At risk of infuriating someone, see A2, A3, B2, B3, C2, C3 on the chart (possibly not in that order). Those three components are pretty good analogs of agent, rational, and choose imo.

I find it more handy to just refer to things directly.

Just state that "in a given causally deterministic system there exist antecedents which will cause rain at determined times t and determined places p."

Then you don't clutter a discussion causal determinism with terminology that does not exist in relevant systems. You can construct if to refer to whichever you want, but the simple reality is that it gives illusions of aspects that are not applicable to the system.

I also don't think your assessment of PTB's original "if"-note is correct. So on that we will have disagree.

I agree that it's possible to restate the proposition to avoid using the word "if", and that's what PTB did, in my opinion

However, what I don't agree with is the idea that conditionals are meaningless in the context of a discussion about a causally deterministic system, whether the conditional is expressed as "if X then Y" or in some other form.

Your restatement is not quite equivalent since it does not contain any explication of which antecedents will cause rain, but it would be simple enough to add it (...there exist antecedents a...) and in so doing you are creating an implicit conditional: if a then it rains.

This is a digression and maybe it's not worth belaboring the point too much further, but I believe you are misunderstanding the relation between conditionals in logic and the consequences of causal determinism, and I think that's clear when you say "terminology that does not exist in relevant systems". The proposition itself does not exist within the system, and so neither does the conditional. Even beyond that, it seems dubious to suggest that the word if (or some equivalent term to denote conditionality) can't exist in such a system. Speaking as a computer programmer, that is trivially false.

I have not in way or form said that the word could not exist in such a system. I have said it was not applicable. Obviously given imperfect knowledge, it would be useful in a model - but we don't need that in hypotheticals.


I was a computer programmer myself once, before I decided to upgrade to humans.

I tend to believe it is the best background possible for discussing free will. Programmers build fixed systems, conditional systems and (pseudo)random systems very easily, which makes understanding the concepts much easier. A simulated causally deterministic system would be fairly simple; Given identical starting conditions, programflow would not differ and final output be identical.

A possible hook, time for one last go. There is a AI theorist, Eliezer Yudkowsky, well known for apply programming concepts to philosophical problems in an effort to dissolve them. He is a compatibilist. Worth pointing out that the two arguably biggest advocates for computational theory of mind (Daniel Dennett and Steven Pinker) are also compatibilists, along with Nobel Laureate in physics Steven Weinberg (who knows a thing or two about causal determinism). Not forgetting the aforementioned Hume, Locke, Hobbes, and Schopenhauer. It should probably concern you that the big names attached to your incompatibilist view are William Lane Craig, Alvin Plantinga, Kant and Sam Harris.

I dont get it, I guess I am just dense.

For example, I am guessing that C is agent. But neither of C2 or C3 give me any sort of agent in a deterministic universe. If you want to argue that we are going to (arbitrarily, or for "best use") define an agent as C2 or C3,then fine, but that seems to be just playing games to me.

tame_deuces: I thought I remembered you having a comp sci background, but then you said psychologist and I figured I was just being dumb

This should make it easier to make the argument I want though.

Think of the definition of the program itself. It contains variables which will be set from the initial starting conditions (either straightforwardly or after as complex of a process as you wish), and it may also contain conditional logic which acts on those variables and produces the output. The program flow is determined, but the internal logic of the program is that which explains why a given input is associated with a given output. The determinism of the program can even be summarized as "If initial condition A then output B". That statement is meaningful, and it's clear that the "if" operates over the set of possible initial conditions, it's not "within" the system but outside of it.

The definition of what "raining" means is exactly the same in its usage of conditional logic. The logic isn't part of the causally determined system, it's part of a description or explanation of it. So it's true that "there is no if in a deterministic system" in terms of how causality is actually carried out, but conditionality is still perfectly meaningful in a description of that system (If A then B) or in a definition that distinguishes one set of phenomena within that system from another (if conditions A are satified, that is defined as "B")

Sure, I have no problems with that. Conditionals are of course perfect for reducing the number of written instructions.

