Does hypothetical knowledge Calvinism have an odd ontology of personhood?

In an ETS paper in 2013, John Laing critiqued Bruce Ware’s model of providence which is very much like my own “hypothetical knowledge Calvinist” model. In a long post on March 10, I explained why John Laing is wrong to think that hypothetical knowledge Calvinism is vulnerable to the grounding objection that Calvinists and Open Theists bring against Molinism. Next, I responded to his second criticism, that it has an “odd notion of necessity/possibility” (pp. 8-11). Laing’s third concern is that hypothetical knowledge Calvinism includes an “odd ontology of personhood,” and I want to deal with that complaint now.

In my conversation with Paul Helm (WTJ 71/2), he pointed out an entailment of my hypothetical knowledge Calvinism that I had not previously discerned. He observed that in my earlier work arguing for Calvinist middle knowledge, in order to remain true to Calvinist principles, the distinction which Molinists make between “would” and ”could” should not be retained. Reflecting on that point, which I have now conceded, Laing writes:

The would/could equation also suggests that, for any given possible set of creaturely actions, there exists a distinct creature. This has the unusual consequence of maintaining that the John who would eat chocolate cheesecake in situation S is a different John from he who would not eat the cheesecake in S (presumably due to some difference in character, desires, disposition, etc.). For any given set of decisions I might make (since all are necessary), there is a distinct me (or should I say, John that is not-me). Thus different possible worlds are populated by completely different beings  (p. 11).

Although Laing notes that Ware “certainly does not” affirm this, and although he is “not aware of anyone explicitly arguing for this position, he suggests that “some of Tiessen’s comments are suggestive of it” (p. 11, citing my discussion of “how God could bring about a different result than that spoken of in the counterfactual” [n. 12]). Laing’s perception of what is entailed in my comments, as unpacked in the statement above, is what leads him to conclude “that the Calvinist is committed to an odd notion of ontology and perhaps modality” (p. 11).

Since I reject the way in which Laing has conceptualized the situation in hypothetical knowledge Calvinism, I deny that the oddness about which he is concerned need trouble us. Here is how the situation should be described. For starters, I grant that there is a truth embedded in Laing’s statement that “different possible worlds are populated by completely different beings,” since what he has in view is the essence of what makes them different worlds. The problem lies in Laing’s speaking in terms of “different beings.” The critical factor that brings about different outcomes in different worlds (e.g., Laing’s eating or not eating cheesecake) is not in the being who acts in S, but in that the long series of circumstances lying behind Laing’s decision at that moment has been different, and that is what brings about a different action on Laing’s part, as anyone who knew all the details of the situation and the principles of causation could discern. Granted, there is a sense in which the Laing confronted with the opportunity to eat cheesecake in world A is “different” from the one in world B, but this is because, in world B, different circumstances have brought Laing to that point, so he has the different desires, dispositions etc. which make him a being who would not eat the cheesecake in world B, though he would have eaten it in world A. The difference is explicable by comparing the complex of events prior to Laing’s choice in each of the worlds which are similar enough that the person in those worlds could be deemed the “same person,” in spite of the personal differences brought about by different prior histories.

What Laing is observing, therefore, is nothing other than what we can all see with regard to ourselves over time. Think of the common statement: “In ten years you will be the same person you are today, except for two things, the books you read and the people you meet.” No one finds that an odd thing to say, and nothing different from this is being asserted in my own construction. So, no “odd notion of the ontology of personhood” is operative here.

The statement Laing quotes from my first piece in WTJ still seems correct to me, so it is worth quoting here. On that occasion, I was responding to comments previously made by Laing about my model in Providence and Prayer, where I talked about how God could have brought about counterfactuals different from those that pertain in this world. Then, as now, Laing was arguing that God’s knowledge of counterfactuals, in a Calvinist framework, should be viewed as part of his free knowledge, not, as I now believe (following the main line of Reformed tradition), as part of his natural knowledge.

As a counterfactual, this is not dependent upon God’s will; it is the way things would be if God chose to actualize the world in which Laing, being exactly who he is now, got into that particular situation. If God had reasons for wanting things to turn out differently, he would not allow the particular circumstances to obtain, or he would have brought into Laing’s personal formation influences that would make him a person who would choose differently than this particular John Laing would do” (WTJ 69 [2007]: 353).

This remains my perspective, and I hope I am able to show why nothing odd, nor the least bit unusual, is being asserted about the “person who would choose differently” in world A than in world B.

Laing wonders if “the Ware/Tiessen model” might be following Alvin Plantinga’s “postulation of world-indexed properties that are essential to objects and beings (he uses the example of being snubnosed in a for Socrates)” (p. 12, citing The Nature of Necessity, 60-61). Might I be suggesting that Laing has the “property of eating chocolate cheesecake when in S in a”? Though I have not read this work by Plantinga, I can say with reasonable confidence that this concept of “world-indexed essential properties” is not where I am headed. As I understand the concept from the description Laing provides, and from his example, I conclude that it works only within synergism, which is where Plantinga lands. Although Reformed by affiliation, he has, like many Reformed philosophers, wandered into synergism, and he has thus been able to make good use of concepts enunciated by Molina, who proposed middle knowledge in order to explain how it is that God knows counterfactuals concerning free creatures, when such creatures are libertarianly free.

