Campbell_Framing_Paul_text

Locating Philemon, Colossians, and “Ephesians”

1. Locating Philemon

This chapter’s deliberations begin with a consideration of Paul’s “short, attractive, graceful and friendly” letter to Philemon.1 Philemon is usually included without demurral in the group of authentic Pauline letters, but determining its exigence opens up a useful area of further interpretative possibility within our developing frame. So its data merits careful scrutiny.

The letter describes an intriguing scenario as its immediate exigence. Like 2 Thessalonians 1:4, Philemon 2 lacks a particular geographical association for its reference to an “assembly,” but its addressees are nevertheless quite specific. Although usually referred to as “Philemon,” the letter was addressed to a community. Paul was sending an unhappy slave, Onesimus, or (from the Latin) “Useful,” back to his master, Philemon. We know this because after the prescript, the letter’s address shifts consistently into a first-person masculine mode. However, the letter begins with a communal focus. Philemon was apparently married to Apphia (she is less probably his sister), and they belonged to an assembly that met in the house of a veteran, Archippus.2 So three figures are named initially, along with a house church. Paul was himself in prison at the time of writing — a state he makes much of (1, 9, 13, 23) — and Epaphras is said to be a “fellow prisoner” (ὁ συναιχμάλωτός μου; 23). Paul is also accompanied by a substantial group of “fellow workers” — Mark, Aristarchus, Demas, and Luke, at the least, along with his co-sender, Timothy.

Scholars have pondered the exact situation that underlay Onesimus’s meeting with Paul. I still find Lampe’s (1985a) suggestion persuasive — that Onesimus was not a runaway (fugitivus) but rather a “truant” slave (erro), appealing to a “friend of the master” (amicus domini) for an intervention into a difficult situation. That a runaway slave would contact anyone associated with his master defies all reason (and that Onesimus would just happen to have been arrested and imprisoned in the same cell as Paul, so to speak, seems highly improbable, not to mention that the letter itself is silent concerning an utterly desperate legal situation). But a deliberate if unofficial appeal by an unhappy slave to an authority figure in relation to his master, a slave therefore operating at the time effectively AWOL as an erro, explains the situation in the prison plausibly.3

As we turn from the letter’s immediate implied exigence to consider its location in relation to our frame, it is helpful to note that Paul’s complex strategy on Onesimus’s behalf vis-à-vis Philemon extends beyond the mere dispatch of a helpful letter, to the somewhat intimidating promise of a visit — apparent in his request for the preparation of a guest room (ξενία). And this data, combined with the immediate exigence of a slave operating as an erro, suggests that Paul was imprisoned relatively close by Philemon’s location and hence most probably in the same region or province, and even the same area.

Paul’s pointed request for the preparation of a guest room would seem to suggest a visit rather soon, perhaps within days. And if Onesimus had been away on his unauthorized visit for much more than a few days, he would probably have been declared a runaway or fugitivus, a fearful status that he evidently wished to avoid by contacting Paul. So the letter implies at the least that Paul’s imprisonment was not very far from Philemon’s community, that is, not too many days’ walk in any direction. (This reasoning seems to exclude the Acts-based locations of Caesarea and Rome,4 by way of anticipation.)5 Unfortunately, the letter does not go on to tell us explicitly where either Paul’s prison or the community in question lay. However, some hints in the data are suggestive.

Philemon’s wife, Apphia, possesses a name frequently found in the province of Asia and attested specifically at Colossae.6 And later tradition associated the letter with Colossae, which was located on the border between Lydia and Phrygia, since this was (at the least) the destination of a letter bearing Paul’s name that is strongly associated with his letter to Philemon. So there are hints that the community lay in Asia. And both hints corroborate one another, essentially independently.

Moreover, the letter presupposes an effective founding visit from some member of the Pauline mission. But Paul himself sends no greetings from the local “brothers” at his location, so he does not seem himself to be imprisoned at the site of a successful mission; no local Christian seems to be named besides itinerant members of his circle of coworkers. And this oddly asymmetrical scenario along with the figures specifically named do not obviously resonate with any of the communities that we already know about from the other authentic letters considered thus far, that is, with Paul’s communities in Galatia, Macedonia, and Achaia, in relation to which we have already treated Romans, 1 and 2 Corinthians, Galatians, Philippians, and 1 and 2 Thessalonians. However, if this seems implausible, we know of successful Pauline communities outside of these regions only in Asia, although even in this region, the letter does not resonate with what we know of Paul’s mission in Ephesus (see 1 Cor and, to a lesser extent, 2 Cor), where Prisca and Aquila were present. So we end up, after tracing through these implications — at least in probabilistic terms — with a location prior to the successful Ephesian mission in Asia.

Clearly, this case is far from decisive. Too many figures traveled too far during the Principate for us to place much reliance on the implications of Apphia’s name (and Lydia springs immediately to mind, to nod in the direction of Acts once more); the letter to the Colossians might have been concocted; and the conjunction of evidence here might have been coincidental. Moreover, Archippus’s house church might have been in Lystra or some similar location, with Paul imprisoned nearby (although later data will speak against this). However, I doubt that all these objections can hold good simultaneously, and any counterproposals are utterly unattested. So we will tentatively locate both Philemon and Archippus’s house church in Asia — more specifically, in the regions of Lydia or Phrygia (although the latter might take us into the western reaches of the province of Galatia) — but will refrain for now from building too much on this judgment.7

If Philemon was written in Asia (or western Galatia) prior to the Ephesian mission, then it would predate 1 Corinthians, which we have already determined in chapter 2 was sent from Ephesus to Corinth in the spring of 51 ce. Moreover, in the spring of 51 Paul is in Ephesus looking back on a communication with Corinth (the Previous Letter to Corinth), as well as on other events that had unfolded in relation to that congregation (i.e., Apollos’s first visit to Corinth). He has been evangelizing Ephesus, and apparently a great but challenging opportunity has just arisen for him there (see 1 Cor 16:9: θύρα γάρ μοι ἀνέῳγεν μεγάλη καὶ ἐνεργής, καὶ ἀντικείμενοι πολλοί). So we may extend his ministry and presence in Ephesus back through at least part of the previous winter. And we can chart Paul’s activities through the calendar year beginning in 51 with considerable precision. The letter to the Roman Christians closes this period, leaving him in the spring of 52 contemplating journeys out of the entire area — to Jerusalem, to Rome, and then on to Spain. So it is unlikely that the events eliciting Philemon unfolded after 52. However, we have a considerable chronological window between the Macedonian and Achaian missions undertaken as early as 40-41 ce and the beginning of Paul’s Ephesian ministry late in 50, within which this letter could fall.

We know, moreover, that Paul traveled to Syrian Antioch and to Jerusalem just before the composition of 1 Corinthians, and presumably before the Previous Letter, hence toward the end of this interval (i.e., somewhere from late 49 through mid-50 ce, 13.x years after his first visit to Jerusalem, in late 36 ce). He seems to have visited the Galatians during his return journey, giving them instructions about the collection (1 Cor 16:1). We have not yet determined decisively whether these were “ethnic” or “provincial” Galatians, but the northern route between Syrian Antioch and Ephesus on the journey back from Antioch and Jerusalem could have taken Paul past the Lycus valley and so through Lydia and western Phrygia, and the southern route almost certainly would have.8 Hence, this seems a plausible if tentative location for Philemon and its particular exigence temporally. If Paul was imprisoned during his return journey to the Aegean from Syrian Antioch and Jerusalem, then the data evident in Philemon would be plausibly explained. And this would locate these events in 50 ce, with the incident in Syrian Antioch and Paul’s second visit to Jerusalem (see Gal 2:11-14, 1-10) taking place just before them.9

In sum, the epistolary evidence locates Paul in this general area at this time — in Asia in 50 ce. There seem to be no obvious difficulties with these judgments (at least yet), although some have been made only tentatively, as the evidence allows.

Stylometric data can get little purchase on Philemon, in part because the letter is so short. Meaningful samples usually require around 500 words, and arguably more, and Philemon contains only 335. (As stylometric techniques advance, the relevant word limits are dropping, so this limit may ultimately relax; however, at present it holds.) So we receive neither confirmation nor disconfirmation of the letter’s authenticity in stylistic terms. But should we entertain suspicions about the letter’s authenticity on more traditional grounds?

In fact, few scholars have doubted Philemon’s authenticity, probably largely because of its brevity, specificity, and apparent innocuousness. Raymond Brown (1997, 502) puts the case for authenticity here succinctly: “The [question] is why would someone bother to create Phlm, a note with such a narrow goal, and attribute it to Paul.” The presumption of innocuousness is not especially accurate, however. Baur managed to wring quite a bit of auspicious content from this letter.10 Further, a case could perhaps be made that the letter addresses the issue of slavery for a later ecclesial generation — essentially conservatively, from a modern liberal vantage point. But this situation and address are not notably distinguishable from Paul’s day. Arguably more decisively, Philemon might have been composed to facilitate the acceptance of a more ambitious pseudepigraphon into a Pauline letter collection, namely, Colossians, with which it is closely intertwined. Having said this, however, it seems equally likely that Colossians, if pseudepigraphic, seized the opportunity afforded by the authentic but more innocuous Philemon. Indeed, we can surmise fairly that the presence of Philemon in any Pauline collection was probably bound up with Colossians; it seems unlikely that this short and quite specific letter would have been included without some weightier assistance.11 But its inclusion as a companion letter to Colossians makes perfect sense (and further evidence will be cited shortly in support of this claim). However, this consideration tells us nothing about the letter’s authenticity.

In sum, then, the argument from anachronistic significance is unproved, and the arguments in terms of brevity and specificity still stand. Therefore, the letter’s authenticity will be affirmed in what follows. And with this judgment made, we can turn to consider Colossians — a letter whose authenticity is much more widely contested but that provides rather more data for the question’s assessment. We should recall in doing so, however, that the investigation of Philemon has created an intriguing interpretative option for the location of Colossians, and perhaps thereby for Ephesians as well — namely, a captivity somewhere in Asia in the year preceding the composition of the “major” letters (i.e., 1 Cor, 2 Cor, Gal, and Rom, along with Phil).

2. Locating Colossians

Locating Colossians in relation to our developing biographical frame is a fascinating but complex task; many of its aspects are hotly debated. As usual, however, at this point in our investigation, we will work from the letter itself outward so to speak. So the key initial features of its immediate implied exigence will be explored before we consider the letter’s fit with our developing frame and then the contentions that surround its authenticity in terms of its arguably distinctive style, its use of possible anachronisms, and its ostensible substantive tensions vis-à-vis other letters, especially in terms of eschatology.

2.1 The Immediate Implied Exigence

(1) Paul’s Incarceration

Paul was in prison when the letter was composed. This is hardly problematic, or even surprising, given the immediate locations of Philippians and Philemon, and the claim of 2 Corinthians 11:23, supported by further references to Jewish and Roman punishments in vv. 24 and 25 (. . . ἐν φυλακαῖς περισσοτέρως. . . . ὑπὸ Ἰουδαίων πεντάκις τεσσεράκοντα παρὰ μίαν ἔλαβον . . . τρὶς ἐραβδίσθην; and if Col proves authentic, this takes the ratio of letters written by Paul during an incarceration to 3/9 or 33 percent, a percentage worth pondering — and Eph and 2 Tim could add to it).12 As we will see in more detail momentarily, a significant sequence of events seems to have unfolded while Paul was incarcerated. But this too is scarcely difficult to comprehend. Incarceration in Paul’s day primarily detained people prior to trial. Its duration depended on the speed of the local judicial process, along with its integrity, and this varied a great deal. People could spend years in confinement waiting for trial if the local organization was poor or corrupt. We cannot yet judge the length of Paul’s incarceration during which Colossians was composed; that judgment must await the introduction of the surrounding frame. We can say at this point that, as suggested already by Philemon, Paul does not seem to have been imprisoned in a place that had had a successful local mission, at least yet. He sends no greetings to the letter’s recipients from local brothers, only from named coworkers.

(2) A Proxy Mission in Colossae

The letter states that Paul was not personally known to the Colossians (2:1: Θέλω γὰρ ὑμᾶς εἰδέναι ἡλίκον ἀγῶνα ἔχω ὑπὲρ ὑμῶν καὶ τῶν ἐν Λαοδικείᾳ καὶ ὅσοι οὐχ ἑόρακαν τὸ πρόσωπόν μου ἐν σαρκί). Their existence as a congregation is attributed to one of their own, an apparent slave called Epaphras (1:7). Thus, quite a complex sequence of events must be posited prior to the letter’s composition, which should now be probed for its details and plausibility. I suggest that the data is explicable, largely — and perhaps only — in terms of the following narrative.

As we just noted, Epaphras was apparently a Colossian and a slave (1:7 speaks of Ἐπαφρᾶ τοῦ ἀγαπητοῦ συνδούλου ἡμῶν, and sun- [συν-] compounds are always referentially significant for Paul;13 in 4:12 he is denoted as “one of you” [ὁ ἐξ ὑμῶν] and, again, as “a slave of Christ” [δοῦλος Χριστοῦ]). Hence, Epaphras had to have traveled first to Paul’s location, which was not Colossae, and converted there, presumably having been sent there by his owner.14

Excursus

An alternative scenario is the possibility that Epaphras was converted by the mysterious congregation at Laodicea, or perhaps even in Hierapolis, and then converted his household to that form of Christianity sometime before Paul’s arrival in the area. He could later have made contact with Paul during a trip to Paul’s location, since he was presumably a well-known Christian leader, and have fallen under Paul’s influence at that point. But although it is worth considering, I do not view this reconstruction as likely.

First, Colossians depicts a set of problematic practices at Colossae that Paul views as an intrusion; the Colossians are exhorted to stand firm and not move from what they were originally taught — suggesting that the foundation of the congregation was Pauline. But this original gospel came from Epaphras. If the institution of the church was Pauline, by way of Epaphras, then it follows that Epaphras was almost certainly a Pauline convert.

In support of this principal inference, we can also note, second, that Onesimus traveled to Paul for an intervention, thereby indicating that Paul was perceived to be the key local figure of spiritual authority, and suggesting as well that he was the foundational spiritual figure for the congregation. However, he was introduced to and represented at Colossae by Epaphras. And this is most easily explained if Epaphras was Paul’s convert. At the least, he would have to have been extensively catechized by the apostle. Third, it seems unlikely that Paul would have affirmed Epaphras’s founding role at Colossae so firmly in Colossians if Epaphras had not been considered a trustworthy mediator of Pauline tradition, and this, again, is most easily explained if he was Paul’s convert. And fourth, it would have been implausibly coincidental for Epaphras, a slave, to have founded the congregation from some other source and then to have met Paul and introduced the Pauline gospel decisively at some later point.

Arguments from explanatory parsimony must be used carefully in historical reconstructions of this sort; they can oversimplify complicated situations for which we simply lack the necessary amount of data, overriding the way that the extant data can gesture toward underlying complexities. But in this instance, it seems to be safely applicable. The most economic explanation of the data is the supposition that Epaphras was a convert of Paul’s.

After Epaphras’s conversion, he had to have returned to Colossae and converted some Colossians, who now presumably formed the nucleus of that congregation, if they did not constitute the congregation in its entirety. And Paul must have been imprisoned either during these events or shortly after them, preventing his own travel to Colossae, so that the Colossians had not yet set eyes on him.

But Epaphras is with Paul at the time that Paul writes Colossians to the congregation (4:12-13), so he had returned to Paul’s location at some point. Moreover, although he is characterized positively (4:12-13), he is somewhat surprisingly not returning to Colossae with the letter, which is being sent with Tychicus and Onesimus (4:7-9). So some questions now arise (noting additionally that this narrative will be complicated in one further respect by Philemon). How did Epaphras convert if Paul was incarcerated? Is his conversion in these circumstances plausible? And why did he not return to Colossae with the letter? But these queries do seem to have plausible answers.

Paul might not have been incarcerated when Epaphras converted, but even if he was, Colossians attests to a group of coworkers operating outside Paul’s confinement who could have made contact with Epaphras and converted him as they worked. This circle comprised no less than seven figures, some of them presumably quite experienced missionaries in their own right: Timothy, Aristarchus, Mark (denoted as the relative of Barnabas), Joshua-Justus, Luke, Demas, and Tychicus.15 That this circle might have made contact with Epaphras in some way and then converted him seems entirely plausible. Perhaps they showed inordinate kindness at some point to a slave from out of town, as did Paul to Onesimus, as evidenced by Philemon.

That Epaphras remained with Paul when Colossians was dispatched is also explicable in a number of different ways. Paul’s short letter to Philemon that apparently accompanied Colossians (and here we must anticipate a fragment of information that will be addressed in more detail shortly) suggests that Epaphras had been imprisoned at some time (Phlm 23), and this might help answer this query. If Epaphras had been incarcerated at Colossae for his offensive new piety, then it might have been dangerous for him to return. Having said this, he may simply not have been the most appropriate person to accompany Onesimus and to read out and interpret Paul’s letter. That this slave would have been unable to read seems probable simply on statistical grounds.16 Alternatively, he might have been gifted to the Pauline mission by his master and so now have been operating with the Pauline group and no longer at Colossae, as Paul seems to have hoped in the case of Onesimus (here again anticipating information from the frame).17 So once again, a plausible explanation of the data seems possible, although, unfortunately, it is hard to know which scenario actually obtained.

With the role of Epaphras clarified — or at least plausibly explained in various possible ways — we should turn to consider another odd suggestion within the letter concerning its implicit prior events.

(3) The Congregations at Laodicea and Hierapolis

The letter to the Colossians indicates the existence of congregations at Laodicea and Hierapolis (2:1; 4:13, 15, 16), which seems at first glance even stranger than the prior foundation of the community at Colossae by proxy. But we do not need to suppose that members of Paul’s circle, assisted by Epaphras (or vice versa), had founded these congregations in the time available in addition to the congregation in Colossae, something that is perhaps implausible, although it probably should not be excluded outright. Nothing in the data necessitates this. That they had had time to visit these communities in the interim, learning Nympha’s name, for example (Col 4:15), is plausible.

If we lean on the Asian (provincial) provenance of Colossians at this point (again anticipating a contribution from the frame by way of Philemon), this phenomenon is even more understandable. The existence of pockets of converts in the Lycus valley ca. 50 ce independently of the Pauline mission is by no means impossible. As we have already noted, the area was on the main southern east-west route through Asia Minor and was studded with large Jewish communities (see esp. Trebilco 1991). Christian growth was intimately related to Jewish networks, at least in its early years (see esp. Stark 1996). Hence, that converts had “leaked” into this area independently of Pauline missionary work is plausible (and later data will speak to this question as well).

Paul’s work had hitherto been much farther to the west, in Europe, and to the east, in Galatia and beyond (i.e., Cilicia, Syria, the Decapolis, and Arabia). Only now do we have epistolary evidence of him circling back to the huge area lying between these regions, namely, Asia, where these mysterious converts were encountered — and apparently much had happened in the intervening decade. Paul’s mission had no monopoly on conversions, much as he might have wanted one. So rather than being problematic, it is simply somewhat fascinating to see congregations coming suddenly into view from another missionary trajectory, perhaps in combination with some migration (i.e., the origins of the Laodicean congregation need not have been in Laodicea; Christians may have converted elsewhere, possibly in contact with Jewish Christians, and then migrated to this prosperous town).

These three broad prior narratives — Paul’s incarceration, the existence of a successful proxy mission at Colossae, courtesy of the convert Epaphras, and the further existence of two congregations of unknown etiology at Laodicea and Hierapolis — distantly inform the composition of the letter to the Colossians. But the letter suggests that particular events flowing from these broader situations actually elicited its composition. More specifically, one principal event brought two distinguishable issues to Paul’s attention and thereby seems to have catalyzed the composition of Colossians and Philemon; and another immediate sequence of events, which is very important for the subsequent interpretation of Colossians in relation to Ephesians, becomes apparent at this moment as well. We will note these features of the situation briefly here; they will be described in more detail in later subsections, since they spill over into considerations of framing, dependence, and possible pseudonymity (see §§2.2, 2.3, and 2.4).

(i) The arrival of Onesimus with news of Colossae. The key event in immediate terms that catalyzed the composition and dispatch of Colossians was the arrival of Onesimus at Paul’s place of imprisonment. Paul’s letter to the Colossians implies that Onesimus brought information that made Paul concerned about the presence of a new bundle of conceptual, linguistic, and cultural practices at Colossae. In large measure, Colossians presents itself as having been written to correct these.18

The perceived problematic teaching is often dubbed “the Colossian heresy,” but this is slanted and unhelpful; it subliminally suggests the anachronistic origin of this material after the time of Paul when the later church was identifying orthodoxy and heresy more precisely (and even the manner in which this complex development took place is much debated).19 It may be that this teaching does ultimately have to be identified with later heresies, thereby identifying Colossians as pseudonymous. But this question should not be prejudged. In the first instance, then, as usual, we will give the benefit of the doubt to the letter in question and dub this material “the Colossian teaching,” noting that it seems to have included various practices, theological and otherwise.20 We will press this teaching for anachronistic details in our final subsection here (§2.4). For now, we need to ask whether this sort of epistolary response by Paul is plausible, and of course it is.

The problematic material at Colossae might well have been introduced by a third party. The congregation was founded by Paul’s convert, Epaphras. Much of the letter’s rhetoric then resists a perceived deviation from what was initially taught, exhorting its recipients to remain firm and steadfast, and to grow from their initial location (see esp. 2:6-7),21 a location tied to Epaphras and in turn to Paul. Moreover, it occasionally gestures toward the presentation of an alternative point of view by someone else (see esp. 2:4, 8, 16, 18).22 So it seems fair in the first instance to attribute the problematic Colossian teaching tentatively to an interloper, although not much turns on this judgment at this point.23 And Paul wrote almost all of his other letters to congregations that had been affected by third parties in ways that he thought were unhelpful — the perennially important insight of F. C. Baur (2003 [1845]). Of the eight letters we have analyzed thus far, only two do not evidence this function, at least to some degree — 1 Thessalonians and Philemon.24 And this suggests the fundamental plausibility that an incarcerated Paul would write a letter, here Colossians, to correct perceived problems at Colossae introduced by someone else. Moreover, a degree of urgency is now quite possibly imparted to the letter’s exigence; the rot at Colossae would need to be stopped, and any further spread prevented.

But Onesimus’s importance for the composition of Colossians was not exhausted by his report concerning this intrusive teaching. He had not traveled to Paul’s location to inform him of subtle congregational aberrations; this information seems to have arisen inadvertently. He had traveled on a more desperate and less subtle errand — to seek intervention into a difficult domestic situation, the exigence that elicited Philemon, although this aspect of the situation is best addressed momentarily, in §2.2, when framing considerations are explored.

These, then, are the two key issues brought to light by Onesimus’s arrival that need to be appreciated, both of which elicited letters in a partly overlapping response. His information about the new teaching in the Colossian congregation elicited an epistolary correction — Colossians — which was possibly relatively urgent. And his own difficult situation elicited a short epistolary response — Philemon — possibly accompanied by a subtle response in various parts of Colossians, which we will explore shortly. That Onesimus even traveled to Paul, however, bringing him information about the Colossian congregation, was caused ultimately by the proxy mission, which the letter attributes to Epaphras. And Paul’s response in the form of letters, over against a visit, was necessitated by his incarceration.

(ii) The writing and dispatch of a letter to the Laodiceans. One final feature of this complex implicit situation needs to be appreciated before we move on. Paul instructs the Colossians at the end of their letter, in 4:16, to have the letter read to the Laodiceans, and vice versa, to read the Laodiceans’ letter themselves. This is a nice confirmation (if it proves authentic) of our earlier supposition that Paul’s letters were circulating freely to congregations beyond their named recipients.25 Moreover, although the point is not overt, it probably entailed the making of copies; it seems unlikely that the congregations would have exchanged their original copies of these precious letters. When Paul tells the Colossians “to make it so that it [i.e., this letter] should be read in Laodicea,” ancient letter readers would have thought first of making and sending a copy — and this might suggest that the sequence of towns away from Paul’s place of imprisonment began with Colossae. The first town reached would have been responsible for making any initial copies of multiple letters, here apparently Colossae. The Colossians would have had to copy the letter to the Laodiceans for themselves before the original went on to Laodicea, and equally importantly, to make a copy of Colossians to send on as well. Paul’s short letter to Philemon was understandably not involved in this duplication. (We will return to this question momentarily.) But because of this rationale, we cannot determine from Paul’s instruction the order of the letters’ composition — an important realization (to which we will also return shortly). Nevertheless, this telltale instruction is of course interesting not merely because of the light it sheds on the exchange of ancient letters. It reveals that another letter existed in immediate proximity to Colossians and Philemon, written to the Laodiceans, who also had not met Paul personally.

