You haven't given an argument! You haven't explained why "actual-world possibility" should be a criterion for a range of options.

I don't know whether I would agree with Quine, but he sounds brilliant and keeps getting brought up when I express my positions.

Anyhow, I think it's meaningful to speak of responsibility because of how human social behavior works. I don't think it has a metaphysical basis, which is what Quine seems to be talking about (could be wrong, haven't read him).

Only because you also defined "range of options" to mean something completely different from what it meant in context.

You agreed with the words of my definition (while defining those words in a completely different way than I did), but you never agreed to the meaning intended by my definition.

When the analysis consists of "a range of options requires possibility, checkmate," the only way I know to "go after" it is to say "no, a range of options doesn't require possibility."

There's nothing else to go after.

That depends. Is claiming that a range of options requires possibility a feature of the method (ie defining terms) or is that the content of your analysis?

I assume it's the former, because I can't think how the concept of choice could imply anything like what you're saying. But if it's the latter, then you need to flesh that out.

Here's a line from a book that includes a chapter on free will/responsibility:

"A forced choice is no choice at all."

The context was that in determinism there are no 'options' to agents since there's only ever one action that they could take. Thus, "it no longer makes any sense to talk about 'decisions' or 'choices.'"

This is not only my position. It's a widely understood problem for determinism/compatibilism.

I haven't done anything weird in taking your definition and discussing "range of options." An option is a disjunct. In order for there to be a range of options, there must be >1 disjunct. In order to select from a range of options, the agent must be able to select either disjunct and not just be forced to ever take one.

This is a "forced" choice and is no choice at all!

I assume you are asking why I think you haven't shown a contradiction in madnak's definition of "choice." Part of the frustration for you here is, I think, due to the fact that madnak doesn't seem clear on what conceptual analysis is. Now, as I said (and as seems to be the case judging from his latest comments), this might be because he rejects the picture of language that makes conceptual analysis possible. So you won't get far in convincing him using this method.

However, I (and most linguists and philosophers of language) do accept the view that there is a formal structure to language that can be described in logical and semantic terms. So if I were madnak, I would respond by stipulating a definition of choice that is consistent with determinism. That is, I would say that "choice" means choosing from a range of options, even if you only have one actual option.

Now, you seem to think that you can show this to be incoherent by demonstrating that it is a logical implication from the idea of a range of options that there be at least two. Thus, when I say choosing from a range and there is only one, then I am not referring to a "range of options." But I can still here just stipulate the meaning I prefer--I can say, by "range of options" I mean a range of both actual and epistemic or counterfactual options. Now, you might think this is a counter-intuitive way to understand "range," but I have at least removed the incoherence.

Essentially, I think this is a really difficult line for you to prove. It seems to me much more promising to show that this definition of choice is not adequate to ground moral responsibility than to show that it is incoherent. And to do this you'll have to do more than say that you are using a sense of moral responsibility that requires choice. Presumably, it is only some understandings of choice that you are allowing--more accurately some feature of the concept of choice you are using that you think is required to groung moral responsibility. If madnak's coherent definition of choice doesn't have that feature, then madnak has failed to defend commpatibilism.

For example, Gettier didn't prove that the JTB analysis of "knowledge" was incoherent (it would be easy enough to just accept the counterintuitive claims in his examples), he showed that it didn't match with our intuitive sense of the meaning of knowledge. I think madnak (and most other compatibilists) are willing to accept some counterintuitive uses of "choice" as long as it is able to ground moral responsibility.

OK, I can accept that. At the end of the day nearly all philosophy is about a battle of intuitions. The hope is to show that the implications of one's argument goes against their own intutions. So, the force of my argument depends on someone sharing the intutions about what a 'range of options' means as not allowing the move to counterfactuals or having only one option.

Yup, thanks for the explanation of your position.

So far, I think that you're really the one person with which I'd enjoy talking philosophy.

That's a quip. Possibly even a platitude. Not an argument.

This is almost an appeal to authority, still not an argument.

I don't know that >1 is necessary for a "range". But I'll go with that for the sake of argument.

I have already provided a clear example of a disjunct that doesn't involve possibility - the set of moves under consideration by a chess computer.

That only one option is possible doesn't imply that we are "forced" to take that option. For example, I think it is extremely strange to say that a chess computer is "forced" to make a particular move.

My overall position includes the claim that deterministic notions of choice are more intuitive than indeterministic notions.

My claim isn't that one actual option is enough - I abandoned that in my last post because I don't consider it relevant.

