An Unfortunate Appraisal of Molinism: A Response to’s “What Is Molinism and is it Biblical?”

By Jonathan Thompson


March 24, 2017

Summary: CARM, a Reformed ministry, has given a negative appraisal of Molinism. I lay out 9 issues with their explanation and critique of the doctrine, undercutting/rebutting their claims by showing that they either rely on comprehensional errors, dictional ambiguity, or question begging (or some medley of these).1


Issue 1: CARM Misrepresents Libertarian Free Will
CARM’s article starts out well enough by noting the fact that Molinism attempts to harmonize God’s omniscience with human beings possessing libertarian free will (LFW), but then there’s this gem:

Libertarian free will is the teaching that fallen man is still sufficiently free to be able to equally choose or reject God if only presented with the right information.

This is not the libertarian view. LFW is instead the position that the freedom we possess that is necessary for moral responsibility is incompatible with determinism. Molinists aren’t using a peculiar definition here either. Just consider the definition given by the atheist philosopher Mark Balaguer. He writes,

Libertarianism is the view that human beings are [libertarianly]-free, where a person is [libertarianly]-free if and only if she makes at least some decisions that are such that (a) they are both undetermined and appropriately nonrandom, and (b) the indeterminacy is relevant to the appropriate nonrandomness in the sense that it generates the nonrandomness, or procures it, or enhances it, or increases it, or something along these lines.2

Alternatively, consider the definition given by the non-Molinist philosopher Robert Kane; a man who has spent his entire career advancing the debate on free will. In his book “A Contemporary Introduction to Free Will”, Kane summarizes the libertarian thesis as “the view that (1) free will and determinism are incompatible (incompatibilism), (2) free will exists, and so (3) determinism is false”.3 Kane’s explanation further includes the notion of ultimate responsibility, specifically that “. . .an agent must be responsible for anything that is a sufficient reason, cause, or motive for the action’s occurring”.4

Finally, in an essay entitled “How to Think About Free Will”, philosopher Peter Van Inwagen. delineates a thesis for free will which Libertarians and Compatibilists would both agree upon. Van Inwagen then proceeds to define each view, explaining the libertarian position as “the conjunction of the free-will thesis and incompatibilism.”5 It’s also noteworthy to mention that Van Inwagen is theologically open theist.

Also, it’s striking that CARM doesn’t provide any sources revealing where they derive this notion that their definition of LFW is how Molinists (or philosophers, in general) typically understand the concept. This, taken in conjunction with the previous facts suggest that CARM’s definition of LFW is purely idiosyncratic.

Now, at this point one might respond that CARM wasn’t actually trying to define LFW, but instead were presenting what they believe to be an entailment or implication of the view. Any argument in favor of this position, however, will have at least four difficulties to overcome. First, with respect to the structure of the article, the relevant definition is placed in the opening paragraph which purports to explain what Molinism is, not evaluate it. Second, the author explicitly states that the relevant definition is itself the teaching of LFW. Third, the author doesn’t attempt to offer any reason to believe that this is an implication or entailment of LFW. Finally, (as of this writing) the article includes a hyperlink to another essay written by the same author, where, once again, he incorrectly defines LFW, explaining it as the view that a “person’s will is not restricted by his sinful nature, and that he is still able to choose or accept God freely”.6

Thus, if on one hand, we accept the point that CARM is providing a definition LFW, then, as I have argued, this definition will simply count as misrepresentation. But, if on the other hand we accept the idea that CARM is presenting what they think to be an entailment or implication of LFW, then CARM’s explanation will still be misrepresentative, but only this time by them confusing a view with its alleged entailment or implication.7

Issue 2: CARM Gives an Ambiguous Example of Natural Knowledge
Following the author’s explanation of LFW, he goes on to define and provide examples of the three logical moments in God’s knowledge. One issue here involves the author giving an ambiguous example of natural knowledge. Specifically, he describes God’s knowledge of “all potential events that might occur under different circumstances“ as an instance of it. This statement, when formally understood, is simply mistaken. Natural knowledge does not disclose to God, which, of the possibilities He knows, might occur. Instead, to quote the author, it informs God of “all things that are possible and logically necessary” – that includes what any libertarianly free creature could do under any set of freedom permitting conditions.

