Well, sorry, some chess-playing computers are. Unless we get into really weird epistemological territory.

That's a function of the decision-making process. Pseudo-random selection is sufficient, other things may play a role (and sometimes do - actually utility calculus probably plays less of a role in most decisions than priming and the like).

It will apply a PRNG. This happens in the real world all the time. Thankfully, many effects in the brain are also "very nearly" random in that they appear to be determined by electromagnetic anomalies, variations in the exact number of molecules in a synapse, etc. And even "bad" pseudorandom and "pseudopseudorandom" will do the job. And some computers use even wonkier systems to break ties.

I don't think the "tiebreaker problem" has ever been a serious practical problem for any competent computer programmer in the last 20 years (unless they're working with severely restricted hardware or forced to use software written by incompetent programmers).

I'm just going to assert that I'm quite confident that I know a lot more about Dennett than you...we can have a dick-measuring if required

I think this is pretty important. Are the components that comprise cognition (in the deterministic view that you are arguing for in this thread... and as you noted you don't necessarily agree with it) based on deterministic laws of physics or not? And if not, what else comprises cognition (again, according to determinism)?

Are you suggesting there is something beyond deterministic physics which comprises the determinism that you are arguing for?

And when you say "deliberate"... what is physically going on in the agent that is making the choice? And if you say there is nothing "special" going on in the brain of someone making a choice (but "there are subtleties"... what subtleties exactly?), do you mean that the choice is fundamentally determined by the physical processes going on in the nervous system? And if they are not, again, what else is there according to the deterministic position?

How does this have any bearing on responsibility at all?

Why is that? What reason do we have to talk about anything other than possible worlds? And since you claim to be talking about metaphysical possibility, if you aren't talking about possible worlds then what are you talking about? Can you provide a reference to a philosopher or a system of modal logic that uses "your version" of metaphysical possibility?

You're right that I don't know David Lewis and I'm not aware of many developments in the last 60 years or so. But I'm not about to take it on faith that there is a consensus against the "possible worlds" view of possibility. Particularly since a quick Google search yielded this as well as many other contemporary articles that references the view as common (admittedly many are critical).

Agreed.

To me "necessarily false" means "false in all possible worlds." So again, this statement reads as nonsensical to me. How do you define "necessarily false" without reference to possible worlds?

I disagree. I don't see how "actual world" possibility is relevant in the slightest, nor do I see how two things can be possible in the "actual world" in the first place. It seems to me that there is only one reality in the "actual world" - the reality that will actually happen. Outside of a multiverse, two mutually exclusive options can't both actually happen, only one will actually happen. Either I will actually eat the toast, or I will actually eat the bagel, not both. And the only option that is part of reality is the one that I will actually choose. If the future is viewed as "not existing yet," then I would say that neither option is possible in this reality, as neither option will be part of reality until the choice is made. If the future is viewed as existent, then one of the options is already defined despite being indeterministic - and in that case the choice is what it is, even though we can't know it yet. Either way, only one course is possible in actual reality, and I don't see how indeterminism can possibly change that.

But even if I did agree that multiple options can be possible in actual reality, I would still say that it's completely irrelevant whether they're possible in actual reality or in a counterfactual reality. Responsibility is the same either way, choice is the same either way, our experience of the world is the same either way, in fact, I can't see how anything is different depending on whether the possibility is an "actual" or a counterfactual possibility. Seems like an empty distinction to me.

I don't deny that one bit. But I still stand by the statement I made about his style (which I notice you didn't actually deny).

Well, in theory, they can be any deterministic laws. Laws of physics, "laws of spirit," or some other kind of laws entirely. My determinism doesn't imply materialism or physicalism - the state of the universe may include spiritual planes, experiences, or energies, souls, god(s), etc.

But just above I decided for the sake of argument to assume physicalism, so I suppose I'll do that (for the sake of argument).

Deliberation, well, I'm assuming for the sake of argument that processes are physical, but that doesn't mean that I can actually describe them. Even if we say that it happens based on neural pathways, and that the pathways are traveled based on the strength of connections among neurons, we're still not describing how it's happening very well.

In terms of subtleties, well, same difference really. The brain is extremely complex and we don't understand it well. But trickiness includes, it has to deal with sophisticated information-processing elements, there are questions of how it deals with problems of self-reference, the issue of models-within-models and the basis for construction of models, how different options are selected as such in the first place, what durka brought up about "ties," and so on.

And yes, I'm assuming for the sake of argument that the choice is fundamentally (well, depending on what exactly you mean by "fundamentally") determined by the physical processes.

But he does believe it's informative whether an agent could or could not have done otherwise in the actual world?

No, I don't understand how "possible in the actual world" has any meaning. I can understand "possible with respect to a certain standard" (as in, possible given a set of laws or axioms), I can understand "possible according to what I know," I can understand "in some possible world," but I've always assumed that possibility by its nature never applies to the actual world, and I don't even know how to make sense of that.