We should note immediately that it is not at all unlikely for Paul to have written a cluster of three letters in prison at this time. Unfortunately, it is well attested through human history that incarceration is a fertile space for reflection and literary work, especially by (literate) activists. This space affords them time like no other to reflect, think, and compose. The result can then be work of a truly definitive nature, provided that the material can be written down in some way and retrieved from the prison. Two famous and more recent examples of such activity are Dietrich Bonhoeffer’s letters and reflections written from prison during the last stages of World War II (collected and published posthumously in Bonhoeffer 2010 [1953]) and the “Letter from Birmingham Jail” written by Martin Luther King Jr. in 1963 (see 2004 [1963], 1896-1908). But these well-known instances barely ruffle the surface of the sea that is prison literature. (A particularly interesting example in this relation is the seventeenth-century Anabaptist anthology The Bloody Theater, collated originally by Van Braght 1951 [1660].)

The existence of this letter to the Laodiceans immediately alongside Colossians opens up a critical interpretative possibility within our developing frame that we will have to assess carefully in due course, namely, whether the letter we now know as Ephesians should occupy this locus — whether Ephesians equals Laodiceans. If Ephesians does not fit this space or, alternatively, if it fits another space better, then we need simply to note that Paul’s letter to the Laodiceans is no longer extant, like the Previous Letter to Corinth and most of the Previous Letter to Philippi, and any inquiries here in relation to Colossians will largely have to stop. But if Ephesians does equal Laodiceans, so to speak, then the implications are potentially significant. Clearly, the whole question will be best considered after Ephesians is introduced into our discussion, in this chapter’s third major section. For now it suffices to note that another letter was written by Paul at the same time as the composition of Colossians and Philemon, a letter addressed to the congregation in Laodicea (2:1; 4:13, 15, 16). And this observation rounds out all the key matters of immediate implied exigence.

The situation surrounding the composition and dispatch of Colossians was clearly layered and complex. But at this stage in our discussion, it is more interesting than problematic. Nothing in the letter’s immediate implied exigence indicates that it was pseudonymous, so we can now turn to consider its location within our developing frame.

2.2 Framing Considerations

We have already used fragments of information derived from the frame when discussing the immediate implied exigence of Colossians, but this letter is so embedded in its local situation, with its complex interwoven narratives and communications, that isolating it entirely is both impossible and undesirable. However, we have yet to engage in detail with Colossians in relation to our broader frame, a task that should begin by exploring the subtle relationship between Colossians and Philemon.

It is perhaps seldom appreciated just how tightly Paul’s letters to Colossae and to Philemon fit together. Most obviously, the co-senders, principal actors, and final greeters are all almost identical. (Possible aberrations in the greetings will be addressed momentarily.) Paul and Timothy are writing to figures identified — if the two letters accompany one another — as Colossian. Archippus is the head of the house congregation of which Philemon, Apphia, and now Onesimus are evidently members (Philemon 1–2); and he is addressed directly in Colossians 4:17. Onesimus is identified as a Colossian in Colossians (4:9), and is clearly returning home in Philemon (i.e., at least in some sense). Aristarchus, Mark, Epaphras, Demas, and Luke, at the least, send their greetings in both letters. So the basic elements in the letters match almost perfectly, although this would perhaps not be surprising in a careful pseudepigraphic work. However, even more impressive is the subtle substantive interlacing of the two letters.

Paul’s letter to the Colossian congregation as a whole, as we have just seen, is principally concerned with the new Colossian teaching, news of this having apparently been brought to Paul somewhat fortuitously by Onesimus. The shorter letter to Philemon (as well as to Apphia, Archippus, and his house congregation) addresses Onesimus’s difficult domestic situation directly. But it is intriguing to note how the longer congregational letter, Colossians, in a secondary rhetorical dynamic, also participates subtly within a broader strategy focused on ameliorating Onesimus’s plight.

The letter to Philemon addresses the situation directly with a number of specific pleas. It also promises a visit from Paul himself as soon as circumstances permit. But Colossians reveals that Tychicus might have been part of Paul’s plan to manage the conflict. There is no suggestion that he was a Colossian convert, although he seems to have been a slave (see σύνδουλος in 4:7). But he was an outsider, a member of Paul’s circle, and presumably literate, and so possibly he would have been able to exert more leverage on various Colossians than one of their own slaves, Epaphras.

More overtly, the letter to Colossians instructs the congregation to “tell” Archippus to complete his “ministry” (4:17: καὶ εἴπατε Ἀρχίππῳ· βλέπε τὴν διακονίαν ἣν παρέλαβες ἐν κυρίῳ, ἵνα αὐτὴν πληροῖς). And this direct address in Colossians 4:17, along with the inclusion of Archippus in the address of Philemon, has often puzzled interpreters. But this is perfectly comprehensible as part of Paul’s broader strategy designed to manage the conflict between Onesimus and his owner, Philemon. Philemon is evidently a member of Archippus’s house congregation; Archippus is his minister. So Colossians and Philemon draw Archippus subtly but firmly into the domestic conflict, placing local congregational and rhetorical pressure on him to “complete his ministry.” Indeed, by 4:17 the letter has carefully constructed an account of ministry in relation to Paul, Epaphras, and (to a lesser extent) Tychicus (see 1:7, 23-25, 28-2:7; 4:7-8, 12-13). Ministers are to be trustworthy, and to exhort, to pray, and even to struggle to present everyone perfect and complete in Christ. So Archippus is drawn remorselessly into a proper ordering of the relationship within his house church between the owner, Philemon, and his slave, Onesimus, and is instructed in how to behave as he carries this out, while the community is simultaneously reminded of his authority and called upon to ensure his faithful action.

Like many NT texts, Colossians provides instructions about community order in terms of the Haustafeln. But it is interesting to note that these instructions are disproportionately concerned in Colossians with the relationship between owners and slaves. The four key family relationships — wives, husbands, children, and parents — receive one brief verse of instruction each. But slaves and masters are instructed for five rather longer verses, which is almost twice as much exhortation as the rest of the Haustafeln combined. By way of comparison, Ephesians provides twenty-one verses of admittedly distinctively expanded exhortation, but only five are directed to owners and slaves. The disproportion is striking, but it is fully comprehensible if Colossians has one eye on the conflict between Philemon and Onesimus. The baptismal instructions in Colossians are then equally suggestive.

Colossians 3:9-11 famously echoes 1 Corinthians 12:13 and Galatians 3:26-28 (although, according to our developing frame, these last texts would not have been composed when Colossians was written). However, the binary couplets transcended in this instance are “pagan [lit. Greek] and Jew, circumcised and uncircumcised, barbarian, Scythian, slave, free.”26 And these have consistently puzzled interpreters. Unlike Galatians 3:28 (but like 1 Cor 12:13), Colossians 3:11 contains no mention of the transcendence of “male and female.” And Colossians adds an otherwise unattested couplet, “barbarian, Scythian,” which is decidedly odd. Its members are opaque, especially the category “Scythian,” which refers literally to a feared nomadic people inhabiting the great Eurasian steppe (although they were, strictly speaking, long extinct by this period, their place taken by other nomadic peoples). Moreover, this does not seem to be arranged in any obvious contrasting opposition as the other couplets are (i.e., Jew versus pagan, slave versus free). Barbarians and Scythians seem to come from the same category (i.e., barbarians). However, these problems are eliminated once it is grasped that the entire series of couplets in Colossians 3:11 has been arranged chiastically, and furthermore, that it thereby speaks directly to the conflict at Colossae between Philemon and Onesimus.

The category “Scythian” was not just a traditional Greco-Roman designation for “the nomadic barbarian other” from the steppes. It was frequently applied to those who had been enslaved from the region, presumably by means of raids along the shores of the Black Sea. Hence, “Scythian” was an attested slave name. (This was probably not just typical slave nomenclature; members of this savage race would have fetched a higher price.) And the largest local emporium for the sale of these slaves was Ephesus. It seems significant, then, that the category “Scythian” is coordinated chiastically by Colossians 3:11 with the category “slave,” and the category “barbarian” with the category “owner” (technically, the word used here is “free,” which would encompass more than just owners; contextually, the use of the word “free” would be especially pointed for owners).

These coordinations map directly onto and further illuminate the situation at Colossae that is principally disclosed by Philemon. Onesimus may have been, it seems, one of the unfortunate “white Russians” enslaved by a Black Sea raid, and was very likely sold in Ephesus or one of the other great port emporiums on the western Aegean coast (although his place of acquisition does not matter overmuch). That he is now associated with Colossae is therefore plausible. His master, Philemon, is identified here as a barbarian, and his wife’s (or sister’s) name attests plausibly to this identification as well; he is probably a Phrygian or Lydian, something that is again not surprising for a citizen of Colossae. The carefully crafted baptismal chiasm in Colossians 3:11 thereby suggests that the difficulties between Philemon and Onesimus have an ethnic as well as a class component, a doubly difficult situation to navigate. So Colossians maps in detail here the unfolding situation addressed directly by Philemon. And it is worth noting that without this illumination, these textual details in Colossians remain opaque; they are literally uninterpretable.

Enough has probably been said by this point to suggest that the problem of slavery and its pastoral management identified by Philemon are woven into the warp and woof of Colossians (although even more material could arguably be introduced in support of this claim). The involvement of the local deacon Archippus in both Philemon and Colossians, the unusual emphasis in the Haustafeln on the relationship between owners and slaves, and the inordinate applicability of the unique baptismal material in 3:11 lace these two letters and their situations tightly together. We might want Paul’s advice concerning slavery to be rather more radical than it is, but that is not our current concern; we need only to recognize here that the issue is writ large through the text of Colossians, if in a somewhat regrettable sense.27 Yet unlike the overlaps between the prescripts and postscripts, this interlacement is not especially noticeable; it is easy even for modern commentators to miss these clues. And this serves to corroborate a judgment of authenticity as against pseudepigraphy concerning Colossians, at least at this stage in our discussion. As Paley put it some time ago: “This result is the effect either of truth which produces consistency without the writer’s thought or care, or of a contexture of forgeries confirming and falling in with one another by a species of fortuity of which I know no example” (1790, 111).

With some confidence, then, we can turn to consider the immediate implications of the interlacement of Colossians and Philemon for the rest of the frame. The first implication worth noting is the reinforcement of the letters’ Asian provenance.

We learned from Philemon primarily that Paul was most probably imprisoned quite close by Onesimus’s place of ownership. Onesimus had reached him as an erro, and part of Paul’s response was to request the preparation of a guest room. There were hints that these locations were in Asia, but they were very tentative (i.e., Apphia’s name; later tradition’s association of Phlm with an Asian provenance by means of Col; and the general suitability of this location against what we know of Paul’s situations elsewhere). We learn from Philemon, in short, something about the distance that Paul was from his recipients but not definitively where those recipients were located.

Paul’s letter to the Colossians, however, clearly identifies its recipients as Colossian. And the tight connections between these two letters now allow these sets of data to be confidently combined, suggesting that Paul was somewhere in the Lycus valley when he wrote these letters, or nearby. That is, we now know with some confidence that he was not far from his recipients, and his recipients were Colossian. And this corroborates our earlier chronological estimates as well, made somewhat tentatively in relation to the hinted Asian provenance of Philemon. So we can now confirm more confidently that the imprisonment in question was during 50 ce, apparently occurring while Paul was returning from Syrian Antioch and Jerusalem by way of Galatia. But Colossians now adds further important details to this developing picture.

Paul was not imprisoned in Laodicea or Hierapolis or Colossae itself. Moreover, he seems to have been positioned east of Colossae and these other communities, looking west toward them, so somewhere most probably on the important route between Laodicea and Pisidian Antioch by way of Apamea, the important town located on the eastern frontier of the Roman province of Asia.

Modern people tend to think about locations and distances and to plan any travel accordingly “from above” (i.e., as if they were in a helicopter or plane), largely because of the cartographical revolution that took place from the Renaissance onward that depicts our environments in this fashion. And most modern scholars tend to think about ancient travel in the same terms — topographically, and in fairly accurate, two-dimensional and three-dimensional terms, a purview doubtless assisted by the provision of accurate maps in modern printed editions of the Bible. But this is of course anachronistic. Modern topography in these terms, based on spatially accurate cartography, became possible only after the discovery in the fifteenth century of accurate surveying techniques based on telescopes, magnetic compasses, and sextants, along with the invention and standardization of various mapping codes that modern readers take for granted — the introduction of the Mercator projection in 1569, for example. The mental landscape of ancients was very different.

It is becoming increasingly clear that Romans and other ancient peoples thought about distance and travel primarily in terms of sequences or itineraries, like beads on a string, that could lead them reliably from their current location to a destination (ironically, rather in the way that programs like Google Maps or MapQuest produce sequenced travel itineraries today, thereby returning us to ancient mental maps; see esp. Scheidel, Meeks, and Weiland 2012; Scheidel 2013). This purview is represented well by the Peutinger Table (i.e., the Tabula Peutingeriana; see esp. Talbert 2010), whose archetype probably dates from the fifth century ce. It is highly significant, then, that several times the postscript of Colossians suggests that Paul is thinking in terms of an underlying itinerary or sequence that runs from east to west.

In 4:13 Paul thinks of Colossae, then Laodicea, then Hierapolis (μαρτυρῶ γὰρ αὐτῷ ὅτι ἔχει πολὺν πόνον ὑπὲρ ὑμῶν καὶ τῶν ἐν Λαοδικείᾳ καὶ τῶν ἐν Ἱεραπόλει). In 4:15 he thinks of Laodicea, then of Nympha’s house congregation, which seems to have been located in Hierapolis (Ἀσπάσασθε τοὺς ἐν Λαοδικείᾳ ἀδελφοὺς καὶ Νύμφαν καὶ τὴν κατ’ οἶκον αὐτῆς ἐκκλησίαν). And in 4:16 he thinks first of Colossae and then of Laodicea, before reversing the sequence (καὶ ὅταν ἀναγνωσθῇ παρ’ ὑμῖν ἡ ἐπιστολή, ποιήσατε ἵνα καὶ ἐν τῇ Λαοδικέων ἐκκλησίᾳ ἀναγνωσθῇ, καὶ τὴν ἐκ Λαοδικείας ἵνα καὶ ὑμεῖς ἀναγνῶτε). And the three primary sequences all run consistently in the same direction — east to west.

If Paul was thinking in terms of itinerary or sequence, then, as was most common in the ancient world, we may infer with considerable confidence that the envisaged itinerary ran west from Paul’s current location to Colossae, then to Laodicea, and finally to Hierapolis, the community westernmost from him. This is not necessarily to suggest that these communities were on a straight line or direct route with respect to one another. But the most practical order in which to visit them from Paul’s place of imprisonment would have been in this sequence, and this strongly suggests locating Paul’s imprisonment to the east.

We should now combine this inference with the implications from Philemon concerning the relatively short distances involved and the additional suggestion that Paul is not apparently writing from a place where a successful congregation had been established. It seems likely in the light of this evidence that Paul was imprisoned somewhere between Colossae and — let us say for the moment — Pisidian Antioch, the gateway to south Galatia, where a successful congregation had been founded (although little changes here if we switch the city of origin to a gateway to north Galatia such as Dorylaion/Dorylaeum).28

Pisidian Antioch was around 6.4 days’ journey on foot from Laodicea (utilizing ORBIS here). And Colossae was less than half a day’s walk from Laodicea back toward the east (ca. 16 km; note that Colossae is too unimportant to be mapped by ORBIS, at least yet). So we may infer fairly that Paul was imprisoned somewhere to the east of Colossae and no more than six days away, and there is one particularly appealing candidate for this imprisonment: Apamea.

The strategic city of Apamea was one of the gateways to Asia from the east and was located only 2.9 days’ walk from Laodicea (105 km), and even closer to Colossae (2.4 days’ walk or ca. 89 km). There are no records of a successful Pauline congregation being established there, although isolated conversions were presumably possible. And it was certainly a sufficiently large administrative center to have had officials willing and able to detain someone like Paul. It seems fair to suppose, then, that Paul was imprisoned in Apamea when he wrote Colossians and Philemon; either this, or he was being held in some similar location and situation between Colossae and Pisidian Antioch (or, alternatively, Pessinus), although the other candidates do not fit the parameters in the data quite as well as Apamea. Dorylaion is a candidate, but is almost ten days’ journey distant, which seems a little far. However, nothing rests on a precise location at this point, so we will with due caution suggest Apamea as the best current candidate for the locus where both Philemon and Colossians were composed, remaining open to other possibilities. We can now turn to consider potential questions within the data that might challenge our initial judgment here of authenticity concerning Colossians:

(i) Is there time during Paul’s imprisonment for all the implied sequences of events to have taken place?

(ii) Is there an unacceptable disjunction between the addressees in Colossians and those in Philemon?

(iii) Is there contradictory information in the two letters’ lists of greetings?

(iv) Does additional data concerning other figures in the Pauline mission contradict the scenario currently being advocated — that is, an early date for these letters’ composition from somewhere in Asia prior to any missionary work in Ephesus taking place?

(v) And why are these letters silent about the collection for the poor in Jerusalem, which was unfolding at this time?

These could pose difficult challenges, but I suggest that there are plausible rejoinders.

(i) The duration of the incarceration during which Colossians was written is limited, at its maximum possible extent, only by the events that took place at either end. Is there time between these events for a relatively lengthy imprisonment during which the events presupposed by Colossians could have occurred? Or is there too much time compression evident here, suggesting that Colossians is a forgery?

As we have already noted briefly, this period is bound on its early end by Paul’s departure from Galatia after traveling back from meetings in Syrian Antioch and Jerusalem, and those took place as early in the relevant chronological interval as late 49 ce (i.e., 13.x years after Paul’s first visit to Jerusalem, which occurred in late 36 ce). The late end of the period is bound by Paul’s arrival in Ephesus in time for the events presupposed by 1 Corinthians to have taken place, and these could have unfolded as late as the final months of 50 ce. This establishes the widest possible interval for the period in question — roughly December 49 through November 50 ce. We must reckon with the appropriate travel times within this interval, however.

A sea journey from Jerusalem to Antioch would have taken about a fortnight, and a journey from Syrian Antioch to Apamea by way of Galatia on foot about 21 days, or three weeks, so approximately five weeks total.29 So if Paul left Jerusalem in January and arrived in Ephesus the following fall, around September, then he could theoretically have been incarcerated in the Lycus valley from March through August, a six-month sojourn (and this estimate is not pressing the limits of the situation). He need not have been in prison this long, but there is clearly enough time for the events implied in Colossians to have unfolded around him during an incarceration of this potential duration. Time compression is not a problem.

(ii) There is possibly a problem, however, buried in the prescripts of Colossians and Philemon. Paul’s letter to Philemon is addressed to Philemon, to his wife Apphia, and to Archippus “and the assembly in your [masc. sg.] house” (Phlm 1–2). Paul’s letter to the Colossians addresses the “holy and trustworthy brothers who are in Colossae” (Col 1:2). Later in the letter, as we have already seen, it also exhorts the letter’s recipients to (paraphrasing) “tell Archippus to get on with his appointed spiritual task” (4:17). And this does not make a lot of sense if the addressees in the prescript of Philemon were identical with the congregation at Colossae.

Yet while this skepticism is warranted in general terms, it must give way in the face of specific data, which seems to suggest here that the entire Colossian congregation was larger than the house church addressed in Philemon. Arguably, fragments of epistolary data confirm this. Paul never addresses the recipients of Colossians as an ἐκκλησία, or assembly. Only assemblies in Laodicea and Hierapolis (the latter presumably being a single house gathering) are named as such. (The references in Col 1:18 and 24 are generalized and distinctive.) The prescript to Philemon also names a single house gathering. This suggests that the Christians in Colossae were gathering in more than one house congregation. And when this data is combined with the probable presence of a rival teacher at Colossae, it seems possible to imagine that conversions at Colossae numbered significantly more than the single household gathering named in Philemon 1b-2. This figure or group may have had some success, too, analogous to the situation at Corinth. But a further problem lurking in the two lists of greetings is more difficult to deal with.

(iii) Epaphras is characterized as a “fellow prisoner” in Philemon 23 (ὁ συναιχμάλωτός μου) but not in Colossians, while Aristarchus is characterized as a “fellow prisoner” in Colossians 4:10 but not in Philemon 24. Consequently, it seems at first glance as if Aristarchus was in prison with Paul when Colossians was written, but Epaphras was in prison with Paul when Philemon was written. And this just looks like a mistake. It makes little sense in practical terms (i.e., the release of one figure and then the arrest of the other). Moreover, the explanations offered for this — where the difficulty is noted — often lack plausibility.

Rapske (1996 [1994], 238), following Kümmel and others, suggests that various figures assisted and stayed with Paul during his house arrest on different occasions and thereby merited the moniker at different times (see Acts 28:16, 30). But first, nothing suggests here that Paul is under house arrest; second, he is not in Rome (as in Acts); and third, such figures are hardly thereby actually imprisoned and so accurately described as fellow prisoners.

Some commentators have argued, rather differently, that the designation is metaphorical (so Wilson 2005, following Moule and ultimately Lightfoot), but this is an implausible reading in its own right. Paul’s sun- compounds elsewhere are never metaphorical but fundamentally concrete. Moreover, a metaphorical meaning is somewhat opaque. In what sense is a Christian coworker a “fellow prisoner” metaphorically? Paul’s gospel speaks of the very opposite, that is, of the liberation of its participants from any slavery or impediment. Nevertheless, the line of interpretation opened up by Lightfoot and Moule is ultimately helpful.

They note that the word used did not, strictly speaking, reference a prisoner in a straightforward sense; it denoted a “prisoner of war,” or POW. (“Prisoner” was denoted by δέσμιος and, more rarely, by δεσμώτης, from “chain” or “bond” or “fetter” [δεσμός]; αἰχμάλωτος, however, denoted someone captured during a conflict.) Hence, Paul’s use of the word αἰχμάλωτος in Colossians 4:10 and Philemon 23 did not necessarily imply a shared imprisonment, as its application in Romans to Junia and Andronicus confirms (Rom 16:7). It denoted, rather, the special status of previous imprisonment, although here described in the apocalyptic terms of a war between the gospel and the evil powers opposing it in the cosmos (this being the truth of the metaphorical reading suggested by Moule). So this signifier functions much as the designation “fellow soldier” does elsewhere for Paul (συστρατιώτης). Such a figure is no longer in the army or serving in a war but is honored by that recollection, the moniker now being interpreted additionally in terms of Christian service (and this is the characterization of Epaphroditus in Phil 2:25, and Archippus in Phlm 2). The result of this realization is that we learn rather less from the data concerning Epaphras and Aristarchus than we might first have thought we did. It seems that they both have been imprisoned at some time because of the gospel. And clearly, no contradictions are now apparent in the texts, since this is quite possible.

However, we do now have to insert an imprisonment into our previous narrative concerning Epaphras, although there are several appropriate locations for this. It is striking that he has been incarcerated already for his new piety, but this is not necessarily impossible. He could have been imprisoned when he converted at Paul’s location. And he could have been detained on his return to Colossae, or on his return to Paul’s location, and both scenarios would explain why he is not accompanying the letter back to his hometown (if we still need to). It seems unlikely that Aristarchus and Epaphras were both actually imprisoned at the time when Colossians and Philemon were written; we would expect Paul to be a little more consistent with his designations if this were the case, and also a little more pointed (i.e., not use the metaphorical reference; Paul never describes his own current incarcerations in these terms). Ultimately, then, I tilt toward an earlier Colossian imprisonment for Epaphras, since this helps to explain his continuing presence with Paul when Colossians was written and dispatched. So this possible inconsistency between Colossians and Philemon seems well explained. And another possible anomaly in the greetings is also worth noting in passing.

