Pauline Soterilogy Interviews: Doug Campbell (Part 1)

In the coming weeks I shall be interviewing Douglas Campbell Associate Professor of New Testament at Duke Divinity School and Michael Gorman of St. Mary's Seminary about their forthcoming books on Pauline soteriology (both with Eerdmans).


I first met Doug at SBL in 2005 (Philadelphia). I had just brought his book The Quest for Paul's Gospel, walked about 30m down the book exhibit, and then bumped into him. So I got him to sign it for me and we had a brief chat. Doug is great because he confirms my prejudice that the best biblical scholars in the world are antipodian! Doug is also the most prominent champion of the subjective genitive interpretation of the pistis christou construction in Paul's letters these days. Anyways, here is the first installment of my interview with Doug Campbell on his forthcoming book The Deliverance of God.

1. Can you tell us about your intellectual journey in Pauline soteriology? What were the major moments in your research and what has influenced you the most?

That’s an excellent question to begin with. (Incidentally, I explain all this in more detail in the Preface to The Deliverance of God—hereafter DOG.)

My doctoral work took place in the 80s in Toronto, mostly under the aegis of Dick Longenecker, but influenced by John Hurd, Schuyler Brown, and Peter Richardson as well—all wonderful scholars in their own ways. Longenecker was writing his Word commentary on Galatians at the time, as well as lecturing on Romans, so that conjunction of events really shaped the future of my life. He was heavily engaged with Sanders’s work. (Sanders began his teaching career in Canada and so the two scholars knew each other quite well; moreover, Longenecker had anticipated many of Sanders’s celebrated claims about Judaism in his earlier book Paul, Apostle of Liberty). Longenecker was a strong advocate of participatory approaches to Pauline soteriology in ultimate dependence on Deissmann; of the correctness of the faithfulness of Christ reading of Romans 3:22 etc.; of a positive approach to Paul’s Jewish background; and of the broader importance of Jewish martyrological thinking for Paul’s development. So my work is clearly just a fairly direct continuation of his agenda!

The difficulty I was left with after my doctoral work—during which I focused eventually on Romans 3:21-26—was that Longenecker left most of these dynamic interpretative trends unresolved in relation to one another. One didn’t need to be too stringent interpreting a proto-Rabbi in his view. And this didn’t really satisfy me.

The next event that really changed my life occurred during my first teaching job, at the University of Otago in Dunedin, New Zealand. The local Presbyterian seminary had just appointed a very young and brilliant theologian, Alan Torrance, to its theology chair (son of J.B., and nephew of T.F.). Alan and I became great friends, and during the course of our many conversations about fishing, music, sport, and university politics, I received an excellent theological education. His father’s work on the role of contractualism in Scottish Presbyterianism over against an unconditional covenantalism was especially important for me at this time—something I learned about after Alan had passed on this material to me (and explained it!). I noticed that it coordinated and articulated with enviable clarity many of the debates that we were struggling with in Pauline studies—concerning the role of Jews and Judaism, the atonement, faith, and so on, that is, many of the very questions that Longenecker’s teaching had left me with. The application of James Torrance’s categories to the concerns and texts of the Pauline interpreter seemed to promise an exciting moment of clarification and theological progress—although this was a task that proved harder to fulfill in detail than to envisage in broad prospect!

In 1996 I accepted a lecturing position in New Testament at King’s College London, partly to continue to work alongside Alan, who had left NZ in the interim, and partly to learn how to do rigorous NT work from scholars like Graham Stanton and Francis Watson. (Graham moved shortly after to the chair in Cambridge, and was replaced by the equally adept Judith Lieu; I also learned a lot at this time from my long-time friend and colleague at Kings, Eddie Adams.) But conditions were very difficult in UK universities in the late 90s and I struggled to make real progress on my project. A year in Germany got it going, but there was still a long way to go. So I moved to Duke in 2003, and finally got the support I needed to get the treatment finished. But this move also enriched the analysis in some additional ways.

I learned in particular from Stanley Hauerwas—who was admittedly in certain respects reinforcing many of the things I had learned at King’s from Colin Gunton—that I needed to incorporate explicitly the political and ethical dimensions in the reading strategy that I was criticising. Putting things at their bluntest, there was an important connection between exegesis and execution that I had not really grasped. But how could I overlook it when I had moved, partly unawares, to a state that still killed criminals, supported all the while by its surrounding, heavily churchgoing populace? Articulating these connections slowed me down and expanded the project still more, but seemed important. I also continued to press against any thought-act and being-act dichotomies in interpretation, and away from universality and “principles” toward particularities. These concerns continued to open up both the primary text and the secondary literature in some surprising ways.

So that’s basically how I ended up writing The Deliverance of God in the broader setting of my academic career. It’s been a long journey, but hopefully it will have been worth the wait.

2. In your last book, The Quest for Paul's Gospel, you championed an approach that you abbreviated PPME. What is PPME and how does it differ from the justification by faith model and salvation-history model?

PPME is just a teaching rubric I use that expands on Sanders’s formula for the heart of Paul’s soteriology for the purposes of greater precision. He—quite rightly in my opinion—viewed the centre of Paul’s gospel in terms of “participationist eschatology,” so in my abbreviation, PE. (He was drawing to a degree on W. D. Davies at this point.) But I have a couple of difficulties with leaving things at this level. I worried in particular that some of my students might be prone to misunderstanding or overlooking some key issues. So paying the price of complexity, I expanded this formula in two ways.

