By: Dr. Simon Gathorcole (Article)

“This article uses British English spelling. While the information presented on this topic will align with similar articles over time, I am sharing it here for convenience and to benefit fellow Christians.” –  AC

Ancient Authors and the Use of Third-Person Narrative: Cases of Initial Anonymity and Ambiguous Attribution

1. Xenophon – Written entirely in the third person, possibly pseudonymous; debates exist about initial anonymity.
2. Julius Caesar – Commentarii de Bello Gallico and Commentarii de Bello Civili: Written in the third person, attributed to him but lacking self-reference.
3. Plutarch – Parallel Lives: No self-reference within the text, leading to initial perceptions of anonymity.
4. Josephus – Antiquities of the Jews: Lacks self-reference in this work, though he names himself in Jewish War.
5. Lucian of Samosata – Alexander the False Prophet, The Passing of Peregrinus: Third-person narratives with indirect self-references.
6. Nepos – Lives of Eminent Commanders: No self-reference in the preface, leading to a perception of initial anonymity.
7. Suetonius – Lives of the Caesars: Initial pages lost; unknown if self-reference existed, leading to ambiguity about initial attribution.
8. Herodotus – Histories: Brief self-reference at the beginning but primarily third-person narrative.
9. Thucydides – History of the Peloponnesian War: Minimal self-reference; uses third-person narration.

Command/Ctrl + F to find the individuals above faster in the article

Abstract 

The apparent anonymity of the Gospels is a neglected topic in NT studies. The present  article offers an investigation of it. It will survey the work that has been done specifically on the subject, as well as how it is treated in Gospels scholarship more  broadly. The main body of the argument is in two parts. First, anonymity cannot be  defined by the lack of reference to a name in the body of the work, and therefore the argument that the Gospels are anonymous because they do not contain the authors’  names is invalid. Secondly, and more positively, while the titles contained in the  earliest Gospel manuscripts may well in their present form be secondary, this does not  exclude attributions of authorship made in some other way. Aspects of practical  necessity make the presence of author’s names very likely. Second-century Christian  literature is replete with references to Matthew, Mark, Luke and John as authors of  Gospels, and there is never any sense that the Gospels were anonymous or written by  others. The most likely conclusion to be drawn is that the attributions of authorship  are original. 

It has been said by Terry Eagleton that ‘all literary works are anonymous, but some  are more anonymous than others.’¹ The canonical Gospels are usually classed very  much on the ‘more anonymous’ side, but surprisingly this apparent anonymity of the  Gospels has received relatively short shrift. What little discussion there has been has  taken place mainly in Germany, and among those discussions treatment of anonymity  has often been lumped in with its bigger sister, pseudonymity. As David Aune has noted, the topic of the anonymity of biblical books ‘has been almost completely  neglected’.² 

This article will begin (§1) with a brief review of scholarship on the subject over the past half-century, both in specialised studies and more broadly. From this it can be clearly seen that the great majority of scholars, including leading figures in current Gospels research such as Michael Wolter and Francis Watson, hold to a view of the Gospels as anonymous. It is the intention of the present article to take issue with this position. The argument against the majority proceeds with negative and positive arguments. The negative argument (§2) consists in an assessment of the significance of there being no author mentioned within the body of the work, concluding that it is  of no significance at all as evidence for anonymity. The more constructive argument (§3) explains how even though the full εὐαγγέλιον κατά (the Gospel according to) … titles may be later, this  does not permit the conclusion of anonymity because there is considerable positive evidence for the antiquity, even the originality, of the names. The aim of the article is  hence to show that the standard reasons for considering the Gospels anonymous cannot support the claim that they are, and to contend instead that the most likely  conclusion from the evidence is that the attributions go back to the first century. The                                                          

¹ T. Eagleton, ‘Unhoused’, London Review of Books 30.10 (22 May 2008), p. 19. 

² D.E. Aune, ‘Anonymity’, in idem (ed.), The Westminster Dictionary of New Testament and Early  Christian Literature and Rhetoric (Louisville: WJK, 2003), p. 35. I owe this reference to the article of  Baum discussed further below.

1 

aim here is not to say that Matthew, Mark, Luke and John wrote the Gospels, still less to argue for a fourfold Gospel canon. Nor is the argument concerned with the form or antiquity of the εὐαγγέλιον κατά (the Gospel according to) titles in toto. The point is to argue, in part drawing on evidence already adduced by others, but with new arguments and new testimonia, 

that as far back as we can go, and probably from the beginning, these were the names  attached to the Gospels. 

1. The Anonymity of the Gospels in Current Gospels Scholarship 1.1 Specialised Studies 

One of the most important articles to spark off interest in the subject is that of Kurt Aland, in an essay covering both anonymity and pseudonymity.³ Aland takes it as read that these two topics need to be considered together (p. 40). He includes in the category of the ‘anonymous writings’ of the New Testament Hebrews, 1 John and the Gospels (p. 41): ‘In my opinion it is beyond doubt that all the gospels were published  anonymously. Our present opinion about their authors dates from information which  derives from the time of Papias or later’ (p. 42).⁴ As one might expect from Aland, he  makes reference here to some of the manuscript evidence. The reason he offers for the anonymity of the Gospel writers lies in the fact that the authors regard themselves as mere quills of the Holy Spirit. When Luke wrote, for example, his Gospel ‘just passed  through him as a person… He was but the pen moved by the Spirit.’ We would need  an explanation if the author had provided his name (p. 45). The written Gospels, Aland avers, are simply the transposition of spirit-inspired oral discourse into writing. The shift towards named authors (like Justin) of Christian works takes place in the second century, when prophetic inspiration (along with expectation of the parousia) wanes (p. 47). 

A second article by Aland repeats many of the concerns of the former piece.⁵ These  include the needs to treat the two topics together, to take a broader view encompassing Christian literature of the first two centuries, and to treat Christian literature as distinctive – not out of a Christian exceptionalism, but because of the  special factors of the prophetic spirit and the delay of the parousia which Aland sees as having a particular impact on authorship. Aland begins by noting that anonymity has attracted much less attention than pseudonymity (p. 122). Overall, not a great deal is added for our purposes: only one paragraph deals with the anonymous literature of the NT, and that is taken up with the epistle to the Hebrews. 

                                                        ³ K. Aland, ‘The Problem of Anonymity and Pseudonymity in Christian Literature of the First Two  Centuries’, JTS 12 (1961), pp. 39-49, reprinted in idem, The Authority and Integrity of the New  Testament (London: SPCK, 1965), pp. 1-13. German version: ‘Das Problem der Anonymität und  Pseudonymität in der christlichen Literatur der ersten beiden Jahrhunderte’, in idem, Studien zur  Überlieferung des Neuen Testaments und seines Textes (Berlin/ New York: De Gruyter, 1967), pp. 24- 34. Page references in the text above are from the JTS publication. ⁴ Aland considers it unclear, however, whether John’s Gospel is really anonymous or pseudonymous  (p. 41). 

⁵ K. Aland, ‘Noch einmal: Das Problem der Anonymität und Pseudonymität in der christlichen  Literatur der ersten beiden Jahrhunderte’, in E. Dassmann & K.S. Frank (eds.), Pietas: Festschrift für  Bernhard Kötting (Münster: Aschendorff, 1980), pp. 121-139.

2 

Partly in response to Aland’s former piece, an article by Horst Balz also provides a  treatment of the two topics.⁶ Balz also considers that they ought to be treated together because ‘Pseudepigraphie ist … als Extremfall der Anonymität zu werten’ (p. 434; cf.  406). The article is more wide-ranging than Aland’s focus on Christian literature of the first two centuries, but on the other hand, the article is taken up with pseudonymity to a much greater extent than with anonymity. In the eight-page  treatment of the pagan context only the last sentence really discusses anonymity, although Balz does cover instances where originally anonymous material becomes pseudepigraphic, such as the works attributed to Homer and Aesop (pp. 408-416).  These examples in his view provide an analogy or a background to what happened with the Gospels (p. 417). When it comes to explaining anonymity in early Christian literature, Balz, in contrast to Aland, focuses not so much on the inspired process of composition, but on the fact the author is only passing on traditional material, and so can – indeed must – remain anonymous (pp. 433-434). The discussion of the Gospels runs to just over one page, and stresses that the authors are focused simply on reworking tradition, with their products entirely marked in Balz’s view by their function of presenting in a stable form the gospel as it was preached in their particular communities (pp. 428-429). These two factors – the author’s modest role as a redactor, and the work’s close connection to its community – rule out for Balz the possibility of the author including his name (p. 429). 

The most important treatment of the subject, and the first to treat anonymity as a separate topic in its own right, was an article by Michael Wolter in 1988.⁷ Wolter agrees with the previous authors on the fact of the Gospels’ anonymity. The Gospels, along with Acts (which forms, with Luke, a single Doppelwerk), Hebrews and 1 John,  are works ‘die ursprünglich überhaupt keinen Verfassernamen genannt haben und  darum als anonym zu bezeichnen sind’ (p. 1). The idea that the titles are original ‘steht heute ausser Frage’ (p. 4). According to Wolter, anonymity differs from  pseudonymity in that the latter seeks to establish continuity with apostolic tradition,  while the anonymous literature of the NT seeks to ground Christ as the sole authority (p. 6). Wolter offers a sociological context for this contrast: the pseudepigraphical writings of the NT assume a settled Christian identity in need of preservation, while the anonymous works feel the need to provide a legimitation of Christian identity, especially in relation to Judaism. John’s Gospel, for example, is an instance of an (originally) anonymous work in which Jesus rather than the author of the Gospel is the exclusive guarantor of Christian identity. (John 21, however, is a later  pseudonymizing addition.) In Mark, too, only Jesus authorises the Gospel and  legitimates a Christian identity which extends beyond Judaism (pp. 11-12). Luke-Acts  is a slightly different case in that Luke 1.1-4 does tie the authority of the work to  apostolic tradition, and Acts presents Jesus as the proclaimed rather than proclaimer;  on the other hand, the aim of legitimizing a Law-free Gospel over against Judaism  remains a constant (p. 13). Notably, Wolter remarks in his Fazit that the anonymity of  the Gospels is not an anonymity in the sense that the particular author wants to remain  unknown: ‘das wäre viel zu modern gedacht’ (p. 15). 

