How reliable is Jerome’s testimony for the pericope adulterae?

In an earlier post I pointed to some striking parallels between the Vulgate edition of the pericope adulterae (= PA) and the circumstances of Jerome’s expulsion from Rome, apparently under accusations of sexual impropriety. Considering various anomalies in Jerome’s testimony concerning the PA, I raised the question as to whether Jerome himself may have contributed to its final form.

I had already drafted the above post when last Saturday I attended Amy M. Donaldson’s paper, “‘What Was Spoken through the Prophet Asaph’ (Matt. 13:35): Textual Evidence from Jerome, or Conjectural Emendation by Origen?”, at the Pacific Northwest Regional SBL meeting on the variant “through Asaph” (ασαφ) in Matthew 13:35. [1] Since the reading lacks any surviving continuous manuscript support, the external evidence rests entirely on Jerome’s remarks. [2]

Fortunately, since Jerome comments on the reading on two separate occasions, we are able to compare his respective accounts and assess his reliability. In fact, the results are not favorable to Jerome’s credibility. When he first mentions the reading in his Commentary on Matthew (398), it is still conjecture [3]:

I have read in several manuscripts [Legi in nonnullis codicibus], … that … it is written as ‘through Isaiah the prophet, saying.’ Because the text is not at all found in Isaiah, I think it was later removed by prudent men. In my judgment [Sed mihi uidetur], it was originally published as follows: ‘[in order that what was written] through Asaph the prophet, saying.’ (Comm. Matt 13:35). [4]

Yet just a few years later, in his Homily 11 on Psalm 78 (77 LXX) (401), Jerome confidently asserts that what had formerly been in his judgment “is found in all the ancient copies” (in omnibus ueteribus codicibus). [5] It is unlikely of course that Jerome had discovered any actual MSS in the few years between these remarks. Presumably he would have mentioned such favorable evidence! So we are forced to ask, is Jerome extrapolating on the basis of a self-assured conjecture to evidence that he simply never saw? [6]

Whatever his motives, Jerome’s lack of inhibition in ascribing his personal conjecture to the Greek MS tradition should give us pause in assessing the reliability of his remarks elsewhere. Imagine that Homily 11 had been our only surviving source for the reading ασαφ in Matthew 13:35. We would quite reasonably infer that, had Jerome found this reading in all of his copies, it must have been at least in a great number of Greek copies. Yet we would be utterly mistaken!

Fast-forwarding to the year 415, Jerome asserts that there then existed “many Greek and many Latin codices” with the PA (Pelag. 2.17). But surely Jerome is being somewhat disingenuous in his appeal to “many” Latin copies. After all, this figure included copies of the Vulgate that he had himself revised to include the PA! And why is it necessary to cite Latin evidence at all for a reading that exists in many Greek MSS? Are we to wonder then whether Jerome is extrapolating in this case also to Greek evidence he never saw?

[1] University of Idaho, May 21, 2016.

[2] The evidence is still cited in NA28 and (apparently) UBS5. Jerome’s starting point seems to be Origen’s conjecture.

[3] Date according to Gryson et al., Répertoire général des auteurs ecclésiastiques latins de l’antiquité et du haut Moyen Âge (2007) 1:540.

[4] Cited from Dr. Donaldson’s handout, which cites FC 117:160-161;  SC 242:284.

[5] Date according to Gryson et al., 1:545. Cited from Dr. Donaldson’s handout, which cites FC 48:81-82 in modified form and CCSL 78:66-67.

[6] Dr. Donaldson plausibly suggests that Jerome simply assumed that his inference was correct and hence present in the old copies. See also, Amy M. Donaldson, “Explicit References to New Testament Variant Readings among Greek and Latin and Church Fathers.” (PhD Diss.; University of Notre Dame, 2009) 369-372, where she observes, “The homilist therefore assumes that Asaph is the predominant and oldest reading.” (p. 370) “Jerome especially emphasizes that his conjectured original  reading, Asaph, is theologically correct …” (p. 372)

Are there autobiographical elements in Jerome’s pericope adulterae?

