# The Case for Lukan Priority

The case for Lukan priority is primary comparative and textual, although a strong linguistic case can be made based on Luke’s Hebraic character. The evidence corroborates a broader pattern seen in the Synoptic parallels: Luke repeatedly preserves the simpler, more restrained, less polished, and less sensational form of the tradition, while Mark and Matthew often display revision, expansion, dramatization, clarification, or theological development. The subsections that follow set out the scholarly precedent for Lukan priority, including a substantial concession from a foundational advocate of Markan priority who proposed that Luke’s primary source predates Mark. The primary evidence of Luke not using Mark as a source is addressed including the minor agreements, the broader Markan Cross-Factor of which the minor agreements are one face, the εὐαγγέλιον (*euangelion*; “gospel, good news”) test case showing that Mark’s signature keyword left no trace in Luke’s Gospel, and the body of Markan-Matthean pericopes and sayings without Lukan parallel. These evidences develop the comparative textual case by applying the criteria established above to the Synoptic parallels in primitive wording and comparative restraint, progressive embellishment and redactional expansion. This is compounded by all textual evidence for Matthean posteriority. Finally the confirmatory evidence in the Hebraic syntax of Luke is also addressed along with the statistical validation of the Mark-in-the-middle pattern, and the unreliability of patristic tradition on Gospel origins.

## Lukan Priority in Modern Scholarship

The modern scholarly case for Lukan priority has a longer history than is often recognized. From the late eighteenth century through the nineteenth, advocates of the two-gospel (Griesbach) hypothesis developed a substantial body of arguments for Mark’s posteriority to Luke, including the work of H. U. Meijboom and others cited in the discussion of standard arguments below.[^1] Although these scholars held that Matthew rather than Luke stood first in the synoptic sequence, their detailed case for Mark’s secondary character, which covers Mark’s brevity, redactional style, vivid-imagery argument, and conflational tendencies, is independent of the order between Matthew and Luke and transfers directly to the Lukan-priority position.

Lukan Priority, that Luke is the first of the Synoptic Gospels to have been written, is not a recent innovation. In 1922, William Lockton advanced the theory in his article “The Origin of the Gospels,” published in *The Church Quarterly Review* two years before B.H. Streeter’s *The Four Gospels* (1924) crystallized the Markan-priority consensus in the English-speaking world.”[^2] Robert Lisle Lindsey arrived at substantially the same conclusion independently in the early 1960s, having searched the literature for any predecessor and finding only Lockton’s article. The theory became more commonly known as the Jerusalem School hypothesis, advocated by several scholars from the Jerusalem School of Synoptic Research, which has a history of over 50 years as a group of “Jewish and Christian scholars collaborating in the land and language of Jesus; bringing historical, linguistic and critical expertise to bear on the Synoptic Gospels.” Through their current online journal, *Jerusalem Perspective*, they have contributed much evidence and analysis in support of Lukan Priority.[^3]

While translating Mark into Hebrew, Lindsey found that Mark contains hundreds of non-Semitic expressions, such as the often-repeated phrase “and immediately,” that are absent from Lukan parallels. This suggested that Mark may have been revising Luke rather than Luke revising Mark. Since Lindsey published these findings, David Flusser, David Bivin, Halvor Ronning, Richard Stegner, Brad Young, and other scholars associated with the Jerusalem School have argued strongly for Lukan priority. Lindsey describes how he came to discover the primacy of Luke over Mark and Matthew:

> “I therefore turned to a story-by-story, word-by-word, study of the Synoptic Gospels of Matthew, Mark, and Luke… To my surprise it turned out that Luke’s Gospel contained almost none of the non-Hebraic expressions so common in Mark! On the other hand Matthew, when copying Mark (or if copying Mark), appeared to reject about half the non-Hebraisms of Mark completely, to accept others without question and repeat them in exact Markan contexts, and to reject still others in the earlier chapters of his Gospel only to accept them in later portions…

> “Having long supposed that Luke, as the non-Jewish companion of Paul, tended to modify his text to make it more understandable to Greeks of pagan background, I was even more surprised to note that the Lukan text was almost always easier to translate to idiomatic Hebrew than was Mark. After several more years of study in which this observation has been confirmed again and again, I today find my early supposition amusing, but the point is that I was quite unprepared to suppose that of all the Synoptists, Luke should prove to be the best in preservation of earlier texts...

> “Without quite realizing it and quite without intending it, I thus found myself questioning whether our Mark could, in fact, be the principal narrative source standing equally behind Matthew and Luke in their so-called Markan portions. It looked as if Luke had universally copied more faithfully whatever Greek sources he had and that these had been translated earlier from a Hebrew source or sources, or at least from some Semitic document or documents so much like Hebrew that in retranslation it was impossible to tell the difference. It also looked as if Matthew had indeed used Mark’s Gospel with all its didactic expressions but had rejected many of these for some reason I could not yet explain.”[^4]

It is worth noting that B.H. Streeter, who popularized the view of Markan priority in *The Four Gospels: A Study of Origins* (1924), nonetheless granted substantial independent weight to a pre-canonical Lukan source. Streeter proposed the existence of Proto-Luke, an early gospel-like document independent of Mark and roughly contemporary with it, which he held to be the framework into which extracts from Mark were later inserted to produce canonical Luke. In Streeter’s own framework, Proto-Luke functioned as Luke’s “primary” source and Mark as his “secondary” source: Luke started from Proto-Luke and preferred it to Mark where the two diverged. Streeter further described Proto-Luke as “a kind of half-way house between Q and Mark,” meaning a stage in the evolution from sayings collections to the biographical gospel form Mark inaugurated, and assigned it independent authority comparable to Mark’s, sometimes superior and sometimes inferior in historical value, but “on the whole of approximately equal value” (p. 222). Although Streeter himself maintained Markan priority, the scope of his Proto-Luke concession significantly qualified Mark’s role as the primary narrative authority for the life of Jesus.[^5]

The scope of Streeter’s Proto-Luke concession is more substantial than is often appreciated. On Streeter’s own analysis, the majority of Luke’s content, including large portions of chapters 3 through 18, the passion narrative, and the resurrection accounts, was attributed to Proto-Luke rather than to Markan dependence. Only a handful of Lukan verses were identified as derived from Mark. This means that on the very hypothesis Streeter advanced to preserve Markan priority, Mark’s influence on Luke is minimal, and the bulk of what is characteristic of Luke stands on an independent source prior to Mark. Several later scholars extended this line of argument: Vincent Taylor reinforced the Proto-Luke hypothesis in *Behind the Third Gospel* (1926), and German scholars Joachim Jeremias and Friedrich Rehkopf in the mid-twentieth century also affirmed the existence of a substantial pre-Markan Lukan source.[^6] A detailed treatment of Streeter’s Proto-Luke argument and its development is given in the companion article “Proto-Luke and Lukan Priority.”[^7]

Streeter went further than the question of quantitative scope. He explicitly argued that Proto-Luke was, and was originally intended as, a complete Gospel, rebutting the objection that it was an ‘amorphous’ document insufficient to stand on its own.[^8] On Streeter’s own analysis, then, Mark is not the framework into which Luke fitted special material but a later supplement to a Gospel that was already complete.

Vincent Taylor’s *Behind the Third Gospel* extended Streeter’s hypothesis with detailed verbal-statistical and structural analysis.[^9] Taylor’s most distinctive contribution is what may be called the ‘insertion’ argument: within Luke’s Passion narrative (Lk. 22–24), the Markan elements appear as isolated insertions slotted into a non-Markan framework rather than as the underlying base. Removing the Markan elements yields a narrative with improved internal sequence; the variations of order between Luke 22–24 and Mark are explained as displacements caused by inserting Markan material into a previously assembled document. Taylor concludes that ‘the substance of Lk. xxii.-xxiv. … existed as a document before St. Luke.’ Mark is therefore not the framework into which Luke fitted his special material; Mark is a later supplementary source. Taylor dates this Proto-Lukan source to approximately 60–65 AD, a few years earlier than Mark, as compared to Streeter who dates Proto-Luke to the two years of Paul’s captivity at Caesarea (57–59 AD).[^10] Taylor develops a verbal-statistics methodology (percentage of common words, distribution patterns, order of narratives) that anticipates the quantitative approach Ronning later refined.[^11] Taylor defended the hypothesis in two further publications, in 1943 against C. S. Petrie’s criticisms and in 1955 against the broader critical reception, indicating that the Proto-Luke hypothesis remained a live and defended scholarly position well into the mid-twentieth century rather than a discarded early proposal.[^12]

Other Scholars, who advocate the two-source hypothesis, affirm that Luke more often than Matthew preserves the more primitive reading of the lost source.[^13] The simpler explanation is that Luke (Proto-Luke) is itself the primary source for narrative sayings common to Luke, Mark, and Matthew.

## The Minor Agreements as Evidence for Lukan Priority

There is a pattern in the Gospels that the two-source hypothesis has never fully managed to explain. When Matthew and Luke tell a story that also appears in Mark, they sometimes agree with each other in small details that are not in Mark. They use the same unusual word where Mark uses a common one. They put the same phrase in the same order. They leave out the same detail from Mark. These shared departures from Mark, known in synoptic studies as the minor agreements, number in the hundreds across the triple tradition.[^14]

The difficulty these agreements pose is simple. If Matthew and Luke used Mark independently and never saw each other’s work, their small changes to Mark should not line up. Each author would make his own choices. The shared departures should be scattered and occasional. Instead, they are concentrated and systematic, often involving rare vocabulary or distinctive word order that is unlikely to have arisen twice by accident.

The two-source hypothesis has offered three responses. The first is that the agreements are coincidences: Matthew and Luke happened to make the same changes in the same places. The second is that scribes introduced them later, copying Luke toward Matthew or Matthew toward Luke in the manuscripts we now possess. The third is that Matthew and Luke are drawing on Q in places where Q overlapped with Mark, so the agreements are really shared Q readings. None of these answers has persuaded all parties. The more agreements accumulate, the less coincidence will bear. The manuscript evidence does not support the scale of scribal contamination required. And the third answer extends Q so far beyond its original purpose, which was to account for sayings material not found in Mark, that Q becomes a second source nearly as extensive as Mark itself.[^15]

The Farrer hypothesis is one solution to the minor agreements problem. It postulates that where Matthew changed Mark, Luke simply followed Matthew. This is the strongest single argument for Farrer, since the hypothesis is simpler than the two-source view.[^16] However, the Farrer hypothesis fails on closer inspection, including the editorial fatigue argument as applied to Luke, addressed later in this study in the subsection “Editorial Fatigue Proves Markan Priority.”

Lukan priority handles the agreements differently. If Luke came first, Mark rewrote Luke, and Matthew then rewrote Mark with Luke also in view, then Matthew’s agreements with Luke against Mark are exactly what one would expect. At many small points where Mark had substituted a synonym, added an explanation, or rearranged a phrase, Matthew reverted to the earlier Lukan wording. He did not coincidentally arrive at Luke’s readings; he chose them. No appeal to chance or to lost sources is needed. The pattern is a natural consequence of the sequence.

The Hebrew retroversion evidence confirms this reading. The places where Matthew and Luke agree against Mark tend to be phrases that translate cleanly back into Hebrew or Aramaic, while Mark’s divergent readings contain the Latinisms and the Greek dramatic idioms characteristic of Mark’s own style. Matthew’s agreements with Luke, in other words, are not just agreements with Luke; they are agreements with the Semitic substratum that Luke preserves and that Mark has paraphrased away. Table 2 compares how the five hypotheses handle these two bodies of evidence.

**Table 2.** How the five contemporary synoptic hypotheses handle the minor agreements and the Hebrew retroversion evidence.

| **Synoptic** **Hypothesis** | **Minor Agreements** | **Hebrew Retroversion** |
| --- | --- | --- |
| Two-Source (Mark first, plus Q) | Treated as an anomaly. Explained as coincidence, as overlap between Mark and Q, or as later scribal copying. None of these scales to the observed count. | Cannot account for why Luke, viewed on this hypothesis as a later Gentile composition, preserves the most easily retroverted Semitic syntax of the three. |
| Farrer (Mark first, no Q) | Solved by Luke’s direct use of Matthew. But this direction of dependence is the one the editorial fatigue evidence fails to support. | Cannot account for Mark’s non-Semitic Greek syntax compared to Luke’s. |
| Two-Gospel (Matthew first) | Accommodated by Luke’s use of Matthew and Mark’s conflation of both. Contradicted by the evidence of of progressive embellishment and Matthean Postoriiorty, which runs the opposite way. | Cannot account for why Luke, on this view a middle stage between Matthew and Mark, preserves more primitive Semitic syntax than the Gospel it is supposed to have copied. |
| Matthean Posteriority (Mark → Luke → Matthew) | Explained as Matthew’s partial conflation of Luke against Mark. Matches the directional evidence but unnecessarily gives Mark priority. | Cannot account for why Luke preserves the most easily retroverted Semitic syntax and is more primitive in wording than Mark. |
| Lukan Priority (Lindsey), Luke → Mark → Matthew) | Predicted. Matthew reverts to Lukan readings that were revised and expanding in Mark | Fully aligned. Luke translates directly from a Semitic source; Mark paraphrases using Latinisms and Greek idioms; Matthew partially restores Lukan readings. |

## The Markan Cross-Factor

The minor agreements are one face of a broader structural pattern that Lindsey termed the Markan Cross-Factor.[^17] The name captures an inversion: two variables, pericope order and wording, trade places depending on whether Mark is in the comparison. In the seventy-seven Triple Tradition pericopae (self-contained Gospel passages), where Mark stands alongside Matthew and Luke, Matthew and Luke share order with Mark but diverge sharply in wording, both from Mark and from each other; verbal agreement rarely exceeds half the words, even in episodes describing the same event. In the forty-two Double Tradition pericopae, where Mark is absent, the relationship flips: Matthew and Luke share wording closely, often verbatim or nearly so across whole sentences, but differ widely in narrative placement. Mark’s presence yields order-agreement and wording-divergence; Mark’s absence yields the opposite. All of this is an indication that Mathew is third and he often preferred the structure of Mark while depending on the text of Luke when there is no Markan text to rely on.

