If the subject (me, in my example) is aware of the change, then my brain state is not the same. So, of course I might choose coffee instead of tea (well, not me specifically, I don’t like coffee) – put me back in time, will you? Aha, I’ll make a different selection! That will show you!
ETA: That is, the RitterSport the first time through has had a different life experience than the RS the second time, since the second time, I’m aware that I’m choosing again. Given a different life experience, of course my choice can be different.
Whether we live in a non-deterministic world with free will, or a deterministic world with the illusion of free will, it doesn’t affect me one way or the other, and it won’t change my outlook, or the way I live my life. I believe our perception of reality is an illusion. Our understanding of quantum mechanics makes that quite clear. We live in the macro-world of classical Newtonian physics, but we’re built from the wacky, probabilistic world of subatomic particles.
Of the 3 leading interpretations of QM, the Copenhagen Interpretation posits an indeterministic universe, where probabilities govern quantum events. The Many-Worlds Interpretation suggests a deterministic multiverse, where all possible outcomes are realized and the universe’s evolution is rigidly deterministic. The Hidden Variable Interpretation is deterministic, but Bell’s theorem challenges this interpretation by demonstrating that any local hidden variable theory must adhere to certain constraints, which have been violated in experiments. Whether determinism at the micro level can be scaled up to the macro level is anyone’s guess.
While quantum mechanics has reshaped our understanding of reality, suggesting that at a fundamental level, reality may be probabilistic or even branching into multiple universes, the implications for our macroscopic world remain a subject of ongoing scientific inquiry and philosophical interpretation. But whether real, or illusion; deterministic, or non-deterministic—I’m just along for the ride, and I’ll enjoy the ride while I’m here.
It illustrates that there are certain assumptions that you make, but don’t question. The same leeway extended to free will makes it just as reasonable as the other options.
Again, what is it that connects sound waves with brain processes? You assert that things happen a certain way, but give no accounting, no mechanism behind this. Sure, if you assume that things occur that way, and everything must be accounted for by these measures, then there is no room for free will, because it can’t be reduced to causal processes and remain free will. But that just says something about the primitive notions you’re ready to accept.
That physical laws constrain anything already is an assumption. On a Human conception of laws, they are just an abbreviated description of what happens, but have no power of making anything happen. Think about the weaver creating a tapestry: it’s not that the pattern dictates the weaver’s actions, but rather, those actions create the pattern.
Equally, that they chose tea produces the conjunction between the state of their brain and their actions we call ‘causation’.
Sure. And like setting a chess board back to a certain configuration, you can then make a different move, should you choose to do so.
It’s a fairly widespread conception in the philosophy of science. (According to the 2020 PhilPapers survey, shared by around 31% of respondents.) For instance:
Humeanism about laws of nature is an approach to scientific laws, which denies that laws imply necessary connections between distinct existences. Roughly, the idea is that laws of nature reduce to the patterns of occurrent, nonmodal, events that occur in the world. The laws of nature are just patterns, or ways of describing patterns, in the mosaic of events.
It’s akin to the difference between prescriptivism and descriptivism in linguistics: the former claims that proper language is governed by a set of rules, while the latter recognizes that it is in fact language use that determines the rules as effective descriptions of this use.
There is nothing nonphysical implied; in fact, Humeanists probaby tend to be naturalist in outlook.
This is not a road I’m interested in going down. Describing the laws of nature as laws, and doing experiments assuming those laws constrain things, has been remarkably successful.
I have no idea how to even parse that paragraph that you quoted. What does it imply for orbital mechanics, nuclear physics, cosmology?
You don’t need to answer, because there’s an unbridgeable gap between us if you don’t think that the brain operates under the laws of physics, or that the laws of physics constrain the brain’s operation.
Which is perfectly consistent with a Humean conception of laws, of course.
Why would it have any special implications there? In those areas, as everywhere else, laws form a compressed description of the pattern of events. Such a description will exist whenever things are not totally random.
Sure. Once you dogmatically constrain the allowed options, alternatives are closed off to you.