Now, lest I be accused here of making a distinction without a difference it’s important to understand that although the words ‘could’ and ‘might’ are often used synonymously in an informal context, – and CARM might be doing that here – there are nevertheless formal modal differences between the two that need to be respected. William Lane Craig explains,

“Might” counterfactuals should not be confused with subjunctive conditionals involving the word ‘could’. ‘Could’ is taken to express mere possibility and so is a constituent of a modal statement expressing some possible truth. The distinction is important because the fact that something could happen under certain circumstances does not imply that it might happen under those circumstances. “Might” is more restrictive than “could” and indicates a genuine live option under the circumstances, not a bare logical possibility.8

To provide an illustration highlighting the differences between ‘could’ and ‘might’ counterfactuals consider the following scenario: Suppose that Sakura is a vegan who has been practicing this lifestyle for her entire life. Moreover Sakura has never had any type of cravings for meat and also knows herself to be allergic to most types of meat. One day Sakura decides to go to her favorite restaurant whose menu has vegan options but also happens to serve explicitly meat-based dishes. In this scenario although it could be the case that Sakura orders for herself something which she knows is meat-based and which she is allergic to, nevertheless, we will not want to say that such a scenario might obtain. All things being equal, given what Sakura knows about herself, ordering a meat-based meal will likely not even be taken in to consideration.

To give a more controversial example, consider how the method of inference to the best explanation is employed throughout assessing the historical evidence for the resurrection of Jesus. In examining the evidence historians will first be confronted with certain data to be explained (e.g. Jesus’ empty tomb, his post mortem appearances, the origin of his Disciples belief in his Resurrection, etc.) From there a pool of live options will be assembled consisting of various explanations for the data in question (e.g. Apparent death theory, Displaced body theory, Hallucination theory, etc.). Finally, the historian will choose a theory from the the pool of live options which, if true, best explains the data.

Now, it’s possible that the theory preferred by the historian in this scenario is false. In fact, it could be the case that every theory allowed within the pool of live options is incorrect and that the facts surrounding the Resurrection are actually best explained by an extravagant theory involving extraterrestrial shape-shifters using highly advanced technology. Now although such a theory could obtain, nevertheless, given what we know about the world, no intellectually responsible historian would take such a theory seriously and therefore wouldn’t include it within the pool of live options.

In short, these examples serve to highlight that although might counterfactuals entail could-counterfactuals, nevertheless, this relation is asymmetrical. As a result, it cannot be said that God having knowledge of any particular could counterfactual entails that He knows, as a part of His natural knowledge, whether or not the relevant counterfactual might occur. The question then raised is this: Since Molinists agree that God knows the truth-value of might-counterfactuals, where, in His knowledge, are these counterfactuals located?

The answer lies is in middle knowledge. As Craig has noted, “in the customary [Stalnaker-Lewis] semantics for counterfactuals conditionals ‘would’ counterfactuals logically imply ‘might’ counterfactuals”.9 This is the result of might counterfactuals being defined on this theory as “not-would-not”.

To demonstrate the point, if we use the modal operator “→” to denote the might counterfactual, then the statement “P ◇→ Q” will be read as “If P were the case, then Q might be the case”, which is equivalent to the consequent that “If P were the case, then it’s not the case that Q would not be the case”.

Now let the modal operator “→” symbolize the would counterfactual. The statement “P □→ Q” would be read as “If P were the case, then Q would be the case”. Well, if it’s true that Q would be the case if P were the case, then that entails that if P were the case, then it’s not the case that Q would not be the case, which just is the double negative statement we saw before.

We can now use “⊢ “ to symbolize this entailment relation as (P □→ Q) ⊢ (P ◇→ Q), Q.E.D.