The standard two-source hypothesis offers no principled explanation for the asymmetry. If Matthew and Luke independently used Mark and Q, the two sources should produce comparable verbal fidelity, modulated only by each writer’s editorial habits. Instead, Matthew and Luke treat Q with high verbal fidelity and Mark with low verbal fidelity, and they do so in concert. Two independent writers approaching two written sources do not ordinarily converge on systematically divergent treatment of those sources; that is both treating one with high fidelity and both treating the second source (Mark) with lower verbal fidelity. The simpler explanation is a dependency on Luke by the author of Matthew as a primary source where there was no parallel in Mark.

On Lukan priority, the Cross-Factor is exactly what would be expected. Luke wrote first. Mark then reworked Luke, substituting synonyms, intensifying details, and adding redactional embellishment. Matthew used both Luke and Mark, choosing between them at each point. Where Mark was absent (the Double Tradition), Matthew had only Luke to draw on. He transcribed Luke’s wording closely, producing the verbal agreement that characterizes the Double Tradition. The locational divergence reflects Matthew’s freedom to relocate Lukan material within his own narrative arrangement.

Where Mark was present (the Triple Tradition), Matthew faced a choice at each point between Mark’s redacted text and the older Lukan wording behind it. The result was a hybrid Matthean text that often agrees with Luke against Mark in single words and short phrases (the minor-agreements phenomenon) but rarely sustains that agreement across whole sentences, because Matthew was synthesizing rather than transcribing. Pericope order, by contrast, propagates cleanly through the chain: Luke set the order, Mark followed it, and Matthew followed Mark, so all three Synoptists agree on Triple Tradition order even where their wording diverges.

The Cross-Factor and the minor agreements are therefore not two independent observations. They are two scales of the same structural fact: Mark is the disturbing element in the synoptic transmission, having reworked the prior Lukan tradition, and Matthew preserves Lukan wording wherever he chose Luke over Mark.

## The Euangelion (Gospel) Test Case

A particularly sharp lexical case for Lukan independence from Mark concerns the noun εὐαγγέλιον (*euangelion*; “gospel, good news”), which Mark uses to frame his entire Gospel and which Luke entirely omits from his. Mark uses εὐαγγέλιον seven times, opening his Gospel with the phrase “the beginning of the gospel of Jesus Christ” (Mark 1:1) and placing the term in Jesus’s mouth at Mark 1:14–15 in a context where the Lukan parallel material at Luke 4:14–15 makes no use of it. The other Markan occurrences are at Mark 8:35, 10:29, 13:10, and 14:9. Luke’s Gospel, by contrast, contains the noun zero times. The complete absence is striking given Luke’s evident familiarity with the term in Acts, where he uses it twice: once in Peter’s speech at the Jerusalem council (Acts 15:7) and once in Paul’s farewell at Miletus (Acts 20:24).[^18]

The asymmetry between Luke’s Gospel (zero occurrences) and Acts (two occurrences) is methodologically decisive. Luke is plainly willing to use εὐαγγέλιον when his source material warrants it, since he uses the term in reporting the speech of Peter and Paul. His absolute omission of the term from his Gospel cannot therefore be explained as personal stylistic aversion. The natural inference is that the noun was not present in the sources Luke used for his Gospel. If Luke had written with Mark in front of him, the seven Markan occurrences (one of which functions as the opening words of Mark’s entire Gospel) would have produced at least a few Lukan retentions or substitutions in parallel passages. The fact that not one appears in the entire Gospel of Luke is direct evidence that Luke’s Gospel-source pool did not contain Mark.

Matthew’s handling of εὐαγγέλιον corroborates the directional implication of Matthean posteriority. Matthew uses the term four times, but with one exception always in the expanded phrase “the gospel of the kingdom” (Matt 4:23; 9:35; 24:14), as if the bare Markan term required elucidation. This is the editorial signature of an evangelist who inherited the word from a source he found insufficiently clear and added a clarifying genitive. The pattern fits the Luke-to-Mark-to-Matthew sequence: Paul coined the term and used it densely in his epistles, Mark drew on the Pauline corpus and made the noun a structural keyword of his Gospel, Matthew inherited Mark and accepted the term while expanding it for clarity, and Luke, who did not use Mark, never adopted the noun for his Gospel narrative even while using it in Acts where it belonged on the lips of Peter and Paul.

The redactional pattern within Mark itself sharpens the inference. James D. G. Dunn has shown that Mark’s references to “gospel” appear consistently as the evangelist’s own additions to received tradition rather than as inherited material. Where Mark summarizes Jesus’s preaching as “repent and believe in the gospel” (Mark 1:15), Matthew at the parallel point retains only “Repent” (Matt 4:17). In the saying about losing one’s life, only Mark reads “for my sake and the gospel’s” (Mark 8:35); Matthew and Luke at the parallel points read only “for my sake” (Matt 16:25; Luke 9:24). In the promise to those who have left family and possessions, only Mark adds “and for the sake of the gospel” (Mark 10:29); the Matthean and Lukan parallels lack the addition (Matt 19:29; Luke 18:29). Matthew and Luke thus omit Mark’s references to εὐαγγέλιον even where they otherwise stay close to Mark’s wording. The most natural explanation is that Matthew recognized those references as a Markan peculiarity rather than as part of the inherited tradition. Despite Dunn holding to Markan priority, he draws the conclusion reading of the same data that Luke preserved the original tradition without εὐαγγέλιον in Jesus’s mouth, and Mark inserted the term across the relevant sayings.[^19]

The εὐαγγέλιον evidence also extends a parallel line of literary dependence already documented in this study. Mark’s compositional method systematically appropriates vocabulary from outside the Lukan stratum: the Aramaic “*Talitha cum*” command at Mark 5:41 echoes “Tabitha, arise” at Acts 9:40; the rare word κράβαττος (*krabattos*, “pallet”) is shared between Mark 2:4 and Acts 5:15 in healing contexts, absent from the rest of the Synoptic tradition; and the αββα ὁ πατήρ (“Abba, Father”) formula at Mark 14:36 is lifted from Romans 8:15 and Galatians 4:6. Mark’s distinctive deployment of εὐαγγέλιον reflects the same pattern, drawing on Pauline rather than Jesus-tradition vocabulary. Sixty of the seventy-six occurrences of the noun in the New Testament appear in the Pauline corpus, and Paul is plausibly the figure who introduced the noun into Christian usage on the basis of Isaiah 52:7 and 61:1–2. There are no indications that Jesus himself used a Semitic equivalent of the noun, with the possible exception of Mark 14:9, where Mark’s hand is again evident. Mark’s, “the beginning of the gospel of Jesus Christ” (Mark 1:1), goes further still: it marks the transition by which εὐαγγέλιον moves from the content of a message to the genre of a written text, with Mark’s opening line effectively introducing the literary category later called “Gospel.” Both features locate Mark within the Pauline mission’s vocabulary and within a second-generation move to convert apostolic preaching into written text. That fit explains Mark’s posteriority to Paul’s letters as well as to Acts, and it explains why Luke, working closer to the apostolic generation, preserves the more primitive distinction. Luke’s complete omission of the noun from his Gospel narrative, even as he uses it twice in Acts on the lips of Peter and Paul (Acts 15:7; 20:24), is therefore not stylistic aversion but a positive indicator of Lukan primitivity: As a post-Resurrection Pauline term, Mark retrojected what was absent in Luke into the speech of Jesus.[^20]

## Markan-Matthean Agreements Without Lukan Parallel

The εὐαγγέλιον test case identifies a Markan keyword absent from Luke that Matthew adopted. A parallel pattern emerges at the level of whole pericopes and sayings: passages present in both Mark and Matthew that have no equivalent in Luke, or that appear in Luke only in a restrained and undeveloped form. The pattern is significant because Matthew incorporates virtually all of Mark, retaining approximately ninety four percent of Mark’s verses, while Luke’s relationship to Markan content is much sparser and uneven. On the Markan-priority model, the asymmetry requires Luke to have made a series of editorial choices to drop coherent and theologically substantive material from his source, and the conventional explanations for those choices (defective Markan exemplar, scroll-length constraint, deliberate compression) do not converge on a consistent editorial principle. On Lukan priority the same asymmetry is what is to be expected: the material was not in Luke because it had not yet entered the tradition; Mark composed it, and Matthew received it from Mark.

The asymmetry is also quantitative. Mark contains 661 verses, of which approximately 606 are reproduced in Matthew and approximately 350 in Luke. The three Synoptics share a triple-tradition base of approximately 326 verses found in all three. Beyond that shared base, Matthew extends Mark’s content by 280 verses absent from Luke, while Luke exhibits only 24 verses absent from Matthew, a ratio of more than ten to one (Figure 1). The remaining 31 verses of Mark are unique to Mark and absent from both Matthew and Luke. The 280-verse Matthean extension and the 24-verse Lukan extension are the central asymmetry: Matthew’s Markan adoption beyond the triple tradition is more than an order of magnitude larger than Luke’s, and the 280-verse block has substantive narrative and theological coherence rather than being a scatter of small omissions. Forty-five percent of Matthew’s adoption of Mark consists of material absent from Luke, while only seven percent of Markan Material in Luke consists of material absent from Matthew. That block corresponds substantively to the Great Omission, the cursing of the fig tree, the anointing at Bethany, the Eloi cry, the Markan-Matthean expansions of Lukan saying-kernels, and many other Markan-Matthean agreements without Lukan parallel.[^21]

![Stacked bar chart comparing the Markan content reproduced in Matthew (606 verses) and Luke (350 verses), each split into a triple-tradition base of 326 verses shared with the other gospel and the additional Markan material absent from the other gospel: 280 verses for Matthew and 24 verses for Luke.](https://ntcanon.com/assets/figures/figure1.svg)

*Figure 1. Markan Content in Matthew and Luke, Partitioned by Overlap*

The Great Omission is the largest single example. Mark 6:45–8:26, a block of roughly seventy-five verses, is almost entirely absent from Luke and almost entirely paralleled in Matthew. The block contains the walking on the sea (Mark 6:45–52 // Matt 14:22–33), the healings at Gennesaret (Mark 6:53–56 // Matt 14:34–36), the discourse on defilement and the tradition of the elders (Mark 7:1–23 // Matt 15:1–20), the Syrophoenician woman (Mark 7:24–30 // Matt 15:21–28), the feeding of the four thousand (Mark 8:1–10 // Matt 15:32–39), and the extended yeast-warning discussion of the disciples’ incomprehension (Mark 8:14–21 // Matt 16:5–12). Several of these are precisely the kind of dramatic, theologically loaded material that an evangelist working from an earlier source would have strong incentives to preserve. The walking on the sea is one of the most striking Christological displays in the Synoptic tradition. That Luke would omit a Christological theophany of this magnitude if it lay before him in Mark is difficult to defend. That the entire surrounding block reads as a Markan composition unit which Matthew received and Luke did not is the straightforward explanation.

Beyond the Great Omission, three further Markan-Matthean episodes stand out for their narrative weight and theological prominence. The cursing and withering of the fig tree (Mark 11:12–14, 20–25 // Matt 21:18–22) is a public destructive miracle by Jesus that Luke does not record, though Luke has a parabolic fig-tree narrative at Luke 13:6–9 with comparable thematic content and no miracle. The anointing at Bethany (Mark 14:3–9 // Matt 26:6–13), with Jesus’s striking declaration that “wherever the gospel is preached in the whole world, what she has done will be told in memory of her,” is paralleled in Luke only by the differently-set anointing at Luke 7:36–50, where the gospel-preaching saying is absent. The cry from the cross “*Eloi, Eloi, lema sabachthani*” (Mark 15:34 // Matt 27:46) appears at one of the most theologically charged moments in the passion narrative, wherein Luke reads a much subdued, “Father, into your hands I commit my spirit” (Luke 23:46). These are not minor omissions on Luke’s part if Luke is working from Mark. They are the kind of things the author of Luke would have preserved if indeed he had Mark in front of him.

A related pattern appears at the level of individual sayings, where Luke exhibits the core saying in restrained form, Mark expands it with additional material, and Matthew receives the Markan expansion. The millstone saying is representative. Luke 17:1–2 contains the core warning about causing one of the little ones to stumble. Mark 9:42–48 expands it with three pendant warnings about the hand, the foot, and the eye, each glossed by the threat of Gehenna. Matthew accepts the Markan expansion at 18:6–9 and reuses parts of it at 5:29–30. The “if anyone would be first, he must be last of all” amplification at Mark 9:35, accepted by Matthew at 20:26–27 against the simpler Lukan form at Luke 9:48, follows the same directional pattern, as does the faith-and-mountain expansion at Mark 11:22–24, accepted by Matthew at 17:20 and 21:21–22 against the simpler Lukan core at Luke 17:5–6. The directional reading on Markan priority requires Luke to have systematically pared back Markan expansions to their kernels in case after case. The directional reading on Lukan priority requires only that the kernels are what Luke received and that Mark expanded them.

