The Amanuensis Hypothesis in New Testament Scholarship: Its Origin,Evidential Basis,and Application

The Amanuensis Hypothesis in Antiquity?

The origin of the amanuensis hypothesis is commonly traced back to antiquity, where its initial formulation is attributed to Jerome. On two occasions, Jerome acknowledges that the letters of 1 and 2 Peter were written in notably different styles. The first time that this compositional feature is observed is in his description of illustrious ecclesiastical writers and the documents they produced (De Viris Illustribus, ca. 392/393 CE). When recounting the life and written accomplishments of Simon Peter, Jerome explains how the stylistic variation between 1 and 2 Peter had led many to question the authenticity of the latter: ‘He [Peter] wrote two epistles that are called Catholic; the second of which is considered by many not to be his due to its stylistic difference from the first (quarum secunda a plerisque ejus esse negatur, propter styli cum priore dissonantiam)’ (Vir. ill. 1.3 [PL 23:638]).

Although Jerome does not specifically identify who was among the ‘many’ (plerique) dissenters, some have speculated that he may have been referring to his teacher, Didymus of Alexandria (so, e.g., Westcott 1889: 448; Gregory 1907: 277). Whether or not this was the case, it is clear that the authenticity of 2 Peter had been challenged prior to this point. The earliest doubts are reported by Origen (Commentaries on the Gospel of John cited in Eusebius, Hist. eccl. 6.25.8), and such sentiments appear to have continued up through the time of Eusebius (Hist. eccl. 3.1.4; 3.3.1; 3.25.3). Even after the letter’s inclusion in numerous fourth century canonical lists (e.g., Cyril of Jerusalem, Catech. 4.36 [ca. 350 CE]; Athanasius, Ep. fest. 39 [367 CE]; Epiphanius, Pan. 76.5 [ca. 375 CE]; Const. ap. 8.47.85 [380 CE]; Gregory of Nazianzus, Carm. 12.31 [390 CE]; Jerome, Ep. 53.9[8] [394 CE]), some remained skeptical. The epistle is lacking in the Old Latin (see Westcott 1889: 263-65) and some strands of the Syriac tradition (e.g., the catalogue of St. Catherine’s monastery [ca. 400 CE]; Doctrina Addai 44; Aphraates, Homilies). Further, various church fathers from the fourth through sixth centuries rejected either the Catholic Epistles as a group (see Leontius of Byzantium, C. Nestorianos et Eutychianos III.14 [PG 86:1365]; Cosmas Indicopleustes, Christian Topography 7:292 [PG 88:372-373]) or only a certain portion that included 2 Peter (see Amphilochius of Iconium, Iambi ad Seleucum 310-313; John Chrysostom, Synopsis Scripturae sacrae [PG 56:317]; Junilius Africanus (De partibus divinae legis 6 [PL 68:19]). So even if ‘this formulation [of Jerome] may be somewhat exaggerated’ (Grünstäudl and Nicklas 2014: 221), there was obviously a large enough constituency that Jerome felt it necessary to mention.

What is particularly important about Jerome’s description of this group is that he identifies the reason for their skepticism. Up to this point, church fathers had merely noted the debate surrounding the epistle; but here we learn, for the first time, that the doubts were connected to the letter’s (stylistic) deviation from 1 Peter. Whether the earlier questions surrounding the authenticity of 2 Peter were driven by such comparisons, we can only conjecture. But this reference at least provides definitive documentation of an interpretive decision that would continue to grow in popularity.

Although Jerome mentions the basis for the skepticism surrounding 2 Peter, he does not attempt to address the stylistic problem in De Viris Illustribus. Nearly a decade later, however, we find his explanation within a personal correspondence. It seems that Jerome was in the habit of answering questions posed to him by Christians from around the empire (see, e.g., Jerome, Ep. 119, 121, 123, 125, 129). On one such occasion, he received a letter (ca. 407 CE) from a widow name Hedibia, who was seeking answers for various problems related to New Testament interpretation. For centuries, this letter was read as a real correspondence written by a woman from Gaul who was interested in the specifics of New Testament exegesis. Whether this is the case (Cain 2003: 15-33), or whether the letter was a fictive account designed to provide Jerome with a platform to pose and then answer interpretive questions (Burgon 1871: 51-57), the theory of Jerome represents an attempt to resolve what was perceived as a problem by many interpreters of 2 Peter.

In total, Hedibia posed twelve questions to the learned scholar. The eleventh related to the meaning of Paul’s statement in 2 Cor. 2.15–16, in which he declares, ‘We are the aroma of Christ to God among those who are being saved and among those who are perishing, to the latter the smell of death leading to death, and to the former the smell of life leading to life. And who is adequate for such things?’ As part of this response, Jerome’s discussion of Paul leads him into a brief aside about Peter and the epistles attributed to him. He states,

Even though Paul had a perfect knowledge of the holy Scriptures, though he was naturally eloquent, and though he possessed the gift of speaking in tongues, as he prides himself in the Lord, saying, ‘I thank my God that I speak with the gift of tongues more than all of you’, he was not able to express the majesty of the divine sentiments in eloquent Greek words. Therefore, Paul had Titus as an interpreter, just as Saint Peter had Mark, whose gospel was composed as Peter narrated and Mark wrote it down. Finally, also, the two epistles ascribed to Peter differ from one another in style, character, and word order (duae epistulae, quae feruntur Petri, stilo inter se et caractere discrepant structuraque uerborum), from which we can discern that it was sometimes necessary for him to employ different interpreters (ex quo intelligimus pro necessitate rerum diuersis eum usum interpretibus) (Ep. 120.11 [CSEL 55:507-508]).

The most important question that arises from this passage is the meaning of the term interpres. It could denote a ‘secretary’ who received the apostle’s spoken message, originally delivered in Greek, and then recorded it (in Greek) on a written medium. In this case, any variation between the words of Peter and the words recorded in the letters would represent the compositional license taken by the secretary. On the other hand, interpres could denote a ‘translator’ who received the original Aramaic (or Hebrew) message of Peter, converted it into Greek, and then recorded the translation on a written medium. In this case, any variation between the words of Peter and the words recorded in the letters would represent the (translational approach involved in the) conversion of a message from one language into another. Although there is obviously overlap between these two functions, they have been purposefully distinguished here in order to highlight the mechanism behind the stylistic divergence.

The evidence—while admittedly not overwhelming—seems to suggest that Jerome had the latter in mind. This is evident from the fact that reference is made to the Petrine epistles in the context of a discussion about the limitations of Paul’s linguistic abilities and his need for a translator. Even though Jerome lauds the various gifts and communicative abilities of Paul, he states that the apostle ‘was not able to express the majesty of the divine sentiments in eloquent Greek words’ (diuinorum sensuum maiestatem digno non poterat Graeci eloquii explicare sermone, Ep. 120.11). After the aside about the letters of Peter, Jerome returns to Paul’s speaking abilities. He describes Paul as being in a state of unrest while in Troas (cf. 2 Cor. 2.12-13) because he had not yet found ‘the pipe of his preaching, the organ through which he would sing to Christ’ (quia praedicationis suae in praesentiarum fistulam, organumque per quod Christo caneret, non invenerat, Ep. 120.11). By this, he was referring to Titus, who joined Paul’s team and thus solved his need for an interpreter. This seems to suggest that Jerome is thinking in terms of translation, not merely transcription.

At the same time, it is difficult to deny that when Jerome describes the composition of Mark’s gospel the written recording of oral speech was involved. He explains how Mark wrote while Peter narrated. This reflects an early and long-debated tradition of Mark serving as Peter’s ‘interpreter’ (see Papias [= Eusebius, Hist. eccl. 3.39.15; Irenaeus, Haer. 3.1.1; 3.10.5; Clement [= Eusebius, Hist. eccl. 6.14.6-7]; Tertullian, Marc. 4.5; Origen [= Eusebius, Hist. eccl. 6.25.5]; Jerome Vir. ill. 1, 8). When this tradition is discussed in modern scholarship, the question normally focuses on whether Mark served as an amanuensis, a translator, or a messenger. Yet it is important to recognize that the meaning attached to Mark’s role seems to have changed as the tradition evolved. Jerome stands at the end of this development, wherein the transcription of Peter’s speech allows the gospel to have direct apostolic authority in ways that earlier forms of the tradition do not (a fact noted long ago, see Davidson 1894: 1:463). However, even when Jerome supplements the Mark-as-interpreter tradition to include dictation and transcription, this does not rule out the possibility of language transfer; that is, Jerome’s reference is consistent with Mark recording (in Greek) as Peter dictated in Aramaic. This consideration, in connection with the evidence discussed above, suggests that, for Jerome, the stylistic discrepancy between 1 and 2 Peter should not merely be attributed to the compositional freedom of an amanuensis, but to the range of possibilities involved in translation.

Other pieces of evidence lend weight to this conclusion. First, this interpretation is consistent with Jerome’s use of the noun interpres elsewhere. Generally, he employs this term to denote one who explains the meaning of a word or phrase (Ep. 48.14, 17), or more commonly, someone who translates a spoken or written account from one language into another (Ep. 57.5, 11; 80.1; 106.3, 30; 116.3; 140.6). By contrast, when Jerome refers to one who transcribes the words of another into written form (without any translation involved), he tends to use the term notarius to describe a secretary (see Ep. 34.6; 57.2; 117.12). Consequently, his choice of interpres in Ep. 120.11 is most likely intended to highlight the impact of translation.

