Wednesday, November 25, 2015

Hard and Soft Determinism

Soft and hard determinism are the same kind of determinism. The difference is that hard determinists say that since determinism is true, we aren't responsible for our actions. Soft determinists say that even though determinism is true, we are still responsible for our actions. The ultimate causes of our actions are outside our control, but the immediate cause of our action is our desire to perform the action, and that makes us responsible.

Or does it? Shouldn't the ultimate cause be what counts?


oozzielionel said...

VR: " The difference is that hard determinists say that since determinism is true, we aren't responsible for our actions." It seems more likely that critics of those who hold to some form of determinism claim that it implies we aren't responsible for our actions. Are you saying that there are determinists that claim we aren't responsible?

Victor Reppert said...

Are you saying that there are determinists that claim we aren't responsible?


oozzielionel said...

Dawkin's view seems to be catching on in the popular media as well. I have notices that every time there is a mass murder or other vicious crime the talking heads try to analyze the underlying cause. Psychotic disorder; societal structures; chemical imbalance; mental break; abusive childhood; economic disadvantage. It is never attributed to a self-serving decision. The immediate turn is to make sure this never happens again by making a law or improving mental health programs. Yet there is also a call to "hold somebody accountable."

Joseph Hinman (Metacrock) said...

That is hardly Dawkin's invention. One need not be a genetic determinist to hold it. When I was a kid conservatives blamed that sort of analysis on sociologist, who are notoriously environmental determinists.

Dr. R the thing is I don't accept the premise. The question you asked implies that we all agree with the latter sort of determinist (soft) but I don't agree with either. I'm a free willer. Of course there are deterministic aspects but For the most part we can chose moral in matters of ethics and morality.

If I did agree with moral determinism I would say that variables are too complex to understand the ultimate cause.

oozzielionel said...


When there are deterministic aspects to a free will position, is that soft free will?

Unknown said...

It seems to me that soft determinism is the same as compatibilism which, in turn, is no different than determinism, interpreted from the ultimate cause of actions. Hence, one is either a libertarian or a determinist, and there is no in-between, a la Peter van Inwagen.

Look at psychology. People's actions are no longer the result of one's decision to be rational or irrational, ethical or unethical, but because of one's psychological disposition. That presupposes a deterministic view of free will, even if it is partly true on libertarianism. I can easily say the actions of X were evil because of A, B, and C vices and because X suffered from narcissistic personality disorder, but can the determinist say the same?

Angra Mainyu said...

Hi Doug,

A compatibilist does not need to be a determinist. Some compatibilists (like me) take no stance on whether determinism is true. Rather, our position is that we act of our own free will, are morally responsible, etc., regardless of whether determinism happens to be true.

Also, I would disagree that one's actions aren't the result of one's decision. They are, regardless of whether they're also the result of a myriad of other causes that aren't morally relevant.

Angra Mainyu said...


Hard or soft determinists are not always committed to there being such thing as "ultimate" causes. They may well have no stance on whether there is a first cause, or there are infinitely many ones, so no ultimate one as far as I can tell (unless you have a definition of "ultimate cause" that would always yield an ultimate cause, in which case I'd like to ask what that is).
That aside, you ask: "Or does it? Shouldn't the ultimate cause be what counts?"
I don't see why, or why there would be a burden on the part of the compatibilist - who needn't be a determinist; she might take no stance on whether determinism is true.
The compatibilist might say: As long as Joe decided to kill Bob of his own free will, then he's responsible. But why would his actions not be of their own free will just because they were also caused by previous events?

Victor Reppert said...

But it is implied that, if determinism is true, given circumstances outside his control, he could not have done otherwise. If you aren't responsible for the circumstances, and the circumstances guarantee your actions, wouldn't that mean you aren't responsible for your actions either.

Angra Mainyu said...

Assuming determinism, he could have chosen to do otherwise if he had decided to do otherwise, and that was under his control. The fact that his decision to do otherwise had earlier causes does not seem to preclude that he acted of his own accord (or his own free will, which seems to mean the same).
That the circumstances "guarantee" your actions uses the ambiguous term "guarantee", and it might give the impression of bypassing - i.e., that you don't cause your actions.
But the circumstances cause your existence, etc., and then you cause your actions, so I don't see the problem.
In other words, I don't see why just because you're not responsible for circumstances that cause your existence - and so, of distant causes of your actions - you wouldn't be responsible for your actions - of which of course you are a more proximate cause, and regardless of whether those distant causes are sufficient causes or not.
Of course, I disagree with the analysis of the concepts of free will, moral responsibility, etc., that defenders of the Consequence Argument implicitly or explicitly assume or defend.

Incidentally, given that you defend both theism and the Consequence Argument (CA), what do you think about the following parallel?

Let's say that God knows in the distant past what you will do X in the future. Then, the fact that God has the state of mind he has - whatever that is; he has a belief if God's has beliefs, or some sort of intuitive apprehension of truths, or whatever it is, but it's a concrete state of the world in the distant past - guarantees that you will do do X in the future. So, given circumstances outside your control (namely, God's state of mind in the distant past), you could not have done otherwise.

You might have to pick among the following options:

1. The Consequence Argument against compatibilism is not successful.
2. It's not the case that God knows the future.
3. God is timeless.
4. Only circumstances that causally determine the future preclude free will, but not circumstances that determine the future non-causally.
5. We're not responsible for our actions.
6. Any alternatives?

Option 4. seems to be preferred by Craig, but does not seem compatible with usual versions of the CA, and in particular the one you seem to be pressing, because it remains the case that the circumstances in the distant past are circumstances you're not in control of, and given those circumstances, you couldn't have done otherwise.
Option 3. probably commits you to tenseless time, but even then, it can be argued you're not responsible for God's timeless state, either.

Steve Lovell said...

Hi Angra,

Haven't got time to respond fully, but it's nice to have some proper philosophical content on here.

It should be pointed out that the "parallel" argument seems to work just as well without reference to God's foreknowledge ... If that argument is a good one, the mere existence of truths about the future will also work.

The resulting argument for determinism (or fatalism if that is different), doesn't have many defenders ... and for good reason.

Victor Reppert said...

Oh yes, I have always considered this to be one of the stronger arguments for open theism.

Angra Mainyu said...

Hi Steve,


I would argue that there is a subtlety that makes the God-based argument relevantly different from the argument based on future truths, for the following reasons:

Let's say that God is essentially omniscient, and his knowledge includes knowledge of the future.

So, for every possible world W, and for every t and t+s, at time t God knows what the state of the whole W at t+s will be. For example, at W0, God knows that at t+s, E will happen.
Let's now consider W1, which has the same initial as W0 segment up to t. That includes all concrete states, including God's assessment that E will happen at t+s. Whether that's a belief, an intuitive apprehension, etc., the point is that it's a concrete state of the world at time t. So, given that God is essentially omniscient, it follows that at t+s, E will happen in W1 as well, and the same goes for every other future event, so W0 and W1 are identical up to t+s, for any s, and thus at every time.

