Introducing Calvinism and Middle Knowledge: A Conversation

I am happy to report that a new book is hot off the press, Calvinism and Middle Knowledge: A Conversation, to which I contributed two chapters and half of a third one, which was co-authored with Paul Helm. I’ll give a brief introduction to the book, and then I will trace the history of my own theological journey in relationship to middle knowledge, and finally I’ll briefly sum up my current beliefs about God’s knowledge of counterfactuals of creaturely freedom. This will provide context for understanding my contributions to this book.

The concept of “middle knowledge” may seem rather esoteric, perhaps reminiscent of Tolkien’s “Middle Earth,” but this is a very important subject, so I’ll be happy if this book generates more interest in it. Along with other authors in the book, I want to make Christians more aware of these concepts, and to lead them to examine seriously their own understanding of the nature and content of God’s knowledge. This is extremely important to our beliefs about God’s way of working in the world, and these beliefs inevitably have far-reaching effects within our daily lives and our ministries, as God’s co-workers in the world.

The contents of this book

Following a Preface and an Introduction, the book is divided into four parts, as follows:

Part I. Molinism, Evolution, and Intelligent Design

1.   Intelligent Design, Middle Knowledge, and the Problem of Creaturely Flaws (John D. Laing)

2.   The Impossibility of Evolution Apart from a God with Middle Knowledge (Kirk R. MacGregor)

3.   The Evolution of Molinism (Greg Welty)

Part II. Calvinist Concerns with Molinism

4.   Molinist Gunslingers: God and the Authorship of Sin (Greg Welty)

5.   Molinist Gunslingers Redux: A Friendly Response to Greg Welty (Kenneth D. Keathley)

6.   Molinist Gun Control: A Flawed Proposal? A Reply to Ken Keathley’s “Friendly Response” (Greg Welty)

Part III. Calvinist Appropriation of Middle Knowledge

7.   Does Calvinism Have Room for Middle Knowledge? (Paul Helm and Terrance L. Tiessen)

8.   Middle Knowledge Calvinism (Bruce A. Ware)

9.   Middle Knowledge and the Assumption of Libertarian Freedom: A Response to Ware (John D. Laing)

10. A Response to John Laing’s Criticisms of Hypothetical Knowledge Calvinism (Terrance L. Tiessen)

Part IV. Calvinism and Molinism: The Ongoing Conversation

11. On Parsing the Knowledge and Will of God, or Calvinism and Middle Knowledge in Conversation (John D. Laing)

12. A Calvinist Perspective in the Conversation about Middle Knowledge (Terrance L. Tiessen)

13. “Lord Willing and God Forbid”—Divine Permision, Asymmetry, and Counterfactuals (Guillaume Bignon)

14. Calvinism vs. Molinism (Paul Helm and William Lane Craig)

The book concludes with a proposal by the editors concerning “Prospects for Further Discussion.”

I’d like to interact with chapters of this book in coming months because, though I read a few of them previously (when they were in the form of academic papers), and I have interacted with the authors of some of them, I expect to benefit from reading (or rereading) all of the contributions from my current perspective, which has gradually been modified in the past couple of decades of thinking and writing about this topic. Here, as on other theological topics, I join Anselm in the life-long practice of faith that is seeking understanding.

The Purpose of this Book and the Story of Its Coming to Be

The Preface explains that this is “an anthology of essays and dialogues presented in or directly related to the 2012-14 Molinism/Middle Knowledge Consultation of the Evangelical Theological Society,” which had three major goals:  

1) to move the discussion of Molinism/middle knowledge out of the philosophical arena and into the broader theological community,

2) to “spark an ongoing conversation between Calvinists and Molinists about Calvinist concerns with middle knowledge and Calvinist appropriation of middle knowledge without embracing libertarian human freedom,” and

3) “to see what explanatory role middle knowledge may or may not play in Calvinist and Molinist accounts of providence and practical theology” (p. ix).

I had been invited to participate in the Consultation but, by then I was retired and no longer had funding from a school to attend academic conferences, so I regretfully declined. Nonetheless, I was able to interact with some of the papers presented, and participants in the consultation were aware of things I had written.