The cumulative force of the pattern is the central point. Matthew’s adoption of nearly all of Mark, paired with the content being sparse in Luke, is an indication that Mark revised and expanded upon a simpler Lukan core according to the Luke-to-Mark-to-Matthew sequence in a way that the reverse sequence does not. The εὐαγγέλιον evidence shows the pattern at the level of one Greek word; the present evidence shows it at the level of pericopes, sayings, and entire literary blocks.

## Primitive Wording and Comparative Restraint

A core argument for Lukan priority rests on repeated comparison of parallel passages. In many Synoptic parallels Luke gives the less polished, less expansive, and less sensationalized wording, while Mark and Matthew add vividness, explanatory detail, literary polish, or theological emphasis. This is not an argument from mere brevity. A shorter reading can be secondary when it omits inconvenient material. The argument is from directional pattern: Luke repeatedly preserves wording that is more restrained, less narratively heightened, and more naturally retroverted into Hebrew or Aramaic.

Direction of dependence best explains the actual sequence of wording, omission, addition, clarification, and theological development across shared episodes. This criterion focuses on the comparison. If Luke repeatedly preserves the more restrained form, if Luke more often carries the Semitic substratum of the tradition, and if Mark and Matthew repeatedly display more developed narrative or theological features, the direction of dependence is more plausibly Luke to Mark to Matthew. These claims should not be reduced to the claim that Luke is more Hebraic. The stronger claim is that Luke’s wording is repeatedly more primitive in direct comparison with Synoptic parallels, and that the Hebraic character of Luke helps explain why this primitive wording appears in the first place. The primitive wording and the linguistic substratum reinforce one another.

The pattern can be seen directly in parallel sayings of Jesus, where Luke repeatedly preserves the simpler, less developed form. Table 3 shows numerous quotations of Jesus that are more rudimentary in Luke and improved in Mark.

**Table 3.** Jesus Sayings Exhibiting a More Primitive Form in Luke than in Mark

| Luke (AICNT) | Mark (AICNT) |
| --- | --- |
| “Truly I say to you, no prophet is accepted in his hometown.” (Luke 4:24) | “A prophet is not without honor except in his hometown [and among his relatives] and in his household.” (Mark 6:4) |
| “Do not fear, from now on you will be catching men.” (Luke 5:10) | “Come after me, and I will make you fishers of men.” (Mark 1:17) |
| ““No one tears a piece from a new garment [and sews it] on an old one; otherwise, the new will tear and the patch from the new will not match the old.” (Luke 5:36) | “No one sews a patch of unshrunk cloth on an old garment; if he does, the [new] patch tears away from the old, and a worse tear is made.” (Mark 2:21) |
| “The ones that fell among thorns are those who, when they have heard, [go out and] are choked with cares, [and] wealth, and pleasures of life, and bring no fruit to maturity” (Luke 8:14) | “And there are others who are sown among the thorns; [these are the ones] who hear the word, and the worries of the age and the deceitfulness of wealth and the desires for other things entering in, choke the word, and it becomes unfruitful.” (Mark 4:18–19) |
| “Therefore, be careful how you listen; for whoever has, to them will be given; and whoever does not have, even what they think they have will be taken away from them.” (Luke 8:18) | “Pay attention to what you hear. With the measure you use, it will be measured to you who hear, and more will be added to you. For whoever has, to him more will be given; and whoever does not have, even what he has will be taken away from him.” (Mark 4:24-25) |
| “For what does it profit a man to gain the whole world, but lose or forfeit himself?” (Luke 9:25) | “or what does it profit a man to gain the whole world and forfeit his soul? For what can a man give in exchange for his soul?” (Mark 8:36–37) |
| “Put these words into your ears: for the Son of Man is about to be delivered into the hands of men.” (Luke 9:44) | “The Son of Man is to be handed over to the hands of men, and they will kill him, and [after being killed,] he will rise after three days” (Mark 9:31) |
| “Whoever receives this child in my name receives me; and [whoever receives me, receives] the one who sent me; for the one who is least among all of you, this one is great” (Luke 9:48) | “If anyone wants to be first, he will be last of all and servant of all… Whoever receives one of such children [in my name], receives me; and whoever receives me, does not receive me but the one who sent me.” (Mark 9:35-37) |
| “And everyone who speaks a word against the Son of Man, it will be forgiven him; but he who blasphemes against the Holy Spirit, it will not be forgiven.” (Luke 12:10) | “Truly I say to you, all sins will be forgiven the sons of men, and blasphemies [, whatever they may blaspheme]; but whoever blasphemes against the Holy Spirit does not have forgiveness [forever], but is guilty of an eternal sin.” (Mark 3:28-29) |
| “I say to you, Peter, the rooster will not crow today until you deny three times that you know me.” (Luke 22:34) | Truly I say to you [today,] that on this very night, before the rooster crows [twice], you will deny me three times.” (Mark 14:30) |
| “[Therefore] Salt is good; but if [even] the salt becomes foolish, with what will it be seasoned? Neither into earth nor into dung is it suitable; they throw it out.” (Luke 14:35) | “Salt is good; but if the salt becomes unsalty, with what will you season it? Have salt in yourselves, and be at peace with one another.” (Mark 9:50) |
| “If you had faith like a mustard seed, you would say to this mulberry tree, ‘[Be uprooted and] Be planted in the sea;’ and it would obey you.” (Luke 17:6) | “[And] l Jesus, answering, said to them, “Have faith [in God]. Truly I say to you, whoever says to this mountain, ‘Be lifted up and thrown into the sea,’ and does not doubt in his heart but believes that he says is happening, it will be done for him. Therefore I say to you, whatever you pray and ask for, believe that you have received it, and it will be yours.” (Mark 11:22-24) |
| “And they will say to you, ‘Look, there! [Or,] Look, here!’ Do not go away, nor pursue. (Luke 17:23) | And then if anyone says to you, “Look, Here is the Christ,” or “Look, There he is,” do not believe it; for [false christs and] false prophets will rise and will give signs and wonders in order to lead astray, if possible, the elect. (Mark 13:21-22) |
| “Truly, I say to you, [that] there is no one who has left house or wife or brothers or parents or children, for the sake of the kingdom of God, who will not receive many times more in this time, and in the coming age, eternal life.” (Luke 18:29–30) | “Truly I tell you, there is no one who has left [house or] brothers or sisters or mother [or father] or children or fields [for my sake and for [the sake of] the gospel], who will not receive a hundredfold [now] in this time [houses [and brothers] and sisters and mothers and children and fields, with persecutions] [and] in the age to come, eternal life.” (Mark 10:29–30) |

These points indicate that Luke is often less polished than Mark, and in textual-critical terms, less polished means more primitive. In areas where there are no parallels between Luke and Mark, such as the prologue to Luke, the Greek syntax and grammar are highly advanced. However, where Luke incorporates primitive source material into the gospel narrative, the syntax and grammar are likewise primitive, as would be expected of someone staying true to his sources with minimal alteration.

## Progressive Embellishment

Progressive embellishment (redactional expansion) is not merely an aesthetic judgment. It is a directional indicator within redaction criticism. When one version of a shared episode is plain, restrained, and narratively simple, while its parallels repeatedly add vivid detail, theological explanation, heightened conflict, or harmonizing clarification, the expanded versions are more plausibly secondary revisions of the plainer source.

The point is cumulative. A single parallel might be explained in more than one way, but a repeated pattern of expansion and intensification in the same direction strengthens the case for dependence. The companion articles document the wider body of examples; the study gives representative patterns rather than reproducing every parallel in full.[^22]

The pattern can be catalogued under four recurring categories. The first is dramatic intensification. Mark frequently heightens narrative energy through vivid movement, emotional immediacy, crowd pressure, conflict escalation, and repeated use of εὐθύς (euthys), commonly translated “immediately.” The Lindsey citation above is important because it shows how this Markan habit came into view through direct Hebrew retroversion and comparison with Luke.

The second is narrative polish. Later parallels often smooth the sequence of a scene, clarify who is speaking, add explanatory transitions, or reshape a spare episode into a more finished story. Such polish may make the later Gospel more readable, but it does not thereby make it earlier.

The third is theological clarification. Mark and Matthew often make implicit meanings more explicit through stronger Christological cues, sharper anticipation of the passion, intensified conflict with opponents, or developed narrative devices such as the Markan messianic secret. Matthew then adds further fulfillment formulas, ecclesial concerns, and teaching structures that reflect later community formation.

The fourth is harmonizing addition. Matthew in particular often appears to combine, regularize, or liturgize earlier material. Distinctive Matthean features such as the fuller ecclesial material, the Peter-and-church passages, and the developed Great Commission are best read within this broader tendency toward expansion and theological consolidation rather than as the simplest recoverable form of the tradition.

An examination of sixty-four parallel passages between Luke and Mark shows that Mark is predominantly an expansionist work. In thirty-six triple-tradition parallels (Luke-Mark-Matthew) and twenty-eight additional Luke-Mark parallels, the Markan text is, on average, more than 46% longer than the corresponding passages in Luke. These sixty-four parallels account for 65% of the entire Gospel of Mark, demonstrating that the expansionist pattern is not incidental but characteristic of Mark as a whole.

The figure below provides a quantitative measure of Mark’s expansionist nature. Not only does Mark contain many more words than the parallels in Luke, but the author of Mark also further embellishes the narrative by adding dialogue, intriguing elements, and sensational details.

![Bar chart comparing Greek word counts of tradition parallels in Gospels of Luke, Mark, and Matthew. Left side shows counts for 36 triple tradition parallels with Matthew highest at 4,639 words, while right side shows counts for 28 double tradition parallels with Mark highest at 3,922 words. ](https://ntcanon.com/assets/figures/figure2.svg)

*Figure 2. Progressive Embellishment of Gospel Parallels in Luke*

David Flusser, associated with the Jerusalem School of Synoptic Research, provided further evidence for the view that Luke’s Gospel preserves the most accurate account of Jesus’ life and teachings. Flusser highlights instances in which Luke appears to preserve a more reliable tradition, whereas Matthew, following Mark, is sometimes influenced by Mark’s redactional (editorial) tendencies rather than preserving the earliest form of the tradition.[^23]

One illustrative case is the account of the Man with the Withered Hand (Mark 3:1–6 and parallels in Matthew and Luke). In Mark’s version, the event ends with the Pharisees, along with the Herodians, plotting to destroy Jesus. Matthew, following Mark, preserves this severe conclusion. However, Luke’s version offers a more restrained account, stating, far short of plotting to kill Jesus, that Jesus’ opponents, “discussed among themselves what could be done to Jesus” (Luke 6:11).

This distinction is significant because Luke’s more neutral phrasing is less embellished, reflecting a more restrained and historically plausible account. Mark, by contrast, appears to retroject later hostilities and developments into an earlier stage of Jesus’ ministry. By stating that the Pharisees, along with the Herodians, immediately conspired to “destroy” Jesus (Mark 3:6), Mark attributes to this moment a level of opposition that better fits the intensified conflicts leading up to Jesus’ eventual arrest and crucifixion. This editorial tendency in Mark aligns with what some scholars identify as a theological and dramatic stylization, in which events are framed to heighten tension and opposition to Jesus. Matthew, following Mark, retains this heightened portrayal.

