Add to that the typical journalistic fault of forcing one to read through paragraph after paragraph of narrative before actually explaining what the thesis they are presenting is. I’d much prefer to read a journal article where they state their central thesis upfront
A new proposal for how mind emerges from matter
61–70 of 240 posts
Re: A new proposal for how mind emerges from matter
#62The interesting thing about mind (as far as I'm concerned) isn't intelligence, but consciousness. Intelligence isn't well defined. According to our current understanding of physics, everything is cause and effect, which throws free will out the window, so what do we even mean by intelligence? Is it just when chains of cause and effect become so complex we can't understand them any more? And if so, why does consciousn…
> The only consistent theory of consciousness I'm aware of is panpsychism, which seems very unsatisfying. Integrated Information Theory seems interesting , at least, but far from flawless. Like panpsychism though, I don't think it's falsifiable.
To me, it seems like a bit of a trick - take a philosophical question, propose an answer, stuff some mathematics into the answer, maybe even a few minor empirical predictions, make people think those added bits make the answer more likely to be correct and more respectable than competing answers which don’t do that - I don’t think that actually works though
> Like panpsychism though, I don't think it's falsifiable.
Well, I don’t think any theory in this area is falsifiable. Either you give up demanding falsifiability or embrace agnosticism (we don’t know and we maybe never will)
Maybe give up on falsifiability, since presenting it as an absolute must-have is self-defeating (the idea that falsifiability is a must-have is itself unfalsifiable)
Re: A new proposal for how mind emerges from matter
#63Re: A new proposal for how mind emerges from matter
#64Earlier quoted context omitted.
You're welcome to read the paper. Tonini's work is well-known within CogSci and its not quackery by any stretch.
I skimmed it. There's one mention of "qualia" and I didn't spot anything to connect their theory with the actual experience of consciousness besides them saying they think so.
Re: A new proposal for how mind emerges from matter
#65Articles like this annoy me: it seems to want to comment on philosophy of mind, but shows zero awareness of the classic debates in that discipline - materialism vs idealism vs dualism vs neutral monism, and the competing versions of each of those, e.g. substance dualism vs hylemorphic dualism, eliminativist vs reductionist/emergentist materialism, property dualism, epiphenomenalism, panpsychism, Chalmers’ distinction…
Re: A new proposal for how mind emerges from matter
#66The interesting thing about mind (as far as I'm concerned) isn't intelligence, but consciousness. Intelligence isn't well defined. According to our current understanding of physics, everything is cause and effect, which throws free will out the window, so what do we even mean by intelligence? Is it just when chains of cause and effect become so complex we can't understand them any more? And if so, why does consciousn…
We don’t know whether the apparent indeterminism of QM ultimately reduces to determinism or not. It depends on which interpretation of QM you prefer, and none of them has any strong empirical support
> which throws free will out the window,
As well as assuming deterministic QM, you also assume incompatibilism. Reject incompatibilism and you can still have free will even if physical reality turns out to be 100% deterministic
> so what do we even mean by intelligence?
People disagree greatly as to what “intelligence” means, but this is the first time I’ve seen anyone suggest that the issue turns on free will. Usually people present “definition of intelligence” and “does free will exist?” as questions having a significant degree of orthogonality
> Is it just when chains of cause and effect become so complex we can't understand them any more?
Mediaeval philosophers commonly viewed intelligence or intellect as a power, the power to engage in conceptual thought. A “power” is a casual concept - it is the possibility of causing something - but possibility, not necessity. It is orthogonal to the question of free will, so long as our position on free will enables us to meaningfully speak of “things we could have done but never actually did”, “thoughts we could have thought but never actually thought”…
It was our dog’s birthday the other day. She’s a smart dog, but she’s never going to understand the concept of a birthday… she enjoyed some special treats but she’ll never understand why she got them on that occasion. She has intelligence, but not the specifically conceptual sort of intelligence mediaeval philosophers were talking about. And free will doesn’t enter into that
> And if so, why does consciousness even exist at all?
I think intelligence and consciousness are distinct issues. Consciousness is having subjective experiences; intelligence is having thought processes of a certain kind. One entity might have consciousness without intelligence, another might have intelligence without consciousness.