I think we've reached the limits of your ability...sad to say.

How can you not know the relevance of this discussion to responsibility?! Geebus!

I've already tried to explain that counterfactuals and possible worlds semantics is only ONE interpretation of "metaphysical possibility." You've defined yourself into a hole by only understanding 'necessarily x' as 'true in all possible worlds.' You keep bringing up this way of thinking and yet you know nothing of the principal author of the discipline! Furthermore, you appear to not know about where Lewis takes possible worlds semantics for epistemology: he's a contextualist. So, "necessarily x" may be true for some set of relevantly close accessible possible worlds. So, you're apparently not even aware of that development of understanding "necessarily x" as not naively as "true in all possible worlds."

But, once again, you don't get it: switching to possible worlds semantics LOSES the distinction between compatibilism and libertarianism. Don't you see that it's impotent here? If it can't distinguish between libertarianism and compatibilism, then it isn't the right analysis of possibility.

PLEASE avoid these walls of text. It may be a valuable skill for you to learn to be more concise and succinct in distilling your argument.

It does distinguish between libertarianism and compatibilism. Libertarianism is the assertion that free will can't exist under determinism, combatibilism is the assertion that it can.

You seem to be suggesting that because possible worlds semantics support compatibilism (and more or less refute libertarianism) that we need to switch to a system in which libertarianism (the position you happen to be defending) is more credible. I don't exactly see a great incentive to do that, even if I could.

And of course, I can't. Maybe there is something wrong with possible world semantics. I am not, and never will be, a philosophy graduate student. But from my perspective it looks like you're just retreating into territory where I can't follow you.

You're probably right, but that's no fun.

No, it has nothing to do with avoiding possible worlds semantics because it makes libertarianism more plausible...it's because the libertarian can assent to 'could have done otherwise' in other possible worlds but this doesn't distinguish the libertarian from the compatibilist!

It does NOT distinguish between them. Libertarianism is the claim that in the actual world agent S could do either A or B. Compatibilism is the claim that in the actual world agent S could only do A (if they were determined to do A in the actual world). For the compatibilist, the only way that S could have done otherwise than they did in the actual world is to be able to do B in some other possible world but not the actual world.

Libertarians don't care that in another possible world S could have done B instead of A...they care about the actual world. This is why possible worlds semantics can't distinguish between libertarianism and compatibilism.

ducy?

Possible worlds semantics NEITHER supports compatibilism nor libertarianism.

I have never heard the terms defined that way.

Which?

As regards choosing the toast or bagel; what forms of possibility does choice imply? What forms of possibility does determinism negate?


PairTheBoard

Actual world and not possible world.

And I remain convinced that possibilities in the actual world are an immaterial distinction.

Choice implies probably logical, metaphysical, counterfactual. Determinism only negates temporal.

How is it immaterial? It's the central point of disagreement between compatibilism and libertarianism.

Determinism implies the impossibility of such 'actual world possibilities' but libertarianism asserts that they exist and they're the source of 'responsibility.'

Now, do you disagree that it's meaningful, or do you merely assert that the deterministic thesis is true and that such possibilities are therefore impossbile?

I disagree that it's meaningful, and I disagree that there is any legitimate distinction. Looking around, I don't see this distinction anywhere. The fact that the Stanford Encyclopedia doesn't mention it anywhere, that Wikipedia doesn't appear to, and that web searching reveals no such thing, while virtually everything I find mentions possible worlds, tells me that this actual world/possible world distinction is either obscure or nonexistent (and probably not well defined).

Even if it were well defined, it wouldn't be relevant unless it spoke to the issues at hand (ie systems of modal logic, accepted definitions of choice, and the problem of future contingents). Yet googling for these hasn't given any mention of your semantics, either. Furthermore, what is written most recently seems to be all over the place, with a wide variety of different systems and approaches - but much of it references straight back to the idea of possible worlds (leading me to believe that this is the closest thing to an accepted standard that exists - and not a discredited early model as you suggest).

So possible worlds is perfectly adequate for a definition of "possibility." The fact that you want to define "choice" as an explicit reference to your specific system of possibilities, when the actual definitions of "choice" rarely contain any mention of modalities in the first place (and when they do they are wholly compatible with possible-world notions), strikes me as an attempt to handwave.

You realize that a google search probably won't return the sources where this stuff gets discussed...right? Also...I've already tried to indicate that wikipedia is lol (it's NOT writen by experts). Also, SEP is not updated very often and isn't always complete. That aside...don't make appeals to authority: discuss the concepts on their merits.

First, everyone understands a distinction between the actual world and possible worlds: are you kidding me? How does this not make sense to you?