In an earlier post I pointed out some anomalies concerning Jerome’s involvement with the pericope adulterae (= PA) as it appears in the Vulgate at John 7:53-8:11:

  • Jerome’s remarks on the PA’s wide attestation come some thirty years after the presentation of the Vulgate gospels
  • Jerome does not cite the Vulgate version of the PA in his later remarks on the PA
  • Jerome’s PA is very different from attested pre-Vulgate Greek forms, such as that of his mentor Didymus or the account in Didascalia Apostolorum
  • Jerome seems to be exaggerating the representation of the PA in Greek manuscripts to make a point against the Pelagians
  • The appearance of the PA in the Vulgate is followed by an explosion of interest in the story within the Latin tradition

Considering the above anomalies and Jerome’s reputation for inventing material when it suited his purposes (Vigil. 3 and see Nautin [1983] on fabricated letter exchanges), the question needs to be asked whether Jerome had a hand in the composition of the final form of the PA. The following parallels between the story as it appears in the Vulgate and Jerome’s own circumstances may be illuminating (apologies for the Victorian English in the NPNF quotes):

  • like the woman in the story, Jerome was accused of sexual misconduct in his capacity as spiritual advisor to a circle of ascetic women in Rome:

It often happened that I found myself surrounded with virgins … Our studies brought about constant intercourse, this soon ripened into intimacy, and this, in turn, produced mutual confidence. … No; my sex is my one crime, and even on this score I am not assailed, save when there is a talk of Paula going to Jerusalem. … Men call me a mischief-maker … There hath no temptation taken me but such as is common to man. … Men have laid to my charge a crime of which I am not guilty [i.e. incontinence] (Epist. 45.2,6 “To Asella”; trans. NPNF 2.6, 59, 60)

  • as the accused woman was brought before the Pharisees, so Jerome was called before a tribunal of Roman clerics (whom he calls “Pharisees” elsewhere) to answer charges of sexual misconduct:

As to what judgment was formed of me at Rome, or what was written afterwards, you are quite welcome to speak out, especially since you have writings to trust to; for I am not to be tried by your words … but by the documents of the church. (Ruf. 3.22; trans. NPNF 2.3, 530)

  • given the PA’s interest in hypocrisy, it is telling that Jerome painted his accusers as hypocrites, referring to the tribunal as a “senate of Pharisees” (Pharisaeorum … senatus) (preface to Did. Spir.)
  • Jerome’s attachment to Paula, whose plans to follow Jerome to Palestine initiated rumors of an indiscretion, placed him in a particularly sympathetic position towards the woman accused of adultery:

Of all the ladies in Rome but one had power to subdue me, and that one was Paula. … The only woman who took my fancy was one whom I had not so much as seen at table. But when I began to revere, respect, and venerate her as her conspicuous chastity deserved, all my former virtues forsook me on the spot. (Epist. 45.3 “To Asella”)

  • concerning Jesus’ writing on the ground, we should only note that the written word was Jerome’s own weapon of choice against those he considered his hypocritical accusers

Why does Jerome not cite the Vulgate of the Pericope adulterae?

Jerome’s remarks on the story of the adulterous woman in his tract Against the Pelagians (415) are sometimes cited as evidence for its existence at the time of his revision of the gospels in 384. In an earlier post though, I pointed out that Against the Pelagians was written some thirty years after the Vulgate gospels and hence probably cannot be considered evidence for the state of the text in the early 380’s, at the time of Jerome’s Vulgate project.

The question remains of course whether Jerome was looking back to this earlier time. In light of this possibility, it is noteworthy that Jerome does not cite the Vulgate text in his remarks to the Pelagians. Consider:

Against the Pelagians (CCSL 80) Vulgate (R. Weber)
At Iesus inclinus digito scribebat in terra (v. 6)

Qui sine peccato est uestrum, primus mittat super eam lapidem (v. 7)

Vbi sunt? Nemo te condemnauit? Quae ait: Nullus, Domine. Responditque ei Iesus: neque ego te condemno. Vade, et amodo noli peccare. (vv. 10-11)

Iesus autem inclinans se deorsum digito scribebat in terra (v. 6)

qui sine peccato est vestrum primus in illam lapidem mittat (v. 7)

ubi sunt nemo te condemnavit quae dixit nemo Domine dixit autem Iesus nec ego te condemnabo vade et amplius iam noli peccare (vv. 10-11)

Clearly Jerome’s three citations of the pericope adulterae in Adversus Pelagianos attest some significant discrepancies in comparison with the Vulgate. The differences in vv. 10-11 seem hardest to explain. After all, it seems odd that Jerome would default to a different form of the woman’s two-word response, “No one, Lord,” or the memorable pronouncement, “Neither do I condemn you,” even if he were paraphrasing or citing from memory.