Second, Jerome offers a similar explanation to defend Pauline authorship of the epistle to the Hebrews. He acknowledges that others at the time doubted that Paul was the author of that letter, ‘because it differs (from the other Pauline letters) in style and language (propter styli sermonisque dissonantiam)’ (Vir. ill. 5). Nevertheless, Jerome accounts for this discrepancy by suggesting that Hebrews had been translated. The process that he envisions is noteworthy: ‘Paul, being a Hebrew, wrote in Hebrew, that is, in his most eloquent expression, so that those things that were eloquently written in Hebrew could be more eloquently turned into Greek (ut ea quae eloquenter scripta fuerant in Hebraeo, eloquentius verterentur in Graecum). This is the reason why it seems to differ from other epistles of Paul’ (Vir. ill. 5). It is clear, therefore, that Jerome was aware of the impact that translation could have on the style of documents written by the same individual (cf. Jerome, Ep. 57; Chron. preface; Sit. preface), much like he postulates in the case of 1 and 2 Peter.

A third consideration that points toward translational divergence rather than secretarial freedom is that Jerome elsewhere employs style as a criterion by which to judge authorship claims. On numerous occasions, appeal is made to stylistic considerations as a way to differentiate a work attributed to a given author from other writings by the same individual. In doing so, Jerome is thus able to challenge the authorship claim of the work in question (e.g., Jerome, Vir. ill. 25, 58; Ep. 84.10; Ruf. 2.19). This practice is applied to a variety of genres, including epistolary literature. At times, he rules out the possibility of common authorship between two letters because their style is different (e.g., Jerome, Ep. 102.1; Vir. ill. 15). In other cases, the identity of a letter writer is postulated due to the fact that the epistle shares the same style as others written by a known author (Jerome, Ruf. 2.24; 3.25). Since Jerome uses style to distinguish one epistolary author from another, it seems unlikely that he would allow for the possibility that a secretary could play a determinative role in the composition process. Otherwise, the stylistic distinctions that he notes would have no force.

Finally, it is important to note that ‘many’ (plerique) at the time of Jerome rejected 2 Peter based on its stylistic difference from 1 Peter (Vir. ill. 1.3). This raises an important question about the prevalence of the negative evaluation: If, as Jerome is commonly thought to do, one could appeal to the freedom of an amanuensis to explain the discrepancies between two epistles, why was the style of 2 Peter such a widespread problem? Why did others not postulate this same explanation? Was Jerome the first biblical interpreter in nearly three hundred years to devise such a solution based on ancient letter-writing practices? And once the dissenters became aware of this formulation, why did the postulation of multiple amanuenses not alleviate any and all doubts? As it stands, it is much easier to explain the environment of skepticism to which Jerome is responding if his solution relates to the translational practices of an interpreter and not the compositional freedom granted to an amanuensis.

In conclusion, then, the frequent appeal to Jerome’s ‘solution’ by modern proponents of the amanuensis hypothesis appears to be a case of mistaken identification. While Jerome may be the source of the amanuensis hypothesis, in the sense that his explanation may have inspired the eventual formulation of the theory that has become so popular today, he seems to have been advocating for a slightly different explanation. Jerome actually put forward what could be labeled the ‘translator hypothesis’, wherein the differences in style between 1 and 2 Peter are explained by the translational decisions involved in converting the Aramaic message of Peter into Greek.

But if this is the case, who first postulated the amanuensis hypothesis, and why was the view originally formulated? The answer to these questions is found nearly a millennium after Jerome, during the time of the Protestant Reformation.

The Original Formulation of the Amanuensis Hypothesis

The doubts that some ancient interpreters expressed about 2 Peter continued to be echoed at the start of the Reformation. Not only were claims made about canonical distinctions, wherein 2 Peter was located within a less significant category of scripture (see Karlstadt 1520: 4; Hutter 1609: 8; Serarius 1612: 40), but prominent interpreters seemed hesitant to fully endorse the letter’s authenticity. Luther expressed some doubts, although he never rendered a final and absolute verdict. When addressing 2 Pet. 3.15 and its reference to ‘all of [Paul’s] letters’ (πάσαις ταῖς ἐπιστολαῖς), he claimed that this ‘sufficiently shows that this epistle was written long after St. Paul’s epistles’ (gnugsam anzeygt, das dise Epistel lang nach S. Paulus Episteln geschrieben sey; Luther 1524: e i). Further, he noted that this is one of the passages ‘that might persuade someone to conclude that this epistle was not St. Peter’s’ (die da yemand möchten bewegen, zu halten, das diese Epistel nicht S. Peters were; Luther 1524: e i). Yet he seems to remain open to the possibility that the letter was penned by the apostle.

Doubts were also expressed by Erasmus. As he began his Paraphrases of 1–2 Peter and Jude, the renown philosopher and humanist drew attention to the fact that, in antiquity, ‘there was never any doubt about 1 Peter, but there was for 2 Peter’ (de priore Petri numquam fuit dubitatum; de posteriore fuit; 1520: 5). Erasmus himself expressed some reservations about the letter’s authenticity based on its style and content, noting that the only way he could explain it was either that 2 Peter was pseudonymous, or that, like the (now) lost letter that he envisioned from his interpretation of 1 Pet. 5.12 (see Erasmus 1527: 684; cf. also Grotius 1650: 36; Pott 1790: 152; Hottinger 1815: 145), Peter had commissioned Silvanus to transcribe the letter (Erasmus 1520: a3; although, cf. 1520: 53, where he seems to attribute the letter to Peter). While this suggestion was put forth merely as a possibility, without any consideration being given to ancient letter-writing or to the role(s) of amanuenses in the process, it represents the first documented occurrence of the amanuensis hypothesis.

A few decades after the initial—albeit cursory—postulation of the amanuensis hypothesis, Calvin would suggest a similar explanation of 2 Peter’s composition. Like Erasmus, Calvin was aware that some in the early church had questioned the letter’s authenticity. In fact, he mentions Eusebius and the doubts that the historian records. This, however, appears to have done little in the way of persuading Calvin. Much more influential, it seems, was Jerome’s report that some in his day had rejected 2 Peter because it differed from 1 Peter. This stylistic argument was one that Calvin could not deny. He notes, ‘What Jerome writes influences me somewhat more, that some, induced by a difference in the style, did not think that Peter was the author. For though some affinity may be traced, yet I confess that there is that manifest difference which distinguishes different writers (fateor... manifestum esse discrimen, quod diuersos scriptores arguat)’ (Calvin 1551: 140; trans. Owen).

The reason why this presented a problem for Calvin was because he felt the need to balance historical and literary considerations (which pointed away from the letter’s authenticity) with 2 Peter’s placement in the canon of scripture (which would not allow the epistle to be pseudonymous). He reasoned, ‘according to the consent of all, it has nothing unworthy of Peter, as it shews everywhere the power and the grace of an apostolic spirit. If it be received as canonical, we must allow Peter to be the author, since it has his name inscribed, and he also testifies that he lived with Christ: and it would have been a fiction unworthy of a minister of Christ, to have personated another individual’ (Calvin 1551: 140; trans. Owen).

On the assumption that the letter is genuine, Calvin believed that a more nuanced description of the compositional process was required. As such, he hesitantly proposed the amanuensis hypothesis. To account for the stylistic differences between 1 and 2 Peter, he suggested the following:

… not that he [Peter] himself wrote it, but that some one of his disciples set forth in writings by his command, those things which the necessity of the times required (non quod eam scripserit ipse: sed quod vnus aliquis ex Discipulis, ipsius mandato, complexus fuerit quæ temporum necessitas exigebat). For it is probable that he was now in extreme old age, for he says, that he was near his end. And it may have been that at the request of the godly, he allowed this testimony of his mind to be recorded shortly before his death, because it might have somewhat availed, when he was dead, to support the good, and to repress the wicked (Calvin 1551: 140; trans. Owen).

Even though Calvin confessed that he did not recognize the language of Peter, this proposal allowed him to assuage his own reservations about doubting a letter that he felt was pervaded by the Spirit of Christ (see Calvin 1551: 140).

Such a compromise is significant in that Calvin’s postulation of the amanuensis hypothesis seems to have arisen out of theological necessity rather than historical inquiry. In other words, when Calvin appealed to the secretary theory, he made no attempt to ground his view in letter-writing practices from antiquity or the role(s) of ancient amanuenses. The hypothesis was merely a useful expedient to avoid the (perceived) contradiction of a canonical pseudepigraphon. Little about this situation would change even after the time of Calvin. Those who espoused the amanuensis hypothesis made their case apart from ancient comparative evidence, and this would be the trend for almost four centuries thereafter.

