Greg Welty Sat, 29 Apr 2017 16:13:37 +0000 en-US hourly 1 71920641 Freethinking About Molinist Gunslingers – A Response to Stratton Sat, 29 Apr 2017 16:13:37 +0000 To contribute a chapter to a book entitled Calvinism and the Problem of Evil just is to stoke controversy and invite replies. This is normal, and should be welcomed. How else are we to gain well-earned beliefs on important matters unless we interact with those who disagree with us? Tim Stratton, a fellow Christian brother who has read my “Molinist Gunslingers: God and the Authorship of Sin,” has written up some thoughts in reply. Here are my responses to what I think are his central points.

“That is, if Calvinistic exhaustive determinism is true, then God is a moral monster and responsible as the author of every evil action. After all, on this omni-causal view, God made the Holocaust happen — not Hitler!”

Stratton seems to think that if God exhaustively determines all events (‘omni-causality’), then “God made the Holocaust happen – not Hitler!” But this last phrase – “not Hitler!” – is unmotivated by divine determinism. For there is no reason to think, either logically or biblically, that divine causation precludes human causation, such that if God makes something happen then human agents don’t make it happen. Logically, if God causes X to cause Y, then obviously X is a real cause. Perhaps Stratton has some subtle argument according to which unless an agent satisfies incompatibilist sourcehood conditions, he cannot be a genuine cause. But in the absence of such an argument, there’s not much for the Calvinist here to worry about, logically speaking. Able defenders of libertarian free will, such as Robert Kane, make this point for me: “Determinism… does not imply that we have no influence on how things turn out, including the molding of our characters. We obviously do have such an influence, and determinism alone does not rule it out” (Kane [summarizing a point made by John Stuart Mill], A Contemporary Introduction to Free Will, Oxford Univ. Press, 2005, p. 20).

Biblically speaking, Stratton must have overlooked the final section of my chapter, entitled “Scriptural testimony to double-agency or double-causation”. (He never refers to it in his presentation.) There I give examples where the Scriptures teach that both God and man bring about the same event. For instance, the Sabeans and Chaldeans “struck down the servants with the edge of the sword” (Job 1:15, 17). And yet Job described these events as a case where “the LORD has taken away” (1:21). His friends confirmed that this was “evil that the LORD had brought upon him” (42:11). The verbs used here are regularly used throughout the Hebrew Bible to describe ordinary, mundane cases of causation, whether by God or man. (And notice that in saying that God caused it all, “Job did not sin or charge God with wrong,” 1:22. That is, Job wasn’t sinning in attributing this high view of providence to God.)

Or again: “Therefore Saul took his own sword and fell upon it… Thus Saul died…” (1 Chronicles 10:4, 6). Who killed Saul? Saul killed Saul, by committing suicide. “Therefore the LORD put him to death…” (1 Chronicles 10:14). Who killed Saul? God killed Saul. God put him to death. We might want to say that God killed Saul by way of Saul killing Saul, but saying that doesn’t magically convert two causal claims into one. If God killed Saul – that is, “put him to death” – then God caused his death. (After all, if you don’t cause someone’s death, you can scarcely be said to have killed him.) It follows that God’s causing an event doesn’t preclude humans causing that same event (confirming the logical point I made above). Many more texts like this can be given (and were given). So the idea that if Calvinism is true then it follows that bad moral agents can’t be causes is both a non sequitur in logic and refutable from the Bible.

“I have argued that if Calvinistic determinism is true, then God causally determines the thoughts and actions of all creatures, and thus is implicated in the evil actions of creatures because the creature’s action was “up to” God and not the creature. God is the responsible puppet master on this view.”

One doesn’t have to appeal to my ‘gunslinger’ argument to effectively push back against the above argument. When Stratton talks about “the evil actions of creatures,” he is speaking of their actual actions in history, what they in fact do. And it is clear that on Molinism which human actions actually occur in history is up to God. For, knowing would-counterfactuals about how humans would act in various circumstances, God acts on that knowledge, ensuring what they will do (as a matter of history) by actualizing the appropriate circumstances. In this way, while God has no control over would-truths about human actions (he knows these passively in his middle-knowledge), he does have control over will-truths about human actions. That is, it is up to God whether or not “the evil actions of creatures” actually occur in history with all of their specificity. For it was up to God whether or not to create creatures, and to actualize circumstances which – together with the truth of the counterfactuals – are sufficient for the evil actions occurring. (It is not up to creatures whether or not they exist and are placed in the appropriate circumstances!) As Alfred Freddoso puts it, with respect to “the created universe with its entire history,” God plays “an active causal role sufficient to ensure its exact realization” (p. 3 of his Introduction to Molina’s On Divine Foreknowledge). Undoubtedly, this includes the realization of human sin in history.

“Before proceeding, it is important to note that these “dimensions of the case” seem to assume libertarian free will — that is, this crime was “up to” the gunman (and not something or someone external to the gunman). This is because the gunman’s thoughts and actions are a necessary condition for the trajectory of the bullet, but his thoughts and actions are not sufficient for a bullet to the head because the laws of physics are also necessary. That is to say, if the gunman did not (freely?) choose to pull the trigger, then although nothing changes in regards to the laws of nature, the other man does not get shot by the gunman.”

This is exactly right. Now reread the above paragraph, except that God is the gunman. It is precisely the fact that God has libertarian free will that exacerbates the problem for Molinism. God’s ‘pulling the trigger’ – his creating human agents and placing them in specific circumstances – was up to God and not up to something or someone external to him. If God did not actualize agents and circumstances, then no sins come to pass, but if God does so actualize them, then he ensures that all their sins come to pass exactly as they do come to pass. And yet God was free to either actualize or not actualize. So if the ordinary gunman is culpable for what he brings about, then so is God.

“He argues that the gunman is still responsible for Mario’s death even though Bullet Bill possessed the libertarian freedom to choose to kill Mario or not. This is akin to suggesting that parents who know that if they have a child, their child will freely choose to sin, then the parents should be held responsible for the child’s free choice to sin later in life. My parents knew, for example, that I would not be a perfect human being long before they chose to start a family. If I get a speeding ticket should they get one too? Of course not — that is absurd!”

I agree that the consequence would be absurd, but thankfully the situations are not relevantly analogous. Avoiding these kinds of disanalogies is precisely why I laid out the Molinist gunslinger situation in such detail. Here, Stratton can only avoid my conclusion by suppressing the Molinist details that generate it, and putting in its place something akin to open theism. But this will not do. Undoubtedly the mere fact that I conceive a child is not sufficient to make me culpable for what he later chooses to do. But that’s because my choice to conceive, combined with relevant knowledge I had at the time, is not sufficient to ensure any future sins on the part of my offspring. By way of contrast, given his exhaustive and infallible knowledge of would-counterfactuals, God’s actualization of the circumstances is sufficient for the agent to sin. Yes, there are worlds in which God creates them and they don’t sin, but in those worlds the relevant counterfactuals are false. As we all know, knowledge matters for culpability, and in the actual world God infallibly knows that his actualization combined with the true counterfactuals ensures the future sinful actions of agents, in all their specificity. There is nothing remotely analogous to this in Stratton’s example of mundane conception.

Or look at it the other way around. Why do we regard the ordinary gunman to be guilty? All he did was choose to pull a trigger, and surely a trigger-pull – by itself – is not sufficient for the killing of anyone. So why do we hold him culpable for what happens ‘down the line,’ so to speak, after the trigger releases the hammer which hits the primer which explodes and ignites the propellant which creates the gas pressure which sends the bullet down the barrel, into the heart of the victim? So many things had to happen after the trigger pull, in order for the bullet to kill! So maybe he’s… not guilty? After all, so many things had to happen after conception, in order for the offspring to kill! So if parents aren’t guilty for mere conception, then ordinary gunmen aren’t guilty for mere trigger-pulls either. But clearly something has gone wrong with this analogy, and it is this: in the ordinary gunman case, the gunman has highly justified beliefs about the specific outcome – the victim will die, if the gun doesn’t misfire. He doesn’t just surmise that something or other will happen; he knows what will happen. Not so for the parents. Their choice to conceive does not ensure or guarantee any future sins. Indeed, it’s compatible with what the parent knows at the time of conception, that the child not go on to do any of the bad things he in fact goes on to do. Not so for God at the moment of creation, on Molinism. I suggest it is obvious that Molinist creation is aptly modeled by the ordinary gunman, and only misleadingly modeled by the conceiving parent, and one must temporarily forget one’s Molinism to not see this.