But I don't think the same argument works with truths, because of what we mean by saying "the same past", or "the same initial segment", etc.

For example, let's say that determinism is false, and the universe works more or less as an indeterministic interpretation of QM holds (with some adjustment for gravity, but roughly).
It's true that Obama is not going to quantum tunnel through a wall before 2016.
There is a possible world W2 with the same initial segment up to today as our world in which he will do so.
So, I know today that Obama is not going to quantum tunnel through a wall before 2016, but in W2, I (or my counterpart) falsely believes today (rather than knowing) that Obama isn't going to quantum tunnel, etc.

In my assessment, that shows that when we say that two worlds have the same initial segment (or more informally, that the past is the same up to some point, etc.), we don't mean to include truths about the future, or the epistemic status of our beliefs. Otherwise, we would be assuming determinism whenever we do that.

On the other hand, it seems clear to me that we do include any concrete states of the world, like the state of mind of an agent, so in particular, the state of mind of God is included (if we assume he exists), and given that God is essentially omniscient, the epistemic status is also included as a consequence (even if that inclusion wouldn't hold in regular cases, like our beliefs).

Granted, one might disagree with the analysis of the concept of "same past", "same initial segments" that I sketch above, but I think the example show it matches how we usually use the words.

Unknown said...

Victor Reppert please contact me here > < Thanks!

Unknown said...

Angra, I hope you pardon me for not responding after you once referred to my Aristotelian-Thomism as bull****.

Victor, I'm sorry for bringing this up on your blog, but there are few venues I can point this out. If you feel it's inappropriate, you are well within your rights in deleting this post.

Angra Mainyu said...


No problem, if you want not to respond to my substantive points, that's your choice of course.
I don't know why you would raise a charge of inappropriate behavior here (it seems implicit in your post that I shouldn't have called it "bullshit"), but in any case, I don't recall calling Aristotelian-Thomism "bullshit". I searched the forums where we interacted, but I haven't found it, either.

So, in order to address your reply, I would like to ask where I said that, so that I can give some context to my words.

Angra Mainyu said...

Addition: if you don't want to post a link here, you can contact me by PM at TFT. But since you brought that up here, I'd like to ask what it's about.

Angra Mainyu said...


I've just done a considerably more thorough search of my posts at FRDB and other places.
While I have found a good number of posts in which I criticize some aspects of Thomism (especially some of the moral claims about sex and/or reproduction), I still haven't been able to find any post in which I call Aristotelian-Thomism "bullshit".
Google searches failed as well. I'm not sure where else to search.
I'll keep searching, but since you brought that up, I think a link would be useful, so that I can look at the context - or at least, I would like to know on which forum, blog, etc., I said that.

Unknown said...

Is there a new website? The one we conversed on is some medical site.

Unknown said...

There are other reasons I won't debate you unless you change your debate style. You spent hundreds of words refuting Platonism, even though my position was AT. I think you also have a tendency to filibuster. It's nothing personal. You have a much better attitude than many, if not most, on that forum. However, I'm trying to distance myself from anything that reminds me of the forum and focus instead on professional philosophers.

Angra Mainyu said...

Yes, there is a new website, called "talkfreethought" (which is pretty much the same), though I don't post very often and ad length anymore, partly because debates are usually too hostile and I got tired of that, and partly because I'm spending less time debating these days, due to meatspace commitments. But you can still send me an email or PM through the site if you register, if you want to. Also, the old forum is archived at , but I think one needs to register to TFT in order to access the archives.

Regarding Platonism, I argued against it on some occasions, but I don't recall focusing on that in the context of our exchanges. If I did that, I'd say chances are it was clear in context I wasn't attributing that position to you. I've again been unable to find any such exchanges, though maybe you're saying I was arguing against Platonism without calling it so?
Still, without a link, I can't address your claim in more detail.

That aside I don't agree with the suggestion I have a tendency to filibuster, either. I sometimes write long posts (or did) but that's in order to be thorough - I don't avoid discussing the issues that you raise.

I'm not asking you to debate me, by the way. If you're just not interested for whatever reason, that's your call and I won't insist. But I was (and I am) surprised by the charges.

As far I remember, our exchanges were usually civil in spite of our moral disagreements - surely a lot better than most exchanges there or generally on the Internet.
However, without something more concrete (i.e., some links that you could provide in support of what you say, or at least some context I might use to narrow it down), I don't have a way to address them any further.

Unknown said...

Good, then we're agreed. We won't debate any further.

Steve Lovell said...

Hi Angra,

I don't think your attempt to put distance between the argument from Omniscience and the argument from Future Truths works.

First, it assumes that God is in time, and while you couldn't see how that might be relevant before, it surely is here.

Second, while you've presented an alleged disanalogy, it isn't at all clear how that disanalogy would undermind the argument for determinism from the existence of truths about future contingents. The implication seems to be that free will would require us to be able to change what God foreknows. But even if that were the case, why would it be any more problematic than changing the truth value of statements about future contingents? Because one is "concrete"? I don't buy that.

Third, as a mere matter of logic I'm not at all sure that any logically or theologically dubious consequences would follow from our having the ability to change what God foreknows. If we believe in "libertarian" freewill then we think that for any free action we could have done otherwise. Of course we in fact don't do otherwise from whatever we do, but we could. So we could do the thing which is contrary to what God foreknows, we just don't. One could take this to imply that we have the power to change what God foreknows.

Perhaps we need to set out the argument from foreknowedge in a structure form and the parallel argument from future truth, and see if we can make sense of denying premises in the latter without denying the equivalent premises in the former?

Unknown said...

Angra, I owe you an apology for bringing this up on a public blog. If anything, I should have emailed you about it. I don't plan on joining any more forums. My experience on, among many others, was a disaster. Be well.

Unknown said...

By the way, how are you searching when all of the posts on are gone? Have they been transferred elsewhere?

Angra Mainyu said...


In re:

1. Time:

That argument is based on the hypothesis that God is temporal at least with the universe. I actually pointed that out in my reply to Victor (see above), when I added the option "God is timeless." to his potential replies. But the option "God is timeless" has other difficulties, and furthermore, one can make a similar parallel argument saying one is not responsible for God's timeless state. But I'll leave that aside, since I didn't make an argument for it.

2. Truths about future contingents.

My argument works by means of analysing the concepts of having the same past and/or the same initial segments. My assessment is that those expressions do not mean that the truths about the future are the same, nor do they imply it.
For example, it makes perfect sense to say that if something like an indeterministic interpretation of QM is true, then there is a possible world W5 with the same past as our own up to yesterday, in which Obama quantum tunnels through a wall before 2016. In that case, W5 has the same past as the actual world A up to yesterday, but not the same truths about the future.
In short, my assessment (after analysis how we are disposed to use those words) is that having the same past up to a point does not mean having the same truths about the future.
Similarly (though I didn't make it explicit before for reasons of brevity, but the conceptual analysis is similar), expressions (in definitions of "determism") like "a full true description of a world W at some time t" do not include future truths in what counts as a full description.