Perhaps part of the reason why the historical basis of the amanuensis hypothesis was not addressed at this time was because of how rarely the view was posited. Despite the influence of scholars like Erasmus and Calvin, few in the 16^th and 17^th centuries espoused this solution to the stylistic problem of 2 Peter (one exception: Winckelmann 1705 [1608]: 1049, who suggested that Peter wrote 1 Peter himself, but employed a secretary to compose 2 Peter due to his advanced age). Many Reformation interpreters continued to defend the letter’s authenticity without appealing to secretarial license (Lefèvre d’Étaples 1527: 35-36; Pellikan 1539: 739; Hemmingsen 1566: n.p.; Strigel 1566: 302; Weller 1567: 166-68; Aretius 1581: 232-33; Herlin 1605: 328; Vorstius 1631: 544; Ames 1635: 121-22; Piscator 1638: 752; Dickson 1659: 298; Escobar y Mendoza 1667: 263). Various explanations were offered to account for the stylistic differences. Some followed Jerome and appealed to the use of different translators for each letter (e.g., Bellarmine 1587: 94). Others argued that the style of an ancient author could and would vary from composition to composition, especially over time and with regard to changing circumstances, such as the aged condition of Peter (cf. 2 Pet. 1.13-15). The primary example used to support this claim was the letter of ‘Paul’ to the Hebrews, which differs from the other letters in the Pauline corpus despite (at the time) being considered a letter from Paul (Lapide 1627: 354; Adams 1848 [1633]: 1; Cajetan 1639: 4:381). But even with the proliferation of such solutions, many who defended the authorship of the epistle simply refused to acknowledge that 1 and 2 Peter were stylistically distinct (e.g., Laurentius 1641: 1-5; Carrière 1663: 710; Alexandre 1710: 181; Bullinger 1534: 85; cf. Simon 1689: 196-97).

Aside from those who defended the authenticity of 2 Peter, there was also a small minority of interpreters who attributed the letter to someone other than the apostle Peter. One whose work garnered some strongly negative responses from subsequent interpreters was Grotius (1650: 38; for responses, see Hammond 1845 [1659]: 4:426-29; Le Clerc 1699: 598-99; Nietzsche 1785; Bertholdt 1819: 3124-27). Claiming that 2 Peter must have been written after the destruction of the temple in 70 CE, he postulated that the letter was composed by Simon (or Simeon), the bishop of Jerusalem, following the death of James. So rather than being a forgery, the epistle had mistakenly been attributed to the apostle by later Christian readers. To defend this thesis, Grotius was required to posit numerous interpolations and to divide 2 Peter into two letters: chapters 1–2 being the first epistle and chapter 3 being the second (cf. Weber 1798: 153-54, who also partitioned 2 Peter into two letters). Because of the complicated nature of this theory, few (if any) were ultimately persuaded. Others did, however, question the authenticity of 2 Peter. But when such challenges were raised, they were often connected to the content of the letter just as much as its style (see, e.g., Scaliger 1667: 71-72; Saumaise 1641: 14).

The Early Development of the Amanuensis Hypothesis

During the post-Reformation period, a growing number of scholars began to view 2 Peter as pseudonymous. In fact, near the middle of the 19^th century, one review of the interpretive landscape boldly declared that since the time of the Reformation, ‘the voices of all discerning authorities have united in doubting and rejecting it’ (haben sich die Stimmen aller Einsichtigen in seiner Anzweiflung und Verwerfung vereinigt) (Schwegler 1846: 1:491; cf. Augusti 1801–1808: 2:5). While this assessment represents an exaggeration, in that many competent interpreters continued to defend the authenticity of 2 Peter based on detailed treatments of the relevant issues (see below), a sizeable shift in scholarship can be discerned. At this time, a large number of interpreters came to conclude that the letter (or portions of it: see Ullmann 1821; Bertholdt 1819: 3157-61; Bunsen 1847: 175; Lange 1853–1854: 1:152-54) was spurious (e.g., Wetstein 1751–1752: 2:698; Semler 1784: praef. D1, 167-68; Schmidt and Schwarz 1798–1803: 5/2:263; Welcker 1805: 181-83; Eichhorn 1804–1827: 3:624-56; Richter 1810; Mayerhoff 1835: 149-217; Magnus 1835; Credner 1836: 1:660-61; Jachmann 1838: 176-79; Audemars 1838; Daumas 1845; Schwegler 1846: 1:490-518; Neander 1847: 2:598-99 n. 1; de Wette 1847: 77-78; Volkmar 1867: 4-6, 94; Reuss 1887: 268-69; Ewald 1870: 105-12; Renan 1873: vi-vii; Hilgenfeld 1875: 765-70; Abbott 1882: 49-63, 139-53; 1903: 25-41; Lechler 1885: 440; Bleek 1886: 752-64; Davidson 1894: 2:529-51; von Soden 1899: 211; Chase 1900: 799-818; Mitchell 1911: 64-66; Strachan 1910: 106-109; cf. Schott 1830: 425, who claimed that a close friend of Peter composed the letter after the apostle’s death using the things that he heard Peter preach). What is noteworthy about this shift, at least for our purposes, is that the stylistic differences between the Petrine epistles was only one among many reasons for the rejection of 2 Peter. Critiques also pointed to the problems created by the intertextual relationship with the epistle of Jude, the reference to ‘your apostles’ (2 Pet. 3.2), and the existence of a collection of Paul’s letters (2 Pet. 3.15-16), all of which seemed to suggest that the letter was written after the lifetime of the historical Peter.

Despite this mounting skepticism, many post-Reformation interpreters continued to defend the authenticity of 2 Peter (e.g., Lange 1734: 215; Michaelis 1788: 2:1475-83; Morus 1794: 214-15; Augusti 1801–1808: 2:5-19; Feilmoser 1830: 519-28; Hug 1847: 2:475-87; Stier 1850: 10-13; Dietlein 1851: 1-74; Reithmayr 1852: 736-40; Maier 1852: 419-23; Schaff 1854: 365-66; Fronmüller 1862: 68-70; Wiesinger 1862: 5-30; Güntner 1863: 317-19; Schott 1863: 154-88, 284-94; Steinfass 1863: 83-90; Weiss 1866: 255-308; Danko 1867: 476-78; Bisping 1871: 100-102; Hofmann 1875b: 128-41; Luthardt 1875: 16; Haneberg 1876: 705-707; Hundhausen 1878: 19-100; Thiersch 1879: 204-207; Spitta 1885: 475-82). In fact, numerous works from this period were specifically dedicated to demonstrating that the epistle was genuine (e.g., Flatt 1806; Dahl 1807; Olshausen 1822–27; Kern 1829; Moutier 1829; Armand-Delille 1835; Windischmann 1836; Heydenreich 1837; Henkel 1904; Grosch 1914; Snyman 1923). Part of these defenses focused on addressing questions of style. Contemporary scholarship on the New Testament had recently begun to refine the stylistic comparison of ancient documents. As part of this general trend, interpreters of 1 and 2 Peter sought to devote much closer attention to issues such as vocabulary, syntax, word-order, intertextuality, and rhetoric. Consequently, some of the fullest accounts of the differences between the Petrine epistles were produced at this time (e.g., Magnus 1835: 9-16; Mayerhoff 1835: 161-70; Farrar 1882b: 1:188-203; Davidson 1894: 2:546-51; Chase 1900: 3:807-809, 812-13; Strachan 1910: 106-109).

Not everyone was convinced by these efforts, however. Many proponents of 2 Peter’s authenticity continued to argue that the style of the letter simply did not diverge from 1 Peter to such an extent that it would undermine the authorship claims (e.g., Michaelis 1788: 2:1480-83; Dahl 1807: 60; Augusti 1801–1808: 2:13-14; Kern 1829: 16-19; Hug 1847: 2:480-82; Barnes 1875: ccix-ccxi; Hundhausen 1878: 72-80; Lumby 1881: 225-31). Others pushed back by challenging the use of style as a criterion for comparison. This was the approach of Heydenreich, who maintained that ‘one should not rely so much on such affinities or discrepancies of style and expression in investigations into the authenticity or non-authenticity of a text’ (1837: 90; cf. also Hug 1847: 2:487; Gaussen 1860: 1:501). These considerations, he suggested, are ‘always very uncertain’ (immer sehr unsicher). According to Heydenreich, style is an especially unreliable gauge of authenticity in cases like the Petrine epistles, where there is only a very small amount of literary material to compare (1837: 91; cf. Pott 1790: 163-68; Dahl 1807: 59; Gloag 1887: 214-16). ‘It is impossible’, he contended, ‘to form a complete and definite idea of an author’s style from so little’ (Heydenreich 1837: 91; cf. Michaelis 1788: 2:1480; Spitta 1885: 482). What is more, he also pointed to the vastly different subjects treated in 1 and 2 Peter, noting that this distinction makes stylistic comparison all the more difficult (Heydenreich 1837: 91; cf. Morus 1794: 214-15).