The theological proposal known as Molinism

I assume that many readers of my blog are much more familiar with Calvinism than with Molinism. So, I’ll briefly outline the distinctive features of this school of thought, but readers with an interest in this subject would benefit greatly from reading the “Introduction” to this book (pp. xi-xxxi), which was written jointly by the editors. It offers a fine outline of the history of the work of Jesuit theologian, Luis de Molina (1535-1600), which triggered, within the Roman Catholic Church, a serious controversy over efficacious grace.

The theological difference was greatly complicated by a power struggle within the church, between the Dominicans and the Jesuits. On the theological level, the two groups charged each other with heresy. The Dominicans dubbed Molina’s work “Pelagianism,” and the Molinists charged the Dominicans with “Calvinism.” The Popes feared that the dispute within the Roman church would work to the benefit of the Protestant movement, so both sides were ultimately prohibited from making heresy charges, and the church never did decide which group was theologically correct. Consequently, within Catholicism, Thomism and Molinism both survive.

The “Introduction” states succinctly Molina’s proposal that God has “middle knowledge,” but also describes his theology of social justice, which addressed particularly the issues of private property and the African slave trade. That is an aspect of Molina’s work about which I was largely ignorant, having focused on his major contribution to the discussion of God’s knowledge, especially as it relates to the respective roles of God and human beings in God’s governance of the world and in salvation. Molina’s middle knowledge proposal was an “attempt to reconcile God’s specific providence with human freedom” (p. xi).

Molina’s lengthy work, Concordia, was originally intended to be a commentary on Dominican theologian Thomas Aquinas’s (1225-1274) massive Summa Theologiae, but it did not get beyond discussing the relationship between God’s grace and human freedom in salvation. To those of us who are Protestant, this immediately brings to mind the never-ending dispute between Calvinists and Arminians regarding this topic.  John Calvin’s (1509-1564) theology aligned fundamentally with that of Aquinas, in that both of them followed Augustine’s emphasis on God’s sovereign control. Dutch Reformed theologian, Jacob Arminius (1560-1609), was a contemporary of Molina and, within different church contexts, both of them taught that the decisive factor which made God’s grace effective in salvation was the decision of humans.

Although there is this very basic similarity between the Thomist/Molinist and the Calvinist/Arminian disagreements, there is a very significant difference between Aquinas and Calvin with regard to the nature of human freedom. They both agreed that God’s eternal decree determines everything in created history, but Aquinas believed that humans are libertarianly free (having the power of “contrary choice” or of “alternative possibilities”), whereas Calvin asserted that humans are morally accountably free so long as they act voluntarily, without coercion. Molina and Arminius both concurred with Aquinas that God created humans libertarianly free. Thus, both Aquinas and Calvin were “compatibilists,” which is to say that they believed that God’s determination of all things is compatible with human freedom (though they differ concerning the nature of that freedom). This makes Aquinas a hard compatibilist and Calvin a soft compatibilist. Molina and Arminius were “incompatibilists,” in that they both believed that God’s comprehensive determination of all events is incompatible with authentic human freedom, which is libertarian in nature. It is significant, however, that Molinism gives God more control than Arminianism does. I see the divide between incompatibilism and compatibilism as the great watershed between Christian theological models or “schools.” (In this regard, the comparative chart of the 11 models of providence which I described in Providence and Prayer may be helpful.)

 Before I move on to my own theological story, we need to get a handle on the essence of Molina’s theological proposal, which has been very helpful to me. Aquinas, and the Christian tradition in general, had posited two logical (though not chronological) moments in God’s knowledge. God has natural or necessary knowledge, pertaining to all metaphysically necessary truths. This knowledge is independent of God’s will, or prevolitional. God also has free knowledge, which is the knowledge of his own will. “Thus, free knowledge is postvolitional. By God’s natural knowledge, God knows what could (and couldn’t) be the case, and by free knowledge, he knows what will (and won’t) be the case” (p. xx). In between those two logical moments, therefore, stands God’s decree, or eternal purpose, the logical moment in which he decides to create a world and decides its history, including both the decisions and actions of creatures and God’s own actions, in bringing his decree into effect.

