This is how philosophers should greet each other: “Take your time!”
I want to recapitulate some of the arguments I recently presented in a discussion I had with a number of people at once on Skype (all of whom are philosophically astute). I found myself defending a position so unpopular that every person in the room seemed not only to disagree, but to do so adamantly. I will thus do my best to summarize and run through the points I made, along with the responses I tried to give to certain objections raised.
I should begin by tacitly admitting that I may be wrong about this, and it would be good news for me if I were wrong about this. What I want to argue is that libertarian free will is compatible with the Principle of Sufficient Reason such that a commitment to the PSR does not logically entail the modal collapse which Spinoza’s system falls prey to explicitly (and which Leibniz’ philosophy does implicitly). If I were wrong, then I could adopt a weakened version of the PSR, the WPSR, which states that any contingent fact which possibly has an explanation has an explanation, and then I could add what Timothy O’Connor says to it; namely that a fact without an explanation is not a brute fact just in case some reason can be given for why an explanation cannot be given (such as would be in the case of necessary truths like mathematical truths, and as would be the case in cases of libertarian freedom, and perhaps even cases of Quantum Mechanical events). Furthermore, I could get past one of my three main obstacles to Molinism, which would work miracles, so to speak, for solving theological problems more parsimoniously. So, I would be quite happy to find that I am wrong, but unfortunately I do not think I am.
Here’s what I am trying to maintain. The Principle of Sufficient Reason (PSR) states that there is no contingent fact for which there is nothing we can say by way of explanation of it’s being true. I believe that the PSR is correct. I also believe that having something to say by way of explaining a contingent fact’s truth can, in some cases (at least one case) be the very same thing as having something to say by way of explaining it’s being true rather than not. Thus, I claim that the contrastive question adds absolutely nothing to the original question in some cases or at least one case. Why do we think it does? I suspect that we think we’re asking a deeper question when we ask the contrastive question because what we are looking for is a causal explanation. By a causal explanation, I’ll say that what we mean is an explanation of the following variety: that identifying all the relevant antecedents will of necessity give us the consequent(s). I think that this very question is antithetical to Libertarian Free Will (LFW) – it is literally question-begging.
Now, consider the case of Susie, a libertarian free agent, making a libertarian free choice. Let’s say she has the choice between two options A and B. Now, suppose she chooses A. What is the sufficient reason of her having chosen A? Well, the sufficient reason of her having chosen A is just the set of all the reasons she had for choosing A, along with the set of reasons she had for not choosing B (which here act indirectly as reasons for choosing A). That certainly is some kind of explanation for why she chose A. Somebody may object “yes, but why did Susie choose A rather than B?” To my mind the person asking this question is either looking for a causal explanation, or else they aren’t asking anything other than what they would have been asking had they said “yes, but why did Susie choose A?” If they are looking for a causal explanation then we should say they are looking for an explanation of a kind that cannot be had in this case, by the very nature of the case, since the whole idea of libertarian freedom precludes there being such an explanation. However, if they think that because there isn’t an explanation of this variety, there is no explanation of any variety, I would object by saying that of course there is an explanation of a more modest variety. The explanation of why Susie chose A rather than B just is the explanation of why Susie chose A, which is all the reasons she had for freely choosing A.
The idea I have in mind when I talk about Susie libertarian-freely choosing something goes like this: the antecedents play some explanatory role in leading to Susie’s choice, but they don’t play a causal role if by causal explanation we mean what I’ve stipulated above (that given X antecedents, Y consequents follow). They provide the boundary conditions, as it were. Without the antecedents, Susie could not have made the choice at all, since the option to chose would not have become ‘live’. We can thus talk about these antecedents as providing intentional impetus for at-least-two mutually exclusive options. Without this intentional impetus (i.e., these antecedent conditions) a libertarian-free choice cannot be taken/made.
If you ask ‘what is the reason Susie chose A’ the answer is going to be the set of all the reasons Susie had for choosing A (i.e., the antecedent conditions providing impetus for A). If you ask “yes, but what caused Susie to chose A?” – well, now you’re looking for that causal explanation which is antithetical to libertarian free will, and you’re no longer looking for an explanation simpliciter. Suppose you say “ok, but what is the reason Susie chose A rather than B” – the answer is going to be the very same as it would have been if you had asked “what is the reason Susie chose A?” The reason Susie chose A rather than B is just the set of all the reasons she had for choosing A. If one thinks that’s not an explanation then one is looking for a causal explanation and doesn’t want to accept anything less, but the PSR does not commit us to thinking that all explanations are causal explanations of this variety. The PSR just commits us to saying that there is some ‘Sufficient’ explanation. The Libertarian believes that citing all the reasons Susie had for choosing A is sufficient to explain why Susie chose A, even if her choice could not have been predicted by merely citing all those reasons. We might ask, “what is the criteria for sufficiency?” Fr. Copleston once said that the Principle of Sufficient Reason requires total explanation to which nothing more could be added. I think that is a little too strong, but at least the PSR requires explanation to which nothing more need be added – nothing has been left out.