Others admitted that there were enough differences between 1 and 2 Peter to require an explanation. Numerous reasons were given for such a discrepancy, and very often interpreters offered more than one possibility. Echoing earlier approaches, some argued that stylistic variation is a natural and expected part of the writing process (Horne 1856: 606-607; Wordsworth 1867: 73), and as such, the literary style of an individual can and will change depending on certain factors. One crucial factor frequently mentioned in this discussion was subject matter or content (so, e.g., Macknight 1806: 5:514-15; Bloomfield 1855: 730; Wiesinger 1862: 17-21; Wordsworth 1867: 70-73; Fausset 1875: 515; Alford 1878: prol. 4:156-57; Hofmann 1875b: 128-39). Those defending the authenticity of 2 Peter pointed out that the letters addressed vastly different topics, with 1 Peter focusing on persecution and 2 Peter being directed toward ‘false’ teaching on eschatology. In the latter, it was noted, there was no occasion on which to mention topics such as baptism or Christ’s role as shepherd of his people. Related to this, scholars argued that different situations would have given rise to varying emotions or moods that are displayed in the two letters (so, e.g., Bloomfield 1855: 730; Wordsworth 1867: 70-73; Lumby 1881: 230; Alford 1878: prol. 4:156-57; Hundhausen 1878: 73-74). Another factor that was sometimes mentioned is the impact that time plays on a person’s writing style. How an author communicates often changes over time—and especially over an extended amount of time. This, they argued, could be relevant in the situation of 2 Peter, wherein the author anticipates his imminent death, which might thus imply that he had aged since the composition of 1 Peter (e.g., Guerike 1854: 479; cf. also Pott 1790: 163).

Rather than appealing to the unintentional effect of factors related to the writing process, Schott (1863: 185) maintained that the differences between 1 and 2 Peter were intentional acts of the Petrine author. He argued that these stylistic changes were made with a view toward the situations addressed by the letters. Since 1 Peter was written to Gentile Christians in areas that were familiar with Paul, the apostle Peter wrote that letter in a way that suppressed his own individuality and in a way that would have sounded familiar to his readers, whom he had never met. Schott accordingly describes 1 Peter as ‘the product of no small self-denying labor of Peter’ (das Erzeugniß einer nicht geringen selbstverleugnenden Arbeit des Petrus; 1863: 185). Then, once a connection had been established between the apostle and these communities, and since the persecution of the readers had dissipated, Peter was able to write in his own style and using his own personality in 2 Peter.

Although the translator hypothesis of Jerome was not popular at this time, there were a few who continued to promote it. Advocates were nonetheless divided over whether this translation was made from a written document or from the oral speech of the apostle. Those who advocated the former view maintained that 2 Peter was translated from an earlier Hebrew or Aramaic original (e.g., King 1871; Wohlenberg 1923a: xxxv-xxxvi; cf. Sherlock 1725: 199-230; Benson 1756: 322-23; Whytehead 1854: 174, who claim that both Peter and Jude are quoting from a Hebrew source). This latter view was defended by Farrar (1882a: 419; cf. Falconer 1902: 56). He recognized that while the translator hypothesis held out the benefit of confirming the authenticity of 2 Peter, it also required sacrificing the close connection to the apostle (see Farrar 1882b: 1:180). Nevertheless, Farrar maintained that the spirit of the apostle Peter was still evident in the letter (1882b: 1:206-207).

It was against the backdrop of (and in competition with) these variegated attempts to explain the stylistic differences between 1 and 2 Peter that the amanuensis hypothesis continued to take shape. Just as it was in earlier periods, this view was rarely posited in the 18^th and 19^th centuries, and there was little progression in its theoretical underpinnings beyond the proposal of Calvin. A major reason why the theory was so uncommon in Reformation and post-Reformation scholarship was because of how interpreters understood the role of Silvanus in 1 Pet. 5.12. In this passage, the author claims to have written the letter ‘through Silvanus’ (διὰ Σιλουανοῦ). While the reference to compositional activity (ἔγραψα) opens the possibility that Silvanus might have functioned as an amanuensis, few adopted this interpretation (one exception: Ewald 1870: 6; cf. others who claimed that 1 Peter had been composed by Silvanus shortly after the death of Peter: Seufert 1885: 350-71; Usteri 1887: 345-47; Spitta 1885: 530-32; von Soden 1899: 115-18; Bornemann 1919–1920: 160); instead, the vast majority of scholars understood διὰ Σιλουανοῦ as a reference to Silvanus’ role as letter-carrier (e.g., Erasmus 1520: 52; Pellikan 1539: 737; Winckelmann 1705 [1608]: 1048; Dickson 1659: 298; Estius 1666: 1159; Froidmont 1678: 593; Benson 1756: 308; Nonnen 1758: 8; Macknight 1806: 5:505-506; Doddridge 1808: 225; Hensler 1813: 231; Jachmann 1838: 173-74; Wiesinger 1856: 334; Besser 1857: 405; Schott 1861: 334-36; Webster and Wilkinson 1861: 50; Wordsworth 1867: 67; Demarest 1851: 276-77; Lillie 1869: 331; Barnes 1875: 206; Fausset 1875: 514; Hofmann 1875a: 200-201; Hundhausen 1878: 464; Cook 1881: 219; Keil 1883: 173; Mason 1884: 435; Johnstone 1888: 407; Kühl 1897: 284; Monnier 1900: 244; Bennett 1901: 255). Consequently, Peter was assumed to have written 1 Peter with his own hand and through his own linguistic abilities.

This approach began to change near the end of the 19^th century. An important impetus in this shift came from the work of Zahn, who noted the significance of this interpretation as it related to the authenticity of the Petrine epistles. He pointed out that, ‘[a]s long as one proceeded from the premise that 1 Peter was written by the hand of the apostle, and that 2 Peter was intended for the same group of readers as 1 Peter, the great difference in thought and language between the two letters was necessarily strong evidence against the authenticity of 2 Peter’ (Zahn 1899: 96). Such a line of reasoning, he maintained, was folly. According to Zahn, ‘[t]his argument falls to the ground, because both of these presuppositions have proved to be wrong’ (1899: 96). It was Zahn’s contention that Silvanus served as the amanuensis of 1 Peter and that Peter himself composed 2 Peter for a different audience than the one who received the first letter. Such a reconstruction was believed to possess strong explanatory power:

It goes without saying that Peter in a letter to the Gentile-Christian communities across Asia Minor, which Silvanus wrote on Peter’s behalf and in his name, would be different from a handwritten letter from Peter to Jewish-Christian communities, who owed their Christian faith to him and his companions. In particular, it is clear that in the latter case Peter would show himself to possess a quite different consciousness of his apostolic calling and professional relationship to the readers (Pt letzteren gegenüber sich ganz anders von dem Bewußtsein seines apostolischen Berufs und seines berufsmäßigen Verhältnisses zu den Lesern erfüllt zeigen mußte). While in the former he is represented, and as it were, introduced by a missionary prominent in the communities of Asia Minor, in the latter the apostle of the circumcision speaks to the flock over which Christ had made him shepherd (1899: 96).

But while the proposal of Zahn marked a significant development in the formulation of the amanuensis hypothesis, it did not address how or why the use of Silvanus would have impacted the style of the letter so significantly. The ancient epistolary mechanism upon which the entire amanuensis hypothesis rests (viz. secretarial freedom) is assumed, but unexplored. It was in connection with the Pauline epistles that this key feature was later added.

The Pastoral Epistles and the Beginnings of Stylistic Analysis

In antiquity, the letters of 1–2 Timothy and Titus were occasionally omitted (or rejected?) from consideration alongside other Pauline writings (see Tertullian, Marc. 5.21; Epiphanius, Pan. 42.9.3-4; Jerome, Prologue to Titus; cf. also Clement of Alexandria, Strom. 2.11.52.6; Origen, Comm. Matt. 117); however, the small amount of extant evidence provides no indication that this situation was based on stylistic evaluations. It was not until the early-19^th century that the authenticity of any of these letters was challenged on stylistic grounds. The first modern scholar to question the genuineness of a letter within the Pauline corpus on this basis was F.D.E. Schleiermacher (1807; although, cf. Eichhorn 1804–1827: 3.1:318-19 n. 1, who claims this accomplishment for himself). Although his critique was limited to 1 Timothy, the study of Schleiermacher opened the door to further exploration into the authenticity of the Pastoral Epistles as a group.

Schleiermacher’s doubts about the authenticity of 1 Timothy were expressed in a letter-essay—written in German, rather than the traditional academic language of Latin—addressed to J.C. Gass. In this epistle, he hoped to provide a fuller defense of the theory that the two had only begun to discuss in passing (Schleiermacher 1807: 3-4). Aside from pointing out certain historical inconsistencies with the Acts of the Apostles, much of Schleiermacher’s case was grounded in the fact that 1 Timothy contains numerous terms and phrases that do not occur elsewhere in Pauline writings (1807: 27-76). At the same time, he noted a close correspondence with 2 Timothy and Titus. From this evidence, Schleiermacher concluded that 1 Timothy was a pseudonymous composition, written sometime in the second century, which borrowed from 2 Timothy and Titus.

As one of the earliest applications of higher criticism within New Testament scholarship, the theory of Schleiermacher was—as might be expected—met with strong opposition (e.g., Planck 1808; Beckhaus 1810; Wegscheider 1810: 9-36). But as others built upon this foundation over the ensuing decades, the strength of the position would eventually grow to become the majority position by the mid-20^th century. One such study appeared only a few years after Schleiermacher published his groundbreaking work. In his introduction to the New Testament, J.G. Eichhorn rejected the authenticity of all three Pastoral Epistles (1804–1827: 3.1:315-410). While this view never rose above a minority position during the 19^th century (e.g., Baur 1835: 8-39; Credner 1836: 1:448-87; Mayerhoff 1838: 11-19, 25-27, 32-34, 37-39; Neudecker 1840: 542-89; Schwegler 1846: 2:138-53; Bauer 1850–1852: 3:77-88; Saintes 1852: 47-154; de Wette 1848: 297-313; Plitt 1872: 1-7; Pfleiderer 1872: 832-38; Hilgenfeld 1875: 744-65; Huther 1876: 53-73; Bahnsen 1876: 1-8; Schenkel 1879: 357-63; Hincks 1897), it served to shift the discussion away from a focus solely on the genuineness of 1 Timothy to a consideration of all three letters as a group.