So, can the Earth decide to stop rotating? Can Mercury decide to stop orbiting the sun? Can hydrogen decide to pair up with 30 oxygen atoms instead of one or two? Can Uranium decide to have a shorter half-life?
Or, does this compressed description of the pattern of events only affect the human brain for some reason, and not all the other physical and chemical interactions that we observe all the time?
It seems that laws of physics do an exceptional job saying that everything will follow the rules laid out, in experiment after experiment, and in everyday experience. A chair sitting on the floor will not suddenly start to float for no reason. A ball thrown in a vacuum will follow a certain path. We all agree that those initial conditions will lead to certain outcomes.
Now, here’s the brain, a physical object, but for some reason, it’s now in question whether initial conditions will lead to a certain outcome. Seems like special pleading to me.
Avoiding a wall of text here. The brain is complicated and we have no way of predicting in advance what a person will do when asked whether they want coffee or tea.
These new large language models are also complicated, and again we don’t know what it will output when asked a question. For example, the Go machine made all kinds of unexpected and surprising moves. If the Go computer or the LLM had its state saved and, after it put out an answer (or made a move), that state was reloaded, would you argue that it could have another output?
There’s nothing at all special about the brain; it’s a physical system like any other, save that it has cognitive processing capabilities. The problem you’re having is that you expect there to be some sort of ‘pattern-maker’ that makes it such that the Humean mosaic of events is describable by the laws we find; but that’s just like saying, but there must be some laws that make it so that this pattern exists—but that just denies the premise.
On the other hand, you’re not asking how it could be that laws of physics—abstract objects—have power over the way things happen in the world. In the words of Stephen Hawking, what breathes fire into the equations? What makes it so that the laws have power to force themselves upon atoms and galaxies alike? This you accept as a brute fact of the universe. Accept the Humean mosaic on the same terms, and you’ll get an explanation for the laws, at the expense of having no answer for the mosaic—it’s a symmetric situation.
Or, you can go the occasionalist route, and explain that mosaic by the intention of some agency—some entity that forms the mosaic to exhibit a certain pattern, without being constrained to do so, like the weaver with their tapestry. That way, you again get an explanation for the mosaic, at the expense of introducing an unexplained intentional action.
None of these yields an explanation ‘all the way down’. It’s just that one of them seems palatable to you, while the others don’t. But this is just habit.
On the contrary, I have asked repeatedly in this thread for a description of free will in a universe that has free will by definition. And I even said that such a universe could have souls, and magic, and gods, and whatever one thinks is necessary for free will to exist.
And still, when it comes to basic questions of how an entity in such a world makes its choices (does it even take into account past experiences, predilections it was “born” with, etc?), the answer has been crickets, because determinism is completely a red herring here.
The problem with free will is that the concept itself is incoherent.
Sure, our knowledge isn’t complete. However a key point here is that cause and effect not only has predictive power, but is something we all make use of every day. It’s an incredibly useful model.
It’s very much a false equivalence / bait and switch to say that because one model cannot yet explain everything, then it’s on the same footing as a concept which isn’t even concretely defined, let alone formed into a model yet.
I mean this is basically: we don’t yet understand everything about the human body, therefore cthulhu.
Given what metaphysics? In the world governed by natural laws, no. In the Humean mosaic, depends on whether there is genuine intent or will in these devices (which there probably isn’t). In the occasionalist world, depends on the choice of the entity weaving the mosaic.
I’m not arguing for any of these possibilities in particular. I’m just pointing out they exist.
It is, if you require it explicated in terms of other primitives, such as causation, natural laws, or randomness. Finding an account of how any of those work is just as nonsensical as with free will: it’s the question of the unmoved mover, of the fire breathed into the equations. That you see this as a show-stopper for free will, but not for any of the other accounts of how stuff happens, only tells us something about what sort of inexplicables you’re ready to accept without question.