The greetings in Colossians mention in 4:11a “Jesus/Joshua who is also called Justus” (καὶ Ἰησοῦς ὁ λεγόμενος Ἰοῦστος), but he seems to have dropped from sight in Philemon (see only Μᾶρκος, Ἀρίσταρχος, Δημᾶς, Λουκᾶς, οἱ συνεργοί μου). Is this another telltale slip by a pseudepigrapher? Not necessarily, if we suppose that Paul used only this person’s first name in Philemon 23b, “Jesus/Joshua,” and that this was subsequently folded by reverent scribes into an extended designation for Christ, thereby erasing the reference to the figure of “Jesus/Justus” (see Ἀσπάζεταί σε Ἐπαφρᾶς ὁ συναιχμάλωτός μου ἐν Χριστῷ Ἰησοῦ). Any emendation here is conjectural, being unattested in the manuscript tradition, but seems highly plausible (i.e., Ἀσπάζεταί σε Ἐπαφρᾶς ὁ συναιχμάλωτός μου ἐν Χριστῷ, Ἰησοῦς κ.τ.λ.), a view that Raymond Brown (1997, 614 n. 31) notes in passing. This conflation would have been assisted by the μου just three words previous, which might have suggested the genitive rather than the nominative ending for Ἰησοῦς. In addition, the characterization of Epaphras in Colossians 4:12 varies, with the longer reading — “of Christ Jesus” — possibly reflecting the emendation of Philemon 23b. Alternatively, if original, it might have prompted or reinforced that emendation. So this further potential anomaly seems explicable. But other data might still cause trouble for our developing reading, which roots Colossians and Philemon in the early stages of an Asian mission.

(iv) Romans 16:5 suggests that Epaenetus was Paul’s first convert — literally, “first fruits” — in Asia (see 1 Cor 16:15). Moreover, we know of Epaenetus only in conjunction with Prisca and Aquila. He seems to be part of their house gathering at Rome when Paul writes Romans in the spring of 52 ce (16:3-4). And it seems plausible to infer from this data that Epaenetus was converted through some connection with Prisca and Aquila and so most probably at Ephesus (1 Cor 16:19). But this would preclude earlier conversions of Epaphras and Tychicus in Asia, as supposed by Colossians and our developing reconstruction (later tradition identifying Tychicus with Asia, almost certainly unexceptionally; see Acts 20:4). That Paul would have been speaking in local, regional rather than broader, provincial terms about Epaenetus’s Asian origin, thereby avoiding this conundrum, also seems unlikely, given his usage in 1 Corinthians 16:19.

But there is insufficient data here to build a decisive case. Epaenetus might have been part of Prisca and Aquila’s house gathering in Rome and a product, earlier on, of their work in Ephesus, but he could have been living with them in Rome simply as a Pauline convert enjoying their hospitality. Or he might simply have been named in Romans immediately after them by way of association with the Asian mission and not have been staying with them at all. We should infer that he was converted prior to Epaphras in Asia, although he no longer seems to have been with Paul when Colossians was written. But Apamea was in Asia, along with any settlement to the west of it, so he could have converted in one of those locations, where Paul was imprisoned. And his presence in Rome as attested by Romans 16:5 attests directly to his mobility. Tychicus might have converted subsequent to Epaenetus in Asia as well, or he might have converted somewhere else, outside Asia. So resolving this challenge thickens our description concerning the early Asian mission helpfully, if we add the conversion of Epaenetus to Paul’s current incarceration in addition to the conversion of the Colossian slave Epaphras, rather than undermining its plausibility. Similar considerations hold for Tychicus. Moreover, the existence of these congregations in the Lycus valley helps us explain Paul’s otherwise somewhat opaque plural in 1 Corinthians 16:19, written the following spring: Ἀσπάζονται ὑμᾶς αἱ ἐκκλησίαι τῆς Ἀσίας. And at this point one last query remains.

(v) Both Colossians and Philemon are silent about the collection for Jerusalem, a major project that Paul is currently administering. According to our developing scenario, he has just instructed the Galatians about its prosecution (during a visit) and will shortly engage in a long and complex negotiation with the Corinthians about the same, will gesture toward it subtly in his letter to the Galatians, and will mention it at its conclusion in Romans. Does Paul’s silence about this important undertaking in Colossians and Philemon suggest that they were not written at this time?

This inference is doubtful. Paul’s letter to the Philippians attests to the fact that he does not always refer to the collection during this period. Its mention is helpful — if not critical — for sequencing when it does take place, but the converse does not apply. That is, it does not follow from the silence of a letter concerning the collection that it was not composed during this period. Moreover, that Paul would resist introducing a request for money into letters written to gatherings who had not yet set eyes on him is entirely understandable. The ancient world was full of charlatans (see Anderson 1994); hence, one of Paul’s most constant struggles was the battle for perceived financial integrity, a struggle discernible from his earliest correspondence to the Thessalonians (1 Thess 2:3, 5). So the silence concerning the collection in Colossians and Philemon is not only not a difficulty; in these circumstances, it is to be expected.

It seems, then, that any difficulties arising as we insert Colossians into our developing frame alongside Philemon are more apparent than real. They are worth exploring, and some, in their resolution, have deepened our knowledge of the original situation, but none are fatal. And as a result, our earlier judgment that Colossians was authentic has been strengthened; a web of overt and subtle connections ties it into our existing frame, which should now be augmented by its confident insertion (that is, pending the outcome of later challenges).

The existing principal sequence of letters, running through Paul’s (calendar) year of crisis, can now be expanded with the addition of Philemon and Colossians in the previous year:

// 50 //
. . . Phlm / Col . . . [PLC] —
// 51 //
1 Cor — 2 Cor — [PLP]Phil 3:2–4:3 / Gal — Phil —
// 52 //
Rom . . .

With this useful expansion of the biographical frame tentatively in place, we can turn to consider some of the weightier methodological challenges to the authenticity of Colossians — in terms of dependence and style.

2.3 Questions of Dependence and Style

(1) Dependence

A specific and quite important question of dependence clearly arises here, namely, the relationship between Colossians and the prior letter to the Laodiceans (Laod). Not much turns on this if Laodiceans is simply lost; however, if Ephesians is identified with Laodiceans, then the situation waxes in importance. In fact, a matrix of possible results exists for this relationship. But only one of these can potentially alter our developing judgment here — and this presupposes that dependence in one direction can be definitively demonstrated.30 If this option is eventually affirmed — that Ephesians demonstrably precedes Colossians and is inauthentic — then we will return and revise our judgment concerning Colossians. But in view of its unlikelihood, we can safely postpone consideration of this issue until the letter to the Ephesians is assessed in detail in §3. All the other options leave the question of Colossians’ authenticity open.

	Ephesians authentic	Ephesians inauthentic
Ephesians precedes Colossians	Colossians could be authentic	Colossians inauthentic
Ephesians follows Colossians	Colossians authentic	Colossians could be authentic

But this still leaves other possible instances of dependence in Colossians unaccounted for.

Recently, Leppä (2003), looking back to a classic analysis by E. P. Sanders (1966), has argued that Colossians is inauthentic, because of its overt dependence on various indubitably authentic Pauline letters — a significant suggestion, because it is made independently of any judgments concerning Ephesians.31 But we have already entered a strong objection concerning this type of inference, and must introduce a caveat to Leppä’s argument as well.

Paul evidently kept copies of his letters; the letters circulated through multiple congregations (see Col 4:16); and copies seem to have been made and studied in those locations. So he seems to have echoed his letters in various ways within further correspondence when he felt the need to do so.32 Second Thessalonians 3:8 provides an especially good example of a small specific echo when it quotes but also pointedly reapplies 1 Thessalonians 2:9 (2 Thess is discussed in Leppä 2006; and here in ch. 4), and we have seen the presence of broad structural similarities between 1 Thessalonians and 1 Corinthians as well (also in ch. 4). Hence, the mere fact of dependence has already been shown to be unremarkable, as Sanders in effect concedes. So this aspect of Leppä’s case can actually be granted, but it suggests nothing further in terms of authenticity — the false inference in her analysis (i.e., dependence necessitates pseudonymity). Indeed, this was our principal objection to this line of reasoning earlier. But a caveat ought to be entered here as well.

Leppä’s data is indecisive even with respect to sequencing Colossians, since all of the putative echoes that she assembles can be explained plausibly in the reverse direction, as the echoes of Colossians by later Pauline letters. That is, echoes of undisputed letters appearing overtly later in Colossians would break apart our existing frame, pushing Colossians later in the broader letter sequence at the least — an important result. But Leppä never demonstrates that the echoes must proceed in this single direction, from another letter to Colossians, thereby falsifying our frame. Indeed, demonstrating direction or causality is a common problem in the analysis of perceived dependence. Leppä assumes the inauthenticity of Colossians and so does not feel the need to address this problem of reversible dependence, but it is a critical question for our framing project.

Having said all this, it is important not to overlook the hurdle that advocates of authenticity face in this relation, namely, the supply of plausible contingent accounts of any identified echoes, something that was arguably done here on a small scale for 1 and 2 Thessalonians, and on a larger scale for the echo of 1 Thessalonians by 1 Corinthians.33 But arguably good contingent accounts can be supplied of the echoes that Leppä and others have detected between Colossians and the undisputed Pauline letters.

So, for example, Sanders argues that

(i) Second Corinthians 4:4, Romans 1:20, 1 Corinthians 8:5-6, and Romans 11:36 are echoed in Colossians 1:15-16; but Sanders’s second instance here (Rom 1:20) is forced, and the others are arguably understandable echoes of an important preexisting hymn or confession;

(ii) Second Corinthians 5:18, 1 Corinthians 8:5, and Romans 5:10 are echoed in Colossians 1:20-22a; but the front end of this set of echoes in Colossians is again explained plausibly by the hymnic origin of that material, and the second is a plausible reprise of a key Pauline theological narrative — the reconciliation of a hostile humanity (and Sanders underplays the presence of this discourse elsewhere in Paul, omitting other parallels in 2 Cor 5:16–6:1);

(iii) First Corinthians 2:7, Romans 16:25-26, and Romans 9:23-24 are echoed in Colossians 1:26-27; but this appeal to “the revelation schema” informed by Jewish Wisdom traditions is again arguably unremarkable for Paul (i.e., because it is so important substantively), and any leverage from the doubtful Romans 16:25-26 is questionable; and

(iv) Romans 6:4, Romans 4:24, Galatians 1:1, Romans 6:11, and Romans 8:32 are echoed in Colossians 2:12-13; but Romans 8:32 is a false parallel, and Galatians 1:1 vestigial, leaving overtly traditional verses (Rom 4:24) and material (the Rom 6 account of baptism) in close relation to a baptismal discussion in Colossians 2.

In sum, Sanders’s echoes are less impressive than they first appear, and arguably, they combine to tell us only what theological motifs are important to Paul — useful information in another setting, to be sure, but not here.34 Hence, we will set this interesting challenge to one side for now and proceed to the next challenge to the authenticity of Colossians, namely, its much-touted difference from the authentic letters in terms of style. Of course, some of these questions of dependence will have to be revisited in earnest during the assessment of Ephesians, when we will have to consider the significance of the virtual web of intertextuality that holds between Ephesians and Colossians, and with the undisputed letters in turn, and will have to at least try to determine the flow of the obvious dependence in view. But that discussion can wait.

(2) Style

The stylistic challenge to the authenticity of Colossians began as early as 1838, when it was one important aspect of Ernst Mayerhoff’s case for the letter’s pseudonymity. And it reached an apparent climax with the soul-destroying tabulations of Walter Bujard in 1973. Thus, Ehrman cites Bujard’s analysis and comments that it is “widely thought to be unanswerable” (2013, 175). But is it?

Bujard’s data is impressive. However, it needs to be measured against the statistical and (where relevant) computer-assisted evaluations of style that were first noted in chapter 4. And we should recall the initial methodological observation made there — that mere tabulations of difference are, when uninterpreted, largely meaningless. All of Paul’s undisputed letters contain stylistic differences from one another, and “style” itself is an immensely complicated and largely undefined notion — a combination of dozens, if not of hundreds, of different features of the language used in a given set of texts. So interpreters of Paul need to be very careful, first, not to “cherry-pick” the data, tabulating variables that indicate differences between undisputed and disputed letters, thereby omitting those variables that either are unremarkable in a disputed text or that suggest the oddity of an undisputed text. And second, they need to avoid assuming blithely that an instance of difference is an automatic marker of alternative authorship, developing pseudonymous readings in the light of this conviction.

Both problems are apparent in Bujard’s analysis and in those like it. Hence, we need to ask, are tabulations of difference like his helpful at all? Is style a meaningful element in this debate? And we need to turn in response to those who know how to evaluate the meaningfulness of differences appropriately, something that ought to take place in the standard statistical sense, that is, in terms of distances from the mean for given variables in relation to the measurable standard deviation of a sample. And I would suggest that Kenny’s older analysis (1986, esp. 27, and 117-18) remains helpful in this regard.

It seems quite important that after tabulating 96 stylistic features and then correlating them, in order to assess their significance, Kenny concludes that the author of the Pauline corpus was extremely versatile stylistically. Without this observation, his and most other tabulations simply collapse into absurdity (i.e., the undisputed letters prove incommensurate). Once this versatility has been acknowledged in stylistic terms, only Titus “must be under suspicion” (1986, 95-100, quoting from 97). The sum of the coefficients (although he notes that they are not, strictly speaking, “additive”) also generates an indicative ranking of the letters in terms of their fit with one another, that is, in terms of cumulative differences. So, in descending order of “comfort”: Romans, Philippians, 2 Timothy, 2 Corinthians, Galatians, 2 Thessalonians, 1 Thessalonians, Colossians, Ephesians, 1 Timothy, Philemon, 1 Corinthians, Titus. In addition to the difficulties of Titus, the relative comfort of 2 Timothy and discomfort of 1 Corinthians here are noteworthy, not to mention the locations of Colossians and Ephesians as more “comfortable” than 1 Corinthians and Philemon.

In short, the calculations of Kenny directly contradict the significance (not the fact) of Bujard’s tabulations, and of those like them. The latter are unreliable indicators of significant differences between the style of Colossians and the stylistic characteristics of the rest of the Pauline corpus. Can we go further than this, however, drawing especially on the more recent computer-assisted evaluations of stylistic data in Paul to generate some more aggressive insights about authorship?

I would suggest — tentatively — that at present we cannot, although we should not rule out the possibility that more refined methods and experiments will deliver more decisive results in the future. (Indeed, I expect that they will.) But the most sophisticated analyses we currently possess — Neumann, Mealand, and Ledger — in my view lack decisiveness. Although they do seem to be able to discriminate successfully between significant blocks of material within the NT (like the Gospels, Acts, the book of Revelation, and the letters; see Ledger 1995, figs. 3 and 4, pp. 88 and 89), when they turn up the focus, so to speak, and generate plots of the Pauline letters alone, the resulting results lose plausibility (see figs. 5, 6, and 7, pp. 90, 91, and 92; also Mealand 1995, fig. 8, p. 83; and the plots below, unpublished but which Mealand kindly shared with me in personal correspondence). It becomes very difficult to detect whether the plots generated are supplying meaningful information or mere noise. So, for example, Ledger concluded rather confidently that 1 Thessalonians is doubtfully authentic, because it plots close to Ephesians, Colossians, and, to a certain degree, the Pastorals. But this is unlikely to be a meaningful result. Rather, it calls his plot into question. And the plots for Romans and 1 Corinthians are as widely spread in Mealand’s analyses as the plots for Ephesians, while only the sample drawn from the first half of Ephesians tends to be an outlier.35

I suggest, then, that we must wait for the next generation of stylometric analyses for more decisive results at the front end of the framing process — and perhaps this is not surprising. Stylometrics has always had more success identifying an author if it already knows that one of two or more authors is responsible for the text in question.36 And this indicates the correct location for this method within framing in terms of its more positive suggestions — later on, as a corroborative contention. That is, if the pseudonymity of any Pauline letters can first be established with some confidence on traditional grounds, such as demonstrable anachronism, then stylometric analyses can be introduced under this control and potentially corroborate the judgment. But this analysis cannot be pursued yet. It is a far more difficult business to determine whether multiple authors are present within a given sample, especially given that the sample size is not particularly large (and this is exacerbated in the case of Paul, because he is not working in English, or even in a Romance language, the language(s) on which most computer-assisted analyses of style are premised).

However, the negative result of this brief detour into stylometrics is still significant for our current question, as it was in chapter 4. If a statistical analysis of the data like Kenny’s can detect almost no meaningful variation within the extant Pauline sample, then analyses like Bujard’s are revealed to be less decisive than they purport to be. Moreover, if the most sophisticated computer-assisted analyses of multivariate data arrays cannot yet achieve consensual mapping of differences beyond this, positive claims by NT scholars like Bujard on the basis of mere tabulations of particular variations are doubly judged to be insignificant. Stylometrics suggests, in short, that many arguments from style against authenticity in the past have been considerably overcooked. It seems, conversely, that we must relearn how to interpret stylistic differences in Pauline texts when framing.

Instead of allowing a mental organization of the data of difference alone to reinforce an underlying hypothesis of pseudonymity for a given text, we must learn to resist this siren’s call. We must, rather, treat the differences in style apparent between Pauline texts in the first instance as possible evidence of spread within the authentic Pauline sample, and as indicative of contingency.37 And in terms of our current question — the challenge to the authenticity of Colossians on grounds of perceived differences between its style and those of the undisputed letters — we must simply say that these differences exist (the truth of Bujard’s analysis and of those like it) but that they are not demonstrably meaningful in terms of authorship. In basic statistical terms, there is currently no evidence that these exceed the parameters of a normal single sample, lying so far from the mean in terms of the sample’s standard deviations that another author has to be posited.38 Instead, these differences should be explored vigorously in due course for what they can tell us about the situation to which Paul was specifically responding. In short, differences in style speak initially of particularity and not of pseudonymity. And with this decision made, and the stylistic challenge to authenticity abandoned — at least for the moment — we should turn to consider the challenge to the authenticity of Colossians in terms of substance.

2.4 Substantive Challenges

One of the most important indicators of pseudonymity is the presence of small mistakes in a text — telltale signs that a pseudepigrapher cannot sustain the depiction of a supposedly original contingent situation that has in fact been long lost to view. But we have already noted the most important candidates in Colossians in this respect — apparent inconsistencies between the greetings in Philemon and Colossians and so on — and found them all to be plausibly explicable in original terms. The case for pseudonymity must consequently turn to larger problems of possible anachronism if it is to prove successful. And the interpretation of Colossians has a long history in this regard. Throughout the modern period, scholars have opined that the teaching it combats is later than the time of Paul. We need now to consider the probity of this contention in a little more detail.

(1) The Problematic Teaching at Colossae

For much of the early modern period, that is, the nineteenth century, scholars viewed the teaching at Colossae as a variant on Gnosticism and hence the letter’s situation as necessarily later than the time of Paul. And this attack may have been abetted in the case of Protestant scholars by the perception that Colossians was overly concerned with the church and hence unacceptably close to Catholicism. Gnosticism was known during this period principally through the mirror reading of patristic figures like Irenaeus and so was thought to have flourished in the second century ce and beyond as a particular Christian heresy. But its evident fondness for the language of “emanation” and “overflow” resonated with use of the same by Colossians (see πλήρωμα in 1:19; 2:9;39 and πληρόω in 1:9, 25; 2:10; 4:1740), suggesting a late and possibly even second-century date for the latter.

However, the Gnostic reading of Colossians has since been overtaken by historical events — principally, the gradual dissemination since their discovery in 1945 of the Nag Hammadi texts. Gnosticism is now viewed as a rather earlier phenomenon than was first thought and is no longer isolated within the Christian tradition, being viewed as interacting with Judaism extensively as well. This has undermined any case for the pseudonymity of Colossians on the grounds of possible resonances with Gnosticism. Indeed, it has facilitated a Gnostic reading of Paul (see Pagels’s well-known but marginal assertions in 1992 [1975]).

The classic approach to the interpretation of the Colossian data is far from exhausting its anachronistic interpretation, however. More recently, scholars have recognized the presence of the language of emanation within the Platonic tradition, leading to readings of Colossians in terms of Middle Platonic teaching that was again often assumed to be late. This reading of Colossians has enjoyed greater prominence in recent years, thanks in large measure to the work of Eduard Schweizer (1982 [1976]; 1988), whose approach to Colossians is ably advocated by DeMaris (1994). But the importance of Philo for the development of Middle Platonism directly undermines any appeals to this reading of Colossians as evidence of Pauline anachronism.41 (It might still be the correct construal of the data, but simply from the time of Paul.) In a similar vein, Troy Martin (1996) advocated a Cynic construal of the opponents, with no necessary implications for authorship.

However, Sappington (1991) advocated another major approach, namely, a construal in terms of Jewish apocalyptic, repristinating in his thesis the important insights of Fred Francis (1962; 1967; 1975a; 1975b; see now also esp. Stettler 2005 and I. Smith 2006). Here the rediscovery of Jewish apocalyptic texts during roughly the same period that the Gnostics were being reappraised in the wake of Nag Hammadi led to a new appreciation of the importance of angels within late Second Temple Judaism (see, i.a., Stuckenbruck 1995, summarized usefully in Hurtado 2009). And this opened up a significant new angle on the critical data in Colossians 2:18, where the text had been thought in the past to speak — rather problematically — of the “worship of angels” at Colossae (i.e., an objective genitive construal). It also seemed to speak of some sort of initiation or mystery rite on the basis of the text’s disparaging reference to “what he has seen, by way of initiation,” this being the preferred translation of the rare participle ἐμβατεύων (μηδεὶς ὑμᾶς καταβραβευέτω θέλων ἐν ταπεινοφροσύνῃ καὶ θρησκείᾳ τῶν ἀγγέλων, ἃ ἑόρακεν ἐμβατεύων, εἰκῇ φυσιούμενος ὑπὸ τοῦ νοὸς τῆς σαρκὸς αὐτοῦ). But a greater sensitivity to the concerns of Jewish apocalyptic allowed the construal of this data as a reference to “worship by [or like] the angels” (that is, a subjective or characterizing understanding of the genitive), and to “what he has seen by way of heavenly visions [i.e., after a journey or ascent, or a revelation],” both of these notions being recognizable motifs within apocalyptic strands in Judaism in Paul’s day.

Furthermore, we must add to these two key modern approaches the opinion of Arnold (1996 [1995]; see also 1989; 1992), who persuasively advocated an analysis of the problem at Colossae in terms of local syncretism, at least to some degree, informed by street religion or magic.

Fortunately, we do not need to resolve this fascinating but complex debate here for the purposes of framing (although my money is on the Jewish apocalyptic reading, perhaps including traces of Judaized Middle Platonism). It suffices to note that of the four major candidates within past NT scholarship for the construction of the problematic Colossian teaching noted here, not one is necessarily post-Pauline and thereby demonstrably anachronistic.42 Gnostic, Middle Platonic (or even Cynic), Jewish apocalyptic, and local syncretistic readings of this data are all possible during the mid-first century ce. So any pressure on Pauline authorship from this teaching’s infiltration to Colossae can be seen to have largely dissolved. And perhaps for this reason, a later case for the letter’s pseudonymity such as Ehrman’s (2013, 171-82) barely mentions this significant element within the earlier tradition in relation to claims of pseudonymity. Ehrman advocates instead the implausibility of the theological case by the author of Colossians, not the inappropriateness of the Colossian heretics themselves. We need now to consider the cogency of this type of argument.