                                                        ⁶ H. Balz, ‘Anonymität und Pseudepigraphie im Urchristentum: Überlegungen zum literarischen und  theologischen Problem der urchristlichen und gemeinantiken Pseudepigraphie’, ZThK 66 (1969), pp.  403-436. ⁷ M. Wolter, ‘Die anonymen Schriften des Neuen Testaments. Annäherungsversuch an ein literarisches  Phänomen’, ZNW 79 (1988), pp. 1-16. Wolter criticises Aland’s lumping of the two topics together (p.  3).

3 

A further significant contribution to the discussion stems from Armin Baum (2008),  who brings anonymity further out of the shadow cast by pseudonymity.⁸ He talks of  ‘the striking fact that the NT Gospels and Acts do not mention their authors’ names…  only the NT letters and the Apocalypse were published under their authors’ names  while the narrative literature of the NT remained anonymous’, citing at the beginning  of his article an assertion about Luke by François Bovon: ‘The absence of the author’s  name in Luke’s prologue remains mysterious to me’ (p. 120). He further notes it as a  ‘fact which has not sparked much interest among New Testament scholars’ that the  Gospels ‘were written and published anonymously’, especially given that they have  prologues (pp. 121, 122). Baum, considering the Gospels and Acts to be ‘history  books’ (so Baum’s title), pays special attention to the fact that in Greco-Roman  history writing, ‘the Greek historian would mention his name and his provenance’ (p.  125). As with the authors already considered, however, the absence of names is no  accident, for the absence of titles was a common feature of ancient Near Eastern, and  especially Old Testament, history writing. In the Old Testament, as in its literary  environment, it was the subject matter which was important, not who wrote it. The  same is true of the Gospels, where the evangelists ‘regarded themselves as  comparatively insignificant mediators of a subject matter that deserved the full  attention of the readers’ (p. 142). 

There have been a few minor treatments which have not added significantly to what  has been said by the protagonists expounded above.⁹ Despite the differences among  these studies, some general points emerge. (i) These scholars highlight the topic as a  

neglected one. (ii) Aland and Balz treat anonymity in relation to pseudonymity, or  even within it: certainly the latter dominates. In Wolter and especially in Baum,  however, anonymity comes of age as a free-standing topic. (iii) All these scholars see the anonymity of the Gospels not merely as a publishing convention but as a function  of the self-effacing posture of the evangelists. (iv) The principal reasons adduced for  the Gospels’ anonymity are the obvious absences of names within the text, as well as  the lateness of the current titles: this is what they mean when they say that the  Gospels were published anonymously. 

1.2 The Dissenting Few 

There have been a few dissenters from this line of scholarship. Kirsopp & Silva Lake,  for example, said of the titles: ‘Why should this testimony not be accepted? No reason  has ever been shown, for the view that antiquity tended to anonymous books is  contrary to evidence, and it is most unlikely that the second Gospel, for instance, ever  circulated without the name of Mark attached to it.’¹⁰ Dibelius and Moles comment  

                                                        ⁸ A.D. Baum, ‘The Anonymity of the New Testament History Books: A Stylistic Device in the Context  of Greco-Roman and Ancient Near Eastern Literature’, NovT 50 (2008), pp. 120-142. He deals with  pseudonymity in his monograph Pseudepigraphie und literarische Fälschung im frühen Christentum:  Mit ausgewählten Quellentexten samt deutscher Übersetzung (Tübingen: Mohr, 2001). ⁹ See e.g. the interesting discussion of the nature of authorship (especially as it relates to Acts) in A.J.  Droge, ‘Did “Luke” Write Anonymously? Lingering at the Threshold’, in J. Frey, J. Schröter & C.K.  Rothschild (eds.), Die Apostelgeschichte im Kontext antiker und frühchristlicher Historiographie (Berlin: De Gruyter, 2009), pp. 495-518. ¹⁰ K. Lake & S. Lake, An Introduction to the New Testament (London: Christophers, 1948), p. 4. 

4 

that the dedication to Theophilus in Luke’s Gospel means that ‘the name of the author  could hardly be omitted from the title’, and Luke’s prologue demands a title with  Luke’s name, or else the first person references in Luke 1.1-4 ‘would be left  hanging.’¹¹ Bauckham’s surgical distinctions between the two Johns attested in the  literature of the second century conclude of John’s Gospel that ‘that there is no  evidence that the Gospel was ever regarded as anonymous’;¹² indeed, more positively,  ‘The fourth gospel was never anonymous.’¹³ The most extensive protest against the  anonymity theory is found in various studies by Martin Hengel, who argued that all  the Gospels were initially circulated with titles in the form εὐαγγέλιον κατὰ Μᾶρκον etc.¹⁴ Hengel’s arguments, some of which we shall revisit in due course,  focus mainly on the practical necessity of the titles especially in the institutions of the  library, the book-trade, and early Christian worship. 

1.3 Gospels Scholarship at Large 

Nevertheless, Gospels scholarship at large very much endorses the picture painted by  Aland and the other studies devoted to anonymity. One scholar states with sovereign  confidence: ‘we know that the original manuscripts of the Gospels did not have their  

authors’ names attached to them.’¹⁵ More modestly, Mary Ann Tolbert says that the  view that the Gospels were ‘originally anonymous’ is ‘the most probable’,¹⁶ and Bock  and Wallace say the Gospels ‘almost surely were anonymous when penned.’¹⁷ After  their composition, the four Gospels were ‘anonym herausgegeben’¹⁸ and then  ‘anonym überliefert’.¹⁹ The reasons for these judgments are the same two principal  reasons that emerged in our survey of the specialist studies. (1) The first is the lack of  inclusion of authors’ names in the Gospels: ‘they themselves do not tell us who their  authors were.’²⁰ Or as some more crudely put it, the Gospels are not ‘signed’.²¹ (2)  

                                                        ¹¹ Respectively, M. Dibelius, Studies in the Acts of the Apostles (London: SCM, 1956), p. 148, and J.  Moles, ‘Luke’s Preface: The Greek Decree, Classical Historiography and Christian Redefinitions’, NTS  57 (2011), pp. 461-482 (467). ¹² R.J. Bauckham, ‘Papias and Polycrates on the Origin of the Fourth Gospel’, JTS 44 (1993), pp. 24-69  (25). This essay is also reprinted, largely unchanged, in idem, The Testimony of the Beloved Disciple:  Narrative, History, and Theology in the Gospel of John (Grand Rapids: Baker, 2007), pp. 33-72.  References to the essay henceforth will follow the pagination of the JTS article. ¹³ Bauckham, ‘Papias and Polycrates’, p. 65. ¹⁴ See M. Hengel, The Four Gospels and the One Gospel of Jesus Christ (London: SCM, 2000), p. 50:  they were ‘not secondary additions but part of the Gospels as originally circulated’. That this refers to  the individual Gospels (not the collection) is clear in his statements about Matthew (p. 77; cf. p. 97)  and Luke (102-103). See also the questioning of anonymity in ‘Die Evangelienüberschriften’ (orig.  1984), in Hengel, Jesus und die Evangelien: Kleine Schriften V (Tübingen: Mohr, 2007), pp. 526-567  (542-543), and 565 on the origins of the titles at the point of the final editing and earliest circulation of  the Gospels. 

¹⁵ B.D. Ehrman, Jesus: Apocalyptic Prophet of the New Millennium (Oxford: Oxford University Press,  1999), p. 42.  

¹⁶ M.A. Tolbert, Sowing the Gospel: Mark’s World in Literary-Historical Perspective (Minneapolis:  Fortress, 1989), p. 27. ¹⁷ D. Bock & D. Wallace, Dethroning Jesus (Nashville: Thomas Nelson, 2007), p. 127. ¹⁸ A.F.J. Klijn, ‘Die Entstehungsgeschichte des Neuen Testaments’, ANRW II 26.1 (Berlin/ New York:  De Gruyter, 1992), pp. 64-97 (79). ¹⁹ W. Reinbold, Der Prozess Jesu (Göttingen: Vandenhoeck & Ruprecht, 2006), p. 41. ²⁰ D. Senior, P.J. Achtemeier & R.J. Karris, Invitation to the Gospels (New York/Mahwah: Paulist,  2002), 328. Cf. F.B. Watson, The Fourfold Gospel (Grand Rapids: Baker, 2016), p. 61, on Luke: ‘The 

5 

Added to this lack of self-reference is a claim about the εὐαγγέλιον κατά X titles: initially there was what Watson refers to as ‘the anonymizing tendency of the earliest  tradition’;²² after that ‘the present titles probably were not added until sometime in the  second century,’²³ or even ‘late in the second century’.²⁴ They were added when the  Gospels began to circulate together and ‘express the traditions that were current at  that time.’²⁵ Freed rightly assesses the majority view, indeed near-consensus, among  scholars: ‘Most NT scholars agree that the gospels are anonymous and that the present  titles probably were not added until sometime in the second century.’²⁶ 

1.4 The approach in this article 

We are therefore left with two seemingly indisputable facts about the Gospels: (1)  ‘they themselves do not tell us who their authors were’, and (2) ‘the present titles  probably were not added until sometime in the second century.’ The first is  indisputable, and although Hengel has made a fascinating case against (2), he has not  really persuaded many others. Rather than disputing these two facts, then, what I wish  to dispute in the rest of this article is their significance, and to argue that they are in  fact entirely irrelevant to the question of the Gospels’ anonymity. 

2. The Significance of the Absence of the Author’s Name 

The first issue to address, then, is the significance of there being no named author in  the prologue or epilogue to the text (‘fact 1’ noted just above). Rather than looking at  the entirety of antecedent and contemporaneous literature from the Graeco-Roman  world and the Near East, we can narrow the scope by sketching how names were  employed in three of the main candidates in NT scholarship for the genre of the  Gospels. Some have claimed that the Gospels are sui generis, but if this is the case  then one cannot have any assumptions one way or the other. We will explore here the  conventions about the presence or absence of the author’s name in works which can  be classified as technical treatise, history or biography. 