My claim is that an option doesn't have to be possible (certainly not according to some special notion of possibility) in order to be an actual option.

I do think that people in Western culture are taught that there's a conflict here. "How can it be an option if it's not possible?" seems to have some clout (though I can't see why) in American culture. But I don't believe that there is a rational basis for this "intuition," I don't think it's universal (I often repeat that I've never had any such intuition), and I don't think it is fundamental.

I think it is learned. Much in the sense we "learn" that dogs are cute and cows are not, and are therefore horrified at the thought of dogs being killed but comfortable with the thought of cows being killed, we also "learn" that possibility is related to choice or "having options." I expect that this notion is far less common in cultures that haven't been infiltrated by Western European notions (just as the view of dogs as "cute" is less common in those cultures) - probably somewhat common but not overly so in cultures that have had limited contact with the West, and probably almost unheard of in areas with none.

I think it's an element extraneous to any common intuition of choice, that gets conflated with the notion of "choice" due to habit and indoctrination.

(But this isn't the claim I'm defending here in the thread, I'm only defending against the assertion that my position is internally contradictory here.)

My overall position includes the claim that deterministic notions of choice are more intuitive than indeterministic notions.

My claim isn't that one actual option is enough - I abandoned that in my last post because I don't consider it relevant.

My claim is that an option doesn't have to be possible (certainly not according to some special notion of possibility) in order to be an actual option.

I do think that people in Western culture are taught that there's a conflict here. "How can it be an option if it's not possible?" seems to have some clout (though I can't see why) in American culture. But I don't believe that there is a rational basis for this "intuition," I don't think it's universal (I often repeat that I've never had any such intuition), and I don't think it is fundamental.

I think it is learned. Much in the sense we "learn" that dogs are cute and cows are not, and are therefore horrified at the thought of dogs being killed but comfortable with the thought of cows being killed, we also "learn" that possibility is related to choice or "having options." I expect that this notion is far less common in cultures that haven't been infiltrated by Western European influences (just as the view of dogs as "cute" is less common in those cultures) - probably somewhat common but not overly so in cultures that have had limited contact with the West, and probably almost unheard of in areas with none.

I think it's an element extraneous to any common intuition of choice, that gets conflated with the notion of "choice" due to habit and indoctrination.

(But this isn't the claim I'm defending here in the thread, I'm only defending against the assertion that my position is internally contradictory here.)

Edit - Unfortunately there doesn't seem to be much research on how different cultures view the question of possibility and choice. We know that other cultures often view consciousness or "the soul" very differently than we do. But I think fundamentally the question of what makes up a common intuition is an empirical one, and I think that all humans have slightly different intuitive takes on any subject in the first case (as I said above, I don't think any two people can literally have the same idea).

So even if it is intuitive for durka (and yourself) that "actual options" involve possibility in some way, that's irrelevant because it's my own framework we're discussing and that notion is not at all intuitive to me.

So you removed cognition from the equation, that's fine.

Take a simple circuit with a switch, a resister, and a light bulb. If the switch is closed, the light bulb comes on. If it is open, the light bulb is off. Is the circuit choosing whether the bulb is lit or not?

It is "taking a course of action" (electron flow --> light bulb turns on). The process is internal (to the circuitry, like a computer). It is enumerating and evaluates modeled actions (if switch = closed, light bulb = on. if switch = open, light bulb = off).

Since you explicitly removed cognition from the discussion, your view leads one to the conclusion that basic circuits "choose" and "make decisions", right?

I would say that you're the one evaluating and modeling the action. The switch, as far as I can tell, merely performs the action without any process of evaulation or of modeling.

Computers do. Circuits, I wouldn't say so.

What exactly is different between a computer evaluating and modeling the action and a simple circuit evaluating and modeling an action? The circuitry of a computer is what is doing the "modeling" when it selects from your "range" of possible chess moves.

Surely your position doesn't just boil down to the circuitry is "more complicated" in a computer thus it is choosing?

Well, for one thing, the model of a chess move is not the same as the actual move. The computer generates new data structures to represent what the board would look like under move 1 versus move 2, and so on.

Second, the computer typically evaluates each move, associates some "rank" with each one, and selects the winner on the basis of this "ranking."

So both the modeling and the evaluation are separate from the action itself. They are processes run before the action is selected and on the basis of which the action is taken.

In your circuit example, it seems like the only "model" or "evaluation" of the action is the action. Does the electricity "evaluate" whether or not to flow? Ehhh, that's a stretch.

I'm thinking about this in the context of following a set of instructions. Perhaps you are building something with Legos. You're just following the process of read the instructions-follow the instructions.