Jerome and the Pericope adulterae

Jerome’s remarks on the story of the adulterous woman in Adversus Pelagianos (415) are sometimes taken as a virtual textual commentary on his Old Latin revision completed more than three decades earlier (384):

In Euangelio secundum Iohannem in multis et Graecis et Latinis codicibus inuenitur de adultera muliere, quae accusata est apud Dominum. (Jerome, Adversus Pelagianos 2.17)

The story of the adulterous woman who was accused before the Lord is found in many Greek and many Latin manuscripts of the gospel of John. (Jerome, Against the Pelagians 2.17, my translation)

From these remarks, it is sometimes assumed that at the time he revised the text of John in the early 380’s, Jerome already had at his disposal many Greek and many Latin manuscripts both with and without the pericope adulterae (= PA). But this assumption overlooks a gap of over thirty years between Jerome’s dispute with the Pelagians in 415 and his Vulgate revision of the gospels in 384. In short, it is questionable that Jerome’s remarks in Adversus Pelagianos can be cited (as they often are) as evidence that the PA was represented in many Greek and many Latin copies of John before its appearance in the Vulgate.

But even in 415, it is not easy to rule out a degree of exaggeration on Jerome’s part. Certainly, given the polemical context, Jerome has every reason to present the Greek evidence in the most favorable light. After all, his argument from the Greek meaning of αναμαρτητος (v. 7) as “without sin” rests somewhat precariously (it would seem) on the existence of Greek copies. But if Greek copies had been easy to find, why mention their number or (for that matter) bring up the problem at all? It seems that by calling in the Latin evidence, Jerome anticipates an  objection concerning the scarcity of Greek copies, which suggests that in 415 there were still very few Greek copies of John with the PA, but apparently plenty of Latin copies.

Stepping back a bit, there is a disconcerting aspect to Jerome’s appeal to the Latin tradition given his own contribution to the dissemination of the PA in Latin. By all appearances, the Vulgate contributed immensely to the story’s popularity. We must wonder at the sudden explosion of interest in the PA in the Latin tradition after its first appearance in the Vulgate. (Consider, for example, Ambrose’s reference in Epistle 68 from 385-387, Rufinus’ apparent reinterpretation of Papias through the PA from 401-402, and Augustine’s two references from after 399. The other allusions in Latin seem to follow this pattern as well.)

On the other hand, it is odd that besides Jerome’s citation of a single Greek word, the only Greek evidence occurs in a manuscript that also contains a Latin column (Codex Bezae), while additional allusions to the PA continue in Latin writers for several centuries before the first surviving reference in a Greek writer. Even in the fifth century then, we still struggle to find the pericope standing as an independent Greek tradition at its canonical position in the Vulgate, leading us to wonder what Greek evidence Jerome might have found in the early 380’s outside of extra-biblical traditions. Why does Jerome’s mentor Didymus still know a very different form of PA (for all its impressive similarity)? Why is he still unwilling or unable to identify which gospel it came from?

Of course, this may all be a coincidence. But it leaves open a number of unanswered questions concerning Jerome’s involvement in establishing the PA in its “canonical” position between John 7:52 and 8:12.

The Pericope Adulterae and Protoevangelium of James

In a 1997 essay, W. L. Petersen argued for a literary connection between the Pericope adulterae (= PA) and the Protoevangelium of James on the basis of the common phrase “neither do I condemn you” (ουδε εγω σε [κατακρινω]). [1] While in the PA Jesus utters these words to the accused woman (John 8:11), in the Protoevangelium the high priest pronounces them over Mary and Joseph, declaring them innocent of the charge that Mary’s virginity has been compromised.