“This passage illustrates that although the father brought the son into existence the son’s free choice to sin broke the “chain of causation,” and thus, the son is independently accountable for his own sin.”

I find it curious that Stratton reads into Ezekiel 18:20 concepts that are nowhere to be found: that “the son’s free choice to sin broke ‘the chain of causation’.” The text speaks neither of the son’s ‘freedom’ nor of a ‘chain of causation’ (which then gets ‘broken’ by freedom). I fear Stratton is reading into texts the theses he needs rather than finding them there.

Here’s one reason why ordinary fathers don’t inherit the guilt of their sons, but God according to Molinism might well indeed inherit the guilt of created agents: ordinary fathers don’t govern the lives of their sons by way of a Molinist providential scheme. They don’t engage in acts of actualization that ensure or guarantee their son’s sins.

“The son’s father is not responsible for the sin of the son, God is not responsible for the sin of the son — the son is responsible for his sin (it was up to him)!”

The ambiguity of “it was up to him,” on the Molinist scheme, is what needs to be disambiguated if we are to have clarity on this issue. In one sense (a culpability-irrelevant sense) the murder of the victim wasn’t up to the ordinary gunman and in another sense (a culpability-relevant sense) it was. It wasn’t up to the gunman, because mere trigger-pulls are not sufficient for murders. A lot of other stuff has to happen; nature has to cooperate with the trigger-pull. But surely that doesn’t get the gunman off the hook! And in another sense the murder was up to the ordinary gunman, because – holding fixed the stuff over which the gunman had no control (the laws of nature) – it is obvious that the trigger-pull will ensure the murder. So the gunman is culpable. He is exploiting his knowledge of the laws of nature for his providential purposes, as it were. Likewise for Molinism: God is exploiting his knowledge of the counterfactuals for his providential purposes, as it were, and so the analogy here with Molinism is precise. Holding fixed the stuff over which God has no control (the truth of the counterfactuals), it is obvious that God’s actualization will ensure the creaturely sins in all their specificity. The only true sense in which it wasn’t up to the gunman, is a sense that doesn’t absolve him of responsibility: mere trigger pulls don’t ensure murders (strictly speaking). Likewise, the only true sense in which it wasn’t up to God, is a sense that doesn’t absolve him of responsibility: mere circumstance-actualization doesn’t ensure creaturely sins (strictly speaking). But in both cases, the part that is up to the gunman/God is sufficient to generate culpability. Stratton still hasn’t given a reason to see these two cases as relevantly disanalogous. Ordinary fathers conceiving sons simply isn’t analogous to the details of Molinist providence, and so the appeal to Ezekiel 18:20 fails.

“Bullet Bill is not causally determined to kill Mario (it is up to him) if Molinism is true. But if Calvinism is true, Bullet Bill is causally determined by the gunman (God) to kill Mario. That is to say, on Molinism Bullet Bill can freely choose to avoid murder; on Calvinism Bullet Bill has no choice but to kill Mario as God causes Bullet Bill’s thoughts, intentions, and actions.”

This is all true, but irrelevant. My argument never affirms that Bullet Bill is causally determined to kill Mario. And it never affirms that Bullet Bill’s choice is not up to him in some relevant libertarian sense. In fact I take great pains to avoid characterizing the situation in that way, because if I affirmed those things I would not be talking about Molinism, and my critique would be dead on arrival (because an obvious strawman). Nevertheless, the conclusion follows. For my argument is not: “On Molinism, God causally determines agents to sin and their choices aren’t up to them. Therefore, God on Molinism is culpable for human sin, and Molinism is as bad as Calvinism.” My argument instead is that Molinist providence is sufficiently analogous to sufficient causation, so that Molinism inherits the Calvinist liabilities with respect to authorship of sin, responsibility, and blame. I don’t believe I could have been any clearer on this point.

Stratton’s response so far appears to be a threefold act of misdirection. First, speak of a situation that is not relevantly analogous to the particularities of Molinist providence, namely, parents conceiving children. Second, find theses about freedom and chains of causation in a text of Scripture that speaks of neither. Third, announce the gunslinger analogy is dead, and then prove it by misstating the analogy.

Perhaps a diagnostic question will help to clear the air. Assuming that Stratton believes that the ordinary gunman case is a case of culpability, which aspect of the ordinary gunman case – relevant for generating his culpability – is missing from the case of Molinist providence? In my view, the analogy is so tight that I don’t believe Stratton will be able to identify a missing feature. This is all the more the case if we put a Bullet Bill gun in the gunman’s hands.

“2- On Molinism, the gunman (God) desires Bullet Bill to freely choose to love Him (the gunman) and all people — from his neighbors to Bill’s enemies. In fact, the gunman (God) actually loves both Bill and Mario perfectly (omnibenevolence)! The gunman (God) actually desires the best for both Bill and Mario for all eternity (John 3:16; 1 Tim 2:4; 2 Pet 3:9) and does not want Bill to choose to kill Mario, Luigi, or anyone else.”

This is all very interesting, but wholly irrelevant to undermining my conclusion. I concede these as features of Molinism. (Indeed, I am happy to describe my Calvinism in these terms too.) But how does affirming an ethical truth about God undermine my inference from the metaphysical implications of Molinism? If the ordinary gunman ‘desires’ good things for everybody, does that suddenly mean he is not culpable for murdering his victim? No, because his culpability is grounded in the fact that he acts in such a way that ensures the murder of his victim. His feelings, desires, and loves, are irrelevant. (If they are relevant, then Calvinists have a great comeback to the charge that they make God the author of sin: “God loves everybody, so Calvinist providence is perfectly fine!)

“Of course, given God’s omniscience, the gunman (God) must know that Bullet Bill will freely choose to sin and kill Mario (although he genuinely could have done otherwise). The significant difference here is that in the “ordinary gun” scenario (Calvinism), the gunman (God) actually desires the death of Mario as well as the Nazi Holocaust, Islamic terror, and all the damned who suffer in the eternal fires of hell for choices they were powerless to make.”

Stratton seems to have overlooked my reply to the seventh and final objection that I consider in the chapter. There, I note the basic moral fact that intentions are not closed under known entailment. If S intends that p, and S knows that p implies q, it does not follow that S thereby intends q. So if God intends the universe he creates because it will manifest some intrinsic value, or promote his glory, or reveal his attributes, and God also knows that the actualization of such a universe implies human sin, it does not follow that God thereby intends or ‘desires’ human sin. So there is no “significant difference here,” for if Molinists get the doctrine of double effect, so do Calvinists. If Stratton’s strategy is sufficient to deflect culpability for God on Molinism, I’m all for it! For this strategy is perfectly available to Calvinists as well. It appeals to no resources distinctive to Molinism.

In addition, by focusing our attention on “God’s omniscience,” on what God “must know,” Stratton leaves out what God does. What God does is perfectly analogous to the trigger-pull: he actualizes circumstances. The fact that the ordinary gunman does this with knowledge only increases his culpability. He pulls the trigger knowing what will result. It’s not the gunman’s knowledge, of course, that ensures the result. It’s the trigger pull (combined with the realities over which the gunman has no control, the laws of nature). Likewise, the fact that God actualizes circumstances with knowledge only increases his culpability. God actualizes circumstances knowing what will result. It’s not God’s knowledge, of course, that ensures the result. It’s the circumstance-actualization (combined with realities over which God has no control, the counterfactuals). Stratton’s focus on divine knowledge doesn’t pry these cases apart. Instead, it contributes to their similarity, and thus to my conclusion.

“On Molinism, however, the gunman also knows that all of these freely chosen evil actions will eventually be used for the ultimate eternal GOOD (Gen 50:20; Rom 8:28; 2 Cor 4:17). The Calvinistic determinist cannot make the same case (click here for more).”

