Living Footnotes in the Gospel of Luke: Luke’s Reliance on Eyewitness Sources

By Luuk van de Weghe

Did Luke interview eyewitnesses to write his Gospel? Living Footnotes in the Gospel of Luke provides a careful, thorough examination of Luke's claims (Luke 1:1-4), demonstrating that he not only claims to use living sources but also did so. It builds a corroborative evidence case towards this end, not merely by accumulating unrelated strands of evidence, but by showing the interconnectedness of independent lines of subtle clues in Luke's text. These historically rich, unintentional features weave together to generate a robust impression upon the reader: Luke not only relied on living informants but in fact sifted his sources in preference of eyewitness testimony.

• Language: English
• Publisher: Pickwick Publications
• Release date: Aug 29, 2023
• ISBN: 9781666765380

Luuk van de Weghe

Luuk van de Weghe (PhD, University of Aberdeen) is a New Testament scholar whose research has been published in in preeminent peer-reviewed journals in biblical studies, including New Testament Studies, Tyndale Bulletin, and Bulletin for Biblical Research. He is the author of The Historical Tell: Patterns of Eyewitness Testimony in the Gospel of Luke and Acts (2023) and For People Like Us: God's Search for the Lost of Luke 15 (2023). He lives near Seattle, WA with his wife and five daughters.

Book preview

Introduction

Beginning in the early Middle Ages, a robust tradition emerged within the church that Luke the Evangelist was a painter.¹ The hymnographer Andrew of Crete (b. 660 CE) is the earliest extant source to the legend that Luke painted Jesus and his mother Mary, and by the thirteenth century it was claimed that Luke had completed as many as seventy portraits of Jesus and/or the Mother of God.² What made Luke the prime candidate for this legendary persona of God’s portraitist? According to Rebecca Raynor, It was the combination of his faith, his writings, and the observational skills associated with physicians at that time [that is, the time of the Middle Ages] that gave grounds for his elevation as Christ’s artist.³ As we see through our investigation, the medieval church was not completely wrong about Luke. Leaving aside the conjecture of Luke’s traditional vocation, his observational and rhetorical skills clearly informed his portraits of Jesus, albeit it that these, contra the beliefs of the medieval church, were literary in nature and confined within the bounds of the Lukan papyri or manuscripts.

The Argument

The argument of the present book concerns the topic of eyewitness testimony in Luke’s writings. The key contention is that several streams of data within the Gospel of Luke are explained by positing historical, living sources, such as the author of the Fourth Gospel, behind Luke’s literary portraits. Our focus will primarily be on hitherto unexamined lines of evidence. Our interest, secondly, is in explaining these lines of evidence. Often, we will argue that eyewitness testimony provides the best explanation of a given set of data. Moreso, we will argue that these sets of data create a corroborative evidence case in favor of Luke’s use of eyewitness testimony.⁴

This argument is not merely what one would call cumulative, like a rope made up of many strands. It is corroborative. That is, every piece of evidence not merely strengthens the overall case, but often also corroborates—that is, strengthens—another piece of evidence at the same time. It is as if, while adding another strand onto a rope, a piece of thread also gets tangled up with another strand to make it thicker. This corroboration, as we will see, increases as we move through our case; evidences do not merely cumulate but also interlock and strengthen.

Again, we could use a criminal case to illustrate the distinction. Demonstrating that a suspect had both motive and opportunity to commit a crime strengthens the case that the suspect is guilty, but neither element in the case increases the probative strength of the other. Nevertheless, they form part of a cumulative case towards the suspect’s guilt.

Now let us take a case in which a set of independent witnesses testify that they each saw the suspect commit the crime. In the process, they provide incidental details that line up with one another’s testimonies. This set of witnesses also increases the likelihood of the suspect’s guilt, but their testimony is more than cumulative; not only do their testimonies confirm the guilt of the suspect, but their testimonies, in the process, strengthen the reliability of the other independent witnesses by corroborating their accounts.⁵ I will provide some specific examples of how seemingly inconsequential features of Luke’s text accomplish this later in this chapter.

