Do you have a view on the fairness of (ultimate) punishment given a least damage compatibilist view?

Perhaps you can help me understand the metaphilosophy of the free will debate because I'd rather be wrong than naive about what's going on (also if you recall, I have hangups about the lack of progress in philosophy). I think it's a straightforward and dead debate; I don't see any basis for incommensurability of intuitions among the various camps. My view is that libertarians have no tenable conception of free will because they cannot explain how indeterminism combines with one's intent to choose in such a way that people can be held responsible for their actions. The optimal libertarian stance is just mysterianism about our ability to understand the problem of free will and causation (Searle, Nagel), but if I may reprise an earlier comment, this is Kantian noumenalism in a different dress.

Compatibilism is almost fine because there's little overhead in what determinism is taken to mean. And I don't object in general to compatibilist-style accounts of agency. But, as a looming principle, I do not see how any compatibilist account of making choices is compatible with holding agents morally responsible for what they do. For example, I found Frankfurt's paper to be exceptionally irritating on moral responsibility.

On the other hand, what exactly is philosophically wrong/insufficient about pessimism? Where is the overreach in that argument? Yes, it's quite annoying to be forced to say things like "Hitler was not morally responsible for what he did" and even more annoying when you have no philosophical defense for judging others when they aggrieve you, but if pessimists are right then there's just no way around that.

Anyway, my point is not just that pessimism is right, but that it's an example of an easy and decisive philosophical result, which stands out because philosophy has few decisive results.

I have no idea what you're talking about. I'm on the topic of madnak's definition of "emergence." You cut out most of my post and said that I was saying something or another about not knowing:

And I'm reiterating the point that it's not epistemological. That is, there's nothing to "figure out" regarding madnak's concept of "emergence."

Knowing or not knowing how many parts we can remove from the brain before it ceases to function isn't particularly important when talking about what madnak means by "emergence."

It seems as if you're looking for an empirical answer to a non-empirical question.

That's not what I'm talking about. I'm just talking about whether the properties exist in the first place (in whatever form).

We can't (yet) figure out the properties of a dog just by looking at the atoms. So fair enough.

In this case it's different; we can (and do) calculate the properties of "a gas" and the properties of "water" based on the properties of the individual molecules.

But this isn't even important. What's important is that "water" has properties different from "a molecule of H2O." In a similar sense, "the brain" has properties different from "a neuron."

This isn't on the topic.

The only thing I need to do in order to cut you off at the pass is to show that a brain doesn't necessarily have the same properties as a neuron.

In other words, just because a neuron is incapable of choice doesn't imply that the brain is incapable of choice.

No, choice was defined by you as: ""to take a course of action as a result of an internal process that enumerates and evaluates modeled actions."

I've shown that both a computer built to simulate chess and a simple circuit have these properties. You haven't shown there's anything "special" about the computer other than it is modeling chess moves rather than the on/off state of a light bulb. So what is special about chess?

In my example, the "action" is to turn the light bulb on or off. This is modeled by the circuitry. In your example the "action" is a chess move. This is modeled by the circuitry. Why is the chess move any more special with respect to the circuitry choosing or not chossing?

No, I didn't say the computer is the same as a circuit. I said it is built entirely on circuitry, and the way it models phenomena is through utilization of this complex circuitry. The same as my light bulb circuit. The light bulb circuit can't model chess moves simply because it isn't complicated enough. Add enough circuit components and it indeed can model chess moves.

If you're so hung up on chess. Then hook my circuit up to a simple program that receives inputs from a chess board. The program states, "if the game is in state X1, then close the switch and turn on the light bulb. And if the light bulb is on make move Y1. If the game is in state X2, then open the switch and turn off the light bulb. If the light bulb is off make move Y2."

What EXACTLY is different with my light bulb circuit modeling chess moves vs. your computer (that is made up of more complicated circuitry) which simply takes a greater range of inputs and gives an output, but rather than turning on/off a light bulb it turns on/off specific pixels on a computer screen?

Here's a hint: I used to have a chess board that had internal circuitry (a computer) which modeled chess. Guess how it would tell me to make the next move for it? It would light 2 bulbs up corresponding to locations on the chess board.

But you don't understand how trivial this sense of "emergence" is. It's not a meaningful characterization in any real sense. Using the exact same sense, we have the emergence of "water" from "water molecules." Big deal.

I am not denying that there are characterizations of objects that have new properties. I'm just saying that your understanding is so broad and vague that it doesn't actually mean anything.

Edit: You're treating "emergence" like a magic wand that you can wave and assert whatever you want at whatever level you want to assert it.

This isn't where I was going.

Right. Simple emergence is not controversial.

I brought up emergence is that this is where I thought you were going.

Turning the light bulb on and turning the light bulb off are not modeled by the circuitry. Each chess move is modeled by the circuitry.

With Aaron I've already defined a "model" as a "representation or simulation," and "evaluate" as "weigh and rank." This isn't happening. There is no representation of the actions nor simulation of the actions, and it's a stretch to claim the actions are being evaluated.