This position was given further stimulus by the two most thorough examinations of the stylistic problem in the Pastorals. In 1880, H.J. Holtzmann laid out, in a careful and comprehensive manner, a detailed case against Pauline authorship of the Pastorals. His critique addressed the problems associated with situating the letters within the lifetime of the apostle (1880: 15-53) and the key differences that distinguish the Pastorals from the undisputed Pauline epistles (1880: 126-252). An important part of his argument, however, was the fact that the style of the Pastorals diverged significantly from other Pauline writings (1880: 84-118). Not only are hapax legomena prevalent in the Pastorals, the letters lack certain characteristic features of Pauline style, such as his argumentative rhetoric and even Paul’s typical syntax.

Even more attention was devoted to the linguistic question by P.N. Harrison (1921; cf. also 1955). In an exercise in what is now referred to as stylometry, Harrison employed statistical analysis to reveal how and to what extent the style of the Pastorals diverged from the other Pauline letters. What is more, he communicated the results through numerous charts and graphs. This unique visual element allowed Harrison to further illustrate the stylistic variations of the Pastoral Epistles. However, style and grammar are addressed in the study (1921: 38-44), much of the focus is on vocabulary (1921: 20-38). As others had previously done, Harrison points out that the Pastorals contain a large number of hapax legomena. But through comparative analysis, he takes this observation further by demonstrating that many of these terms appear in the Apostolic Fathers and some early Christian apologists (1921: 67-86). On this basis, Harrison places the composition of the letters sometime in the second century CE (1921: 8).

Despite the growing number of scholars in the 19^th and early-20^th centuries who read the Pastoral Epistles as pseudepigraphic compositions, the majority of interpreters remained unconvinced by the historical and linguistic arguments (e.g., Bertholdt 1819: 3491-630; Heydenreich 1826–1828: 1:5-19; Curtius 1828; Schrader 1830: 177-257; Flatt 1831: 421-615; Mack 1836: 524-34; Böttger 1837; Leo 1837: vii-xxiv; 1850: ix-xxxviii; Matthies 1840: 3-48; Wieseler 1848: 286-315, 329-55, 461-78; Conder 1850: 191-213, 380-95; Oosterzee 1864: 2-6; Alford 1878: prol. 3:70-97; Conybeare and Howson 1860: 2:449-90; Laurent 1866: 104-105; Langen 1868: 116-24; Glaire 1868–1869: 5:3-17, 109-31; Herzog 1872: 1-39; Fairbairn 1874: 1-19; Hofmann 1874: 3-4, 55-62, 208-20, 315-22; Lewin 1875: 2:336-58, 375-409; Thiersch 1879: 147-49, 172-73; Koelling 1882–1887; Ellicott 1883: xx, 108, 178; Plummer 1891a: 3-16; Davidson 1894: 2:21-75; Stellhorn 1899: 1:1-10; Bernard 1899: xi-xlv; Belser 1907: 1-14; Brown 1917: xxiii-xx; Hillard 1919: xxiv-xxxv; Wohlenberg 1923b: 15-67; Meinertz 1931: 15-24). In fact, various studies from this time were specifically dedicated to defending the authenticity of the letters (e.g., Böhl 1829; Baumgarten 1837; Good 1848; Doumerge 1856; Mangold 1856; Dubois 1856; Otto 1860; Märcker 1861; Ginella 1865; James 1906; Bruston 1910; Maier 1920).

One of the key points that these works were forced to explain was why the Pastorals diverge stylistically from the other Pauline writings. To account for this situation, interpreters commonly appealed to the fact that stylistic differences are also present among the other Pauline epistles; as such, they maintained that the same should be expected in this instance as well (Bertholdt 1819: 3510; Heydenreich 1826–1828: 13; Böhl 1829: 8-9; Wiesinger 1850: 237-57; Meinertz 1931: 17). Further, it was stressed that the language by which Paul expressed himself in the undisputed letters would not exhaust his vocabulary, and therefore, scholars must be open to the possibility that he could employ new words and different expressions (Flatt 1831: 462-65).

Aside from this appeal to the breadth of Paul’s linguistic abilities, other factors were also noted when accounting for stylistic variation. One was the genre of the letters. The Pastoral Epistles represent Paul’s direct communication to individuals whom he had known for many years. By implication, scholars maintained that the style through which the apostle would have communicated to these co-workers would have been different from a missive directed toward a Christian congregation, which was intended to be read as a public document (e.g., Hort 1894: 131; Parry 1920: cxii-cxiv; Lock 1924: xxix). Others pointed to the variations that would have arisen simply on the basis of changes in subject matter (so, e.g., Ellicott 1883: xx; Plummer 1891a: 7; Alford 1868: 31-36) or those resulting from new circumstances (e.g., Böhl 1829: 10-11; Hillard 1919: xxxii-xxxiv). Proponents also noted that the ‘heretical’ ideas addressed in the Pastorals had not been otherwise encountered by Paul; therefore, the situation would have required a fresh vocabulary (e.g., Hemsen 1830: 197-210; Belser 1907: 13-14; Brown 1917: xxvi-xxvii). Finally, some appealed to the apostle’s advanced age to account for these stylistic differences. If the Pastorals are placed at the end of Paul’s ministry, with 2 Timothy being composed just prior to his death, this would have provided an opportunity for the apostle to gain further proficiency in the Greek language and would thus account for the improved stylistic quality (Böhl 1829: 10; Bernard 1899: xxxvi). At the same time, if Paul was imprisoned for a long period in Rome, as many proponents of authenticity claimed, this would also explain the influences of Latin on the letters (e.g., Wohlenberg 1923b: 55; Hitchcock 1928; cf. also Delitzsch 1851: 726).

Even though explanations like these were the common refrain among those who defended the Pastoral Epistles, the case set forth by Schleiermacher, Eichhorn, Holtzmann, and Harrison was at least enough to cast doubt on the adequacy of explaining these stylistic discrepancies as the standard variation within the works of a single author. For those willing to acknowledge the strength of their case, the amanuensis hypothesis provided an alternative explanation for the linguistic divergence. The first interpreter to move in this direction was H.A. Schott. Recognizing the difficulties created by the style of the Pastorals, but also wanting to avoid labeling the epistles as pious frauds, he viewed the secretary hypothesis as a mediating solution. Schott proposed that an associate composed the Pastorals due to the fact that Paul did not want to write out the letters by hand or to dictate them verbatim (Schott 1830: 324-25; cf. Warren 1894: 406-407). This companion—whom Schott conjectured to have been Luke—was granted license to write about issues that he and Paul had previously discussed and even to offer his own elaboration on such topics. At the end of these letters, Schott maintained that the apostle would have added his own salutation as a way to confirm their contents.

By the early 20^th century, the amanuensis hypothesis had grown in popularity and had become a relatively common means of explaining the style of the Pastoral Epistles (e.g., Harvey 1888: 10; Warren 1894; Bernard 1899: xli; Bennett and Adeney 1899: 414; Laughlin 1905: 22-24; Scott 1909: 329-71; White 1910: 71-72; Rackham 1912: 384; Milligan 1913: 102-103; Headlam 1913: 7; 1920: 49; Allen and Grensted 1913: 216; Cobern 1929: 102-103; Bisseker 1920: 881; Gore 1922: 214; Sayce 1922: 488). Perhaps this was because, unlike the case of 2 Peter, the Pastorals allowed interpreters to identify a specific secretarial candidate, whose writings could be used to compare their stylistic similarity. Based on 2 Tim. 4.11, wherein the author claims, ‘Only Luke is with me’, most who adopted this position attributed the composition of the letters to Luke.

Variations of the amanuensis hypothesis were also put forth at this time. One alternative was proposed by R.A. Falconer. He maintained that while Luke composed all three of the Pastoral Epistles, they were not written at the same time. According to Falconer, 2 Timothy was transcribed during Paul’s imprisonment; 1 Timothy and Titus, on the other hand, were not composed until sometime after Paul’s death (Falconer 1922: 593). This represents a combination of a secretary theory and pseudonymity. A different type of amanuensis hypothesis was suggested by F.W. Farrar. Since Paul regularly employed secretaries in his other extant letters (cf. Rom. 16.22; 1 Cor. 16.21; Gal. 6.11; Phlm 19), Farrar suggested that the Pastoral Epistles were autographic, being directly composed by Paul himself (Farrar 1880: 745). This theory allows for the possibility that amanuenses could have influenced the style and content of individual letters, without assuming that they were involved in every composition.

The Modern Formulation of the Amanuensis Hypothesis

In the early part of the 20^th century, a survey of contemporary trends in New Testament scholarship noted that a proper understanding of secretarial activity in the ancient world held out important implications for the authorship of the Pauline epistles. In particular, it was noted that this issue ‘may eventually prove to have considerable effect upon criticism based upon the language and style of the various Epistles’ (Jones 1914: 287; cf. Milligan 1913: 26).