Between these two classically agreed upon moments in God’s knowledge, Molina posited a third class of knowledge that informs God’s decision about what sort of world to create, and this includes truths that are contingent and independent of God’s will, which he called scientia media (middle knowledge). He suggested that propositions about what all possible free creatures would do in all possible situations are independent of God’s will because the nature of free will requires no outside determinants, but contingent because the creatures can act or not-act if truly (i.e., libertarianly) free. It has become customary to refer to these propositions as counterfactuals of creaturely freedom (even though they are not all counter-to-fact, if they are actualized). For any given creature in any particular set of circumstances wherein he is faced with a choice to act, there is a true proposition about what s/he will or will not do (of the form “If person P were in circumstances C, P would freely perform action A”; of course, there is a corresponding false proposition of the form “If P were in C, P would not freely perform A”). Thus, Molinism is the belief that God’s creative decisions are informed by his knowledge of all possibilities via natural knowledge and of how all possible free creatures would act in all possible situations via middle knowledge. Since the truths known by both of these types of knowledge are independent of God’s will, they limit the kinds of worlds (i.e., complete descriptions of the ways things could be) he can create/actualize. His act of will in creating then grants him foreknowledge via free knowledge, and also affords him extraordinary providential control without violating creaturely freedom. It does so by allowing him to actualize a world where the counterfactuals of freedom, along with other relevant factors, result in the outcome(s) he desires. (Calvinism and Middle Knowledge: A Conversation,  xx-xxi).      

Now that you know the concept which is essential to Molinism, namely, “middle knowledge,” I’m ready to tell my theological story in regard to this idea.

My theological journey to “hypothetical knowledge Calvinism”

My conversion to Calvinism

After finishing high school in India, where my parents were missionaries, I went to London College of Bible and Missions, in London, Ontario, and enrolled in a BTh degree program. I was just 16 years old at that time, so my theological formation had scarcely begun. During my first year at LCBM, I took a course on “Personal Evangelism,” a subject required of all of us. The teacher (Donald Oakley) was a Calvinist, and he described for us how God saves people, and how we can participate in his saving work. As I recall, his presentation was met with virtually universal protest. We freshmen students may not have known much, but we knew that human beings are free, and that it is they who decide whether or not to believe the gospel and be saved by God’s wonderful work in Christ. The suggestion that God had sovereignly decided before he created the world who would believe and be saved, and who would not, seemed horrible to me. It was not a totally new concept to me. It had come up during my high school years in a Christian school, but I doubt that I had ever before heard it laid out so clearly, with biblical support. (Arminians or Molinists reading my story will doubtless share my sentiments at that time.)   

I finished my BTh, went on to do a BA and then to Wheaton graduate school, for an MA in New Testament studies. Near the end of the program I had to do comprehensive exams. In preparation for the theology exam, I asked the prof (Charles Horne) if he could give me a short list of books which would be particularly helpful to me. Among the books on his list was John Murray’s Redemption Accomplished and Applied. It is a short book, in which Murray essentially walks through the list of salvific events given by Paul in Romans 8:28-30, unpacking them as the logical and chronological order of salvation. A stunning thing then happened to me. I started the book as a somewhat eclectic Calminian, believing that God is sovereignly determining in some areas, but not in salvation. By the time I was done with Murray’s book, however, I had become a convinced Calvinist, rejoicing in the wonderful “doctrines of grace.” Eureka!

In the next few years, I did a lot of reading in Puritan theology (three cheers for Banner of Truth publications and Puritan and Reformed [now P&R] Publishing ) and then, two years later, I (with my wife and first son) headed for the Philippines on a month long trip by freighter. On that trip, I read through John Calvin’s Institutes of the Christian Religion. I was richly blessed by Calvin’s writing, and I still believe that he was essentially correct theologically, including some points at which the tradition later departed from his position (such as the intent of the atonement, but that is a different story).

The writing of Providence and Prayer

Many years later, I was on the faculty of Providence Theological Seminary, in Manitoba, and I was due for a sabbatical, provided I could convince the administration that I would make good use of it. The prospect of a year spent reading and writing was delightful, but about what should I read and write? I had then been teaching theology for years, and I was regularly frustrated by meeting students whose theological beliefs were self-contradictory. Commonly, when it came to matters of God’s general providence, they were compatibilists, believing that God was meticulously in control. But when it came to salvation, they were sure that we had to be libertarianly free and be the ones who determined that we would be saved by receiving God’s gracious offer of salvation. (I had once been right where they were!)