The light bulb circuit doesn't model anything.

A conditional statement is not a model of a chess move. The chess moves aren't being evaluated at all. You could argue that the condition of the switch is being evaluated in the loosest sense, but the moves are not being evaluated. The connection between the circuit and the chess moves is wholly arbitrary. The circuit does not contain any internal representations of the chess moves - if it doesn't represent the chess moves then there is no modeling going on.

While I might be willing to play on "evaluate," what's being evaluated is not actions (or models thereof) but the specific switch condition. And no, I don't buy that a switch in the "on" position is a model of "knight to c3." That's like saying a blank piece of paper is a map of New York City.

The difference isn't in the inputs and outputs, but in the process (evaluating modeled actions).

Does Madnak understand multiple realizibility?

And meaningless. You gain nothing by bringing it up.

When you brought up emergence, I had a feeling you were going to go with some nebulous statement that doesn't really say anything at all. I was right.

You are trying to get to a point where you have the "emergence" of a "model." You have that "automatically and robotically following instructions" is not a decision. You can say this of and/or gates and consequently of an if-then statement.

Does the existence of an if-then statement give you a model?

Mad: the very topic of models and modelling is highly controversial and is a big contemporary topic in the philosophy of science (though it's a topic that has been around in metaphysics for quite some time).

I never said that turning the light bulb on and turning the light bulb off were modeled by the circuitry. I said that the light bulb turning on/off is the ACTION being taken per the definition of choice that you gave. What is being MODELED by the circuit is the if-then statement "if switch = closed, light = on", "if switch = open, light = off".

The circuit is REPRESENTING the if-then statement. It is EVALUATING whether the switch is open or closed. The definition you gave for "evaluate" as "weigh and rank" is ridiculous. You can evaluate something without "weighing and ranking" it.

I can evaluate f(x) = x^2 given any value of x without "weighing and ranking" it. I can evaluate if-then statements without "weighing and ranking" them.

And if you really really insist on this silly definition of evaluate, then fine, my circuit is "weighing and ranking" the state of the light bulb and switch:

1) If switch = closed, then light on is weighed higher/ranked higher than light off.

2) If switch = open, then light off is weighed higher/ranked higher than light on.

Yes there is. I've shown this multiple times in multiple ways.


Yes, it clearly does.


Yes they are. The moves are being ranked based on the initial state of the chess board.

If you are going to show that this is wrong, it is pretty clear you are going to have to explicitly show what is different about the circuitry in each case, and why in one situation the circuitry is "modeling" something and in another case it is not.

Yes they are.

As it is in a CHESS SIMULATION. The programmer is arbitrarily making a connection between the circuitry and the program to model the chess game because that's his goal. It's not like there is some magical set of circuits that are "chess simulation" circuits that possess special madnak free choice electrons.

OK -- show me as explicitly as you can possibly muster how the circuitry in YOUR chess simulation computer contains an internal representation of the chess moves, while MY circuit that is explicitly linked to a chess board does not.


OK, show me how the circuitry in your chess computer is "modeling actions", while my circuitry is not. They are both based on the EXACT same principles of electronics.

It can be part of the model if you link the switch being in the "on" position to a chess move. This should be obvious because this is how computers work, and given our discussion it is clear you believe computers can model chess moves.

Either both circuits are "evaluating modeled actions" are neither of them are. Why? Because they are both circuits which utilize the same principles of electronics.

If you are going to show this is false, you need to show how one configuration of circuitry components "evaluates modeled actions" while another one cannot.

Matt, you're giving a good line of argument pushing on Madnak's position.

The worry is that his use of 'model' and what it is for something to model something is probably not very precise and may lead to a number of problems for his position.

As before: No more than a blank sheet of paper is a map of New York City.

So what does it take to get a "map"?

Or more important to the conversation, what does it take to get a "model"?

Sure, anything like this is controversial. As a monist, I am in a position in which everything I believe in can be viewed as "the same stuff." People who believe in fundamental differences between "things" can define things on the basis of those fundamental differences. The fact that I don't believe in those differences means I don't have that luxury.

Still, I don't think it has much bearing on this debate. I think we're headed for Sorites territory, if we're not already there. Since I don't believe in the "reality" of boundaries between things, any boundary I accept can be pointed out as arbitrary. "Why did you draw the line here, instead of a little bit to the left or right?"

I don't need to invoke those beliefs here, but it would be silly of me to define my terms in a manner inconsistent with those beliefs. So I'm going to be inundated with "why does this count as a heap, but not that?"

A sufficient level of detail and fidelity, in both cases. A geographic map is probably a kind of model, come to think of it - a model of a particular geographic area.

There is a correspondence between the model and the thing it models. Typically this is the structure (spatial, physical, logical, informational) of the thing being modeled, some element of that structure will be preserved in the model itself.