This observation was especially pertinent given that, at this time, there was a wide variety of opinions on secretarial freedom. Many believed that the amanuenses of Paul functioned like stenographers, trying to keep pace with the rapidity of the apostle’s spirited and sometimes disjointed dictation (Moffatt 1911: 50 n. ¶; Pelly 1917: 10; Jenkins 1925: 169-71). While accident and inability sometimes hindered these aspirations, the intent was to write down every word of Paul exactly as it was spoken. To facilitate the task, it was sometimes maintained that the secretary employed shorthand (e.g., Kent 1916: 127; Allworthy 1918: 559). In certain instances, however, interpreters viewed it as necessary to posit a freer role for secretaries. However, the secretary used for the letter to the Romans may have transcribed Paul’s words very close to (if not identical with) the way they were dictated, in other letters (e.g., Colossians; Ephesians; Pastoral Epistles) the secretary may have been given simply a plan or a few ideas and allowed the freedom to write up the letter on their own (Harris 1898: 402; Nairne 1920: 61; Cobern 1929: 584-85). Most variations in vocabulary and style were thus explained by the manner in which Paul used secretaries.

The problem was that, even among those who adopted the amanuensis hypothesis, the impact of secretaries on the composition of the Pauline epistles was more often assumed than demonstrated (see Jones 1914: 288). What scholarship (still) lacked was a definitive treatment of the processes of ancient letter-writing, one devoted to discerning the level(s) of compositional freedom granted to amanuenses. This situation was remedied in the 20^th century. It was at this time that scholars began to focus attention on ancient letter-writing, and secretarial activity was a natural point of interest. This shift in focus resulted in the modern formulation of the amanuensis hypothesis.

The Amanuensis Hypothesis and the Epistle of James

Among the various reasons why the authorship of the epistle of James has been challenged, none are perhaps as significant as its literary style. Over the years, some have questioned whether James would have been capable of writing in Greek (e.g., Reicke 1964: 4). But for most, objections have not been based simply the expectation that a brother of Jesus would have written in Aramaic (or even Hebrew). The problem is the elevated level of Greek: the letter represents some of the highest quality Greek in the New Testament, surpassed only by Luke-Acts and Hebrews. Reservations of critics have thus centered on whether a person from the social, economic, and geographic location of James would have been able to compose Greek at the level reflected in the epistle. The following explanations are representative of the types of clarifying comments that surround this discussion (cf. also Pratscher 1987: 210-11; Tsuji 1997: 41-42; Painter 2012: 21):

The issue is not whether someone in Palestine could speak and read Greek. That is amply demonstrated by inscriptions and other data. Nor is it unlikely that some members of the Jerusalem Christ group could write Greek or that even that the brother of Jesus, who surely spoke Aramaic as his first language, could speak and write some Greek. The issue is whether James, from a family of artisans, could have acquired the level of Greek displayed in the letter of James…. The question is whether James, from a non-elite family, without access to expensive grammatical and rhetorical schools, could have acquired the facility in Greek that is evidenced in the letter of James (Kloppenborg 2022: 25; original emphasis).

Despite such caveats, some have been strongly opposed to any claims that challenge the linguistic and/or literate abilities of James, the brother of Jesus. According to Blomberg, ‘this view represents about as blatant a classist and ethnic prejudice imaginable’ (2016: 467). Similarly, McKnight describes this as ‘the logical fallacy of applying what may have been the general situation statistically to a particular person’, noting that ‘[t]here are always exceptions to the average’ (2011: 31; cf. Bernier 2022: 207-208). However, this line of reasoning is neither prejudiced nor fallacious; it is simply the outworking of the historical method. Investigations into the linguistic environment of first century Roman Palestine and the educational opportunities that were available to groups and/or individuals living at that time reveal certain information about what was typical for most people or for particular groups of people. It is not a stretch to then apply those conclusions to a given member of that society. One could grant that there may have been exceptions to these general trends; but due to the fact that such cases were atypical, critics have merely argued that sufficient evidence is required to posit that any given individual should be considered such an exception.

Some who defend traditional authorship claims have argued that James the Just would have possessed the ability to directly and independently compose the letter (e.g., Sevenster 1968: 3-21, 190-91; Roberts 1984: 13-15; Motyer 1985: 18-19; Stulac 1993: 13-14; Johnson 1995: 116-18; Richardson 1997: 23-24; Wall 1997: 7; Moo 2000: 14-15; Doriani 2007: xiv; Witherington 2007a: 399-400; Blomberg and Kamell 2008: 33; Besançon Spencer 2020: 37-39). This confidence is based on the spread of Hellenism within Jewish culture following the conquests of Alexander the Great (see esp. Hengel 1969). To demonstrate the extent of Hellenization across ancient Palestine, interpreters commonly point to the use of Greek loanwords in rabbinic literature as well as the preference for Greek on Jewish epigraphy (esp. ossuaries). What such defenses have not produced, however, is any evidence for widespread Greek literacy among rural Galilean craftsmen, or further information about the specific literate and/or linguistic abilities of James. In the absence of such evidence—and given the questions that surround the extent of Hellenistic influence in early Roman Galilee (see, e.g., Chancey 2002; 2005)—many interpreters have explained the stylistic quality of the letter in ways that (to varying degrees) eliminate James’ direct involvement in the composition process.

It is common to interpret the letter’s elevated style as evidence that it is a pseudonymous composition (e.g., Kugelman 1980: 10; Martin 1982: 11-12; Scheef 1983: 58; Vouga 1984: 17-18; Pratscher 1987: 211; Schrage 1993a: 11; Perkins 1995: 84; Burchard 2000: 3-6; Popkes 2001: 64-68; Hartin 2006: 8; Allison 2013: 25-26; Cargal 2016: 655; Metzner 2017: 3-4; cf. Theißen 2003: 55; Konradt 2009: 578). In fact, this view has become the majority within modern scholarship (see Hahn and Müller 1998: 62-63), although the position was widely represented within an earlier generation as well (e.g., von Soden 1899: 175; Bauer 1910: 15; Ropes 1916: 47-52; Marty 1935: 243-49; Windisch 1951: 3; McNeile and Williams 1953: 203-205; Thyen 1955: 14; Williams 1965: 93-95; Michl 1968: 19-20; Dibelius and Greeven 1976: 17). Only occasionally will one come across a downplaying of the stylistic evidence among those who interpret the letter as an early pseudepigraphon (e.g., Laws 1980: 40; Painter 2004: 238).

Another way that scholars have resolved the stylistic problem is through the translator hypothesis, a view that originated as far back as the 18^th century (Faber 1770: 4; Schmidt 1804: 307-308; Bolten 1805: 17; Bertholdt 1819: 3033-3040; Wordsworth 1885: 136-50; Burkitt 1924: 65-70). Proponents of this view maintain that the canonical letter of James was originally written in Hebrew or Aramaic (perhaps by James, the brother of Jesus) and was later (freely) translated by someone fluent in the Greek language. Since the letter does not read like a wooden translation, one would have to attribute stylistic features such as syntax, word choice, rhetoric, etc. to the Greek translator, indicating that he (or she?) played a fundamentally determinative role in the shape of the epistle. If this evidence were pressed further, the active role of the translator could also imply—although it need not—that James was unable to compose Greek at this level on his own.

A similar but slightly different view has been advocated more recently by Davids (1982: 10-13, 57-59) and Martin (1988: lxix-lxxvii). They maintain that the epistle of James represents two compositional layers that were added a different times: an original piece of teaching/tradition, written in a Semitic language (perhaps by James), which was subsequently revised in Greek by one skilled in that language (see also Kittel 1942: 79; Knox 1945: 10-17; Blackman 1957: 26-27; Schneider 1961: 4; Mitton 1966: 232; Niebuhr 1998: 431; cf. others who allow for this possibility: Wright and Bird 2019: 735). Much like the previous proposal, in this redactor hypothesis the elevated Greek of the letter is attributed not to James but to the reviser. What distinguishes the two is that the redaction is thought to have occurred at a time and place that was separate from the actual author (i.e., James), which represents a serious problem in the eyes of some interpreters (see Sleeper 1998: 40).

Finally, numerous interpreters seek to circumvent the stylistic problem presented by the letter of James through the postulation of a secretary. Among these proponents of the amanuensis hypothesis, some are hesitant to affirm that James would have been able to independently compose the letter attributed to him (e.g., Klijn 1967: 150; Mussner 1981: 8; Palmer 1997: 3-4; Bateman 2013: 55-56). These doubts usually derive from the widespread illiteracy levels that existed in Roman Palestine or from the quality Greek in which the letter was written. By positing that an amanuensis was afforded considerable compositional latitude, interpreters are able to account for the present form of the epistle without severing all connections to the historical James. Many who adopt the theory, however, believe that James would have been quite capable in Greek, and thus they argue that any assistance he received from a secretary would have consisted merely of editorial improvements (Bennett 1901: 22; Mitchell 1911: 34-35; Mayor 1913: cclx-cclxviii; Hayes 1921: 96-97; Chaine 1927: xci-cviii; Beasley-Murray 1965: 19; Sidebottom 1967: 19; Hengel 1987: 251-52; Harrison 1992a: 4; Bernheim 1997: 226-27; Bauckham 1999: 21-24; Brosend 2004: 4-5; Guthrie 2006: 200-201; Burge, Cohick, and Green 2009: 381; McCartney 2009: 27-29; Jobes 2011: 150-58; McKnight 2011: 31-34; deSilva 2012a: 48; Vlachos 2013: 3-5; Coker 2015: 61-62; Blomberg 2016: 467; Anderson 2017: 3; Cheung and Spurgeon 2018: 3).