In short, while it’s true that Molinists believe that God knows the truth-value of might counterfactuals, this knowledge will nevertheless be located in middle rather than natural knowledge.10

Issue 3: CARM Gives a Potentially Misleading Definition of Free Knowledge
Next, a quick word of caution concerning CARM’s definition of free knowledge is in order. They define it as God’s necessarily knowing “in totality all that actually exists that He has freely chosen to create”. This definition, however, can be suggestive of the idea that free knowledge pertains to a world which God has already created as opposed to one that He has merely made the intellectual decision to create. Molinism denies the former view and affirms the latter. Kirk MacGregor further elaborates on this distinction in his biography of Luis de Molina. MacGregor writes,

Only after apprehending his middle knowledge does God choose one particular world composed of various logically compatible (i.e., compossible) agents and circumstances to be the actual world. It should be emphasized that Molina did not say God creates the actual world at this point, but rather that God decided or decrees which feasible world will be actual. In fact, Molina did not believe that God brought the actual world into being simultaneous with his divine creative decree. For the divine creative decree was issued from eternity past (and God’s free knowledge was held from eternity past), whereas the actual world came into being at the first moment of time.11

So, while CARM’s definition of free knowledge is technically correct, still, it will be helpful to provide a more explicit and comprehensive definition of free knowledge in order to offset some of the ambiguity that permeates CARM’s explanation.

Issue 4: CARM Gives an Inaccurate Summary of Molinism
If the previous issue weren’t enough, another word of caution is in order with respect to CARM’s summary of Molinism. They summarize it as follows:

Basically, we can see Molinism as the teaching that God knows what the potential free will choices of people will be and chooses who will be saved based on that knowledge. In other words, God sovereignly predestines and saves those whom He knows will choose Him.

First, the word ‘potential’ requires some disambiguation. The standard use of the term can denote mere possibility or likelihood12, but not both simultaneously. If the author is using the word to denote mere possibility, then CARM’s summary will be inaccurate. Why? Because God’s saving individuals solely based on his knowledge of the bare possibilities available to these persons is an exemplification of natural but not middle knowledge.

Alternatively if we understand the word ‘potential’ to denote those possibilities which are likely to occur, then CARM’s schema won’t fully incorporate natural knowledge since the schema fails to include God’s possessing knowledge of those possibilities which aren’t likely to occur.

The last way CARM’s use of the relevant term could be understood is that they are using it to denote both mere possibility and likelihood. This route, while being able to circumvent my previous criticisms, is, as noted before, an idiosyncratic use of the term and so is misleading.

That aside, perhaps the most glaring issue with CARM’s summary of Molinism is that the author presents this doctrine as if it were a soteriological theory when it’s not. Instead Molinism is a theory of omniscience and providence that can be applied to make sense of soteriology just as it has been applied to make sense of other areas of theology such as Biblical Inspiration or Christology.13 14 That in mind, notice then how wrong-headed it would be to describe Molinism, for example, as the teaching that God knew what the potential choices of the Biblical authors were and then sovereignly orchestrated the Bible in accordance with that knowledge. Such an appraisal would clearly be confusing Molinism itself with its being incidentally employed in to further theological reflection.

Issue 5: CARM Misrepresents The Molinist View of Divine Cognition
Following this, the author begins his critique of Molinism. He writes, “First, [Molinism] means that God looks into the future to see what people will do and saves them based on their choices.”

The author provides no justification for the idea that Molinists believe this or else for the notion that this is somehow an implication or entailment or of Molinist theology. In fact, Molinists have been very forthright in their denial of God’s foreknowledge being acquired via Him “looking in to the future”. Before I elaborate on this, however, it will be helpful to distinguish between perceptualist and conceptualist models of divine cognition. Craig explains the differences,

The perceptualist model thinks of God’s cognition on the analogy of sense perception. This model is implicitly presupposed when people talk. . . of God’s “foreseeing future events.” He somehow looks ahead in time and “sees” what is there.15

Craig follows this up by explaining the conceptualist model in that it “thinks of God’s knowledge more on the analogy of innate ideas.”16 Craig further explains that “[a]s an essentially omniscient being God has the property of believing only and all truths. He didn’t get this knowledge from anywhere; He just has it innately.”17

So, when Craig, is asked a question related to the Molinist position on divine cognition, he clarifies it by saying,