If you have a selection of objects from which to choose, yes. Part of my brain simply asks "How can this *NOT* be a choice?"

And the gap widens.

I don't know whether you know what this means. You're once again claiming that you can redefine the word to be whatever you want it to be. And this is the whole problem.

The characteristic tactic is to take your words and see what the implications are. If you use a phrase like "cognition" then we have to determine what you mean by "cognition." The same is true of all these other words that you're using. As it turns out, your idea of a "process" is broad enough to also include collections of numbers. (If you follow the logic that leads you to that conclusion, it's hard to see how anything can *NOT* be a process. All you need is a proper encoding scheme for whatever it is that you're looking at.) Is it any wonder that nobody is really sure what you're talking about?

And it does this by utilizing the internal circuitry present in the computer. The exact same way my simple circuit does. It's just more complicated.

And the circuitry is what is doing this. Like with my simple circuit. The only difference is it is more complicated.

The processes are exactly the same in my simple circuit vs. the computer. One is just more complicated and evaluating chess moves. The other is just less complicated and evaluating whether to turn the light bulb on or not. If the modeling/evaluation are separate in one example, they are separate in the other as well.

As it is in the computer, because they are based on the exact same principles. The computer is just more complicated.

If you want to separate the "modeling" from the "action" in my example, the model is "if switch = closed, light = on"; "if switch = open, light = off". The modeling is switch open/closed. The action is light on/off.

If you want to separate the modeling from the action in your example, it's "if chess board is in position X1, then make move Y1". If it's in position "X2, then make move Y2.", etc. through the total number of possible positions. And this modeling is based on the exact same, just more complicated circuitry, as my light bulb circuit.

They are exactly the same aside from the complexity and the purpose of the circuitry design. If you want to say the "action" and "model" is separate in one case, they are separate in both.

Your position reduces to the claim that my light bulb circuit is choosing and making decisions.

I was actually thinking of Wittgenstein's view of language here. Quine also criticizes conceptual analysis, but I am not sure that his criticism is relevant to your own defense.
Yeah. The claim is that there are universal features of language such that if we are using the same concept and have a correct understanding of the logical and semantic features of language we will agree on results. How Platonic this gets depends on the particular version.

I'm suspicious of non-linguistic ideas, but maybe you are right. As for language, while you are correct that some features of language are arbitrary, I think Chomsky pretty conclusively showed that others are universal.

Eh. Concepts are not ideas. If you want to have a completely nominalistic account of what goes on in our minds...good luck. Here I would recommend reading Frege's essay, "The Thought" for some problems with claiming that ideas have no propositional content.
No one is interested in what goes on in your mind alone. If you view concepts as referring to discrete ideas in your brain--then of course durka is being foolish in trying to analyze this concept. Of course, you are being foolish as well--if you want to analyze it you should be looking at the actual chemistry of your brain, rather than trying to explain or define what you mean to other people using "concepts" they don't have. On your account, there is nothing in common between our concepts, and so we are stuck within ourselves.

The general view is that conceptual analysis is supposed to be about thoughts, or propositions, or specific features of language that are held in common. What durka is trying to do is analyze these parts of our ideas--the conceptual part, the part that is implicated in propositions, that "timelessly mean."

If no internal process is guiding your actions at each step, but you are automatically and robotically following the instructions, then no, you are not choosing.

If a monkey is presented with a rock and a banana, and grabs the banana, do you believe that the monkey is necessarily choosing?

If no, what prevents a human being from doing whatever the monkey is doing?

If you can conceive of something completely random as a "choice," then yes.

No, I'm not. If I go to Google Maps and center the view on 360 Adams Street, being able to zoom in or out is not the same as being able to manipulate the map however I want. The important thing is that I'm focused on the right address. How big a surrounding area I include is incidental (though "too close" and "too far" make for some practical problems and are to be avoided).

It's important that a definition be consistent once a boundary has been defined, because that boundary is important in moving forward. But during the process of clarifying definitions and establishing boundaries, it's only necessary that a single definition is used (it doesn't matter which, depending on the territory).

I'm not switching from my modeled evaluated action definition, I'm just mentioning "selection" as a definition that I could use equally well (in theory). That's not the definition we're discussing, and I don't intend to change the definition we're discussing.

I view "process" as a pretty broad term. Almost as broad as "thing." But I also think numbers can fit the definition of "apple" and "skyscraper." I don't think the "ontological medium" is relevant to the definitions of these words. I figured that you assumed the same thing, so yeah, it got messy. Apparently you don't.