2.1 Technical Treatise 

We can treat this case briefly as it has only been applied to Luke. Alexander, in her  arguments for the prologue of Luke’s Gospel as a ‘scientific’ (in the sense of  wissenschaftlich) preface mentions the absence of the name as a standard (non-) 

                                                                                                                                                               third of the canonical gospels did not begin life as the work of “Luke,” nor does it identify itself as a  “gospel.”… he does not give his own name.’ 

²¹ G.E. Ladd, New Testament and Criticism (Grand Rapids: Eerdmans, 1967), p. 128; P. Achtemeier,  J.B. Green & M.M. Thompson, Introducing the New Testament: Its Literature and Theology (Grand  Rapids: Eerdmans, 2001), p. 143. 

²² F.B. Watson, Gospel Writing: A Canonical Perspective (Grand Rapids: Eerdmans, 2013), p. 350. ²³ E.D. Freed, The New Testament: A Critical Introduction (Belmont: Wadsworth, 2000), p. 123. ²⁴ M.E. Boring & F.B. Craddock, The People’s New Testament Commentary (Louisville: WJK, 2009),  p. 284. 

²⁵ Boring & Craddock, People’s New Testament Commentary, p. 284. ²⁶ Freed, The New Testament, p. 123.

6 

feature of technical handbooks.²⁷ In the appendix to an article on the subject, she gives  in full four examples of scientific prologues: those to Diocles, Letter to Antigonus (iv  BCE), Demetrius’s Formae Epistolicae (i BCE), Hero of Alexandria’s Pneumatica I (i/ii CE), and Galen’s De Typis (ii CE).²⁸ Of these authors, none mentions his own  name except Diocles who begins his treatise with an epistolary address. Absence of  the name from such treatises, then, is unremarkable, and if Luke’s preface belongs in  this tradition, then the absence of a name from it should not arouse comment. 

2.2 History 

As we saw in our survey of the specialist studies above, Baum particularly maintained that histories written in Greek included their authors’ names, and so the absence of  such names from the Gospels is a startling datum. The fountain-heads of Greek  history-writing, Hecataeus of Miletus (vi-v BCE), Herodotus (v BCE) and  Thucydides (v-iv BCE) announce themselves as authors in the prefaces to their  works.²⁹ Thereafter, there is a stronger preference among Greek historians for  including names, than in, say, technical treatises. Baum’s statement that Greek  historians generally did this, however, stands in need of correction. There are three  weaknesses in Baum’s treatment of Greek and Roman literature that need to be noted.  

First, there is a conflation in Baum’s discussion of (a) the inclusion of the name  within the work and (b) publication under one’s own name. Baum’s statement that  Jason of Cyrene and Justus of Tiberias (about whose works we know almost nothing)  along with Eupolemus, Artapanus and other Jewish historians did not write their  works anonymously is probably true, but confusing following a statement about  Josephus’s inclusion of his actual name in a preface which is part of his Jewish War. We do not know if those other Jewish historians included their names in their works. 

The second, more substantive difficulty concerns his assertion: ‘At the beginning or  end of his prologue, the Greek historian would mention his name and his  provenance.’³⁰ Leaving aside the exaggeration about the frequency of provenance, the  statement is also an over-generalisation about the presence of the name.³¹ There are  plenty of counter-examples, instances of which are by no means obscure figures.  There is a certain degree of unevenness in Baum’s account: he remarks that Xenophon (iv BCE) used a pseudonym for the Anabasis, but Xenophon (if indeed he  did publish the Anabasis under another name) did not include this name as the author  

                                                        ²⁷ L. Alexander, The Preface to Luke’s Gospel: Literary Convention and Social Context in Luke 1.1-4  & Acts 1.1 (Cambridge: CUP, 1993), pp. 98-99. ²⁸ L. Alexander, ‘Luke’s Preface in the Context of Greek Preface-Writing’, NovT 28 (1986), pp. 48-74  (72-73). ²⁹ Hecataeus, FGrHist 1 F 1; Herodotus, Hist. 1, Preface; Thucy. Hist. 1.1.1. For later examples with  the name included, see D. Fehling, ‘Zur Funktion und Formgeschichte des Proömiums in der älteren  

griechischen Prosa’, in K. Vourveris & A. Skiadas (eds.), ΔΩΡΗΜΑ: Hans Diller zum 70.  Geburtstag. Dauer und Überleben des antiken Geistes (Athens: Griechische Humanistische  Gesellschaft, 1975), pp. 61-75 (65); J. Marincola, Authority and Tradition in Ancient Historiography  (Cambridge: CUP, 1997), pp. 271-275. ³⁰ Baum, ‘The Anonymity of the New Testament History Books’, p. 125. ³¹ Also regarded too much as the norm in E. Herkommer, Die Topoi in den Proömien der romischen  Geschichtswerke (Diss. Tubingen, 1968), pp. 46-52.

7 

within the text, and it should be added that his other works are unnamed.³² Baum  notes Josephus’s self-reference in the preface to his Jewish War, but not the absence  of his name from Antiquities of the Jews. Nor are there prefatory self-references in  such notable Greek historians, roughly contemporaneous with Luke, as Polybius (ii  BCE) and Diodorus Siculus (i BCE).³³ To these we can add Arrian (late i–ii CE),  though admittedly in his case the absence is ostentatiously self-effacing.³⁴ Therefore  although one could talk of a possible tendency to include the name as an element of ‘the prologue-form in ancient historiography’, the title of Earl’s article on the subject,  he is undoubtedly correct that ‘minor variations such as the position or even the  inclusion of the author’s name were allowable’.³⁵ 

Thirdly, although Baum’s treatment of the name appears in a section on Greco Roman historiography, he mentions no histories in Latin, at least of a non biographical character. Among Roman historians, one can find major figures failing  to include their names, including Sallust (i BCE),³⁶ Livy (i BCE – i CE), Tacitus (i-ii  CE) and Florus (i-ii CE).³⁷ Indeed, Herkommer notes that this is the norm.³⁸ 

2.3 Biography 

In some respects biography is a sub-genre of historiography, but it is treated  separately here because of the different ways in which the sphragis may or may not be employed.³⁹ Moreover, it is probably true that the most popular view of the  Gospels’ genre (though not a consensus position) is that they most closely  approximate to bioi. In this connection it is important to note that Philo omits his  name from his biographically oriented works on Abraham, Joseph and Moses.  Josephus’s autobiography does not contain a preface mentioning the subject’s name,  

                                                        ³² That Xenophon did publish the Anabasis pseudonymously (a point first made by Plutarch) is  questioned by e.g. M. Flower, Xenophon’s Anabasis, Or The Expedition of Cyrus (Oxford: Oxford  University Press, 2012), pp. 53-54. Further, C. Pelling, ‘Xenophon’s and Caesar’s third-person  narratives—or are they?’, in A. Marmodoro & J. Hill, eds. The Author’s Voice in Classical and Late  Antiquity (Oxford: Oxford University Press, 2013), pp. 39-73 (41 n. 8): ‘Unlike Herodotus and  Thucydides, Xenophon does not name himself at the beginning of his text in Hellenica or in Anabasis:  might this suggest a shift in paratextual conventions, with the author’s name now stated along with the  title before the text?’ 

³³ D. Earl, ‘Prologue-Form in Ancient Historiography’, ANRW I 2 (Berlin/ New York: De Gruyter,  1972), pp. 842-56 (843), notes Diodorus and Polybius. S.A. Adams, ‘Luke’s Preface and its  Relationship to Greek Historiography: A Response to Loveday Alexander’, JGRChJ 3 (2006), pp. 177- 191, 181, also notes Dionysius, but see Ant. Rom. 1.8.4. ³⁴ Moles, ‘Luke’s Preface’, 467. See further S. Swain, ‘The Hiding Author: Context and Implication’,  in A. Kahane & A. Laird (eds.), A Companion to the Prologue of Apuleius’ Metamorphoses (Oxford:  Oxford University Press, 2001), pp. 55-63 (56-58). ³⁵ Earl, ‘Prologue-Form in Ancient Historiography’, p. 843. ³⁶ Moles, ‘Luke’s Preface’, p. 467, noting Livy and Sallust. ³⁷ Earl, ‘Prologue Form in Ancient Historiography’, p. 843. ³⁸ Herkommer, Topoi in den Proömien, 48-49 notes that unlike some Greek historians, Roman  historians did not usually include their names in their works. Similarly, Alexander, Preface to Luke’s  Gospel, p. 27. ³⁹ In Classical scholarship, the author’s reference to his name in the text is often called, following  Hesiod, the σφράγις, or ‘seal’. The classic study is W. Kranz, ‘SPHRAGIS: Ichform und Namensiegel  als Eingangs- und Schlußmotiv antiker Dichtung’, Rheinisches Museum für Philologie NF 104.1  (1961), pp. 3-46. See now I. Peirano, ‘“Sealing” the book: the sphragis as paratext’, in L. Jansen (ed.), The Roman Paratext (Cambridge: Cambridge University Press, 2014), pp. 224-242.