I wondered what I might be able to do on sabbatical which could be helpful to students who did not have a coherent or self-consistent theological framework. I had often heard them pray in a way which did not cohere with their theological affirmations, especially when I knew that they were Arminian in their theology. They would ask God to do, or thank him for, things which were outside of the power he had retained, if he had given creatures libertarian freedom. So, I decided to study and write about the various models of God’s providence, and the way in which prayer functions within each of those models. I did my work at Wycliffe Hall, in Oxford, as a visiting scholar, on my sabbatical year, in 1995-96. What a marvelous time that was. In the course of the year, I read widely in the literature on divine providence, and I identified 10 models of providence, each of which should produce distinctive forms of petitionary prayer. The fruit of that work was Providence and Prayer, published by IVP in 2000.

In my book, I laid out the ten models I had encountered, but then I described my own model as it had evolved in the course of the year, which I then dubbed “middle knowledge Calvinism.” In doing my study of providence, I was eager to refine my own model. I was hoping that I might be able to construct a biblically supported model in which God has meticulous sovereignty (as per Calvinism), but in which moral creatures (human and angelic) have libertarian freedom. Some years before, I had told Clark Pinnock that, when I read the book he edited, The Grace of God, the Will of Man: A Case for Arminianism, I had wanted to be converted, but it had not worked.

We live in a society where people intuitively assume that authentic freedom is libertarian, the power of alternative possibilities (PAP). This makes Calvinism a very hard sell. When I studied Thomas Aquinas’ doctrine of providence, I was greatly intrigued by the way he put together meticulous divine sovereignty and libertarian freedom. Might that work? I concluded that it would not. Essential to Thomas’s construct was a belief in God’s absolute timelessness, and the proposal that God’s own action is, therefore, not prior to the creatures’ action but simultaneous with it. The model is certainly intriguing, but I did not believe in the features which were essential to it.

Prior to that sabbatical, I had been only vaguely familiar with the work of Molina, and it first caught my interest through reading the work of William Lane Craig, who is a prominent evangelical Molinist. I read other Molinists (including Molina himself, though not extensively), and I could understand why Thomas Flint wrote:

It seems to me that the prima facie case made here in favor of the Christian’s endorsing both the traditional notion of providence and the libertarian account of freedom—that is, for embracing what I shall call libertarian traditionalism—is a potent one. Absent insurmountable problems which its acceptance might engender, libertarian traditionalism seems, if not the only, then at least by far the best game in town” (Divine Providence: The Molinist Account , 34).

If Molinism were coherent, I would agree with Flint. Molinism would then provide the strongest account of divine sovereignty compatible with the most widely plausible account of libertarian freedom. It improves on classic Arminianism, so I’m not surprised at the number of Arminians who incorporate Molinism into their own model. I’m aware, however, that some Arminians (such as Roger Olson) protest that hybridization, because they consider Molinism’s account of God’s control to be as problematic as Calvinism’s.

Given my appreciation for Flint’s enthusiasm about the Molinist construct, why am I not a Molinist? That is a path which is solidly blocked to me by the grounding objection. I do not think that knowledge of counterfactuals of libertarian freedom can be known, even by God. God naturally knows all necessarily true propositions, which include all truths about the actual future, whether these are made true by divine determination (as Thomists and Calvinists believe) or by creaturely decision (as Molinists and Arminians believe). But, in the classic Arminian model, the truths God knows about the actual future are made true by creaturely decisions and actions. By contrast, in the case of future counterfactuals, if creatures have the power of contrary choice, it is impossible to predict with certainty what any hypothetical moral creature would do in hypothetical circumstances. From the Open Theist perspective, Greg Boyd has been helpful in drawing attention to God’s knowledge of “might counterfactuals,” but this gives God only knowledge of probabilities. Considering the huge number of creaturely decisions made at every moment in time, even if God predicted what a creature would probably do with extremely high accuracy, the cumulative effect of all the improbable actions would make “might counterfactuals” minimally useful to God, preventing him from having the degree of control that the metanarrative of Scripture presents to us.