The linguistic and rabbinic context further supports Luke’s primitivity in this pericope. Luke’s phrasing, “they discussed among themselves what they might do to Jesus,” closely echoes a first-century B.C.E. rabbinic story preserved in m. Ta‘anit 3:8, in which Pharisees rebuke the unconventional miracle-worker Honi the Circle-Maker with the question, “What can I do to you?” The implication in the rabbinic context is that nothing legally could be done to Honi because he had not transgressed the halakhah. Luke’s wording reflects exactly this rabbinic idiom of recognizing that Jesus, like Honi, had committed no halakhic violation: healing by command did not violate Sabbath law. Mark, working in Greek and removed from the rabbinic linguistic context, took the Lukan phrase to mean that Jesus’s opponents “took counsel together” to destroy him, and in doing so retrojected motives belonging to the later high-priestly conspiracy. The directional inference is straightforward: Luke preserves an authentic Hebrew/rabbinic idiom; Mark misreads it and inflates the conclusion.[^24]

Flusser concurs that Luke’s Gospel, being less shaped by Markan editorial tendencies, may preserve a more original account of the incident. This example supports the broader argument that Luke often provides a more accurate historical record, whereas Matthew, following Mark too closely, sometimes inherits and amplifies Mark’s redactional elements:

> “The first [lesson] is that any normal philological reasoning would indicate the priority or greater authenticity of Luke’s accounts. Water does not flow uphill. It is simply impossible to believe that the Matthean-Markan account could be changed secondarily into the Lukan form… The second lesson is that understanding of the language usages of Jesus’ time can quite often throw immediate light on questions of originality in our Gospels. The third is that Matthew is indeed secondary to Mark and Mark to Luke, for only in such an order of dependence can we see how Matthew can accept the secondary oddity of the Markan text.”[^25]

A second pericope-level case is the so-called “trial” of Jesus, where Luke’s account again preserves the more primitive form. Luke records no condemnation of Jesus by the Jewish authorities at his hearing before the Sanhedrin: there is only an interrogation, after which Jesus is delivered to Pilate (Luke 23:1) without any verdict by the Jewish council. The three Lukan passion predictions (Luke 9:22, 44; 18:31–33) likewise lack any reference to a Jewish verdict. Mark, by contrast, reports that “all judged him worthy of death” (Mark 14:64). Paul Winter observed in his classic study that the absence of a Jewish condemnation in Luke is a sustained pattern, not an isolated omission, and that the Markan formula reads as an expansion of the Lukan substrate, in which the high priests reach a conclusion (“What further testimony do we need? We have heard it from his own lips,” Luke 22:71) but pronounce no formal sentence. As with the Withered Hand pericope, the directional inference is more economical from Luke to Mark: Mark expands a non-condemning interrogation into a formal condemnation; the reverse direction would require Luke to systematically excise the formal verdict from his three passion predictions and from the trial scene itself, but editorial moves almost always fall in the direction of enhancing a narrative rather than diminishing it.[^26]

A third pericope-level case is the Temple Cleansing (Luke 19:45–46; Mark 11:15–17; Matthew 21:12–13), where Luke’s brief, non-violent account contrasts sharply with Mark’s expanded narrative. Luke’s version is twenty-five words; Mark’s is sixty-five; Matthew’s is forty-five. Luke records that Jesus “began to take out” (ἐκβάλλειν [*ekballein*], in the sense of Hebrew lehotzi, “to take out, remove”) those who were selling, with no mention of overturning tables, no driving out of buyers, no shutting down of Temple traffic, and no violent confrontation. Mark adds the overturning of the money changers’ tables (Mark 11:15), the shutting down of all Temple commerce (Mark 11:16), and the expanded citation “a house of prayer for all the nations” (Mark 11:17). Mark also brackets the Temple incident between the cursing and withering of the fig tree (Mark 11:12–14, 20–25), a literary frame that scholarship has identified as a Markan sectarian agenda presenting the Temple as cursed. Mark 11:16’s claim that Jesus “did not allow anyone to carry anything through the Temple” is historically problematic: m. Berakhot 9.5 forbids carrying a purse on the Temple platform, the merchant stalls were located outside the Temple courts proper, and no Synoptic parallel attests this Markan claim. The directional pattern is striking. Luke contains no violence; Mark introduces violence; Matthew follows Mark; and John (2:13–16) extends the pattern further with Jesus braiding a whip and driving out animals. The escalating violence across the four-Gospel sequence Luke→Mark→Matthew→John supports the directional inference of Lukan primitivity.[^27]

A different kind of evidence comes from the floating transitional phrase “For they no longer dared to ask him another question” (Luke 20:40; Mark 12:34; Matthew 22:46). All three Synoptists place this phrase at the conclusion of a series of Jerusalem-week disputes between Jesus and the religious leaders, but only in Luke does the phrase fit its narrative context. Luke places it at the end of the Sadducean dispute about the resurrection, the climax of a sustained series of exchanges with the priestly authorities (Luke 20:1, 19–20, 27); the phrase fits naturally as a summary of that escalating conflict. Mark places the same phrase after the Great Commandment pericope (Mark 12:28–34), where Jesus and his interlocutor have actually agreed and the scribe has been commended for his answer; the phrase fits the conclusion awkwardly. Matthew places it after the Question about David’s Son (Matthew 22:41–46), where it likewise does not naturally cap the exchange. The directional inference matches the previous cases: the phrase has its native home in the Lukan narrative context, and Mark and Matthew preserve the phrase but have separated it from the material it originally summarized. David Flusser characterized this example as “the clearest illustration of the Synoptic relationship,” since the phrase’s misplacement in Mark and Matthew makes most sense if Mark drew on a Lukan source and rearranged the surrounding pericopae.[^28]

## History and Evidence of Matthean Posteriority

The thesis that Matthew was the last of the three Synoptic Gospels and used Luke as a source is known as Matthean Posteriority. The position is not novel: in addition to the Jerusalem School scholars cited above, it has been advocated by numerous biblical scholars over the past two centuries.

The first modern advocate of Matthean posteriority was Gottlob Christian Storr (1746–1805), who concluded from his attempts to harmonize the Synoptic Gospels that Mark and Luke preserve a more accurate chronological order while Matthew gives a more thematic arrangement, indicating that Matthew was written later.[^29] Johann Gottfried Herder defended a similar position emphasizing the role of oral tradition.[^30] Christian Gottlob Wilke (1786–1854) developed the most detailed nineteenth-century case, arguing that Matthew wrote third and used both Mark and Luke. Wilke identified four lines of evidence: seams in Matthew’s text showing that the double tradition material was oriented on the basis of Luke, modifications resulting from Matthew’s relocation of material out of its original Lukan contexts, vocabulary characteristic of Luke appearing in Matthean parallels, and Lukan redactional interests visible in the double tradition.[^31] Gustav Schlager extended Wilke’s vocabulary argument by demonstrating that distinctive Lukan expressions appear in Matthew at parallel passages in the double tradition and in the minor agreements of Matthew and Luke against Mark, evidence that, on the two-document hypothesis, would require an unmotivated coincidence of Matthew and Luke in distinctively Lukan style.[^32]

The early twentieth-century revival of Matthean posteriority began with William Lockton as expressing the Lukan primary position. Lockton’s argument that synoptic differences are best explained by tradition-development in the order Luke → Mark → Matthew, that Matthew exhibits the most developed and artificial structure of the three, that the infancy narratives of Matthew and Luke share more material than can plausibly be coincidental, and that double tradition vocabulary is more characteristic of Luke than of Matthew, anticipated the later Jerusalem School position.[^33] Ernst von Dobschütz argued independently in the early 20th century that Matthew depended on Luke “at least in a secondary way,” with Matthew generally more developed in arrangement, in the expansion of double tradition material, and in the construction of discourses. Dobschütz also placed Matthew historically late on the basis of its “catholicizing” character, comparable to James, the Pastoral Epistles, and the Shepherd of Hermas, suggesting a date around A.D. 100.[^34]

Taking a step beyond B.H. Streeter, who proposed a “Proto-Luke” (Q + L material) as a source lying behind canonical Luke,[^35] H. Philip West Jr.’s 1967 article in *New Testament Studies* argued that Matthew also used a “Primitive Luke,” one that resembled Marcion’s Gospel in content and arrangement and that filled the role assigned to Q in the two-document hypothesis.[^36] Ronald V. Huggins reintroduced the position to late-twentieth-century scholarship with “Matthean Posteriority: A Preliminary Proposal,” arguing that Matthew’s use of Mark as primary source and Luke as secondary source removes the need for any hypothetical document like Q.[^37] Evan Powell defended Matthean posteriority in two monographs, observing that Matthew makes the same kinds of editorial changes to Luke in the double tradition that he makes to Mark in the triple tradition, and that the high verbatim agreement between Matthew and Luke in the double tradition is explained by Matthew conservatively copying Luke in the same manner he copies Mark.[^38]

George A. Blair’s *The Synoptic Gospels Compared* presents a parallel-text synopsis of the three gospels with interspersed commentary that defends Matthean posteriority in each pericope having synoptic parallels.[^39] Alan Garrow’s 2016 *New Testament Studies* article ‘Streeter’s “Other” Synoptic Solution: The Matthew Conflator Hypothesis’ argues the case from Matthew’s high verbal agreement with both Mark and Luke and from Matthew’s complex scribal operations consistent with codex-form access to both gospels, presenting what he calls the Matthew Conflator Hypothesis.”[^40] James R. Edwards, in The Hebrew Gospel and the Development of the Synoptic Tradition, argued that the canonical Matthew is the latest synoptic and was composed after both Mark and Luke, distinguishing it from the earlier Hebrew Gospel attributed by the church fathers to the Apostle Matthew.[^41]

Martin Hengel, longtime professor of New Testament at the University of Tübingen and one of the most influential New Testament historians of the late twentieth century, offered a mediating position holding that Matthew used Luke directly while retaining a sayings source shared by Matthew and Luke. Hengel’s general approach to early Christian origins has been broadly accepted in critical scholarship, making his support for Matthew’s direct use of Luke a significant scholarly endorsement of the directional claim, even though he stops short of the full Matthean-posteriority position.[^42] David A. deSilva, drawing on Martin Hengel, observes that while Luke’s use of Matthew “remains problematic in the extreme,” the alternative possibility of Matthew’s use of both Mark and Luke has been comparatively neglected and may prove the most elegant solution to the synoptic problem.[^43]

Robert MacEwen’s *Matthean Posteriority: An Exploration of Matthew’s Use of Mark and Luke as a Solution to the Synoptic Problem* provides the most comprehensive recent defense of Matthean Posteriority, surveying the prior scholarship and developing the textual and verbatim-agreement evidence in support of the hypothesis.[^44] MacEwen identifies seven principal arguments for the position: higher verbal agreement between Matthew and each of Mark and Luke than between Mark and Luke; the comparative ease of explaining Matthew’s arrangement of the double tradition if he used Luke than the reverse; Matthew’s general developmental priority in literary structure and theological elaboration; the adequacy of Matthew’s redactional tendencies to explain his omissions of Lukan special material (Sondergut); historical factors including Hengel’s argument for a later first-century Matthean setting; evidence in the infancy and resurrection narratives that Matthew knew Luke’s accounts; and the argument from characteristic vocabulary.[^45]

The most striking quantitative evidence is the asymmetric expansion pattern documented by Evan Powell. Although the Gospel of Luke is approximately seven percent longer than the Gospel of Matthew overall, Matthew contains a higher concentration of material across all seven categories of tradition that Powell examined. Luke contains seventy-seven percent as many references to supernatural events as Matthew, seventy-one percent as much eschatological content, seventy-three percent as many ethical sayings, seventy-five percent as many references to Jesus as Christ, eighty-three percent as many to Jesus as Son of Man, seventy-five percent as many to the Kingdom of God, and only thirty-six percent as many to God as Father. The pattern is consistent and one-directional: Matthew develops nearly every prominent category of Jesus tradition more elaborately than Luke does. The proposition that Luke would have systematically reduced Matthew across seven independent categories while simultaneously producing a longer gospel overall strains plausibility, whereas the proposition that Matthew expanded Luke fits the data without modification.[^46]

Specific Matthean revisions to Luke illustrate the editorial pattern. Where Luke instructs forgiveness of a brother who repents “seven times” in a day (Luke 17:3–4), Matthew amplifies the obligation to “seventy-seven times” (Matt 18:21–22). Where Luke records that any disciple must “hate” father, mother, wife, children, and siblings (Luke 14:26–27), Matthew softens the demand to a question of comparative love and worthiness, removing the difficult word and reducing the listed relations to those of closest natural bond (Matt 10:37–38). Where Luke directs John the Baptist’s “brood of vipers” rebuke at the crowds at large (Luke 3:7–8), Matthew redirects it specifically to the Pharisees and Sadducees, sharpening the polemic against the religious elite (Matt 3:7–9). Where Luke and Mark both describe Joseph of Arimathea as a member of the Sanhedrin who had not consented to Jesus’s condemnation (Luke 23:50–53; Mark 15:43), Matthew reshapes him as a rich disciple of Jesus, removing the implication that a member of the Jewish ruling council had honored the body of Jesus (Matt 27:57–60). Each revision points the same direction: Matthew clarifying, softening, sharpening, or theologically reshaping a Lukan original to suit later editorial aims.[^47]

Matthew’s conflation of Mark and Luke is the most distinctive feature of the directional case, because conflation is a textual signature that one document drew on two earlier ones. The Calling of the Twelve at Matthew 9:35–10:4 weaves together material from Mark 6:6, Luke 8:1, Mark 6:34, Luke 10:2, Luke 9:1b, Mark 3:14–19, and Luke 6:13–16, producing eight Matthean verses drawn from six different chapters across Mark and Luke. The Beelzebul controversy at Matthew 12:22–30 combines near-verbatim material from Mark 3:22–27 and Luke 11:14–23, where Mark and Luke themselves share only minimal verbatim agreement, indicating that Matthew is the conflating party rather than either Mark or Luke. The community discourse at Matthew 18:1–22 likewise stitches together Markan and Lukan units into a single coherent passage. The methodological significance is that Luke exhibits no parallel pattern of conflating Mark with Matthew. If Luke had used Mark and Matthew, the textual signature would be detectable in similar weaving. Its absence in Luke and its abundance in Matthew is one of the strongest single arguments for the Matthew→Luke direction of dependence.[^48]

Deeper textual evidence further establishes Matthew’s specific knowledge of Luke. Matthew uses the form of Jerusalem Ἰερουσαλὴμ (*Ierousalēm*) exactly once, at Matthew 23:37, in the lament “Jerusalem, Jerusalem, the city that kills the prophets,” exactly where Luke uses the same form at Luke 13:34. Elsewhere Matthew uses the Markan form Ἱεροσόλυμα (*Hierosolyma*). The single Matthean appearance of the Lukan form, in the same passage where Luke uses it, points directly to Lukan influence on this verse. The Aramaic loanword Mamon appears three times in Luke 16:9–13 but only once in Matthew, at Matthew 6:24, in a verse that summarizes Lukan content. Matthew’s treasures-in-heaven passage at Matthew 6:19–21 reorganizes material drawn from Luke 12:32–34 and surrounding verses, with the editorial choice to omit Luke’s instruction to sell possessions and give alms. The Sermon on the Mount’s law-and-prophets prologue at Matthew 5:17–20 shows clear engagement with the apparently contradictory pair at Luke 16:16–17, where Matthew resolves Luke’s tension by inserting Jesus’s affirmation that the law and prophets remain in force while being fulfilled, with the otherwise unmotivated reference to scribes and Pharisees in Matthew 5:20 reflecting awareness of the broader Lukan context at Luke 16:14–15.[^49]

