Thank you for agreeing to do this interview. Let’s start with some questions about your early life. You grew up in Crosby, North Dakota, a small town in a very rural area of the United States. I even see cowboy and ranch themes in some of your academic work and décor. Will you tell us a little about what your childhood was like?

Yes, I grew up in North Dakota, before moving to North Carolina for life as a teen. The memorable experiences in the Dakotas were summertime stays on grandparent ranches, where I developed my love of horses and for life outside of cities. That interest led us to country life in Texas, and to training and showing quarter horses in the 1980s and some of the 1990s. My love of sports also developed in my early years, in part because it isn’t hard to be one of the best athletes in your grade in a town of about 1600 people! It was a disappointing discovery by about the time I was twelve to realize that I wasn’t going to be a major league baseball player. By the time I was a teenager, I still played several sports, but had also developed musical interests, in large part because of my seventh-grade band teacher, Miss Sykes, who thought I had promise as a trombone player. She found a trombone teacher for me, Mrs. Nelson, and talked my parents into paying for private lessons, which continued throughout my teenage years and into college.

You went on from North Carolina to attend Evangel University in Springfield, MO, for your undergraduate degree. What influenced your decision to attend Evangel?

I went to Evangel on a music scholarship, where the chair of the music department had a degree in trombone performance. I grew up in the Assemblies of God denomination, and planned on becoming a minister, so I double majored in Biblical Studies and Music, with the idea of going to seminary after graduating. By the end of my sophomore year, I had become interested in philosophy by taking a philosophy of religion class and had grown dissatisfied with dabbling in academia. So I began disciplined study practices and stopped calculating how little I could get by with, and became a much better student.

From Evangel, you went straight into a Master’s degree in Philosophy at Mizzou, then a Ph.D. at Notre Dame. What led to your interest in Philosophy enough for you to pursue a Ph.D.?

As graduation neared from Evangel, one of my professors, Don Baldwin, asked me what I planned to do after I graduated. I said I was unsure, because personal reasons led me to want to stay in Springfield, while academic interests counselled against it. It was April of that year, too late to apply to graduate programs, but Don called some friends at the University of Missouri and somehow I was admitted and offered an assistantship. I went and was immediately assigned to teach two sections of Introduction to Philosophy, and I still feel bad for students who have to take such courses from first-year grad students! I don’t think the practice is common anymore, and I know that Mizzou quit that practice several decades ago.

In the process, my thoughts were not directed at degrees or a career to follow but instead simply on the rewarding experience of getting to pursue a discipline that I loved. I have always loved coming to understand positions and arguments in philosophy and seeing where the frailties of human cognition appear in the process. So I went to graduate school to get to pursue the academic study I found the most rewarding, and that continued when I went to Notre Dame as well. Once again, I was just doing what I loved doing, rather than aiming at a degree or career. Things changed, though, while dissertating in January of 1982, when it became clear I was going to graduate and needed a job for that fall. Uncertainty was short, however. I don’t know the details, but I’m pretty sure my visiting position at Notre Dame was the result of efforts by my dissertation director, Dick Foley, still one of my most cherished friends. It is among my most prized badges of honour to have been his very first Ph.D. student. He was and is a superb philosopher, a spectacular administrator, and a marvellous mentor.

Several of your monographs seem to engage in a reflection on, or response to, some of the doctrines one would learn at Evangel. Are there any particular evangelical doctrines that you would describe as influencing the direction of some of your scholarly work? If so, which ones in particular have influenced your work?

Well, with some chagrin, I acknowledge being a philosophical iconoclast. I balk at sacred images, and one can see this tendency in the variety of ways that I distance myself from the traditional understandings I grew up learning. Moreover, before developing a positive view, my writings tend to begin by finding fault with extant approaches. But I guess on this score I have become unrepentant, viewing our liability to error as something to be acknowledged and embraced, leaving open the possibility of thinking hard and carefully about philosophical matters while at the same time developing and displaying intellectual humility in the process. The theology I learned in my youth provides the targets for reflection, but mostly it works as a launching pad to develop and build something new rather than something to be restored by patient reconstruction. Probably the best example of this tendency is my work on hell, where the annihilationist direction I pursue is one that provoked blank stares in my own Christian communities. (There is a bit of irony here, though, since the view I develop has Lutheran roots, and it is the Lutheran tradition that my grandparents left in favour of the Assemblies of God, so one might have expected some familiarity with Lutheran positions on the afterlife.)

