- [atoms] <-> [consciousness] is Dualism/Interactionalism
- [atoms] <- [consciousness] is objective idealism (at least insofar as I understand it, this theory confuses me)
- [atoms] ≙ [consciousness] is panpsychism
- [atoms moving] -> [consciousness] is Dualism/Epiphenomenalism
- [atoms moving] ≙ [consciousness] is functionalist reductionism aka "what an algorithm feels like from the inside"
- [atoms] is illusionism
- [consciousness] is subjective idealism

the ones missing are (1) all with an incoming arrow to [atoms moving] because this doesn't make sense, and [atoms] -> [consciousness] which is logically coherent but I've never heard anyone defend it.

[silverspawn]

Although, if you're going to entertain removing the right box, and assert that there is no-one there to observe anything, wouldn't it then be strictly simpler to just

[silverspawn]

There doesn't seem to be a theory by that name. I'm disappointed. I only found isolated instances of people arguing the idea.