8 

so if absence of such a self-reference were a criterion for anonymity we would be left  with the absurd result that Josephus’s Vita was anonymous. Among well-known  biographers Plutarch (i-ii CE) makes no mention of his name in his Parallel Lives. Of  the biographical writings of Lucian (ii CE), his Passing of Peregrinus has his name,  but as part of an epistolary prescript (‘Lucian to Cronus, with best wishes’);  otherwise, Alexander the False Prophet, the Toxaris and the biography of Demonax  have no mentions of Lucian’s name in a preface. In the Alexander, Lucian’s name  appears towards the end merely because he is a participant in the drama (Alex. 55),  and similarly Porphyry’s Life of Plotinus (iii CE) only features its author when  Porphyry’s relationship with his subject intrudes into the narrative. Accordingly, his Life of Pythagoras does not mention Porphyry by name. The Life of Apollonius by  Philostratus (ii-iii CE) and two works entitled Lives of the Sophists, one by him and  another by Eunapius (iv-v CE), do not. Similarly, on the Roman side, Nepos (i BCE)  makes no mention of his name in his preface. The Agricola (late i CE) makes no  mention of Tacitus himself, although the mention that the author is the subject’s son in-law means that the writer’s identity is not in doubt (Agr. 3). The opening pages of  Suetonius’ Lives of the Caesars (ii CE) are lost, and with them any possibility of our  knowing whether he mentioned his name.⁴⁰ Lucian’s Passing of Peregrinus is one of only two examples of a biographical work which I have found with a prefatory  mention of the author, and this in a very different form from the self-descriptions of  Herodotus and Thucydides. The other instance is a fictional one, in the biography of  Aelius in the Historia Augusta. This second biography in the sequence uniquely but  spuriously begins: ‘To Diocletian Augustus, his Aelius Spartianus, greetings’. 

2.4 Assessing the Significance of the Absence of the Name 

This article is not the place for arguing a particular case for the genre of the Gospels,  hence the spread of treatments here. What emerges is that the absence of the name  from a technical prologue (if that is what Luke’s is) would be entirely unremarkable. If belonging to the historiographical genre, the Gospels might be regarded as rather  coy examples, but are certainly within the acceptable spectrum of possibilities. Finally, if – as most maintain – they are closest to being bioi, the absence of the  evangelists’ names should excite no comment at all. Such an absence is not remotely  a ‘curious feature’.⁴¹ 

The absence of a name within the body of an ancient work is entirely understandable  because of all the other ways in which the author might be identified. There were of  course numerous ways of indicating an author’s name in or on a roll or codex, outside  of the work itself. To name a few examples,⁴² the name might appear (i) in a  superscription or initial title above the main body of the work proper,⁴³ (ii) in a list of                                                          

⁴⁰ According to M. Meckler, ‘The Beginning of the “Historia Augusta”’, Historia 45.3 (1996), pp. 364- 375 (365 n. 5), Suetonius’s work is reported by John Lydus as having been dedicated to the praetorian  prefect Septicius, but this of course does not necessitate a reference to Suetonius’s own name. The  works of Marius Maximus (ii-iii CE) are lost in their entirety, although quotations appear in the  Historia Augusta. ⁴¹ Pace Droge, ‘Did “Luke” Write Anonymously?’, p. 495. ⁴² For a study of (i) to (iv) with reference to Gospel manuscripts, see S.J. Gathercole, ‘The Titles of the  Gospels in the Earliest New Testament Manuscripts’, ZNW 104 (2013), p. 33–76. ⁴³ See e.g., with respect to rolls, M. Caroli, Il titolo iniziale nel rotolo librario greco-egizio: Con un  catalogo delle testimonianze iconografiche greche e di area vesuviana (Bari: Casanova, 2007).

9 

the contents of the work (the capitula list) preceding the main body of the text, (iii) in  a running header, (iv) as an end title, in a subscriptio or longer colophon appended  after the end of the work.⁴⁴ Other methods included (v) title pages,⁴⁵ or (vi) the  inscription of the author’s name on the back of the roll, or (vii) a ‘name tag’ inserted  into the roll, in a form rather like the modern book-mark (called variously silluboi or  sittuboi).⁴⁶ Some of these seem intuitively improbable, only coming in later in the  transmission of the Gospels (such as ii) along with (iii) and (v) which are more allied  to the codex form. Moles seems to suggest the last option (vii), i.e. a sittubos: ‘his [sc.  Luke’s] name must have been inscribed on the titulus attached to the physical book.’⁴⁷ We do not of course know if or how any of these were used in the first copies of the  Gospels, but the point here is merely to emphasise the irrelevance of the fact that the  author’s names are absent from the prefaces in the Gospels. As Herkommer noted  years ago, the introduction of titles and authors’ names separate from the work itself  rendered superfluous (even if it did not eliminate) the personal self-introduction.⁴⁸ 

In overall conclusion to this second part, one can at least say that it is a category  mistake to say that a work is anonymous because it does not contain within it the  name of the author. As Simon Swain has commented about Greek and Latin literature  in general: ‘It is perfectly normal for literary works to begin without a reference to  their author. The author’s name should already be known to the reader or hearer from  the usual devices.’⁴⁹ 

3. Were the Titles or Names Later Additions? 

If the first item of evidence for the Gospels’ anonymity is insignificant, what then of  the second challenging datum: ‘the present titles probably were not added until  sometime in the second century’ (the formulation of Freed quoted above, in his  representation of the majority view). This again may well be true. It is not clear that  Mark added ευαγγελιον κατα Μαρκον as a prefix or a suffix to his autographon. This observation, however, is also irrelevant to the question of whether there was  some other indication of authorship (not in the εὐαγγέλιον κατά X form) in the  paratexts to the Gospel compositions. Irrespective of whether the full title was absent,  the author’s name might in any case have been appended somehow. 

There are three ways to address this question of whether it was probable that there  were indications of authorship from the beginning, namely (3.1) by arguments from  practical necessity, and (3.2) by the evidence from the earliest period for names being  assigned to Gospels. A final section (3.3) will offer some further corroborating  evidence. 

                                                        ⁴⁴ See e.g. F. Schironi, ΤΟ ΜΕΓΑ ΒΙΒΛΙΟΝ: Book-ends, End-titles, and Coronides in Papyri with  Hexametric Poetry (Durham, NC: American Society of Papyrologists, 2010). ⁴⁵ See S.J. Gathercole, ‘The Earliest Manuscript Title of Matthew’s Gospel (BnF Suppl. gr. 1120 ii 3 / ?�4)’, NovT 54 (2012), p. 209–235. ⁴⁶ T. Dorandi, ‘Silluboi’, Scrittura e Civiltà 8 (1984), p. 185-199. ⁴⁷ Moles, ‘Luke’s Preface’, p. 467. ⁴⁸ Herkommer, Die Topoi in den Proömien, p. 48. See also the statement by Pelling in n. 32 above. ⁴⁹ Swain, ‘The Hiding Author’, p. 55.

10 

3.1 Practical Necessity 

We can return to Martin Hengel’s three suggested settings requiring titles mentioned  earlier. In two cases, namely the importance of titles for (i) the book trade and (ii)  libraries, it is hard to see the significance for the period at the beginning (when the  Gospels were first distributed) with which we are concerned here. It is not until the  second century that appeal is made to the public availability of Christian books (although it is quite early in the second century).⁵⁰ Similarly, Christian libraries proper  probably began around the same time, although it is likely that some teachers (like the  evangelists) and churches possessed a number of Christian books, and it is easy to  imagine that there they had means of identifying the different volumes.⁵¹ Potentially  more significant is (iii) the setting of Christian worship: Hengel adduces, for example,  the comment at the beginning of Melito’s Peri Pascha that ‘Exodus has been read’  and Luke 4, in which a scroll identifiable as that of Isaiah is handed to Jesus. Even  here, however, a nameless title ‘Gospel’ would be sufficient as long as only one  Gospel was in use – just as ‘Exodus’ was sufficient. 

Perhaps more important is the common-sense speculation that when Gospels left their original contexts and were read elsewhere, it is hard to imagine at least some hearers  not thinking ‘Says who?’ Hengel’s point about anonymous writings potentially  inviting suspicion is relevant here. Even before the presence of multiple Gospels in  one congregation, Christians might well have wanted to know where a εὐαγγέλιον came from. Hence it is not necessarily the case that names must only have become a  concern at the earliest when two or more Gospels were gathered together, as Wolter  avers.⁵² 

What has not, to my knowledge, been considered in the literature on this topic is the  relevance of this point to the use of Mark by the authors of Matthew’s and Luke’s  Gospels.⁵³ One could, conceivably, argue that both these later evangelists were struck  by the ‘ring of truth’ in Mark’s Gospel, and so for that reason followed it closely.  More probably, Matthew and Luke, for all their sense that Mark needed adaptation  and expansion, followed Mark so closely in order and content because they had  received reports about it which inspired some confidence.⁵⁴ Minimally, this is likely to  have been a name. It is one thing to assume that Christians in a first-century  congregation would simply have taken on trust what they heard read out, without  asking the ‘says who?’ question. It is a considerable step beyond that – and beyond  what I at least can believe – to suppose that a professedly conscientious investigator  like Luke would, without any accompanying testimony, follow (by ancient standards,  very closely) a naked anonymous account. It seems far more likely that Luke and  Matthew would have received, alongside Mark’s Gospel, some statement – oral or  paratextual – about what and whose it was. The first readers and audience of Mark’s  Gospel would have known who wrote it, and the distance in time between Mark and  

                                                        ⁵⁰ Aristides, Apol. 15-16 presuppose the availability of the Gospel(s) and Christian writings to the  emperor Hadrian around 125 CE. 

⁵¹ See further M. Hengel, Studies in the Gospel of Mark (London: SCM, 1985), pp. 77-78. ⁵² Wolter, ‘Die anonymen Schriften’, p. 4. ⁵³ I leave aside here views according to which Luke also used Matthew (or vice versa). ⁵⁴ For an illustration of Matthew’s and Luke’s close adherence to Mark’s order, see Mk 8.27— 9.48//Matt. 16.13—18.9//Lk. 9.18-50: a helpful table is presented in M.S. Goodacre, The Synoptic  Problem: A Way through the Maze (London: Continuum, 2001), p. 18.

11 

his Synoptic successors was not very great (Mark is rarely considered to be much  more than about 15 years earlier than Matthew and Luke). As we saw Wolter noting  earlier, it is not the case that the author wished to remain unknown – there is no  evidence for a desire for self-effacing secrecy.⁵⁵ 

We can look at this another way, not from the perspective of Matthew’s or Luke’s use  of Mark, but from the angle of the transportation of Mark’s Gospel. How did Mark  reach Matthew and Luke? Books in the ancient world did not simply float around.  They were copied by trained scribes, and would have been carried transprovincially  by trusted people. For Matthew in (let us say) Antioch to have got hold of Mark  written in (let us say) Rome, the text must come either directly or indirectly from  Rome to Antioch. It is likely that not only the initial audience and the first readers of  Mark would have known the author’s identity, but also the earliest copyists and  carriers. Again, the distance in time between Mark on the one hand, and Matthew and  Luke was not very great.  