You’ve had a storied career. From Notre Dame, you accepted a position at Texas A&M, chaired the department at Missouri, then took a distinguished professorship at Baylor, then a full professorship at WashU St. Louis. During part of that time, you started and ran an epistemology blog that was very popular in the profession, Certain Doubts. You’ve been a co-PI on high-impact grants in which you’ve organized annual professional conferences and seminars, and you were the editor for Oxford Studies in Philosophy of Religion for 14 years. Although there are probably too many to mention, could you mention a few of your co-workers or colleagues who have made you a better philosopher over the course of your career and the effect they’ve had on your life or work?

First and foremost, I want to mention Hugh McCann, who was department head at Texas A&M when I was hired. Hugh’s writing benefited from his wife’s input – Janet is a first-rate poet – and Hugh rekindled my interest in literature on pragmatic grounds: if you want to write better, read great writers. So I did, and in the process recovered my love of literature that had been lost through the experience of high school English classes (e.g. by having to write a term paper on Melville’s use of the colour red in Moby Dick).

Hugh is a good example of the colleagues who have had the biggest impact on my life, since they always involve social times at meals and happy hours where conversations range from ordinary topics to figuring out difficult issues in philosophy. The regulars for such occasions were Hugh as well as Michael Hand at Texas A&M, Robert Johnson at Missouri, and Alex Pruss at Baylor. Many others deserve mention as well, and one of the costs of answering a question like this by identifying some is the risk of leaving out others that ought to be mentioned as well. For that I apologize to the unmentioned!

It’s a great idea to read outside of the philosophical discipline to become a better writer in the discipline. What are some examples of great authors or books you’ve read that have made you a better writer?

I began by reading classic novels (let’s stipulate that the end-point for these is Faulkner) that I’d always wanted to read but didn’t take the time to do so, in part because graduate school leaves little time for such. I then turned to contemporary authors, and there the task of finding great writers is harder, though among my favourites are Marilynne Robinson, Annie Proulx, Jane Smiley, Richard Russo, and Anthony Doerr. I read everything they write! (I’ve also read everything by Larry McMurtry, but that’s more just a Texas thing – it is reported that, at his bookstore in Archer City, he used to wear a shirt with the inscription ‘Third-rate Regional Author’, a humble and befitting sentiment, I think.) The one that astonishes me the most, though, is Cormac McCarthy, with a caveat. After All the Pretty Horses, he went Hollywood, and subsequent writing tended to look more like script writing rather than the writing in his earlier work. The earlier work, including Suttree, Outer Dark, The Orchard Keeper, and the magnificent Blood Meridian, has a lyrical quality to the writing that is somewhere between prose and poetry.

Hemingway, though, is who I try to emulate. It is simple, minimal prose, and I wish more of philosophy were written like that. Figure out how not to waste your audience’s time when telling them what you’ve figured out, and do that.

Moving on to your academic work, in some of your work – your work on faith and understanding, for example – you describe your methodology before you engage in theory. Your methodology seems important; it’s what you describe as an axiological approach, which focuses on what’s valuable, as opposed to an approach focused on linguistic analysis or metaphysical fundamentality. Why do you opt for the axiological approach?

I can’t remember when this happened, but in the process of developing my views about the nature and value of understanding, I was having lunch with Eleonore Stump, and she mentioned the medieval heritage in epistemology where the great intellectual goods are knowledge, understanding, and theoretical wisdom, with understanding being knowledge of causes and wisdom involving knowing what matters. The general idea resonates with my axiological approach, since one of my recurring complaints is the amount of time that can be wasted on matters of insignificance. One can see this in philosophy and also in the theological controversies that plague Protestant Christianity. Some have thought that the Gettier problem devolved into such, and I’ve had colleagues who thought that most all of current metaphysics could be ignored on such grounds. I’ll not mention my favourite targets for Christian disputes, since the examples don’t matter for the general point, which is that a presupposition underlying the topics we think and write about are things that matter, and if we keep this point clearly in focus, we can avoid mistakes that are quite common. I have articulated this point of view regarding knowledge and understanding, but it is perhaps easiest to appreciate concerning the topic of faith, which is a virtue both inside and outside religious contexts. Once the importance of faith is in full view, the idea that faith is a kind of belief should look wrongheaded from the very beginning. The same point can be made when it comes to the Gettier literature, since proposals for how to solve the Gettier problem did not always keep in mind the idea that knowledge is supposed to be more valuable than its proper subparts. The central idea of the approach is not that accounts of the nature of something important have to explain its importance but rather that if an account of a thing’s nature undermines or threatens its importance, something is amiss.