Relevance of the Project

Within the canonical writings, as noted by Dennis Nineham, an explicit focus on testimony—when this relates to one who has seen/experienced Jesus (especially his resurrection)—is relegated to the works of Luke and John, with one exception, throughout the whole New Testament.⁶ Nineham, nevertheless, goes on to assert:

It would seem that for both writers their interest in eye-witness testimony was a specialized, reflective, and apologetic concern, related to their peculiar circumstances and forced upon them, in part at least, by contact with sophisticated inquirers and opponents; and the same is obviously true of similar claims in later writings. It must not too readily be assumed that Luke or John had, or even claimed, direct contact with the eyewitnesses whose existence at an earlier stage they regarded as so vital to their case.⁷

That eyewitness testimony played a secondary, legitimizing role for the theological concerns of Luke is mirrored in the conviction of modern scholarship. In Howard Marshall’s significant contribution, Luke, Historian and Theologian, he devotes but a single chapter to Luke’s role as historian—and this only to usher in a six-chapter contribution to Luke’s role as theologian.⁸ In a 1975 paper titled Luke the Historian, Colin Hemer reflects on the title of Marshall’s book and comments:

The theology of Luke is currently in the forefront of scholarly interest. My purpose here is to offer some remarks on the unfashionable half of a fashionable topic. This may seem an unworthy offering, for surely the whole importance of a book like Acts is in its interpretation and teaching, and certainly not in points of antiquarian detail. Are these not the prerogatives of the popular apologist, who has not read his Haenchen or Conzelmann? I hope to make it clear that my chosen topic is a vast and important one, which we neglect to our peril . . .⁹

This imbalance of interest has not abated since that November of 1975. Are Lukan studies, therefore, in peril? As we discuss shortly, more work could certainly be done on the topic of Luke’s historicity, especially on his proximity and reliance on eyewitness testimony—a topic containing the potential to anchor and to elucidate Luke’s theology rather than to merely legitimize its creativity. This, then, marks the backdrop for this study.

Unity and Uniqueness of the Lukan Writings

Luke claims to have written a narrative of historical events, to have received accounts from eyewitnesses, to have structured his composition with care, and to have followed everything closely from the beginning (Luke 1:1–4)—claims we will consider in-depth in our next chapter. What we find there, however, can only mark the beginning of an inquiry into Luke’s historical interests. Josephus, for example, makes similar claims to Luke but occasionally reveals a terrific penchant for exaggeration. As Hemer points out, Josephus is much more likely to embellish narratives and to inflate figures than Luke is.¹⁰ Speeches in Acts, likewise, are much shorter and show less rhetorical license than speeches among near-contemporary historians such as Plutarch and Dionysius; Josephus, again, as a case in point, places a lengthy speech on the lips of Abraham as he is about to sacrifice Isaac, and this is based only on a few words in Gen 22:8 (cf. Ant. 1.13.3). Luke also resists pairing speeches—the rhetorical creation by an ancient author of a debate-like scenario between two speakers—and he places appropriate language on the lips of speakers at appropriate times (see, for example, the correspondence between Peter’s kerygmatic preaching in Acts, esp. 10:37–41, and the outline of Mark’s Gospel).¹¹

Such fidelity extends to Luke’s treatment of names on the lips of characters. For example, Luke consistently writes Simon Peter’s given name in its Hellenized form throughout his Gospel (i.e., as Σίμων: Luke 4:38; 5:3; 5:4; 5:5; 5:8; 5:10; 6:14; 22:31), yet in Acts 15:14 he appropriately switches to the Semitic form of Simon’s name, Συμεών, when placing it on the lips of Jesus’s brother, James (Is this a sign of Luke’s personal awareness of how James referred to Peter in this instance?). The relevance of these points, however, assumes that Luke and Acts have the same author, a point which should be briefly defended.