For example, let's consider the following conception of determinism:

D1: Determinism is true in world W iff, from a full true description of W up to time t, plus the laws of nature or whatever laws hold in W works by if any (including supernatural, etc.), a full description of W at any time t' later than t can be derived.

Under that conception, neither future truths nor the epistemic statues of our beliefs (i.e., whether they constitute knowledge or not) are included in a "full true description". But on the other hand, the state of mind of an agent A at time t is included; that includes an agent's beliefs, predictions about the future, expectations, etc., and so does the nature of the agent in question. In the case of God, given that his nature includes omniscience, the epistemic status of his beliefs end up being included as well.

3. "The implication seems to be that free will would require us to be able to change what God foreknows."

Actually, the implication is that if God is essentially omniscient and that includes knowledge of future contingents, determinism is true. The further implication that we don't have free will is only a problem for incompatibilists.

4. The argument from future truths for determinism doesn't work because future truths are not included in a description of the world. If you think the matter is still unclear, I can write down the argument, but in order to compare it, I would ask for your argument from future truths (including a definition of "fatalism", or "determinism", or whatever the argument you have in mind supports).

Angra Mainyu said...


Apology accepted.

With regard to the posts, they're not really gone, though it looks like that to non-members.
If you're registered and logged in, the number of posts still appears to be zero, but you get a message on the first page saying that the archives are available, and they are - you can search through all of the old posts.
I understand you don't want to register, so I guess you can't search, though if you're interested in some of your old posts, you can always register under a new user name that nobody knows, never post, and search all you want.

For example, in the case of AT, I searched as follows:

1. I searched for all posts by Angra Mainyu containing the word "Thomism", searched each of them (by reading them or with the browser's own "find" function) for the words "bullshit", "BS", "bull", "bullcrap", or "bull****" (with the asterisks).

2. I searched for all posts by me containing the word "zzzz" (where I write your user name on that forum instead of "zzzz"), and check them as well.

3. I search for all posts by me containing the word "Thomist", check them, then "Aristotelian", and check them too.

4. I searched for all of the posts by me containing the words "bullshit", "bull", and "bullcrap", ("BS" was too short for an independent search, and "bull****" has asterisks that seem to mess with the search function, so I got those covered by searching for the other words (i.e., "Thomism", etc.), and then check whether they contained the words "BS" and "bull****" as described in points 1., 2., and 3. above). (by the way, there were very few posts in which I used the word "bull..." etc., and always in cases in which I was replying to someone else using the word in question first).

I also searched for "Platonism", etc., though one can argue against something like Platonism without calling it so, and the search would take a very long time, so I can't be certain in that case.

Unknown said...
This comment has been removed by the author.
Unknown said...

Fair enough. I'm not able to find the citation in which you refer to AT as bull****. However, do you remember the formal discussion we had over Thomas' argument from De Ente et Essentia? You spend a lot of time on Platonism, even though my argument had to do with AT.

Angra Mainyu said...

Right, I remember the debate.
It's from 2010, and memory isn't always reliable, but luckily I found the debate.
If you send me a PM or email, I can give you a link. Alternatively, I can email you a copy of the whole debate if you give me your email (I tried to send you a PM from the archived forum; I'm not sure you'll receive that).

Anyway, you offered the example of Platonists as evidence against my analysis of the meaning of "explosiveness exists", and my use of that example. But while my assessment would be a problem for Platonism, I didn't present that example as a challenge to Platonism. Rather, it was part of a parody parallel intended to challenge the coherence of some of the AT claims (whether it succeeds or not is another matter, but either way, I wasn't trying to debunk Platonism).

In reply to your example, I said (in 17 words) that Platonists were mistaken about the meaning of the words, and then spent 81 words arguing for a conceptual point I had already argued - and which I didn't originally intend as an argument against Platonism -, but in the latter case, I did so while replying to one of your Platonist examples.
So, if we count the second one as arguing (albeit incidentally) against Platonism, that would make a total of 98 words. Even so, I was only addressing your objection, which used Platonism as an example - I didn't introduce the issue of Platonism.

On the other hand, I did spend lots of words challenging the coherence of some of the AT claims, and I get the distinct impression that you may have thought I was targeting Platonism, rather than AT. Generally, there was a considerable amount of miscommunication in that thread. I also get the impression that neither of us was very happy with how the debate went down, and even that it was somewhat more heated than most of our interactions, though still definitely civil.

In my assessment, our more recent exchanges (from 2013) generally went better.

Unknown said...

You actually used at least 302 words, starting in your first post. Moreover, I stopped counting about halfway through.

This is evidenced by the fact that I respond: "Moreover, you have committed a category mistake in analogizing 'murderousness' with existence. Murderousness, I take it, is an abstract object, some kind of generalization we make about the set of murders." So, it wasn't even me who started talking about Platonism.

I do agree that our later exchanges went better.

Angra Mainyu said...


I said I used 98 words at most arguing against Platonism, not on the example of explosiveness, or in total in my posts.
As for the parodies, I get that you think they - and the rest of my argumentation - fail, so we disagree on that, as well as on the claim that I incurred a category error, etc. However, what I'm saying here is that regardless of whether the parodies and/or the rest of my argumentation succeed , it would remain the case that the parodies were aimed at your argument (and so, at AT), not at Platonism.

In other words, I was not arguing against Platonism. I was arguing against your argument - (I still believe successfully so, even though I could improve on the replies if I were to argue the matter today).

Unknown said...

So you used Platonic parodies to argue against AT? Pardon me if that doesn't make any sense. Essentially, you're arguing that Platonism is wrong, and since AT has that much in common with Platonism, AT must also be wrong.

Unknown said...

In fact, you didn't even argue that Platonism was wrong. What you did conclude was that murderousness and explosiveness don't exist, but that's not an argument. I'm no Platonist, but I did provide conceptualism as an alternative.

Unknown said...

One more thing, as I implied, "explosiveness" on Platonism and Aristotelianism are two different things, so there's a disanalogy. On the former, explosiveness, like all abstract objects, exist as mind-independent realities. On the latter, explosiveness is a universal that is instantiated in concrete particulars. Hence, I don't see how the parody was supposed to work in the first place.

Angra Mainyu said...

I wasn't arguing that Platonism was wrong at first, when I wrote the parodies. I only argued very briefly against Platonism after you gave the example of Platonists. But it was incidental.
And I don't agree with your claim that I used Platonist parodies. The parodies were not Platonist. The parodies were meant to mirror your argument, not to target Platonism.
Granted, you don't agree that the parodies or the rest of my argumentation succeeds. So, we have a disagreement, and it seems we're not going to convince each other. Do you have any suggestions?