Now, in the case of Susie Libertarian-freely choosing A, what we’re dealing with is a causal story in which there is a first mover or first cause. Obviously there isn’t going to be a causal background story for the first cause (for then it wouldn’t be a first cause at all). However, there can still be an explanation of another sort for the first cause (eg. all the antecedent conditions in place which made it ‘live’ in the first place). This will work even with God, where the antecedent condition in place is simply the formal cause of creation (i.e., God’s nature). So, we can give an explanation of the first mover in the Libertarian sense (we say God is an unmoved mover meaning that he isn’t moved by anything external to himself, but we don’t say that of Susie, because the antecedent conditions include more than just her nature). So then, Susie Libertarian-freely chooses A given options A and B. Ex Hypothesi that choice represents a first-mover’s move, such that the kind of causal story we’re dealing with is one in which we stipulate a ‘first cause’. Certainly those kinds of causal stories are logically possible (I see no reason to think they wouldn’t be). By the nature of the case one cannot ask about the causal story leading up to the first-cause, since that is just confused. However, admitting that there is not a causal explanation of the first-cause is not to say that there is NO explanation of the first cause. In fact, we can give an explanation for such a first cause to which nothing more need be added (nothing is left out), in terms of citing the antecedent conditions which ground the possibility of the first cause in the first place. In a logically possible world where Susie chooses A given the choice between options A and B, the sufficient reason of her having chosen A is all the reasons she had for choosing A. If one is asking for a causal story then we can say that “Susie chooses A” is where the causal story begins. Is there a sufficient reason for that first cause? Sure, but it must be in terms of the antecedent conditions which made it possible for Susie to choose A. Suppose in the world where Susie chose A one asks “why did she choose A rather than B”, what they are asking is why the actual world is the world in which Susie chose A rather than the one in which she chose B, and the answer to this kind of question is that Susie chose A (that’s where the causal story begins). It is be-cause Susie chose A that that world is actual. If you ask for a causal story leading up to that point then you’re just confused about what it means for Susie to libertarian-freely choose A (i.e., to act as a first-cause in a causal story). If you ask for a non-causal explanatory story then I can give you all the antecedent conditions which Susie acted upon as the sufficient conditions given which she could possibly choose A (they were obviously sufficient for her to act libertarian-freely as a first mover). In the world where Susie chose B the same answer could be given. In either world-scenario there is an explanatory story to which nothing need be added for the first-cause event in the causal story. If one asks why Susie chose A rather than B, and one is not looking for a causal story (since that would be confused given that Susie acts libertarian-freely in the case we’re considering), then the explanatory story we can provide is one in terms of antecedent conditions which made possible the first-mover act Susie actualized. These conditions were sufficient for her to act, which just is to say that they are the sufficient reason of her acting. No causal story can be added by the nature of the case (it isn’t logically possible to have a causal story leading up to a first-cause, or a causal story whose narrative begins with a first-cause). At this point I’m just repeating myself hoping that repetition will help elucidation. So obvious does this point seem to me, and so hard to get across, that I am not sure what to do but to explain the same idea from as many directions as I can.
- “What is the cause of Susie’s Libertarian-freely choosing A?” – There is no answer to this question, the question is just confused.
- “What is the sufficient reason of Susie’s acting libertarian-freely by choosing A?” – All the antecedent conditions which were sufficient for her to make a libertarian free choice.
- “Why is the actual world the one in which Susie chose A, and not the one in which Susie chose B?” – Because Susie actualized the first by freely choosing it.
- “What caused Susie to freely choose A?” – There is no answer to this question, the question is just confused.
- “What is the sufficient reason for Susie freely choosing A?” – all those antecedent conditions which were collectively sufficient for her to choose A.
- “What is the sufficient reason for Susie freely choosing A rather than B?” – All those antecedent conditions which were collectively sufficient for her to choose A. Since the word “choose” in question five presumes that there was at least one option other than A this later question is just asking the very same thing, and so the answer must be the very same answer.