(2) The Implausibility of “Paul’s” Theology in Colossians

The case against the authenticity of Colossians made in substantive terms tends to appeal to its incompatible eschatology, as well as to other substantive elements that are ostensibly discordant with what Paul says elsewhere. In essence, the Paul of the undisputed letters could not have said the things that we find in Colossians. I would suggest, however, that few substantive incursions into questions of preliminary framing are more risky and less helpful than “incompatibilist” contentions in terms of Paul’s eschatology. An entire series of problems is detectable within this line of argument. However, there is no need to flesh these out fully. A brief discussion should suffice.

(i) A major initial objection in methodological terms is sufficient to call the entire contention into question. Preliminary framing cannot appeal to substantive configurations of Paul’s thinking, because we do not know at this early methodological stage what letters that thinking should be reconstructed from or what shape that thinking takes. We do not yet know what Paul’s position on eschatology — or any other substantive matter — was in full or whether his position, along with his thinking and reasoning more generally, was coherent. Hence, any appeal seeking to exclude a letter from the authentic sample in terms of evident consistency begs the key questions concerning Paul’s substantive concerns during framing, in terms of both content and nature. Put slightly differently: we need the data of Colossians, if it is authentic, in order to complete a full (extant) picture of Paul’s thinking on any given topic. If we exclude it on the basis of a supposed reconstruction of his thinking that appeals only to other letters, we risk introducing a vicious circularity into the heart of our description — a description that it will be impossible to correct and a circularity from which it will be impossible to recover.

This alone is sufficient to rebut Ehrman’s contention at this point, along with any like it. But in view of their popularity, some further objections to this line of argument are probably worth noting, although we will have to trace Ehrman’s case out in a little more detail before this can be done.

Ehrman’s case against the eschatology apparent in Colossians begins with the judgments that the authentic Paul was a fairly strict “futurist” and “imminentist.” That is, the apostle was oriented primarily by the future coming of Christ, which he thought would be very soon, and this led to the composition of texts like 1 Thessalonians, with its strong sense of the Lord Jesus’ parousia, and more broadly, to the urgent sense of mission apparent in its background. This eschatological reconstruction is complemented, however — as it must be — by the claim that Paul’s soteriology and ethics were rooted in something of an intermediate state: Christian believers have died with Christ (so, e.g., Rom 6:1-11), thereby moving beyond the flesh, but have not yet been raised with him in any strict sense. Paul viewed believers as dead but not yet resurrected. Ehrman buttresses this claim by a reconstruction of the Corinthian situation in terms of an overenthusiastic, overrealized eschatology, which Paul corrected in 1 Corinthians 15, this last text being mirror read in this relation. Paul’s resistance to overrealization at Corinth thereby seems to bear out the soteriological and ethical account of Romans in terms of an intermediate soteriological and ethical state. And Ehrman suggests that the soteriology, ethics, and eschatology in Colossians are too different to be plausibly from the same thinker.

The letter to the Colossians emphasizes the importance of the ascension of Christ for these theological topoi, and hence of the presence of the risen Christ in some vigorous and real sense within believers. And this incompatible emphasis on the resurrected presence of Christ is said to subtract from the futurism and imminence of Christ’s parousia; those notions are evidently diluted. Hence, two completely different eschatological and soteriological views are apparent, one in the undisputed letters and one in Colossians, leading to a plausible judgment of pseudonymity for the latter, more developed view, and for Colossians itself.

What should we make of this case, in addition to the basic methodological problems previously noted?

(ii) I would suggest briefly that we do not need to know definitively what Paul’s eschatology, soteriology, and ethic were — a full discussion of which would require respective monographs (after framing) — to be suspicious that the previous configuration of the data is too selective and somewhat oversimplified. Its omissions and simplistic assertions create a stark picture of difference and apparent incompatibility between the two sets of texts. More accurate presentations of the specific data, however, suggest a more complex set of views in Paul that overlap significantly across all the letters currently in play. A more nuanced reading of the data suggests, in short, that there is little or no evident incompatibility. The resurrected and ascended Christ is present in the lives of believers in Paul’s undisputed letters (a view supported by a reasonable construal of the Corinthian data); and the future, coming Christ is present in Colossians. Any further differences between the letters can then be plausibly ascribed to contingency.

Excursus

The risen Christ is present in the two major undisputed letters to which Ehrman appeals for Paul’s futurist eschatological views — Romans and 1 Corinthians. The ascended Christ, who is seated at the right hand of God, intercedes for suffering believers in Romans 8:34, a claim that develops the analysis of divine prayer for believers and creation in 8:22-23 and undergirds the famous claim in vv. 28-31 and vv. 36-39 that the divine benevolence is invincible. This location and its associated roles are noted at both the beginning and the end of the letter as well, in 1:4 and 14:10-12. And this last text reads like a cryptic summary of the ascended Christ’s role in 1 Corinthians 15:24-28 — which resonates strongly in turn with Philippians 2:10-11 — in the final judgment and triumph of God (to which Rom 16:20a probably alludes as well).

But this living presence “on high” is complemented by an equally astonishing claim made throughout Romans and 1 Corinthians that Christians somehow live within this presence. Christians are, according to Paul, simply “in Christ” (a situation that can be spoken of reversibly; see Gal 2:20). So baptism, appropriately interpreted, speaks of the manner in which Christians have died and been resurrected in Christ (see Rom 6:1-11; 1 Cor 12:13; Gal 3:26-28). They are now clothed with Christ (Gal 3:27), and hence beyond the categories that might seem to structure their lives in any determinative fashion (Gal 3:28). Most importantly, it is this location that allows them to live beyond the flesh, which has been executed, and in the power of the Spirit (see also Rom 8:1-14), thereby undergirding Paul’s controversial claims that the Christian ethic is beyond Torah.

Any suggestion, then, that Paul understands believers to be dead but not yet risen, and so living in some intermediate state, is misguided. This is not to suggest that everything has been resolved and Christians have been perfected. Clearly, Paul thinks that much has yet to happen. But his conceptualization of this broader situation is something we will address momentarily. Suffice it for now to note that if Christians are not presently living in some concrete sense within the resurrected and ascended Christ — most obviously, by way of the Spirit — then all of Paul’s central theological claims collapse.

Complementing this realization concerning the present force of the resurrection and ascension in the undisputed letters, we ought to note that any claims concerning Paul’s theology are dubiously buttressed by strong claims concerning the error at Corinth he combats in 1 Corinthians 15. The Corinthian situation was inordinately complex, and its reconstruction is fraught with difficulty — the limitations in the data, the perils of mirror reading, and so on.43 It is rather more plausible to proceed from Paul’s argument (here in 1 Cor 15) to the Corinthian error rather than vice versa. And that argument again seems premised significantly on the living Christ. Admittedly, some of its orientation is future. But earlier arguments in the letter suggest that the future grammar of this discussion is often argumentative, not temporal (see esp. vv. 22 and 49). So, for example, Paul’s anguish over the Corinthian use of prostitutes in 6:9-19 is premised not just on future accountability but on the present unity of the Corinthians’ bodies with Christ’s by way of the Spirit (see also 3:16-17). This understanding of bodiliness is then extended significantly in chapters 11 and 12 when Paul addresses abuses within the worship service. In the midst of this extended and subtle set of contentions, the present unity between Christians as members of the body of Christ is entirely concrete; if it is not, Paul’s arguments (again) collapse. This all suggests that 1 Corinthians is deeply committed, at least in some sense, to the presence of the resurrected Christ within the lives of believers.

And with these realizations concerning the data in Romans and 1 Corinthians (not to mention in parts of Philippians and Galatians), the arguments of Colossians no longer look so incompatible. Rather, they seem to be contingent variations on much the same set of basic convictions concerning eschatology, soteriology, and ethics.

The resurrected and ascended Christ is obviously important in Colossians (1:18; 3:1), as Ehrman notes, but equally clearly, Christians presently die and rise in him; it is this that leads to their freedom from the need for bodily circumcision in 2:11-13 (see esp. 2:12a: συνταφέντες αὐτῷ ἐν τῷ βαπτισμῷ). And this looks like (i.a.) the depiction of baptism in Romans 6:1-11 and the arguments against prostitution and liturgical disruption in 1 Corinthians 6 and 11–12. Still more significantly, the baptized are told in Colossians 3:2-4 to wait in some sense for their perfection ([2] τὰ ἄνω φρονεῖτε, μὴ τὰ ἐπὶ τῆς γῆς [3] ἀπεθάνετε γὰρ καὶ ἡ ζωὴ ὑμῶν κέκρυπται σὺν τῷ Χριστῷ ἐν τῷ θεῷ [4] ὅταν ὁ Χριστὸς φανερωθῇ, ἡ ζωὴ ὑμῶν, τότε καὶ ὑμεῖς σὺν αὐτῷ φανερωθήσεσθε ἐν δόξῃ). So Colossians is still to some degree “futurist.” And it is only fair to note that the question of imminence is never addressed in Colossians — the Colossians apparently never having asked Paul for instructions concerning the timing of the parousia. So we might ask whether the eschatological scenario adumbrated in part by 1 and 2 Thessalonians is incompatible with anything said in Colossians. Could the Colossians not have held those views without difficulty while adhering to everything that was taught in this letter? It seems possible.

Thus, it is apparent that the actual configurations of the data from various Pauline texts used by Ehrman to create a case for the pseudonymity of Colossians on the basis of conceptual incompatibility are dubious (or at least questionable). A more careful reading of the argumentation in parts of Romans, 1 Corinthians, and Colossians, assisted by relevant fragments from Philippians and Galatians, suggests that these texts all plausibly draw from the same basic conceptual configuration, differing from one another only by way of circumstantial emphasis. And Ehrman’s case for pseudonymity in these terms consequently deflates. Even granting the validity of an argument in terms of conceptual incompatibility, none is really apparent. Indeed, if anything, the plausibility of including Colossians within the authentic Pauline sample seems enhanced.

Before moving on, one final observation on this type of case might be useful.

(iii) Ehrman’s case concerning different ostensible soteriologies and eschatologies in both the undisputed Paul and Colossians rests in part on the application of zero-sum relationships to salvation in terms of being and time. And these assumptions are worth querying.

That is, Ehrman assumes that the relationship within Christians between a present, difficult state, which Paul refers to as “in Adam,” and a liberated, joyful state, which Paul refers to as “in Christ,” operates in a zero-sum, or essentially quantitative, manner. And in the case of quantities, people can experience more of one only as they experience less of the other. To be in Adam is also, to the equivalent inverse degree, to be lacking Christ, and to be in Christ is to be, to the same inverse proportion, lacking and beyond Adam. Hence, it is almost as if non-Christians are 100 percent Adamic. When Christ comes and converts them, a part of them is now located in him, but because they are still Adamic to some degree, they still lack Christ to some degree. So we could speak of Christians being, say, 50 percent still Adamic and 50 percent in Christ. This is apparently helpful, because the present weakness and difficulties of Christians can be understood in terms of the Adamic material that lingers. And future expectations of Paul can be understood in terms of the elimination of the remaining Adamic elements and the arrival of Christ in fullness, at which point Christians will presumably be 100 percent in Christ. Everything Adamic will have been eliminated. (There is a further assumption operative here in terms of time, but addressing the ontological conceptualization in terms of space and quantity will be more than enough to deal with for now.)44

It is in part this conceptuality that allows Ehrman and others arguing as he does to play certain claims in the Pauline texts off against one another.

An initial emphasis on texts proclaiming the future coming of Christ and his resolution at that time of any cosmic struggles and problems suggests in quantitative terms that Christians must presently be living in a mixed condition characterized by the ongoing presence of a significant percentage of Adamic material, so to speak. If fullness and perfection lie ahead, then they cannot be present now. And any ongoing emphases on sin or struggles would seem to affirm this view. So texts drawn from any letters emphasizing the perfections inaugurated by the parousia will set up “mixed” expectations concerning the present state of Christians.

The citation of fulfillment texts from other letters will then create an apparent problem. Any emphases on the effectiveness of the resurrected and living Christ within the lives of believers in the present will tilt the perceived quantities in a stronger Christian direction. Instead of a strongly mixed condition, then, perhaps 50 percent Adamic and 50 percent Christ, such texts suggest a balance strongly tilted toward Christ and perhaps even completely so. Christians are 90 percent or even 100 percent Christ, and hence presumably only to this degree dwelling in Adam. (Perhaps they are only 10 percent Adam or even 0 percent Adam.) And this condition clearly creates interpretative tensions with future texts, and with data concerning early Christian sinfulness. It is immediately apparent that Adam is alive and well in both Paul and his recipients in these texts. Hence, any such ludicrous perfectionism is best attributed to later church movements known for their overly high estimation of their own sinlessness — various Gnostics and the like.

Ehrman presupposes this quantitative analogy and utilizes the resulting difficulties when he builds his case against Colossians in terms of incompatibility. He emphasizes the future nature of certain undisputed texts — of 1 Corinthians and the like. (First Thessalonians lies in the background of his case here, too.) And he emphasizes the absence of the resurrected Christ from the present lives of Christians even in Romans — as he must. This move is supported, as we have seen, by an appeal to the Corinthian situation. Paul is said there to be resisting the Corinthians’ endorsement of the present resurrected Christ in their lives, because this would not be mixed enough; they would be embracing a foolish perfectionism. The present Christian condition, according to these undisputed letters, is characterized by a decidedly mixed ontology. A significant percentage of Adamic material remains, and Christians will receive a decisive dose of Christ only after Christ returns sometime in the future. They are presently 50/50 Christians, and possibly not even that.

Ehrman then turns to Colossians and emphasizes the ascended Christ. This Christ is clearly present to the church in power and glory “now,” so he is not, for this reason, future. Ehrman also emphasizes the present resurrected state of Christians in Colossians. And these emphases create a very different picture of the underlying Christian situation if they are interpreted quantitatively. The author of Colossians is not looking forward especially strongly to some future infusion of Christian being; it is here already. The Christians presupposed by Colossians, then, are 10/90 Christians, or perhaps even 0/100, and seem as a result to be very different from the benighted 50/50 Christians presupposed by (Ehrman’s account of) the undisputed letters. The two depictions seem, in short, to be incompatible.

What are we to make of this case when it is understood in these terms — with its key ontological assumption identified? A little ironically, it is now apparent that Ehrman’s case rests in part on a presupposition that is probably anachronistic.

It is, in the first instance, just an assumption — specifically, that a certain metaphor or analogy that seems self-evident for basic physical quantities and their interactions (although it does not actually hold good even there) applies to divine action on humanity. But we do not need to make this prior decision and go on to support its reductionist interpretative consequences.45

Besides being reductionist and unsupported, this assumption also seems characteristically Western and hence quite possibly anachronistic. These spatial and mutually exclusive perceptions of states are familiar to the intellectual traditions of western Europe, but Paul was not an inhabitant of these traditions, many of which built up after his epochal period and within the Latin-speaking church, however anomalously. In direct complement to this concern, properly theological categories, drawing on the reflections of Nicene orthodoxy, tend to contradict this assumption. Divine-human interactions must in the first instance be understood self-interpretatively, and emphatically not by way of analogy with creation; the latter is one of the mistakes that an emphasis on divine transcendence is supposed to guard against. Moreover, any divine self-interpretation begins by affirming the full mutual indwelling of quite distinct realities — three persons in one God; a God fully present in a person; and so on. This is not necessarily to affirm the truth of those claims but merely to note that completely different ontological assumptions are possible that eliminate the difficulties generated by Ehrman’s application of crude spatial analogies to Paul’s texts — although those assumptions are more in tune with the central commitments of the church.46 However, it might be easier to appreciate these objections if we simply switch metaphors.

Rather than thinking of ontologies in terms of mutually exclusive spaces, it might be helpful to think in sonic or musical terms and of ontologies as sounds or musical discourses. So we can conceptualize an Adamic ontology present within humanity as a particular piece of music, perhaps playing loudly and somewhat discordantly. And any ontology present in relation to Christ can be thought of as a completely different piece of music, perhaps playing in a captivating and melodic fashion. It is immediately clear that both discourses can be fully present in Christians, occupying the same human space, at exactly the same time. If we play two pieces of music on the opposite sides of a room, there is no place in that room where the two pieces of music are not both penetrating — although which one is dominant depends on which one is being played with the highest volume. But irrespective of whether we can hear both pieces of music, both are fully present.47

If Paul’s ontological categories are best understood sonically and musically and not visually and spatially, then all of the “incompatibilities” enumerated by Ehrman — even granting both the legitimacy of the coordination at this analytic stage and the characterization of the data — disappear.

This analysis should be sufficient to indicate how cases for pseudonymity in substantial terms like Ehrman’s are problematic. His claim that a fundamental incompatibility is observable between the eschatological material in the undisputed letters and Colossians is a methodologically invalid argument during preliminary framing; it rests on a series of oversimplifications of the data; and it presupposes an anachronistic and arguably false analogy concerning the ontologies involved (which map, we should recall, interactions between the human and the divine).

It follows that claims of substantive incompatibility in any other terms — concerning the Haustafeln, the developments of bodiliness, the role of the assembly — should probably be rejected as well. These claims too are methodologically inadmissible, and they tend to rest on selective interpretations of the data in question.48

With these realizations, it can be seen that the last significant element within the case for the pseudonymity of Colossians has collapsed. We should turn, then, to our final judgment — bearing in mind that all judgments made during preliminary framing are provisional.

2.5 Judgment and Further Implications

It will be apparent by this stage in our discussion that Colossians fits well into both its own implicit exigence and our developing frame, augmenting the latter in intriguing ways. It reinforces the hints in Philemon that Paul was imprisoned in Asia near the Lycus valley, in 50 ce, most probably a few days’ travel to the east of Colossae and hence possibly in Apamea. Moreover, all the usual charges against the authenticity of Colossians have largely wilted under cross-examination — that is, assertions that the text contains small, telltale errors of fact; that its style is unacceptably distinctive; that the teaching it opposes is anachronistic; and that some of its substantive positions, especially in terms of eschatology and soteriology, are incompatible with what we know of Paul from his undisputed letters (even conceding the admissibility of this last type of argument). Indeed, the cumulatively increasing strength of the letter’s apparent authenticity — at this preliminary stage in the historical reconstruction of Paul, framing his basic story — has been rather surprising. It remains only to add that later investigation into the original Pauline letter collection, insofar as we can determine its contents with any certainty, will marginally corroborate this judgment, although this determination is most profitably undertaken at the end of the next chapter.

If Colossians is judged authentic, taking our group of authentic letters to nine, then what are the implications for framing? The main consequences have already been introduced, but some broader implications have yet to be noted.

We should recall first that Philemon and Colossians were written during an Asian captivity, probably in Apamea (or, at least, somewhere like this), in the year previous to Paul’s (calendar) year of crisis in 51, hence in 50 ce. We can infer, moreover, that Paul was eventually released from this imprisonment, his anticipation of this in Philemon 22 thus proving correct. (He was an experienced detainee.) He then almost certainly visited Colossae, Laodicea, and Hierapolis en route to Ephesus, since those settlements lay on the direct path to the Maeander valley and the road to both Miletus and the great Asian metropolis. And doubtless he spent some time in those small towns on the way, catechizing his new converts, along with any other pagans he could persuade to join them.

We cannot say at this point, however, just how long Paul worked in Ephesus over the winter or what events might have transpired to open a door but also to challenge his mission there. And we do not know what the front end of this entire sequence of events contained beyond Paul’s important visits to Antioch and Jerusalem. The letters are silent here (i.e., we have none from this period). Behind this point in the frame, there is simply the long interval back to the Thessalonian correspondence, written in or after 40 ce. Furthermore, while we nurse suspicions that 1 and 2 Thessalonians were written relatively early within this long interval — say in 41 ce itself — we cannot yet confirm this with complete confidence. Beyond this, we know principally of only two specific events that fell into this window, although we know of several more general activities.49

Fragmentary epistolary evidence suggests that at some point between 40 and 50 Paul visited Corinth for a second time. His visit anticipated in 1 and 2 Corinthians, and unfolding when Galatians, Philippians, and Romans were written, was his third. And we could be forgiven for surmising a second visit to Thessalonica as well. There is no direct evidence of this, but it does seem likely given Paul’s announced wishes in 1 Thessalonians 2:17-18 and 3:11, as well as the fact that of the regions he probably subsequently evangelized, three lay along the Ignatian Way, Illyricum being named specifically.

In fact, Paul’s almost inadvertent claim in Romans (15:19) that he evangelized during this period “as far as Illyricum,” hence up the eastern coast of the Adriatic Sea, is highly significant. It seems, in the light of this, that Paul evangelized quite distant regions during this interval, although apparently without much success. We have no evidence of any successful church plants in any other areas from this time (with the possible exception of data from Titus, but we are not considering that data yet). But we can still ask how far Paul seems to have reached in this long and apparently fruitless missionary work.

We have already calculated a rough estimate of the time that Paul spent in a given region on the basis of Galatians 1:17-18, that is, one to two years. If we suppose that Paul spent close to two years in Corinth as against one, given the complexity and success of the Achaian mission, this leaves 43-49 ce unaccounted for in specific terms, assuming the maximum possible extent of this temporal window for the sake of argument. (By the end of 49 or thereabouts he is visiting Antioch and Jerusalem.) And it follows that, moving at his fastest rate, Paul might have evangelized as many as seven regions during this time, one of those being Illyricum.

The regions within the Roman Empire that lie east of Illyricum and west of Jerusalem and Judea (i.e., west of Judea, Arabia/Nabataea, the Decapolis, Cilicia, and Syria; see Rom 15:19), included — enumerating west to east — Moesia, Thrace, Bithynia, Pontus, and Cappadocia. So, astonishing as it appears at first glance, it seems possible that Paul tried at some point to visit and to evangelize all of these regions — six in total, including Illyricum. Moving at his slower speed of two years per region, he could still have evangelized four, so, Illyricum, Moesia, Thrace, and Bithynia. But the slower rate probably held when the missions were successful, so we might think more in terms of the upper limit here of six areas. Moesia and Thrace were quite disturbed areas during the early 40s but might have seemed providentially ordered from around 44 ce onward.50 However, perhaps we should not exclude the possibility of Paul working outside the borders of the Roman Empire periodically, most probably in Dacia (see esp. Col 3:11 and Rom 1:14). Moreover, Galatia could have been evangelized at this time as well. Whether Paul’s Galatians were “ethnic” Galatians from the northern part of the province or provincial Galatians from the southern part (or both) is hard to say; neither option can be obviously excluded at this point from our developing frame. Paul’s most relevant terminology seems to be klimatic or regional, potentially denoting areas larger than (Rom 15:23), equal to (2 Cor 11:10), or smaller than (Gal 1:21) provinces, hence encompassing any of the foregoing options. Acts might eventually help us here, but that intervention must wait. In the meantime, fortunately, not a great deal changes in specific terms in relation to any of these options. We simply cannot envisage where Paul’s Galatian addressees were in absolutely precise terms — whether located on the main northern east-west route from Syria across what is now Turkey, on the southern east-west route, or scattered across both (see Syme 1995).51 And Paul’s remark in 2 Corinthians 11:25 now becomes especially pertinent.

There he writes, τρὶς ἐραβδίσθην, ἅπαξ ἐλιθάσθην, τρὶς ἐναυάγησα, νυχθήμερον ἐν τῷ βυθῷ πεποίηκα.52 Paul implicitly but obviously affirms here a great deal of travel, and especially of sea travel, much of it apparently at inauspicious times of the year. But this remark nicely confirms the implications of the regional data just noted concerning Paul’s activities during this interval. Paul was traveling, as much as he could, by sea, this being by far the easiest and fastest way to travel at the time, although not without risks. And we must envisage him traveling enough to be shipwrecked three times during this period.53 Furthermore, it is clearly not unreasonable to suppose that from 43 to 49 ce he covered a great deal of ground. Hence, his remarks and implications concerning regional missionary work and his remarks concerning travel confirm one another — and Romans 15:19b now makes rather more sense than is often thought, in what might be a final flourish of confirmation (ὥστε με ἀπὸ Ἰερουσαλὴμ καὶ κύκλῳ μέχρι τοῦ Ἰλλυρικοῦ πεπληρωκέναι τὸ εὐαγγέλιον τοῦ Χριστοῦ). Paul really had evangelized all the areas around the Mediterranean between the two points of Jerusalem and Illyricum.