One significant obstacle that lies in the way of the amanuensis hypothesis in this case is that the letter of James does not (explicitly) indicate that a secretary was involved in the composition (noted, e.g., by Sevenster 1968: 3; Tsuji 1997: 43; Metzner 2017: 4). To address this issue, some appeal to the very stylistic qualities they are attempting to explain. McCartney, for instance, argues the high-quality Greek is a sign of an amanuensis:

James has no explicit marks [indicating the presence of an amanuensis], and given that we have only one letter from James to consider, it is difficult to do more than hypothesize regarding implicit indications. The one consideration that might apply is that the language of James is stylistically more like what one would expect from an educated Hellenist than from a Palestinian Jewish whose knowledge of Greek would have been only secondary (2009: 28).

But not only does this argument assume what needs to be demonstrated (viz. that ancient amanuenses were able to extensively shape the style of the letters they transcribed), some have reached the opposite conclusion about secretarial involvement based on stylistic considerations (see Holloway 2010: 574, who argues that there is evidence within the literary character of the letter that points away from the presence of an amanuensis).

Others base their conjecture of secretarial assistance on the regularity of this practice. In what seems like a slight departure from his earlier redactor hypothesis, Davids notes, ‘Since it was rare for a person, especially an important person, to write his or her own letters, some sort of amanuensis was certainly used, and a trusted amanuensis would likely improve the style of a letter’ (1999: 43 n. 28). But this raises an important question: How common was the use of a secretary in epistolary composition? Ever since the time of Roller, proponents of the amanuensis hypothesis have claimed that the practice was quite common based on clues in the literary sources. Research on epistolary subscriptions in the documentary papyri seem to challenge this assertion, however (see Sarri 2018: 192, who claims that ‘the vast majority of the very large number of surviving private letters that circulated and have been preserved to us were probably written by their authors themselves’).

The Amanuensis Hypothesis and the Epistles of 1–2 Peter

Discussions of authorship in 1 Peter, as with the epistle of James, must explain letter’s elevated style. Some account for this situation by attributing to the apostle Peter a high level of fluency in Greek (Spicq 1966: 17-26; Kelcy 1972: 7-9; Barbieri 1977: 18-22; Grudem 1988: 25-31; Hiebert 1992: 11-20; Miller 1993: 57-75; Clowney 1994: 18-21; McKnight 1996: 26-29; A. Black 1998: 14-16; Waltner 1999: 176-78; Jobes 2005: 6-8; Powers 2010: 26-31; Keating 2011: 18-19; Lee 2018; Schreiner 2020: 14-16). Since the predominant language in early Roman Galilee is generally agreed to be Aramaic (see Poirier 2007: 56-64), the defense of this position normally begins with an attempt to establish that Peter was bilingual—not merely displaying a basic acquaintance with the Greek language but possessing a considerable degree of proficiency in it (for a full defense, see Lee 2018: 158-81). Consistent with this task, some contend that Bethsaida, the hometown of Peter, had been especially impacted by Hellenism. Around the time of Jesus’ death, Philip the tetrarch renamed the village ‘Julias’, in honor of Augustus’ daughter, and afforded it the status of a polis (see Josephus, Ant. 18.28; cf. War 2.168). In this environment, proponents argue, it would have been difficult to have escaped Greek influence. Further, the fact that Peter’s brother was known by a Greek name (Andrew) is thought to indicate some level of Hellenization within the family (cf. Mason 1884: 385; Ebright 1917: 82; Spicq 1966: 23; Kelly 1969: 31). Others maintain that the commercial interactions inherent in Peter’s occupation as a fisherman would have given rise to a need for fluency in the Greek language as well (e.g., Witherington 2007b: 52; Judd, Jr. 2014: 252-53; Schreiner 2020: 14).

Even among those who seek to defend the letter’s authenticity, however, it is generally acknowledged that the language of 1 Peter rises above what could have been accomplished by the apostle and thus presents a significant problem to traditional authorship claims (see Cranfield 1960: 13-14; Green 2019: 85; Keener 2021: 10). While this was not the primary reason why the authenticity of the letter was first challenged (on the early challenges, see Williams and Horrell 2023: 1:121-22), stylistic considerations have become an important factor in this discussion. In recent scholarship, the language of 1 Peter continues to be one of the main reasons why the letter is understood to be pseudonymous (e.g., Frankemölle 1990: 9-11; Goppelt 1993: 48-53; Brox 1993: 43-47; Schrage 1993b: 64-65; Krodel 1995a: 63-64; Schweizer 1998: 15-18; Boring 1999: 30-37; Seethaler 2003: 13-16; Bony 2004: 11-14; Feldmeier 2008: 32-39; Gunsalus González 2010: 4-5; Vinson 2010: 1-7; Donelson 2010: 15-17; Schlosser 2011: 35-37; Vahrenhorst 2016: 9-16; Bartlett 2016: 667; Wagner and Vouga 2020: 8; Ostmeyer 2021: 17-19; Williams and Horrell 2023: 1:122-31, 149-62). Even some who defend the authorship claims of the letter find it difficult to attribute a high degree of literate and linguistic competency to the apostle Peter. As a result, they are forced to postulate alternative compositional scenarios.

One way of explaining the elevated style of 1 Peter is that the letter was the product of a Petrine circle (Best 1971: 59-63; Elliott 1980: 250-67; Elliott 1990: 267-95; Elliott 2000: 127-30; Blevins 1982: 402-403; Soards 1988: 3827-49; Knoch 1990: 22-25; Knoch 1991: 105-26; Perkins 1995: 13; Achtemeier 1996: 42; Dschulnigg 1996: 175; Richard 2000: 9-11; Prasad 2000: 36-46; Böttrich 2001: 247-48; Senior 2003: 4-6; Klauck 2006: 340; Chatelion-Counet 2006: 403-24; Prigent 2006: 10; Hartin 2006: 29; Le Roux 2018: 15-21). According to this view, a group of disciples had formed in Rome around the influence of Peter. Following his death, the teachings of the apostle were used to compose the letter of 1 Peter and then were dispatched under his name. While this view maintains that someone other than Peter composed the epistle, it nonetheless allows proponents to maintain a connection with the apostle, for it is his ideas and teachings that are being employed. Where defenders of the letter’s authenticity take objection to this theory is that it assumes that the letter was composed without the direct commissioning or consent of the apostle (for a critique of this view, see Horrell 2002: 29-60).

Although not as widely posited in modern scholarship, another way to explain the style of 1 Peter is by appealing to a translator. In other words, some maintain that Peter had little-to-no aptitude in Greek and thus was required to communicate his message in a Semitic language. In this case, Peter orally dictated the letter to an amanuensis in either Aramaic or Hebrew, and the amanuensis rendered the contents of Peter’s discourse into Greek (e.g., Olshausen 1838: 145; Schonfield 1955: 420; Picirilli 1992: 93, 212). Alternatively, a few have proposed that 1 Peter was originally a written document composed in Aramaic and that the secretary’s role was simply to translate the text into Greek (e.g., Bertholdt 1819: 3078-79; cf. also Baronio 1864: 1:297, who claims that it was written in Hebrew and translated by Mark). In both instances, the translator would ultimately be responsible for the final product. The biggest obstacle for this view is that the letter provides no indication that it was translated, much less from a Semitic language. While exhibiting some degree of Semitic interference (see Jobes 2003), the style reflects a moderately polished form of Greek composition.

The only interpretation that allows scholars to attribute the letter to the historical Peter, regardless of his literate abilities or his level of proficiency in Greek, is the amanuensis hypothesis. What is more, 1 Peter is the only one of the Catholic Epistles that might potentially refer to a secretary. In 1 Pet. 5.12, the author states, διὰ Σιλουανοῦ ὑμῖν… ἔγραψα (‘I have written to you through Silvanus’). Many modern interpreters take this phrase to mean that Silvanus performed the role of secretary for Peter (for a full defense of this position, see Keener 2021: 393-402; against this interpretation, see Richards 2000; Williams and Horrell 2023: 615-24). Proponents of this view differ slightly on the reason why Silvanus functioned in this role. Some claim that in antiquity it was customary to employ amanuenses to transcribe letters, even for those who could perform the task themselves; therefore, the secretarial duties of Silvanus were merely undertaken with a view toward editorial improvement (e.g., Hart 1910: 12-15; Blenkin 1914: xxiii-xxiv; Selwyn 1947: 9-11, 25-27; Stibbs and Walls 1959: 20-30; Wheaton 1970a: 1236; Blum 1981a: 211-12; Holmer and de Boor 1986: 16-17; Marshall 1991: 22-23; Bénétreau 1992: 36-41; Hillyer 1992: 2; Witherington 2007b: 41-42; Osborne 2011a: 131-32; Watson 2012: 4; Rodgers 2017: 2-3; Keener 2021: 9-13; cf. Helyer 2012: 109-11; Judd, Jr. 2014: 251-55, 257-59; Gäckle 2015: 12-13; Green 2019: 85-93). Others maintain that the decision to employ a secretary arose out of necessity. Since it is unlikely that a Galilean fisherman would have possessed the skills to compose such an epistle, Peter was required to seek out someone with the requisite abilities (e.g., Robinson 1929: 1339; Willmering 1953: 1177; Cranfield 1960: 13-14; Reicke 1964: 69-71; Beasley-Murray 1965: 51; Fitzmyer 1968: 2:362; Thompson 1983: 80-81; Mounce 1983: 3-4; Michaels 1988: lxii; Davids 1990: 4, 6-7; Skaggs 2004: 3; Bosetti 2004: 15; Charles 2005: 278; Green 2007: 6-7; Warden 2009: 12). In either case, as Peter’s amanuensis, Silvanus is assigned an active part in the composition of the letter.