Angra Mainyu said...

Regarding the "One more thing" (I hadn't seen that post when I posted my latest reply prior to this one), the parodies were meant to mirror the reasoning in the argument. The fact that properties are abstract objects is not relevant to whether they are successful. For that matter, someone might conclude that Explosiveness is a concrete being (or substance, etc.), and I would say the problem remains the same.
I gave also arguments independent of the parodies, which are not aimed at Platonism, either.
But in any case, as I mentioned, I don't think we're likely to agree. Also, this thread does not seem to be an adequate place to keep going back and forth discussing a past debate most people here don't have access to.
So, I would like to ask you if you want to continue discussing that debate (or starting again), and in that case, if you could pick a place where we can have a thread about it, post the arguments, etc. Any forum or blog you might choose is okay with me in that case. If you prefer not to discuss it further, of course that's okay with me too.

Unknown said...

Angra, thank you for your congeniality, minus the AT is bull**** thing. ;) I don't have any interest in debating this issue any further. You can have the last word here. If you find this interesting at all, my second Master's thesis (this time in Philosophy) is over Thomas Aquinas' First Way. Afterwards, my doctoral dissertation will likely be over God's relationship to time or God's relationship to abstract objects. When it comes to natural theology, I almost exclusively focus on Thomas' Five Ways.

Angra Mainyu said...


Thanks for the congeniality, minus the saying I said AT is bull**** thing.;-)

And thanks for the offer. While I don't think I can dedicate the time to read an entire thesis on the First Way at the moment (unless it's short), if I you give me a link (or send me a PM, email, etc.), I'll read it later.

Steve Lovell said...

Hi Angra,

I agree with you about the argument from future truth to the lack of free-will and the truth of determinism. It doesn't work. In this guise I call determinism "fatalism" because if such arguments worked, determinism would not be an empirical thesis but a logical/metaphysical one about the nature of tensed facts.

I quibble with your definition of determinism: Whether things can be derived is a separate question from whether they are entailed (see Godel). Determinism only requires the latter, but this is a minor matter.

You agree with my basic response to the argument from future truths to (hard-)determinism, namely that the truth-makers exist at the point of the action and their existence at that time is sufficient to make statements about that time true at all times.

You're saying that the same response isn't available to the person who believes in God's foreknowledge, since his knowledge and it's status as knowledge is supposed to be a concrete state of affairs which obtains at all times. But I think the important thing is the direction of explanation, not the chronological order. God "believes" (ahem) these things because they are true. They are not true because he "believes" them. If it helps, you may consider God's foreknowledge to operate via backwards causation.

Here's your conclusion from way up in this thread:

(C) So, given circumstances outside your control (namely, God's state of mind in the distant past), you could not have done otherwise.

I can't see much of an argument for it. But the idea seems to be that one's future actions have the same modal status as the items in the distant past and so ... um, what? Here's the crucial question about this general direction of thought here: what is meant my "could not have done otherwise"? Even the distant past certainly could have been otherwise.

I could take a decent guess as to how the argument will go from here, but I don't want to put words into your mouth.

Angra Mainyu said...

Hi Steve,

Godel showed that some formulas that are true according to a standard interpretation (in arithmetic), can't be derived from the axioms by some specific procedures.
If so, it seems to me the formulas aren't strictly logically entailed, either, but you might say the truths are entailed because they're mathematical truths and are entailed by everything; is that what you have in mind?
If so, it seems to me that's a broader sense of "entail" than the sense of "derived" in which they can't be derived. But that sense also seems to me to correspond to a broader sense of "derived", and the latter should be the one relevant in D1 (for example, if we can establish what happens in the future from a statement of what happened in the past plus some mathematical truths, that would count as "derived", even if we're using mathematical truths that can be expressed by statements that aren't strictly logically necessary in first order logic).

But I concede the definition was at least unclear; I was trying to capture common usage.
How about the following variant?

D1': Determinism is true in world W iff a full true description of W up to time t, plus the laws of nature or whatever laws hold in W works by if any (including supernatural, etc.), entail a full description of W at any time t' later than t.

That takes into consideration your quibble, though I'm not sure this is precise enough, either (there are issues with the conception of "entails" as well). If you have an alternative definition that captures common usage better, please let me know.

As to your substantive point, my argument from God's foreknowledge to determinism doesn't assume that things will happen because God believes that they will, nor does it assume that the correct description of God's state of mind is given by the concept of belief. Whether it's a belief or another form of intuitive apprehension, it's a concrete state of the world at some time t0, and as such, I argue it counts with respect to the definition of determinism D1 or D1' (but if you have another definition, please let me know and we can consider that one).
So, in short, I think my argument shows that if God exists and has essentially infallible foreknowledge, determinism is true (but I'm a compatibilist, so I hold that's not a problem for human freedom on theism).

Regarding your question about my conclusion, I was mirroring Victor's argument (i.e., where he states "But it is implied that, if determinism is true, given circumstances outside his control, he could not have done otherwise."), and saying that given circumstances outside your control (namely, God's state of mind in the distant past), you could not have done otherwise.

Granted, you might say that you didn't do X at t1 because of God's state of mind at t0 << t1, but rather, God's state of mind at t0 was what it was because you were going to do X at t1.
I would say that that does not change the fact that God's state of mind at t0 is not something you could control at t1 (it had already happened), and given God's state of mind at t0 (i.e., fixing that past up to t1), you couldn't have done otherwise.
Now, to be clear, I hold you were free at t1 because you could have done otherwise if you had intended to, and that's what matters - I'm a compatibilist -, but I don't think that's an argument available to someone committed to Victor's argument, which holds (implicitly) that if you couldn't have done otherwise given something outside your control.

I hope that clarifies my argument sufficiently; if not, please let me know.

Steve Lovell said...

Hi Angra,

I realise this may sound disingenuous, but the way you're using it, I don't know what the meaning of "couldn't have done otherwise given something outside your control".

God's necessary omniscience gives us the conditional:

(Omni) Necessarily, if God "believes" that P, then P.

It does not give us the conditional

(Omni+) If God "believes" that P, then necessarily P.

Except on the basis of something kinda like the second, I don't see how you get "couldn't have done otherwise than X" from God knowing that someone will do X.

Angra Mainyu said...


My argument for determinism (D1' definition, or relevantly similar ones) from theism does not require anything like the second. I'm not arguing for something like modal collapse. I'm arguing that theism (including essential omniscience about future events) entails determinism.

D1': Determinism is true in world W iff a full true description of W up to time t, plus the laws of nature or whatever laws hold in W works by if any (including supernatural, etc.), entail a full description of W at any time t' later than t.

Let's say that in W, at t, God believes/whatever that X will happen at t'>t. Then the description of W up to time t, plus whatever, entails that X will happen at t'. Since that's for every X at t', determinism follows.
If you have a different definition of "determinism", please let me know and we can consider that.