But Paul also details a series of official punishments near his travel text, which are worth considering again (see 2 Cor 11:24: Ὑπὸ Ἰουδαίων πεντάκις τεσσεράκοντα παρὰ μίαν ἔλαβον [25] τρὶς ἐραβδίσθην, ἅπαξ ἐλιθάσθην κ.τ.λ.). Almost all of these would have presupposed an arrest and probable period of detention prior to trial and sentencing. And this seems to confirm the plausibility of the Asian imprisonment presupposed by Philemon and Colossians. Indeed, if Paul visited six more large regions during this period, including Illyricum, and experienced only three of his five synagogual whippings and two of his three beatings with rods, he could still have been punished officially in some way in almost every one of these regions. He was a troublemaker, or at least frequently perceived to be one.

In short, the “years of shadow,” as I call them, that fall between the details of the frame discernible in the late 30s and early 40s ce and the details that come into view from 50 ce onward should be filled out with these inferences. It is frustrating that we do not know more about this period, but we do not know nothing. Moreover, the hints that we do have nicely confirm the scenario developed from this chapter’s analysis of Philemon and Colossians — extensive travel, frequent incarceration, and evangelistic struggle, Paul’s preferred term for this activity being “weakness,” or perhaps better, “vulnerability” (ἀσθένεια).

We should now augment our developing frame in the following terms.

. . . mission to Achaia . . .
// 43- //
. . . second visit to Thessalonica . . .
. . . mission to Illyricum . . .54
. . . mission to Moesia . . .
. . . mission to Thrace . . .
. . . mission to Bithynia . . .
. . . mission to Galatia? . . .
. . . mission to Pontus . . .
. . . mission to Cappadocia . . .
. . . second visit to Corinth . . .
// -49 //
. . . Antioch incident — second visit to Jerusalem55 —
// 50 //
mission to Asia begins — imprisonment56 — Phlm / Col —
visit to Colossae57 — visit/s to Laodicea/Hierapolis58 — visit to Ephesus59 . . .
// 51 //

And with these further framing implications noted, we can turn finally to consider the location and authenticity of Ephesians.

3. Locating “Ephesians”

3.1 The Immediate Implied Exigence

If doubts have not arisen by this stage in our broader analysis concerning Paul’s authorship of a letter, they will almost certainly arise now. Paul’s letter to the Ephesians, as it is usually known, has numerous features that at first glance trouble any judgment of authenticity. Its instructions are highly generalized and even abstract; few if any specific details or problems are apparent in relation to a particular community. Coupled with this, the style of the letter, whether judged grandiose or ponderous, seems quite distinctive. Various substantive emphases are distinctive as well — most famously, a sustained emphasis on the Christian community usually articulated in terms of “the church.” The letter consequently seems to belong to a later, calmer, more ecclesial era than to that of Paul’s gritty struggles with small groups of people, many of whom could be named if necessary. Hence, a judgment of pseudonymity is invited quite quickly. However, certain similarities between Ephesians and Romans should probably give us pause.

While the opening and closing of Romans contain a great deal of particular information (not unlike Col 4:7-17), the letter body is similar in style and generality to Ephesians. We find a series of carefully crafted units that seem to have little direct contextual relevance or specificity. But I would suggest that Romans had a highly practical and specific exigence for which its sets of generalized discussions proved necessary — the pending arrival at Rome of the Teacher with his sophisticated Judaizing agenda. This necessitated an extended, somewhat generalized response by letter from Paul, hence Romans. The question here will be whether a plausible particular exigence can explain the composition and dispatch of Ephesians in similar terms. That is, we will here, as usual, assume authenticity at first and see whether any developing hypothesis in these terms breaks down under close scrutiny.

We should note immediately that the text of Ephesians presents itself as grounded in some particular Christian community or sequence of congregations, because it expects Tychicus to bear the letter, to explain it, and to model Paul’s ethos and ministry (see 6:21: Ἵνα . . . εἰδῆτε καὶ ὑμεῖς τὰ κατ’ ἐμέ, τί πράσσω, πάντα γνωρίσει ὑμῖν Τύχικος ὁ ἀγαπητὸς ἀδελφὸς καὶ πιστὸς διάκονος ἐν κυρίῳ). He will also bring news of “us” and encourage the letter’s recipients (so 6:22: ὃν ἔπεμψα πρὸς ὑμᾶς εἰς αὐτὸ τοῦτο, ἵνα γνῶτε τὰ περὶ ἡμῶν καὶ παρακαλέσῃ τὰς καρδίας ὑμῶν). So a specific arrival by the letter’s bearer and interpreter, Tychicus, is articulated, in relation to specific recipients and from a specific sender. But whom was he actually supposed to visit? Can a plausible account of the letter’s recipients be found in the text?

We would ordinarily expect these sorts of details to be provided first in an ancient letter’s prescript, and that is the case here in Ephesians, although we now step into murky waters. Identifying the addressees in Ephesians is complicated by a set of textual variations. Yet there is no way forward without engaging with this question. Nestlé-Aland27 provides most of the necessary data; five variations must initially be considered.

1. τοῖς ἁγίοις τοῖς οὖσιν ἐν Λαοδικείᾳ καὶ πιστοῖς ἐν Χριστῷ Ἰησοῦ . . .60

2. τοῖς ἁγίοις οὖσιν καὶ πιστοῖς ἐν Χριστῷ Ἰησοῦ . . .61

3. τοῖς ἁγίοις τοῖς οὖσιν καὶ πιστοῖς ἐν Χριστῷ Ἰησοῦ . . .62

4. τοῖς ἁγίοις τοῖς οὖσιν ἐν Ἐφέσῳ καὶ πιστοῖς ἐν Χριστῷ Ἰησοῦ . . .63

5. τοῖς ἁγίοις πᾶσιν τοῖς οὖσιν ἐν Ἐφέσῳ καὶ πιστοῖς ἐν Χριστῷ Ἰησοῦ . . .64

It can be seen immediately from this tabulation that the case for the letter’s Ephesian addressees is rather weak. The main uncials, p46, and the earliest traceable patristic comments are all against this identification. It cannot be attested in the manuscript tradition before the fifth century, although Tertullian protests in its behalf at the beginning of the third. Corroborating a suspicion of later emendation, many of its witnesses seem to have harmonized the addressees further in the light of Romans 1:7 (πᾶσιν τοῖς οὖσιν ἐν Ῥώμῃ ἀγαπητοῖς θεοῦ, κλητοῖς ἁγίοις . . .). So an Ephesian address should be tabled initially in favor of variations 1, 2, and 3.

But in variant 2 the sole witness, p46, has apparently, somewhat typically, made a mistake and elided the article (although not much turns on this). And this reduces the variants essentially to two: one with a Laodicean address — 1 — and one lacking a specific address altogether — 3, supported to a degree by 2. A decision between them is not that difficult, however, at least at this stage in our analysis, with the authenticity of Ephesians being assumed. Paul uses a participle of being on four other occasions in his prescripts, and they are invariably both arthrous (here speaking against p46 and variant 2) and possess a local geographical object.

Rom 1:7: πᾶσιν τοῖς οὖσιν ἐν Ῥώμῃ ἀγαπητοῖς θεοῦ, κλητοῖς ἁγίοις. . . .

1 Cor 1:2: τῇ ἐκκλησίᾳ τοῦ θεοῦ τῇ οὔσῃ ἐν Κορίνθῳ, ἡγιασμένοις ἐν Χριστῷ Ἰησοῦ, κλητοῖς ἁγίοις, σὺν πᾶσιν τοῖς ἐπικαλουμένοις τὸ ὄνομα τοῦ κυρίου ἡμῶν Ἰησοῦ Χριστοῦ ἐν παντὶ τόπῳ, αὐτῶν καὶ ἡμῶν. . . .

2 Cor 1:1b: τῇ ἐκκλησίᾳ τοῦ θεοῦ τῇ οὔσῃ ἐν Κορίνθῳ σὺν τοῖς ἁγίοις πᾶσιν τοῖς οὖσιν ἐν ὅλῃ τῇ Ἀχαΐᾳ. . . .

Phil 1:1b: τοῖς ἁγίοις ἐν Χριστῷ Ἰησοῦ τοῖς οὖσιν ἐν Φιλίπποις σὺν ἐπισκόποις καὶ διακόνοις. . . .

So, as Lincoln (1990, 3) observes, the construction in Ephesians 1:1b “demands a subsequent geographical location,” and we should try to supply one, thereby avoiding the odd situation of a general address created with an unnecessary participle of being. But we are in the happy position of having such a location supplied in what is easily our earliest attested variant by way of Marcion, and so, at the latest, deriving from the mid-second century — namely, Laodicea.65 Moreover, we can even go on to offer a plausible explanation for the elimination of this address in the next phase of scribal work as Paul’s letters were transmitted. It could well have been thought inappropriate that an apostolic letter should address a church that the book of Revelation exposed later to be such an abject failure (see Rev. 3:15-18 in the context of 3:14-22 — Harnack’s view, which I find plausible; the much later insertion of Ephesus into the address might then be attributable to the influence of Rev as well [see the glowing account of the Ephesian church in Rev. 2:1-7] or might have been an inference from 2 Tim, as Van Kooten [2003, 199] suggests). Alternatively, this emphatically ecclesial letter could simply have been generalized, as Romans was.

Irrespective of the exact reasons for its later modifications, it is difficult to object cogently to an original address in terms of variant 1.66 And so we will proceed for the time being on the assumption that the original addressees of our canonical Ephesians were actually the Laodiceans, and will consequently refer from this point onward to canonical Ephesians as Laodiceans (or, alternatively, as “Ephesians,” thereby indicating the improbability of this address).67 And with this initial judgment in place, we need to consider what information Laodiceans gives us about its recipients besides their location in the ancient town of Laodicea, which lay in the mouth of the Lycus valley astride an important crossroads. Does this specific address help us interpret the letter plausibly in contingent terms, or is it a crafty fiction?

According to the letter, its addressees do not seem to have known Paul personally when it was composed. Laodiceans 3:2 suggests that they had only “heard” of his special benefaction for pagans — although arguably not to the depth and degree that Paul would have liked — and apparently, he had only “heard” of their faith and love in turn (1:15).68 Paul’s address throughout the letter is nevertheless highly consistent in general terms; the letter’s addressees are identified as pagan converts in the specific sense of being Christian converts from paganism. In 2:11-12 this is especially clear: [2:11] Διὸ μνημονεύετε ὅτι ποτὲ ὑμεῖς τὰ ἔθνη ἐν σαρκί, οἱ λεγόμενοι ἀκροβυστία ὑπὸ τῆς λεγομένης περιτομῆς ἐν σαρκὶ χειροποιήτου, [12] ὅτι ἦτε τῷ καιρῷ ἐκείνῳ χωρὶς Χριστοῦ, ἀπηλλοτριωμένοι τῆς πολιτείας τοῦ Ἰσραὴλ καὶ ξένοι τῶν διαθηκῶν τῆς ἐπαγγελίας, ἐλπίδα μὴ ἔχοντες καὶ ἄθεοι ἐν τῷ κόσμῳ.

Hence, the addressees are never included by the text within Judaism, nor are they placed outside the Christian community within paganism proper at Paul’s time of writing. Their construction in these terms is quite generic but consistent, although their inextricable relationship with Judaism is plainly in view as well, a matter to which we will later need to return.69 Moreover, the letter supplies a persuasive reason for the lack of personal acquaintance between Paul and the Laodiceans.

Paul is currently incarcerated close by, something he mentions three times overtly (3:1, and see also 13; 4:1; 6:19-20). This is of course a plausible location for Paul when composing a letter. Of the nine letters currently judged authentic, three were written from an imprisonment. If Paul had been incarcerated prior to reaching Laodicea, then clearly the letter’s recipients would not have known him personally, although they might have had some distant knowledge of his reputation. This is all fundamentally plausible.

And with these initial observations in place, we should attempt a slightly more detailed description of the letter’s immediate implied exigence. Three further, particular, and quite significant dynamics discernible in the text need to be probed.

(1) The Definition of the Addressees’ Identity

We do not know how the Laodiceans had converted, but as we just saw, they are defined by the letter consistently as converts from paganism. And that this group of unknown etiology had converted is intriguing but not necessarily problematic. We know of other missionary work among pagans in addition to Paul’s from his letters alone. The letter suggests that Paul had been imprisoned not far from these converts, before he had reached their town. So he had had no personal contact with them yet and no influence on their formation. However, technically these converts still fell under his apostolic jurisdiction; Paul was a recognized authority on pagan conversion (Gal 2:1-10, esp. 8-9), and certainly he would have viewed matters in this way. It seems to be this situation that creates the rather fascinating and distinctive exigence for Laodiceans.

Paul had a clear reason to shape the Christian existence of these unknown pagan converts in terms of his own missionary and apostolic agenda, that is, in terms of his coherence. The Laodicean converts knew none of this but clearly ought to have, in his view. This was their basic deficiency. But because of Paul’s imprisonment, they would have to be instructed by letter in lieu of a personal visit. Hence, Paul composed Laodiceans — and thereby created a largely unparalleled situation for the modern interpreter. If this reconstruction holds good, then we would observe in Laodiceans as in no other epistolary situation (at least yet) a relatively straightforward account of Paul’s missionary agenda in relation to pagan conversion — a presentation, he might say, of his gospel, and that Beker would define in terms of coherence. It was the existence of the mysterious Laodicean converts not far from Paul’s incarceration that created this unusual rhetorical “space.” We have encountered in this situation, essentially for the first time in our letter sequence thus far, a contingent reason for a largely coherent text. The letter to the Roman Christians would be the closest approximation, but those recipients had some Pauline converts among them and, moreover, were under threat from a third party, complicating any direct presentation of Paul’s coherence. No such third party is evident in relation to Laodiceans.

This, then, is the fundamental explanation that the letter provides of its own immediate exigence, and it seems plausible, although certain pitfalls become apparent as we try to articulate it further.

Scholars have often suggested reductionist accounts of this coherence when Laodiceans presents a complex orchestration of themes and symbols. Alternatively, problematic and/or anachronistic organizations of this material have been suggested — for example, into doctrine and ethics. But fortunately, we do not need to press much further at this stage for the purposes of framing. It suffices for now to note that this exigence would have been filled rhetorically by Paul with what we might call an account of pagan Christian identity.

Admittedly, “identity” is a notion fraught with difficulty. It is contested, complex, and unstable.70 But even admitting its important postmodern qualifications, the notion that groups and their leaders engaged (and engage) in attempts to define themselves is plausible and arguably even ubiquitous. Suitably qualified, then, it seems fair to suggest, following some key hints in Lincoln, that Laodiceans constructs a pagan Christian identity — in Pauline terms! — for its Laodicean recipients.71 But this account does not exhaust the letter’s implications concerning its situation. We need now to note two further, less important but still significant aspects of its genesis.

(2) The Rhetorical Dimension

Paul’s letter to the Laodiceans might well be the most straightforward account that we possess of his gospel as it constructs the identity of converts from paganism, but it is not for this reason simply a coherent statement. It still sought to move a group from a location perceived to be deficient to one perceived to be better, that is, it still possessed rhetorical goals and strategies, and these need to be identified, although Laodiceans presents peculiar challenges in this regard. In the first instance, it seems that Paul himself did not know a great deal about the Laodiceans. Moreover, their lack of acquaintance with him, and possibly even lack of much knowledge of him, entailed that he could exert little or no leverage on them in terms of prior personal knowledge, relationship, and friendship. The letter itself had to establish an appropriate account of his authority, so this was a rhetorical goal rather than a point of rhetorical leverage. And second, we are one further step removed from this shadowy situation, lacking even Paul’s sketchy knowledge of his recipients. We must therefore make certain assumptions about the letter’s rhetoric and then qualify them heavily in terms of certitude.

Paul would probably have appealed to views that the Laodicean Christians already held to build toward his own positions; he would presumably have sought to suggest that, correctly understood, some of their own existing commitments entailed ultimately a Pauline construction of their identity.72 Moreover, he would probably have effected as much solidarity with those existing commitments as he could, affirming and reiterating them where that was possible. Hence, it would be unwise to understand any affirmations in merely propositional or cognitive terms, although the letter would have included those. The Laodiceans were, like all human groups, located within a set of existing practices (in their case, early Christian), so it is not surprising to find Laodiceans (at least arguably) echoing elements of early Christian worship — eulogy, doxology, songs, and prayers.

In short, then, insofar as Paul knew their views and was able either to build from or simply to affirm them, Laodiceans contains non-Pauline Christian material already assumed to be operative at Laodicea, which he is using as a rhetorical resource. We have little or no way of distinguishing this material from Paul’s at present; this exercise must wait (at least) until framing. But we do need to take this dimension of Laodiceans into account when assessing it.

And with this realization about the probable presence of traditional early Christian material in Laodiceans, a second subordinate aspect of the situation now needs to be recognized.

(3) The Compositional Dynamics of Incarceration

If Paul was imprisoned for some time during the composition of Laodiceans — perhaps for a number of months — then several further aspects of the letter might need to be taken into account.

We have already noted that incarceration is often a fertile space for reflection and composition by literate activists (principally by affording time), yielding definitive work, provided that appropriate ways of preserving and disseminating it can be found. So King’s “Letter from Birmingham Jail” was written initially on the margins of a newspaper and smuggled out of his cell in a shoe. It is salutary to note further that it was written in stages; other parts were composed on scraps of paper and on a pad used by his attorneys when visiting him, and the final letter was then edited together by someone else, the Rev. Wyatt Walker, at the local campaign headquarters.73 Consequently, we should note, first, that the letter to the Laodiceans could have been a deeply considered piece, worked over time and time again. We should not regard it as a manifesto or systematic summary per se; we have just seen that a practical exigence elicited its reflections. But Paul’s incarceration entailed that these could have been highly considered reflections that were eventually committed to writing in Laodiceans. They also could have been largely oral/aural reflections in their initial stages, something we will need to consider further momentarily.

Second, Paul’s incarceration might have restricted his knowledge about the Laodicean situation. We must therefore reckon with the possibility that he possessed very little information about his addressees, which might have generated in turn a certain blandness in his characterizations of them.

Third, Paul might then have filled out his reflections, as all people do, with material from his own emotional and conceptual situation.74 This would not have been a matter of deliberate, subtle instructions for any Nebenadressat so much as the subliminal introduction of aspects of Paul’s own context of imprisonment into the constructions of the text. We should be alert to these possibilities, and aware of their implications for any assessment of framing.

This, then, is the account that Laodiceans supplies of its immediate exigence. The letter suggests that the principal situation underlying its composition was created by the presence of a mysterious group of pagan converts at Laodicea and Paul’s incarceration not far from them, along with his knowledge of their existence. Paul understandably wanted to shape this group’s Christian identity in his own terms, and was forced to do so by letter. But in that letter it seems certain that as part of his broader rhetorical strategy he appealed to and affirmed their own existing Christian location to some degree, although just where and to what extent is difficult to determine; and he also likely subliminally projected aspects of his own incarceration into his highly reflective epistolary account of Christian identity. There is nothing immediately objectionable in this exigence, although it is atypical, but then most of Paul’s letters have distinctive aspects to their exigencies. As usual, we will find this account either confirmed or denied as we turn to evaluate where Laodiceans might fit in these terms into our broader developing frame — but this is not a difficult judgment now that we have identified “Ephesians” as Laodiceans.

3.2 Framing Considerations

Like the disciples at Jesus’ last meal in Jerusalem, Paul’s letter to the Laodiceans already has a place prepared for it within our developing biographical framework. His letter to the Colossians notes its existence in 4:16b (. . . καὶ τὴν ἐκ Λαοδικείας ἵνα καὶ ὑμεῖς ἀναγνῶτε), so it simply slots into our existing biography in this place without further ado. Moreover, this verse and two others in Colossians confirm the prior existence of the Laodicean congregation (2:1; 4:15, 16a), along with their lack of any direct personal acquaintance with Paul (see 2:1 in particular: Θέλω γὰρ ὑμᾶς εἰδέναι ἡλίκον ἀγῶνα ἔχω ὑπὲρ ὑμῶν καὶ τῶν ἐν Λαοδικείᾳ καὶ ὅσοι οὐχ ἑόρακαν τὸ πρόσωπόν μου ἐν σαρκί). These are important confirmations by Colossians of the key elements in the account that the letter to the Laodiceans supplies of its immediate exigence: the existence of the Laodicean converts, their lack of personal acquaintance with Paul, and Paul’s current incarceration close by.75 The existence of Laodiceans alongside Colossians and Philemon now deepens our understanding of the entire situation within which all these letters were composed in several interesting ways, enriching our frame at this point.

First, Colossians in combination with Laodiceans suggests that what was probably a more reflective situation had become characterized by some urgency. The time pressure was generated by the arrival at Colossae of a problematic teaching, probably by way of an equally problematic third party, information that was conveyed to Paul by Onesimus. The situation was therefore urgent at Colossae, necessitating the dispatch of a brief corrective letter. But Colossae was not far from Laodicea and Hierapolis. So it would have become apparent at this moment as well that the introduction of Paul’s apostolic agenda at Laodicea would potentially be confused and hindered if it reached the Laodiceans after the teaching and/or evident third party at Colossae did. Given the situation at Colossae, catalyzed for Paul by the arrival of Onesimus, it followed that Paul’s intervention at Laodicea by way of Laodiceans now needed to be made as soon as possible. And certain relationships between the letters seem discernible in the light of this new urgent exigence.

It seems likely that Laodiceans was primarily composed before Colossians (if of course it is authentic). The letter to the Colossians clearly responds to the urgent exigence in that small town. And it seems unlikely that Paul would have written it and then composed a highly crafted letter in the light of Colossians for the Laodiceans, to be sent with the same messenger, Tychicus — one apparently eliminating many of the specific references in Colossians to the approaching problematic teaching, despite its approach, but expanding elaborately on much else. Such dallying seems less likely than the reverse sequence.

Paul was probably crafting Laodiceans in a more leisurely manner, and then Onesimus interrupted its final dispatch with his urgent news. Paul then composed Colossians rather more quickly, apparently informed significantly by the composition he had just undertaken so carefully for the Laodiceans (and this economy is understandable within both an urgent and an incarcerated situation). But he clearly responded in Colossians as well to the specific information about the Colossian situation brought to him by Onesimus (see esp. the largely unique material in Col 2:8–3:4).76 He then wrote Philemon, addressing Onesimus’s particular difficulties, and sent all these letters off together with Tychicus. But before noting the challenges implicit within this reconstruction, we ought first to note some of its further helpful implications.

Reconstructed in this way, the situation entails that both Laodiceans and Colossians are now to some degree components in a single extended epistolary event and at more than one location. Paul is explicit in this regard in Colossians 4:16. The letter to the Colossians is to be read to the Colossians but then copied and sent to Laodicea; in this way, the specific teaching in Colossians addressing the problematic Colossian teacher will be heard by the Laodiceans and by Nympha’s house gathering. Moreover, the letter to the Laodiceans is to be read to the Colossians (and presumably copied as well) so that they will benefit from its considered account of pagan Christian identity. Thus, the letter to the Colossians no longer addresses just the Colossians, its explicit recipients, but will address the audience of Laodiceans as well; and vice versa, the letter to the Laodiceans will address the audience of Colossians. (Phlm will not apparently be part of this exchange.) Paul is aware of this strategy and its intended results before he authors Colossians 4:16.

This important level of integration between the two main letters before us is confirmed by the phenomenon of the letter bearer. Tychicus is to carry all three letters to Colossae, and two onward to Laodicea and Hierapolis, bearing and reading out at least Colossians and Laodiceans in each location. Paul is clearly envisaging this process as well when he finishes Colossians. So we should not now think in overly discrete terms about the recipients of Colossians and Laodiceans, despite the named addressees. Because of the peculiarly intertwined and urgent situation that elicited these letters, the Colossians heard Colossians and Laodiceans (and some of them, Phlm as well); and the Laodiceans and Nympha’s house gathering heard Laodiceans and Colossians. It seems to me that these two phenomena — Paul’s new strategy of multiple letters and their delivery by way of a single visit from Tychicus — now greatly complicate any precise assessment of dependence. Indeed, the situation must now be understood in a rather more complex way than is arguably often the case. (And it suggests a slightly different approach to any commentary on these letters; isolating the letters risks producing artificial historical-critical results.)