Petersen assumed that the Protoevangelium must depend on an earlier form of the PA tradition without so much as mentioning the possibility that the PA, if it were a late tradition, might in fact depend on the Protoevangelium. As we might expect, Petersen’s essay is typically cited in discussions of the PA as a second-century tradition. But if we regard the PA in its fullest form as a fourth-century development, there is no reason the dependency must go in this direction. In fact, if we regard this saying as one of the final features of the PA, the only possible connection (if we accept a connection) is from the PA to the Protoevangelium. But if such a connection is plausible, we have a potentially valuable clue as to the interests of the fourth-century compilers of the PA in the antidicomarianite and related controversies concerning the virginity of Mary.

[1] ΟΥΔΕ ΕΓΩ ΣΕ [ΚΑΤΑ]ΚΡΙΝΩ, John 8:11, The Protevangilum Iacobi, and the History of the Pericope adulteraein Sayings of Jesus: Canonical and Non-canonical (ed. W. L. Petersen, J. S. Vos, and Henk J. de Jonge; Leiden: Brill, 1997), 191-221.

Major Late Textual Changes in Codex Bezae?

On April 14, Prof. Larry Hurtado published a notice on his blog about a new book on the Pericope Adulterae (= PA): The Pericope of the Adulteress in Contemporary Research, eds. David Alan Black and Jacob N. Cerone (London:  Bloomsbury T&T Clark, 2016). In the notice, Prof. Hurtado questions a common assumption in the field of NT textual criticism that the vast majority of significant changes to the NT text were introduced during a “wild” period of uncontrolled textual development in the second and third centuries, after which a period of stability reigned beginning in the fourth century.

Why am I Excited about this Question?

First of all, I am delighted that the question is being asked. Several areas of my research on Codex Bezae and the so-called “Western” text explore evidence of systematic changes to the text of the gospels and Acts in the latter part of the fourth century. These topics, particularly as they relate to the phenomenon of layering in Bezae’s text (Holmes, 1996), are addressed here on my blog and in my small but growing list of publications and papers.

(Note that Prof. Hurtado’s own contribution to the book explores these questions in further detail, though I have not seen it as of this post. Therefore my remarks here are based on the question raised in the blog post.)

The Challenge

The question Prof. Hurtado raises (as two questions) is as follows:

Is it possible that the common view of the transmission-history of NT writings (however intuitively it appeals) is wrong, or at least seriously defective? More specifically, were there factors and dynamics in the later period that facilitated the inclusion and wide acceptance of these sizeable variants?

The question of major textual change in the later period (i.e. the fourth century and beyond) is the crux of the challenge to the common view, which in general assumes that all major variations were introduced by the end of the third century, after which the text reached a point of essential stability. But if the PA appears in its fully-developed form (or forms) only in the fourth century as the evidence seems to suggest, we are led to inquire as to whether the notion of a stable fourth-century text must be abandoned. On the other hand, recognizing the diversity within the tradition, perhaps we can accommodate major change in certain parts of the tradition, while acknowledging a degree of control in others.

The Place of Codex Bezae

Whatever we believe about its history, it is clear Codex Bezae represents one of the extremes of diversity in the tradition. But the unique place of Bezae’s tradition points to one area where we can further nuance the question. Bezae’s highly divergent text reminds us that we can expect different parts of the tradition to be more closed or open to change than others. We cannot expect the tradition of Codex Vaticanus to be as accepting of change as the early Byzantine tradition of John Chrysostom. At the same time, we are not as surprised to find major variations in the Old Latin tradition as we are in the Latin Vulgate (which makes the PA’s appearance in the latter all the more remarkable). In fact, if we suppose that the full form of the PA first appeared in Bezae’s tradition, it is difficult to see how this opinion alone would challenge the common view that essential stability was achieved in the fourth century. Bezae’s exception while perhaps typical of its own tradition, would merit a mere footnote in relation to the rest of the tradition.

Major Late Textual Change in Bezae’s Tradition?

It is probably no accident then that the PA occurs first in a manuscript like Codex Bezae, a witness well-known for its high number of apparently singular and sub-singular readings. With regard to similar character and size, Bezae’s singular readings present us with other substantial changes, such as Jesus’ encounter with the Sabbath worker (Luke 6:5) or the harmonization of Luke’s genealogy to Matthew. While Bezae’s singular and sub-singular readings are often assumed to date to the second century, this is obviously a pure guess. After all, how can we be sure if the readings are otherwise unattested? While we cannot of course assume that all such apparently singular readings belong to Bezae’s copyist or even the copyist’s immediate tradition, this possibility is not at all implausible (Holmes, 2003).