You could for example do
instead of
We don't disagree on this. We also both know that any highlevel instruction will in the end (through a compiler or interpreter) be translated into fixed order machine language. In this sense a conditional makes a fine model for the "laws" that govern a causally deterministic system.

I, however, still think using "if" in this context carries the danger of masking what we are really discussing. Most people don't have background of your type, and intuitive understanding is not really enough here.

I'm actually fairly certain we could could argue that imperfect mental modelling of the world around us leads to an intuive grasp of the world as "conditional" - which would in itself adequately explain a sense of free will (regardless of whether we actually possess it or not).

That is another discussion however.

Hey, just to probe a bit further with this - if my view of moral responsibility is a building, do you think your objections are equivalent to requiring more storeys or stronger foundations (or both)? Thanks for the great discussion btw, its been very helpful.

I suspect it's an imperfect mental modelling of the world around us that leads others to an intuitive grasp of the world as causally deterministic (or deterministic).

PairTheBoard

Fair enough. For what it's worth, I am not aware of Coyne saying that we should reject the use of conditionals. As far as I know, he rejects compatibilism for essentially pragmatic reasons--he thinks it is confusing and a waste of time to attempt to come up with coherent ways of talking about free will that are compatible with determinism (I've only read his blogposts on the issue).

As for tame deuces, he also eventually concedes that there are in fact coherent definitions of free will that are consistent with causal determinism, but claims that they are "not interesting." Since this is a value judgement, obviously others can (and do) disagree with him about what is interesting.

Beyond that, I have no clue what tame_deuces is trying to say about "if" here, although I obviously disagree with the claim that conditional statements cannot be true if causal determinism is true.

But, since we are giving illustrations in this thread, here is one for my what I was saying over in that other thread. Well named says above that the OP is about whether there is a definition of "free will" that is compatible with causal determinism. But if that is the issue, I wonder how anyone who understood the issue could ever not be a compatibilist. The answer to that question is trivially yes. I'll concede that the issue is still worth talking about here, as even questions with trivially correct answers sometimes need to be discussed, but if anyone serious argued that the answer is no I would presume that they just didn't understand the issue (e.g. both Coyne and tame_deuces grant that the answer is yes, but they still seem to view themselves as incompatibilists).

Actually no, there is no concession. I made it clear from the start that I rejected "definitions of free will that are not meaningful" and that one could resolve my protest by "stating that free will is causally deterministic". The last one is what PTB did.

The form of "compabilitism" that goes in this thread are the equivalent of claiming alligators are cuddly because they might be sedated, toothless, partially paralyzed from the neck down and have muscular twitches that happen to be very similar to hugs.

This has not been said. I don't even know whether conditional statements can exist in a causally deterministic world, so why would I have an opinion on that?

I have said they are not applicable, which is fairly trivial. A causally deterministic universe consists of a chronologically linear causal chain of determined events. Everything is caused by antecedent. Programflow is a good analogy; You can have high-level structures that seem conditional, but at the bottom everything is actually a fixed chain of binary events.

I suspect what you are doing here is (like some others in thread) treat causal determinism as if our universe is or is not causally deterministic. This (for obvious reasons) is a fairly bad approach.

The best way to discuss this is probably completely through the use of virtual structures. A causally deterministic world is a world of a chronologically fixed causal chains of determined events.

Thus, In causal-deterministic world F:
World starts: Foo and Qux exists
Antecedent Foo causes determined Bar
Antecedent Bar causes determined Baz
Antecedent Baz and Qux causes determined Quux.

etc.


It is unproblematic to write this as:
if (foo) bar.

And as a model, this would actually work perfecly for an observer:
if (foo) quux

If the events were similar enough (let's say both foo, bar and baz all were variants of "rain"), this might even model all of the above phenomena sufficiently to be good for everyday use ("I will get wet").