Although this reconstruction does envisage Laodiceans as coming first in principal compositional terms, that does not necessarily entail that it was completed first — an important distinction to grasp. The three letters being sent with Tychicus would have received final readings and Paul’s signatures together, along with an explicit notation from Paul of at least two words in Philemon 19 (ἐγὼ Παῦλος ἔγραψα τῇ ἐμῇ χειρί, ἐγὼ ἀποτίσω), and their dispatch would then have been simultaneous. Hence, that different copies of each letter might have received final modifications and “tweaks” from Paul just prior to their finalization cannot be ruled out; indeed, it seems likely. So while we might assign a principal compositional priority to Laodiceans over Colossians, we cannot assign an utterly unidirectional sequence in every respect. It seems more likely that Laodiceans, Colossians, and Philemon were finally written up by a scribe together, signed together by Paul, and then dispatched together with Tychicus — to be read largely together on arrival. Fundamentally interactive relationships therefore seem likely in the final stages of their composition, despite the unidirectional nature of the initial composition of the main letters, and that interaction would have begun with the arrival of Onesimus at Paul’s place of imprisonment. This does not entail that matters of dependence cannot be sorted out to some degree; indeed, any account of dependence should be helped by these realizations. But it qualifies sharply any judgments based on a rigidly unidirectional account of the causality between Laodiceans and Colossians (and a further caveat will be introduced here shortly, in the next subsection).

With this nuanced account in place concerning the probable compositional dependencies between the letters — of initial principal unidirectionality making a transition into interactivity — we should turn to evaluate whether this reconstruction can respond adequately to the serious challenges offered by numerous scholars to the authenticity of “Ephesians.” If “Ephesians” was in fact Laodiceans, how might this affect its frequent designation as pseudonymous?

3.3 Challenges

It seems fair to claim at the outset that although the immediate implied exigence of Laodiceans is distinctive, it is not for this reason implausible in any respect, and one of its advantages is that the relationship of the letter to our developing frame is immediately indicated plausibly as well. It is the letter identified by Colossians 4:16 and consequently fits straightaway into a part of the frame that has already been developed and tested, both clarifying and complicating this locale in several ways. It is now time to place pressure on some of those complications. Any challenges here will tend to fall into the usual categories — concerns in terms of dependence (here especially noticeable), style, and substance, since simple anachronisms or other mistakes by a pseudepigrapher have not been identified decisively.

(1) Dependence

If the dependence of Laodiceans on Colossians could be demonstrated, then it should in my view be designated as pseudepigraphic and we should revert to calling it Ephesians, since it should no longer be identified with the letter described in Colossians as being written simultaneously to Laodicea. We have already noted how unlikely it would have been for Paul, in a situation of some urgency occasioned by the arrival of Onesimus, to have composed this rather generalized letter in complex but indirect dependence on Colossians — after it — as the third part of a set of three letters being composed at the time.77 It would seem more likely, if the dependence of Laodiceans on Colossians could be decisively proven, that Laodiceans was a later generalization based on Colossians by a cunning pseudepigrapher, who thereby found a brilliant point at which to insert this pseudepigraphon into a Pauline letter collection. It was introduced in place of the lost letter to the Laodiceans, a location further facilitated by its deliberate resonances with the other main letter composed at that time and still extant, Colossians. But the advocacy of the pseudonymity of “Ephesians” on grounds of its demonstrable dependence on Colossians faces some serious challenges. Three particular difficulties within this case need to be appreciated.

(i) As in any attempted resolution of the Synoptic Problem, it is difficult to demonstrate definitively the direction of the relationships of dependence in view. Almost all the arguments used in favor of one relationship can generally (and even invariably) be countered with equally plausible explanations that run in the opposite direction — the recurrence here in another form of the problem of multiple plausible narrativity. That is, that the letters exhibit some sort of dependence is obvious. But we cannot tell in relation to Laodiceans and Colossians — as in many of the relationships between the Gospels — whether Colossians is a compression of the material in Laodiceans that summarizes and reorients it, or whether Laodiceans is an elaborate expansion of selected and more cryptic material found in Colossians. Both accounts of any given textual relationship between Laodiceans and Colossians seem almost invariably to be plausible, rendering the entire contention indecisive.

(ii) The situation is further complicated by the possibility that the relationship between Laodiceans and Colossians was fundamentally oral/aural and not written, a point where we must resist the anachronistic tendency to conceptualize Paul’s ancient oral culture in terms of our own written privileges (so of course Ong, esp. 2012 [1982]; also 2002). That is, most scholars seem to conceptualize the dependence in view as textual in the specific senses of texts being written and copied, and at points this seems definite (i.e., Laod 6:21-22 // Col 4:7-8;78 and quite possibly Laod 1:1-2 // Col 1:1-2 as well79). This predilection is evident in the production of various synopses of the problem, which assume that it is soluble in these terms.80 But the texts just noted are small, distinctive segments within a much richer pattern of dependence, and they occur in the letter opening and closing, where we might expect a more standardized approach. Elsewhere the relationship between the two letters is by no means synoptic in the sense that it maps obviously in terms of the use of a prior written source by a later redactor. Some sequences are preserved, but in general, motifs are mixed up and recombined (depending on the assumed flow of dependence) — something that causes the various attempted synopses a lot of difficulty. This situation seems more explicable in terms of a fundamentally oral process of composition as against a written one, in which one deeply reflective and carefully memorized text has filtered through into another compositional “performance,” much as improvising actors return to and reuse stock stories, jokes, and verbal sequences in subsequent performances (see here esp. Wells 2004). But when interpreters concede the orality of the process, they lose the right to make strong claims about causality. The direction and causalities of oral reproductions cannot be reconstructed with any confidence simply in their own terms; this would be not unlike trying to eat soup with a knife.81

An oral/aural understanding of the process lying behind Laodiceans and Colossians possesses the further virtue of being the most plausible account of Paul’s practical compositional situation, in prison. His ability to compose a text orally and aurally would likely not have been obstructed, but any use of written materials might have been much more difficult. Indeed, as was noted earlier, prison literature is often a fundamentally oral process, committed to writing at late stages and in fragile ways.

In short, if we recognize a primarily oral process of composition lying behind both Laodiceans and Colossians, then the relationship of dependence between them was rather fluid and complex, defying any obvious mapping in terms of causality. The reutilization and redeployment of some material from Laodiceans by Colossians seems just as plausible as the same process running in the other direction; and a significant interaction between the two letters in these terms cannot be ruled out, with Paul perhaps composing and memorizing both penultimate drafts of the letter bodies prior to syllabic dictation to a scribe.

(iii) The situation is further complicated in the case of any putative dependence by Laodiceans on Colossians by the probable interactivity of the later stages in the composition of these texts. Hence, even if an oral process of composition had a principal causal direction early on, it nevertheless seems likely that as the three letters were committed to print, presumably by a scribe or a scribally adept member of Paul’s circle (although none is designated as such), cross-checking and what we might call textual “cross-pollination” might have taken place. This seems to be a plausible explanation for the written dependence in view in the letter openings and closings. In essence, because the final production and dispatch of the letters was simultaneous, any minor moments of dependence and adjustments to the texts during their final stages of composition could have run in multiple directions. We should probably imagine Paul listening to penultimate written versions of all three letters and offering minor comments. These could then have been introduced either in the margins of the letters or in a new copy, after which he would have signed them personally and sent Tychicus on his way.

It is arguably easier to imagine the extra material apparent in Laodiceans 6:21-22 as an adjustment to Colossians 4:7-8 — although this is by no means certain (see Best 1997b). But this sort of addition could well have come from the final stages in the letters’ composition and consequently suggests nothing about the principal direction of dependence between them. The text in Colossians could have preceded Laodiceans in this instance, but Laodiceans could still have largely preceded Colossians, after a phase of reflection and oral composition in prison.

In view of these dynamics, it seems to me that any definitive case for the pseudonymity of “Ephesians” because of its evident dependence on Colossians is doomed. The establishment of this conclusion in defiance of all alternatives is effectively impossible. Moreover, it now seems that many advocates of the dependence of “Ephesians” on Colossians have been caught up by broader disciplinary flows of plausibility that have left them vulnerable in the context of our particular method and resulting reconstruction. If it has already been decided that Ephesians is pseudepigraphic but Colossians possibly authentic, then it follows that Ephesians is dependent on Colossians — and with sufficient determination, this construal of the data is possible. But in the context of this framing exercise, with both “Ephesians” as Laodiceans and Colossians currently being treated as authentic, such considerations beg the key question. Indeed, arguably it is now becoming apparent that our whole approach to the assessment of the dependence between Laodiceans and Colossians needs to be reversed, and made to depend on framing, not to dictate it. With preliminary framing judgments made, we will be able to return and investigate the dependence in view knowing already, from the frame, what the principal direction was, and what the broader circumstances were that informed its nature. Perhaps only responsible framing will allow us any real hope of reconstructing the main flows within the complex relationship of dependence evident between Laodiceans and Colossians.

So much, then, for arguments in favor of the pseudonymity of “Ephesians” in the light of its evident dependence on Colossians. That relationship is possible, but so are many others. But do other echoes of “Ephesians”/Laodiceans in the undisputed letters call its authenticity into question?

(2) Dependence on Undisputed Pauline Letters

Numerous echoes of Laodiceans can be found in the other undisputed Pauline letters, and this might suggest an influence on the later composition of “Ephesians” from the authentic Pauline letters, and hence pseudepigraphy. But the peculiar circumstances of Laodiceans arguably speak to this concern quite satisfactorily.

Of all the Pauline letters examined thus far, Laodiceans was elicited by particular circumstances that called most directly for an exposition of Paul’s most coherent concerns — in effect, for his construction of the Christian identity of a community of pagan converts. Moreover, this letter fits into the frame just prior to Paul’s year of crisis, that is, in 50 ce, preceding the troubles of spring 51-spring 52, when 1 Corinthians, 2 Corinthians, Galatians, Philippians, and Romans were composed. It seems possible that the echoes of Laodiceans found in the undisputed letters are not evidence of the later influence on “Ephesians” of the authentic Paulines so much as the filtering through of many of Paul’s most significant concerns from Laodiceans to the letters he was forced to write in 51 and the spring of 52 in various rather more difficult and specific circumstances — and its especially concentrated set of echoes with Romans might further encourage confidence in this view. Preliminary framing can proceed, in short, with the hypothesis that Laodiceans articulates many of Paul’s central concerns. This would explain the heavy interdependence evident between the undisputed letters and Laodiceans.82

It will of course pay a price for doing so. Ultimately, this position will result in a more “Ephesiocentric” account of Paul’s thought than might otherwise be the case, and some scholars will be happier paying this price than others. But it is not cogently falsifiable at this stage in our discussion, when framing. That is, any falsification of the authenticity of Laodiceans in terms of evident dependence would have to demonstrate the nature and content of Paul’s coherence, and then proceed to show how the positions found in “Ephesians” in echo with the undisputed letters are a demonstrably impossible account of that coherence. This case can be made only after preliminary framing and not during it (because it presupposes a biographical frame within which Paul’s coherence can be reconstructed and assessed). So framing can proceed for now, with the caveat that — probably — Laodiceans may ultimately have to figure centrally in any account of Paul’s coherence. (I say “probably” because Paul’s coherence could ultimately be deemed incoherent or markedly developmental, with Laodiceans representing the concerns of 50, and his other letters the rather different concerns of 51 and 52 ce.)

But we have raised substantive concerns here and so should now consider the many serious objections that arise to the authenticity of “Ephesians” in these terms.

(3) Unacceptable Generality

Perhaps the principal initial difficulty that many scholars have with the authenticity of “Ephesians” is its generality, which can be described in different ways. One can speak of its level of abstraction, and even of its blandness. What is generally being articulated here is the sense that “Ephesians” lacks any of the particularity or specificity that we invariably find elsewhere in a genuine Pauline letter. One line of defense offered against this criticism in the past has been the hypothesis that “Ephesians” was actually a circular letter or, in a variation on this view, a treatise or manifesto that was freighted in an epistolary form — and this might explain some of the textual confusion in the prescript as well. But we have already briefly noted why this rejoinder is inadequate.

Most forms of the letter have addressees, and those that lack them are grammatically odd and stylistically atypical. Moreover, the letter presents itself as being carried and delivered by Tychicus, who has been entrusted with quite concrete personal duties; one of them is to remind his recipients of how Paul lives. So the letter is not really a circular (except perhaps in a highly localized sense) or a manifesto, and the letter’s blandness consequently cannot be explained in this way. However, the identification of “Ephesians” as Laodiceans does seem to speak to this perceived difficulty adequately.

Once it is grasped that the Laodiceans had not been converted by Paul, and barely knew him, and that Paul was prevented from instructing them by his incarceration, and may in turn have known very little about them, the scene is set in quite specific terms for a highly generalized discussion — an account of Christian identity for pagan converts. The key to Ephesians in this relation is then its hidden identity as Laodiceans, along with the location of the Laodiceans in relation to Paul. The letter’s level of abstraction seems well explained in these practical terms. However, having made this general observation about the genesis of the letter’s distinctive approach, we can still ask whether certain substantive emphases within it are unacceptable and/or anachronistic in relation to Paul.

(4) Unacceptable Substantive Emphases

Scholars have suggested that the realized eschatology evident in “Ephesians” is as unacceptable in relation to Paul here as it was in Colossians. But we have already noted a series of difficulties with this objection, not least of which is its own anachronistic interpretative projections. The letter to the Laodiceans affords several other themes susceptible to this objection, however, most notable of which is its emphasis on “the church.”

The ekklēsia (ἐκκλησία) is named specifically nine times in Laodiceans, and always in its general, universal sense (1:22; 3:10, 21; 5:23-25, 27, 29, 32). But this usage is rare elsewhere in the Pauline corpus, occurring only seven times in other letters (see esp. Col 1:18 and 24; but also Gal 1:13; 1 Cor 10:32; 12:28; 15:9; Phil 3:6). Paul usually uses this signifier to refer to a local gathering or congregation, a usage entirely absent from Laodiceans. This seems to be a peculiar configuration, although it could be explained quite easily in terms of a later pseudepigrapher’s desire to provide an account of “the church.”

But while this challenge identifies an important emphasis within the data, it does not seem fatal to Pauline authorship. The presence of the universal sense for ekklēsia in four other authentic letters might reinforce the hypothesis that Laodiceans articulates Paul’s most central, coherent concerns. And we may now add a secondary argument to this rejoinder.

We noted earlier that, although it is difficult to detect, Laodiceans doubtless uses traditional Christian material at Laodicea to generate leverage on its auditors. And arguably, the universal sense of “the assembly” is one such traditional point of purchase. This motif is poorly translated as “the church,” a rendering that will instantly generate a spectrum of anachronistic connotations. It is actually a reference to an assembly or gathering, and hence, in its singular usage, to the assembly, drawing here most probably on the OT notion of the assembly of all Israel (qahal Israel). This was thought by many in Paul’s day to be eschatologically consummated and hence could well have been appropriated by the first followers of Jesus to define themselves. Paul’s use of this meaning in three references to his pre-call persecutions seems tentatively to corroborate this reconstruction (Gal 1:13; 1 Cor 15:9; Phil 3:6), as does its occurrence in this sense in Colossians 1:18, a section of material widely viewed as traditional, if not quoted directly from tradition. These realizations combine to respond more than adequately to the objection that Laodiceans is overly ecclesial.

Not only is this emphasis on the ekklēsia possibly a part of Paul’s coherence, but it might, in either complementary or alternative terms, have been a part of the Laodiceans’ traditions to which Paul was appealing for rhetorical leverage as he sought to construct his richer account of Christian community — glossing this image in turn with connotations in terms of a temple (which might also have been traditional material), a new humanity (here Adamic, and so traditional in some sense as well), a family, a bride, a body, and a place characterized by fullness or overflow (i.e., πλήρωμα). It seems an entirely plausible candidate for this role; hence, its heavy representation in Laodiceans is, a little ironically, plausibly explicable in contingent terms.

Moreover, we learn from this that any argument for the pseudonymity of Laodiceans in terms of its unacceptable substantive emphases must demonstrate not only that these could not have inhabited any plausible reconstruction of Paul’s coherence (something that must, strictly speaking, take place after framing in any case); it must demonstrate in addition that these could not have been traditional resources within the early church to which Paul might have been appealing in Laodiceans for rhetorical leverage. And this type of challenge now seems to be a very tall order. But the defender of the letter’s authenticity can make a further appeal in this relation that we should note before moving on.

It might be objected that Laodiceans contains an implausible account of Judaism, a challenge that can be couched in slightly more aggressive and comprehensive terms than the contention in terms of the church. On the one hand, certain recognizably Jewish motifs found frequently elsewhere in Paul are almost entirely absent. “Works of Torah” and “faith” are found only vestigially, in 2:8-10,83 while the language of justification (i.e., using the verb δικαιόω or its substantive cognates) is absent altogether. However, there is an evident concern to locate the Christian community in a tightly unified way alongside converts from Judaism. These two groups convert from the same desperate location (2:1-5; see also vv. 11-12) and then form one single body and temple (2:15b, 18), having had their mutual hostility put to death (2:14, 16). Moreover, this unity between Jew and pagan is reinforced by the letter’s later calls for communal unity (4:3-6). The letter breathes a certain sort of ecumenicity.

The ostensible absences from Laodiceans, which raise slightly different methodological issues, will be addressed in the next subsection. Here we will concentrate on the rather programmatic account of Jewish-pagan relations articulated in Laodiceans, along with its emphasis on unity. Certainly, it can be suggested that a later pseudepigrapher, concerned at “the parting of the ways,” might have crafted this impressively interreligious material. But is it implausible to suggest that Paul might have voiced these concerns? Addressing this challenge will in fact highlight the way in which the identification of “Ephesians” as Laodiceans within our developing frame can provide distinctive rejoinders to various challenges to its authenticity.

According to our frame, Laodiceans was written during an incarceration in 50 ce, while Paul was returning to the Aegean from a critical meeting in Jerusalem that took place in turn on the heels of a dramatic confrontation in Syrian Antioch over the status of Paul’s pagan converts. Thus, it is possible that the carefully nuanced account of Christian unity found in Laodiceans may reflect in part these meetings that had just preceded it. (Certainly, they would have informed it.)84 And it is important to recall that the immediate outcome of Paul’s second visit to Jerusalem was positive. The leaders of the two main missions in the early church — to Jews and to pagans — had shaken hands in a binding mutual acknowledgment of their participation in a society together (Gal 2:1-10, esp. vv. 8-9), and the collection was now tangible evidence of Paul’s and his pagan converts’ commitment to that society. Hence, arguably, the carefully nuanced account of communal unity found in Laodiceans, embracing the execution within the assembly of the enmity often existing in general between Jews and pagans, fits this situation like the proverbial hand in a glove — and may in turn shed light on some of the content of those earlier meetings, which was so critical and yet whose attestation is so limited.85 That is, the presence of both the Jewish and the “ecumenical” material in Laodiceans seems well explained by its location in our broader letter sequence just after the Jerusalem conference.

This response should also serve to indicate how challenges to the authenticity of Laodiceans on grounds of any inappropriate substantive emphasis are quite difficult to make cogently, and reasonably easy to rebut. Such a defense can appeal, first, to the distinctive contingency of Laodiceans (with countervailing accounts of Paul’s coherence lacking viability or validity at this point); second, to the way Paul might be utilizing traditional material from the early church that was shared by the Laodiceans for rhetorical leverage; and third, to the location of Laodiceans in a distinctive window within Paul’s life, after important discussions in Syrian Antioch and Jerusalem that had affirmed the unity of the early church and the validity of Paul’s distinctive mission to paganism. These are powerful interpretative resources for any advocate of authenticity against any charges of inappropriate substantive emphases in Laodiceans. But if distinctive substantive emphases can be accounted for, does the same apply to distinctive absences, one of which we have already noted?

(5) Unacceptable Substantive Omissions

In a mirror image to the preceding concern, it might be objected that certain key emphases found elsewhere in the authentic Paul are missing from Laodiceans, thereby calling its authenticity into question. We have already noted briefly the main candidate in this relation, namely, Paul’s “justification discourse,” as I refer to it elsewhere (2009). This is a distinctive cluster of words and phrases centered on the phrase “works of Torah” (ἔργων νόμου) and words using πιστ- and δικαιο- stems, usually translated in terms of faith and belief or of righteousness and justification, respectively. (Other words and motifs are associated with this central group but need not detain us here.) Protestant interpreters of Paul have frequently argued that this cluster of material is either the heart of his coherence or a significant part of it. So the vestigial acknowledgment by Laodiceans of just two of the three critical motifs in this cluster, in 2:8-10, is troubling to such interpreters, and they tend to query the letter’s legitimacy accordingly.

However, this challenge in terms of absence is as inadequate as the preceding challenges in terms of unacceptable presence. In the first instance, it should be recalled that all such substantive assertions are moot during preliminary framing. But even granting them here for the sake of argument, it is questionable simply on its own terms that this cluster is necessarily coherent for Paul.

It occurs only in parts of Romans (principally 1:16–5:2 and from 9:30, fading by 10:17) and Galatians (from 2:15, fading through to 3:26, recurring in 5:5-6), and much more briefly in Philippians (see 3:6 and 9 in the context of 3:1-11, i.e., it seems to have characterized the Previous Letter to Philippi). This is not a promising distribution for material supposedly coming from Paul’s properly basic convictions. Hence, when we turn to Laodiceans, its passing acknowledgment in 2:8-10 might well be an appropriate reflection of its relative importance alongside the rest of Paul’s concerns, viewed in the broader context of his authentic letters — and this is, in any case, a stronger representation of the material than we find in 1 and 2 Thessalonians, Colossians, and Philemon. (There are fragmentary occurrences in 1 Cor 1:30; 6:11; and 2 Cor 5:21; although none of these instances is as full as Laod 2:8-10.) But Laodiceans has an even stronger card to play in its own defense at this point, indicating here how it can defend itself against any or all charges of substantive silences concerning material well represented in the major letters.

Because of its location prior to the composition of these letters, Laodiceans can turn the tables on these objections and suggest that when this distinctive material occurs in a letter written later, we have a good prima facie case for that material’s fundamental contingency. The boot, it seems, is on the other foot. Advocates of the importance of material deemed absent from Laodiceans but present only in parts of 1 Corinthians, 2 Corinthians, Galatians, Philippians, and/or Romans must now make a case for the importance and centrality of that material — a job that will have to take place after preliminary framing. Conversely, Laodiceans itself might suggest that this material was highly circumstantial. So, for example, Paul’s justification discourse might have been mobilized in depth only when the Teacher was in view, an overtly circumstantial exigence. Otherwise, Paul barely mentions this distinctive material, indicating thereby that it should perhaps be excluded from any significant role in his coherence, or, at least, positioned toward the edge of it.

In short, this challenge in terms of a supposedly pregnant substantive silence in Laodiceans should be laid to one side — although by considering it, we have identified another strength in the location of Laodiceans in 50 ce within our broader frame. Not only does this allow us to explain certain emphases within Laodiceans (and Col) as possible echoes from the meetings that had just taken place in Syrian Antioch and Jerusalem, but it removes the need for Laodiceans to justify any of its silences concerning material that occurs in the later letters. Because these letters did not exist when Laodiceans was written, any silences in relation to their concerns are hardly a problem automatically. Still, if various substantive silences are no longer threatening to Laodiceans, certain initially more innocuous personal silences might be.

(6) An Absence of Particular Features

One of the most troubling features in the case for the authenticity of “Ephesians” as Laodiceans is not its set of substantive absences so much as its cluster of personal omissions.