You are right as well that I contrast this approach with that of linguistic analysis and at least some ideas concerning metaphysical fundamentality. The details of this contrast can be found in my work, so I’ll limit my response here to the source of the approach I favour, which characterizes the highest of the great intellectual goods in terms of understanding what matters and what doesn’t. I owe the idea to an offhand comment from Eleonore, and I thank her for it.

It seems like you’re saying that one has to have the wisdom to discern which things matter in order not to waste time doing philosophy about things that don’t matter. And in some of your works (e.g. on understanding and faith), your approach seems to be a call to philosophers to refocus our efforts on what matters. If so, how does one develop that wisdom, if not by reading philosophical work in general, which might focus on issues that don’t matter?

Well, more cautiously, we should aim to have such wisdom and let that aim guide our choices about where to focus. Though I won’t name any, it is easy to find examples of work that suggest failure on this score. There are obviously no guarantees for how to achieve the goal, and the more convinced one is that we are all fallible, flawed, and full of foibles, the more seemingly opaque the path to take to get to the goal. Having grown up in a part of the world where winter tries repeatedly to kill you, I tend towards a quite pessimistic view about our capacities, but not about the possibility that we can try to do better.

Concerning methodology again, you also seem to take a somewhat pluralistic approach that’s tempered by acknowledgement of our cognitive limitations. For example, in Depicting Deity, you argue for an approach that incorporates many metatheologies, and in Rationality and Reflection, you argue that it’s epistemically permissible to reflect on our doxastic attitudes using different principles than we use at the non-reflective level. You advocate for epistemic humility in other work, e.g. Faith and Humility and Scepticism and Fallibilism. Your approach seems at odds with approaches that begin with a first principle and ‘follow the argument where it leads’, where one might stick to the argument’s conclusion in the face of objections. Why do you opt for your approach over this latter approach?

I think you’ve pinpointed one of the reasons, which involves epistemic permissivism, according to which there is typically more than one cognitive attitude that can be rational in the face of any given body of evidence. This position dovetails with the holistic character of justification that coherentists have emphasized and which is often overlooked in religious epistemology, especially. Here, the key point to note is that the story of rationality is pretty thoroughly that of defeasible reasons and reasoning, which keeps bodies of information from pointing monotonically towards a given target proposition. The position that fits best with these points is permissivist in nature, and the account I prefer of this permissivism hinges on the range of acceptable risk-taking when it comes to the cognitive task of attempting to get to the truth and avoid error. As William James noted, at one extreme are those for whom the possibility of error predominates, and at the other extreme are the epistemically cavalier, those for whom the slightest indicator of truth blossoms into full doxastic commitment. Perhaps positions at the extreme are hard to defend as rational, but even if they are indefensible, that leaves lots of room in the middle where rational weightings of the competing values of getting to the truth and avoiding error can be found.

It is also worth noting the way in which the methodology you describe can flout the Moorean wisdom of never giving up the more obvious for the less. When one starts with some first principle and just follows its implications undeterred, cognitive catastrophe can easily ensue, since one way of testing our ideas is to ferret out where they lead and be able to recognize what should lead us to reject them.

On your view, can it be rational for us to be, in a way, epistemically mercurial, in which we believe a proposition confidently, but whenever we reflect on our belief, we’re epistemically more risk averse and so don’t believe we’re justified in believing that proposition, but then when we jump back down to the object level and don’t reflect, our epistemic risk aversion switches, and we believe it confidently again? Is the rationality of this kind of epistemic vicissitude compatible with your view? This kind of mercuriality can seem like an epistemic conflict, and some religious adherents, for example, might be in this kind of epistemic state about whether they really are believers.

I think you’ve located a couple of really important issues in epistemology here. One of them concerns the degree to which the normativity to be characterized has cross-temporal dimensions. One view is that, once you reflect, there’s no going back to the pre-reflective level, since you can’t escape responsibility for conclusions you’ve come to on the basis of reflection, even if you forget them. I suspect such a view is too strong, though it may be that such information is still knowledge-undermining. The other extreme is a current time-slice view, according to which every new moment is one in which, to put it metaphorically, the sins of the past are always forgiven. In internalist language, it’s only information currently encoded in relevant mental states that makes a difference.