That said, as I mentioned, the "couldn't have done otherwise given something outside your control" part was my argument mirroring Victor's. It's independent of my argument for determinism on theism.
So, the "couldn't have done otherwise", etc., is meant to mean whatever Victor meant with "could not have done otherwise", which seems to be equivalent to saying that given the past/circumstances beyond one's control (i.e., given that that already happened), the future is entailed, or that in any possible world with the same past up to that time, including the laws (rather than in every possible world) , you do the same (not otherwise), or something like that. At least, that seems to be what his argument says.

Another way to see that is as follows: his argument is an argument for the incompatibility of determinism with the ability to do otherwise (and free will, etc.). I showed that if God exists and has necessary omniscience, then there is no ability to do otherwise (or free will, etc.).
You might think Victor means something else by "could not have done otherwise" and/or determinism. But that seems improbable, given the context of his arguments. In fact, when I raised that worry, he reply is that he considers this to be one of the strongest arguments for open theism, not that I was somehow misconstruing his argument.

So, if I need something like the second to conclude that given the past/circumstances beyond your control you could not have done otherwise (in the sense relevant in this context), then the argument I'm mirroring (i.e., Victor's) fails as well, and that's it. My argument from theism to determinism (see above) remains unaffected, as it's independent of this part.

If you have any other definitions in mind, please let me know.

Steve Lovell said...

Commenting from my phone, so I'll keep this brief ...

I don't take issue with much of your last comment, the only question is given the direction of explanation what makes you think that the particular content of God's foreknowledge is beyond our control?

Angra Mainyu said...

That's good question.

Briefly, I would sy that for an event to be within our control in this context, one needs to have have the capability to bring it about/cause it, or to refrain from bringing it about.
But if you say - for example - "I'm going to bring it about that 1000 years God believed(or whatever) that today, I will kill the mosquito on the wall", and then you kill the mosquito, it seems intuitively clear that you don't bring that about. Assuming God exists, has foreknowledge, etc., 1000 years ago he already believed/whatever "On December 4, 2015, Steve will kill the mosquito on the wall" (not saying God would be thinking in English, etc., but I'm human and have to express it humanly), and you're not engaging in retrocausation.

A question is how God knows what you will do. But however it is, it does not seem that it's because you have the power to cause God's beliefs by retrocausation.

Still, I guess you might insist that maybe it's because of that.

For now, I can say the following:

a. The argument that God's foreknowledge entails determinism (D1') still succeeds, so any theist who says that moral responsibility, free will, etc., requires indeterminism would still be committed to open theism. In the case of Victor's position, it seems the result is achieved anyway.
b. If they accept the compatibility of freedom, etc., and D1'-determinism (but Victor does not), saying that "maybe" retrocausation is how God knows, etc., would not be enough: If a theist believes that having control of the events in the past or indeterminism is a requirement for freedom, responsibility, etc., and the person holds that we have freedom, responsibility, etc. but accept determinism (because of the foreknowledge argument), they are committed to the hypothesis that we indeed engage in retrocausation and bring about today that, say, 1000 years ago God believed that today one will kill a mosquito on the wall, etc. That seems very counterintuitive and would need some explanation (in addition to counterintuitiveness, there are other funny results with retrocausation, which they would have to deal with).
c. Additionally (though this might not be a difficulty, depending on the case), it seems they would be committed to the falsity of both presentism and also growing block theory of time, because in 1015, 2015 didn't exist on presentism/growing block, and so God's belief could not have been caused by something that does not exist.

Steve Lovell said...

Apologies for the delay in responding, Angra.

Hmm, I guess that according to your definition, I am indeed a determinist ... though oddly enough this now seems to me compatible with "liberatarian" free will, which suggests that the definition is wrong. I think the "complete description" needs to circumscribed somehow. I need to think a little about that. My first thought was that the description should be limited to a naturalistic ontology. But that isn't right, since there is no particular reason why an opponent of naturalism couldn't be a determinist.

I don't have any particular views on how God foreknows. I guess I imagine this as a direct apprehension ... something like how a Platonist imagines our awareness of "the forms". I have no problem with it being imagined as retro-causation ... though to say that we bring it about that God "believes" (or "believed") things does indeed sound odd. Though I think the oddity is mostly in attributing this power to us rather than to God. In any case, I don't think it is, as you say, "intuitively clear that [we] don't bring that about".

Also, in relation to your point (c) about my being committed to the falsity of both presentism and growing block theory of time ... I'm so committed if you are (by your response to the argument from future truths to determinism).

Angra Mainyu said...


No problem.

Regarding the conclusion that you would be a determinist but determinism would be compatible with libertarian free will, I don't think that suggests the definition is wrong. There seem to be different conceptions of both determinism and libertarian free will. But at least to the best of my knowledge, D1' matches (at least, roughly) a most common usage of the expression "determinism" in philosophy discussions about free will, moral responsibility, etc.
Now, I've also seen a distinction (e.g., made by W. L. Craig, though he's not specialized on that) between whether something is specifically causally determined or not. W. L. Craig - for example - also considers himself a libertarian, but does not hold that if A freely does X at t in W, there is a world W' at which A freely does not X at t in W'. In fact, and while Craig does not say so as far as I can tell, he too is committed to the truth (or at least high probability) of determinism, even though he rejects causal determinism.

Even so, D1' seems to capture the sort of determinism relevant in this context, since it seems to cover the sort of argument Victor defended in the exchange with me (but see below).

With regard to presentism and growing block, I don't agree that I said anything that commits me to that. I'm undecided about growing block, and I hold that presentism is false, but for other reasons. I'm not sure how you're interpreting my reply to the argument from future truths. Maybe there is a misunderstanding here, but I don't see how that would be similar to my point c).
Also, on point c.), I didn't say that you were committed to the falsity of both presentism and growing block theory of time. My point was that having control over what God believed in a distant past would commit one to retrocausation, and retrocausation would commit one to the falsity of both presentism and growing block theories. Since you're not committed to retrocausation, that does not apply to your case, but it would apply to someone who holds that the way in which God knows is by retrocausation.

In re: retrocausation, you say that oddity is in attributing this power to us, rather than to God. I would say that retrocausation brings about a number of oddities, but in this particular context, what matters is whether we have control over the event in the distant past. In order for that to be the case, we would have to have the power to bring about that God believes X in the distant past (i.e., it's causation but under our control). So, if that's where the oddity comes from, the person committed to control over the distant past seems committed to that oddity.

All that said, in the OP Victor was talking about ultimate causes vs. proximate causes, which indicates a different argument against causal compatibilism, but not against D1'-compatibilism, or entailment-compatibilism. My reply to that argument is in my first post in the thread, and his exchange with me ended with his point that he considers this to be one of the strongest arguments for open theism, which would not follow if he only is against causal compatibilism. So, it seems that he's pressing two different arguments, one against causal compatibilism, and one against something roughly like entailment-compatibilism.