The letter ending is abrupt, and it contains a number of distinctive features. It reproduces material from Colossians on the role of Tychicus almost exactly in 6:21-22 (without pronouncing on the underlying causality here). A somewhat elaborate peace wish follows in v. 23 (Εἰρήνη τοῖς ἀδελφοῖς καὶ ἀγάπη μετὰ πίστεως ἀπὸ θεοῦ πατρὸς καὶ κυρίου Ἰησοῦ Χριστοῦ), and then a grace wish in v. 24. But this last is directed in atypical fashion, in the third person, to “those who love our Lord Jesus Christ,”86 as against in the second person, to those who are receiving the letter, Paul’s invariable practice elsewhere. And it is combined with a troublesome phrase — “through immortality” (ἐν ἀφθαρσίᾳ).87

Admittedly, we have seen many times already that distinctiveness alone is not enough to call Paul’s authorship of a text into question, but here it combines with a puzzling omission. Alone among the letters considered thus far — all of which have been judged authentic — Laodiceans omits any greetings. This is odd in and of itself. But it is especially troubling when we note that the two letters addressing congregations personally unknown to Paul — namely, Romans (16:3-16, 21-23) and Colossians (4:10-15, also 18a) — seem to go out of their way to craft especially elaborate greetings. To make matters worse, both of those sets of greetings emphasize Jewish connections where possible, something that would presumably have been especially useful in Laodiceans, which is evidently concerned to promote unity between pagan and Jewish converts. This is an argument from silence, but the silence in Laodiceans does seem to be a pregnant one. We have strong expectations that Laodiceans would send numerous personal greetings to the Laodiceans, and emphasize any Jewish converts in doing so, but it does not, prompting doubt about the letter’s authenticity.

It is also slightly odd that Colossians emphasizes Jewish connections in its elaborate greetings when the letter itself is not especially concerned with Jewish-pagan relations or unity in the way that Laodiceans is. But herein may lie the first important clue to a possible solution to this problem. Perhaps tellingly, the Laodiceans figure prominently in these greetings. Epaphras is said to struggle for them as well as the Colossians and those in Hierapolis in v. 13; the Colossians are told to greet them in v. 15; and the letter is to be sent to them in v. 16, and their letter read to the Colossians as well. And this last set of instructions by Paul recalls the claim made earlier in our initial consideration of the immediate implied exigencies of both letters.

By the time Paul writes Colossians 4:16, he is envisaging an integrated epistolary event at Colossae and Laodicea; both letters are to be read out at both locations. So the poverty of greeting at the end of Laodiceans is potentially explicable in terms of the wealth of contact and information that Paul provides at the end of Colossians. All of this would have been heard by the Laodiceans, and they are explicitly engaged by the material. This explanation has the added virtue of explaining why the Jewish credentials of some of Paul’s companions are emphasized so strongly in Colossians when this seems otiose in the context of the Colossian difficulties. These emphases speak directly to the situation at Laodicea. Paul’s circle of coworkers, which labors vigorously together and includes converts from both Judaism and paganism, is an instantiation of the community that he advocates in Laodiceans. (It is not inconsistent with the ecclesiology advocated by Col, but such concerns are not especially overt there.)

Once all these pieces of evidence have been grasped, any doubts in this relation tend to evaporate — although in confronting them, we do seem to learn more about the urgency gripping the final stages of these letters’ preparation.88 Time, access, and/or energy seem to have been so lacking that parts of the letter openings and closings were mechanically reproduced, and the extensive greetings and practical instructions occurring at the end of one of the letters were effectively made to do double duty in relation to the recipients in both key locations. An epistolary process that seems to have begun with nothing but time apparently ended under pressure from an acute lack of it.

And with this concern addressed — in my view one of the most serious — we can turn to our final challenge in this relation.

(7) Style

The authenticity of “Ephesians” has often been challenged on grounds of its ostensibly distinctive style. But what we have said before remains relevant. It is true that tabulations can be made of various stylistic features that are particularly distinctive to “Ephesians,” most notably, sentence length. But many stylistic features are shared unremarkably with other undisputed letters.89 So the key question is, again, not so much whether “Ephesians” has distinctive stylistic features, because it does, as whether those are different enough from the style of the other Pauline letters to suggest a judgment of pseudonymity. The statistical analysis of this question is at best ambiguous, and at worst negative.

Those statistically astute analysts who have addressed the issue differ in their results. Ledger and Mealand view the differences apparent in “Ephesians” at certain points as significant, while Neumann does not. But not all the samples used by Ledger and Mealand are in fact significantly different — only those taken from the first part of the letter (see the plots introduced earlier in relation to Colossians). And Kenny does not detect any statistically significant variations. However, we can add some further dimensions to this rejoinder in the case of Laodiceans, looking in particular at Paul’s crafted rhetorical engagement with his audience.90

I have already suggested that Paul utilized traditional material to appeal to the Laodiceans. That is, Laodiceans is a sophisticated account of Christian identity that offers far more than merely cognitive suggestions. It was a rhetorical intervention. As such, it probably sought to affirm where it could the Laodiceans’ existing Christian identity, as well as to build on that. And it seems to have done so by working its claims into the language of Jewish worship. The letter is strongly infused with eulogy and prayer. There are moments of doxology, and probable hymnic fragments.91 And this all stands to reason. The Laodiceans, although consistently described as pagan converts, most likely came from the fringes of Judaism. If they were familiar with the synagogue and its worship practices, then these appeals on Paul’s part are quite understandable. That they have interacted previously with Judaism seems, moreover, to be probable.

Paul’s mission seems to have had a radical reputation, garnered by engaging with pagans directly and then refusing, after any conversions, to teach a Christian ethic based on full Jewish Torah observance. This mission and ethic caused conflict in the early church, and he had to defend it cleverly and vigorously — which he seems to have done successfully (at least for the time being) when Laodiceans and Colossians were composed. But it follows from this that most other evangelizing of pagans was probably less radical than Paul’s. We gain a sense of the resulting different ethical locations of various pagan converts when Paul addresses the practices of the weak and the strong in Rome and Corinth (see esp. Rom 14; 1 Cor 8, 10). Some pagan converts were clearly comfortable with abandoning more of the Torah than others. Hence, the weak apparently continued to observe Jewish dietary and temporal regulations. All of this combines to suggest that any pagan converts not converted by Paul or one of his proxies would have been rather more likely to have come from a more observant Jewish context than not. They would, in short, most probably have been God-fearers. And if this was the case, then it goes some way toward explaining the distinctive style of Laodiceans.92

Recalling Paul’s probable rhetorical strategy of affirming and building from the Laodiceans’ own traditions, as well as the distinctive commitment of the text to different aspects of the language of worship, we might surmise that in Laodiceans Paul deliberately wrote a text that a gathering of former God-fearers would recognize and appreciate by evoking the language of the synagogue or some similar context involving Jewish worship. And this language, with its special rhythms and ritual uses, would have affected his vocabulary and syntax. It is unnecessary to press any deeper at this point to an identification of where the Laodiceans might have fallen on a broad map of Judaism in Paul’s day.93 Such specificity is not necessary for our present purposes. We need only to note here that the distinctive features of the letter’s style are plausibly explained by a deliberate evocation on Paul’s part of some of these forms. The Laodiceans’ location and his rhetoric, moreover, would have made these appeals quite distinctive within his corpus as a whole (as it is extant). With this realization, any concerns about the authenticity of Laodiceans in terms of style should dissipate, at which point we are ready finally to reach a judgment concerning its authenticity.

3.4 Judgment

Our consideration of this question began, as usual, by assuming the letter’s authenticity, something that led in this instance to the identification of “Ephesians” as Paul’s “lost” letter to the Laodiceans (Col 4:16b). This led quickly in turn to the location of Laodiceans alongside Colossians and Philemon within our developing frame, in 50 ce. Hence, the letter now falls, like Colossians and Philemon, in Paul’s mission to Asia that took place just after the important events surrounding his second visit to Jerusalem, but prior to the tempestuous year of crisis that unfolded through 51 and the spring of 52 ce, the year in which he wrote most of his extant letters. And in the light of these initial framing decisions, all the different challenges to the authenticity of “Ephesians” have proved to be indecisive or have simply collapsed.

The contention of pseudonymity can point to no decisive evidence in terms of incontestable anachronisms or the like. Nor are any problems evident in the letter’s later transmission (although we have yet to discuss this question fully; but the next chapter will supply further grounds for our confidence in the letter’s authenticity in these terms). So the case for pseudonymity had to rely on different challenges, and consequently on the usual suspects — purported problems in the letter’s dependence, substance, and style.

These were by no means inconsiderable. But closer scrutiny has suggested that they are all answerable. It is worth emphasizing, moreover, that this defense has drawn frequently on the particular location of Laodiceans in the frame, in 50, after Paul’s second visit to Jerusalem but before Paul’s year of crisis. This location offered a number of quite specific insights into the composition of Laodiceans, many of which went on to defuse concerns with the letter’s authenticity.

Hence, it follows that the letter should be judged authentic. But it also follows that it has been judged so only in these terms and in this particular location — with “Ephesians” identified as Laodiceans and located in 50 ce. That is, I am not convinced that my defensive rejoinders hold good in any other location (i.e., later). Pseudonymity seems to me to be a better explanation for this letter’s dynamics than its authentic composition either as a circular or in a later imprisonment in Rome (although, strictly speaking, we are not considering this location yet). But this is possibly not surprising. It seems that when we remove Laodiceans from this particular identification and from a relatively early location in our frame, we generate a number of the interpretative difficulties that have elicited a judgment of pseudonymity. That is, many modern scholars seem to have reacted to other improbable provenances for Laodiceans understandably, with a judgment of pseudonymity. But my principal suggestion here is that these difficulties can be resolved by the biographical expedient, achieved by accurate preliminary framing, of an alternative provenance.

Hence, we will proceed here on the assumptions that the designation and identification of “Ephesians” as Laodiceans is correct, along with its location in 50 ce, and that the letter is authentic. We have now affirmed the genuineness of ten Pauline letters in total, although the name of one of the letters so identified will probably strike most interpreters as a little unexpected. Nevertheless, we are now working with a “ten-letter canon” — in chronological sequence, 1 Thessalonians, 2 Thessalonians, Laodiceans/Colossians/Philemon, 1 Corinthians, 2 Corinthians, Galatians, Philippians, and Romans. And if this judgment proves correct, I recommend that Bibles and New Testaments be amended accordingly. (There is of course nothing sacred about a letter’s later canonical designation.) Paul did not write a letter to the Ephesians, and it causes a lot of trouble to think that he did so. The letter so designated was sent to the Laodiceans as referenced by Colossians 4:16b.

1. Baur (2003 [1845], 2:80).

2. Somewhat strangely, this feature of the data is much overlooked. The leadership of the house church is usually attributed to Philemon. However, the syntax of v. 1 makes the most sense if the church meets in Archippus’s house, and his presence in the prescript is otherwise inexplicable. (Col, if authentic, supports this identification as well; see 4:17.) Apphia is explicable as Philemon’s wife or, less likely, his sister. The only reasonably extensive analysis of Archippus is by John Knox (1959). He theorizes that Archippus was Onesimus’s owner as well as the host of the church addressed by the letter known as Philemon. But this reading leaves the first two names in the prescript opaque and hence is as unsatisfactory in immediate terms as the traditional reading that it seeks to supplant.

3. Bellen (1971, 18, 78) arguably made this suggestion earlier. Its subsequent qualification does not affect our case here. Having said this, I have not found its attempted refutation plausible: see Harrill (1999); Wengst (2006, 31-32). For its subsequent — learned — endorsement, see Arzt-Grabner (2003; 2010). Useful general treatments of slavery during the Roman era may be found in Bradley (1987 [1984]; 1994); more controversial is Patterson (1982). An important work treating slavery in relation to the NT is Glancy (2006 [2002]).

4. For some useful points of comparison, ORBIS (May 24, 2013) suggests that a fast journey between Ephesus and Laodicea (ad Lycum) would have taken just under 5 days (4.8) at roughly any time of year (winter barring travel rather than slowing it in these terms), and from (Asian) Apamea or Pisidia to Laodicea, just under 3 (2.9).

5. It might be suggested in support of a more “traditional” location at Caesarea or Rome that Paul is writing as an “old man” (πρεσβύτης; 9). This is the only ostensible reference to Paul’s age that we have from his writings and so is worth considering carefully. But (1) it is indecisive for dating purposes, since we do not know from the letters how old Paul was when he began his ministry, and he could have been an “old man” in 50 ce, albeit an active one (Acts data might stand against this but should not be appealed to yet; see νεανίας, “youth” in 7:58); (2) it is not necessarily a reference to old age in any case, referring in some instances, like πρεσβεύτης, to ambassadorial status and functions (Polyaenus 8.9.1; BDAG) and in others, as a technical term, to status in a group so designated (i.e., one of “the elders” is not necessarily old); and (3) this last reading is possible in place of πρεσβύτης in v. 9 (Bentley’s conjectural emendation; see BDAG). Ambassadorial references occur elsewhere in Paul (see esp. 2 Cor 5:20; see also Eph 6:20), whereas πρεσβύτης is unattested elsewhere. Scribal confusion between the two signifiers would have been understandable, and the diplomatic reading makes excellent sense in the immediate context. That this important ambassadorial figure, who is enduring such humiliation, nevertheless implores Philemon on Onesimus’s behalf is a powerful rhetorical appeal in terms of both ethos and pathos. (This is not to discount the appeal of an imprisoned old man petitioning Philemon.)

6. BDAG citing CIG 3:1168, 4380k, 3; this is helpfully discussed by Barth and Blanke (2000, 254).

7. This takes our developing frame close to Riddle’s (1940) position, as well as, to a degree, to Duncan’s (1929), and to some of Knox’s (1990) comments made late in life. But most scholars sensitive to the importance of Paul’s Asian mission for the letters are preoccupied with his activities in Ephesus, presumably being influenced here — somewhat inconsistently — by the Acts narrative. And I see no decisive epistolary evidence for this preoccupation, at least yet.

8. ORBIS suggests that even a northern east-west route across “Turkey,” passing through Ancyra, could have gone through Pisidian Antioch/Apamea if a southerly route from Dorylaeum to Ephesus was taken, via Pisidian Antioch and the mouth of the Lycus valley, as against a more westerly route through Sardis and Smyrna. A more southerly east-west route across Turkey would almost certainly have gone past the Lycus valley. Having said this, a north Galatian provenance for that letter might open up the possibility of an imprisonment in the Hermus River valley to the north, somewhere in the vicinity of Sardis — the outlet for the second major (more northern) east-west route through Anatolia (the Lycus valley being on the major southern east-west route). A third alternative might be some location farther west from the Lycus valley, in the valley of the Maeander River, hence closer to Miletus and Ephesus. But later data will confirm the likelihood of a provenance to the east of the Lycus valley location, at which point composition prior to the visits to Antioch and Jerusalem seems highly unlikely as well. (And even if Acts data is introduced, we still cannot suppose with any confidence that Paul took an inland route from Ephesus to either Antioch or Jerusalem; a sea route seems much more probable. ORBIS (March 27, 2013) estimates for a journey to Antioch in October a cautious sea voyage of 12-15 days (12.3-15.1; i.e., traveling only by day and along the coast) but a journey overland, on foot, of around 37.

9. Their specific window extended, according to previous data, from late 49 ce through mid-50. But a cutoff date of early 50 for this window looks likely if our suppositions here about Phlm are correct. Paul’s letter to Philemon and its associated events need to fall in an imprisonment between the Antioch incident / second visit to Jerusalem and the composition of the Previous Letter to Corinth in the fall of 50, by which time Paul is apparently unfettered and writing from Ephesus. And this looks like too much of a squeeze if Paul is in either Jerusalem or Syrian Antioch in mid-50. But there is no pressing need for this to have been the case, i.e., for the second Jerusalem visit and/or the incident at Antioch to have taken place late in the relevant chronological window. If they fell earlier, then there is plenty of time for all the implied events to have unfolded during the broader interval.

10. Baur (2003 [1845], 2:84) argues eloquently that the letter is a Christian romance that conveys “a genuine Christian idea” expressed most clearly in v. 15 (τάχα γὰρ διὰ τοῦτο ἐχωρίσθη πρὸς ὥραν, ἵνα αἰώνιον αὐτὸν ἀπέχῃς) — “that what one loses in the world, one recovers in Christianity, and that for ever; that the world and Christianity are related to each other as separation and reunion, as time and eternity.” It is an ingenious reading, but it fails to account for most of the text — illustrating nicely that with sufficient determination and imagination, a “pseudonymous” reading can be supplied for virtually any text.

11. I am grateful to Yuriy Golota for drawing this argument to my attention.

12. First Clement 5:6 suggests that Paul was imprisoned seven times, although this number is clearly quite possibly legendary. For general background on Roman law, provincial administration, and justice, see the essays of Bowman, Frier, Galsterer, and Treggiari in Bowman, Champlin, and Lintott (1996); for earlier background under the republic, the essays of Cloud, Crook, and Richardson are also useful in Crook, Lintott, and Rawson (1994). Millar (1984) provides a typically learned account of imprisonment during this period — an underresearched area. For analyses specifically of Paul in prison, see Rapske (1996 [1994], which is focused particularly on Acts data); and Wansink (1993/1996). Brent notes the helpful remark of Lucian, Peregrinus 12-13 (2007, 50-51), which suggests that Christian leaders could be relatively well supported during imprisonment, although this refers to conditions in the second half of the second century ce.

13. The scholarly discussion of this phenomenon is generally disappointing. A relatively recent bibliography can be found in Thiselton (2000, 59); it is indicatively short. Ellis (1971) is a classic introduction to the data, updated and summarized in 1993.

14. [1:7] καθὼς ἐμάθετε ἀπὸ Ἐπαφρᾶ τοῦ ἀγαπητοῦ συνδούλου ἡμῶν, ὅς ἐστιν πιστὸς ὑπὲρ ὑμῶν διάκονος τοῦ Χριστοῦ, [8] ὁ καὶ δηλώσας ἡμῖν τὴν ὑμῶν ἀγάπην ἐν πνεύματι.

15. The “Pauline circle” has been discussed by scholars, although not always satisfactorily; Ollrog (1979) is precise. This useful designation derives from the title of a brief treatment by F. F. Bruce (2006 [1985]).

16. Generalizations about ancient literacy have to be ventured carefully. A standard introduction is W. Harris (1989). Further nuancing can be found in (i.a.) W. Johnson (2000; 2010).

17. This would have increased the number within the Pauline circle to eight, with the recruitment of Onesimus taking the figure to nine, three of whom would have been slaves — Tychicus (Col 4:7), Epaphras, and Onesimus.

18. Some variations on this construal of the data are possible at this point and are worth noting but will be excluded later, so I have not introduced them as major interpretative variations here. In particular, we do not know from Col that it was Onesimus who brought this disquieting information to Paul. It could have been Epaphras who brought the information on his return visit. Or both figures might have contributed to a composite picture. But when Col is coupled tightly with Phlm (introducing framing considerations again), it becomes apparent that Paul wrote two letters at largely the same time to respond to the situation at Colossae. And we know that Phlm responds to the arrival of Onesimus. This would suggest that the information lying behind Col came from Onesimus’s arrival as well. Certainly, this is the most economic explanation of the situation.

19. A seminal analysis is Bauer (1971 [1934]), although it is now much developed and disputed.

20. To dub this “the Colossian philosophy” would be almost as problematic again, since it would assume a philosophical influence. This might be correct, but it is debatable.

21. [2:6] Ὡς οὖν παρελάβετε τὸν Χριστὸν Ἰησοῦν τὸν κύριον, ἐν αὐτῷ περιπατεῖτε [7] ἐρριζωμένοι καὶ ἐποικοδομούμενοι ἐν αὐτῷ καὶ βεβαιούμενοι τῇ πίστει καθὼς ἐδιδάχθητε κ.τ.λ.

22. 2:4: Τοῦτο λέγω, ἵνα μηδεὶς ὑμᾶς παραλογίζηται ἐν πιθανολογία κ.τ.λ.; 8: Βλέπετε μή τις ὑμᾶς ἔσται ὁ συλαγωγῶν διὰ τῆς φιλοσοφίας καὶ κενῆς ἀπάτης κατὰ τὴν παράδοσιν τῶν ἀνθρώπων κ.τ.λ.; 16: Μὴ οὖν τις ὑμᾶς κρινέτω ἐν βρώσει καὶ ἐν πόσει ἢ ἐν μέρει ἑορτῆς κ.τ.λ.; 18: μηδεὶς ὑμᾶς καταβραβευέτω θέλων ἐν ταπεινοφροσύνῃ καὶ θρησκείᾳ τῶν ἀγγέλων, ἃ ἑόρακεν ἐμβατεύων, εἰκῇ φυσιούμενος ὑπὸ τοῦ νοὸς τῆς σαρκὸς αὐτοῦ κ.τ.λ.

23. On the one hand, it is unlikely that just one figure was involved; people generally traveled and worked in groups in the ancient world. On the other hand, it is likely that the intellectual impetus for this group came from a single figure, rather as the impetus for the Pauline circle seems to have come from Paul. So alternations between singular and plural designations in what follows do not signify referential alternations in the numbers of people involved.

24. Paul’s letter to the Philippians is partly concerned with a third party, although I have suggested that it quotes a previous letter that seems to have been primarily concerned with that intrusion. Both of Paul’s extant letters to Corinth are concerned with third parties, 1 Cor significantly so and 2 Cor almost entirely. His letter to the Galatian congregations is obviously so concerned, and I have argued elsewhere that Rom is ultimately comprehensible only in these terms (2009, 469-518). Hence, a significant portion of Paul’s corpus — Phil, 1 and 2 Cor, Gal, and Rom — is dominated by this function. Paul’s second letter to Thessalonica was probably prompted by the malicious actions of a third party, in a fake communication suggesting that the day of the Lord Jesus had already come. And this leaves only 1 Thess and Phlm addressing issues that seem to have arisen from their congregations largely internally, although Phlm is of course something of a special case, since it is a short letter accompanying another letter and hence is not an entirely independent rhetorical act on Paul’s part.

25. Presumably, Nympha’s house church was at Hierapolis but was to be folded into the instructions concerning Laodicea, thereby indicating the latter’s greater size and importance; see Col 4:15 and 13.

26. [Col 3:9] μὴ ψεύδεσθε εἰς ἀλλήλους, ἀπεκδυσάμενοι τὸν παλαιὸν ἄνθρωπον σὺν ταῖς πράξεσιν αὐτοῦ [10] καὶ ἐνδυσάμενοι τὸν νέον τὸν ἀνακαινούμενον εἰς ἐπίγνωσιν κατ’ εἰκόνα τοῦ κτίσαντος αὐτόν, [11] ὅπου οὐκ ἔνι Ἕλλην καὶ Ἰουδαῖος, περιτομὴ καὶ ἀκροβυστία, βάρβαρος, Σκύθης, δοῦλος, ἐλεύθερος, ἀλλὰ τὰ πάντα καὶ ἐν πᾶσιν Χριστός κ.τ.λ.; see 1 Cor 12:13: καὶ γὰρ ἐν ἑνὶ πνεύματι ἡμεῖς πάντες εἰς ἓν σῶμα ἐβαπτίσθημεν, εἴτε Ἰουδαῖοι εἴτε Ἕλληνες εἴτε δοῦλοι εἴτε ἐλεύθεροι, καὶ πάντες ἓν πνεῦμα ἐποτίσθημεν; Gal 3:27: ὅσοι γὰρ εἰς Χριστὸν ἐβαπτίσθητε, Χριστὸν ἐνεδύσασθε [28] οὐκ ἔνι Ἰουδαῖος οὐδὲ Ἕλλην, οὐκ ἔνι δοῦλος οὐδὲ ἐλεύθερος, οὐκ ἔνι ἄρσεν καὶ θῆλυ· πάντες γὰρ ὑμεῖς εἷς ἐστε ἐν Χριστῷ Ἰησοῦ. For further details, see my 1996.

27. Having said this, the implicit engagement of Col with the discourse of slavery is complex and in certain respects at least arguably subversive. Barclay (1991) is an especially interesting analysis.