The other issue I notice in your question concerns the issue of how stable our intellectual personalities are. (There is also the issue of whether it’s OK to be more risk-averse about some topics and issues than about other topics and issues.) If we think of normativity as falling into various kinds, only one of which abstracts from everything except purely theoretical interests and purposes, I’m inclined to think that instability on this score implies a normative defect to be located in some domain other than that of the purely intellectual or theoretical.

So, take Hume as the poster child for the kind of case you are describing. He’s in his study, coming to radically sceptical conclusions, all the while being firmly convinced that once he leaves his study he’ll be as dogmatic as the rest of us. Ignore the fact that his sceptical conclusions might themselves not be supported by his total evidence and focus on his intellectual stance when outside his office. He doesn’t forget what he was thinking in his office; he just ignores it. Moreover, it is hard to see why we’d want to posit some purely intellectual personality change for the two contexts, where his ‘private, preponderant horror of being duped’ (to quote William James) was present in his office but disappeared after leaving it. Given all that, it is hard to avoid the conclusion that Hume’s ordinary attitudes are inadequate from a purely intellectual point of view.

Suppose though that we consider a contextual possibility of the following sort. Suppose Hume, in his office, really does have the horror of being duped, but he returns to normal when leaving it. It’s clear that there is some sort of foible or flaw in being such a person, but I doubt it is a purely intellectual flaw. After all, we find a (perhaps less exaggerated) form of such contextuality if we impose different standards of statistical significance for theoretical physics than we do for sociology. Hume’s descriptions, though, don’t seem to support such a contextual path for avoiding normative infelicity of a purely intellectual sort. That’s a much too brief response, and I’d love to see more work done on it. It strikes me as a really important set of issues.

Readers might be interested in some of the theological positions you’ve argued for. Let’s start with your work in Depicting Deity on metatheology. A metatheology is an approach one takes to figuring out what God is like. You examine three metatheologies: Creator Theology (CT), Perfect Being Theology (PDT), and Worthiness of Worship Theology (WWT). What’s the difference between these approaches, how do we figure out which one(s) to adopt, and which ones should we adopt?

These metatheologies arise from quite intuitive, but importantly different, ways people start when asked to say what God is like. Many start with being the Creator, others with being the most perfect possible being, and still others focus on being worthy of worship (in such a way that doing so is required of any being capable of worship). From these initial starts, one can articulate three different ways of characterizing what is fundamental to our idea of deity: one descriptive, one evaluative, and one normative or deontic.

The latter territories exhaust the logical space in question, but the three main options I present are, of course, only one possibility within each of the categories noted. That leaves room for other possibilities to be suggested, so CT (Creator Theology), PBT (Perfect Being Theology), and WWT (Worthiness of Worship Theology) should not be thought of as the only options available. Still, each of them has considerable traditional standing, and so investigating how each of them fares in comparison with the others is a nice way to begin thinking about the metatheological question concerning the fundamental nature of God.

The approach I recommend is to see what each of these starting points tells us about the nature of God, and then to consider the prospects of alternative approaches for explaining the worthwhile implications of alternative starting points. For example, how might a perfect being theologian try to show that the most perfect possible being must also be the sole creator of all else (the fundamental tenet of CT)? Alternatively, how might a creator theologian try to argue that God is singularly worthy of our most exalted worship?

I also argue for a presumption in favour of metatheologies that imply monotheism and divine personhood. The presumption is weak, however, allowing a methodology to explain away the considerations I cite in favour of these two attributes.

One can then move on to the attributes that separate classical theism from neo-classical theism: impassibility, eternality, simplicity, and immutability. One must be cautious in doing so, not assuming that failure to account for such attributes is a negative for a metatheology. Instead, the way I see things is that any metatheology should report its capacity to sustain items on this list and then present arguments for the conclusion that their approach gets things exactly right. Thus, the approach I favour treats the disputes between classical theism and neo-classical theism as metatheologically neutral, needing resolution in a way independent of the question of which fundamental metatheology to adopt.

The surprising result I argue for is that the favourite metatheology in recent philosophical theology – PBT – achieves the worst score among the three competitors I mention. The best results are achieved by combining aspects of CT and WWT, using the combination to explain why a being that is both ultimately worthy of worship and the sole creator of all else is also the most perfect possible being.

One by-product of this result is a new insight into why creator theologians are suspicious of the ontological argument. So one can think of the conclusions I draw as vindicating a Thomistic conception of God over against an Anselmian one.