A final indicator of Matthean lateness is the cluster of explicitly Judaizing interpolations that appear in Matthew but not in Luke. These include the warning that whoever relaxes the least commandment will be called least in the kingdom (Matt 5:19–20), the saying that only the doer of the Father’s will enters the kingdom (Matt 7:21–23), the fiery-furnace fate for “law-breakers” (Matt 13:36–43), the recompense by deeds at the Son of Man’s coming (Matt 16:26–27), the commandment-keeping requirement for eternal life (Matt 19:17), the directive to obey what the scribes and Pharisees teach (Matt 23:2–3), and the eternal-punishment-versus-eternal-life judgment scene (Matt 25:31–46). Matthew also displays markedly higher frequency of vocabulary characteristic of later Jewish-Christian polemic, including “lawlessness,” “righteousness,” and to “be perfect.” If Luke had used Matthew, the omission of this entire stratum of unambiguous teaching about commandment-keeping and final judgment would require an explanation that no proponent of Lukan dependence on Matthew has supplied. The interpolations are far more naturally explained as Matthean additions reflecting the theological commitments of a later Jewish-Christian Torah-observing community. The cumulative force of the expansion data, the editorial-revision pattern, the conflation textual signature, the demonstrated knowledge of specific Lukan vocabulary and context, and the late Judaizing interpolations is that Matthew drew on Luke as a secondary source alongside its primary Markan base.[^50]

## The Hebraic Syntax of Luke

The comparative case is strengthened by a further linguistic feature: Luke’s Greek exhibits an unusually heavy reliance on grammatical patterns and idioms that come from Hebrew rather than native Greek. Scholars call these patterns “Semitisms.” This evidence should be read as corroborating the comparative argument. If Luke repeatedly preserves the more primitive wording in Gospel parallels, then Luke’s stronger Hebraic character helps explain why its wording often appears closer to an earlier Semitic source.

> “Hebraisms proper are special characteristics of Luke. There is reason, therefore, for a closer scrutiny of the style of this evangelist with its wealth of Hebraisms.”[^51]

This fact can be attributed to the theory that Semitisms derive from an original Hebrew Gospel authored by an apostolic witness. James Edward tested this hypothesis in *The Hebrew Gospel & The Development of the Synoptic Tradition*. His approach was to chart Luke’s individual Semitisms verse by verse, to see whether they occurred in statistically greater numbers in passages unique to Luke. Edwards found that in the half of Luke that is “Special Luke,” Luke exhibits approximately a 400% increase in Semitic character. This bears witness that Luke did not attempt to diminish the Semitisms in his Hebraic source material by altering them to conventional Koine (common Greek) standards. Luke is faithful to the extrinsic literary standards of his source material. Edwards likens the ability to identify the Semitic character of the Gospel of Luke to seeing faint Hebrew characters underlying the Greek text:

> “Reading the Greek NT with a knowledge of biblical Hebrew is like reading a palimpsest. The Hebrew thought world, like a subtext, often lies faintly beneath the Greek surface. But in the Gospel of Luke – or at least in parts of it – the subtext became much more visible. The Hebrew words seem to have been erased less completely than elsewhere in the Gospels. They are more evident, intrusive, and inescapable. Like rocks and coral reefs, they lay barely submerged beneath Luke’s Greek. Nor did Luke seem to make an effort to tame or camouflage the Hebraisms. Their primitive and alien dignity seem to be consciously retained without Hellenizing or harmonizing to Lukan style. They give every appearance of coming from a source that the author valued and attempted to preserve.”[^52]

The Hebrew Gospel is not merely a hypothetical reconstruction. Its historical existence is attested by a long and geographically broad patristic tradition. Twenty church fathers across the first ten centuries of Christian history reference the Hebrew Gospel or quote from it, including Ignatius, Papias, Irenaeus, Clement of Alexandria, Hegesippus, Hippolytus, Origen, Eusebius, Ephrem of Syria, Didymus of Alexandria, Epiphanius, John Chrysostom, Jerome, Theodoret, and the Venerable Bede, with Jerome alone citing it twenty-two times. The aggregate corpus amounts to approximately seventy-five references spanning geographical regions from the western Roman Empire to India. The Hebrew Gospel occupied a distinctive canonical status in early Christian deliberations: Eusebius, in his threefold classification of recognized, disputed, and rejected books, placed the Hebrew Gospel in the disputed category alongside Revelation rather than among the rejected writings, indicating that it was accepted as authoritative in many Christian communities even as it was contested in others.[^53] Edwards concludes that no other non-canonical document occupied the disputed category as long and consistently as the Hebrew Gospel, and that no other non-canonical text was cited as frequently and positively alongside canonical texts in early Christian exegesis.[^54]

The paradox that has long puzzled scholarship is that patristic witnesses ascribe the Hebrew Gospel to the apostle Matthew even though its content is demonstrably closer to Luke than to canonical Matthew. The resolution is that the Hebrew Gospel (Hebrew Matthew) and canonical Greek Matthew are distinct compositions. The Hebrew Gospel is the primitive apostolic source preserved most faithfully in Luke, while canonical Greek Matthew is a later, stylistically polished, and heavily redacted work that draws on Mark rather than on the Hebrew Gospel directly.

The case for Semitic influence is further strengthened when clusters of Semitisms occur in portions of a document that are otherwise drafted in conventional Koine Greek. Davila noted:

> “If we found blocks of text containing a high density of Semitisms alongside blocks of good Greek… we could conclude that the writer was either incorporating translated Greek passages into the work or translating passages from a Semitic source in some places while writing in his or her normal style in others.”[^55]

Edwards’ analysis shows that Semitisms in Luke are not uniformly distributed, contradicting the idea that Luke deliberately imitated the Septuagint’s style throughout his gospel. Instead, the uneven concentration of Semitic features suggests that certain portions of Luke were based on a Semitic prototype rather than being an artificial stylistic choice. The presence of unusual or awkward Semitic words, idioms, and expressions in an otherwise highly cultivated Greek text indicates that these linguistic anomalies stem from Hebraic conventions embedded in Luke’s principal source. This supports the premise that Luke incorporated Semitic material rather than merely emulating the Septuagint.

Luke has often been considered the most learned in the Greek tongue among all the evangelists. It makes no sense that an author would craft the prologue in the highest form of Greek and then adhere to a style that is highly Semitic in much of the gospel and less so in other places unless he was doing his best to convey his sources in the truest way he could. Edwards affirmed that the foundational Hebraic source for Luke’s special material is the Hebrew Gospel:

> “This Semitic source apparently functioned as a primary source for Luke, into which other sources were integrated or to which they were supplemented according to Luke’s overall purpose. That Luke did not try to expunge and blend his sources, and particularly, his Semitic source, is indicated by stylistic differences… which are particularly evident in the high-caliber Greek of the prologue, the basic Koine in passages shared with Matthew and/or Mark, and the distinctly Semitically- flavored Greek of Special Luke. As suggested in the prologue, Luke endeavored to produce a full and final narrative while leaving vestiges of the sources that comprise it.”[^56]

The suggestion that Luke translated Hebrew texts may surprise some. A common characterization is that Luke was a Greek, given his proficiency in Greek. However, patristic sources do not provide a definitive answer about Luke’s ethnic background. Some modern scholars argue that Luke was a Hellenistic Jew, citing his intimate knowledge of Jewish customs and the Temple, as well as the detailed accounts of Jewish rituals in his writings.

There are additional clues that Luke was competent in Hebrew. Epiphanius (c. 310–403 AD), in his work *Panarion*, indicates that Luke was a Jew:

> “Here is a third Gospel, Luke’s, they said (for Luke was given this commission. He too was one of the seventy-two who had been scattered because of the Savior’s saying).” [^57]

Eusebius writes that Luke translated Hebrews from Hebrew to Greek:

> “And as for the Epistle to the Hebrews, he says indeed that it is Paul’s, but that it was written for Hebrews in the Hebrew tongue, and that Luke, having carefully translated it, published it for the Greeks; hence, as a result of this translation, the same complexion of style is found in this Epistle and in the Acts.”[^58]

Jerome, in *On Illustrious Men* (5), also acknowledged that some claimed Luke translated Hebrews, though he maintained that Luke was Greek (7).

Regardless of whether he was Greek or a Hellenized Jew, there is no clearer indication that Luke was proficient in Hebrew if, in fact, he is the translator of Hebrews into Greek. Thus the Hebraic evidence functions as one part of a cumulative case. It supports the same direction already suggested by the parallel texts: Luke stands nearest to the earlier Semitic and apostolic source material, while Mark and Matthew reflect later stages of revision and adaptation.

## Statistical Validation of Lukan Priority

Halvor Ronning, founding member and past director of Jerusalem School of Synoptic Research, published a multi-part series in which he conducted statistical and comparative analysis of all the potential options regarding the sequence of the Synoptic Gospels.

In *A Statistical Approach to the Synoptic Problem: Part 1*, Ronning conducted statistical analysis on the material with parallels in Luke, Mark, and Matthew (triple tradition material), considering all six possible theoretical relationships:[^59]

- Matthew→Luke→Mark
- Mark→Luke→Matthew
- Luke→Matthew→Mark
- Mark→Matthew→Luke
- Luke→Mark→Matthew
- Matthew→Mark→Luke

The method counts words that are identical in form and sequence (IFS) between any two Gospels: identical Greek words appearing in the same word order. For each of the six theoretical orderings, the percentage of IFS words flowing from the first Gospel through the middle Gospel to the third can be predicted, then compared to the actual percentage observed in the texts. Four of the six orderings produce predictions that diverge sharply from the observed counts; only the two orderings with Mark in the middle position fit the data:

- Matthean Priority Scenario: Matthew→Mark→Luke
- Lukan Priority Scenario: Luke→Mark→Matthew

In both cases, the actual word counts came within about 0.1% of the theoretical predictions. The indication that Mark must be the middle term between Luke and Matthew has often been termed the Markan cross-factor: statistical word-pattern analysis rules out four of the six possible Gospel sequences, leaving only the two scenarios with Mark in the middle position.

The scale of these contrary agreements is what gives the Markan cross-factor its weight. In the triple tradition, 17.4% of Matthew’s text and 17.9% of Luke’s text consists of words that Matthew and Luke share against Mark, totaling 1,163 words across all categories of agreement. Beyond these positive agreements, Matthew and Luke also agree to drop approximately 2,000 of Mark’s words at the same parallel positions, which on the standard Markan-priority story would represent roughly 30% of additional potential text that both evangelists supposedly elected to omit. Taken together, the positive and negative agreements between Matthew and Luke against Mark affect more than 45% of the potential triple-tradition text of each Gospel. Lindsey expressed the same pattern in summary form: in the Double Tradition where Mark is absent, Matthew and Luke can share as much as 90% of their wording, but in the Triple Tradition where Mark is present, their shared wording falls to roughly 25%. Mark’s intervening presence is therefore the variable that controls the Matthean-Lukan agreement rate. Ronning’s preferred name for this body of evidence is therefore not “minor agreements” but “Key Agreements,” since their cumulative weight is what unlocks the Synoptic Problem.

A second observation undermines a long-standing argument for Markan priority. Markan-priority advocates traditionally appeal to Mark’s overall brevity, reasoning that the shorter Gospel must be the original. Within the triple tradition itself, however, Mark is the longest, not the shortest: 7,889 Markan words tell the same parallel stories that Matthew tells in 6,682 and Luke in 6,482. Mark’s overall brevity reflects his omission of much non-overlapping material (no infancy narrative, no extended Sermon, a much shorter resurrection account); within the parallels themselves, Mark consistently expands. The brevity argument inverts when measured against the very parallel material it is supposed to explain.

In *A Statistical Approach to the Synoptic Problem: Part 2*, Ronning conducts additional analysis to also include Double Tradition text (text that is attested by two of the three gospels but not found in one of the three).[^60] Most of the double tradition material is instances where Luke and Matthew agree on words, phrases, and sentences contrary to Mark. These instances are called “minor agreements” and were the key to determining which of the two remaining scenarios is statistically valid.

The result favors Lukan Priority on two fronts: consistency and prediction. The consistency check observes that under Lukan Priority, Matthew’s percentage of word identity with Mark stays steady (around 50%) whether Luke is also parallel or not. Matthew treats Mark as a single source and remains uniformly close to it. Under Matthean Priority, by contrast, Luke would have to relate to its parallel sources at three different rates (32%, 39%, and 50%) depending on which other Gospel is present, an unmotivated inconsistency for which no editorial rationale has been proposed.

The predictive check is sharper still. If Lukan Priority is correct, then the percentage of Matthew’s text identical in form and sequence with Luke, when Mark is also parallel, must equal the percentage of Mark’s text identical with Luke (32%) multiplied by the percentage of Matthew’s text identical with Mark (50%). The arithmetic predicts 16%. The observed count, after subtracting the minor agreements (which by definition could not have reached Matthew via Mark), is 16%. The deviation between prediction and observation is less than one-tenth of one percent. Matthew’s triple-tradition relationship to Luke is the precise mathematical residue of his relationship to Mark multiplied by Mark’s relationship to Luke, exactly what Lukan Priority requires and what no competing scenario predicts with the same precision.