Well, when I “click here for more,” I don’t find an argument that God cannot ensure that all “freely chosen evil actions will eventually be used for the ultimate eternal good.” I find an argument that some Calvinists have speculated that divine desires for glory may preclude universal salvation. But even if this were true, Stratton’s inference wouldn’t follow. It wouldn’t follow that all evils wouldn’t be worked out by God for an ultimate eternal good. It would only follow that there is a distinction among goods, and that not all evil actions are used for the same kind of good (i.e., salvation).

“The gunman, on Molinism, never desired for Bill to kill. In fact, the gunman never shot the bullet in the direction of Mario; rather, the gunman merely placed the bullet on the table. The gunman did not project the bullet in either direction, both directions were perfect alternatives and neither of Bill’s choices contained deterministic factors. God allows Bill to make His choice freely.”

I fear that Stratton’s rhetoric is getting the best of him. To say that the Molinist gunman doesn’t shoot bullets in particular directions – he just places bullets on tables – is to fundamentally misunderstand the analogy. On Molinism, God isn’t passive. He actualizes circumstances. And he does so in a way that ensures particular outcomes – indeed, ensures the particular history of the universe in exhaustive detail! This is the fundamental point. ‘Placing bullets on tables’ isn’t a theory of providence. So here are the questions:

  1. Does the ordinary gunman pull the trigger, or not? Analogously, does God actualize circumstances, or not?
  2. In pulling the trigger, does the ordinary gunman ensure a particular outcome (as long as the laws of nature are in place)? Analogously, in actualizing circumstances, does God ensure a particular outcome (as long as the counterfactuals are in place)?
  3. In pulling triggers and ensuring outcomes, is the ordinary gunman culpable for the outcome? Analogously, in actualizing circumstances and ensuring outcomes, is God culpable for the outcome?

To be successful, Stratton has to cite a factor that is present in the ordinary gunman scenario, but absent in the Molinist scenario, that is relevant for culpability. But simply changing the analogy will not do. In his talk of ‘merely placing bullets on tables,’ is Stratton denying circumstance-actualization on the part of God? If so, then he’s not a Molinist. But if he’s not denying this, then given the rest of Molinist providence the conclusion still goes through.

“He reasons that the Molinist’s view of God is just as bad as the Calvinistic view because God created a world in which He knew how free creatures would freely choose to act.”

I’ve cited a sentence that is typical of Stratton’s lengthy section, “Predestination != Determinism”. I cite it because of how badly it represents the argument I in fact gave. If my inference were that God (on Molinism) is culpable simply because “He knew how free creatures would freely choose to act,” I would be of all philosophers most to be pitied. It is impossible to infer culpability from mere knowledge. I know of many bad actions by many people; it doesn’t follow I’m culpable for them. I might even know of someone’s bad actions ahead of time (perhaps I read of their plan); it still doesn’t follow that I’m culpable for them, for my knowledge doesn’t involve me in any way with their sin. But contrast all this with the following: I act in such a way that ensures and guarantees that everyone in history has the sinful intentions and commits the exact sins that they in fact have and commit. This changes the picture considerably.

Again and again in this section, Stratton seems to assume that I am making an inference from mere knowledge on God’s part. Nothing could be further from the truth. I am making an inference from God’s providential actions on Molinism, and that is an active thing, not a role to be played by a mere passive observer.

“Sure, on Molinism, God is “responsible” for creating a world in which creatures can freely choose to do good or evil (which is a logically necessary “side effect” of creating a world where eternal true love can be obtained), but if creatures are not forced or causally determined to do evil, then they are responsible agents and guilty of their evil deeds — not God.”

Here Stratton underdescribes Molinism and combines it with a straw man. What generates my conclusion is not some innocuous truth that God creates a world “in which creatures can freely choose to do good or evil.” What is crucial is the central Molinist claim that Stratton leaves out: God ensures and guarantees that human freedom will be used in the exact ways it is used in human history, by actualizing circumstances in light of his middle-knowledge. There is no need for scare quotes around ‘responsible’. God acts in ways that ensure outcomes, in as meticulous a way as we can imagine. That’s far different from giving a gift of free will and then hoping for the best with one’s omnibenevolent desires. And if you act in such a way as to ensure outcomes, ordinarily you are culpable for those outcomes. It is so in the ordinary gunman case. Why not in the Molinist case?

I concede that, on Molinism, “creatures are not forced or causally determined to do evil,” and so “they are responsible agents and guilty of their evil deeds.” That is the point of Bullet Bill: he is a sentient encased in steel, possessing libertarian free will. But as I pointed out in the chapter, the fact that Bullet Bill is a responsible agent is neither here nor there when it comes to the responsibility of the one who shoots the Bullet Bill gun. Let Stratton answer the question directly: if I murdered someone with a Bullet Bill gun, would he hold me culpable? Why not? Could I plead innocence because Bullet Bill wasn’t “forced or causally determined”?

“God is a necessary condition for the death of Mario (after all, God created a world in which Mario lives and can be killed), but He is not the sufficient cause because Bullet Bill has a real and genuine free choice to make.”

Again, Stratton underdescribes the content of Molinism. Notice that an open theist could endorse what is said above: God is a necessary condition for the murder of any person, for God created a world in which persons live and can be killed. Well, yes, of course. But God’s being a merely necessary condition isn’t Molinism. For being a merely necessary condition doesn’t ensure outcomes, and that’s precisely what God does in actualizing circumstances.

Notice that if Stratton’s defense here stands, then no ordinary gunman is ever culpable for murdering people with his gun. In pulling the trigger, he is ever and always only a necessary condition for the death of his victims, never the sufficient cause! Since nature has to cooperate with the trigger-pull in order to get the bullet into the victim, and whether nature cooperates isn’t up to the gunman, it follows that the gunman isn’t the sufficient cause of the murder. So either Stratton has provided us with a good reason for never holding persons culpable for the outcomes of their (insufficient!) actions, or he needs to rethink his strategy. Unless he can show a relevant disanalogy between the ordinary gun and Molinist cases, he will be stuck with something he doesn’t want: letting all murderers off the hook, or letting Calvinism off the hook.

Finally, Stratton makes much of his distinction between predestination and causal determinism, in the closing paragraphs of his presentation. I am happy to concede that on Molinism, although God does not causally determine human choices in the world, he nevertheless “predestines the evil in that world” (as Stratton puts it). Stratton finds this distinction significant with respect to claims about divine culpability, whereas I do not. But we can easily put this stance to the test: should I be held culpable for murdering people with a Bullet Bill gun? Since the gun is constructed according to the principles of Molinist providence, not Calvinist providence, then if Molinism involves ‘predestination’ rather than ‘causal determinism,’ let’s stipulate that’s an apt description of the relation between the gunman and Bullet Bill. Does that fact absolve me of culpability when I murder five people in the room with the Bullet Bill gun? I don’t see how. The options seem to be these:

  1. Deny that there can be Bullet Bill guns. But that would be saying that Molinism is incoherent.
  2. Deny that gunslingers are ever culpable, because they merely pull triggers and merely perform necessary conditions for outcomes. But that would spell the end of human culpability for outcomes.
  3. Deny that ‘Molinist predestination’ is different from ‘Calvinist causal determinism,’ for the purposes of assessing culpability for outcomes. But that would mean Molinists should drop this claim of difference from their repertoire of alleged advantages of Molinism over Calvinism. And that is my view.
Calvinism and the Problem of Evil Wed, 10 Aug 2016 15:05:04 +0000 My chapter, “Molinist Gunslingers: God and the Authorship of Sin,” was recently published in David E. Alexander and Daniel M. Johnson (eds.), Calvinism and the Problem of Evil (Wipf and Stock, 2016). That chapter grew out of a couple of conference presentations I made at the Molinism Study Group at ETS in 2010 and 2013, and was revised for publication. The contributors to the volume include:

David E. Alexander and Daniel M. Johnson – Introduction
Daniel M. Johnson – Calvinism and the Problem of Evil: A Map of the Territory
Greg Welty – Molinist Gunslingers: God and the Authorship of Sin
Heath White – Theological Determinism and the “Authoring Sin” Objection
James E. Bruce – Not the Author of Evil: A Question of Providence, Not a Problem for Calvinism
David E. Alexander – Orthodoxy, Theological Determinism, and the Problem of Evil
Paul Helm – Discrimination: Aspects of God’s Causal Activity
Hugh J. McCann – On Grace and Free Will
Alexander R. Pruss – The First Sin: A Dilemma for Christian Determinists
James N. Anderson – Calvinism and the First Sin
Christopher R. Green – A Compatibicalvinist Demonstrative-Goods Defense
Matthew J. Hart – Calvinism and the Problem of Hell
Anthony Bryson – Calvinism, Self-Attestation, and Apathy toward Arguments From Evil

Reason and Faith: Themes from Richard Swinburne Wed, 10 Aug 2016 14:37:55 +0000 My review of Reason and Faith: Themes from Richard Swinburne (Oxford Univ. Press, 2016) can be found in Themelios Volume 41 Issue 2 (August 2016). The book “consists of ten papers presented by distinguished philosophers of religion at a conference held in September 2014 at Purdue University, in order to reflect upon Swinburne’s work and honor him on the occasion of his eightieth birthday.” Contributors included: Jonathan Kvanvig, John Schellenberg, Paul Draper, Hud Hudson, Dean Zimmerman, Alvin Plantinga, Eleonore Stump, Nicholas Wolterstorff, Peter van Inwagen, and Marilyn Adams. Swinburne was my doctoral supervisor, and I was glad to be there for the conference.

The Case Against Reality Fri, 13 May 2016 16:14:36 +0000 Donald Hoffman is a Professor of Cognitive Science at the University of California, Irvine. He is known for his studies in visual perception and optical illusion, as documented in his 1998 book Visual Intelligence: How We Create What We See, and at his webpage. These optical illusions got him interested in thinking about whether our visual perceptions are veridical at all.

Recently, The Atlantic magazine published an interview with Hoffman entitled “The Case Against Reality”. According to the article,

Hoffman has spent the past three decades studying perception, artificial intelligence, evolutionary game theory and the brain, and his conclusion is a dramatic one: The world presented to us by our perceptions is nothing like reality. What’s more, he says, we have evolution itself to thank for this magnificent illusion, as it maximizes evolutionary fitness by driving truth to extinction.

Hoffman explains his change toward this point of view in this 2008 piece, “What Have You Changed Your Mind About? Why?”:

I now think that perception is useful because it is not veridical. The argument that evolution favors veridical perceptions is wrong, both theoretically and empirically. It is wrong in theory, because natural selection hinges on reproductive fitness, not on truth, and the two are not the same: Reproductive fitness in a particular niche might, for instance, be enhanced by reducing expenditures of time and energy in perception; true perceptions, in consequence, might be less fit than niche-specific shortcuts. It is wrong empirically: mimicry, camouflage, mating errors and supernormal stimuli are ubiquitous in nature, and all are predicated on non-veridical perceptions. The cockroach, we suspect, sees little of the truth, but is quite fit, though easily fooled, with its niche-specific perceptual hacks. Moreover, computational simulations based on evolutionary game theory, in which virtual animals that perceive the truth compete with others that sacrifice truth for speed and energy-efficiency, find that true perception generally goes extinct.

Hoffman argues for this conclusion in more rigorous fashion in his peer-reviewed article:

Justin T. Mark, Brian B. Marion, Donald D. Hoffman, “Natural selection and veridical perceptions,” Journal of Theoretical Biology 266 (2010) 504–515.

The abstract:

Does natural selection favor veridical perceptions, those that more accurately depict the objective environment? Students of perception often claim that it does. But this claim, though influential, has not been adequately tested. Here we formalize the claim and a few alternatives. To test them, we introduce ‘interface games,’ a class of evolutionary games in which perceptual strategies compete. We explore, in closed-form solutions and Monte Carlo simulations, some simpler games that assume frequency-dependent selection and complete mixing in infinite populations. We find that veridical perceptions can be driven to extinction by non-veridical strategies that are tuned to utility rather than objective reality. This suggests that natural selection need not favor veridical perceptions, and that the effects of selection on sensory perception deserve further study.

Interestingly enough, therefore, Hoffman seems to come to a conclusion similar to the one Alvin Plantinga argues in ch. 10 of Where the Conflict Really Lies: we should not expect — in the absence of further argument — that creatures formed by a naturalistic evolutionary process would have veridical perceptions. Hoffman is aware that he may be playing into Plantinga’s critique of naturalistic evolution as epistemologically self-defeating. In his most recent publication on the topic (Donald Hoffman and Chetan Prakash, “Objects of consciousness,” Frontiers in Psychology, published 17 June 2014), Hoffman responds to a criticism of his argument:

(9) You argue that natural selection does not favor true perceptions. But this entails that the reliability of our cognitive faculties is low or inscrutable, and therefore constitutes a defeater for belief in natural selection. See Alvin Plantinga’s argument on this (Plantinga, 2002).

Hoffman replies:

Evolutionary games and genetic algorithms demonstrate that natural selection does not, in general, favor true perceptions. But this entails nothing about the reliability of our cognitive faculties more generally. Indeed, selection pressures might favor more accurate logic and mathematics, since these are critical for the proper estimation of the fitness consequences of actions. The selection pressures on each cognitive faculty must be studied individually before conclusions about reliability are drawn.

But there are two problems with this reply. First, even if Hoffman’s argument were restricted to visual perception, and not to our cognitive faculties more generally (e.g., memory, introspection, a priori rational insight, testimonial belief, inferential reasoning, etc.), the conclusion that our visual perceptions would be wholly unreliable given natural selection would be sufficient for Plantinga’s conclusion of self-defeat. After all, reliance upon the veridicality of our visual perceptions was and always will be crucial for any scientific argument for the truth of evolution. So if these perceptions cannot be trusted, we have little reason to think evolutionary theory is true.

Second, it’s not clear that Hoffman’s application of evolutionary game theory is only specially applicable to visual perception, rather than being relevant for our cognitive faculties generally. If “we find that veridical perceptions can be driven to extinction by non-veridical strategies that are tuned to utility rather than objective reality” (2010, p. 504, my emphasis), then why wouldn’t veridical cognitive faculties (more generally) be driven to extinction by non-veridical strategies that are tuned to utility rather than objective reality? After all, evolutionary theory purports to be the true account of the formation of all of our cognitive faculties, not just our faculty of visual perception. If evolutionary game theory proves that “true perception generally goes extinct” when “animals that perceive the truth compete with others that sacrifice truth for speed and energy-efficiency” (2008), why wouldn’t there be a similar sacrifice with respect to other cognitive faculties? In fact, Hoffman regards the following theorem as now proven: “According to evolution by natural selection, an organism that sees reality as it is will never be more fit than an organism of equal complexity that sees none of reality but is just tuned to fitness” (Atlantic interview). But then wouldn’t it also be the case that an organism that cognizes reality as it is will never be more fit than an organism of equal complexity that cognizes none of reality but is just tuned to fitness? On the evolutionary story, every cognitive faculty we have was produced by a process that was tuned to fitness (rather than tuned to some other value, such as truth).

Won’t Get Foiled Again – A Rejoinder to Jerry Walls Wed, 27 Apr 2016 21:21:55 +0000 A couple of months ago (February 2016), Philosophia Christi published my and Steve Cowan’s rejoinder to Jerry Walls’s reply to our response to his original article on Christian compatibilism. Our abstract:

Jerry Walls has attempted to make the case that no orthodox Christian should embrace compatibilism. We responded to his arguments, challenging four key premises. In his most recent response, Walls argues that none of our rebuttals to these premises succeed. Here we clarify aspects of our previous arguments and show that Walls has not in fact undermined our defense of Christian compatibilism.