3.2 Early evidence for named evangelists 

As far as I am aware, there is not a clear account collecting together the testimonia to  the four evangelists as Gospel-writers in the first and second centuries. This is a  different kind of exercise from that of offering arguments for a fourfold Gospel  canon, because attestation of content is not sufficient here. Hannah’s discussion of the  fourfold Gospel canon in the Epistula Apostolorum, for example, is based on the  Epistula’s knowledge of some of the contents of the four canonical Gospels.⁵⁶ The  same applies to Stanton’s arguments for Justin.⁵⁷ For the purposes of the arguments  here, by contrast, named testimonia are what are needed, although in some cases the  arguments rely also on indirect evidence for a work’s attestation of a named Gospel  author. Working backwards from the rather arbitrary date of the end of the second  century, it is potentially misleading to present Irenaeus as a novum, because he is  merely one of several second-century writers noting evangelists’s names.⁵⁸ We will  also introduce here two testimonia not regularly noted in these discussions. 

                                                        ⁵⁵ Wolter, ‘Die anonymen Schriften’, p. 15. ⁵⁶ D.D. Hannah, ‘The Four-Gospel “Canon” in the Epistula Apostolorum’, JTS 59 (2008), pp. 598-633. ⁵⁷ G.N. Stanton, Jesus and Gospel (Cambridge: CUP, 2004), pp. 92-105. ⁵⁸ Others have proposed some additional works as containing potential testimonia. See e.g. M. Hengel,  The Johannine Question (London: SCM, 1989), p. 74 and R.J. Bauckham, Jesus and the Eyewitnesses  (Grand Rapids: Eerdmans, 2006), p. 464, on Ep. Ap. 2.1. I am not quite convinced that this is sufficient  evidence, though it is quite plausible that the author would assume an identification. Some e.g. Hengel,  Four Gospels, 21, have argued that Gaius and the “Alogoi” in the second century claimed that the  fourth Gospel was not written by John but by Cerinthus. However, both the dating of Gaius’s floruit to  the second century and the authenticity of his attribution of the Gospel (and not just the Apocalypse) to  Cerinthus are questionable: see S. Manor, Epiphanius’ Alogi and the Johannine Controversy (Leiden:  Brill, 2016), pp. 60-74, on both the date of Gaius’s work (early iii) and the authenticity of the Cerinthus  attribution to the Gospel. See also as a possibility Epiphanius’s Asian source, probably from the late  second or third century: C.E. Hill, The Johannine Corpus in the Early Church (Oxford: Oxford  University Press, 2004), p. 138. Since these are all potential testimonia to John, which are plentiful  elsewhere, they would not greatly affect the picture in this article. Tertullian could also have been  included, but his works mentioning the evangelists lie just outside the time-frame: John is first  mentioned in Praescr. 22.5 (c. 203), and they all appear in Marc. 4.2.2 and thereafter. The Adversus  Marcionem as we have it in its third edition dates to c. 207-208, although its first edition appeared  ‘perhaps as early as 198’, according to Tertullian: Adversus Marcionem, ed. and trans. E. Evans

12 

The first two testimonia we can note which might well derive from the second century  are two manuscripts: 

(i) �66 (late ii-early iii) attests the title of John’s Gospel, ευαγγελιον κατα [ι]ωαννην.⁵⁹ 

(ii) Secondly, a flyleaf amongst the �4 fragments may also date to the late second  century, though it may also be from the third.⁶⁰ It reads ευαγγελιον κ̣ατ̣α μαθ’θαιον. 

The rest of the evidence comes from second century literature, broadly understood to  include letters and other documentary sources. 

(iii) Fairly securely dated right at the end of the second century (and in one case, into  the third) are Clement of Alexandria’s statements of the evangelists as Gospel writers.  The earliest references to authors of the Gospels are Matthew in Stromata I (c. 198), Mark in Quis dives salvetur (c. 203), Luke in Paedagogus (c. 197), and John in  Protrepticus (c. 195).⁶¹ 

(iv) There are several other references specifically to John from around this time, which identify John as the beloved disciple who reclined at Jesus’ side (Jn 13.25;  21.10), and therefore as the author of the Gospel (21.24). One of these has, to my  

knowledge, hitherto been missed. In a scene in the Nag Hammadi Acts of Peter and  the Twelve (second half/ end of the second century⁶²), Jesus gives the apostles a pouch  of medicine, and tells them: ‘Heal all the sick of the city who believe in my name.’  Not being medically trained, the disciples are baffled by this. Peter does not want to  ask Jesus how this could be possible, so he signals to John, who is next to Jesus, and  tells him to ask Jesus about it. John then does so: 

ⲁϥⲣ ϩⲟⲧⲉ ⲛϭⲓ ⲡⲉⲧⲣⲟⲥ [ⲉ]ⲟⲩⲁϩⲙⲉϥ ⲉⲣⲟϥ ⲙⲡⲙⲉϩⲥⲉⲡ [ⲥ]ⲛⲁⲩ· ⲁϥⲕⲓⲙ ⲉⲡⲏ ⲉⲧϩⲓⲧⲟⲩ ⲱϥ ⲉⲧⲉ ⲓⲱϩⲁⲛⲛⲏⲥ ⲡⲉ ϫⲉ ϣⲁϫⲉ ϩⲱⲱⲕ ⲙⲡⲓⲥⲟⲡ· ⲁϥⲟⲩⲱϣⲃ ⲛϭⲓ ⲓⲱϩⲁⲛⲛⲏⲥ ⲡⲉϫⲁϥ ϫⲉ ⲡϫⲟⲉⲓⲥ ⲧⲛⲣ ϩⲟⲧⲉ ϩⲁ ⲧⲉⲕ ⲉϩⲏ ⲉϫⲉ ⲟⲩⲙⲏⲏϣⲉ ⲛϣⲁϫⲉ… (AcPetTwelve 11.1-8) 

Peter was afraid to reply to him again. He motioned to the one who was beside Jesus, which was John:  ‘You speak this time.’ In response, John said, ‘Lord, we are afraid to say many words in your  presence….’ 

The scene seems to recall John 13.22-25, where the disciples are also baffled by what Jesus has said. There too Peter does not ask Jesus himself for an explanation, but  motions to the beloved disciple – who is next to Jesus – to ask him, which the beloved  

                                                                                                                                                               (Oxford: Clarendon, 1972), p. xviii. ⁵⁹ Gathercole, ‘The Titles of the Gospels’, p. 37-38. ⁶⁰ Gathercole, ‘The Titles of the Gospels’, p. 38; see further, idem, ‘Earliest Manuscript Title of  Matthew’s Gospel’, pp. 209–235. ⁶¹ Strom. 1.21.147.5; Quis 5.1; Paed. 2.1.15.2; Protr. 4.59.3. Dates in J. Ferguson, Clement of  Alexandria (New York: Twayne Publishers, 1974), pp. 16-17. See Hill, Johannine Corpus, pp. 121-122  for possible earlier dates. 

⁶² See P. Nagel, Codex apocryphus gnosticus Novi Testamenti (WUNT 326; Tübingen: Mohr, 2014), p.  348 (mid-late second century); M. Scopello, ‘Introduction’, in M. Meyer (ed.), Nag Hammadi  Scriptures (New York: HarperOne, 2007), p. 359 (end ii/ beginning iii); A.L. Molinari, The Acts of  Peter and the Twelve Apostles (NHC 6.1) (Atlanta: Society of Biblical Literature, 2000), pp. 201-233  provides a helpful overview of the various scholarly attempts to date the work, and an argument (which  is by no means decisive) for a post-Decian date. 

13 

disciple then does. Seeing John as the one ‘beside Jesus’, the Acts of Peter and the  Twelve Apostles thereby sees John as the beloved disciple and therefore the author of  the Gospel. 

(v) In the Quartodeciman controversy, Polycrates of Ephesus wrote a letter (190s⁶³) to  Victor of Rome stating that the paschal feast should always be celebrated on the 14th  day. He writes of various local luminaries, buried in Asia, who observed this practice.  One was John, here identified as the beloved disciple who reclined at the Lord’s side,  and therefore also as the author of the fourth Gospel:⁶⁴ 

ἔτι δὲ καὶ Ἰωάννης ὁ ἐπὶ τὸ στῆθος τοῦ κυρίου ἀναπεσών· ὃς ἐγενήθη ἱερεὺς τὸ πέταλον πεφορεκὼς, καὶ μάρτυς καὶ διδάσκαλος· οὗτος ἐν Ἐφέσῳ κεκοίμηται. (Eusebius, EH 3.31.3;  5.24.3-4) 

In addition there was also John, who reclined on the Lord’s chest, and who became a priest wearing the  mitre, and a witness and a teacher. He sleeps at Ephesus. 

(vi) The Muratorian Fragment (end ii⁶⁵) refers to Luke and John by name, but since it  names them as the third and fourth evangelists, it is all but certain that Matthew and  Mark are mentioned in the lost section preceding. 

(vii) Irenaeus (writing c. 174-189 CE⁶⁶) refers several times to Matthew, Mark, Luke  and John (e.g. AH 3.1.1). In Irenaeus we also have clear connections between these  authors and the Gospels which we know under these names because he refers to the  beginning of each Gospel in connection with each name (AH 3.11.8). 

(viii) An indisputable testimonium to John’s authorship of the fourth Gospel appears  in Theophilus of Antioch’s Ad Autolycum (c. 180 CE):⁶⁷ 

ὅθεν διδάσκουσιν ἡμᾶς αἱ ἅγιαι γραφαὶ καὶ πάντες οἱ πνευματοφόροι, ἐξ ὧν Ἰωάννης λέγει·  Ἐν ἀρχῇ ἦν ὁ λόγος, καὶ ὁ λόγος ἦν πρὸς τὸν θεόν. (Autolyc. 2.22) 

Therefore the Holy Scriptures teach us, as do all those inspired by the Spirit, one of whom, John, says,  ‘In the beginning was the Word, and the Word was with God.’ 