In *A Statistical Approach to the Synoptic Problem: Part 3*, Ronning addresses single tradition material (passages found in only one gospel) as well as double and triple tradition material with respect to Semitic influence.[^61] Here, Ronning highlights Raymond Martin’s prior analysis. The value of Martin’s data lies primarily in providing a quantitative comparison of Matthew’s use of Semitic indicators (Hebraic syntax) with Luke’s.

The numerical data indicated that Luke’s single traditional material is much more Semitic than Matthew’s, and that both Luke’s double and triple traditional material are slightly more Semitic than Matthew’s. In both material classifications, Luke is more Semitic and thus more likely than Matthew to be an earlier, more primitive text. The data also showed that Mark’s triple tradition material is the least Semitic among the Synoptic Gospels. This evidence strongly supports Lukan Priority. In short: Luke’s material is more Semitic in character, which is what one would expect from the earliest gospel most closely reflecting its Hebrew sources.

The Luke-versus-Acts contrast within Luke’s own corpus is decisive on this point. Luke’s single-tradition material in his Gospel is the most Semitic of all the data Martin examined; the second half of Acts, where the same author writes from his own autoptic experience as Paul’s companion, is among the most purely Greek material in the corpus. The same author works at opposite ends of the Semitic-Greek spectrum within his own two-volume work. This rules out the suggestion (advanced by H. F. D. Sparks among others) that Luke’s Hebraic syntax is an authorial mannerism or stylistic affectation imitating the Septuagint. Luke can write smooth Greek when his sources permit it; the Semitic syntax in his Gospel reflects sources, not authorial style.

In *A Statistical Approach to the Synoptic Problem: Part 4*, Ronning addresses common hypothetical scenarios in which more than one primary source is used for a particular gospel, examining them in respect to the ease of back-translating the Greek text into Hebrew.[^62]

Several patterns from the broader analysis sharpen the case against the chief alternatives. Within the triple tradition Luke is both the easiest to retrovert into Hebrew and the shortest of the three Gospels, inverting the brevity argument traditionally offered for Markan priority. The minor agreements, problematic on every other hypothesis, become the predicted residue of Lukan source material reaching Matthew through Mark. The Markan-Conflation hypothesis (Griesbach: Matthew first, Luke uses Matthew, Mark conflates both) faces a particularly acute difficulty: Mark accepts roughly 48% of Matthew’s text and drops the remaining 52%, but the material Mark accepts is precisely the Matthean material that resists Hebrew retroversion, while the material Mark drops is the Matthean material that retroverts easily. Under Griesbach, Mark would have had to systematically prefer non-Hebraic Matthean content while discarding the Hebraic, an editorial pattern for which no rationale has been proposed. The pattern is naturally explained the other way around: Mark used Luke and Matthew later combined Mark with a Hebraic source independent of Mark.

Ronning concludes with the most plausible scenario:

> “For Lukan Priority the Matthean-Lukan agreements against Mark… are no longer a problem, but part of the elegant solution made possible by these “key agreements.” When Mark is placed in the middle the Matthean-Lukan agreements against Mark can now be easily understood as the “corrections” of the final Gospel writer. The final writer—the author of the canonical Greek Matthew—is to be imagined as working with Mark, his immediate source, in front of him. But out of his awareness of Luke’s Hebraic source, Matthew reinserted many of the exact words that Luke had copied from that first Gospel, words that Mark had omitted. Matthew reinserted them right back into the exact same positions as in Luke. Not only did Matthew make many reinsertions, but he also frequently omitted words and phrases that Mark had inserted, words and phrases that were not present in that earlier Gospel that stands behind the Gospel of Luke. This all falls into place as we see the statistical support for the mediating position of Mark between Luke and Matthew…We can judge that Luke was more steady and less innovative in relation to his sources. Even though we do not have copies of Luke’s sources, we can observe how Luke preserved whole blocks of material that are more consistently easy to translate into Hebrew than the parallel material in Mark or Matthew. Luke does not share the same degree of erratic character with respect to Hebrew retroversion as does Mark. Like Luke, Matthew is also generally easier to revert to Hebrew than Mark, except where Matthew has a Markan parallel. Where Matthew has a Markan parallel, Matthew is just as difficult to revert to Hebrew as Mark. These observations are the origin of Lindsey’s insights regarding the dependence of Matthew on Mark, and the independence of Luke from either.”[^63]

## The Unreliability of Patristic Tradition on Gospel Origins

The evidence presented thus far establishes Luke’s priority through comparative textual analysis, primitive wording, comparative restraint, Hebraic Greek syntax, and statistical analysis. So, what defense remains for the traditional view that Mark was written first? The usual appeal is to patristic tradition, though here the appeal is awkward, since the Fathers almost unanimously affirm Matthean rather than Markan priority. Where they are invoked in support of Mark, it is typically to defend Mark’s apostolic origin (via Peter) and early composition rather than its priority among the Synoptics. But the patristic tradition about the origins of the gospels cannot bear the weight it has historically been asked to carry. Patristic writings cannot reliably validate the canon or accurately reconstruct the origins of the gospels and the early history of the Church. Many of these writings reflect theological agendas, post hoc rationalizations, and retrojected traditions rather than firsthand apostolic testimony. These failures include contradictions among Papias, Irenaeus, and Clement of Alexandria, transmitted with further inconsistency by Eusebius and Jerome, concerning the composition of Mark; the erroneous conflation of Papias’s reference to a Hebrew *logia* of Matthew with canonical Greek Matthew; and the pattern by which apostolic authority was retroactively assigned to texts whose actual origins were unknown or disputed.[^64]

The traditional account of Mark’s composition begins with the following statement from Papias, preserved by Eusebius:

> “And the Presbyter used to say this, ‘Mark became Peter’s interpreter and wrote accurately all that he remembered, not, indeed, in order, of the things said or done by the Lord. For he had not heard the Lord, nor had he followed him, but later on, as I said, followed Peter, who used to give teaching as necessity demanded but not making, as it were, an arrangement of the Lord’s oracles, so that Mark did nothing wrong in thus writing down single points as he remembered them. For to one thing he gave attention, to leave out nothing of what he had heard and to make no false statements in them.’”[^65]

This account of an account of an account is anything but the truth, as Mark is anything but a transcript. The following statement by Irenaeus affirms that Mark was written after Peter’s departure, which conflicts with Eusebius’s account of Papias:

> “After their departure, Mark, the disciple and interpreter of Peter, did also hand down to us in writing what had been preached by Peter.”[^66]

Irenaeus’s quote indicates that Mark wrote his Gospel after Peter’s death, implying a date sometime after 64 AD, which contradicts Papias, who implies that Mark wrote while Peter was alive. The quote from Irenaeus further conflicts with Clement of Alexandria, according to Eusebius:

> “And again in the same books Clement has inserted a tradition of the primitive elders with regard to the order of the Gospels, as follows. He said that those Gospels were first written which include the genealogies, but that the Gospel according to Mark came into being in this manner: When Peter had publicly preached the word at Rome, and by the Spirit had proclaimed the Gospel, that those present, who were many, exhorted Mark, as one who had followed him for a long time and remembered what had been spoken, to make a record of what was said; and that he did this, and distributed the Gospel among those that asked him. And that when the matter came to Peter’s knowledge he neither strongly forbade it nor urged it forward.” [^67]

Now, according to Clement, Mark wrote his Gospel in Rome at the request of Peter’s listeners. Peter did not explicitly authorize it, but he also did not reject it. This contradicts Irenaeus, who claims Mark wrote after Peter’s death. Irenaeus’ statement also does not make sense, since Peter’s own disciples would not need a written account if they had heard Peter firsthand. Furthermore, the idea that Peter was neutral about Mark’s Gospel seems unlikely if Peter was Mark’s direct source. Around 393, Jerome makes a claim of Peter’s approval against Clement:

> “Mark the disciple and interpreter of Peter wrote a short gospel at the request of the brethren at Rome embodying what he had heard Peter tell. When Peter had heard this, he approved it and published it to the churches to be read by his authority.”[^68]

What patristic writers are doing here, as elsewhere, is making historical claims most favorable to orthodoxy. The probable reality is that Mark was composed after Peter’s death and was not a transcription of his testimony, but a composed gospel drawing on earlier written and oral sources. Irenaeus, in fact, says as much (*Adv. haer.* 3.1.1), placing Mark’s composition “after the departure” of Peter and Paul. Clement of Alexandria’s softer claim that Peter “neither hindered nor encouraged” the work (Eusebius, *Hist. eccl.* 6.14.7) is itself the kind of post hoc retrojection that patristic testimony repeatedly produces: a tradition adjusted to secure apostolic standing for an otherwise anonymous gospel. When Papias and his successors styled Mark “the interpreter of Peter,” the relationship was exaggerated, and over time embellished, to anchor the Gospel of Mark to apostolic authority.

The evidence and conflicting testimonies above establish that the early Patristic writers cannot be trusted to convey the actual circumstances surrounding the composition of Mark and, by extension, the truth about the origin and acceptance of the other gospels. Statements attributed to Papias, Irenaeus, and Clement of Alexandria cannot be regarded as reliable, nor can Eusebius’s church history be relied upon to convey the truth accurately. Furthermore, Papias is possibly the least reliable of all patristic witnesses, as Eusebius says of Papias:

“For he was a man of very little intelligence, as is clear from his books.”[^69]

Similar unreliability characterizes other patristic witnesses whose writings have come down to us in altered or expanded form. The Ignatian corpus, for example, exists in multiple recensions (revised versions), and several of the letters traditionally attributed to Ignatius are recognized as fabrications rather than authentic compositions.[^70] The broader point is that the surviving writings of the apostolic fathers are not necessarily original to the figures whose names they bear, and cannot be treated as transparent witnesses to the circumstances surrounding the composition of the New Testament writings.

Another major misunderstanding rooted in patristic tradition is that Greek Matthew is the first gospel written. Early church tradition holds that Matthew wrote first in Hebrew/Aramaic and that Greek Matthew is a translation of Hebrew Matthew. Patristic sources agree that Matthew was written in Hebrew/Aramaic. However, Jerome writes that the originator of Greek Matthew is unknown:

> “Matthew, also called Levi, apostle and aforetimes publican, composed a gospel of Christ at first published in Judea in Hebrew for the sake of those of the circumcision who believed, but this was afterwards translated into Greek, though by what author is uncertain.”[^71]

Jerome acknowledges that the originator of the Greek Matthew is unknown, and that it is a later development, which complicates claims of direct apostolic authorship. It can be demonstrated that the Gospel of Luke has more in common with Hebrew Matthew than Greek Matthew does.[^72] Thus, the claim that Greek Matthew is a translation of Hebrew Matthew is a false assumption of patristic tradition.

A related methodological question concerns why the study engages Papias, Irenaeus, and Clement of Alexandria as the principal patristic witnesses to gospel origins. Even the earliest extant patristic witnesses are substantially removed from the events they describe. Papias, composing in the early second century, stood seventy to eighty years past the ministry of Jesus. Irenaeus, writing around 180 AD, stood a century and a half past. Neither had living access to the first-generation church, and the oral chains on which they reportedly relied were already long enough for legendary accretions and mythological reshaping to set in. Their accounts of gospel origins are not firsthand reports but retrospective constructions shaped by second-century apologetic priorities: securing apostolic attributions against Gnostic and Marcionite rivals, and stabilizing a coherent tradition of gospel authorship where the historical record itself was fragmentary. Writers after Irenaeus sit further still from the events, and their testimony is derivative: Jerome draws heavily on Eusebius, who in turn drew on Papias and Irenaeus. The patristic tradition on gospel origins begins already at several generations’ removed. It is built on oral tradition and apologetic need, and later writers refine or expand what their predecessors had fixed rather than add independent information.

Similar unreliability characterizes the patristic tradition about John’s composition, which the section on John below will address directly. What these three test cases jointly demonstrate is that the patristic tradition cannot be used to defend the apostolic origin of any of the canonical gospels. Attributions and compositional accounts were assigned retroactively, often generations after the texts were written, by writers who did not have access to the actual circumstances of authorship and whose own testimonies contradict one another. When textual evidence is added to this picture, the traditional apostolic attributions cannot be sustained. That textual evidence is what the remaining sections of this study examine.

Mark and Luke had already attained sufficient circulation and authority that no direct apostolic attribution was needed, unlike later documents such as Matthew and John, which were assigned to apostles outright.[^73] The retroactive assignment of apostolic authority to the second-century anonymous works that became “Matthew” and “John” supports an inference whose force is easy to miss: once apostolic naming became the dominant route to credibility (roughly mid-second century onward), only writings already established in the churches could survive without direct attribution to an apostle. Mark and Luke fall in this earlier stratum. Later apologists nonetheless felt it necessary to anchor them to apostolic authority *indirectly*, by claiming alignment with Peter and Paul, respectively. The Luke-Paul alignment has substantial New Testament support (the “we-passages” in Acts, Paul’s references to Luke in Colossians 4:14 and 2 Timothy 4:11, and the broadly Pauline theological orientation of Luke-Acts), but the evidence presented thus far demonstrates that the patristic claim of Mark-Peter alignment does not bear out under scrutiny.