Given my October 2015 post, there are now five articles that compose this particular conversation. In chronological order, they are:

  1. Jerry L. Walls, “Why No Classical Theist, Let Alone Orthodox Christian, Should Ever Be a Compatibilist,” Philosophia Christi Vol. 13 No. 1 (Summer 2011): 75-104. [link]
  2. Steve Cowan and Greg Welty, “Pharaoh’s Magicians Redivivus: A Response to Jerry Walls on Christian Compatibilism,” Philosophia Christi Vol. 17, No. 1 (Summer 2015): 151-173. [link]
  3. Steve Cowan and Greg Welty, “Addendum: More Rebuttals of Walls from Pharaoh’s Magicians’ ‘Bag of Tricks’!” (Evangelical Philosophical Society website, September 16, 2015), 1-12. [link]
  4. Jerry L. Walls, “Pharaoh’s Magicians Foiled Again: Reply to Cowan and Welty,” Philosophia Christi Vol. 17, No. 2 (Winter 2015): 411-426. [link]
  5. Greg Welty and Steven Cowan, “Won’t Get Foiled Again: A Rejoinder to Jerry Walls,” Philosophia Christi Vol. 17, No. 2 (Winter 2015): 427-442. [link]

It is this last item (#5) that is my recent addition to this exchange.

[Blog post edited on 4/29/2016 to include the link to item #4.]

]]> 1 231
The Basis of Counterfactual Repentance Fri, 30 Oct 2015 18:20:25 +0000 A comment on “Chorazin, Bethsaida, and Molinism.”

In Acts 7, Stephen preaches a narrative that culminates in the prophetic witness concerning “the coming of the Righteous One” (v. 52).

In Acts 8:1 Saul was present to hear this, approving of Stephen’s execution. No doubt Saul was conversant with other Christian preaching as well, as he dragged Christians off to prison.

In Acts 8:26-40 the Ethiopian eunuch is converted through a simple Bible study.

But in Acts 9 Saul is converted by means which go beyond simple Bible study. God had to perform physical miracles, including seeing “light from heaven,” hearing Jesus’s voice, being struck blind, and getting healed from that blindness.

On the basis of this narrative, we could set forth a true hypothetical that parallels Jesus’s rebuke to the unbelieving Jews in Mt 11:21, a hypothetical that illustrates asymmetry of hardness of heart and guilt (thus grounding the rebuke), but accommodates the sufficiency of God’s converting grace for each and every conversion. Jesus could have said to Saul in Acts 9:

Woe to you, Saul! For if the Bible study that was presented to you by Stephen in Acts 7 had been presented to pagan Ethiopians, they would have repented.

Fact: Bible study hadn’t converted Saul up to that point, though he was certainly exposed to it (in his upbringing and recent experience).

Fact: Bible study would have been the means of converting at least some pagan Ethiopians, because it was the means of the eunuch’s conversion.

Fact: As an unbelieving Jew who rejected Christian preaching, Saul is therefore more guilty than the eunuch who accepted Christian preaching, and Jesus would be right to say so. Reflecting on his conversion experience, Saul confesses that he was the ‘foremost’ of sinners, and towards him Jesus was gracious to “display his perfect patience as an example to those who were to believe in him for eternal life” (1 Timothy 1:15-16).

Assumption: Both the eunuch and Saul were converted by God’s irresistible grace. The gracious influence is different in both cases because the stubbornness is different in both cases. But in each case, God’s grace is sufficient to ensure the outcome.

Fact: in the hypothetical scenario presented above, the same ‘mighty works’ are referred to: God’s gracious influence by means of a mere Bible study.

Fact: in the hypothetical scenario presented above, the same ‘mighty works’ that fail to convert Saul succeed in converting the eunuch.

Conclusion: the means needed to eventually convert Saul are greater than the means needed to convert the eunuch (thus preserving asymmetry of guilt and wickedness), but in each case God’s grace is sufficient to ensure the salvific outcome (thus preserving God’s irresistible grace). In short, the Bible study that proved irresistible in the eunuch’s context was resistible in Saul’s context, and so stronger means were needed (and in Saul’s case, they were graciously supplied).

Similarly, when Jesus says to the unbelieving Jews of his day:

Woe to you, Chorazin! Woe to you, Bethsaida! For if the mighty works done in you had been done in Tyre and Sidon, they would have repented long ago in sackcloth and ashes. (Matthew 11:21)

Jesus doesn’t need to be taken as implying, “Chorazin, you guys are so bad, that the irresistible grace that I could have used to convert Tyre and Sidon hasn’t worked with you!” On that view, ‘irresistible grace’ is resistible in some contexts (Chorazin) but not in others (Tyre and Sidon). That indeed seems contradictory. But instead Jesus can be taken as implying, “Chorazin, you guys are so bad, that the mighty works which would be irresistible for Tyre and Sidon prove resistible for you.”

God uses stronger means for stronger sinners, because they are stronger sinners. The fact that lesser means would succeed with others, while they fail with us, would be a reflection upon us. If you can’t be converted through a Bible study, but only through miraculous voices and blindness, God’s grace will be up to the task (as always). But then you should praise God that he persevered with you in the midst of your greater depravity and hard-hearted spiritual ignorance (as Paul does in 1 Timothy 1:15-16).

All of this helps us answer an objection to the Calvinistic doctrine of irresistible grace, cited at the link above:

But there is another problem with regards to the folks of Tyre. Neither the people of Chorazin and Tyre actually repented. On Calvinism, we could safely conclude neither were given irresistible grace, because had they being given irresistible grace, they would repent. But the verse gives us the counter-fact: the people of Tyre would have repented, given the same mighty works. So how is it that Tyre would have repented without irresistible grace? On Calvinism, we are left with the contradiction that irresistible grace both is and is not necessary for repentance.

It’s not “that Tyre would have repented without irresistible grace.” It’s that the works that would have been irresistible in Tyre’s context are resistible in Chorazin’s context. Thus the woe pronounced on Chorazin and the other cities in Jesus’s day, for rejecting Jesus’s preaching and miracles.

Are Molinist Distinctives Found in the Bible? Sun, 18 Oct 2015 00:16:41 +0000 RTS philosopher James Anderson has an interesting post which asks the question, “How Biblical is Molinism?” He concludes that there is strong biblical support for two theses held by Molinism: “God’s comprehensive providential control over his creation” and “God’s counterfactual knowledge of human choices.” But he points out that the latter kind of knowledge doesn’t have to be knowledge of libertarian human choices, and a good thing too, since it doesn’t seem that the Scriptures themselves indicate that this is the case.

Anderson closes by asking the question:

If we want to show that Molinism has better biblical support than Augustinianism (or vice versa) then we need to find some proposition p which is affirmed by Molinism and denied by Augustinianism (or vice versa) such that p enjoys positive biblical support (i.e., there are biblical texts which, on the most natural and defensible interpretation, and without begging philosophical questions, assert or imply p).

My working hypothesis with respect to this question is the following: the propositions in Molinism which have explicit or implicit biblical support are not distinctive to Molinism (they are common to other views, such as Augustinianism), whereas the propositions in Molinism which are distinctive to Molinism do not have explicit or implicit biblical support.

On my view, the following five theses are essential to Molinism:

  1. Infallible divine omniscience
  2. Meticulous divine providence
  3. Human libertarian free will
  4. Three logical moments in divine omniscience (natural knowledge, middle knowledge, free knowledge)
  5. God’s decree placed in between middle knowledge and free knowledge

As Anderson points out, theses 1-2 are common to other views of divine providence, such as Augustinianism. So although they are essential to Molinism, they are not distinctive to it. I predict that Anderson will argue that prospects are dim for deriving theses 3-5 from the Bible (unlike theses 1-2).

It is important to point these out as the sine qua non of Molinism, because unfortunately, in my experience, some non-Molinists misunderstand what is necessary for Molinism. In particular, they think that if you’re a Molinist, then it follows that you’re a Pelagian or even a semi-Pelagian, or that you deny divine sovereignty, or that you’re a crypto-Roman-Catholic. I think this is all a mistake. These charges either confuse systems with people, or fall prey to some sort of genetic fallacy (Molina was a Roman Catholic, so Molinism is inherently Roman Catholic.)