(ix) Hill, following Lawlor, notes an interesting passage from a historical epitome  citing Hegesippus (c. 175-180)⁶⁸ which is relevant for our purposes because it names  John as apostle and evangelist in the sense of Gospel-writer:⁶⁹ 

Δομετιανὸς υἱὸς Οὐεσπασιανοῦ πολλὰ κακὰ εἰς τοὺς ἐν τέλει Ῥωμαίους ἐνδειξάμενος τὴν Νέρωνος νικήσας ὠμότητα δεύτερος κατὰ Χριστιανῶν διωγμὸν έποίησεν. καθ᾽ ὃν καὶ τὸν ἀπόστολον καὶ εὐαγγελιστὴν ᾽Ιωάννην ἐν Πάτμῳ περιώρισεν.  

Domitian son of Vespasian displayed many evils against those in office in Rome, and beating Nero in  cruelty he was the second to institute a persecution against Christians. At that time he imprisoned John,  apostle and evangelist, on Patmos…⁷⁰ 

                                                        ⁶³ Bauckham, ‘Papias and Polycrates’, p. 28. ⁶⁴ Bauckham, ‘Papias and Polycrates’, p. 31. ⁶⁵ B.M. Metzger, The Canon of the New Testament: Its Origin, Development, and Significance (Oxford: Oxford University Press, 1997) pp. 191, 194. Arguments for the later date have not proven  convincing. 

⁶⁶ Irenaeus (AH 3.3.3) mentions bishop Eleutherius (174-189 CE) as current, but not his successor  Victor. 

⁶⁷ Theophilus’s chronicle in Autolyc. 3.28 implies composition in 180 CE. ⁶⁸ Hill, Johannine Corpus, p. 88. ⁶⁹ The reference to an evangelist in the New Testament sense of the term (Ac. 21.8; Eph. 4.11; 2 Tim.  4.5) would be redundant after ‘apostle’.

14 

Lawlor argues persuasively that a direct literary relationship between Eusebius and  the epitome is unlikely, and that both are dependent upon Hegesippus.⁷¹ 

(x) In the Preface to the Chronicon Paschale, the Peri Pascha by Claudius  Apollinaris of Hierapolis (c. 175)⁷² is cited: 

Εἰσὶ τοίνυν οἳ δι’ ἄγνοιαν φιλονεικοῦσι περὶ τούτων, συγγνωστὸν πρᾶγμα πεπονθότες· ἄγνοια γὰρ οὐ κατηγορίαν ἀναδέχεται, ἀλλὰ διδαχῆς προσδεῖται· καὶ λέγουσιν ὅτι τῇ ιδʹ τὸ πρόβατον μετὰ τῶν μαθητῶν ἔφαγεν ὁ κύριος, τῇ δὲ μεγάλῃ ἡμέρᾳ τῶν ἀζύμων αὐτὸς ἔπαθεν, καὶ διηγοῦνται Ματθαῖον οὕτω λέγειν ὡς νενοήκασιν· ὅθεν ἀσύμφωνός τε νόμῳ ἡ νόησις αὐτῶν,  καὶ στασιάζειν δοκεῖ κατ’ αὐτοὺς τὰ εὐαγγέλια. (Chronicon Paschale 13-14) 

There are, then, those who out of ignorance stir up disputes about these this, even if what they do is  pardonable. For ignorance does not deserve condemnation but needs instruction. And they say that on  the fourteenth day the Lord ate the lamb with the disciples, and he himself suffered on the great day of  the feast of unleavened bread. They explain that Matthew says this, or so they think. Therefore it is  case both that their opinion disagrees with the Law, and that the Gospels seem to contradict them.⁷³ 

Apollinaris here opposes the view, apparently based on the Synoptic chronology  specifically in Matthew (26.17-19), that Jesus ate the Passover with his disciples on  14th Nisan. Elsewhere he maintains that Jesus died on this date.  

(xi) The Acts of John (c. 150-200⁷⁴), like Polycrates and the Acts of Peter and the  Twelve, shows knowledge of the tradition that the beloved disciple was John (Ac. Jn  89), the son of Zebedee (Ac. Jn 88). 

(xii) Heracleon (c. 150-175⁷⁵) comments that Jn 1.18 is spoken ‘not by the Baptist but  by the disciple’ (Comm. John, fr. 3).⁷⁶ It is not quite as certain as Gunther claims that  Heracleon ‘attributed the Gospel to John, “the disciple of the Lord”’,⁷⁷ but it seems  probable that he avoids calling either the Baptist or the evangelist ‘John’ to avoid  confusion, especially given that ‘John’ is what Heracleon usually calls the Baptist.  

                                                                                                                                                               ⁷⁰ For the text, see J.A. Cramer (ed.), Anecdota Graeca, vol. 2 (Oxford: The University Press, 1835), p.  88 and C. de Boor, Neue Fragmente des Papias, Hegesippus und Pierius in bisher unbekannten  Excerpten aus der Kirchengeschichte des Philippus Sidetes (TU V/ 2; Leipzig: Hinrichs, 1888), p. 169. ⁷¹ See H.J. Lawlor, Eusebiana: Essays on the Ecclesiastical History of Eusebius (Oxford: Clarendon,  1912), pp. 40-56. This whole section is a series of brilliant pieces of detection. For a different view, see  R.J. Bauckham, Jude and the Relatives of Jesus in the Early Church (Edinburgh: T & T Clark, 1990),  p. 97. ⁷² R.M. Grant, Second Century Christianity: A Collection of Fragments (Louisville: WJK, 2003), p.  175, has c. 175. Eusebius (HE 4.19.1—21.1) puts Apollinarius’s floruit in the time of Soter (c. 168- 174), and his composition of an apology to Marcus Aurelius (161-180 CE) is also indicative of his  dates (HE 4.27.1). See further on Apollinaris, R.M. Grant, Greek Apologists of the Second Century  (London: SCM, 1988), pp. 83-91. ⁷³ Tr. mine. On the passage, see further See U. Huttner, ‘Kalender und religiöse Identität: Ostern in  Hierapolis’, ZAC 15.2 (2011), pp. 272-290. ⁷⁴ Acta Iohannis: Textus Alii – Commentarius – Indices, eds. E. Junod & J.-D. Kaestli (CCSA 2;  Turnhout: Brepols, 1983), p. 700, gives a date of 150-200 CE. ⁷⁵ Grant, Second Century Christianity, p. 70, talks of Heracleon being active in the third quarter of the  century. 

⁷⁶ The Fragments of Heracleon, ed. A.E. Brooke (Texts and Studies 1.4; Cambridge: Cambridge  University Press, 1891), p. 55. ⁷⁷ J.J. Gunther, ‘Early Identifications of Authorship of the Johannine Writings’, JEH 31 (1980), pp.  401-427 (425).

15 

(xiii) According to Irenaeus, the exegesis of John 1 by the Valentinian theologian  Ptolemy (c. 150-175) consisted in an exposition of how the Father emanated all things  spermatically: 

Ἔτι τε Ἰωάννην τὸν μαθητὴν τοῦ Κυρίου διδάσκουσι τὴν πρώτην ὀγδοάδα μεμηνυκέναι.  αὐταῖς λέξεσι, λέγοντες οὕτως· Ἰωάννης ὁ μαθητὴς τοῦ Κυρίου βουλόμενος εἰπεῖν τὴν τῶν ὅλων γένεσιν, καθ’ ἣν τὰ πάντα προέβαλεν ὁ Πατὴρ, ἀρχήν τινα ὑποτίθεται τὸ πρῶτον γεννηθὲν ὑπὸ τοῦ Θεοῦ, ὃν δὴ καὶ Υἱὸν Μονογενῆ καὶ Θεὸν κέκληκεν, ἐν ᾧ τὰ πάντα ὁ Πατὴρ προέβαλε σπερματικῶς. (Ptolemy, apud Irenaeus, AH 1.8.5) 

Further, they teach that John, the disciple of the Lord, pointed to the first Ogdoad, saying as follows, in  these words: ‘John, the disciple of the Lord, wanted to describe the origin of all things, that is, how the  Father emanated everything. Therefore he lays down a certain principle, namely that which was first begotten by God, which Being he has called both the only-begotten Son and God. In him, the Father  emanated all things spermatically…’ 

In this passage, Irenaeus is not paraphrasing, but quoting, ‘in these words’ (αὐταῖς λέξεσι). The content of the teaching about the only-begotten Son and God is  sufficient to establish that John’s Gospel is in view. This is confirmed by named  citations of John 1.1-4 and 14: ‘Thus John’, Ptolemy writes, ‘spoke about the first  Ogdoad, the mother of the aeons’.⁷⁸ 

(xiv) Another neglected testimonium is the probable allusion to Matthew the  evangelist in the Gospel of Thomas (c. 140-180):⁷⁹ 

ⲡⲉϫⲉ ⲓⲥ ⲛⲛⲉϥⲙⲁⲑⲏⲧⲏⲥ ϫⲉ ⲧⲛⲧⲱⲛⲧ⳿ ⲛⲧⲉⲧⲛϫⲟⲟⲥ ⲛⲁⲉⲓ ϫⲉ ⲉⲉⲓⲛⲉ ⲛⲛⲓⲙ ⲡⲉϫⲁϥ ⲛⲁϥ⳿ ⲛϭⲓ ⲥⲓⲙⲱⲛ ⲡⲉⲧⲣⲟⲥ ϫⲉ ⲉⲕⲉⲓⲛⲉ ⲛⲟⲩⲁⲅ⳿ⲅⲉⲗⲟⲥ ⲛⲇⲓⲕⲁⲓⲟⲥ ⲡⲉϫⲁϥ ⲛⲁϥ ⲛϭⲓ ⲙⲁⲑ⳿ⲑⲁⲓⲟⲥ ϫⲉ ⲉⲕⲉⲓⲛⲉ ⲛⲟⲩⲣⲱⲙⲉ ⲙⲫⲓⲗⲟⲥⲟⲫⲟⲥ ⲛⲣⲙⲛϩⲏⲧ⳿ ⲡⲉϫⲁϥ ⲛⲁϥ ⲛϭⲓ ⲑⲱⲙⲁⲥ ϫⲉ ⲡⲥⲁϩ ϩⲟⲗⲱⲥ ⲧⲁⲧⲁⲡⲣⲟ ⲛⲁ⟨ϣ⟩ϣⲁⲡϥ⳿ ⲁⲛ ⲉⲧⲣⲁϫⲟⲟⲥ ϫⲉ ⲉⲕⲉⲓⲛⲉ ⲛⲛⲓⲙ⳿ (Gos.  Thom. 13.1-4) 

Jesus said to his disciples, ‘Compare me and tell me whom I resemble.’ Simon Peter said to him, ‘You  are like a righteous angel.’ Matthew said to him, ‘You are like a wise philosopher.’ Thomas said to  him, ‘Master, my mouth is completely unable to say whom you are like.’ 