Indeed, the doctrine of middle knowledge just is a theory of how God can know future contingents without any sort of perception of the world at all. I think that you have mistakenly assumed that, according to the doctrine of middle knowledge, God deduces from the circumstances in which a free person is placed what he would do in those circumstances. . . But that is not the doctrine.18

Kenneth Keathley, a Molinist, has also echoed this. He writes,

Since God is omniscient, He innately knows all things—this means He does not go through the mental processes that finite beings do of “figuring things out.” God never “learns” or has things “occur” to Him. He already knows all truths.19

Finally, there’s the words of Molina himself:

. . .before (in our way of conceiving it, but with a basis in reality) [God] creates anything at all, He comprehends in Himself—because of the depth of His knowledge—all the things which, as a result of all the secondary causes possible by virtue of His omnipotence, would contingently or simply freely come to be on the hypothesis that He should will to establish these or those orders of things with these or those circumstances: and by the very fact that through His free will He established in being that order of things and causes which He in fact established, He comprehended in His very self. . .20

So, it’s simply no part of Molinist theology to claim that God’s foreknowledge is acquired by Him looking in to the future. Of course, CARM could argue that Molinism implies a perceptualist model of divine cognition, but, to repeat, they offer no justification for this assertion.

Issue 6: CARM Misunderstands What It Means To Be Partial
CARM then attempts to draw out the implications of their straw man by inferring,

Therefore, God reacts to man’s choices, and God saves a person based on some quality (the ability to make a right choice) that the person possesses. But this is showing partiality (favoring one person over another based on a quality in/of that person)–something that God condemns (Rom. 2: 11).

CARM’s understanding of partiality seems dubious. For if showing favoritism is a vice, then it’s not at all clear that CARM’s understanding of it captures the essence of this sin. To illustrate the point, all things being equal, are we to really believe that an employer choosing to promote one employee over another based on the first possessing that quality of being the harder worker somehow illustrates a vice in the employer? That seems terribly wrong-headed yet CARM’s understanding of partiality implies that the employer is somehow sinning.

Properly understood, partiality is not merely showing favor towards someone over another, basing one’s favor on some quality that either individual possesses. Instead partiality is bias demonstrated towards some particular over another in such a way that is morally unjust. Moreover, this isn’t to argue that there simply are no cases in which partiality can occur under CARM’s definition. For example, it seems perfectly reasonable to think that under general circumstances it will be wrong to exclusively afford men the right to vote in Presidential elections simply because these individuals are men. However, the point here is that something can satisfy CARM’s definition and yet still fail to illustrate some unfair bias.

Additionally, recall that CARM’s inference concerning God’s reacting to man’s choices and acting partially was inferred from the prior premise that states that Molinism teaches or else implies a perceptualist model of divine cognition. As I showed in the previous section, Molinists do not hold to such a model. Furthermore, it was also shown that CARM didn’t even attempt to argue in favor of the idea that Molinism somehow implies the relevant model. Thus, even if we were to grant CARM’s understanding of partiality, their argument will either apply to a position that Molinists do not hold to – in which case, they are no longer offering a critique of Molinism at this point – or else CARM will still have to prove that Molinism implies the aforementioned partiality since they have not shown the premise from which they derive this inference to be implied by Molinism.

Finally, it’s worth noting that Molinism has no position on the criteria which God uses to judge individuals. This is because, as stated previously, Molinism is a view about divine omniscience, providence and nothing more.

Issue 7: CARM Misunderstands Particular Conceptions of LFW and Begs The Question Against Them
CARM then moves on with their next criticism. They write,

Molinism does not answer why one person chooses God and another does not when it is God who makes the person and puts him in that place and time. In other words, what is it about the human free will that God has made that enables him to choose God or not? Just saying it is up to the individual doesn’t answer the question.

A more succinct way of construing this objection would be to say that Libertarian views are incapable of providing a sufficient reason for why an individual makes one choice over another with respect to salvation. One issue with this objection is that it assumes that there must be a sufficient reason for why an individual chooses one way over another. More precisely, the problem with this assumption is that it begs the question against certain conceptions of LFW since these views maintain that our free choices just are brute facts (incapable of having an explanation) with respect to there being any sort of explanation for a persons choice outside of the libertarian agent being the cause of it.