28. That is, not much changes if Paul is traveling at this time south, down from the main northern east-west route across Anatolia that ran through the “ethnic” Galatian communities. ORBIS suggests that this journey would have proceeded roughly from Germa through Dorylaion/Dorylaeum, then headed south, intersecting the main southern east-west route not far to the west of Pisidian Antioch. (The Antonine table lists Pessinus as sixteen Roman miles from Germa, the latter not existing in Paul’s day.) A journey from Germa to Laodicea would have averaged 12.6 days (454 km); from Dorylaion, 9.8 days (351 km).

29. ORBIS (May 29 and 31, 2013) suggests that a sea journey in November from Antioch to Jerusalem (including travel closer to these destinations on foot and river) would have taken between 8 and 13 days (7.8 and 12.6). A sea journey at that time from Ephesus to Antioch using a coastal route would have taken 7.1 days; using that route only during the day, 14.6. But the clock starts ticking (so to speak) for this marker only from Paul’s arrival in Jerusalem, and his stay there may not have been that long. In January a return journey by sea to Antioch would have averaged between 8 and 14 days, depending on the precise mode (i.e., risking the open sea, traveling by way of the coast, or traveling coastally only during the day). A journey overland from Antioch to Ephesus (1073 km) irrespective of the season would have averaged 30.8 days, but as far as Apamea, 23.1 days (796 km), assuming the Cilician gates were open.

30. I am not convinced that more complex solutions involving multiple stages are particularly plausible — for example, Holtzmann’s suggestion that both Eph and Col derive from an earlier authentic proto-Colossians.

31. Leppä (2003, 17-18) cites other advocates and endorsers of dependence, most notably, Schenk (1983; 1987); and an advocate of an alternative type of dependence, Kiley (1986).

32. Sanders makes an interesting concession in this relation: “Behind the attempt to establish such criteria [i.e., sufficient to distinguish the parallels between a letter plausibly written by Paul and a letter written by a later writer with access to his writings] is the assumption that Paul did not carry about with him copies of his previous letters, or, if he did, that he did not have them memorized nor did he consult them when composing a new letter. If this assumption is false, it would be difficult to distinguish Paul’s work from that of a later imitator” (1966, 29-30). Sanders goes on to argue that this assumption “is probably not false, however, for there is no evidence for the literary dependence of any one of Paul’s seven undisputed letters upon another, much less upon several” (30). But this rejoinder is, in the first instance, a non sequitur; that Paul did not quote earlier material in some cases does not prove that in others he did not. And it is also, second, not quite accurate. A certain dependence between 1 Thess and 1 Cor has already been noted, and many close relationships are discernible between Rom, Phil 3:2-11, and Gal, and between 1 Thess and 2 Thess (although Sanders would probably take this last relationship as evidence of the pseudonymity of 2 Thess). Third, Sanders actually makes three assumptions here, not one, and all have effectively been refuted by the data. Paul did most likely carry about copies of his letters (so Richards 1991, 165 n. 169; 2004, 156-61, 214-23; Murphy O’Connor 1995, 37), did have them memorized (and this is a strange denial for someone like Sanders in any case — an advocate of the importance of rabbinic sources for Pauline analysis), and did apparently consult them when composing new letters, whether to a greater or lesser degree, as his embedded quotations suggest.

33. It is important to appreciate that to have demonstrated the presence of what we might call autodependence by Paul is not necessarily to have explained it plausibly in deeper terms, i.e., to have understood why he did this comprehensively. Several possibilities spring to mind, although this is not the place to analyze them: a rabbinic mind-set; a philosophical mind-set; an “oracular” mind-set; or perhaps, rather differently, what we might call GSL (the ancient equivalent of ESL). I have gestured toward some options already in ch. 4 — see perhaps most auspiciously 2 Thess 2:15.

34. Sanders adds a fifth set of parallels, less persuasive by his own admission, about which little further needs to be said except that baptismal parallels are again cited unremarked: see, to a degree, elements of Rom 13:12b-14, 1 Thess 4:5, 1 Cor 5:11, Gal 5:19, Rom 6:6, Gal 3:27-28, 1 Cor 12:13, and 1 Cor 15:28 in Col 3:5-11. Note that these echoes are interesting, and perhaps very much so. They might shed important light on the critical matter of Paul’s coherence. But used in this way, they tell us nothing decisive about framing.

35. An earlier statistical controversy might remain instructive. Morton (and McLeman 1966; also Morton 1978), relying in the main on data related to sentence length, concluded quite confidently that Paul wrote only Rom, 1 Cor, 2 Cor, and Gal. Kenny (1986, 101-15) placed pressure on this argument in various ways, one of which was the demonstration that according to Morton’s data, 1 Cor struggles to come from the same authentic sample as any other Pauline letter except Gal and Rom (and he demonstrated the same problem in certain statistical analyses of some of Aristotle’s works). This is not to suggest that the later studies of the NT noted here are anything like Morton’s infamous analyses, whether in relation to the NT or to other texts and problems. But Morton illustrates the perennial problem of any statistical analysis of a given set of data — namely, whether the data collected, tabulated, and statistically analyzed reflects a real situation reliably or is either corrupted or incomplete in some way and therefore misleading. I remain hopeful that this subdiscipline will shortly yield vital insights into Pauline authorship — my hopes resting mainly on Mealand and Abakuks — but I am not convinced that corroborative results have yet been achieved entirely reliably.

36. So Juola (2006); Burrows’ Delta test is also apparently effective here.

37. Moreover, some cross-cultural hermeneutical humility might be called for here as well. It seems that modern Western scholars have been overestimating their ability to judge texts written in Koiné and circulated within a Greco-Roman culture that occupied the Mediterranean littoral around two thousand years ago. Indeed, perhaps many of our conclusions have been possible only because all the figures linguistically competent to judge and, if necessary, to correct them are dead.

38. Some questions will have to be asked in due course about whether a bell curve, premised on biological data and genetics, is an appropriate underlying configuration for the mapping of linguistic entities. The important companion notion of a literary “fingerprint” must also be examined. It may be that language is best mapped by another model altogether — perhaps a network that utilizes a “power law” rather than a bell curve. Statistical analyses and results would shift accordingly (and Burrows’s Delta method might capture these somewhat unwittingly). Barabási (2003) introduces the broader issues lucidly.

39. The occurrence of πλήρωμα in the NT is dominated by Paul (see, in addition to the two references in Col just noted, Rom 11:12, 25; 13:10; 15:29; 1 Cor 10:26; Gal 4:4; Eph 1:10, 23; 3:19; 4:13) but is also usefully informed by instances in the Gospels (Mt 9:16; Mk 2:21; 6:43; 8:20; Jn 1:16).

40. The verb is much more frequent in the NT than the noun, occurring 99x, 31x in Paul (counting the cognate form ἀναπληρόω).

41. A classic analysis of Middle Platonism is Dillon (1977), where this point is obvious.

42. T. Martin’s (1996) reconstruction of the opponents as Cynics is learned and fascinating but not representative of a major tradition of interpretation — although clearly it is loosely connected to a philosophical approach. And obviously it does not exclude Pauline authorship; Cynics considerably predated Paul. The possibility of a Neo-Pythagorean resonance is worth noting as well (see Martyn 1985). I have not included here any discussion of Hooker’s (1973) interesting suggestion that the opponents at Colossae did not exist in concrete terms. I think that this is highly unlikely. But more to the point, this approach will struggle to affect a judgment of authenticity.

43. The hypothesis that a principal problem facing Paul at Corinth was “overrealized eschatology” is increasingly unpopular currently, although it has an able recent defender in Thiselton (2000). Against this, see (i.a.) Wright (2003, esp. 277-374).

44. See esp. T. F. Torrance (1976; 1986).

45. That is, unless a prior decision has been made that all historical causality is immanent, and consequently that everything Paul writes about is untrue in the terms in which he understood them. See Kerr’s (2009) useful explication and critique of Troeltsch.

46. T. F. Torrance (1986); see also Zizioulas (1985).

47. See esp. Begbie (2013, 141-75).

48. It is unlikely that the brief and generic instance of the Haustafeln in Col 3:18–4:1 is incompatible with Paul’s advice in 1 Cor regarding various questions of gender and sexuality. Prolonged engagement with these texts has convinced me, rather, that such is the inevitable complication and qualification that takes place when simple binary accounts supposedly structuring social realities encounter those realities on the ground. Actual social reality is rather too complex to be susceptible to straightforward binary ordering. Paul’s navigation of these tensions is apparent in Horrell’s (2005) important treatment.

Arguments against the authenticity of Col in terms of bodiliness suffer from the basic difficulty that they must comprehend a complex discourse in Paul that is utilized in quite diverse ways rhetorically yet must argue nevertheless that certain elements are nonnegotiable and other developments are inadmissible — an implausible undertaking. One such argument, although richly informative in other respects, is Dawes (1998). A lively postmodern treatment of Paul’s engagement with bodiliness at Corinth is D. Martin (1995; and parts of his later collection, 2006, are also informative for this question). On the discourse within early Christianity in general, see the classic account of P. Brown (2008 [1988]). General Jewish and NT background can be found in Wright (2003) within the broader context of a treatment of eschatology. A more recent, controversial analysis is Engberg-Pedersen (2010).

The treatment of the assembly (i.e., ἡ ἐκκλησία) in Col is also arguably in tension with the motif’s treatment in the undisputed letters. However, it can simply be granted that the instances in 1:18 and 24 are different, being rather more universal than the signifier’s occurrences elsewhere (including in Col 4:15 and 16). The universal sense apparent in Col 1 is probably evocative of the qahal Israel, that is, the assembly of Israel, which could also have eschatological connotations. But this signification in Col 1 is not necessarily incompatible with Pauline thinking if Paul in Col is citing an early church hymn or confession in 1:15-20, as many scholars have supposed, and then utilizing the theology of that confession in its immediate context. Moreover, in both 1:18 and 1:24 the notion of “assembly” is overtly qualified by the notion of “body,” which is, as we have just seen, an important Pauline discourse. So this data is hardly fatal to Col’s authenticity.

49. We might also want to infer the presence of a false teacher at Colossae. The hints in the text suggest to me (as we have already seen) the presence there of a third party. Moreover, it might be that further data about such a figure would allow us to decide more firmly about which of the options concerning the intrusive problematic teaching is actually correct — reversing here the usual flow of analysis. But this sort of result is one of the benefits we expect from sound preliminary framing.

50. See Wilkes in Bowman, Champlin, and Lintott (1996, 545-85).

51. That Paul’s designations “Galatia”/“Galatians” (Gal 1:2; 3:1; see also 1 Cor 16:1; also Acts 16:6; 18:23; 2 Tim 4:10; 1 Pet 1:1) could denote all these options has been incontrovertible since Stephen Mitchell (1995; summarized and presaged in his 1992). The advocacy of a “south Galatian,” i.e., provincial, provenance for Gal goes back largely to the work of William Ramsay in the late 1800s, which established it as a possibility (see R. Longenecker 1990, lxi-lxxii, esp. lxiii, lxvi-lxvii). As I have suggested earlier, however, in ch. 3, this debate is largely generated by the need to integrate data from Acts with data from the letters. When faced with letter data alone, Ramsay, Mitchell, and those like them have established that a provincial provenance for Gal is possible as against the more traditional northern, ethnic provenance. But I know of no epistolary data allowing a decisive judgment between them. So, for example, any appeals to “biographical markers” like the prominence of Barnabas in Gal (see 2:1, 9, 13) presuppose appeals to data from Acts and so for now must be set to one side (see Longenecker 1990, lxx-lxxii).

That Gal 4:13 implies a second visit by Paul to Galatia in the interim, i.e., prior to the composition of Gal, and as suggested also by 1 Cor 16:1, does seem likely in our reconstruction: “You know that I proclaimed the good news to you formerly / the first time [I was with you] because of a bodily weakness [i.e., an illness]” (οἴδατε δὲ ὅτι δι’ ἀσθένειαν τῆς σαρκὸς εὐηγγελισάμην ὑμῖν τὸ πρότερον). Certainly, there is no need to resist this implication; rather, the developing frame suggests it. Paul is in Corinth when he composes Gal. He seems to have founded the Galatian communities some time ago — at least prior to the start of the Asian mission. Since then he has journeyed to Syrian Antioch and Jerusalem and returned to the Aegean by way of Galatia, leaving oral instructions about the collection. The implications of 4:13 dovetail with this reconstruction exactly. Moreover, the use of the expression τὸ πρότερον in 4:13 seems redundant otherwise. This observation will have an influence on our later processing of Acts data, however slight.

52. 2 Cor 11:26 is also suggestive: ὁδοιπορίαις πολλάκις, κινδύνοις ποταμῶν, κινδύνοις λῃστῶν, κινδύνοις ἐκ γένους, κινδύνοις ἐξ ἐθνῶν, κινδύνοις ἐν πόλει, κινδύνοις ἐν ἐρημίᾳ, κινδύνοις ἐν θαλάσσῃ. These denote, in the main, the quite considerable challenges of ancient travel.

53. On this dimension within the Mediterranean Sea’s ancient history, see esp. Horden and Purcell (2000).

54. The following missions would not necessarily have been in this order, although they should pivot around the Macedonian cities plausibly.

55. These events might have taken place early in 50 ce as against late in 49.

56. Most likely in Apamea, and more certainly, east of Colossae but west of any town with a successful Galatian congregation. The proxy evangelization of Colossae by Epaphras takes place at this time.

57. First personal visit by Paul to Colossae after his anticipated release (Phlm 22), but not a founding visit in this unusual case.

58. Conflating Paul’s contact with Laodicea and Hierapolis, since Col 4:13, 15, and 16 seem to suggest this.

59. We do not know whether this was Paul’s first visit to Ephesus, although it is the first one for which we have direct evidence from his letters. More significantly, we do not know whether he then founded the church at Ephesus or whether a successful mission was already underway under the leadership (presumably) of Prisca and Aquila (see 1 Cor 16:19; also Rom 16:3-5a).

60. Marcion; see also Col 1:2a: τοῖς ἐν Κολοσσαῖς ἁγίοις καὶ πιστοῖς ἀδελφοῖς ἐν Χριστῷ.

61. p46.

62. a*, B*, 6, 1739 [see p46].

63. B2, D, F, G, Ψ, 0278, 33, 1881, latt, sy, co.

64. a2, A, P, 81, 326, 629, 2464.

65. Marcion’s address is triply attested — by Tertullian and Epiphanius, as often noted (and in both figures several times), but also by Theodoret of Cyrrhus (ca. 393-466 ce; Interpretatio XIV epistolarum sancti Pauli apostoli 82.625C). Full references and quotations of the key claims can be found in Van Kooten (2003, 198 n. 79).

66. Lincoln (1990, 3) resists this final simple solution, but on unconvincing grounds, suggesting that the text is unacceptably awkward because of its syllepsis or zeugma. But he overlooks the decisive relevance of Col 1:2 at this point, which contains almost exactly the same syntax. Hence, Eph 1:1b might not be elegant Greek, but it is certainly acceptable — and rather more so than the introduction of an unattested and more complicated conjectured text. Best (1979; 1982; 1987; 1998) has mounted a vigorous and important campaign in behalf of an original generalized address, but Lincoln (1990, 3) rightly repudiates this on the grounds of its overly complex and unattested early steps in the manuscript tradition. Fortunately, no such steps need to be posited for a Laodicean address.

67. This judgment is rare but not altogether unheralded in previous scholarship; see e.g. BeDuhn 2013, 224, esp. n. 76, 252, 309. Rutherford makes the case tidily (1907/1908). And Coneybeare and Howson — who concur — note the earlier agreement of Paley, although the former combine this view with Usscher’s suggestion that a circular letter to Asia based on Laod was sent simultaneously, and this led to the MSS confusion (1878, 812-16).

68. 1:15: Διὰ τοῦτο κἀγὼ ἀκούσας τὴν καθ’ ὑμᾶς πίστιν ἐν τῷ κυρίῳ Ἰησοῦ καὶ τὴν ἀγάπην τὴν εἰς πάντας τοὺς ἁγίους κ.τ.λ.; 3:2: εἴ γε ἠκούσατε τὴν οἰκονομίαν τῆς χάριτος τοῦ θεοῦ τῆς δοθείσης μοι εἰς ὑμᾶς κ.τ.λ.

69. See esp. 2:11-19; 3:6; and, more implicitly, 4:4-6.

70. Discussions that can usefully introduce some of the sociological and interpretative dynamics in play here are Esler (1987; 1994; 1998; 2003; 2005; with 1994, 1998, and 2003 focusing entirely on Pauline texts). Notions of identity must not be restricted to Henri Tajfel’s account, in my view, nor should the interpretation of Rom be restricted to Esler’s reading in terms of the creation of social identity (and, in particular, as Tajfel constructs this). It is interesting to contrast, even in this limited purview, the work of Horrell (see his 1996; 2005). (Esler and Horrell exchange views in Esler 2000a and Horrell 2000.) I make some remarks on questions of identity in Paul in 1996; 2005b; 2013.

71. So Lincoln helpfully suggests that, among other things, “the letter was intended to reinforce its readers’ identity as participants in the Church” (1990, lxxxvi).

72. Certainly, this strategy is frequently discerned in Rom, often specifically in relation to 1:2-4; see my 2009 (esp. 634-36).

73. See Oates (1982, 222-30; also 342-43: “Letter from a Selma Jail”).

74. Another point where recent advances in neuroscience and psychology prove useful. Gilbert (2005, esp. 83-105) introduces this phenomenon nicely.

75. Data in Acts might arguably confirm this scenario in due course, although this might not have become noticeable without the initial establishment of the scenario on epistolary grounds.

76. This is usefully tabulated by Van Kooten (2003, 262-63).

77. It has already been intimated that the instructions concerning two principal letters in Col 4:16 are not decisive for the order of their composition (although I used to view them this way); they probably refer simply to the first destination on Tychicus’s itinerary, where any copying would have had to take place, and therefore not necessarily to the prior existence of a letter addressed to Laodicea. However, reconstruction of the broader situation on other grounds suggests that Laod came first primarily.

78. Laod 6:21-22:

[21] Ἵνα δὲ εἰδῆτε καὶ ὑμεῖς τὰ κατ’ ἐμέ, τί πράσσω, πάντα γνωρίσει ὑμῖν Τύχικος ὁ ἀγαπητὸς ἀδελφὸς καὶ πιστὸς διάκονος ἐν κυρίῳ

[22] ὃν ἔπεμψα πρὸς ὑμᾶς εἰς αὐτὸ τοῦτο, ἵνα γνῶτε τὰ περὶ ἡμῶν καὶ παρακαλέσῃ τὰς καρδίας ὑμῶν.

Col 4:7-8:

[7] Τὰ κατ’ ἐμὲ πάντα γνωρίσει ὑμῖν Τύχικος ὁ ἀγαπητὸς ἀδελφὸς καὶ πιστὸς διάκονος καὶ σύνδουλος ἐν κυρίῳ,

[8] ὃν ἔπεμψα πρὸς ὑμᾶς εἰς αὐτὸ τοῦτο, ἵνα γνῶτε τὰ περὶ ἡμῶν καὶ παρακαλέσῃ τὰς καρδίας ὑμῶν.

79. Laod 1:1-2:

[1] Παῦλος ἀπόστολος Χριστοῦ Ἰησοῦ διὰ θελήματος θεοῦ τοῖς ἁγίοις τοῖς οὖσιν ἐν Λαοδικείᾳ καὶ πιστοῖς ἐν Χριστῷ Ἰησοῦ

[2] χάρις ὑμῖν καὶ εἰρήνη ἀπὸ θεοῦ πατρὸς ἡμῶν καὶ κυρίου Ἰησοῦ Χριστοῦ.

Col 1:1-2:

[1] Παῦλος ἀπόστολος Χριστοῦ Ἰησοῦ διὰ θελήματος θεοῦ καὶ Τιμόθεος ὁ ἀδελφὸς

[2] τοῖς ἐν Κολοσσαῖς ἁγίοις καὶ πιστοῖς ἀδελφοῖς ἐν Χριστῷ, χάρις ὑμῖν καὶ εἰρήνη ἀπὸ θεοῦ πατρὸς ἡμῶν.

80. So (i.a.) Goodspeed (1933); Mitton (1951); Van Kooten (2003).

81. Attention to oral/aural dynamics has been more prominent in discussions of the Synoptic Gospels recently than it has been within Pauline studies. While a certain amount said in the former relation might be helpful in terms of the latter, I would want to emphasize that quite different temporal, circumstantial, and editorial dynamics are present when we turn from the Synoptics to Paul’s letters or vice versa. In the case of Paul, only one person is at present involved, in a relatively constrained situation, for a relatively brief period of time.

82. Echoes of 1 and/or 2 Thess would involve different explanatory dynamics. But we have already touched on the possible influence of these letters long after their composition, partly since Paul seems to have retained copies of them and referred to them when he thought it useful. So later echoes of these letters will be informative but are unlikely to be fatal to the developing frame.

83. [2:8] Τῇ γὰρ χάριτί ἐστε σεσῳσμένοι διὰ πίστεως· καὶ τοῦτο οὐκ ἐξ ὑμῶν, θεοῦ τὸ δῶρον. [9] οὐκ ἐξ ἔργων, ἵνα μή τις καυχήσηται. [10] αὐτοῦ γάρ ἐσμεν ποίημα, κτισθέντες ἐν Χριστῷ Ἰησοῦ ἐπὶ ἔργοις ἀγαθοῖς οἷς προητοίμασεν ὁ θεὸς, ἵνα ἐν αὐτοῖς περιπατήσωμεν.

84. I do not want to insist on this point but simply to list it as an interpretative possibility. The concern in Laod with Judaism may have arisen from a specific aspect of the exigence in Laodicea — although probably drawing in response to that on the material that had been discussed earlier in Jerusalem. Alternatively, Paul might simply have been projecting material from Jerusalem into his current vision of the identity of pagan converts. At this point, during preliminary framing, we cannot tell.

85. Lincoln (1990, lxiv, xciii) suggests that the account of Judaism found in Laod is irreconcilable with the discussion of Rom 9–11. But I find this contention rather odd. Rom 9–11 is concerned at first with pagan inclusion, and it makes no claims that are in serious tension with the fuller account of this dynamic supplied by Laod; indeed, both appeal strongly to the divine initiative, i.e., to election, properly understood. The rest of Rom 9–11 then addresses the general Jewish rejection of the gospel, providing several interlocking explanations of this tragedy. Conversely, Laod does not specifically address the Jewish rejection of Christian claims. It provides an account of Jewish and pagan conversion that creates a new united community (something that might then arguably look like two types of branches being grafted into one single tree). So the two texts basically address different questions. That they consequently often look different may be conceded, but this hardly establishes any irreducible tensions between them. And arguably when they do touch on the same questions, they say much the same things.

86. ἡ χάρις μετὰ πάντων τῶν ἀγαπώντων τὸν κύριον ἡμῶν Ἰησοῦν Χριστὸν ἐν ἀφθαρσίᾳ.

87. This is probably best linked with the benefaction in view, denoting it to be glorious and immortal; for discussion and a list of supporters, see Lincoln (1990, 467-68).

88. There might also be a suggestion here that Paul composed the letter body first, appending an appropriate address and postscript later, during the letters’ final stages of preparation, and probably leading up to his signature — an interesting compositional practice to reflect on.

89. See Kenny (1986).

90. We noted earlier the probability that Laodiceans was composed initially in an oral/aural fashion, a process that makes particular sense in a carceral setting. And some of the distinctive features of Laodiceans noted by scholars are arguably explicable in terms of the standard practices of oral/aural composition; see Ong (2002; 2012 [1982]).

91. I treat this in more detail in my 2009 (624-38).

92. Cohen (1999) is a helpful orientation in the broader context of a nuanced discussion of ancient Jewish identity. The key restoration of the “God-fearers” to validity, on inscriptional grounds, after their characterization as a fictitious literary construction by Kraabel et al. is Reynolds and Tannenbaum (1987). For further discussion, see Trebilco (1991, 145-66).

93. Lincoln (1990, lxxv, xcii-xciii) makes some preliminary observations here.