[^1]: Johann Jakob Griesbach, “A Demonstration that Mark Was Written After Matthew and Luke,” trans. Bernard Orchard, in J. J. Griesbach: [Synoptic and Text-Critical Studies 1776–1976](https://archive.org/details/jjgriesbachsynop0000joha), ed. Bernard Orchard and Thomas R. W. Longstaff, SNTSMS 34 (Cambridge: Cambridge University Press, 1978), 103–135. Griesbach’s 1789–1790 Latin original, delivered as a Pentecost program at Jena, first articulated the two-gospel hypothesis in its mature form. Henry Owen had previously proposed the same sequence (Matthew first, Luke using Matthew, Mark conflating both) in [Observations on the Four Gospels](https://archive.org/details/bim_eighteenth-century_observations-on-the-four_owen-henry_1764/) (London, 1764). The nineteenth-century development of the hypothesis is surveyed in William R. Farmer, [The Synoptic Problem: A Critical Analysis](https://archive.org/details/synopticproblemc0000farm_w8r4), 2nd ed. (Macon, GA: Mercer University Press, 1976); Hans-Herbert Stoldt, [History and Criticism of the Marcan Hypothesis](https://archive.org/details/historycriticism0000stol), trans. Donald L. Niewyk (Macon, GA: Mercer University Press, 1980); Meijboom, *A History and Critique of the Origin of the Markan Hypothesis, 1835–1866*, trans. John J. Kiwiet (Macon: Mercer University Press, 1993).
[^2]: William Lockton, “[The Origin of the Gospels](https://archive.org/details/churchquarterly14unkngoog/page/n227/mode/2up),” Church Quarterly Review 94 (1922): 216–239.
[^3]: “[Jerusalem Perspective](https://jerusalemperspective.com/),” online journal of the Jerusalem School of Synoptic Research, [https:jerusalemperspective.com](https://jerusalemperspective.com/)
[^4]: Robert L. Lindsey, [A Hebrew Translation of the Gospel of Mark](https://archive.org/details/a-hebrew-translation-of-the-gospel-of-mark/page/n3/mode/2up) (Jerusalem: Dugith Publishers, 1973), 12–13, [Robert Lisle Lindsey, A Hebrew translation of the Gospel of Mark, Jerusalem, Dugith Publishers, 2nd edition, (1973)](https://archive.org/details/a-hebrew-translation-of-the-gospel-of-mark/page/n3/mode/2up).
[^5]: B. H. Streeter, [The Four Gospels: A Study of Origins](https://archive.org/details/bwb_C0-BOB-469) (London: Macmillan, 1924; 4th rev. impression, 1930), 222 (body) and xiii–xv (Preface to the 4th rev. impression). The “primary” and “secondary” terminology and the “half-way house” framing appear in the Preface; the parallel discussion in the body of the work treats Proto-Luke as “another authority comparable to Mark” and notes that Proto-Luke and Mark are roughly contemporary in date.
[^6]: Georg Strecker, *History of New Testament Literature* (Harrisburg: Trinity Press International, 1997), 117.
[^7]: Josiah E. Verkaik, “[Proto-Luke and Lukan Priority](https://lukeprimacy.com/proto-luke/),” LukePrimacy.com, [Proto-Luke and Lukan Priority](https://lukeprimacy.com/proto-luke/).
[^8]: Streeter, [The Four Gospels](https://archive.org/details/bwb_C0-BOB-469), 217 (rebutting the “amorphous document” objection) and 221 (the explicit “complete Gospel” formulation).
[^9]: Vincent Taylor, [Behind the Third Gospel: A Study of the Proto-Luke Hypothesis](https://archive.org/details/behindthirdgospe0000tayl) (Oxford: Clarendon Press, 1926)
[^10]: B. H. Streeter, [The Four Gospels: A Study of Origins](https://archive.org/details/bwb_C0-BOB-469) (London: Macmillan, 1924; 4th rev. impression, 1930), 218–19, locating Luke’s gathering of source notes at Caesarea during Paul’s captivity (Acts 24:27) and the composition of Proto-Luke itself to a later period after Paul’s death.
[^11]: Vincent Taylor, [Behind the Third Gospel: A Study of the Proto-Luke Hypothesis](https://archive.org/details/behindthirdgospe0000tayl) (Oxford: Clarendon Press, 1926), 75 (the insertion argument), 213 (dating of Proto-Luke), and 29–30 (verbal-statistics methodology).
[^12]: Vincent Taylor, “[The Proto-Luke Hypothesis: A Rejoinder](https://journals.sagepub.com/doi/abs/10.1177/001452464305400806),” *The Expository Times* 54, no. 8 (1943): 219–222; and “[Important Hypotheses Reconsidered: I. The Proto-Luke Hypothesis](https://journals.sagepub.com/doi/abs/10.1177/001452465506700104),” *The Expository Times* 67, no. 1 (1955): 12–16.
[^13]: For the principle that Luke generally preserves more primitive wording than Matthew, see Joachim Jeremias, [The Sermon on the Mount](https://archive.org/details/sermononmount0000jere), trans. Norman Perrin (London: Athlone Press, 1961), 17–18; and John S. Kloppenborg, [Excavating Q: The History and Setting of the Sayings Gospel](https://archive.org/details/excavatingqhisto0000klop/page/n5/mode/2up) (Edinburgh: T&T Clark, 2000), 11–111. The position remains widely held by scholars who accept Q; see Bart D. Ehrman, “[The Q Source Used by Matthew and Luke](https://ehrmanblog.org/the-q-source-used-by-matthew-and-luke/),” The Bart Ehrman Blog, [The Q Source Used by Matthew and Luke](https://ehrmanblog.org/the-q-source-used-by-matthew-and-luke/), affirming that Luke is more likely than Matthew to have preserved Q’s original sequence.
[^14]: Frans van Segbroeck, [The Minor Agreements of Matthew and Luke against Mark: With a Cumulative List](https://openlibrary.org/books/OL5255778M/The_Minor_agreements_of_Matthew_and_Luke_against_Mark) (Leuven: University Press, 1974); B. H. Streeter, [The Four Gospels: A Study of Origins](https://archive.org/details/bwb_C0-BOB-469) (London: Macmillan, 1924; 4th rev. impression, 1930), 293–331.
[^15]: The classic formulation is B. H. [Streeter, The Four Gospels: A Study of Origins](https://archive.org/details/bwb_C0-BOB-469) 4th Ed.(London: Macmillan, 1930), 293–331. For the contemporary case on behalf of Q, see Kloppenborg, [Excavating Q](https://archive.org/details/excavatingqhisto0000klop/page/n5/mode/2up), and Tuckett, *Q and the History of Early Christianity*.
[^16]: Mark Goodacre, [The Case Against Q: Studies in Markan Priority and the Synoptic Problem](https://www.markgoodacre.org/Q/) (Harrisburg, PA: Trinity Press International, 2002), esp. ch. 7.
[^17]: Robert L. Lindsey, “[A New Approach to the Synoptic Gospels](https://jerusalemperspective.com/11492/),” *Jerusalem Perspective* (2014), [A New Approach to the Synoptic Gospels,](https://jerusalemperspective.com/11492/), drawing on Lindsey, “A Modified Two-Document Theory of the Synoptic Dependence and Interdependence,” *Novum Testamentum* 6 (1963): 239–263. The Cross-Factor is developed at greater length in Robert L. Lindsey, *A Hebrew Translation of the Gospel of Mark*, 2nd ed. (Jerusalem: Dugith Publishers, 1973), 19–22, 41.
[^18]: For the lexical analysis of εὐαγγέλιον across the Synoptic Gospels and Acts, see Lindsey, “[A New Approach to the Synoptic Gospels,](https://www.jerusalemperspective.com/11492/)” cited above. The seven Markan occurrences are at Mark 1:1, 1:14, 1:15, 8:35, 10:29, 13:10, and 14:9; the four Matthean occurrences are at Matt 4:23, 9:35, 24:14, and 26:13; the two Acts occurrences are at Acts 15:7 and 20:24.
[^19]: James D. G. Dunn, *Neither Jew nor Greek: A Contested Identity*, Christianity in the Making, vol. 3 (Grand Rapids, MI: Eerdmans, 2015), 375–377.
[^20]: Dunn, *Neither Jew nor Greek*, 368–377, on Paul’s introduction of εὐαγγέλιον into Christian vocabulary and Mark’s transition from “gospel” to “Gospel” at Mark 1:1. The same observation was independently advanced by Robert L. Lindsey, who noted that, since there is no Hebrew or Aramaic equivalent for τὸ εὐαγγέλιον, the author of Mark must have picked up the term from Acts and/or the Pauline epistles. The systematic empirical catalog is Joshua N. Tilton and David N. Bivin, “[LOY Excursus: Catalog of Markan Stereotypes and Possible Markan Pick-ups](https://jerusalemperspective.com/13743/),” Jerusalem Perspective, updated 16 January 2026, [LOY Excursus: Catalog of Markan Stereotypes and Possible Markan Pick-ups,](https://jerusalemperspective.com/13743/), summarizing Lindsey’s analysis.
[^21]: B. H. Streeter, [The Four Gospels: A Study of Origins](https://archive.org/details/bwb_C0-BOB-469/) (London: Macmillan, 1924; 4th rev. impression, 1930), 159–160 and 169, on the foundational verse counts: Mark contains 661 verses, of which Matthew reproduces “the substance of all but 55 verses” (Streeter, 169) and Luke exhibits “about 350 verses (i.e. just over one half of Mark)” (Streeter, 160). The figure of approximately 280 verses for Markan material in Matthew but not Luke is derived by subtraction: of the 55 Markan verses Matthew omits, Streeter notes that 24 are found in Luke (p. 169), so the triple tradition stands at approximately 326 verses and Matthew’s Markan content outside Luke at approximately 280.
[^22]: For the extended companion analysis, see [IssuesWithMark.com,](https://issueswithmark.com/) and [IssuesWithMatthew](https://issueswithmatthew.com/), especially “[Progressive Embellishment, Luke→Mark→Matthew](https://issueswithmark.com/progressive-embellishment-luke-mark-matthew/),” “[Embellishments of Mark](https://issueswithmark.com/embellishments-of-mark/),” “[Mark the ‘Re-Write Man](https://issueswithmark.com/mark-the-re-write-man/)’,” and “[Matthean Revision to Mark](https://issueswithmatthew.com/matthean-revision-to-mark/)”; see also the Gospel parallels tool at [The Gospel of Luke: AI Critical Edition](https://corebible.app/).
[^23]: Foreword to Robert L. Lindsey, [A Hebrew Translation of the Gospel of Mark](https://jerusalemperspective.com/product/a-hebrew-translation-of-the-gospel-of-mark/) (Jerusalem: Dugith Publishers, 1973), [A Hebrew Translation of the Gospel of Mark](https://jerusalemperspective.com/product/a-hebrew-translation-of-the-gospel-of-mark/).
[^24]: David Flusser, “Foreword to Robert Lindsey’s [A Hebrew Translation of the Gospel of Mark](https://jerusalemperspective.com/11787/),” Jerusalem Perspective, [A Hebrew Translation of the Gospel of Mark](https://jerusalemperspective.com/11787/).
[^25]: Robert L. Lindsey, [A Hebrew Translation of the Gospel of Mark](https://archive.org/details/a-hebrew-translation-of-the-gospel-of-mark/page/n3/mode/2up), 2nd ed. (Jerusalem: Dugith Publishers, 1973), 5, foreword by David Flusser, [Robert Lisle Lindsey, A Hebrew translation of the Gospel of Mark, Jerusalem, Dugith Publishers, 2nd edition, (1973)](https://archive.org/details/a-hebrew-translation-of-the-gospel-of-mark/page/n3/mode/2up).
[^26]: David Flusser, “Foreword to Robert Lindsey’s [A Hebrew Translation of the Gospel of Mark](https://jerusalemperspective.com/11787/),” Jerusalem Perspective, [A Hebrew Translation of the Gospel of Mark](https://jerusalemperspective.com/11787/), citing Paul Winter, *On the Trial of Jesus* (Berlin: Walter de Gruyter, 1961), 28.
[^27]: David N. Bivin, “[Evidence of an Editor’s Hand in Two Instances of Mark’s Account of Jesus” Last Week?](https://jerusalemperspective.com/4307/)’ Jerusalem Perspective, [Evidence of an Editor’s Hand in Two Instances of Mark’s Account of Jesus” Last Week?](https://jerusalemperspective.com/4307/), originally published in *Jesus’ Last Week: Jerusalem Studies in the Synoptic Gospels*, vol. 1, ed. R. S. Notley, M. Turnage, and B. Becker (Leiden: Brill, 2006), 211–224.
[^28]: Bivin, “[Evidence of an Editor’s Hand](https://jerusalemperspective.com/4307/),” section on “For they no longer dared to ask him another question.” The Flusser observation is reported by Bivin as personal communication.
[^29]: Gottlob Christian Storr, *On the Purpose of the Gospel History [trans. mine]* (1786), surveyed in Robert MacEwen, [Matthean Posteriority: An Exploration of Matthew’s Use of Mark and Luke as a Solution to the Synoptic Problem](https://archive.org/details/mattheanposterio0000mace)*, *Library of New Testament Studies 501 (London: Bloomsbury T&T Clark, 2015), 7.
[^30]: Johann Gottfried Herder, *On the Redeemer of Mankind according to Our First Three Gospels [trans. mine]* (1796), surveyed in MacEwen, [Matthean Posteriority](https://archive.org/details/mattheanposterio0000mace), 7–8.