Unknown said...

Angra, I know this is off-topic, but you once asked me a question about monogenism. I have a developing theory about this. You asked: was it God's will for there to be inbreeding? My answer now is a definitive no. It's just that as a result of the Fall, inbreeding became one of the consequences of sin.

Angra Mainyu said...


That's interesting.
I'm tempted to raise objections already ;-), but I'll refrain because it's off-topic. If you post it somewhere, please let me know.

Steve Lovell said...

Hi Angra,

Perhaps I've missed a nuance. But in response to the argument from future truths, your position seemed to be that if such propositions are true, then they are made true by facts at the time they are about. But that commits one to the existence of those facts, and therefore of those times, does it not?

Did you meant to remain agnostic on whether there are future truths? Or perhaps you have somethign else in mind.

I feel fairly happy with the causal-determinism vs entailment-determinism distinction and instinctively want to say that I reject the former and accept the latter.

Where does this leave us?

(1) We agree that the argument from foreknowledge supports entailment determinism
(2) Entailment determinism is compatible with libertarian free will

What are we arguing about? ;-) Victor's OP? I'll have to re-read and come back to you. We may be agreeing there too!

Angra Mainyu said...

Hi Steve

I meant not to take position on the matter of future truths. Rather, what I meant to reply is that when we talk about two scenarios, worlds, etc., having the same past up to a certain time t, we don't mean to include in that past any future truths. In other words, saying "W1 and W2 have the same past up to t" (or similar expressions) does not mean (in the context of philosophy discussions on determinism) nor conceptually entail that they have the same past including future truths. On the other hand, it does entail (conceptually; i.e., considering only the meaning of the words including logical symbols, and logic) that any concrete states of the world up to that point are identical.
That makes my argument escape parallels from future truths, unless someone challenged the conceptual analysis about what one means in those cases, but you didn't do that as far as I can tell.

That said, regarding my take on the matter of future truths, I am agnostic to a considerable extent.
More precisely, I think it's proper to sincerely say that (e.g.) "it's true that X" (or similar expressions involving terms like "proposition", etc.) whenever it's proper to sincerely say "X" (for adequate X), at least usually (potential exceptions would be situations in which one is discussing precisely the ontological status of future truths, if they have any such status.)
In particular, as long as we have conclusive evidence of X (i.e., it's beyond a reasonable doubt that X; it's [epistemically] improper not to assign X an extremely high [epistemic] probability), one may properly say things like "it's true that X", "I know it's true that X", etc.

For example, I think it's proper to say "It's not the case that Obama will quantum tunnel through a wall before 2016", "I know it's not the case that Obama will quantum tunnel through a wall before 2016", "it's true that it's not the case that Obama will quantum tunnel through a wall before 2016", "The proposition 'it's not the case that Obama will quantum tunnel through a wall before 2016' is true", "I have the true belief that it's not the case that Obama will quantum tunnel through a wall before 2016", and so on.

But I'm (mostly) agnostic on ontological issues about such propositions (or more generally about propositions) beyond that.

Regarding your points (1) and (2), I agree with (1), but my take on (2) is:

(2.1) There is more than one definition of "libertarian free will" at play in philosophy discussions on the matter, and entailment determinism is compatible with at least one but incompatible with at least one of such conceptions.
(2.2) Some of the common arguments against compatibilism are based on the definition of entailment-determinism. If such arguments succeeded (which I reject), then entailment determinism would be incompatible with free will in the sense that matters (i.e., the usual sense of the words, the sense relevant to moral responsibility, etc., regardless of technical definitions).

With respect to Victor's OP, actually he was talking about causation in it, but he seems to have introduced another argument in his reply to my first post in the thread, so he seems to be pressing an argument against compatibilism that is based on something like entailment determinism, plus an argument that is based on causal determinism (though he didn't flesh out the arguments, so this is somewhat tentative).

Steve Lovell said...

I think I see what you're driving at re future truths, though since it sounds like you're undecided, it's difficult to be sure. One of the options you seem to be keeping open is to ground future truths in current propensities. I've heard presentists take that line, but it's always seemed wrong-headed to me ... not least because determinism is not, or at least not necessarily, bidirectional. So while on determinism the present will determine the future, it may radically underdetermine the past. But if you're going for a "probabilistically open" future combined with a growing block theory you may get the best of both approaches.

Does that sound like you? You say you're agnostic on these things, so you may not want to commit.

While it may give you the best of both worlds, I doubt that that best is very good. And perhaps this is where your deflationary/disquotational approach to truth comes in (not trying to use those terms in an evaluative sense merely descriptively if that matters to you). I think for example that either I will get up before 6am tomorrow morning or I will not. On non-determinism (of whatever kind), it may not be possible for me to say with 100% confidence which of those two things is the case, but nevertheless one of them is the case. Moreover, whichever of those things is the case the statement is about tomorrow and not about today, so trying to ground it in propensities seems wrong.

Does that sound like you?

On VR's original post, I'd have read his arguments as being based entirely on causal rather than entailment determinism.

I'm intrigued by your points (2.1) and (2.2) and would like to hear more on both. I can't say that I find entailment-determinism a very interesting position. It seems to be only a technical by-product of God's foreknowledge, and unless there are other reasons to accept it I can't see how it is incompatible with any concept of free will I've come across. Sounds like you think I'm wrong on this ... but you seem to be more familiar with the literature than I am, so you may have something clever up your dialectical sleeve.

Angra Mainyu said...


I was talking about epistemic probability.
That aside, I'm agnostic on many issues. :-)
In particular, I don't have a stance on whether there is a probabilistically open future in a different sense of probability (if there is any different sense; I find the ontological sense rather problematic). I also have no opinion about whether there is a growing block, or whether future truths - or any truth - need grounding (the grounding relation seems rather mysterious to me).

In any case, I wasn't trying to ground truths on propensities or anything else. Rather, I was suggesting that our colloquial talk about the future (e.g., my example about Obama, or if I say that tomorrow I will not go to the supermarket) is not ontologically committed to one theory or another.

On Victor's original argument, I too read the OP as based of causal determinism, but after my first reply to him, he brought up the issue of circumstances outside one's control that "guarantee" that one would behave in one way or another.
That was a bit ambiguous, but that sort of language seems to be used on entailment determinism too.
So - in addition to a reply from a compatibilist perspective that works both on the causal or the entailment views -, I wrote a parallel argument involving God's foreknowledge, and specifically listed (as his potential replies) option "4. Only circumstances that causally determine the future preclude free will, but not circumstances that determine the future non-causally.". Yet, he actually found the argument to be one of the best for open theism. In light of that his incompatibilism encompasses entailment determinism as well.