You’ve also written about divine providence. You argue for a position according to which God both exercises providential control over the world and has infallible knowledge of what occurs – past, present, and future – but we also have libertarian free will – the ability to choose other than what we in fact choose. You reject common views of divine providence: simple foreknowledge, Molinism, and open theism. Will you explain your view, how it’s distinct from the other views, and what advantages it has?

I used to defend Molinism, having done so in the first book I published. But my work in philosophical logic led me to think that the logical space of options available was not properly understood, leaving lots more space between theological determinism and open theism, a space where one could argue for both complete foreknowledge and also libertarian freedom. The space in question involves adopting a deliberative model of God’s reasoning prior to creation, so that the primary work is that done by conditionals. According to Molinists, the distinctive conditionals that leave room for libertarian freedom are counterfactuals of creaturely freedom, and the view is plagued by a variety of objections, including the grounding problem, the falsity of the law of conditional excluded middle, and the troubling independence from God for these conditionals, threatening divine sovereignty.

The view I defend has many of the same features as Molinism but is not threatened by any of these standard objections to Molinism. The view begins by rejecting the idea that the conditionals of deliberation are counterfactuals, favouring instead the idea that conditionals of deliberation can be ordinary, indicative conditionals. Suppose Sally wants to go to the party, but not if obnoxious Bill is going to be there. She says to herself, ‘I will go to the party if Bill doesn’t go’, and so seeks to find out whether Bill is going. She learns from Bill’s roommate that Bill is going home for the weekend and so won’t be going to the party. Sally’s decision-making is completed when she uses this information to decide to go to the party, but in deliberating, she never resorted to the ‘were-would’ language of any counterfactual conditional.

So the first step towards Philosophical Arminianism (as I call the view) is to posit the use of indicative conditionals in one’s deliberative model. The second step I take is to characterize the model in terms of suppositional reasoning that employs a dynamic semantics involving updating rules concerning states of information, in contrast to the more usual truth-conditional possible worlds semantics used to characterize contemporary Molinism. The details are complicated, but the point to note here is just that this is another possibility to explore in the logical space of options that is non-Molinist but aimed at preserving exactly what Molinism aims to preserve.

Philosophical Arminianism shares some features with Molinism, however, including the idea that there are possible worlds that God can’t create. Some may think this a defect, though I don’t see why. So I’m content with embracing this feature, leaving open the possibility that there is an argument yet to be developed to show why this feature should not be embraced. In terms of the most common objections to Molinism, though, Philosophical Arminianism suffers from none of them. The conditionals used in the deliberative model are grounded in epistemic principles that are necessarily true and thus are no more a threat to Divine sovereignty than are the truths of mathematics and logic. Because they are grounded in this way, there is no need for the law of conditional excluded middle in order to explain how complete foreknowledge is possible.

There is also a connection at this point to Open Theism, for Philosophical Arminianism does not imply that God couldn’t create a world in which complete foreknowledge is absent. Instead, the task is to develop a view that explains the possibility of complete foreknowledge with libertarian freedom. Thus, Philosophical Arminianism is also compatible with a weak version of Open Theism, one that doesn’t demand incompatibility between foreknowledge and freedom but merely insists that the world we experience is that of a created order revealing divine risk-taking of a sort that arises only in the absence of complete foreknowledge.

Just for clarification, does your deliberative model of divine providence require that God deliberate or reason discursively, in the sense that God would go through reasoning steps (e.g. taking in information and employing updating rules) prior to choosing which world to make actual? And do you think your answer to that question has any bearing on theological or metatheological debates?

Yes, that’s a feature of the model, one that no one should think mirrors reality. Part of God’s cognitive perfection involves never needing to take any time to figure things out (and that’s true even if one rejects the atemporality of divinity). But when we model such perfection, we have to resort to deliberative models, and those suggest a cross-temporal element you notice. It is of the nature of models to mislead in certain respects, however, and so when presenting a model, it is important to note which parts of the model are meant to generate insight and which should be ignored. I thus tried to be clear about this aspect of modelling in Depicting Deity.

You’ve also written extensively about faith, where you describe faith as ‘affective’ – the disposition to act in service of an ideal, despite difficulties or setbacks. You contrast this with cognitive views of faith (in which faith is simply a matter of one’s beliefs, assumptions, etc.) and views according to which faith is trust. Will you describe your view of faith and why it fares better than other accounts of faith?