Most views of reality will include the question "is there some 'material of being,' some 'stuff' that reality is made out of?" And if the answer is yes, that raises the question "what is this 'stuff?'" Since we were talking about a mathematical universe, I assumed that it went without saying the answers we were using were "yes" and "mathematics." And I figured that if "mathematics" is the bottom rung of reality, then "information" would probably have to be the second rung. Maybe that was too far a jump.

But the point is, it doesn't matter whether it's "mathematics." Or (as you probably believe) "God." Or (as I believe) "my mind." Or (as so many here seem to believe) "physical particles."

Regardless of whether that apple tree is a math-tree, a physics-tree, a mind-tree, or a spirit-tree, whatever its "underlying substance," or even if there is no underlying substance - we can still call it an apple tree. Doesn't matter what it's composed of.

This isn't a matter of my definition of "apple tree" (or "process"), it's a matter of my metaphysics.

One of the hot new fields in philosophy right now is "experimental philosophy." Their view is much like your own. They regard questions about our intuitions regarding concepts to be best answered through empirical studies of how people actually think. They often will compare these intuitions across cultures as well. One of the major areas of research for them has been on people's intuitions about free-will. Here is a link to a some of their research on this topic.

It's complicated in such a way that it contains representational models of the moves and "rankings" for them. Your simple circuit contains nothing of the kind.

This is like saying an operating system is just like a Hello World program, only more complicated. It's not just more complicated, it does different things.

There is no modeling or evaluation going on at all.

That, again, is evaluation that you are performing. The switch doesn't use if/then statements. But you still aren't showing me any models.

The models in my example are data structures. Where are the models in your example?

His "internal process" in a deterministic/physical universe is based ENTIRELY on neurons which are automatically and robotically following the inputs presented to them. Like a circuit. And like a computer. It's just more complicated.

Notice a pattern?

Right. But it could if it was just more complicated. Computers are built via circuits. It is complicated circuitry. THAT'S HOW THEY WORK.

Right. Your circuit evaluates chess moves. Mine turns on a light bulb. For some reason you are giving special status to chess boards. No idea why.


This is an assertion presented without any evidence. I stated why there is modeling/evaluation going on in my circuit. The exact same processes are occurring in your computer. Why? BECAUSE COMPUTERS ARE BASED ENTIRELY ON CIRCUITRY. They are just more complicated.



If you want to go this route then your computer isn't doing any "modeling". The computer programmer which created the system is doing the "modeling". I still don't understand why you are giving your complicated chess-evaluation circuit special priveleges.


If-then statements linked to the light bulb. And here you are giving special status to "data structures" in computers when it is simply based on COMPLICATED CIRCUITRY.

For some reason I think I'm about to go down madnak's definitional rabbit hole. This time with the word "model".

Either that or my complicated (I wish) neural chess simulator in my head is flashing "checkmate".

I consider myself an experimental philosopher (I have a project in the pipes regarding people's use of norms of assertion). However, xphi has already gotten a bad name from some TERRIBLE early experiments that used very bad methods.

At the very least I'm a very empirically minded epistemologist (and decision theorist). ...fwiw

I don't mean that I think everything about language is arbitrary.

Sloppy language, then. Reading the first couple pages of "The Thought," I don't agree with really anything at all here. I'm not sure I want to slog through something when I don't agree with the premises.

This is similar to the criticisms that, as a solipsist, I shouldn't really discuss anything with anybody. I don't accept that, but it's usually not a short thing to talk about.

Suffice it to say, I don't believe there can ever be a "meeting of the mind" between two people, but that doesn't mean that I view conversation as pointless or futile.

Well, it seems clear that while some thoughts, propositions, and features of language may be held in common, some are clearly not so universal. How does conceptual analysis differentiate?

A common tactic seems to be the claim that something is "self-evident." I'm sure everyone has seen this claim made about propositions that they hold to be in error.

So to me, it seems that unless both parties can agree on a proposition, it can't be used as the bedrock of a discussion. Without finding acknowledged common ground, I don't see how a debate doesn't devolve into "this is self-evident," "no it isn't," "yes it is," and so on.

This is interesting, thanks for the link!

Unfortunately, it doesn't seem like anything is especially clear (other than, "people contradict themselves, what else is new").

Uh, yes. It's based on neurons (not entirely, also on glial cells and the circulatory system and so on). That doesn't mean it's the same as a neuron. The particular arrangement of neurons results in properties that a single neuron doesn't have.

Reductionism doesn't imply nihilism.