In addition, at least in my experience, I think that some Molinists have misunderstood what is sufficient for Molinism. I have heard loose talk in defense of Molinism according to which you’re a Molinist if you believe that:

  • There are logical moments of divine omniscience
  • There are unactualized possibilities
  • God knows how humans would behave in various circumstances
  • God uses his knowledge in order to be an intelligent and wise Creator
  • There is such a thing as divine permission

But it is arguable that non-Molinists of all stripes also believe these theses, which are not the same as theses 3-5 above.

To zero in on the point of this blog post, let’s reword things a bit and compress Molinist distinctives into the following three claims:

  • There are true counterfactuals of creaturely freedom, where the freedom is libertarian freedom. (So the CCFs are only contingently true, not necessarily true.)
  • God knows these CCFs logically prior to his decree. (So the CCFs are not based upon God’s will; they are prevolitional truths.)
  • God’s decree makes use of this middle-knowledge. God draws upon this middle-knowledge in order to figure out which possible world to actualize. (So the CCFs are used by God to steer world history to conform to his purposes.)

If even one of these claims is left out, then either middle-knowledge isn’t ‘middle’ or it isn’t relevant to providence. Notice that without libertarian free will, middle-knowledge isn’t ‘middle’. That’s because counterfactuals of creaturely compatibilist freedom constitute either natural knowledge (they are necessary, like the truths of logic) or free knowledge (they are based on God’s will). Therefore, those Calvinists who call themselves ‘compatibilistic Molinists’ are using misleading terminology at best. Notice again that if the counterfactuals aren’t known logically prior to God’s decree, then God can’t draw upon them as background knowledge to be used in formulating his decree. So they wouldn’t be relevant to providence.

With respect to the first claim on the list, that the kind of freedom we’re talking about is libertarian freedom, it does seem implausible to think the Bible speaks to the issue of human freedom with such specificity that it lays down, say, source or leeway conditions on human freedom. If you’re inclined to agree, theologian Alister McGrath has a theory as to where the idea really came from, arguing that:

the introduction of the non-biblical, secular Stoic concept of autexousia or liberum arbitrium in the articulation of the human response to the divine initiative in justification [was one of] two major distortions [which] were introduced into the corpus of traditional belief within the eastern church at a very early stage, and were subsequently transferred to the emerging western theological tradition (1: 18). Alister McGrath, Justitia Dei: a history of the Christian doctrine of justification (Cambridge: Cambridge University Press, 1986), 2 vols.

Among his conclusions are the following:

Indeed, by the end of the fourth century, the Greek fathers had formulated a teaching on human free will based upon philosophical rather than biblical foundations. Standing in the great Platonic tradition, heavily influenced by Philo, and reacting against the fatalism of their day, they taught that man was utterly free in his choice of good or evil (ibid., 1: 19).

It is quite possible that the curious and disturbing tendency of the early fathers to minimise original sin and emphasise the freedom of fallen man is a consequence of their anti-Gnostic polemic (ibid., 1: 20).

The weakness of Pauline influence in the early church may be illustrated from the fact that two non-Pauline, non-biblical terms (autexousia and liberum arbitrium) came to be introduced into the early Christian discussion of man’s justification before God… the ‘self-determination’ of the human free will is not so much a Christian idea, as a philosophical idea of its early Hellenistic milieu (ibid., 1: 22).

As for the second claim on the list, that the CCFs are known by God logically prior to his decree, William Lane Craig seems to agree that it is going to be difficult to derive it from the Bible:

Biblically speaking, it is not difficult to show that God possesses hypothetical knowledge. For example, Jesus affirms before Pilate the counterfactual conditional “If my kingship were of this world, my servants would fight, that I might not be handed over to the Jews” (John 18:36) RSV). The Scriptures abound with examples of such counterfactual conditionals concerning creaturely choices and actions. Unfortunately, this fact does not settle the matter of whether God has middle knowledge. For the scriptural passages show only that God possesses knowledge of counterfactual propositions, and, as I have said, until modern times all theologians agreed that God possesses such hypothetical knowledge. The question remains, when in the logical order of things does this knowledge come? Is it before or after the divine decree? Since Scripture does not reflect on this question, no amount of proof texting concerning God’s hypothetical knowledge can prove that such knowledge is possessed logically prior to God’s creative decree. This is a matter for theologico-philosophical reflection, not biblical exegesis. Thus, while it is clearly unbiblical to deny that God has hypothetical knowledge, those who deny middle knowledge while affirming God’s hypothetical knowledge cannot be accused of being unbiblical (WL Craig, pp. 83-84 of Stanley Gundry (ed.), Four Views on Divine Providence (Zondervan, 2011), emphasis mine).

Finally, what are we to make of the third and last claim on the list, that God in his decree makes use of his middle-knowledge, in order to figure out – intelligently and providently – which possible world to actualize? Consider God’s words to David about the men of Keilah, in a classic proof-text for Molinism:

9 David knew that Saul was plotting harm against him. And he said to Abiathar the priest, “Bring the ephod here.” 10 Then David said, “O Lord, the God of Israel, your servant has surely heard that Saul seeks to come to Keilah, to destroy the city on my account. 11 Will the men of Keilah surrender me into his hand? Will Saul come down, as your servant has heard? O Lord, the God of Israel, please tell your servant.” And the Lord said, “He will come down.” 12 Then David said, “Will the men of Keilah surrender me and my men into the hand of Saul?” And the Lord said, “They will surrender you.” (1 Samuel 23:9-12)

Surely we can concede that God does have the kind of knowledge indicated in this and similar texts. And let’s concede – simply for the sake of argument – that the texts speak of the libertarian free will had by Saul and the men of Keilah. Despite all that, does the inspired author indicate that God used this knowledge logically prior to his decree to create? That he planned creation by way of relying on this counterfactual knowledge? In fact, do any of the key Molinist prooftexts teach God’s providential use of middle-knowledge? Do they teach that God ‘steers’ the course of history by relying on this knowledge? (In the midst of our skepticism, though, let’s not suppress this interesting point: apparently God revealed a bit of his counterfactual knowledge to David in order to get David to do something. God ‘steers’ the universe by way of middle-knowledge at least to that extent! But that is a poor substitute for what Molinists are claiming more globally about God’s use of his middle-knowledge.)

In short, there is no text that says or implies: “From eternity God had knowledge of how free creatures would use their freedom and he used this knowledge in coming up with his decree, in figuring out what kind of universe to create.”

So, is this some sly disproof of Molinism? Not at all. What would follow from the fact that those claims which are distinctive to Molinism don’t find biblical support? That would depend on what we think is the proper relationship between the Bible and theological/philosophical ‘systems’. Here’s one view: I can only allow into my theology that which is explicitly taught in Scripture. Here’s another view: I can allow into my theology views that are implied by Scriptural teaching, as long as I have a good argument that it is an implication. And here’s yet another view: as long as a view is internally consistent (logically consistent with itself), externally consistent (logically consistent with everything else that is taught in Scripture), and has great explanatory power (for instance, its truth would help me to explain many difficult passages of Scripture that are otherwise hard to explain), then I am permitted to believe it because I have good reason for thinking it is true. As in science so in theology: great explanatory power counts toward truth, and therefore grounds justified belief. If I were a betting man, I would bet that Molinists would bet on this third view of the relationship between the Bible and theological/philosophical systems.

If so, they’d be in good company. Inter alia, the doctrine of the Trinity seems extremely plausible precisely because positing the triune nature of God helps to explain how the following passages of Scripture can all be true together:

  1. The “each Person is God” passages (the F is G, the S is G, the HS is G)
  2. The “persons are distinct” passages (the F is not the S, the F is not the HS, and the S is not the HS), and
  3. The monotheism passages (there is only one God).

Likewise, doctrine that the Holy Spirit is a Person is extremely plausible because it helps to explain how the following passages of Scripture can all be true together:

  1. We can grieve the Spirit, and only persons can be grieved.
  2. Jesus’s discourses in John refer to the Holy Spirit with personal pronouns.
  3. The Spirit can be sinned against, and only persons can be sinned against.

Or, perhaps, the doctrine of divine timelessness is extremely plausible because it helps to explain a whole range of otherwise disparate data (Why is God ethically immutable? How could he have created the entire spacetime continuum? Did he traverse an actual infinite to get to the present? As a perfect being, can he suffer continual lack and loss, such as lack of his future, and loss of his present?)