The starting point is to notice that faith is not, in itself, a religious concept: there is faith in humanity, in democracy, and in one’s friends and family, to mention a few non-religious examples. Moreover, it is a virtue, emphasized most strongly in Christian traditions, but easy to appreciate in non-religious contexts as well. As such, we should expect an account of the nature of faith to be fundamentally dispositional, though not in a way that is sub-mental, as some dispositions to behaviour are. Once we recognize that the disposition involves mental elements as well, we should be asking whether the mental elements are fundamentally or exclusively of one particular type. For example, are the mental elements exclusively cognitive? That sounds unpromising, as is shown by the fact that my faith in democracy involves a pro-attitude towards it. It is equally mistaken, though, to think that faith is exclusively non-cognitive, since faith in democracy has to involve in some way some thoughts or others about how democracy fares in comparison with non-democratic ideas. One quickly arrives at the conclusion that the mental elements involved in faith include cognitive, conative, as well as affective elements.

After getting to this conclusion, one might then search for the precise (semantic) contents of the mental elements that are present, but I argue that this approach is mistaken. For dispositional states, we should already be comfortable with the idea of multiple realizability for such states, when it comes to the underlying structure that makes for the disposition in question. This is especially clear when describing artifacts. What makes something be a table or a chair is some underlying structure that grounds the functional role that the object plays. But multiple realizability is the story of the day when asking about the underlying structure. And just so with faith: even if faith requires combinations of mental states, there is no particular combination that is uniquely required as the structural ground.

Moreover, one should expect this multiple realizability point to hold even when one turns to particular kinds of faith, such as saving faith in Christian traditions. Instead of looking for core commitments of Christianity that must be endorsed by anyone having saving faith, one should instead expect saving faith to be capable of being displayed by a wide variety of underlying mental structures. For those of us who build furniture, one can select harder or softer woods and compensate for the amount of sag distinctive of each by increasing or decreasing cross-member support. Just so, when comparing various examples of saving faith, we should expect some to involve utter cognitive certainty, while others make up for uncertainty with greater hope. Similarly, instead of looking to find cognitive commitments to basic theological truths such as ‘God exists’, we should expect to find a variety of cognitive commitments that may imply the theological truth in question but do not explicitly involve that semantic content. Such a functional understanding of faith resists the more standard cognitive and non-cognitive approaches to faith, where the former relies on the idea that the kind of faith involved can be discerned simply from finding out what a person believes, and the latter succumbs to the positivist tendency to find something of value in religious faith by explaining it in emotivist terms.

Using Jesus is Lord as an example of a typical theological truth whose cognitive commitment is typically looked for in people of the Christian faith, does your view make it so that someone could still count as having Christian faith while not believing Jesus is Lord, or even denying that Jesus is Lord, as long as one hopes Jesus is Lord or consistently acts as if Jesus is Lord in one’s authentic behaviour?

First, I’d like to change the issue from whether one has Christian faith to whether one has the kind of faith that is saving faith, according to the good news that comes to us through Jesus. I favour this change because it would be anachronistic to describe Abraham, Isaac, and Jacob (to name a few) as having Christian faith. But part of the good news that comes to us through Jesus is that each of these has saving faith.

So if we are looking for some mental state, either affective or cognitive, that has Jesus is Lord as its content, we’ll struggle with the saving faith of Old Testament saints. We’ll also struggle with what Karl Rahner describes as ‘anonymous Christians’. Finally, a third source of discontent will be those who first showed the kind of allegiance to Jesus and his message, both before and after the resurrection. I expect they came to view Jesus as Lord at some point, but not as early as they came to have faith in Jesus that counted as saving faith. I expect something similar repeats itself in the spiritual developments for all of us who follow Jesus.

Out of the possibilities you mention, the one that these remarks do not touch is the one about a person who denies that Jesus is Lord. As others have emphasized, it’s not always clear whether what is being rejected is the gospel itself or the combination of message and medium. It’s a Gandhi point that some of the worst advertising for Christianity comes in the form of Christians. So any case involving such a denial would have to be investigated carefully to make sure that the link between words used in such a declaration really expresses the relevant propositional content. I don’t doubt that ultimate rejection of the gospel is possible, but I also think that we are rarely in a good enough epistemic position to have much confidence in any judgment about such matters.