Regarding (2.1) and (2.2), I'm not actually so familiar with the literature :-), but for example:

The the IEP entry on free will gives a definition of "causal determinism" that, despite the name "causal", is in terms of entailment. And there is good reason for that, because it then addresses van Inwagen's Consequence Argument (perhaps, the best known incompatibilist argument), which is based on that definition. In fact, the first premise of the argument states that entailment-determinism is true (even if it calls it "causal"), and the conclusion of the argument is that no one has or even had any choice about the anything that happens in the future, and as a consequence, we have no free will; implicitly, this is meant to be in the sense of "free will" that is relevant to morality, i.e., the colloquial meaning of the expression.

Also, here is a comment on Pruss's proof of (an alternative version of) van Inwagen's Rule Beta, which is used in the Consequence Argument, and here is another reply to one of VI's arguments.

If any of those incompatibilist arguments succeeded, then entailment determinism (even if it's called "causal determinism") would be incompatible with free will, and so the libertarian view supported by the defenders of those arguments who also believe in free will, is incompatible with entailment determinism.

Steve Lovell said...

Thanks Angra, I've got some reading to do!

I've read some of VI's stuff, but I really don't remember it. I've presented arguments for incompatibilism myself before now, but while the definition of determinism I've given may have sounded more like entailment determinism, having been shown the error of my ways by you I'd want to rework those arguments now.

As an aside, I didn't take VR to be endorsing Open Theism, but saying that the sort of argument you made are among the stronger considerations in it's favour. If that is his view, I'd tend to agree.

Since I've been confused about your own stance, in case my own has remained unclear (not deliberately), I take myself to be a fairly orthodox Christian. I reject (causal) determinism and accept both God's foreknowledge of future contingents and "libertarian" free will. I take this as incompatible with Calvinism (though I haven't read much on that) and am, at best, sceptical about Molinism (Middle Knowledge).

Angra Mainyu said...


You're welcome, and thanks for explaining your own stance. Just in case parts of my stance are unclear, I'm a non-theist, compatibilist about determinism (causal or entailment) and freedom, power and moral responsibility; I take no stance on whether determinism (of any sort) is true.

Regarding the examples I provided, upon further consideration, I think they might not be the best ones, because while they state the matter in terms of entailment determinism, given that they talk about the laws of nature plus a description of the past, it might be considered that the description of the past only includes nature (whatever "nature" means), so God's mental states do not enter the picture.
However, I think assuming one can properly distinguish between nature and non-nature (which I doubt, but that aside) the restriction to nature would be arbitrary, and is more justified in the context of discussions in which the proponents of compatibilism are also naturalists.
But I just wanted to point out you still have a way to exclude God's mental states from those arguments, and still make the arguments on the basis of some kind of entailment-determinism, as long as the entailment is only limited to nature (as a compatibilist, I don't think the arguments succeed, but that's another matter).

With regard to Victor's take on open theism, he said it was one of the strongest arguments for open theism. That seems good evidence that I had captured the meaning of what he meant by "determinism" - otherwise, I would have expected him to say that my argument wasn't a good parallel of his argument, given that I was talking about a different sort of determinism.
But after looking up some threads on open theism, and in most, he seems uncommitted, so you're probably correct.

Angra Mainyu said...


If you're reading this post, could you clarify a couple of points for us, please?
Do you endorse open theism?
Do you think that if, in W, A at t freely does X, there is a possible world W' with the same past up to t at which she does not do X at t?

Steve Lovell said...

Following up, after reading the relevant portions of the IEP article ...

The consequence argument seems like a good one to me. It's a variant of that argument which I've defended previously. However, if one's reason for endorsing determinism is that one accepts divine foreknowledge, then I think one also has reason to reject what the argument calls the "fixity of the past". So when used as an argument not merely for incompatibilism but for the non-existence of free will, I'll resist the conclusion either by rejecting (causal-)determinism or denying "fixity" (on entailment-determinism).

Angra Mainyu said...


It seems to me you would need a new argument for all intents and purposes, because rejecting that the past is fixed requires rejecting key parts of the argumentation in support of the premises (in fact, the argument goes far enough to say that even if time-backwards causation is possible, the past is fixed), and at least one of the premises.

In any case, leaving aside other reasons, the argument fails because rule Beta is invalid, as van Inwagen has already conceded (see this SEP link), though some potential modifications have been proposed.

Steve Lovell said...

Although I said I'm denying the fixity of the past, that's not because I think we can change the past ... just that we can influence it in some rare cases, including God's foreknowledge. With that in mind I don't think I need to fundamentally revise the argument. If the only escapes from the argument are time-travel, backwards causation or divine foreknowledge, the argument has done plenty of work!

For the record, I think it's only confusion which allows people to think that they can change the future, never mind the past. What was/is/will-be the case at Time t, is not something that it's meaningful to talk about being changed. Things change over time, not at a time which would require being in two incompatible states at the same moment.

Angra Mainyu said...

Why do you think you wouldn't need to fundamentally modify the argument?
The Consequence Argument (i.e., the descripton of the Consequence Argument on the IEP article you're considering) factors in the possibility of time travel and causal influences on the past, but still holds that it doesn't affect the CA - the past remains fixed.
In other words, if the argument is successful, it's not the case that time travel, backwards causation and God's foreknowledge are the only escapes: if it succeeds, those aren't escapes, either.
Moreover, if you accept the premises and the inference rules, the conclusion (i.e., no one has a choice over any aspect of the future) seems to follow from entailment determinism, even if there is time travel or causal influence on the past.
But I might be missing something. What modification do you have in mind?

In any case, one of the inference rules is invalid, so that block the argument independently of other considerations. Do you have an alternative?

As for changing the future, I think one can talk meaningfully about that, in the sense that if we do X, then the future will be different from what it would otherwise be, in a sense that is relevant to us in a certain context, e.g., if we do X, then Y will not come to pass, but if we don't, then it will (i.e., it's extremely probable, etc., that it will happen).

Steve Lovell said...

I think I expressed myself poorly.

First off, I reject the "no-possibility-of-change" = "no-control" equivalence. This rejection is important given that I don't think the future can be changed either. For anyone just reading this comment, that's not to say that I'm a determinist. Far from it. I think the future is "causally open".

The fixity of the past, for the purposes of the consequence argument, consists in our not having control over things which, according to other premises, "determine" our behaviour. But on entailment determinism the determinants can include things over which we do have control (such as God's foreknowledge). I think that's also the case on the possibility of time-travel (if we have control over our use of time travel!?).

While these things allow us to deny the fixity premise, they strike me as being trivial loopholes in the argument, and leave the argument fundamentally unaffected, since they aren't the kinds of determinants which we're most interested in considering.

I think the variant of rule Beta in which uses [](p->q) rather than N(p->q) looks pretty good to me (where "[]p" = "Necessarily, p" and "Np" = "No one has control over whether p").

Angra Mainyu said...