Second, CARM’s objection fails to understand particular conceptions of LFW on the comprehensional level since it supposes that a metaphysically impossible state of affairs could obtain. In order to explain the rationale behind this objection, it’s important to clarify that I’m not claiming that it’s incoherent to ask for a why a person chose A. Moreover, it’s no part of my objection to claim that there cannot be reasons or even sufficient ones explaining why the individual made that choice. Rather, the claim is that on some conceptions of LFW, incoherence arises from asking the contrastive question why a libertarian agent freely chose A rather than B.

To illustrate the point, we can conceive of two otherwise identical logically possible worlds, W1 and W2. In W1 Neji is given the free choice between A and B, and chooses A rather than B. Moreover, he has reason R1 explaining why he chose A. Now let’s suppose that in W2 Neji is presented with the same choice, but instead chooses B rather than A. Further, Neji, again, has reason R1 explaining why he made this choice.

In these scenarios, R1 would seem to explain why Neji chose A in W1. R1 would also serve to explain why he chose B in W2. However, even if we concede that R1 explains why he made these choices in each of these worlds, notice that it does not explain why he chose A rather than B in W1 (or why he chose B rather than A in W2). This is because on the relevant libertarian views, there simply isn’t any explanation of Neji’s free decision beyond his choosing. Thus, it will simply be impossible to provide a reason as to why Neji chose A rather than B since there is no state of affairs in which libertarianism (as it is conceived for this example) obtains and there be a free choice whose explanation is found in anything outside of the libertarian agents free decision.21

Third, CARM’s objection assumes a version of the Principle of Sufficient Reason (PSR) which Libertarian’s aren’t necessarily committed to and so could argue against. To argue this point, we can turn CARM’s objection that there must be a sufficent reason for an agent freely choosing A over B and generalize it so that it states:

(1) every contingent truth or proposition must have a sufficient explanation.

Unfortunately the author merely assumes (1) without providing any justification for it and so begs the question against those Libertarians who aren’t committed to it. But let’s suppose that the author didn’t do this and instead presents justification for thinking there is a problem with Libertarians accepting the existence of brute facts. One way the author could do this is by arguing that (1) is non-inferentially justified by virtue of being an intuitive metaphysical truth. This would then shift burden of proof back on to the the Libertarian so that they would then have to either defeat (1) or else demonstrate that LFW is not at odds with it.

One approach a Libertarian could take in responding to this principle is by simply rebutting it. William Lane Craig, for example, thinks that not every fact is capable of having an explanation. For the Big Contingent Conjunctive fact (BCCF), which is the conjunction of all contingent facts, seems to be utterly incapable of having an explanation. Why? Because, according to Craig, if the explanation of the BCCF is contingent, then that explanation requires an even further explanation, which just is impossible since the BCCF includes all contingent facts there are. Alternatively, if we suppose that the explanation for the BCCF is itself necessary, then the fact to be explained (the BCCF) by this explanation must also be necessary, which is impossible, since the BCCF is contingent.22 Thus, if Craig’s appraisal of the BCCF is correct then CARM’s original objection will be defeated since the underlying principle from which it is implied is false.

Perhaps then, being met with the problem of the BCCF, the author decides to attenuate (1) so that it states:

(2) every contingent truth, proposition, or state of affairs either has an explanation for why it obtains or has a sufficient explanation for why no explanation can be given.

Our previous illustration involving Neji neatly conforms to the latter desideratum of (2). More specifically, it can be said that no explanation can be given for why a state of affairs obtains where a Libertarian agent freely chooses A over B (outside of the agents free decision) because such a state of affairs is metaphysically impossible (on some conceptions of LFW). These alleged states of affairs thus describes nothing at all, and so describe no contingent truth or proposition.

Alternatively the Libertarian could be sympathetic to (1) by accepting it and then offering an account of how LFW could be construed to be compatible with it. One such way this could be done is adopting the view that the sufficient reason that an Libertarian agent makes a free choice over another just is all the reasons the agent had for choosing the former. Another view the Libertarian could hold to is the idea that an agent’s choices are self-explanatory; that just as necessary truths sufficiently explain themselves, so do these contingent truths concerning our choices.23 I’m not particularly persuaded by either of these views and so won’t bother defending them, however, these are options available to the Libertarian should they so happen accept the relevant PSR.