So, I reached the end of my year-long study of providence and prayer, having failed to find or construct a viable model of strong divine sovereignty which incorporated libertarian creaturely freedom. But, I had been impressed with Molinism’s emphasis on the usefulness to God of knowing counterfactuals. This led me to formulate a model in which God knew counterfactuals of soft determinist freedom, the liberty of spontaneity. When God knows those counterfactuals, whether naturally or at a middle moment did not matter to me. In either case, God had that knowledge logically prior to his decree, and he made excellent use of that knowledge in making his decree. I called that model “middle knowledge Calvinism.”

The realization that, if creatures do not have libertarian freedom, God’s knowledge of counterfactuals is not “middle” but natural

It would be nice if I could report that the publication of my model of God’s providence received widespread approval within the Calvinist community, but that did not happen, although I have had very encouraging response to the book as a whole. At the ETS meeting after Providence and Prayer came out, Bruce Ware approached me and told me that he had reached the same conclusion I had, and he had been teaching it in his classes at Southern Baptist Seminary, but he had not published anything about it yet. It was nice to meet a traveler taking the same road. Some of my students found the model helpful, and I am hopeful that Ware’s perspective has been widely approved among his students. I am glad that he has had good opportunities to make his case for the model in publications (e.g., his case for a “modified Calvinist doctrine of God, in   Perspectives on the Doctrine of God), since we spoke at ETS. Sadly, however, I have seen very little in print which indicates that this modification of the Calvinist model of providence has gained favor. Naturally, I am pleased that Ware has contributed to the conversation in this volume, with chapter 8, “Middle-Knowledge Calvinism.” On this particular subject, when I read his work, I always feel that he and I are kindred spirits, working to enhance the formulation of Calvinist theology in our time in a very beneficial way. May our tribe increase!  

I spent a few weeks at Tyndale House in Cambridge, in the Spring of 2005, and I spent my time reading Reformed theologians, to discern their concerns about Molinism. I took heart from a statement by noted Southern Presbyterian theologian Robert Dabney (1820-98) that: the “history of the controversy on scientia media” presents an instance of the rule “that usually mischievous errors have in them a certain modicum of valuable truth. Without this, they would not have strength in them to run, and do mischief” (Systematic Theology, 159). It is that “modicum of valuable truth” that I am trying to extract from the Molinist proposal, correcting it with soft determinist rather than libertarian freedom.

In November, 2005, at the annual meeting of ETS in Valley Forge, PA, using the fruit of my time in Cambridge, I responded to Reformed concerns about middle knowledge, in a paper entitled: “Is God’s Knowledge of Counterfactuals Necessary, Middle or Free?: A Calvinistic Proposal.” I was happy to see Bruce Ware and some of his students at that session. I then revised the paper slightly and submitted it to the Westminster Theological Journal, in hopes of reaching a Calvinist audience. The article was accepted, and it appeared in WTJ (69 (2007):217-28), entitled “Why Calvinists Should Believe in Divine Middle Knowledge, Although They Reject Molinism.”

Some time later, Paul Helm came upon my article and he wrote me with his comments. We had a good conversation, back and forth a few times, and eventually I became convinced that he was correct: God’s knowledge of counterfactuals only needs to be “middle,” if creatures have libertarian freedom. At Helm’s suggestion, we submitted our dialogue, as an article, to WTJ, and it appeared under the title: “Does Calvinism Have Room for Middle Knowledge? A Conversation” (WTJ 71 (2009: 437-54). It was in that article that I suggested that a better name for my model would be “hypothetical knowledge Calvinism,” and that is the name I continue to use. I am happy that our article is reprinted in this volume (chapter 7).

A brief statement of the features of my Hypothetical Knowledge Calvinism

As the name suggests, my model affirms the main features of the Calvinist Model. I believe that God is comprehensively in control in the world, accomplishing the purposes that he determined in eternity, in his decree or eternal purpose. Because God’s will is always accomplished, it is evident to me that his moral creatures (both human and angelic) do not have libertarian freedom. God did not voluntarily limit his ability to achieve all of his desires for creation, by giving moral creatures the power to determine much of what occurs.

My model is a soft compatibilist model (unlike Aquinas’s hard compatibilist model): it affirms both God’s meticulous providence and human freedom which is of a spontaneous or voluntary kind. God freely predetermined every detail of the history of creation, but a very large part of that history is brought about by the decisions and actions of his moral creatures, who are not coerced by God in the making of their decisions, even though their decisions are determined by their character. (John Murray aptly calls this “the dispositional complex,” and he notes that it is essentially what Scripture calls ‘the heart’ [Prov 4:23; Mt 12:34, 35; Mk 7:21, 22] Collected Writings of John Murray. Vol. Two: Select Lectures in Systematic Theology, p. 61).   