Fascinating work can now be achieved to suggest common authorship using stylometrics.¹² Stylometrics is the study of linguistic features, namely those of syntax and vocabulary, which are then used to compare works in search for a common idiolect. Most pertinent is Anthony Kenny’s somewhat neglected monograph, A Stylometric Study of the New Testament. In this work, Kenny applies stylometric analysis to books of the New Testament by cataloguing and comparing ninety-nine stylistic features—usages of particles, conjunctions, noun distributions, etc.—across the various compositions. After analyzing and comparing these features within Luke and Acts he concludes that Luke is closest to Acts in style compared to any other book in the New Testament; furthermore, he writes, Any two texts in Greek will correlate positively with each other at a significantly high level, but it is unusual for correlations to be as high as this.¹³

As Lucian recommended to ancient historians (How to Write History, sect. 55), the author of Luke-Acts creates a literary and thematic overlap between the beginnings of each volume (cf. Luke 1:1–4; Acts 1:1–3) and the adjoining bridge (cf. Luke 24:26–50; Acts 1:1–11).¹⁴ We will survey two more features of the thematic unity of Luke-Acts in chapter 4, but since this unity is conceded by the vast majority of Lukan scholars it is not necessary to press the point further.¹⁵

Regarding the Lukan writings as a whole, we must be cautious to front the content of a work before the claims of its author and likewise to avoid the dangers of what Sandmer famously termed parallelomania: the tendency to perceive similarities and make inferences from comparisons which have no historical basis.¹⁶ Similarity does not equal relationship; because Luke’s claims and interests are similar to those of Josephus or to those of other authors writing in the genre of βίος, does this imply that Luke engages with every proclivity they possess for literary flare or rhetorical embellishment? Of course not. This does not render comparative studies irrelevant, but Luke must be allowed first and foremost to stand or fall upon his own tells.

The concept of a tell marks the theme of a similar work I have written to serve as a popular-level companion piece to this study.¹⁷ Our investigations here also center around this idea of a tell and focus on primary studies involving cues that indicate Luke’s historical interest and care. Our concentration on raw data in most chapters is intentional. It limits presumptions. Our notion of a tell should be distinguished from the archeologist’s tell (or tel), although both similarly indicate riches worth excavating.

Method: The Historical Tell

In their 2003 article, Cues to Deception, Bella DePaulo et al. synthesized the research of 1338 samples of 158 cues to deception during person-to-person dialogues.¹⁸ Following prior studies, their meta-analysis predicted that no single behavior could be found that always occurs when people are lying but which never takes place when people are being honest.¹⁹ Instead, certain cues generally accompany deception more often than they accompany honesty and vice versa.²⁰ The following patterns, according to their study, form cues to honesty and deception:

•Truthtellers are generally more informative and detailed in their accounts.²¹

•Liars’ tales are generally less plausible than truthful accounts and lack the immediacy and personal engagement of truthtellers’ versions.²²

•Liars’ accounts contain less normal imperfections and unusual details, likely resulting from prior rehearsal or from a need to stick to the script.²³

In poker, these types of cues are often referred to as a person’s tell. This is a signal that reveals a player’s bluff and gives away the truth about their hand. Merriam-Webster gives the following entries for tell (noun):

1.An inadvertent behavior or mannerism that betrays a poker player’s true thoughts, intentions, or emotions

The World Series of Poker: earth’s greatest liars gathered together with millions of dollars on the line . . . It’s a blur of action, but the educated spectator ignores these distractions and focuses on the players’ mannerisms—it’s all part of the science of tells, reflexes a player can’t control that, read right, give away his thoughts.

—Seth Stevenson

2.A revealing gesture, expression, etc., that is likened to a poker player’s tell

But his eyes darted fractionally to one side as he said it . . .; the classic liar’s tell.

—Stephen King

broadly: SIGN, INDICATION

I talked to staffers who said that their bosses had two or three flights booked getting out of town, a big tell that there was not going to be a deal.