I don't follow. The philosophical definition of agent in simplest terms is "a person or other entity, human or any living being in general, or soul-consciousness in religion)". I guess I assumed you were asking the less trivial question of under what conditions agent is responsible for a decision. C1-3 give different three different conditions. The aim of the chart is to separate out questions that often get muddled together.

Take the sentence "I chose coffee instead of tea and I was free to do so"

A1-3 addresses "free"
B1-3 addresses "choose"
C1-3 addresses "I"

C2 is roughly equivalent to "I wanted to". So the sentence becomes "I wanted to choose coffee instead of tea and I was free to do so". There is nothing in determinism that contradicts the statement "I wanted to".

Adding in A2 gives us "I wanted to choose coffee instead of tea and would have done otherwise if I had different higher-order volitions". Again, nothing here conflicts with determinism. It is perfectly valid to consider the counterfactual situation where my higher-order volitions were different.

Adding in B2 gives us a final, compatibilist, "Upon reflection, I wanted to choose coffee over tea and would have done otherwise if I had different higher-order volitions". Again, nothing here conflicts with determinism. Conscious introspection, intention and reflection - if anything - requires neural states to be (largely) deterministic.

Now, the strong libertarian version of this would be:

"My transcendental self acting as the initial cause and aided by supernatural reason, chose coffee over tea and was able to do otherwise in identical circumstances."

Clearly the last part is in conflict with determinism and so is the initial cause part. But the compatibilist version of a free will act is clearly not in conflict with determinism. To quote from Original Position (emphasis mine):

Hence my frustration. And it's not like I'm redefining free will as "ducks go quack" and then claiming that free will is compatible with determinism. The sorts of conditions I've talked about are natural, obvious and everyday matters that reflect how we actually talk about choice: I thought about it, I wanted to do it, and if I'd wanted to do otherwise I would have, but I didn't.

What's free about that? Well tame_deuces example of Free Willy is actually perfectly usable. Whenever we use the word "free" in relation to something - be it free speech, Free Willy, a free man - we have in mind various conditions. Speech is free if there is no government censorship. Willy is free if he gets out of SeaWorld (not seen the film, so vamping on this bit). A man is free if he is not imprisoned. Similarly, for incompatibilists, the will is only free if an act of will is not caused by antecedent conditions. Why? Because theologians demand it. And instead of saying "well that sounds mental" and setting out more sensible conditions, the hard determinists go along with the priests. Lunacy imo.

It depends on what “antecedent conditions” entails. I don’t think I’m completely free because I don’t appear to have a choice in the matter when it comes to choosing the good, the greater of two goods, or the lesser of two evils. So that I always do what seems to me "the good," I see as being hardwired or determined, because I'm not free to do otherwise. Where I do believe I’m free is in choosing or deciding what “the good” is, which in the causal determinism schema can, and usually does in my case, function as an antecedent condition to the act.

It is perfectly possible that I dont understand the issue.

I realise that , but just because we talk that way, doesnt mean these things are real. This is what I dont get. We can say "I chose X" but the choosing is also deterministic, its like those games where you drop a marble in the top, and it bounces down some pins and comes out a certain hole, there is no entity choosing which hole the marble comes out of, the marble goes in the top, and comes out the bottom. Even reflection and intention are just part of the same process. There is no "I" that is choosing, and there is no "I" that is reflecting.

What you seem to be talking about is acting as if determinism wasnt true. Which is fine. Acting as if determinism isnt true can be part of the marble game, where the marble may come out of a different hole if one of the pins is "I can choose". But that system is still deterministic. And the agent isnt responsible, either for his good or bad choices.

I realise that I may still be missing your point, sorry if I am, as its probably getting quite annoying.

Obviously if you are trying fit this into _our_ universe, this is a completely different debate. I was under the impression this was a debate on determinism vs free will. I see no point in assuming our universe is causally deterministic, this would be a very bad approach - as we don't know if it is, and it might contain properties that are not present in a causally deterministic universe.