[^31]: Christian Gottlob Wilke, *The Proto-Evangelist [trans. mine]* (1838), surveyed in MacEwen, [Matthean Posteriority](https://archive.org/details/mattheanposterio0000mace), 8.
[^32]: Gustav Schlager, “The Dependence of the Gospel of Matthew on the Gospel of Luke *[trans. mine]*” (article details from the late nineteenth century, surveyed in MacEwen, [Matthean Posteriority](https://archive.org/details/mattheanposterio0000mace), 9).
[^33]: See above, “Lukan Priority in Modern Scholarship,” for the citation of Lockton’s “[The Origin of the Gospels](https://archive.org/details/churchquarterly14unkngoog/page/n227/mode/2up),” *Church Quarterly Review* 94 (1922): 216–239, and the lineage from Lockton through Lindsey and the Jerusalem School. Lockton developed the Luke → Mark → Matthew case further in William Lockton, [The Resurrection and Other Gospel Narratives](https://archive.org/details/resurrectionothe00lock)(London: Longmans, Green, 1924), 131–133, and William Lockton, [The Three Traditions in the Gospels](https://archive.org/details/threetraditionsi00lock_0) (London: Longmans, Green, 1926), 3–4, 32–37.
[^34]: Ernst von Dobschütz, “Matthew as Rabbi and Catechist,” in *The Interpretation of Matthew*, ed. Graham Stanton, Studies in New Testament Interpretation, 2nd ed. (Edinburgh: T&T Clark, 1995), 34–37. Surveyed in MacEwen, [Matthean Posteriority](https://archive.org/details/mattheanposterio0000mace), 12–15.
[^35]: B.H. Streeter, [The Four Gospels: A Study of Origins](https://archive.org/details/bwb_C0-BOB-469/) (London: Macmillan, 1924; rev. 4th impression, 1930), 183, 199–222, 273–92. Streeter held that Luke generally preserves the original order of the double tradition material better than Matthew, who scatters it across his great discourses (pp. 273–75); that the wording exhibits “alternating primitivity,” such that, in his own words, “sometimes it is Matthew, sometimes it is Luke, who gives a saying in what is clearly the more original form” (p. 183); and that this alternating primitivity in wording was itself an argument *for* Q, since neither evangelist’s version could plausibly be derived directly from the other (p. 183). Streeter’s “Proto-Luke” (chap. VIII, pp. 199–222) is a stage between Q and canonical Luke (Q + L = Proto-Luke), not a source posited for Matthew. It should be noted, however, that Streeter offered no sustained argument against Matthew’s direct use of Luke. His two stated reasons for rejecting direct dependence in either direction (p. 183) are (1) the “crank” argument, which targets only the hypothesis of Luke’s use of Matthew (since it depends on Luke’s hypothetical tearing of material out of Matthew’s contexts), and (2) the alternating primitivity argument, which, as later critics have observed, only establishes the presence of a shared source rather than ruling out direct dependence alongside that source. See Alan Garrow, “Streeter’s ‘Other’ Synoptic Solution: The Matthew Conflator Hypothesis,” *New Testament Studies* 62, no. 2 (2016): 207–26, esp. 208–11, who identifies this as “Streeter’s Unfinished Argument” and notes that subsequent Q-defenders have likewise declined to mount a positive argument against Matthean Posteriority.
[^36]: H. Philip West Jr., “[A Primitive Version of Luke in the Composition of Matthew](https://doi.org/10.1017/S0028688500018518),” *New Testament Studies* 14, no. 1 (1967-68): 75–95. Surveyed in MacEwen, [Matthean Posteriority](https://archive.org/details/mattheanposterio0000mace), 15–17.
[^37]: Ronald V. Huggins, “[Matthean Posteriority: A Preliminary Proposal,”](https://www.academia.edu/18603110/Matthean_Posteriority_A_Preliminary_Proposal)*Novum Testamentum* 34, no. 1 (1992): 1–22.
[^38]: Evan Powell, [The Unfinished Gospel: Notes on the Quest for the Historical Jesus](https://archive.org/details/unfinishedgospel0000powe) (Westlake Village, CA: Symposium Books, 1994); Evan Powell, *The Myth of the Lost Gospel* (Las Vegas: Symposium Books, 2006), especially 50, 107–120.
[^39]: George A. Blair, *The Synoptic Gospels Compared*, Studies in the Bible and Early Christianity 55 (Lewiston, NY: Edwin Mellen Press, 2003), surveyed in MacEwen, [Matthean Posteriority](https://archive.org/details/mattheanposterio0000mace), 7–24.
[^40]: Alan Garrow, “[Streeter’s ‘Other’ Synoptic Solution: The Matthew Conflator Hypothesis](https://www.cambridge.org/core/journals/new-testament-studies/article/abs/streeters-other-synoptic-solution-the-matthew-conflator-hypothesis/84D113C8F0CE6E561A9D80442E47E34E),” *New Testament Studies* 62, no. 2 (2016): 207–226. Companion article: Alan Garrow, “An Extant Instance of ‘Q’,” *New Testament Studies* (2016).
[^41]: James R. Edwards, *The Hebrew Gospel and the Development of the Synoptic Tradition* (Grand Rapids: Eerdmans, 2009), 243–262.
[^42]: Martin Hengel, [The Four Gospels and the One Gospel of Jesus Christ: An Investigation of the Collection and Origin of the Canonical Gospels](https://archive.org/details/fourgospelsonego0000heng), trans. John Bowden (Harrisburg, PA: Trinity Press International, 2000).
[^43]: For Hengel’s view as cited by deSilva, see David A. deSilva, [An Introduction to the New Testament: Contexts, Methods and Ministry Formation](https://www.google.com/books/edition/An_Introduction_to_the_New_Testament/ckBvDwAAQBAJ) 2nd Ed. (InterVarsity, 2018), 137. The thesis of Matthean Posteriority is developed at length in Josiah E. Verkaik, “[Evidence of Matthean Posteriority](https://issueswithmatthew.com/matthean-posteriority/)” and “[Survey of Matthean Posteriority Scholarship](https://issueswithmatthew.com/matthean-posteriority-scholarship/),” Issues with Matthew, [https:isseswithmatthew.com](https://isseswithmatthew.com/), which the present subsection summarizes.
[^44]: Robert MacEwen, [Matthean Posteriority: An Exploration of Matthew’s Use of Mark and Luke as a Solution to the Synoptic Problem](https://archive.org/details/mattheanposterio0000mace), Library of New Testament Studies 501 (London: Bloomsbury T&T Clark, 2015).
[^45]: MacEwen, [Matthean Posteriority](https://archive.org/details/mattheanposterio0000mace), 26.
[^46]: Evan Powell, *The Myth of the Lost Gospel* (Symposium Books, 2006), 41–43.
[^47]: For these and other examples of Matthean revision to Luke, see Powell, *The Myth of the Lost Gospel*, 69–94.
[^48]: For the conflation analysis with full text of the parallel comparisons, see Powell, *The Myth of the Lost Gospel*, 83–94, and Verkaik, “Evidence of Matthean Posteriority,” cited above.
[^49]: The detailed textual analysis of these Lukan-source indicators is given in Verkaik, “Evidence of Matthean Posteriority,” cited above.
[^50]: For comprehensive analysis of Matthew’s Judaizing features, see Josiah E. Verkaik, “[Matthew is a Judaizing Document](https://lukanpriority.com/matthew-is-judaizing/),” Lukan Priority, [Matthew is a Judaizing Document.](https://lukanpriority.com/matthew-is-judaizing/).
[^51]: Gustaf Dalman, [The Words of Jesus](https://archive.org/details/wordsjesusconsi00kaygoog/page/n8/mode/2up) (Edinburgh: T&T Clark, 1902), 38.
[^52]: James R. Edwards, *The Hebrew Gospel and the Development of the Synoptic Tradition* (Grand Rapids: Eerdmans, 2009), xx.
[^53]: Eusebius, [Ecclesiastical History](https://archive.org/details/eusebius-ecclesiastical-history-loeb) 3.25.
[^54]: James R. Edwards, *The Hebrew Gospel and the Development of the Synoptic Tradition* (Grand Rapids: Eerdmans, 2009), 103.
[^55]: J. R. Davila, “[How Can We Tell If a Greek Apocryphon or Pseudepigraphon Has Been Translated from Hebrew or Aramaic?](https://journals.sagepub.com/doi/10.1177/09518207057767)” *Journal for the Study of the Pseudepigrapha* 15, no. 1 (2005): 38–39, [(How) Can We Tell If a Greek Apocryphon or Pseudepigraphon Has Been Translated from Hebrew or Aramaic](https://journals.sagepub.com/doi/10.1177/09518207057767).
[^56]: James R. Edwards, *The Hebrew Gospel and the Development of the Synoptic Tradition* (Grand Rapids: Eerdmans, 2009), 153.
[^57]: Epiphanius, *Panarion*, 51.11.6.
[^58]: Eusebius, [Ecclesiastical History](https://archive.org/details/eusebius-ecclesiastical-history-loeb) 6.14.2–3, trans. J. E. L. Oulton, Loeb Classical Library 265 (Cambridge, MA: Harvard University Press, 1932), 47.
[^59]: Halvor Ronning, “[A Statistical Approach to the Synoptic Problem: Part 1](https://jerusalemperspective.com/15295/),” *Jerusalem Perspective*, [A Statistical Approach to the Synoptic Problem: Part 1—Triple Tradition](https://jerusalemperspective.com/15295/).
[^60]: Halvor Ronning, “[A Statistical Approach to the Synoptic Problem: Part 2](https://jerusalemperspective.com/15304/),” *Jerusalem Perspective*, [A Statistical Approach to the Synoptic Problem: Part 2—Double Tradition,](https://jerusalemperspective.com/15304/).
[^61]: Halvor Ronning, “[A Statistical Approach to the Synoptic Problem: Part 3](https://jerusalemperspective.com/15475/),” *Jerusalem Perspective*, [A Statistical Approach to the Synoptic Problem: Part 3—Single Tradition](https://jerusalemperspective.com/15475/).
[^62]: Halvor Ronning, “[A Statistical Approach to the Synoptic Problem: Part 4](https://jerusalemperspective.com/15729/),” *Jerusalem Perspective*, [A Statistical Approach to the Synoptic Problem: Part 4—Non-Linear Hypotheses](https://jerusalemperspective.com/15729/).
[^63]: Halvor Ronning, “[A Statistical Approach to the Synoptic Problem: Part 4](https://jerusalemperspective.com/15729/),” *Jerusalem Perspective*, [A Statistical Approach to the Synoptic Problem: Part 4—Non-Linear Hypotheses](https://jerusalemperspective.com/15729/).
[^64]: On the general unreliability of patristic testimony for reconstructing gospel origins and canon formation, see Walter Bauer, *Orthodoxy and Heresy in Earliest Christianity* (Philadelphia: Fortress Press, 1971); Helmut Koester, [Ancient Christian Gospels:](https://archive.org/details/ancientchristian0000koes) [Their History and Development](https://archive.org/details/ancientchristian0000koes) (London: SCM Press, 1990).
[^65]: Eusebius, [Ecclesiastical History](https://archive.org/details/eusebius-ecclesiastical-history-loeb) 3.39.15, trans. Kirsopp Lake, Loeb Classical Library 153 (Cambridge, MA: Harvard University Press, 1926), 297.
[^66]: Irenaeus, [Against Heresies](https://archive.org/details/antenicenefather01robe), 3.1.1.
[^67]: Eusebius, [Ecclesiastical History](https://archive.org/details/eusebius-ecclesiastical-history-loeb) 6.14.6–7, trans. J. E. L. Oulton, Loeb Classical Library 265 (Cambridge, MA: Harvard University Press, 1932), 49.
[^68]: Jerome, [On Illustrious Men](https://archive.org/details/nicenepostnicene0003unse_l8w3), 8.
[^69]: Eusebius, [Ecclesiastical History](https://archive.org/details/eusebius-ecclesiastical-history-loeb) 3.39.13, trans. Kirsopp Lake, Loeb Classical Library 153 (Cambridge, MA: Harvard University Press, 1926), 297.
[^70]: J. B. Lightfoot, [The Apostolic Fathers, Part 2: S. Ignatius, S. Polycarp](https://archive.org/details/apostolicfathep2v1clemuoft), 3 vols. (London: Macmillan, 1889). Lightfoot’s critical edition established the Middle recension as the scholarly consensus and documented the fourth-century character of the Long recension expansions and pseudepigraphical letters.
[^71]: Jerome, [On Illustrious Men](https://archive.org/details/nicenepostnicene0003unse_l8w3), 3.
[^72]: Josiah E. Verkaik, “[The Hebrew Gospel and Luke](https://lukanpriority.com/hebrew-gospel/),” Luke Primacy, [The Hebrew Gospel and Luke.](https://lukanpriority.com/hebrew-gospel/).
[^73]: Lee Martin McDonald, *The Biblical Canon: Its Origin, Transmission, and Authority* (Grand Rapids, MI: Baker Academic, 2017), 252–253. McDonald observes that nonapostolic works are unlikely to have survived antiquity if produced in a period when apostolic naming was the established route to credibility; the inference runs that Mark and Luke had already attained authority before that pressure took hold.