Do you see the fallacy you made? It's the fallacy of ignorance: since I can't find any reference to what you're talking about, therefore it must be wrong.

I've given you an argument for why possible worlds semantics is not useful for distinguishing between compatibilism and libertarianism. ducy, or not? Stick to this last question.

Also, thank you for coming back with shorter, more manageable, and more succinct posts.

So a "cognitive process" is determined by the same physical laws that durka's set of falling dominoes are governed by. And since it is physically possible in some world (which is how you are defining choice... as physically/logically possible in some world) for the input to the neurons involved in this cognitive process to be different, the agent can choose from a set of possible options. If the inputs were exactly the same and the initial conditions of the nervous system were exactly the same, the agent has only one possible choice (since if we assume determinism and physicalism, there is a one-to-one correspondence between the inputs/initial conditions and the output).

Thus your "freely choosing" agent in this deterministic world is no more "freely choosing" than a domino is choosing to fall over. Because it is logically/physically possible for durka's domino to be superglued to the table in some other world -- the domino is choosing to fall over or not fall over. The same way a set of ion channels in a person's nervous system are "choosing" to open or not open in a deterministic world.

The only difference is that the person is undergoing a "cognitive process" which you defined as a process involving deliberation and cognition. But these processes of deliberation and cognition are determined by the same physical laws that a domino falling is determined by. Thus it completely arbitrary defininition of "choice" because BOTH systems -- a set of neurons and a set of dominoes -- can select a pathway from many possible worlds (both are dependent upon initial conditions).

A set of dominoes does not deliberate and is not self-aware, and as I said both of these processes, given physicalism, are dependent upon the exact same physics that a set of falling domnoes are governed by. Thus your definition of choice is completely arbitrary and meaningless in the context of a debate on determinism/indeterminism... you may as well say "choice" is defined by being able to select from many possible options (such as the option of being superglued to the table or not) and being red in a stack of blue dominoes. The argument for free will in a deterministic universe would be exactly the same in either case.

Matt...would you like an MA in philosophy? nihan sir

That "cognitive process" is the 'redness' in the original domino analogy of mine.

I'm not an expert on this topic (and I haven't read all this thread), but it seems to me that you are not adequately addressing durka's claim. If we understand determinism as the denial of real possibilities and we understand possibility in terms of possible worlds semantics, then determinism implies that there is only one possible world. Since many determinists deny this claim, either they are inconsistent or they have a different understanding of determinism than the one you are proposing. I suppose it is possible that most determinists are inconsistent, but if this is your view you should quickly submit it in essay form to a journal as that is a major result.

A quick way to show why this seems false. It seems like we can imagine completely deterministic possible worlds that differ from each other (e.g. having different values for physical constants). This sense of possibility obviously has nothing to do with a denial of determinism as both of these possible worlds are completely deterministic.

There's only one future possible world given the current states/microstates and the laws of physics, but there are infinitely many past possble worlds with counterfactuals: had things been different, then...

And that's fine...but it's irrelevant.

What do you even mean by "each state is finite"? Are you saying that each state is describable using a finite amount of data and only finite precision?

No, but it really seems that you're requiring an infinite amount of information to be computed given finite data. That's bad news for a Turing machine. It really seems that you're going to need infinite precision or infinite information to make your Turing machine do what you want it to do.

Ummmm... yeah... that sounds like something you thought through in advance. [/sarcasm]

What is your definition of fatalism?

It sounds like you're equivocating on your definitions again. What's the difference between "predictability" and "computability"? And what do you mean by "predict the past"?

So are you taking a "strict" mathematical interpretation of the words that you're using? That's going to make it tough to make sense of the following:

What is the field over which this vector space lies? Is it actually an n-dimensional vector (endowed with all the appropriate vector space properties) or is it an "list" of some sort with n elements?

sucks being on the sidelines because compatibilism is as lol as libertarianism

Wikipedia was roughly as accurate as Britannica, which is written by experts, many years ago. By most accounts the accuracy has steadily increased over time.

Yes, everyone understands the distinction between the actual world (the realm of actuality, not possibility) and possible worlds (the realm of possibility, not actuality). Now you're blurring this common distinction by talking about possibility in actual worlds and actuality in possible worlds, which I think is contradictory.

If we're talking about possibility, then we're talking about possible worlds (or possible futures). We are not talking about actual worlds (or actual futures).

You're claiming that possible worlds has been discredited, that possibilities being "located" within the actual world is an established consensus. If it were an established consensus there would be plenty of references.

Your argument is that libertarians define free will based on a special kind of possibility that wasn't even invented until recently and still hasn't trickled down to sources like the Stanford Encyclopedia. If you want to define free will like that, fine. But no accepted definitions of "choice" and "responsibility" include any such handwaving: determinism is compatible with choice and with responsibility, and that's what is relevant.