—Kelly O’Donnell

A tell does not always reveal deception. As Merriam-Webster’s second entry indicates, the term can be neutral. There are two common characteristics of a tell, however. First, a tell is subtle. Second, it generally occurs as a pattern of behavior.

In the present study, we will survey several historical tells within the Gospels and Acts that would have occurred beyond the purview of the author(s). The first cue is labeled onomastic congruence (discussed further in chapter 2): a pattern of name-dropping which would be insignificant but for the presence of prosopographies only recently catalogued and published. Next, in chapter 3, we will look at patterns of vividness in Acts, especially how these concentrate within the we narratives. We will focus particularly on how the narrative realism and local detail of Acts 27–28 conforms with what we might expect to find in memory rehearsal of traumatic events, especially of events that are remembered and recorded with intentionality.

After this point, we investigate a complex pattern that occurs when Luke at key points redacts his Markan template with information that appears to come from an independent source (chapters 5 and 6).²⁴ Of course, this information could only appear to be such and could instead spring from the author’s own creative redactional tendencies but, as we will see, there are reasons to doubt this.

We must set our analysis against Luke’s standard treatment of Mark, which is very conservative. To press the point, Luke treats Mark with more reverence—i.e., he is less likely to alter, improve, or embellish Mark—than Josephus treats the Hebrew Bible.²⁵ Further, as Luke Timothy Johnson notes, although the Evangelist takes over less Markan material than Matthew, he is more conservative when it comes to redacting Markan accounts.²⁶ Second, our pattern must be appreciated against Luke’s general redactional tendencies of Mark. Joseph Fitzmyer highlights six redactional tendencies: first, Luke frequently improves Mark’s Greek; second, Luke generally abbreviates Markan stories by eliminating circumstantial/anecdotal details; third, Luke economizes his text by removing perceived duplicates; fourth, Luke deliberately omits source-material that does not move the reader toward Jerusalem as the story’s climax; fifth, Luke transposes Markan material for literary effect; sixth, Luke eliminates the violent, the passionate, or the emotional.²⁷ Given Luke’s conservative treatment of Mark in general and the nature of his redactional tendencies, we will focus on several key places wherein Luke redacts Mark by doing the opposite of what he typically does: when he adds details, lengthens Markan accounts, and adds emotional content, especially from the perspective of a named character.

Corroboration

Increasingly throughout our study, and especially in chapter 5 and 6, we will see how these tells build upon one another in terms of a convergence of features that collectively corroborate one another and support the probability that an eyewitness source lay behind Luke’s redactions. These features include:

•Semitisms. We will occasionally discuss the phenomenon of Luke’s Semitisms throughout this study. They converge, as we highlight in chapter 4, especially within accounts centered around named individuals.

•Vividness. Vividness is a feature of authentic recollection (see chapter 3), although it can be mimicked in fictional works. Lydia McGrew’s comments are relevant here: "From a probabilistic perspective, the fact that someone engaging in fakery can do something does not mean that the presence of such a feature is not evidence for reportage. If the mere possibility of a logically possible alternative explanation were enough to destroy evidential relevance, rational empirical and historical inference would be impossible. A fictional storyteller includes vivid detail precisely because such details make his story look true."²⁸ This brings us to another insight based on memory studies. We should expect an increased vividness in events that are recounted from the perspective of eyewitness participants, but further, and yet to be highlighted in our investigation, these recollections might contain vivid details that are contrary to the recollections of others. This is because memory remains a subjective guide even for salient events (cf. chapter 3); the tendency toward remarkable reliability for such events—with variability in minor details—will be demonstrated in the recollections of six women concerning the final months of Anne Frank’s life. These will be analyzed in chapter 6.

•Primitivity. Again, the occurrence of details that reflect the primitivity of a given context do not of themselves provide a justification for taking an account as authentic, but as a pattern they reveal a connection to earlier layers of tradition.