Issue 8: CARM Implies That Molinism Is Relevantly Open Theistic Without Providing Any Justification
CARM then proceeds with their next criticism. They write,

Second, Middle Knowledge means that God learns what the actual choices of people will be only when they occur. God would then be ignorant about man’s future choices. This violates the scripture that says God knows all things. . . [God] doesn’t learn. He knows!

This assertion that God knows people’s choices upon their occurrence is in flagrant contradiction to the author’s previous claim that Molinism means that God’s foreknowledge of people’s free choices is acquired via Him looking in to the future. The author thus seems to be talking out of both sides of his mouth. Moreover the author provides no justification for the idea that Molinists believe this or else for the notion that this is an entailment of Molinism.

Perhaps then, we could extend charity to CARM and interpret their assertion to mean that Molinism is internally inconsistent in so far as certain aspects of the doctrine imply two mutually exclusive beliefs. Where’s the argument for that though? Indeed, the author provides none. Thus, despite our attempt to interpret CARM more charitably, we still have no reason to believe that the author intended his objection to be understood in this way.

Issue 9: CARM Shotgun Prooftexts
Finally, CARM proceeds with their last criticism:

Third, Middle Knowledge (as it relates to human freedom) fails to properly understand the depravity of man. The scriptures do not say that the unregenerate can freely choose God. In fact, the contrary is taught. It is man who is deceitful (Jer. 17:9), full of evil (Mark 7:21-23), loves darkness (John 3:19), does not seek for God (Rom. 3:10-12), is ungodly (Rom. 5:6), dead in his sins (Eph. 2:1), by nature a child of wrath (Eph 2:3), cannot understand spiritual things (1 Cor. 2:14), and a slave to sin (Rom. 6:16-20). Therefore, what is important here is understanding that an unbeliever is incapable of understanding and accepting Christ given the condition of his fallen, unregenerate state. This is why the Bible says such things as it is God who appoints people to believe (Acts 13:48), chooses who is to be holy and blameless (Eph 1:4), calls according to His purpose (2 Tim. 1:9), chooses us for salvation (2 Thess. 2:13-14), grants the act of believing (Phil. 1:29), grants repentance (2 Tim. 2:24-26), causes us to be born again (1 Pet. 1:3), draws people to Himself (John 6:44, 65), predestines us to salvation (Rom. 8:29-30) and adoption (Eph 1:5) according to His purpose (Eph 1:11), makes us born again not by our will but by His will (John 1:12-13), and works faith in the believer (John 6:28-29).

This objection emanates sophistry. First, the author engages in what we shall call ‘shotgun prooftexting’. This fallacy is much like shotgun argumentation where a person gives so many arguments in succession to one another that the sheer mass of arguments overwhelms the opponent. This objection, however, is different in so far as the author gives an exorbitant amount of Bible verses to overwhelm the Libertarian. Obviously, as Christians, we will not want to say that it is fallacious to use Scripture in arguing theological point. However, the problem lies in how the author employs these verses. Note that the author merely assumes that these verses are in conflict with LFW and provides no further explanation as to how these verses prove the point. The author has therefore begged the question against the libertarian view by assuming to be true what he is supposed to be proving (i.e. that Scriptural teaching is in conflict with LFW).

At this point, for the sake of a less monotonous discussion, we’ll construct a brief argument for CARM based on one of the verses that they use and attempt to demonstrate how LFW could be compatible with it.

We’ll choose 1 Peter 1:3 which states that God causes Christians to be born again. If one is a Determinist, then one could understand this verse as teaching causal determinism since the verse explicitly says that God causes the new birth. This would rule out LFW given that the Libertarian denies that God causally determines us to be born again. The principle issue with this objection, however, is that fails to take in to account Molinist distinctions concerning how God causally brings about some state of affairs. Specifically, Molinists distinguish between strong and weak actualization. On Molinism, God strongly actualizes a state of affairs when he directly brings it about through an exercise of His causal power. God causing the universe to come in to existence would be one example of this. Weak actualization, in contrast, occurs when God places an individual in circumstances where He knows how they would freely respond. So, for example, on the Molinist schema one would say that God has caused me to write this article only in the sense that He knew that I would, given the freedom permitting circumstances He has placed me in.