Free will is not any less well defined as causation, laws of nature, or randomness. Neither of these has a mechanism into which they can be decomposed, because they constitute the primitives that make up such mechanisms. You insist that free will must be decomposable into any of the other mechanisms; but equally well, you could insist on a definition of causality in terms of free actions: no causality would remain, so the notion is inconsistent.
Nothing is ever predicted based on causation. Predictions are made by extrapolating from patterns. If I can continue the pattern of a tapestry from having seen half of it, I don’t claim that the first half of the tapestry caused the second; I just continue the pattern.
OK, then why would the pattern of a Go computer or LLM (that you agree, I think, would give the same output if the state were reloaded) be defined by its conditions, but not a brain?
Explain to me the difference between a brain and an LLM, or a star, or an atom, other than complexity.
To answer this, you’ll first have to tell me what sort of world ‘this world’ is. I seem to lack that kind of insight.
How can it be special pleading for the human brain if I explicitly allow for the possibility that non-brains may have the same properties in the very paragraph you quote?
I don’t believe that every event depends on a choice (so I’m not an occasionalist), hence, there must be some conditions under which choice becomes relevant (if it does at all). The brain fulfills these, if anything does. I don’t know if LLMs are sufficiently brain-like to do so, too, but I doubt they do (I don’t believe they have intentionality, because they only have access to structural facts about language, which fails to suffice to pick out unique referents for the symbols they manipulate).
None of this implies anything nonphysical about the brain or the mind.
In a word, because the mere structure of a symbolic domain fails to settle any nontrivial questions about the domain of referents, i.e. their utterances have no meanings—they have no intentionality. But the argument is a little involved, so I’ve spelled it out in an essay.
I mean, this is a metaphysical question, so I’m not sure why that should be surprising.
So at very best, if I really agreed that cause-and-effect is “nonsensical”, this is a whatabout argument.
I mean, if all descriptions are nonsense and have zero explanatory power, then the rational thing would be to take no position at all. It would be to assume all the descriptions are false, until and unless any of them can make useful predictions or inferences.
The reality of course is that claiming cause and effect to be nonsense purely because we cannot explain the very first cause is an incredible reach.
Finally, let me restate again: at no point have I said “assume Determinism”, or “assume cause-and-effect”; I’ve simply asked for any coherent description of free will and how it relates to entities making decisions. Still you thought the best response would be to essentially say “Determinism is no better”. That’s your best defense of free will, and I think it speaks volumes.
It’s a plea for epistemic fairness: you’re asking something from defenders of free will you’re unable to supply for your own position, and use their failure to come up with something you find satisfying as grounds to disregard their position.
It’s not just the first cause that’s a problem, though, but the very notion of causality. If A causes B, what is it about A that facilitates the occurrence of B? And what put it there? Is a world imaginable where A did not cause B? If yes, why aren’t we in that world? If no, what is it that makes this so? And so on. Recall Russell’s cutting remark more than 100 years ago: “[T]he reason why physics has ceased to look for causes is that, in fact, there are no such things. The law of causality, I believe […], is a relic of a bygone age, surviving, like the monarchy, only because it is erroneously supposed to do no harm.”
What you want, though, is a mechanism, a causal story of how the state of any agent’s brain allows them to act freely. Otherwise, there’s no problem with the simple definition of free will as the ability to intentionally act differently in identical circumstances. It’s only once you ask how that works, and then find that neither a causal story nor one based on random events lives up to that explanatory task, that we get problems. But on that front, free will is no worse off than causality or any of the other options.
That doesn’t mean we have to throw in the towel, contrary to what you seem to be suggesting. We are finite beings of finite capacities, and it’s not a great shock that certain domains are epistemically closed off to us. A simple example is mathematical incompleteness: no theory of the natural numbers can account for all their properties. But that doesn’t mean we have to throw out the whole endeavor!
But it does mean that if somebody proposes a theory of the natural numbers (or any mathematical object of equivalent complexity), you can’t hold its failure to account for certain questions against it, because whatever your own theory is, it can’t ultimately do better. (That’s not to say that different theories can’t be judged on other merits, of course.)