In its modern form, the study of Christian origins has been dominated by the study of the apostle Paul. Take, for instance, Johannes Weiss’s précis of nineteenth-century scholarship on Christian origins, written in 1917: ‘’The history of primitive Christianity is usually written as the history of St. Paul’ (Primitive Christianity, 1). Or, one might say, the history of St. Paul’s letters.
On this way of reading, Paul is a series of discrete letters, to which we have discrete access, and about which we can make discrete critical judgments. With the raw data to hand, scholars then chronologically re-shuffle the Pauline pieces around the board marked ‘early Christian history’. This is a pregnant way to read the letters, and has a long and venerable history. Going all the way back to J. D. Michaelis’s 1788 critical Einleitung, and then more robustly in Leonhard Usteri’s 1824 The Development of the Pauline System of Doctrine, scholars have been hard at work at precisely this project: chronological rearrangement of Paul’s letters for the sake of biographical, theological, and historical narration. A century apart, F. C. Baur’s and Ernst Käsemann’s reconstructions of early Christianity, still influential, gain a good deal of their leverage by reading the Pauline collection this way.
There is, of course, another possibility. In his 1950 Chapters in a Life of Paul, a book in which he was otherwise at pains to distinguish between Acts and the letters as sources for Paul, John Knox begins with a methodological observation, made more or less in passing, about what they share in common. He writes, ‘We may appropriately begin by noting that, like the Gospels, the documents on which we must depend for our knowledge of Paul [i.e. Acts and the letters] came into being in their present form for the use of the Gentile churches of the late first and early second centuries in response to their interests and needs’ (16). Knox is here acknowledging a simple historical truth: that when it comes to the letters, our only access to Paul is via a letter-collection, with its origins (likely) in the late-first/early-second century, a product of its own editorial design and concerns. In the end, Knox doesn’t press this insight, and falls back into the fairly traditional way of treating the Pauline data outlined above. Discrete letters, discretely accessed, chronologically rearranged. It’s a perfectly acceptable, not to mention powerfully imaginative, way (not) to read a corpus.
But try as we might, the access to Paul it imagines is precisely the access we don’t have. Nor, as Knox observed, does this modern hermeneutic attend to the textual artefact(s) we do have. Knox’s observation has remained largely untapped amongst Pauline scholars (except for Trobisch  and, somewhat differently, Childs ). But a similar recognition is causing something of a revolution in the study of classical epistolography.
It was, in fact, precisely this chronological obfuscation of ancient hermeneutical concerns in constructing an epistolary corpus that animated Mary Beard’s 2002 call to a ‘radically old-fashioned’ way of reading Cicero’s letters (“Ciceronian Correspondences: Making a Book out of Letters,” in Classics in Progress). That is, it was Beard’s rather simple suggestion that Cicero’s letter-books be read as originally designed—that is, as books—and studied according to their own principle of arrangement, rather than not read and splayed across a chronological edition. And this project has been taken up. In 2012, acknowledging Beard’s inspiration, Roy Gibson studied eleven ancient letter collections, finding that their dominant principle of arrangement was by addressee or theme, and within this by what he called ‘artful variety’ or ‘significant juxtaposition’ (JRS 102 : 56-78). Not surprisingly, much of the subsequent work has focused on Cicero’s and Pliny’s collections, but with the publication in late 2016 of Late Antique Letter Collections, we now have a critical introduction to the majority of the epistolary corpora of late antiquity, and one that treats them along the lines envisaged by Beard and Gibson.
This last-named volume does not treat Paul’s letter-collection, which makes a certain degree of sense: Paul neither lived nor wrote in late antiquity. In another sense, however, it is ironic, for with a few exceptions, the earliest collections of Paul that remain are products of late antiquity. Put slightly differently, before they are anything else, scholars of Paul are, formally speaking, scholars of late-antique letter collections. There are (small) signs that recognition of this may be growing. The last decade has seen renewed interest in the so-called Euthalian edition of Paul (most recently, Blomkvist  and Scherbenske ), and just this year, T.J. Lang and Matthew Crawford have published the first study, in our discipline, of a remarkable edition of Paul’s letters issued by Priscillian of Avila (NTS 63 : 125-45). Both of these editions are likely late-fourth century, and evince a way of reading Paul in many respects different from that outlined above—driven not by chronology or historical narration, but by the hermeneutical frame of the letter-book.
In the name of history, Pauline scholars, of course, will always remain free to dismember these corpora, to treat the individual letters as discrete entities, and to mine them for historical, biographical, theological, and social (et al.) insights into Paul, his communities, and earliest Christianity. But good history also requires a ready recognition of the nature of the available evidence. And to this end, Knox’s observation serves at least something of a caution: our only access to Paul remains the late-antique collections that attest a hermeneutic very different from our own. In the end, the study of Christian origins may depend on such a disassembling and reassembling of Paul, but it is surely a violent way to read the material evidence we actually have, and as the late-antique collections attest, not the only fruitful way to read the Pauline book.