To speak of Laing as having the “property of eating chocolate cheesecake when in S in a,” in the framework offered by Plantinga’s synergistic understanding of possible worlds, is to assume that the critical factor bringing it about that Laing eats cheesecake in S, is Laing’s decision to do so. I am assuming that Plantinga is trying to account for what grounds God’s knowledge that Laing would eat cheesecake in S in world A, but not in world B, even though there is no difference between those two worlds in regard to either Laing or the situation. One possibility might be, I hear him suggesting, that what accounts for the difference in Laing’s choices, where neither Laing nor the circumstances are different, lies in the world in which Laing makes his choice. Thus in world A, Laing would have the property of eating cheesecake in S, but in world B, he would have the property of not eating cheesecake in S. The only reason I can see for making such a suggestion would be if Laing has the power of contrary choice. Since I deny that, I have no use for the concept of world-indexed properties.

As I elucidated in my first post in this series, the principles of causation which God knows naturally, and which enable to him to know what Laing would do in S, are true in all possible worlds. God only needs to know thoroughly both Laing and the situation S, to know what Laing would choose to do. The difference between worlds would not derive from properties Laing possesses by virtue of the worlds themselves, in any way additional to the differences in the complex personhood of the one making the choices and the complex of circumstances.

In the proposal described by Plantinga, certain essential properties of individuals are “only necessary insofar as they are world-indexed, but the property is not necessary in itself, apart from its world-indexed identification.” Obviously, that is quite different from my own proposal. In that scenario, one would have to know what Laing decides in different worlds where both Laing and the circumstances are identical. Despite that sameness, Laing decides differently in world A than he does in world B, and he can do so because of his libertarian freedom. The world-indexed proposal is a form of Molinism and is consequently vulnerable to the grounding objection. My own proposal, as I demonstrated in the first post in this series, does not share that problem, because Laing is not libertarianly free. In my model, what Laing would do in each possible world can be known by God because of his knowledge of the principles of agent causation which enable God to know how a creature like Laing would act in situation S, precisely because Laing could not do otherwise, though he does it voluntarily.

Laing observes correctly that “the placement of counterfactuals in God’s natural knowledge along with the denial of libertarian freedom suggests that I have the essential property of eating chocolate cheesecake in S, no matter which possible world we consider” (p. 13). It seems to Laing, therefore, that what Helm and Ware and I are suggesting “is that all beings’ properties are not only not-accidental (or essential), but that they are equal to those beings’ essences” (p. 13). In that conclusion, however, Laing makes a critical error, I think. He speaks as though there were something personal to an individual like Laing about the necessity that he eats or does not eat cheesecake in S. But there is nothing personal about it in my construct. When it comes to understanding God’s knowledge of counterfactuals, as an individual person, “Laing” is irrelevant within a soft-compatibilist framework. What matters is the kind of person Laing is in each of the possible worlds, and the circumstances in each of those worlds in which that kind of person finds himself. There is nothing “worldbound” about this. The principles of causation which are at work, and which inform God’s knowledge of the counterfactuals of Laing’s action in different worlds, are true in all worlds, they are not distinctive features of the particular world in which Laing might possibly live. I suggest, therefore, that there is no issue of identity or ontology at stake here. Synergists find the soft-compatibilist model odd, but its perceived oddness lies in the suggestion that if Laing were a particular kind of person, and if he were in particular circumstances, he would act in a particular way, and yet he would act with morally responsible freedom. It is compatibilism that generates a sense of oddness for synergists. But it is wrong to identify that perceived oddness as an oddness regarding the concept of personhood or ontology that is at work in the soft-compatibilist, soft-determinist, model.

Plantinga considers the Theory of Worldbound individuals to be “outrageous“ (Nature of Necessity, 102) and consequently self-evidently and obviously false. “If we know anything at all about modality, we know that some of Socrates’ properties are accidental to him, that Socrates is foolish is not necessarily false, and that Socrates exists does not entail every proposition” (Nature of Necessity, 102). I think that Plantinga is right about this, because what makes it possible for God to know counterfactuals about the future acts of free agents, is not about essential properties of the individual, and Laing can safely disassociate my understanding from the problems which are entailed in the concept of both worldbound individuals and world-indexed essential properties. Just as John Laing changes over time, without ceasing to be John Laing, there could be worlds in which Laing would eat cheesecake in S and worlds in which he would not. A difference as slight as his having eaten a piece of cheesecake which made him sick in one world, prior to his getting into situation S, but not having had any bad experiences with cheesecake in other worlds, could account for his eating in S in world A but not eating in S in world B. To suggest that this difference in his tastes constitutes him ontologically as a different person is both counter-intuitive and completely unhelpful.

What particularly concerns Laing about the theory of worldbound individuals is that God would have had only “one feasible world, that God had to create exactly what He did create, and that sounds awfully close to fatalism” (p. 14-15). That is an important concern, and it is, in large measure, why I reject the concept of there being a best possible world. It seems to me that, if there were such a thing, God would have been obligated to choose it, so that he did not libertarianly free choose this one. (I have spelled out my position in this regard in a blog post entitled “Must God maximize his own glory?” so I will not repeat it here).