I’m wondering why you call your view of faith ‘affective’ rather than ‘conative’. On your view, faith is directed towards an ideal, and it requires a disposition to action despite setbacks, which seems more conative than affective. One of your arguments against trust-based views of faith is that they are too passive, or not active enough, for faith. Wouldn’t a view of faith that’s primarily affective seem just as passive as, or not as active as, a view of faith that primarily involves trust?

That’s a good point, and not much would change if I had called it conative rather than affective. When I was considering that possibility, I found it difficult to draw a clean boundary between the two, except in cases where conation involves activity of some sort: striving or trying, for example. Such cases are more appropriately understood in terms of faithfulness on the part of a person of faith, however, and it is fairly obvious that faith doesn’t imply faithfulness. So, when a person’s faith is weak, one should expect failures on the faithfulness score, and thus not much in terms of activity that is consonant with their faith. The stronger the faith, of course, the more it lines up with uniform faithfulness, and the greater the integration of heart, soul, mind, and purpose at which a life of faith aims.

One of your first books is on hell, titled The Problem of Hell. In that book, your view of hell is that it is not eternal retributive punishment, but it’s not universalism either, in which no one is in hell. What is the problem of hell? And how does your view solve the problem better than these other views?

The problem of hell is a special version of the problem of evil, one that arises even after we grant the obvious point that some suffering has a point to it, a point that can justify the suffering caused or allowed. The suffering of hell, however, leads nowhere, at least on standard views of hell. Removing a cancerous tumour causes suffering, but can be justified by leading to, or aiming at, the future health of the patient. But the suffering of hell has no such teleological purpose. So it is a version of the problem of evil that no greater good theodicy seems equipped to address.

In presenting and addressing this problem, I characterize a traditional, and strong, conception of hell in terms of four conditions. I then note that the history of discussion of hell involves various ways of rejecting one of these conditions: Universalists reject the Population Thesis; Annihilationists reject the Conscious Existence Thesis; and Second Chance Theorists deny the No Escape Thesis. But all of these mitigation ideas share with the strong view the core idea of the standard view of hell, which involves adopting a Punishment Model of hell: that those who end up in hell are there because they deserve it, and the justification for such presence in hell appeals to God’s justice.

More recently, discussions of hell have shown an inclination away from a Punishment model, favouring instead a Choice model for hell. C.S. Lewis, for example, claims that if the doors of hell are locked, they are locked from the inside. So what would explain the presence in hell is something about the choices of those who end up in hell. Similar approaches can be found in others, including Richard Swinburne and Eleonore Stump.

I develop the Choice model in an annihilationist direction, though the end state of annihilation is a possibility that may never be realized for any denizen of hell. It is also an account of hell that is compatible with hell being empty, though not with it being empty of necessity, as we find in strong versions of Universalism. I develop the view by reflecting on the competing values of existence and freedom, arguing from cases of rational suicide that sometimes the value of freedom overrides the value of existence in such a way that intervention cannot be justified. If those arguments succeed, then it is possible that intervention to guarantee the presence of all in heaven cannot be justified either and thus that it is possible that some end up in hell.

This position shares a virtue of other versions of a Choice model in that it denies a presupposition present in articulating the problem of hell above. For on any Choice model of hell, presence in hell involves non-intervention by agents capable of preventing it, a non-intervention justified in terms of respect for persons.

To visit the annihilation point, there are some who maintain that even if it would be better for some people not to live than to continue to live in their current state (e.g. in hell), there are deontic constraints on what God can do given the intrinsic value of the individuals God has created, such that not even God can actively put an end to the existence of individuals in hell, even if those individuals want God to do so. What do you say to someone with this view?

The central issue, as I argue in the second half of the book, is about whether freedom or being is of greater significance. It is on this issue that I part company with other choice model theorists who reject the possibility of annihilation. If being (or existence) had greater significance, they’d be right. The point of the discussion of rational suicide is to argue that they are not right. The point of the discussion is to show that intervention can’t be justified in some cases of rational suicide, and that this conclusion shows that freedom has greater significance, even if (though) there is some other sense in which existence is a great good.

In addition to your work on theological topics, you’ve written on topics that fit purely within epistemology: understanding, knowledge, fallibilism, and scepticism, to name a few. Do you see this work as related to your work on theological topics? Should your epistemological work inform our religious beliefs or our cognitive attitudes towards religious doctrines? If so, how?