Maybe the justification for Molinism is like this, despite its lacking explicit or even implicit biblical support. If so, it is in good company… as long as the explanations powering its ‘explanatory power’ are good explanations, of course!

So far we’ve seen reason for my working hypothesis with respect to Anderson’s question, the hypothesis being: the propositions in Molinism which have explicit or implicit biblical support are not distinctive to Molinism (they are common to other views, such as Augustinianism), whereas the propositions in Molinism which are distinctive to Molinism do not have explicit or implicit biblical support.

But all that I have said above is consistent with something we might add to my ‘working hypothesis’: at least some of the propositions in Molinism which are distinctive to Molinism not only do not have biblical support; in addition, they simply don’t have the explanatory power their adherents claim. Since this is probably the most controversial aspect of my post, I’ve chosen to sneak it in at the end so that only persevering readers are rewarded. But it’s not much of a ‘reward,’ since I’m only going to talk about it in a future blog post, not this one.


]]> 1 202
A Response to Jerry Walls on Christian Compatibilism Fri, 02 Oct 2015 16:00:11 +0000 In Philosophia Christi Vol. 13, No. 1 (Summer 2011), Christian philosopher Jerry Walls published “Why No Classical Theist, Let Alone Orthodox Christian, Should Ever Be a Compatibilist.” The abstract (article available here):

I argue that no classical theist, and even more no orthodox Christian, should affirm compatibilism in our world. However plausible compatibilism may be on atheistic assumptions, bringing God into the equation should radically alter our judgment on this ongoing controversy. In particular, if freedom and determinism are compatible, then God could have created a world in which all persons freely did only the good at all times. Given this implication of compatibilism, three issues that are already challenging become extraordinarily more difficult, if not insuperable, namely: moral responsibility, the problem of evil, and the orthodox doctrine of eternal damnation.

Pedagogically, I was struck by the utter awesomeness of Walls’s metaphor for the libertarian/compatibilist debate (which metaphor he attributes to Michael Rea):

This debate has gone through several rounds, with neither side appearing to convince the other. I am very much in sympathy with arguments that defend libertarian freedom, but I feel the force of objections by critics who think the whole notion is mysterious, and at times even seems to be incoherent. Compatibilists, moreover, like Pharaoh’s magicians, seem capable of duplicating in their own terms every power and ability that libertarians claim their view distinctively grants to agents. (Walls 2011, 77, emphasis mine)

The comparison is ingenious, elegantly capturing what has been so characteristic of the free will debate in recent years. In any event, the gauntlet being thrown down – compatibilists are like pagan magicians! – Steven Cowan and I donned our Egyptian robes, authored a reply, and presented it at the November 2013 meeting of the Evangelical Philosophical Society in Baltimore, MD. We finally got around to submitting it this past winter, and it’s now been published in Philosophia Christi Vol. 17, No. 1 (Summer 2015) as “Pharaoh’s Magicians Redivivus: A Response to Jerry Walls on Christian Compatibilism.” You can read it here (with permission from the publisher). The abstract:

Jerry Walls has recently argued that no Christian theist should be a compatibilist because, on compatibilism, it is “all but impossible to maintain . . . the perfect goodness of God.” More specifically, he contends (1) that Christian compatibilism involves God in manipulation that undermines human moral responsibility, (2) that such manipulation makes God morally culpable for evil human actions, (3) that Christian compatibilism exacerbates the problem of evil in a way that Christian libertarianism does not, and (4) that Christian compatibilism entails universalism. In this paper, we argue that Walls is mistaken on all counts.

There is also an addendum to the article, entitled “More Rebuttals of Walls from Pharaoh’s Magicians’ “Bag of Tricks”!” which you can find here. In it, we develop fourteen additional arguments that rebut the case Walls made against compatibilism in his original article.

One potential criticism of our approach has already been passed on to me. The idea is that our arguments – in both the print article and in the addendum – presuppose that non-Molinist freewill theism is false. That is, we assume that, when contemplating his decree, God has knowledge of how his creatures would choose if God decided this or that (for compatibilists, this is natural knowledge; for Molinists this is middle knowledge). But maybe God doesn’t have this kind of knowledge, and if not, then our arguments beg the question against any non-Molinist freewill theist conceptions of divine providence.

I think this objection is quite interesting, and it illuminates what is at stake in this debate. Consider two non-Molinist freewill theist conceptions of divine providence: open theism, and (incremental) simple foreknowledge. I agree with William Hasker and Dean Zimmerman that these involve risky providence. As Hasker puts it, “God takes risks if he makes decisions that depend for their outcomes on the responses of free creatures in which the decisions themselves are not informed by knowledge of the outcomes” (Hasker and Helm 2003, 219). On this view, God’s risk-taking just is God’s providential decision-making in the absence of such knowledge. Since our arguments presuppose that God has such knowledge, they beg the question against these alternative conceptions of divine providence.

On one level, this is exactly right: both open theism and (incremental) simple foreknowledge involve risky providence, whereas Calvinism and Molinism do not. But no questions have been begged, because the question at issue throughout the article was are Walls’s original arguments cogent? rather than is non-Molinist freewill theism true? The task at hand was to assess Walls’s arguments, not to adjudicate non-Molinist freewill theism.

Beyond this, we took it for granted that being a “classical theist” and “orthodox Christian” – two terms from the title of Walls’s article – plausibly involved rejecting both diminished divine foreknowledge* and risky providence. One can hardly find these commitments in classical theism, after all (though noting this is not an argument for classical theism, of course.) If in fact Walls would like to defend his arguments by embracing diminished divine foreknowledge or risky providence, that is certainly a move he could make, but we are not begging any questions by not making that move for him. At best what the criticism offers us is a perspective on these matters that Walls could indeed try out. If Walls had done so in print, we certainly would have written a different article. The upshot is that this criticism raises the stakes for any defense of Walls, and that’s very helpful for anyone interested in this debate.

For what it’s worth, I think that shifting to a risky view of divine providence does little to get around the ethical conundrums raised in the article and addendum, since even on open theism God has ‘knowledge enough’ to generate reasonable facsimiles of these dilemmas. I try to develop and defend such an argument in “Open theism, risk-taking, and the problem of evil,” which will hopefully see print by the end of the year. To be sure, on a risky view of providence, the particular way things will go cannot be known by God at the moment of creation. This point is often stressed by its advocates. But the way things will go will be known by God (or believed with a high likelihood of truth) just prior to the choices themselves, at precisely a time when God is in a position to prevent the choices. Knowing this, God doesn’t prevent the choice, thus ensuring all the suffering that does occur. This kind of fallible divine knowledge is functionally equivalent to the kind of infallible divine knowledge we presuppose in the text of the Phil Christi article and addendum, and it seems to be ‘knowledge enough’ to generate ethical dilemmas for risky providence that are analogous to the ethical dilemmas we raise for risk-free providence.

*Is it a straw man for me to characterize open theism as involving “diminished divine foreknowledge”? I don’t think so. It’s a view of divine foreknowledge that considerably diminishes it from what classical theists and orthodox Christians have always held. As Zimmerman puts it (p. 36 of “Yet Another Anti-Molinist Argument”):

The Open Theists’ picture of foreknowledge and providence includes two theses that conflict with Catholic teaching and most Protestant theological traditions. Open Theism may save the letter of the traditional doctrine of God’s omniscience—God can know all truths, and yet not know what will happen, so long as there is now no fact of the matter about what will happen. Still, most Christians have affirmed something the Open Theist denies: that God has knowledge, at all times (or perhaps from a timeless perspective), of everything that will ever occur. (Zimmerman 2009, 36)

This last sentence strikes me as utterly obvious, and so open theism offers a conception of divine foreknowledge that is considerably diminished from that offered by the standard Roman Catholic and Protestant conceptions – that is, the classical theist and orthodox Christian conception of God referred to in the title of Walls’s article.

Hasker, William, and Paul Helm. 2003. “Does God Take Risks in Governing the World?” In Contemporary Debates in Philosophy of Religion, edited by Michael L. Peterson and Raymond J. Vanarragon, 218-41. Malden: Blackwell Publishing.