•Signs of personal event memories (PEMs). PEMs are memories of events that are meaningful, unique, salient, and generally well-recalled (PEMs form a broader category that includes flashbulb memories (FBMs) and traumatic memories).²⁹ Dale Allison rightly questions to what extent the appearance of PEM features in certain accounts demonstrate that the Gospels are therefore reliable, or even that they contain memories, or to what extent they do.³⁰ Allison is correct to note that because a Gospel account contains an event that would be—if actual—personal, unique, and salient, this is still a step removed from demonstrating that such an event is, in fact, being remembered. Allison quotes Eric Eve regarding Richard Bauckham’s Jesus and the Eyewitnesses: "Bauckham . . . fails to discuss the possibility that fictitious or highly distorted accounts may equally well describe events that appear salient, significant, unusual and emotional. In other words, the fact that a story contains these features does not in itself constitute an argument for its having been based on reliable eyewitness testimony."³¹ Features of PEMs, like vividness in detail, can be fabricated, but unless evidence is brought forth to the contrary, they should be considered as part of a cumulative argument in favor of authenticity.

•Signs of inquiry. We would expect additional materials which result from Luke’s investigation to answer questions that a critical historian might pose of a given event or situation: Why did this happen? Who else was present? What exactly did he/she say? Etc.

We will conduct our investigation into this tell of convergence along three lines. First, chapter 5 will highlight four accounts that reveal a pattern of eyewitness testimony behind Luke’s redaction of Mark. Second, chapter 6 will compare these patterns to those found in the memories of holocaust survivors who witnessed the final months of Anne Frank’s life. This will establish that these similar patterns of variation/confirmation are part and parcel of our general experience of eyewitness recollection. In other words, many features we highlight could individually be accounted for as something other than features of historical remembrance, but their convergence within these accounts cannot otherwise be accounted for so readily.

Thirdly, we bolster this claim by detailing similar historical tells in the works of ancient historians writing about contemporary or recent events. This will place our primary source study into perspective. One might suspect, for example, that written dependence—in our case, the dependence of Luke upon Mark and possibly Q—excludes the legitimacy of positing additional independent sources for the author.³² But even a brief look at historical practice would put this question to rest. From all of Plutarch’s works there is only one scenario wherein he may have relied on only a single source (the biography of Coriolanus), yet even when he used only one written source as his main template it was general practice to supplement it extensively with other written and oral sources and to sift these critically.³³

Before surveying prior research on the topic of eyewitness testimony, it is important to reemphasize the importance of tells and how these fit into our corroborative evidence case. Recall the difference between corroborative evidence versus mere cumulative evidence, as we discussed earlier. Corroborative evidence, while strengthening the main case, also strengthens one or several other pieces of evidence in the case simultaneously.

We will argue, for example, that Luke’s recall of personal names (chapter 2) demonstrates his reliance on eyewitness sources; at the same time, this argument increases Luke’s general reliability (i.e., it is also an argument for authenticity), which then in turn increases our confidence in Luke’s claims to have relied on eyewitnesses (Luke 1:1–4; chapter 1). We will also discuss how Luke’s vividness in recollecting his participation in a shipwreck demonstrates the likelihood that he was a traveling companion of Paul (chapter 3). On one level, this merely proves his proximity to eyewitness sources, but on another level it also proves Luke’s interest in recalling detail accurately. This, then, adds weight to the argument that Luke, whenever he adds additional, even varying, details to an account he borrows from Mark, did so because he likely had access to additional information from an eyewitness source (chapter 5 and 6). When we notice that Luke tends to do this in correspondence with the presence of a named person in the text, it again strengthens our earlier argument about names in the text being indicative of eyewitness sources (chapter 2).

This is just one illustration of how this argument functions. Like an intertangled rope, some strands and strings are stronger than others. Some elements also function more like a tell than others do; that is, some patterns would be more difficult for Luke to craft intentionally than others.

The tells function as part of this corroborative case by supplying independence. Patterns only corroborate one another if they are independent, much like