With respect to 1 Peter 1:3, the Molinist could understand this verse as teaching that God has caused us to be born again in the sense that He has weakly actualized our new birth. So this verse poses no problem for the Libertarian. Further, for CARM to merely assume that this verse teaches causal determinism, in light of the previous distinctions, simply begs the question against Molinism since it assumes without argument that the verse cannot be understood in terms of weak actualization. Of course, the Molinist would also beg the question were he to assume without argument that the verse is referring to weak actualization. However, the point here is that the verse is compatible with Molinistic as well as deterministic interpretations and so does not compel itself either way.

In conclusion, I have shown that CARM hasn’t posed any insuperable challenges to Molinism. Their explanation of Molinism, LFW, the Molinist view of divine cognition, and the Biblical notion of God’s impartiality rely on simple comprehensional inaccuracies. Further CARM’s explanation of both natural knowledge and free knowledge are ambiguous. We saw that the former definition collapses the distinction between natural and middle knowledge while the latter definition is inchoate in that it is unclear whether the relevant knowledge pertains to a world which God has already created or else to one which He has merely made the intellectual decision to create. Moreover, CARM’s claim that Molinism is relevantly like Open theism as well as their claim that LFW is incompatible with Biblical teaching are simply question-begging.

I’ve gone out of my way to be charitable towards CARM. I’ve not only strived to understand their claims in the strongest, most coherent way possible, but have even gone so far as to present arguments in their favor in order to buttress their claims and still have found these arguments lacking. One of the marks of a good debater is their ability to understand viewpoints that separate from their own and treat these views carefully and with charity. Unfortunately, the author of the CARM article was overwhelmingly deficient in both of these areas, which is especially unfortunate given that CARM is a Christian ministry.

My word of advice to CARM is that they restructure their approach with respect to how they research, explain, and critique views that they disagree with. The fact is that they have not painted themselves as an intellectually rigorous ministry in so far as the strongest arguments given against Molinism in this paper are one’s that I, myself, have constructed for them.24


1Special thanks to Jesse Roach for guiding me in formulating my abstract as well as my conclusion.

2Balaguer, Mark. Free Will As An Open Scientific Problem (pg. 7)

3Kane, Robert. A Contemporary Introduction to Free Will (pp 32-33)

4Ibid. (pg. 121)

5Van Inwagen, Peter. How to Think About Free Will (pg. 330)


7For a further explanation of the problems with conflating a position with its entailments see my article Some Reasons To Keep Assertions and Their Implications/Entailments Distinct.

8Craig, William Lane, Philosophical Foundations For a Christian Worldview (pg. 53)


10 Special thanks to Luke Van Horn for helping me understand the Stalnaker-Lewis semantics for counterfactual conditionals.

11MacGregor, Kirk R. Luis de Molina: The Life and Theology of the Founder of Middle Knowledge (pp. 246-247) Zondervan, Kindle Edition


13William Lane Craig has argued for a Molinist model on Biblical Inspiration in his essay Men Moved By The Holy Spirit Spoke From God.

14Thomas Flint has attempted to construe a Molinist Christology in his essay The Possibility of Incarnation: Some Radical Molinist Suggestions.





19Keathley, Kenneth. Salvation and Sovereignty (p. 16). B&H Publishing. Kindle Edition.

20Luis de Molina, Disputation 49.8

21Special thanks to Tyler Journeaux, a Thomist, for allowing to adapt his particular argument in to this paper. Visit his website at

22For Craig’s full response to the BCCF go to

23For a defense of these views see Alexander Pruss’ The Principle of Sufficient Reason: A Reasessment (pp. 135-136)

24Special thanks to Randy Everist and Ricardo Martinez for being minds for which I could ask questions and ask for evaluation. Visit Randy Everist’s website at


About the Author

By Jonathan Thompson