Walls, Jerry L. 2011. “Why No Classical Theist, Let Alone Orthodox Christian, Should Ever Be a Compatibilist.” Philosophia Christi 13: 75-104.

Zimmerman, Dean. 2009. “Yet Another Anti-Molinist Argument.” In Metaphysics and the Good: Themes from the Philosophy of Robert Merrihew Adams, edited by Samuel Newlands and Larry M. Jorgensen, 33-94. New York: Oxford University Press.

[Blog posted edited on 10/29/2015 to include the link to Walls’s original article.]

Before and after Sat, 09 Aug 2014 19:30:46 +0000 How running* can help demonstrate to your metaphysics class the difference between a thick particular and a thin particular:



before4 (july 2008)

before1.1 (july 2008)



after2.1 (july 2011) after1.1 (july 2011)

*for a year in 2011, in my living room, while watching MI-5.

This has been a public service announcement…

(And please, don’t ask me to take the next logical step by trying to demonstrate a bare particular. Really, no.)

Our experience of libertarian free will Wed, 06 Aug 2014 15:45:12 +0000 Flickers of Freedom is a nice blog, with contributors who are among the philosophical movers and shakers in the contemporary free will debate. This month’s guest contributor is Marcus Arvan, and my memory was jogged by a comment he made on one of his recent posts:

Another obvious source of trouble for Kant (the one that Grenberg points to in her paper on the Phenomenological Failure of Groundwork III) is that Kant seems to think that he establishes noumenal freedom through our experience of practical reason (i.e. the manner in which we *seem* to experience ourselves as libertarian-free from a first-personal perspective). As many critics of libertarianism have long noted, this kind of argument seems like super-poor evidence for libertarianism (I would dare say it is no evidence at all). Just because we *seem* to have libertarian free will when we act, that’s no reason to think that we *do* have it. After all, our seeming to have it might well just be a deterministic illusion!

Arvan’s point is that the empirical case for libertarian free will (LFW) – from our alleged experience of such freedom to the reality of such freedom – seems a bit thin. In his (excellent) book on mind-body dualism, The Emergent Self, William Hasker briefly advances the position that Arvan seems to be critiquing. Since I had to work through this issue a bit when prepping for my philosophy of mind class years ago, I’ll share a few thoughts that occurred to me then.

In ch. 4 of his book, on “Free Will and Agency,” Hasker notes that the philosophers Thomas Nagel and John Searle hold that we experience the following things:

  1. “Antecedent circumstances, including the conditions of the agent, leave some of the things we do undetermined: they are determined only by our choices, which are motivationally explicable but not themselves causally determined.” (Nagel)
  2. “My reason for doing it is the whole reason why it happened, and no further explanation is either necessary or possible.” (Nagel)
  3. “There are all sorts of experiences that we have in life where it seems just a fact of our experience that though we did one thing, we feel we know perfectly well that we could have done something else.” (Searle)
  4. “We know we could have done something else… we might have acted on those [other] reasons and chosen that something else.” (Searle)
  5. Human behavior isn’t predictable like the behavior of physical objects, because “we could have often done otherwise than we in fact did. Human freedom is just a fact of experience.” (Searle)

In light of the evident facts of empirical experience summarized in (1)-(5) above, Hasker then asks the question:

“Why shouldn’t we take the descriptions offered by Nagel and Searle at face value, as accounts of the way the world really is?”

That’s a pretty important question. After all, our experience has prima facie evidentiary value for what is the case. The fact that it seems to me that I am typing is as good a reason as any to think that I am actually typing. If how the world seems to us isn’t good reason for thinking that’s how the world really is, there are precious few things we are justified in believing.

Still, I can think of at least three replies to Hasker’s question.

One reply: because, quite simply, they’re not descriptive of our experience.

It is simply not the case that we experience the things that Nagel and Searle say we experience. Libertarian freedom is a power of some sort, a power to do otherwise in the same exact circumstances. Well, I only know about my powers by way of empirical experience, and then I reflect on that retrospectively. And even if you take all of my past empirical experience and add it all up, it simply isn’t fine-grained enough to support the contention that I have the power to do otherwise in the same exact circumstances. The Principle of Alternate Possibilities (PAP), and Hasker’s lengthy definition of ‘free will’ (p. 85), simply isn’t a deliverance of my experience. In fact, I want to say something much stronger here: it couldn’t be a deliverance of my experience.

What if I were to claim that, for every past choice you have made, if you had done something else instead, a little pink unicorn would have floated by and said, “I love you”? My claim is that if you had refrained from doing anything you in fact did, the unicorn would have communicated with you. Now, this is a possibly true thesis. You perfectly well understand what I am claiming. But your first question to me should be, “Why should I believe such a claim?” And the reason you should ask that is because, not only do we not have any empirical evidence for the truth of the unicorn claim, we could never in principle have such evidence.

For we never experience the truth of a counterfactual. We never have empirical access to what we would have done. We only have empirical access to what we have in fact done. (We only have one life to live, and we don’t get to rewind the tape.)

It’s hard to see how PAP, or LFW, is empirically more evident to me than the unicorn thesis. Why should I believe that, for every (free, responsible) choice I have made, I could have refrained from making that choice and done something different instead? What is my empirical evidence for this? How is this a description of my own experience? It isn’t.

That’s not to say we might not have various intuitions supporting PAP or LFW. But it isn’t right to say it’s a deliverance of our empirical experience. Searle claims that “the sense of alternative possibilities, is built into the very structure of conscious, voluntary, intentional human behaviour” (83). I disagree. I don’t experience my power to do otherwise in the same exact circumstances, and neither do you.

Another reply: whatever empirical component we have here is too vague to decide the issue between libertarians and compatibilists.

At best, what happens is that I am able to imagine myself refraining. But this is pretty vague. Maybe I could have refrained in slightly different circumstances (such as when I wanted to do otherwise). Or maybe I could have refrained in exactly the same circumstances. (The latter is actually hard to imagine: same desires, same thoughts?)

But the problem is that I don’t have access to every fact about my circumstances. I can’t keep it all in my head (biochemistry, all of my wants and desires, all of my experiences up to that time?). So I don’t know, when I imagine myself refraining, whether I’m imagining myself refraining in exactly the same circumstances, or not. So this isn’t enough for grounding PAP. It’s too vague.

And, of course, imagining myself having a power is quite different from experiencing myself having the power. I can imagine flying unaided, but that gives no credence to the view that I have the power to fly unaided.

One correspondent of mine put the point like this: “I can easily imagine a possible world very similar to this one in which I choose a different item from the menu than I did in fact choose.  But I can’t imagine a possible world absolutely identical to this one (at least with respect to past states) in which I choose differently, for the precise reason that my imagination just isn’t that fine-grained.  Yet libertarians often treat the first imagining as though it were the second.”

Yet another reply: psychological determinism is actually more empirically grounded than alternate possibilities.

Hasker says that “psychological determinism,” or “determination by the strongest motive… is empirically vacuous, since the ‘strongest motive’ can only be identified retrospectively, by seeing which motive has in fact led to action” (84).

But isn’t this at least superior to PAP? How do you determine PAP retrospectively? After all, at least with psychological determinism, you’ve experienced what seems to you to be your strongest motive at the time, and you identify that retroactively. How do you experience your ability to do otherwise in the same exact circumstances, either in the present or by reflecting on the past? These aren’t “palpable facts of experience” (85).

Hasker says, “Rejecting this [libertarian] understanding of experience ought to be recognized as a major form of skepticism, along with skepticism about the external world, skepticism about other minds, and other varieties” (84). But maybe it is just being cautious about what experience actually delivers? I think there can be some very powerful arguments for LFW, including arguments from intuition that link LFW with moral responsibility. But I don’t think the case for LFW is given any favors by grounding it in empirical experience.

None of this shows, of course, that we don’t have libertarian free will. All it shows is that we should be somewhat cautious about accepting empirical arguments for its reality. (And I think Hasker would agree; his argument for LFW from empirical experience is not the centerpiece of ch. 4.)