Furthermore, your cookie-cutter characterizations are self-defeating. You _are_ aware that many christians do not believe in free will, yes? Free will is actually in opposition to the doctrine of many Christian denominations.

Your post is, however, revealing. Maybe instead of replying based on what you think people are, you should try replying based on what they say.

Hard to answer. I'm a hard/pessimistic incompatibilist, if I knew what intuitively satisfying definitions of fw/mr might look like then I would poooof stop being that. The impasse is that while I recognize and accept the distinctions you make between hurricanes, toddlers, insane people and sane morally responsible adults, I just think that's all ultimately incidental; the morally responsible adult has no more say in being exactly as she is than does the toddler and the hurricane.

Here's that four case thing I found it (with some guy's comments on it which I haven't read). I don't think it will persuade you but it's good to know about if you're interested in this topic.

This all seems fine to me. I'm not claiming that we are "completely free". Far from it. I'm arguing that we are, in certain circumstances, sufficiently free to ground a naturalistic account of free will.

++++++++++++++++++++++++++

Agreed. That is a reasonable, if hyper-simplified, summary of the compatibilist position.

It's fine, you are clearly trying to understand, so the failure is mine.

++++++++++++++++++++++++++

Fair enough.

Depending on how much work the word "exactly" is doing, I agree with this paragraph.

Thanks

++++++++++++++++++++++++++

I laid out my definition in the OP. It says nothing about whether we are talking about this universe, a parallel universe or the Marvel universe. You have been going off at mad tangents the entire time and, despite a number of regs telling you the same thing, you have just continued trotting out the same irrelevancies. It's one thing to stick to your guns when people are telling you that you are wrong, but if numerous people are saying you've misunderstood the issue then it would be seemly to stop and reconsider.

You don't need to assume this. The poll in the OP asks about whether free will is possible IF (causal) determinism is true. You have not read my OP nor my subsequent posts properly and this is why you've had so much pushback.

There are so many things wrong with your argument in this paragraph (over half a dozen unique problems) that I'm going to assume no third party will see it as being worth my time addressing it.

This sounds similar in some ways to Harry Frankfurt's (compatibilist) theory of free will. I will definitely disagree with your claim that I always do what seems to me the good.

Harry Frankfurt (1982) presents an insightful and original way of thinking about free will. He suggests that a central difference between human and merely animal activity is our capacity to reflect on our desires and beliefs and form desires and judgments concerning them. I may want to eat a candy bar (first-order desire), but I also may want not to want this (second-order desire) because of the connection between habitual candy eating and poor health. This difference, he argues, provides the key to understanding both free action and free will. (These are quite different, in Frankfurt's view, with free will being the more demanding notion. Moreover, moral responsibility for an action requires only that the agent acted freely, not that the action proceeded from a free will.)

Yeah, I don’t consider myself maximally free in regard to the will, but I do consider myself a free-agent. And as Frankfurt seems to suggest, free-agency is all we really need to ground moral responsibility, whether theistic or secular.
Reflectively, I do as well. In the moment, however, I justify eating the candy bar by ranking the immediate gratification above the long-term consequences.

I don't know if this was ever resolved, because I don't feel like reading through the comments right now, but determinism is, well, true. Randomness is but a human construct. Everything is physics.

These are the same. "Determinism" only makes sense within the space-time continuum. There was a non-time when time did not exist. I do not know how to describe that...state (if any word can suffice). I do know that determinism could not have existed then, by definition. But as soon as matter and energy and spacetime emerged, then determinism determined the 13.7 billion years that ensued. Us certainly included.

I don't think anyone knows anything for "certain" beneath the planck scale yet, you'll need a working theory of unification for that.

As for the rest, determinism is fine for models - and that's the best we can say for it; As a mental construction it works most of the time.