Let’s address some other concerns. First, is it the case that libertarian-free choices are random? Suppose one wants to say that this choice is random, and something’s being random is antithetical to it’s being libertarian-free. How I respond is as follows: if by random all you mean is that across a set of possible worlds all actualizing all and only the same antecedent conditions different consequents obtain, then the reverse of the objection is true; a choice’s being libertarian-free entails its being random. However, it isn’t random in the sense that it is out of Susie’s control. Susie is the one who acts as a first-mover freely. She isn’t responsible for random events out of her control (or deterministic events out of her control for that matter) and the whole problem of her being determined is that her actions are ‘out of her control’. However, if her acts are random in the sense defined, and entirely within her control, then obviously she is responsible for them (she can be praised and/or blamed for her actions). How is this case of randomness any different from quantum mechanical randomness (on the presumption that all or any quantum mechanical events really are indeterministic)? Well, quantum mechanical events are not in anyone’s control. Thus, the randomness of libertarian-freedom is not the same as the randomness of quantum mechanical events on standard (eg. Copenhagen) interpretations, since there is nothing which acts as a causal first mover in the case of the quantum vacuum (unless perhaps God acts libertarian-freely to bring this or that particle into extension at every moment, having in view counterfactuals in near possible worlds or something like that).
One might go on to raise the following concern in the form of an objection: well, if contrastive explanations are being reduced to explanations of one of the parts to be contrasted with at least one other, then why can’t we do the very same thing with the cosmological argument? Suppose that Set A is a set of all contingent facts. So Set A is the Big Conjunctive Contingent Fact. Suppose I then ask “What is the sufficient reason for Set A?” What my question involves is what is the sufficient reason for Set A rather than Set B (where Set B is not identical to Set A). However, if contrastive explanations are just identical to the simpler questions like “what is the sufficient reason for Set A?” then the answer to one is the answer to the other. Why then can’t we answer the question “what is the sufficient reason for Set A” by pointing to it’s members?
This objection is confused in a few ways I think. First, let’s assume this kind of contrastive question could be meaningful on the supposition that it isn’t a contrastive question about a libertarian-free choice. Remember that the sentence “Susie libertarian-freely chooses A” already presupposes that Susie libertarian-freely chooses A rather than B (or some set ‘B’ of alternatives to A). To ask the one is to ask the other, there is no causal story for either which doesn’t just begin with Susie’s choice, and there is sufficient reason why Susie chooses A [rather than B]. However, the question of why Set A is actual rather than Set B (presuming that it wasn’t the result of a libertarian-free choice on God’s part, or our part, or some combination of libertarian free choices on the part of multiple parties), even if meaningful, 1) should not be answered self-referentially, and 2) has nothing to do with the cosmological argument.
How the cosmological argument goes is: “what is the sufficient reason for there being any contingent fact?” Since the PSR requires that all contingent facts have an explanation, and even an infinite conjunction of contingent facts represents one contingent fact, the PSR can only be satisfied when it posits an explanation involving incontingency. God’s nature or being is incontingent. The explanation of God’s contingently choosing to create the world (as a libertarian first-mover) is itself explained by his being/nature (which is the analogue to Susie’s antecedental impetus [yes, I did just make that word up]) which is incontingent. In a world where Set A is actual the sufficient reason for Set A’s being actual is going to be that God libertarian-freely chose to actualize Set A (or more pedantically that God libertarian-freely chose to actualize a world in which it would be possible for libertarian-free agents to actualize Set A, and libertarian-free agents chose to actualize Set A), and that fact is sufficiently explained by the antecedent conditions which were sufficient impetus for God to act thusly in his capacity as a libertarian-free agent. The PSR is thus satisfied when it reaches an explanation in terms of something incontingent, which it has.
One final concern may be that if one cannot give a causal explanation for a causal story beginning with a first mover then, whatever other explanation one can give in terms of antecedents which were sufficient for a libertarian-free action to be taken, that explanation doesn’t eliminate the bruteness of the whole causal story in which there is a first-cause. Even if an explanation can be given citing the conditions sufficient for libertarian-free action to be taken, and even if it is entirely confused to say that a causal explanation is lacking (since it can’t be ‘lacking’ in the case where it couldn’t logically-possibly apply), still one may want to say that the whole story is a ‘just-so‘ story, a story for which no sufficient reason can be given of it’s being true. However, again, the explanation of this world being the world in which Susie chooses A is that Susie chose A. The explanation of Susie choosing A cannot be a causal explanation, but must be a ‘sufficient explanation’, and we’ve already said that to cite antecedent conditions which are collectively sufficient for Susie to act libertarian-freely is a sufficient reason of Susie’s acting libertarian-freely. Why did she act freely such that she chose A? The sufficient reason for that is all the reasons she had for choosing A – and we’re right back where we started. For somebody to claim that ‘Susie choosing A’ is a brute fact is for them to say that there is no sufficient explanation of it, which is just to say that there were no conditions which were ‘sufficient’ for Susie to choose A libertarian-freely. It is hard for me to overstate how obviously confused such an objection seems to me.
Well, that’s quite a lot, so perhaps I’ll leave it at this for now.