Many theologians in the Calvinist tradition have deemed God to be absolutely timeless, but I doubt this because I am concerned that such a concept does not do justice to God=s highly relational personal being. God was probably timeless before he created the space-time universe, but when he did that, he maintained his transcendence (so that he is not limited by time) as his creatures are, but he also chose to be immanent within created time and space. God is both omnipresent and omnitemporal. He is present to every inch of space and every minute of time. Since God is not absolutely timeless, in his immanence, he therefore experiences events in sequence, not all at once. Being omniscient, God knows all propositional truths, including time related truths. He remembers the past, he experiences the present, and he anticipates the future.

In a significant sense, God is not only determining human history, he is responding to his creatures within it. This divine responsiveness is facilitated by God=s possessing knowledge of how creatures would act in all hypothetically possible situations. He knows the actual future, because he determined it in the logical moment of his decree. But, in order to do this, without turning creatures into mere pawns which he moves around the checker board of creation, God needed to know how creatures would respond in hypothetical situations, including their response to his own persuasions or actions. God can know this because his creatures are not libertarianly free, and he must know this in order to plan how he will act to bring about his purposes. As part of God’s natural or necessary knowledge, he knows what I call the “principles of agent causation.” He knows exactly what a particular sort of person (with a particular genetic makeup, with a particular temperament, who has had particular experiences, and who has particular motives, desires, and goals etc.) would certainly do in a particular set of circumstances. (Recall Murray’s language of the “dispositional complex.”)

With simple foreknowledge, such as Arminianism attributes to God, he would know the actual future, but he would be unable to do anything about it. (In approximately the words of Open Theist philosopher, William Hasker, “by the time God knows the future, it is too late for him to do anything about it.”) With the knowledge of counterfactuals, however, God is able to plan and then to accomplish his plan without violating the responsible freedom that he has given to his creatures.

Hypothetical knowledge Calvinism, like middle knowledge Molinism, finds the concept of “possible worlds” especially helpful. This takes our attention away from incidents in isolation from one another, and it enables us to look at God’s decree in terms of his knowing all possible worlds and choosing which of those possible worlds to actualize. Thus, before creating anything, God did not simply elect individuals to salvation, in an isolated manner. Rather, he chose the world, the complete complex of creatures and events, in which you voluntarily respond to God’s revelation of himself to you, with a faith which pleases him and is instrumental in your justification. I am not quite sure whether the concept of a “best possible world” is coherent or not. But I know for sure that the world God chose to actualize, as a whole complex of creatures and events, is a good world, with a history which, in the end, will be seen to have been for God’s glory and the good of his creatures. But that history will largely have come about through the free decisions (in the sense of being voluntary, without external coercion) of angels and humans, who will give account in the day of judgment for all of their actions, admitting their guilt where they sinned, and praising God for everything good which they did by his gracious enabling.

Traditional Calvinists always believed that God knew counterfactuals of creaturely freedom, so hypothetical knowledge Calvinism asserts nothing different in that regard. Where we differ is in our assessment of the role God’s knowledge of counterfactuals plays in his decree, and in his sovereign governance of creation. Here we happily acknowledge a debt to the insights of Molinism, while improving upon them by avoiding the grounding objection. Traditional Calvinists have often been concerned that to speak in these terms would compromise God’s independence from his creatures. But that is an unfounded fear. God remains independent of his creatures, and his freedom to choose to create and, if he does, what world history to choose, does not make him dependent upon creatures. Hypothetical knowledge Calvinism is helpful because it gives us better understanding of how it is that God was able to establish his eternal plan for the history of creation, in every detail, but to have that plan come into reality largely through the free acts of morally responsible creatures. It is in this context that we are able to best understand why God is not morally responsible for all the evil which occurs within the world history he sovereignly chose, but all good is of God and should elicit our praise of him, not of ourselves or other creatures. (I consider this perspective to be a useful means of unpacking somewhat the Westminster Confession of Faith’s, statement regarding God’s eternal decree [Chapter III, I and II].)