It turns out then that this potential ‘upper’ layer of singular and sub-singular readings in Bezae could just as plausibly offer us a unique window not to the second century, but to the events surrounding Bezae’s production in ca 400! In fact, if we take Bezae’s various suggested tendencies in Acts, it turns out they consistently fit a context at the end of the fourth century as well as, if not better than a second-century context.

  • Bezae’s anti-Judaic tendency is certainly consistent with a second-century context as suggested, for example, by Justin’s rhetoric in the Dialogue with Trypho. But Justin’s display of anti-Judaism pales next to Ambrose’s attempts to lobby the emperor to withhold financing for the restoration of Jewish property destroyed by Christians!
  • Bezae’s favorable portrayal of Roman soldiers and officials, while puzzling in a second-century context, makes better sense at the end of the fourth century with a Christianized Rome.
  • Regarding Bezae’s interest in the Holy Spirit, it is telling to observe that Bezae’s paleographical date of ca 400 comes within twenty years of the Second Ecumenical Council (381), which clarified the place of the Holy Spirit in the Trinity.
  • Like the anti-Judaic tendency, Bezae’s so-called ‘anti-feminist’ tendency is generally consistent with a second-century context. But the readings themselves align particularly well with the conservative anti-ascetic movement at the end of the fourth century, represented by the likes of Jovinian and Ambrosiaster, who argued that women possess the imago Dei only through a man.
  • Bezae’s enhancement of Peter’s role at the expense of Paul is consistent with an increasing assertiveness of the papacy in political and ecclesiastical affairs at the end of the fourth century. But it is even more consistent with an anti-ascetic tendency, which would naturally look to Peter’s precedent, as the first of the married apostles, to that of Paul, the celibate apostle.

It turns out that most of the above tendencies can be seen in light of fourth-century controversies, perhaps suggesting a common impulse for at least some of the significant and seemingly late changes we find in Bezae. It stands to reason that a key source of pressure on the fourth-century text must have been the unbroken succession of theological controversies in which beatifications and anathemas were won and lost over the effective use of proof texts. While the Christianization of Rome no doubt encouraged standardization of the text, we might surmise that it also raised the stakes on which texts became standard. This leads us to wonder whether Bezae might not represent a rather unexpected sort of ‘controlled’ text in which stability and control do not necessarily align.

Article Now Available on Textual Layers in Codex Bezae

I recently published an article on textual layers in Codex Bezae found here:

Analyzing Textual Stratification in the Greek Gospel Text of Codex Bezae: Comparing Three Approaches to Layer Extraction in John 4,1–42 (2015)

It is one thing of course to suggest that Codex Bezae attests a mixed text with readings from multiple sources combined into its final text. But it is quite another matter to identify and extract these sources in a systematic and repeatable way. The basic method was proposed by Michael Holmes in a 1996 essay (“Codex Bezae as a Recension of the Gospels”). Holmes then successfully demonstrated this method on the text of Matthew. The consistency of his results speaks for itself.

My goal in this paper is to identify more accurate and efficient techniques to extract Bezae’s layers based on Holmes key insights, namely

  1. That the same witnesses are often found together in support of distinct subsets of Bezae’s readings and
  2. That any group of readings supported by essentially the same selection of witnesses represents a ‘layer,’ which we can treat as a distinct element of Bezae’s tradition.

The motivation is to repeat Holmes’ proof-of-concept on the text of Matthew with other full-scale applications in other parts of Bezae. The article uses complete IGNTP transcriptions for a small part of John.

From the abstract:

It has been suggested that Codex Bezae’s Greek column (D) attests a stratified text, consisting of distinct layers of readings that reflect its historical contact with different traditions. Using John 4:1-42 as a case study, this paper compares three methods of partitioning D’s readings by layer: first, Holmes’ (1996) method based on patterns of agreement; second, a proposed method based on the levels of D’s readings in local genealogies; and, third, a proposed method based on multivariate clustering.

The result shows that Bezae’s readings tend to bifurcate cleanly between two main layers, a mainstream layer and an Old Latin layer.