Steve, thanks for the clarification.

Regarding beta2 (from the SEP article: "From Np and □(p ⊃ q), we may infer Nq") As a compatibilist, I don't agree with that rule, so that's part of (or all of) our disagreement on this matter.
As for the rest of your variant of the argument, I'm not sure what modifications you introduce, so I don't know whether there are other disagreements.
For example, the original argument uses beta more than once. But beta2 yields (from ☐(P ⊃(L ⊃ F)) and (4) NP) that N(L ⊃ F), but then, beta would seem to be required in order to continue the argument. Yet, we only have beta2 now.
So, it seems more modifications are needed, and there are a number of different ways in which you might want to change the argument, but I'm not sure which one is the one(s) you find persuasive.
If you don't mind, I'd like to ask what the premises in your version of the argument would be.

Angra Mainyu said...

Steve, after thinking about beta2 a bit more, I think I've come up with an argument that might convince you that it's invalid, or at least that some of the other premises of the argument fails.

Q1: N(q&¬q).
Q2: □(q&¬q ⊃ q)
C: Nq.

For some q, C is false - as you agree-, but C follows from Q1 and Q2 by beta2.
So, if you accept beta, you're committed to rejecting either Q1 or Q2.
I'm pretty sure you won't reject Q2, so the only option left would be to reject Q1.
If you do reject Q1 (i.e., if you think for some q, someone has control over whether q&¬q), then I'm not sure how you could justify the premises asserting N(something) in the rest of your argument. But I'll wait to see the rest of your modified argument.

Steve Lovell said...

Hi Angra,

Apologies for the delay in responding. I've not been at a computer for a few days, and don't like doing these things from my phone. I like your little argument, and after a little inner turmoil find that I am indeed convinced. In case you're interested, I'd initially misread your argument as the slightly more colourful:

Q1: N(p&¬p).
Q2: □(p&¬p ⊃ q) [anything follows from a contradiction]
C: Nq.

Which also seems to do the job.

This mostly makes me wonder whether Np is correctly formulated, and here Frankfurt cases are also relevant. Rather than control over the truth value of propositions, the relevant control is of course control over our actions (or other events/objects/states-of-affairs).

This suggests the following general outline for a revised version of the argument:

(1) I don't have control over state of affairs A (deep in the past)
(2) State of affairs A causally necessitates any later state of affairs (determinism)
(3) If A causally necessitates a later state of affairs, and I don't have control over state of affairs A, then I don't have control over that later state of affairs
(4) Therefore, I don't have control over any state of affairs later than A

Which is good for any state of affairs causally necessitated by some state of affairs deep in the past. Clearly, premise (3) is the correlate of rule beta. More formally (my logic is a little rusty, so forgive me any minor slips):

(1) ∃(x)∀(y): Nx & Cxy
(2) ∀(y): Na & Cay
(3) ∀(x)∀(y): (Nx & Cxy) ⊃ Ny
(4) Therefore, ∀(y): Ny

x is any state of affairs
y is any state of affairs later than x
Nx = I don't have control over x
Cxy = x causally necessitates y

Premise (1) asserts causal determinism. Premise (2) instantiates premise (1). Premise (3) is our equivalent of rule beta. I'm pretty sure (3) isn't vulnerable to the sort of argument you offer, but clearly may be vulnerable to other objections. Perhaps most significantly, it's also a rather less "obvious" premise and easier to reject without necessarily having to refute.

If you're interested, this is where I "defended" this sort of argument. This was an introductory talk aimed 6th form students (17/18 year olds = high school?), so it's a bit much to call it a "defense" really.

Merry Christmas, or Happy Holidays or whatever you prefer!

Angra Mainyu said...

Hi Steve

No problem, and thanks for the link.

The alternative against beta2 that you propose works too, but why is it slightly more colorful? ;)
I agree that your argument against the compatibility of causal determinism and having control is not vulnerable to the same kind of objections.
However, in my assessment, the standard compatibilist reply succeeds.
The question is: why should one accept premise 3?
Suppose if I decide to post on this thread, then I will succeed and post. If I decide not to post, then I won't. Also, my brain is working normally, and in particular, nothing compels me to post or refrain from posting.
Then, I reckon I have control over whether I post on this thread, regardless of whether my decision is causally determined by previous events over which I had no control.
So, I conclude that premise 3 is not true if I'm the "I" in the premise (or if you are, etc.), at least if the world is causally deterministic.
If the world is causally indeterministic, it might be that a deterministic scenario makes my brain work abnormally, and that might threaten my control over events - depending on how the indeterministic features of the world happen to normally play a role in my choices. But on the other hand, if the world is causally inderministic, then it might be that a deterministic scenario makes my brain work abnormally, and then that might threaten my control over events - depending on the manner in which the indeterministic scenario operates.

It might be suggested that determining causes compel. But I don't agree. Cases of compulsion seem to be some specific cases of abnormal decision-making processes, e.g., if a person is being threatened with a knife, or if she has a brain tumor that causes pathological mental states and choices, etc. But if the world is causally deterministic, then our normal decision-making processes are deterministic. In the usual sense of the word "compulsion", I see no compulsion.

So, in my view, the argument fails.
Granted, that's - in general - not going to convince incompatibilists. But the argument isn't - in general - going to convince compatibilists, either. It's difficult to find a way out of this particular disagreement. But I'd like to ask a question: do you think there are potential empirical that might count for or against causal determinism?
(btw, this isn't about theism vs. non-theism. For example, a non-theist might believe in indeterministic Buddhist-like souls, and a theist might believe in causally determined humans (e.g., Calvinism seems to be like that, or at least compatible with that, assuming it has no internal contradictions; here I briefly addressed the question of freedom on a Calvinistic scenario)).

Happy Holidays or Merry Christmas to you as well!

Steve Lovell said...

Hi Angra,

Well, this has worked its way a long way down into the archives! I wanted to make the most of time with family over Christmas, so didn't prioritise a rapid response.

Anyway, I agree that yours is the approach the defender of compatibilism should take. The denial of my correlate of rule beta certainly isn't incoherent in the way it which the denial of the other versions looks like it could be. That said, until a version like one of those others is positively demonstrated, it seems that the sort of response you've given to my argument could also be given to them. I guess there is a concern that perhaps entailment determinism and causal determinism are sufficiently different that that's really not appropriate. However, since (in my view) the entailment relation is supposed to map an underlying causal or natural law relation and I don't personally see so much of a dis-analogy between the two.

Anyway, if you're reading this it seems like we may have come to a natural conclusion to this discussion so no response is necessary. Thanks for the enjoyable interaction!

Angra Mainyu said...

Hi Steve

I think we've covered the gist of it. Personally, I'm inclined to think the disagreement between compatibilists and incompatibilists will go on for a very long time, but that's pretty common in philosophy.

Thanks for the enjoyable interaction as well!.