The view that this is a reference to Matthew the evangelist now has considerable  scholarly support.⁸⁰ A lengthy case has been made elsewhere,⁸¹ but the key point is that here Matthew seems to be an authoritative spokesman, alongside Peter, who  needs to be rebutted by the Gospel of Thomas. The reference to Matthew here also  appears in a logion heavily influenced by Matthew’s Gospel.⁸² Some have also  suspected that the reference on Matthew’s part to Jesus as a ‘wise philosopher’ might  

                                                        ⁷⁸ In his Epistle to Flora (apud Epiphanius, Pan. 33.3.6), Ptolemy also calls the author of the Gospel an  apostle. 

⁷⁹ It is not mentioned as a testimonium in Hengel, Four Gospels, for instance. ⁸⁰ A.F. Walls, ‘References to the Apostles in the Gospel of Thomas’, NTS 7 (1960–1961), pp. 266–270; E. Haenchen, ‘Literatur zum Thomasevangelium (Fortsetzung)’, ThR 27 (1961), pp. 306–38 (315);  T.V. Smith, Petrine Controversies in Early Christianity (Tübingen: Mohr, 1985), pp. 115-116; R.  Trevijano Etcheverría, ‘Santiago el Justo y Tomás el Mellizo (Evangelio de Tomás, Log 12 y 13)’,  Salmanticensis 39 (1992), pp. 97–119 (112); E.H. Pagels, Beyond Belief: The Secret Gospel of Thomas (New York: Random House, 2003), p. 47 (tentatively); R.J. Bauckham, Jesus and the Eyewitnesses:  The Gospels as Eyewitness Testimony (Grand Rapids: Eerdmans, 2006), pp. 236–237; G.W. Most, Doubting Thomas (London/ Harvard: Harvard University Press, 2005), p. 93; S.J. Gathercole, The  Composition of the Gospel of Thomas (Cambridge: CUP, 2012), pp. 169-174; idem, The Gospel of  Thomas: Introduction, Translation and Commentary (Leiden: Brill, 2014), pp. 259-266; Watson,  Gospel Writing, p. 230. ⁸¹ See e.g. Gathercole, Composition of the Gospel of Thomas, pp. 169–174. ⁸² Gathercole, Composition of the Gospel of Thomas, pp. 174-177.

16 

reflect a perspective on the christology of Matthew.⁸³ Matthew is known for little else  in early Christianity besides being an evangelist, and so there is a high degree of  probability that this dialogue in Thomas presupposes not just a Gospel attributed to Matthew, but one which has attained a certain level of authority. 

(xv) Papias clearly attests to both Mark and Matthew as Gospel writers. Fragments of  his comments are preserved by Eusebius, who only includes a fraction of what Papias  says about these evangelists, as is evident from the Matthean extract, which begins  with οὖν.⁸⁴ 

Μάρκος μὲν ἑρμηνευτὴς Πέτρου γενόμενος, ὅσα ἐμνημόνευσεν, ἀκριβῶς ἔγραψεν, οὐ μέντοι τάξει, τὰ ὑπὸ τοῦ κυρίου ἢ λεχθέντα ἢ πραχθέντα… περὶ δὲ τοῦ Ματθαίου ταῦτ’ εἴρηται·  Ματθαῖος μὲν οὖν Ἑβραΐδι διαλέκτῳ τὰ λόγια συνετάξατο, ἡρμήνευσεν δ’ αὐτά, ὡς ἦν δυνατὸς ἕκαστος. (EH 3.39.15-16) 

‘Mark was Peter’s interpreter, and whatever he remembered, he accurately wrote down the things said  and done by the Lord, but not in order…’. About Matthew this is what is said: ‘Matthew therefore  arranged the oracles in the Hebrew language, but each person interpreted them according to his lights.’ 

Some have argued further for Papias’s knowledge of John as an author.⁸⁵ There is a  good probability of this, since Papias’s fragments seem to display dependence upon  John: the order in Papias’s list of disciples is striking in this regard, following the  orders in John 1 and 21. Furthermore, the implication of disorder in Mark’s Gospel presupposes knowledge of a Gospel with a different ordering of events, and the list of  disciples suggests Papias’s preference for Johannine sequence. This may be  confirmed by the Muratorian Fragment’s remark that John wrote down Jesus’ deeds  ‘in their order’ (Mur. Fr. 33), a point probably dependent upon Papias.⁸⁶ The  similarity of the accounts of John’s Gospel-writing in the Muratorian Fragment and  Clement’s account of the ancient elders (which may also be dependent on Papias)  suggests the existence of an explanation by the bishop of Hierapolis of the  circumstances of the composition of John.⁸⁷ There are two more clear pieces of  evidence. One garbled fragment of Papias, despite its howlers on many points,  evidently makes reference to John: ‘The Gospel of John was made known and given to the churches by John while still in the body… as one called Papias of Hierapolis, a  disciple dear to John, reports in his five <exegetical> books.’⁸⁸ Another fragment  quite independently states that one of the five books discussed John’s Gospel, with  John’s Gospel containing the pericope adulterae.⁸⁹ Some also argue for Papias having  discussed Luke, though this is more uncertain.⁹⁰ 

                                                        ⁸³ Bauckham, Jesus and the Eyewitnesses, pp. 236–237. ⁸⁴ Bauckham, ‘Papias and Polycrates’, p. 46. ⁸⁵ See C.E. Hill, ‘“The Orthodox Gospel”: The Reception of John in the Great Church Prior to  Irenaeus’, in T. Rasimus (ed.), The Legacy of John: Second-Century Reception of the Fourth Gospel (Leiden: Brill, 2010), pp. 233-300 (286), for a list of other scholars. ⁸⁶ Bauckham, ‘Papias and Polycrates’, pp. 53-56. ⁸⁷ Clement, citing the ancient elders in the Hypotyposes, apud Eusebius, EH 6.14.7. It is possible,  however, that Clement is writing independently of Papias, in which case his testimonium would be  independent. 

⁸⁸ Holmes, fr. 19: The Apostolic Fathers, ed. and trans. M.W. Holmes, 3rd edn. (Grand Rapids: Baker,  2007), p. 756. ⁸⁹ Holmes, fr. 23: see The Apostolic Fathers, pp. 760-61. ⁹⁰ See C.E. Hill, ‘What Papias said about John (and Luke): A “New” Papian Fragment’, JTS 49 (1998),  pp. 582-629 (625-629), and the criticisms in Bauckham, Jesus and the Eyewitnesses, pp. 433-437.

17 

The question arises of Papias’s date.⁹¹ According to Eusebius, he is a contemporary of  Polycarp and Ignatius – he appears sandwiched between them (HE 3.36.1),⁹² and in  another list he appears after Clement, Ignatius and Polycarp (3.39.1): the list Papias— Polycarp—Ignatius also appears in Jerome’s version of Eusebius’s Chronicon, placed  at the end of the 219th Olympiad, i.e. around 100 CE.⁹³ Irenaeus is quoted as calling  Papias and Polycarp ἑταῖροι (3.39.1). Eusebius also records Irenaeus’s statement that  Papias was an ἀρχαῖος ἀνήρ (3.39.2), and notes that he overlapped with Philip’s  prophetic daughters (3.39.9).⁹⁴ Eusebius disputes Irenaeus’s claim that Papias was a  hearer of John (3.39.1-2), and the passage he cites seems to support Eusebius’s point. Even so, Aristion and John the elder managed to be both disciples of Jesus and  contemporaries of Papias (3.39.4): in contrast to finding out from the elders what each  of the the other disciples had said (εἶπεν), he found out from the elders what Aristion  and John were saying (λέγουσιν) in Papias’s own time. Papias appears in a section of  Eusebius (HE 3.32.1 to the end of book 3) devoted to the first dozen years of Trajan’s  principate (98-109 CE); book 4 begins in ‘around the twelfth year’ (109 CE). The  natural conclusion from this is to place Papias’s activity around 100, or very early in  the second century, and his investigations may have taken place a decade or two  earlier. 

(xvi) Eusebius prefaces the quotation about Mark’s authorship in EH 3.39.15 with the  words: καὶ τοῦθ’ ὁ πρεσβύτερος ἔλεγεν (‘And this is what the elder said…’). This  is evidently a reference to John the elder, noted in one of Eusebius’s earlier fragments  of Papias. It is clear that this John is noted as the source of the comments about Mark.  What is less obvious is from where the remarks about Matthew come, because Papias  

begins in medias res with an οὖν, and the same goes for Papias’s statements about  John’s Gospel. It remains a possibility that both the Markan and the Matthean  testimonia go all the way back to the elder. It may even be the most likely conclusion,  given that Papias would presumably have wanted to get such information from  someone in touch with the elder’s ‘living voice’. Notably, Papias writes that the elder  ἔλεγεν; then Eusebius writes that this is what is recorded (ἱστόρειται) by Papias  about Mark; then introducing the statement about Matthew Papias reverts to εἴρηται,  which may well imply the elder as the speaker again. In any case, John the elder pushes back the Markan testimonium back well into the first century. 