A chess knight (the physical piece) can be described as a model horse; it has a flowing mane and a snout in the shape of a horse. A marble cannot.

I think I've made my point clear now and I'm not going to spend any more time discussing why a marble resembles a horse in some way and is therefore a model of a horse.

He's a monist...time to move on.

Everything is information. Why couldn't Leibniz think of that?

Information is able to be 'stuff'...that's amazing. How does it do that?

How does something being a sign (semiotics) lead to the production of matter from mere 'information'?

My definition is based on evaluated modeled actions. If it's not the action being modeled, then it doesn't fit my definition.


Not according to the definitions I've presented. Furthermore, it is not evaluating the actions. My definition is clear on this - the actions are what must be modeled, and these modeled actions are what must be evaluated. Your switch doesn't do that.

I'm not using that sense of the word "evaluate."

It's not weighing and ranking the actions involved. Let me repeat my definition:

"to take a course of action as a result of an internal process that enumerates and evaluates modeled actions."

The actions must be modeled, then the actions must be enumerated and evaluated. The actions. Moving a knight to c3 is an action. Eating toast is an action. The state of a light bulb is not an action.

Yes they are. The moves are being ranked based on the initial state of the chess board.

If you are going to show that this is wrong, it is pretty clear you are going to have to explicitly show what is different about the circuitry in each case, and why in one situation the circuitry is "modeling" something and in another case it is not.



Yes they are.



What the programmer does is irrelevant. The logic of the chess game is contained within the chess program (and within the memory of the actively running program). In particular, the program runs through each legal move, models each one in succession, and evaluates them in order to determine the move to actually make. Your circuit doesn't do these things.

Maybe this isn't the only possible way to program a chess computer - but a computer programmed in another way would not be making choices.

I don't know the terminology. I do know that the computer typically has a tactics "engine" that calculates the board permutations following from a position. It also has a "library" of recognized board positions that have been analyzed prior to play, and it can derive conclusions on the basis of those. Naturally, it has the logic of chess (winning conditions, legal moves, etc) programmed right inside it.

The program then evaluates moves - typically if it can do so, it will determine whether a particular move leads to a necessary win, a necessary draw, a necessary loss, or permutations resulting in multiple outcomes. Wins are preferred to draws are preferred to losses, and heuristics for determining the best move in a mixed scenario are written into the system. Among differing moves in a particular category, "ties" are won either by length of play (for example, the shortest possible win or the longest possible loss) or by a PRNG if there is no other difference - possibly even some kind of psychological algorithm, I don't know how advanced these things get.

Often, many permutations will lead to uncertain outcomes (ie the game will continue on beyond the computer's ability to calculate further). In these cases the computer can use a heuristic to rank the different moves (probably taking into account number of pieces remaining, general positional advantages, and so on).

You circuit, on the other hand, contains nothing resembling chess - it's a switch external to the chess board. And... that's it. It doesn't do the same things.

Uh, no. The fact that they are both composed of circuits doesn't imply that they do the same things.

I don't believe everything is information. That was a position I was using for the sake of argument in attempting to mathematically model a deterministic universe.

But the fundamental premise implies that matter is information, so that question is nonsensical.

Again as I've repeated, I'm a solipsist. Everything isn't information, everything is my mind. But that conversation is far away from this one and not something to get side tracked on.

I'm just trying to get the most basic model I can out of you. As long as you remain deftly vague about what you're talking about, you're going to keep on avoiding the central issues.

There is a correspondence between up and down spin and the two sides of a coin. Does this mean that spin is a "model" of a coin? Does it "model" the states of "on" and "off"? Does it "model" left/right?

Your notion of correspondence is once again vague enough to be basically anything. But sometimes you say it is a model, and sometimes it's not.

Why not? Just a couple posts ago you were talking about mind-trees and spirit-trees. Why can't we have marble-horses?

LOL

Solipsism isn't monism, ducy?

We don't agree on this. (What you're trying to do or otherwise.)

Heh, bull****. You know exactly what I mean by "model." What you're doing is looking for details that appear inconsistent so that you can jump on technicalities.

Sorites talk is inane. It doesn't matter where the exact line between "red" and "orange" is. Some shades are clearly red, others clearly orange, and others in-between. If you want to start a thread on "whoa dude, like, where do you draw the line?" then go right ahead. But don't insult me with a thousand questions in the vein of "what about this color, is it red?" "How about this other one here? Red, or orange?" "Oh, trick question, this one was actually purple." That's bull**** designed to drag down the debate so you can keep it on a rhetorical footing.

If you want to clarify, then ask questions that are conceptually relevant. I already told you I'm not going to bother answering questions in the form "is x level of detail 'enough' detail?"

It is in my case.

No, it's not.

Solipsism: we can't know that other minds exist.

Monism: all there is is 'mental' stuff (which is basically Berkeley-ish).

Your position (?): All there is is YOUR mind.

Totally different to go from "all I can know is what is present to my consciousness" to "all there is is my consciousness."