On pp. 15-17, Laing enters into an interesting discussion which is relevant to his dialogue with Bruce Ware, but I need not go into it because of the position I have just enunciated. What I do need to assert is that Laing’s concluding comparison between hypothetical knowledge Calvinism (where God knows counterfactuals necessarily) and Thomism (where God knows counterfactuals as part of his free knowledge) does not apply to my own construct. Laing posits that his analysis of the scenarios he had discussed

shows that the compatibilist view of freedom and its coherence argument against libertarian freedom, coupled with the claim that counterfactuals of creaturely freedom are part of God’s natural knowledge and therefore, both necessary and independent of his will, results in the Theory of Worldbound Individuals. The result is that the “John” in [world A] is also different from the “John” in [world B], the result being that we really know what John in B would do, at least not from our knowledge of the relevant counterfactual in [world A] (p. 17).

I hope I have explained adequately why my model entails neither world-indexed essential properties nor worldbound individuals. What pertains in every possible world is the principles of agent causation, which God knows necessarily, not essential properties of either the persons who would make decisions in those worlds, nor  essential properties of the circumstances in which those decisions would be made. There is no reason to believe that the complex of conditions pertaining in situation S in this particular world, did not exist in any other possible world. Similarly, there is no reason to believe that a person just like Laing in this world did not exist in any other possible world, nor that such a person might have been called “John” by parents who were “Laings,” in any other world. What we do know is that in any world where a person exactly like the John Laing in this world existed, if such a person was in exactly the situation S which exists at a moment in time in this world, that person would do exactly what John Laing does in this world. God did not have to choose this world, or other worlds which are only possible, in minute detail, to known what a person of that type would do in S. God only needed to know the principles of agent causation which operate in every possible world, which the omniscient God knows naturally or necessarily.

Laing thinks that the Thomist is in a better position than the hypothetical knowledge Calvinist because what makes the counterfactuals true is God’s having determined that they would be. In that scenario, the many possible worlds are all what they are because they are worlds God could have made. God could therefore have created a world in which a person exactly like John Laing in this world, when in circumstances exactly like the situation S in this world, eats cheesecake, but another world in which he does not eat cheesecake. The difference would lie in the libertarianly free will of John Laing, but it would only actually occur because God freely chose the particular world in which Laing freely chose either to eat or not to eat cheesecake. Like hypothetical knowledge Calvinism that is a form of monergism, but it is hard-compatibilism, holding together meticulous divine control and libertarian creaturely freedom. I do not share with Thomists the assumptions necessary for this compatibilism to be coherent (absolute divine timelessness, and divine action that is concurrent with creaturely action), though I think it would be splendid if the proposal were coherent. By contrast, Molinism affirms human libertarian freedom but denies meticulous divine control (hence the middleness of God’s knowledge of counterfactuals). In Molinism, the determinative factor in most decisions made in this world is the will of creatures who could have decided otherwise than they do, rather than the will of God. Unfortunately, the grounding objection makes this proposal incoherent.

Particularly pertinent to this section of his paper, Laing thinks the Thomist position with regard to counterfactuals is superior to that of the hypothetical knowledge Calvinist because the Thomist “may maintain that the ‘John’ spoken of in each possible world is the same person and does not have to resort to claims that the scenarios are different or the definitions equivocal” (p. 17).  I can understand why Laing thinks that the hard-compatibilist model of the Thomist makes the persons less different between worlds than the  soft-compatibilism of hypothetical knowledge Calvinism does. For the Thomist, exactly the same person could choose differently in exactly the same circumstances, so that the difference in that choice need not be explained by different inclinations, habits, experiences etc. in the person’s life. But I suggest that the difference between the soft-compatibilistically free person who chooses to eat cheesecake in S in one world and the one who chooses not to eat cheesecake in S in another world is not significant. As I said earlier, what is at work are the factors which lead people to say very frequently that they or someone else is a “different person” from what they were 10 years ago. I hope to be more like Christ 5 years from now than I am today, but I will still be “me” when I am more (or, God forbid, less) sanctified.

In short, hypothetical knowledge Calvinism’s notion of personhood or ontology is not odd. But it is coherent in ways that neither Molinism nor Thomism is. I would love to be able to assert that humans are libertarianly free. I understand how intuitively people in our culture assume that authentic freedom is libertarian. I think that Thomism’s hard-compatibilism has a more biblical understanding of the minuteness of God’s providential control than the incompatibilism of Molinism does, but I do not hold the tenets which are essential for Thomism’s compatibilism. Molinism would be an easier sell than hypothetical knowledge Calvinism precisely because it affirms libertarian freedom, but the grounding objection prevents me from affirming it as the best answer to the deficiencies of Thomism. I feel keenly the reasons why synergists object to compatibilism, but my reading of Scripture convinces me of the truth of soft-determinist monergism, and I think that hypothetical knowledge Calvinism offers the best explanation for what makes divine sovereignty and human responsibility compatible.