> The only consistent theory of consciousness I'm aware of is panpsychism, which seems very unsatisfying.
I don’t think the competitors to panpsychism are necessarily inconsistent. Of course, if you reject panpsychism, you need a non-trivial criterion to decide what is conscious - and people will ask you to justify that criterion - and maybe all you can say is that it is axiomatic - and I get why proposing axioms can feel unsatisfying, but it isn’t strictly speaking inconsistent, assuming your axioms are consistent with each other. By the Münchausen trilemma, every quest for justification must end in either infinite regress, circularity, or axioms - and if axioms feel unsatisfying, infinite regress and circularity are just as unsatisfying - so maybe we are just doomed to feel unsatisfied.
> I guess the other option is to divorce consciousness from physical master entirely, but then we have kind of opened ourselves up to almost any kind of woo.
I don’t know what you mean by “divorce… entirely”-do dualists (of whatever kind) do that? Do idealists? From my own idealist viewpoint, I reject the claim that idealism necessarily entails “almost any kind of woo”, and I think the thought that it does relies on misunderstanding or misrepresenting idealism, or else confusing certain versions of it which maybe do do that with other versions which don’t
Re: A new proposal for how mind emerges from matter
#67Earlier quoted context omitted.
Code vs. data. The code needed create, train, and perform inference on a Transformer is quite short. How short depends on how you count the `import` statements in https://github.com/openai/gpt-2/blob/master/src/model.py and https://github.com/openai/gpt-2/blob/master/src/sample.py etc. Spreadsheets performing linear regression etc., — Do they learn? Sure! If you accept that Transformers adapt and learn then you must…
I am confused, a spreadsheet running the code for what a neural network does, sure. But a traditional computer program isn't just excel.
What differentiates a “traditional computer program” from a single matmul? How about a bunch of matmuls? How about a bunch of matmuls with some non-linear arithmetic thrown in between them? And if we are adding in sampling from a distribution but set temperature to zero and use random seeds so that it is fully deterministic?
Re: A new proposal for how mind emerges from matter
#68Can we retitle this as there is nothing new here besides long musings
There is, if you read it. > Which brings us to the most striking idea — that some types of electrical oscillations could mediate an experience of self. > In 2021, Hanson found that similar electrical activity — spontaneous low-frequency oscillations — is evident across many different organisms, from E. coli to humans. She concluded that across a diverse range of creatures, the oscillations may have a shared function:…
Hanson, Alison. "Spontaneous electrical low-frequency oscillations: a possible role in Hydra and all living systems." Philosophical Transactions of the Royal Society B 376.1820 (2021): 20190763.
Discussion of some of it here with M. Levin: https://www.youtube.com/watch?v=bZPhqwrSC24
Re: A new proposal for how mind emerges from matter
#69Maybe the definition of what "intelligence" is could be sharpened by having a look at LLMs and "traditional" computer programs and asking what exactly the difference between the two is. Almost all the traditional criteria of intelligence - reasoning, planning, decisionmaking, memory etc - are exhibited pretty trivially by standard computer programs. Nevertheless no one would think of them as "intelligent" in the sens…
I wouldn't say that's true at all for traditional computer programs. They're doing explicitly what they are designed to do, there is no adaptation/learning.
Alternatively, neural networks are also just doing explicitly what they are designed to do… sure there is a larger computational graph with lots of operations, but it’s all deterministic… backprop is not really much different on a procedural level than the simple fitting algorithm that I outlined above, in as much as it is just a specific well-defined algorithm or sequence of steps designed to compute some parameters from data.
Re: A new proposal for how mind emerges from matter
#70Maybe the definition of what "intelligence" is could be sharpened by having a look at LLMs and "traditional" computer programs and asking what exactly the difference between the two is. Almost all the traditional criteria of intelligence - reasoning, planning, decisionmaking, memory etc - are exhibited pretty trivially by standard computer programs. Nevertheless no one would think of them as "intelligent" in the sens…
Just two? You can name so many more terms in this concept cloud, e.g.: personhood, moral agency, consciousness, self-awareness, processing power, wit, autonomy, feeling-and-experiencing capacity, and so on… We don’t seem to agree on what’s separate from what, and yes, it would be useful.