Taken all in all, then, the first two centuries CE are populated by a good deal more  references to Gospel authors than is commonly appreciated. The most common  attestation is to John and Matthew in the top two, with Mark and Luke in the lower  

                                                        ⁹¹ The best account here, with fairly devastating criticisms of a late date for Papias and compelling  reasons for an early date, is that in R. Yarbrough, ‘The Date of Papias: A Reassessment’, JETS 26  (1983), pp. 181-191. ⁹² Appended to this list later is Quadratus, who as we know from Eusebius’s comment on his Apology,  flourished in the time of Hadrian (EH 4.3.1-2). ⁹³ See Eusebius Werke VII/1. Hieronymi chronicon, ed. R. Helm (GCS; Leipzig: Hinrichs, 1913), pp.  193-194; see further on this Yarbrough, ‘The Date of Papias’, p. 186. ⁹⁴ Yarbrough notes further: ‘The force of this appellation for Papias is strengthened when it is  remembered that Irenaeus referred to John as seeing the Apocalypse “no very long time ago but almost  in our own day, towards the end of Domitian’s reign”’ (‘The Date of Papias’, p. 187, referring to AH  5.30.3).

18 

tier, facts which roughly dovetail with the evidence from the papyri as well as with  the relative frequency of biblical quotations from the period.⁹⁵ 

Date	Source	Matt.	Mark	Luke	John
late ii–early iii	�66				✓
late ii–early iii	�4 flyleaf	✓
c. 200	Clement of Alexandria	✓	✓	✓	✓
c. 200	Act of Peter and the Twelve				✓
190s	Polycrates				✓
late ii	Muratorian Fragment	[✓]	[✓]	✓	✓
174-189	Irenaeus, Against Heresies	✓	✓	✓	✓
c. 180	Theophilus of Antioch				✓
c. 175-180	Hegesippus				✓?
c. 175	Apollinaris	✓
c. 150-200	Acts of John				✓
c. 150-175	Heracleon				✓?
c. 150-175	Ptolemy				✓
c. 140-180	Gos. Thom. 13	✓
c. 100/ early ii	Papias	✓	✓		✓
late i	(John) the Elder		✓

3.3 Two Significant Silences 

Two silences constitute important confirmation of a negative kind. First, there is  never any sense among second-century authors that the Gospels are anonymous. Mention by second-century authors of ‘the Gospel’ tout simple does not imply  anonymity or lack of knowledge of author.⁹⁶ (I suspect that the frequent references to  ‘the Gospel’ in the Apostolic Fathers have exercised undue influence here.) Talk of  ‘the Gospel’ goes side by side with references to evangelists widely in second-century Christianity. Justin uses both the singular and the plural (as well as implying that he  

                                                        ⁹⁵ Hurtado lists the number of second-third century manuscripts for each Gospel as ‘Matthew (12),  Mark (1), Luke (7), John (16).’ See L.W. Hurtado, The Earliest Christian Artifacts: Manuscripts and  Christian Origins (Grand Rapids: Eerdmans, 2006), p. 20. The listing of patristic references to the  Gospels in the first volume of Biblia Patristica, which essentially covers the second century, has for  each of the evangelists: Matthew – c. 70 pages (pp. 223-293); Mark – c. 27 pages (pp. 293-319); Luke  – c. 59 pages (pp. 319-378); John – c. 37 pages (pp. 379-415). See J. Allenbach et al. (eds.), Biblia  Patristica: Index des citations et allusions bibliques dans la littérature patristique. I: Des origins à  Clément d’Alexandrie (Paris: Éditions CNRS, 1975). ⁹⁶ Pace F.B. Watson, Gospel Writing: A Canonical Perspective (Grand Rapids: Eerdmans, 2013), p.  254, who assumes that at the time of 2 Clement ‘the term “Gospel” remains essentially anonymous and  is not yet associated exclusively with the direct or indirect testimony of named apostles’. On the other  side, see M. Hengel, Studies in the Gospel of Mark (London: SCM, 1985), p. 71.

19 

knows some attributions to the evangelists).⁹⁷ Irenaeus can also use the singular alongside the plural.⁹⁸ Polycrates knows John as an evangelist, but also talks of  ‘keeping the passover on the fourteenth day, in accordance with the gospel, never  deviating from it, but following the rule of faith’ (Eusebius, EH 5.24.6), the gospel  here not being a specific one. Theophilus talks about a particular evangelist writing,  but elsewhere can refer simply, without reference to evangelists’ names, to ‘the  Gospels’ or (referring to the contents of Matthew) ‘the Gospel voice’ or ‘the Gospel’ (Autolyc. 3.12-14). The absence of any sense in the second century that the Gospels  are anonymous means that Bauckham’s contrast between Hebrews and John applies  to all the Gospels.⁹⁹ No one ever says, to mutilate Origen’s famous remark about  Hebrews, ‘Who wrote the Gospels, only God knows!’¹⁰⁰ 

The second silence is the absence of any other attributions of authorship assigned to  the Gospels. (Even the idea that Gaius attributed the fourth Gospel to Cerinthus has  been questioned; in any case, if it were true, such an attribution would be a clear  parody of the conventional view.¹⁰¹) If the titles simply emerged very late, say in the  mid- to late-second century, we would expect to find diversity among the names, but  we do not.¹⁰² 

That a substantial time of anonymous transmission would probably lead to diversity  of attributions can be illustrated by the case of Hebrews again. In Hebrews, which is  clearly a letter (Heb. 13.24-25), the absence of the author’s name really is striking. As  a result, it seems that already in the second century, several suggestions of authorship  had been made: Pantaenus had proposed Paul (Eusebius, EH 6.14.4), and a claim to  Pauline authorship is also strongly suggested in �46 (where Hebrews is sandwiched  between Romans and 1 Corinthians); Tertullian states that Barnabas was the author,  without any sense of an alternative view, citing the epistle as ‘more widely accepted’  than Hermas (Pud. 2.1-5) in c. 210 CE;¹⁰³ Origen, according to Eusebius, said that in  addition to Paul, some attributed Hebrews to Clement of Rome and some to Luke the  evangelist (EH 6.25.14).¹⁰⁴ By around the beginning of the third century, then, no less  than four different suggestions for the author of Hebrews had been made. This sort of  diversity is exactly what we do not find in references to the authorship of the Gospels. 

These two points are of course arguments from silence, but the silences are  significance. They fit with the hypothesis that the names are attached to the Gospels  very early, but sit uncomfortably with the line that the names are late. The later that  one views the attributions to be, the harder these silences are to explain. 

                                                        ⁹⁷ Justin, Dial. 103.8 refers to composition by apostles and their followers: this distinction suggests that  he knows particular apostles and followers. Justin also does not name Paul, whose writings he certainly  knows. 

⁹⁸ C.E. Hill, Who Chose the Gospels? Probing the Great Gospel Conspiracy (Oxford: Oxford  University Press, 2010), pp. 179-180. ⁹⁹ Bauckham, ‘Papias and Polycrates’, p. 25. ¹⁰⁰ Cf. Origen apud Eusebius, EH 6.25.14. ¹⁰¹ See n. 58 above. ¹⁰² See e.g. Hengel, Four Gospels, p. 54. ¹⁰³ T.D. Barnes, Tertullian: A Historical and Literary Study (Oxford: Clarendon, 1971), p. 47. ¹⁰⁴ The homilies on Hebrews date to c. 239-242, according to P. Nautin, Origène: sa vie et son oeuvre  (Paris: Beauchesne, 1977), p. 411.

20 

Conclusion 

A good deal of the foregoing argumentation has been negative. This is because the first plank in the positive case for anonymity made by so many, viz. the absence of  the author’s name in the work, is completely insignificant (§2). Even in the case of  histories written in Greek, where the name is perhaps most frequently found in a preface in the work, there is by no means anything approach a rule to this effect. In  consequence, it is a pointless exercise to try to give a theological rationale for why there is no self-reference. 

The second plank in the case for anonymity is, as we have seen, the view that the  present titles were only added later, with the often unstated implication that not only  were the titles absent in their present form but that there were also no accompanying  indications of authorship of any kind at all. The anonymity view not only has the  (insufficiently acknowledged) difficulty of arguing for a negative here, it also has to  contend with all the positive evidence against it. In addition to all the early references  to the evangelists’ names (§3.2), there is the difficulty of imagining Matthew and  (perhaps especially) Luke accepting the second Gospel on trust without  accompanying testimony (§3.1), and also finally the important silences (§3.3). 

To follow up on the individual Gospels, we have noted that attribution of the second  Gospel to Mark goes back to John the elder in the first century. This cannot be more  than about 20 years after the composition of the Gospel. In light of this, it seems  extremely unlikely that there was a time when Mark was not associated with the  Gospel. The testimonia to Matthew are both early and scattered: Papias and the  Gospel of Thomas are notable early witnesses. Again, there are only thirty-odd years  between Matthew and Papias. Luke on the other hand is not attested as an author so early, in fact is not so before Irenaeus and the Muratorian fragment, whichever came first. However, of all the Gospels Luke is perhaps the least likely to be anonymous,  given the first-person references (ἐν ἡμῖν, καθὼς παρέδοσαν ἡμῖν, ἔδοξεν κἀμοὶ παρηκολουθηκότι) and dedication (κράτιστε Θεόφιλε) in the preface to the Gospel  (Lk. 1.1-4) as well as in Acts 1.1 (ἐποιησάμην … ὦ Θεόφιλε) and the “we passages”. John the evangelist has the greatest attestation of all the Gospels in the second century, across the theological spectrum from Valentinians to the Acts of John to the Acts of Peter and the Twelve, to Irenaeus, Theophilus and Polycrates, and reaching as far back in time as Papias. It is the purpose of this article to argue not that these figures actually wrote the Gospels, but that these names are probably original. In the light of these arguments, other hypotheses will have to be ventured besides anonymous publication.

21