The importance of the amanuensis hypothesis lies in the fact that it alleviates many, if not most, of the objections used to challenge Petrine authorship. This is acknowledged even by those who attribute pseudepigraphic status to 1 Peter (e.g., Fuller 1966: 157; Stanton 2003: 1495). If secretaries in antiquity were given considerable latitude in a letter’s composition, and if Peter employed Silvanus as his amanuensis, then the quality Greek style in which 1 Peter was written would reveal little about the actual author. Such a scenario would allow for the possibility that a Galilean fisherman ultimately stood behind the epistle (see Michaels 1988: lxii). In addition, the secretarial role of Silvanus could potentially explain why 1 Peter shares so many linguistic connections with the writings of Paul—given that Silvanus was a close associate with Paul (even co-authoring a letter with him)—and why the scriptural citations in the letter were drawn from the LXX rather than the Hebrew Bible.

In the case of 2 Peter, the situation is very similar. The overwhelming majority of modern interpreters reject the authenticity of 2 Peter, with the style being one of many contributing factors (e.g., Schrage 1993c: 126-27; Fuchs and Reymond 1988: 30-32; Schelkle 1988: 179-81; Knoch 1990: 215-18; Frankemölle 1990: 80-82; Paulsen 1992: 89-91, 93-95; Neyrey 1993: 118-20, 134-35; Vögtle 1994: 122-27; Craddock 1995: 92; Danker 1995: 84; Knight 1995: 22-23; Perkins 1995: 160-61; Horrell 1998: 136; Richard 2000: 308-309; Kraftchick 2002: 75-76; Harrington 2003: 236; Skaggs 2004: 86-88; McKnight 2003: 1504; Hartin 2006: 48-49; Wilson 2010a: 269-70; Donelson 2010: 208-209; Aichele 2012: 4-9; Callan 2012: 137-39; Frey 2018: 213-20; Ostmeyer 2021: 107-108). Much of the suspicion regarding the style of 2 Peter stems from the fact that, despite some early assertions to the contrary (Hug 1847: 2:480-82; Weiss, 1866: 286-300; Warfield 1881: 66-68), the vocabulary, syntax, structure, and rhetoric is notably different from 1 Peter (cf. Mitchell 1911: 64; Cranfield 1960: 149). Nevertheless, many of the same explanations used to defend the authorship of 1 Peter are applicable to 2 Peter as well. For even if there is no direct reference to a secretary in the latter, the possibility that an amanuensis was employed in the former would account for stylistic differences between the two epistles. On the one hand, scholars have argued that 1 and 2 Peter were each composed by different amanuenses (e.g., Plumptre 1879: 79-81; Bennett 1901: 65; Falconer 1902: 47-56; Dillenseger 1907: 193-94; Vrede 1921: 149; Heupler 1942: 627; Charue 1946: 478; Holzmeister 1949: 353; Mounce 1983: 98; Hillyer 1992: 10-11; Picirilli 1992: 219-24; M.C. Black 1998: 145; Charles 1999: 260-63; Powers 2010: 162-65; Warden 2009: 297-306; Osborne 2011b: 275-77; Green 2008: 139-50; Bateman 2013: 55-56; Mbuvi 2015: 16-17; Harmon 2018: 366; Bayer 2019: 116; Schreiner 2020: 298-323). This would be consistent with the tradition that Peter employed various other ‘interpreters’ to record his memoirs (e.g., Mark, Glaucias). The other alternative is that Peter employed a secretary in 1 Peter (viz. Silvanus), but that he composed 2 Peter himself (e.g., Spitta 1885: 530-39; Zahn 1897–1899: 2:89-110; Wheaton 1970b: 1249; Bowman 1982: 156-57, 159-60; Holmer and de Boor 1986: 185-87; Hiebert 1989: 1-20; Lucas and Green 1995: 239-40; Moo 1996: 21-26; cf. those allow for either possibility: Blum 1981b: 257-61; Hillyer 1992: 9-11; Harvey and Towner 2009: 13).

The Amanuensis Hypothesis and the Epistle of Jude

Until approximately the mid-20^th century, the elevated style of Jude had little bearing on the question of authorship. For the most part, considerations other than language played the decisive role in evaluations of the letter’s authenticity. These included the identity of letter’s opponents, the non-canonical source materials cited in the epistle, and whether certain internal (viz. the reference to the predictions of the apostles, v. 17) and external (viz. the age of the historical Jude) considerations point to post-apostolic dating. Among those who defended traditional authorship claims during the 19^th and early 20^th centuries, the letter’s style was rarely (if ever) mentioned (e.g., Huther 1877: 245-47; Alford 1878: prol. 4:188-92; Keil 1883: 287-91; Plummer 1891b: 365-76; Kühl 1897: 291-93; Bennett 1901: 92-96; Bigg 1902: 317-22; Mayor 1907: cxlvi-clii; Mitchell 1911: 69-70; James 1912: xxxvi-xxxviii; Wand 1934: 187-90; Chaine 1939: 267-69; Lenski 1945: 597-603). Even within more recent treatments, this question is sometimes left unaddressed (e.g., Harrison 1992b: 327-28; Moo 1996: 27-28; Charles 1999: 278-80; Mills 2003: 1319; Warden 2009: 453-54; Mbuvi 2015: 10-12; Bateman 2017: 11-19).

For those who maintain that Jude did not possess the linguistic and/or literate abilities to compose the letter independently, two options remain. First, it is possible to view the epistle as a pseudonymous composition. This has been the approach of the majority of critical scholars since approximately the mid-20^th century. Most who take this position point to the elevated style of the letter as a primary reason why it could not have been written by the brother of Jesus (Schelkle 1988: 140-43; Fuchs and Reymond 1988: 148; Frankemölle 1990: 129-30; Paulsen 1992: 44-45; Neyrey 1993: 29-31; Schrage 1993d: 227-28; Vögtle 1994: 10-11; Krodel 1995b: 107-109; Richard 2000: 236-38; Kraftchick 2002: 20-22; Seethaler 2003: 125-26; Harrington 2003: 182-83; Hartin 2006: 48-49; Wilson 2010b: 380; Frey 2018: 25-30; Ostmeyer 2021: 163-64).

For those who wish to defend the letter’s authenticity, the alternative is to conjecture that Jude received assistance from an amanuensis. This view has been occasionally espoused because there of doubts about Jude’s ability to compose in elevated Greek. Cantinat, for instance, notes that the letter discloses ‘a thorough knowledge of the Greek language, which only a Hellenistic Jew could possess’ (1959: 607). This is a problem because ‘the Aramaic upbringing of St. Jude, who was from Nazareth, would certainly not allow him to write Greek this well’ (1959: 607). Nevertheless, Cantinat acknowledges that ‘among the disciples of the leaders in the early church, there was no shortage of trained Hellenists who could easily place their talents at the service of the common good’ (1959: 607; cf. Wohlenberg 1923a: xxxix-xl, xliii; Manser 2009: 212). Even for those who maintain that the linguistic environment and missionary experiences of Jude would have provided a sufficient basis for composing the letter, allowance is still made for the use of an amanuensis. This decision is normally grounded in the frequency with which secretaries were employed in the letter-writing process (Chaine 1939: 281; Dalton 1969: 1263; Watson 1998: 474; Brosend 2004: 4; Green 2008: 4-9; deSilva 2012b: 179-83).

Despite these appeals to other compositional theories, there is still a large constituency who maintains that Jude could have composed the letter, even with its elevated Greek style (e.g., Blum 1981c: 381-82; Bauckham 1983: 14-16; Hiebert 1989: 185-92; Green 1987: 48-52; Skaggs 2004: 148-50; Witherington 2007a: 563-67; Green 2008: 4-9; Harvey and Towner 2009: 146; Powers 210: 252-53; Osborne 2011c: 357-58; Schreiner 2020: 484-90; cf. Ellis 1978: 226-35). As in the case of James and Peter, few interpreters have attempted to argue that Jude would have received a literate education from a young age; instead, most seek to justify Jude’s writing abilities through extended missionary work into Greek-speaking areas. The explanation of G.L. Green is representative of such an approach. He argues, ‘we know that Jude, along with the other brothers of Jesus, engaged in itinerant ministry, which presupposes that they either had or acquired fluency in Greek… How they acquired this skill we can only surmise, but that they had gained some mastery of the tongue is hard to deny on historical grounds’ (Green 2008: 8; cf. also Osborne 2011c: 358).