First, I added a P to the P already denoting participationist/participatory to indicate that this all-important process was effected by none other than the Holy Spirit. Our participation in Christ is, in other words, irreducibly and non-negotiably pneumatological. No other sort of participation makes sense, yet this form of participation leads us to the heart of Christian reality. That is, by adding this further P, I was intimating that the very structure of Paul’s soteriological thinking was Trinitarian—not admittedly in a fully developed or articulated form, but irreducibly and inherently so. And recall that consternation about just how participation works or is effected is one of the main criticisms leveled against this particular construal of Paul’s gospel. It is, after all, fundamentally a miracle that we can participate in the new creation of the age to come.

I also worried that an emphasis on participationist eschatology was too oriented toward the resurrection and Christian triumphalism—positions that Paul spent much of his time combatting. So I introduced an M before E, denoting “martyrological,” to indicate that the Christ event in which Christians participated—pneumatologically!—had two critical trajectories spanning the cross as well as the resurrection. The M denotes, in short, Paul’s theology of the cross. And in flagging this up I was also creating a space for the narrative of Jesus’s crucifixion to be told, as elaborated in Phil. 2:5-11 and related texts, that allowed in turn an appropriate emphasis at some point on the way Christ’s fidelity and submission created the space in and through which others are saved. So I was creating a door through which JF terminology could be coherently integrated with what I was already arguing in relation to the heart of Paul’s concerns in PE terms so to speak.

The price paid for these qualifications was of course a degree of complexity. Perhaps it has been a mistake in retrospect, and I should have stuck with a mere name. But the formula PPME was intended primarily to be a teaching rubric, not a scholarly contribution. And I think it has worked pretty well in those terms for me at Duke. Scholars who already know all this should of course feel free to ignore it and to use simpler, more traditional names like “participation,” remaining aware that in any discussion with me I am going to want to supplement that descriptor early on with further qualifications in terms of pneumatology, martyrology—or, perhaps better, cruciformity in Mike Gorman’s insightful word—and eschatology!

Now to address the second part of your question—an equally important set of queries.

The PPME model differs from both Justification and Salvation-historical construals of Paul’s gospel in certain absolutely crucial ways. But before detailing those, let me first note that I use rubrics for these approaches as well in Quest to avoid providing extensive subliminal reinforcement for what I view as deeply problematic translation decisions—especially in relation to “justification by faith.” So I will speak in what follows of the JF and SH models.

The key point to grasp is that these two models—JF and SH—are both variants on classical theological foundationalism, which is to say that, in Sanders’s charming phrase (more or less), they “work” or “think forward.” What he means by this is that their epistemology is established prior to the Christian state, from the ground up so to speak, by human reflection of some sort. The specifics of their reflections and resulting (theological) epistemologies are rather different, but their basic modality is therefore the same. Essentially a phase of human reasoning takes place prior to the proclamation of the gospel to “ground” that proclamation in terms of truths and criteria that have already been clearly established. And this ultimately generates all sorts of problems.

(1) In the entire history of human thought, it has consistently proved untrue. That is, the opening phase that supposedly establishes axioms and criteria that are universally agreed upon invariably fails, creating a question-begging or even deceptive set of preliminary claims.

(2) Those claims tend to turn out instead to be self-reifying (usually in tandem with a sinister othering project). That is, instead of resting on universally demonstrable or perceptible claims, the key axioms and criteria in any foundationalist project invariably turn out to privilege the person or group making those claims in terms of their social and historical location(s)—so the JF model says precious little about regimes in which white, elite, men have (for example) enslaved black, foreign men, women, and children (and so on); it turns rather a blind eye to such situations. (Note Charles Marsh’s devastating account of Douglas Hudgins in God’s Long Summer at this point.) Moreover, Christian foundationalist projects have invariably constructed their notions of soteriological success out of prior “objective” failures in Judaism, generating “necessarily and objectively true claims” about the stupidity and/or immorality of Jews—clearly not a good thing. I am of course painting with a broad brush at this point, but careful and detailed examination of foundationalist projects invariably reveals these sinister tendencies, whether to a greater or lesser extent.

(3) These stringent, if false and self-serving, prior criteria tend to function as a tremendous obstacle to the new categories and truths introduced by the gospel, overruling any attempted evangelical corrections. So, for example, if a conception of the gospel that seems deeply grounded in the nature of Christ, and in the nature of God as revealed by Christ, challenges prior criteria, it tends to get overruled or ignored, or even branded as heretical! (Some of the recent disputes over the atonement spring to mind at this point.)

(4) Of particular relevance to Pauline interpretation is a fourth major point that these models or construals that work forward must intrinsically exist in diametric tension with those construals that work backward. You cannot think forward and backward at the same time—in terms of the derivation of your key theological criteria (i.e., not psychologically)—without being in a terrible muddle. Moreover, the very nature of the axioms and criteria generated foundationally will tend to contradict the very nature of the axioms and criteria revealed through Christ retrospectively. So it does not take much time to detect that the JF and SH models tend to contradict the PPME model at every level, in relation to every major issue—epistemological, theological (i.e., the basic attributes of God), Christological, soteriological, pneumatological, anthropological, ecclesial, ethical, and so on. So we are not talking here about superficial or secondary points of tension, but major faultlines running through the middle of the most important things that Paul says. If Paul thinks both forward and backward at the same time then he is largely useless to the church for theological reconstruction (i.e., in historical terms) because he is so deeply contradictory; different answers could be generated to every major question in terms of the different systems supposedly operative in his work. And at this point I have clearly segued into my suggested solution, that I pursue down one important avenue in DOG, having sketched it out programmatically in Quest—the need to find one basic construal of the gospel, preferably retrospective, that can explain the bulk of Paul responsibly and in a way that eliminates fundamental dependence on basic prospective models.

We need to show in detail and responsibly that Paul primarily thought backward. And this means pushing back against the JF and SH models, in favour of a PPME approach. Enormously important issues are at stake in this comparatively simple set of recommendations.