Well, my working hypothesis is that everything is linked to everything, though we know very little initially about how all of it connects. So I’ve never started a project in epistemology with a specific idea about how it will affect my work in philosophical theology, nor the reverse either. When the connections appear, though, I’m both gratified and unsurprised. The most recent such happy coincidence came when I realized that the underlying explanation for complete foreknowledge that is found in Philosophical Arminianism also can be used by the sceptic to explain the significance of philosophical scepticism in a way that makes the position much more important than it would be if it were only a bare argument having a sceptical thesis as a conclusion.

Such connections are all there waiting to be found by those who search diligently. But I think they are found, not by looking for them directly, but by looking for something else and having them serendipitously appear in the process. Jesus said that when you put your hand to the plow, don’t look back. If we understand synecdoche, we’ll understand that we shouldn’t look up or to the side either. It’s a point about distraction, about focus, and being always on the lookout for connections is likely to lead to work that shows the effects of such failure of focus.

Thanks for your answers so far. A few more questions as we conclude: First, is there an area that you think deserves more attention from philosophers, one that’s underexplored but would greatly benefit from more philosophical interest?

I’ll mention one, though at the risk of passing over others. I think the implications of fallibilism for religious apologetic projects have not been appreciated or investigated seriously enough. A simple example about arguments for theism can illustrate the point. Suppose you think some consideration C counts in favour of the existence of God. Those with infallibilist presuppositions will want you to present a deductively sound argument that has C as one of its premises. Of course, that’s an easy task: just add a linking premise to the effect that if C, then God exists. That’s pointless, and not what the infallibilist is after. But why go down that path in the first place?

If one adopts a fully general fallibilist epistemology, one would approach the matter differently. One would recognize that all epistemic support is defeasible in character, in such a way that any final verdict about the rationality of any position would have to be holistic in character. So understood, deductive arguments might contain information relevant to such a final verdict, but the task of constructing the deduction would be a distraction to be avoided.

The same points can be made for defences of atheism and agnosticism, and more generally in metaphilosophy concerning proper philosophical methodology. The guiding principle here is that aiming for deductive proofs or refutations of any given position is a kind of false consciousness arising from an infallibilist presupposition. The interesting project that can arise from taking this suspicion seriously begins when we consider what defences of a view would look like under fully fallibilistic assumptions.

Yes, that is interesting. I suppose the result would look more like a cumulative case argument, then, right? Typically objections to cumulative case arguments address each of the arguments that make up the cumulative case individually, but I suppose doing that often isn’t sufficient, especially if the proponent of the cumulative case argument admits fallibility – the proponent might think he hasn’t taken into account all the factors when formulating the arguments, so the objections don’t target the strongest versions of the arguments. In this case, perhaps the objector would need to initiate something more like a paradigm shift in the proponent of the cumulative case argument. Is that along the lines of what you’re thinking?

The problem with cumulative case arguments is that they can’t work unless they take account of the total epistemic situation of the individual in question. As I see it, the problem with any argument is this generic lack of specificity. (A caveat: we can of course put in a linking premise to make the argument deductively valid, and then it will have no formal flaw; but it will still be too generic in that the epistemic standing of the linking principle (if the other premises are true, then the conclusion is true) will vary depending on the total epistemic situation of the person in question.) So the best that can be done is to point to relevant considerations and say what valence those considerations have, but what that accomplishes is merely to identify elements common to any total epistemic situation. Engaging in such a project is certainly worth doing, but the final step of drawing a conclusion would have to be hypothetical, always hostage to the assumption that the considerations discussed exhaust all relevant considerations for any total epistemic situation. If we are thinking of the variety of possible total epistemic situations, that assumption will always be questionable: to advert to a Quinean motif, any finite set of considerations can be embedded in any web of belief that preserves a desired conclusion about naturalism and supernaturalism if we make enough adjustments elsewhere in the web.

Last, you’ve had a lot of experience overseeing young philosophers. You’ve been on admissions committees, directed dissertations, led graduate seminars, and seen philosophers ascend from their lowly first year of graduate school to become scholars in their field. Do you have advice for aspiring philosophers of religion as they navigate their way to becoming scholars?

I try to avoid offering unsolicited advice, since it is almost always too generic to apply to specific situations and circumstances. I also note that the advice that has made the most difference to my own life and career is advice that arose out of extended periods of shared experience and long hours of listening without interruption. So, I guess my advice is meta-advice: look for advice grounded in thorough familiarity and understanding of you and your particular circumstances, from people with insight (and good luck trying to figure that out).