Deep Bible
Back to Questions
Deep dive · 13 min read
Scholars debate

Did Q exist?

Matthew and Luke share roughly 235 verses of teaching material, almost word for word, that does not appear in Mark. Most academic introductions explain this with a lost sayings source called Q. A long minority tradition argues no such source ever existed. The Gospel of Thomas, found in 1945, made the question harder to settle.

What's at stake

Read Mark, Matthew, and Luke side by side and a pattern emerges. The three share extensive material in similar wording. Matthew and Luke share Mark's content plus a large body of teaching (the Lord's Prayer, the Beatitudes, the Sermon on the Mount sayings, parables like the lost sheep) that Mark does not have. Since Heinrich Julius Holtzmann in 1863, the standard academic explanation has been the Two-Source hypothesis: Mark is the earliest Gospel, and Matthew and Luke each independently used Mark plus a sayings collection called Q (from German Quelle, 'source'). The Two-Source hypothesis has been the mainstream view in academic New Testament study for over 150 years. But it rests on the existence of a document for which there is no physical evidence. No manuscript of Q has been found. No early Christian writer names it. Q is, at the textual level, an inference. The question is whether that inference is well supported, and whether better alternatives exist.

What the Synoptic Problem is

Mark, Matthew, and Luke are called the Synoptic Gospels because their material can be set side by side (syn-optic, 'seen together') and read in parallel. A standard Synoptic chart sorts the verses into four categories. Triple tradition: material in all three Gospels (about 350 verses). Double tradition: material in Matthew and Luke but not Mark (about 235 verses). Material unique to Matthew (the special M verses, about 330). Material unique to Luke (the special L verses, about 565). The triple tradition is mostly narrative. The Matthew-Luke double tradition is mostly teaching: sayings, parables, the long Sermon material, exorcism debates, woes against the cities of Galilee.

The double tradition is the heart of the Synoptic Problem. The wording is often nearly identical across Matthew and Luke. The same sayings appear in similar form. But the order is different. Matthew gathers Jesus's teaching into five long discourses (the Sermon on the Mount in 5-7, the Mission Discourse in 10, the Parables Discourse in 13, the Community Discourse in 18, the Olivet Discourse in 24-25). Luke distributes the same material across his travel narrative (chapters 9-19), with the central Sermon on the Plain in chapter 6. If Matthew and Luke knew the same teaching, they organized it very differently.

There are also small data points that have to be explained. The minor agreements: places where Matthew and Luke agree against Mark in wording or detail, even within triple-tradition material. The Sermon on the Mount versus the Sermon on the Plain: the same Beatitudes opening, similar central body, different framing and length. Luke's order: Luke often departs from Mark's sequence of episodes, raising the question of where his alternate order came from. Each position has to read the same data and offer a coherent account.

Four positions on the Synoptic Problem

The four hypotheses

Two-Source has been the academic mainstream since 1863. The Farrer hypothesis is the leading challenger. The Two-Gospel hypothesis is a minority revival of an earlier tradition. The oral-tradition position treats the whole question as smaller than the others assume.

Mark is the earliest Gospel. Matthew and Luke each independently used Mark and a second, lost source called Q, which contained the sayings of Jesus they share. The Two-Source hypothesis has been the academic mainstream since the late nineteenth century.
Held by
  • Heinrich Julius Holtzmann, Die synoptischen Evangelien (1863), early formulation
  • Bernhard Weiss, Die Quellen der synoptischen Überlieferung (1908)
  • B. H. Streeter, The Four Gospels: A Study of Origins (Macmillan, 1924), classic English formulation
  • Joseph A. Fitzmyer, 'The Priority of Mark and the Q Source in Luke,' in Jesus and Man's Hope (1970)
  • John S. Kloppenborg, Excavating Q (Fortress, 2000)
  • James M. Robinson, Paul Hoffmann, and John S. Kloppenborg, The Critical Edition of Q (Fortress, 2000)
  • Dale C. Allison, The Jesus Tradition in Q (Trinity Press International, 1997)
  • Christopher Tuckett, Q and the History of Early Christianity (T&T Clark, 1996)
Evidence
  • Markan priority explains the verbal and structural agreements of all three Synoptics where they overlap
  • The double tradition (Matt-Luke material absent from Mark) requires a source other than Mark; if Matthew and Luke did not know each other, that source is a separate document
  • Matthew and Luke never agree in the order of double-tradition material against Mark's order in a way that implies one used the other; this is the strongest argument for their independence
  • The Gospel of Thomas, found at Nag Hammadi in 1945, is a sayings-only gospel; its existence proves that sayings collections of the kind Q is reconstructed to have been did circulate in early Christianity
  • The 'doublets' (the same saying appearing twice in Matthew or Luke) suggest the evangelists drew from two sources, Mark and Q, sometimes preserving both versions
  • Reconstructed Q, when arranged as a sayings collection, shows a recognizable wisdom-prophetic theological profile that fits its proposed historical setting in early Jewish-Christian communities
Challenges
  • No manuscript of Q has ever been found, and no early Christian writer mentions it by name
  • The minor agreements of Matthew and Luke against Mark (small wording overlaps within triple tradition) are difficult to explain on strict independence and have to be attributed to coincidence, textual corruption, or overlapping oral tradition
  • Reconstructing Q from the Matt-Luke overlap requires deciding which version preserves Q's wording; that decision is contested at hundreds of points
  • The Sermon on the Plain (Luke 6) and the Sermon on the Mount (Matt 5-7) share a structure but differ enough that scholars debate whether Q itself contained a 'sermon' or whether each evangelist constructed one
  • Q's reconstruction depends on assuming Matthew and Luke each preserved most of it; if either omitted material substantially, Q is bigger than the reconstruction shows and the case becomes harder to test

How each position reads the four central data points

The four data points

Each position has to read the same four pieces of evidence. The double tradition itself, the minor agreements, Luke's order, and the Sermon on the Mount versus the Sermon on the Plain.

Two-Source
Double tradition (~235 verses)
Matthew and Luke independently drew from Q, a written sayings collection. Verbal agreement reflects shared written wording; differences reflect each evangelist's editing.
Minor agreements
Small Matt-Luke agreements against Mark are attributed to coincidence, scribal harmonization in the manuscript tradition, or overlapping oral tradition. This is the position's weakest point.
Luke's order
Luke followed Q's order for double-tradition material, which often differs from Matthew's. The contrast in arrangement is evidence Luke did not know Matthew.
Sermon on the Mount vs Plain
Q contained a 'sermon' that Matthew expanded into the Sermon on the Mount and Luke shortened into the Sermon on the Plain. Both preserve different selections of the same Q material.
Farrer-Goulder
Double tradition (~235 verses)
Luke read Matthew and selected the teaching material he wanted. Verbal agreement is direct literary dependence. Differences reflect Luke's editorial choices.
Minor agreements
Direct evidence that Luke knew Matthew. The position handles the minor agreements as the strength rather than the weakness of the case.
Luke's order
Luke restructured Matthew's five discourses into the travel-narrative format. Goulder argues this is consistent with Luke's general editorial method.
Sermon on the Mount vs Plain
Luke abridged Matthew's Sermon on the Mount into the Sermon on the Plain. The selections Luke made align with his theological emphases on the poor and on reversal.
Two-Gospel
Double tradition (~235 verses)
Luke used Matthew; Mark later abridged both. The double tradition is what Mark left out when he picked common narrative material from his two predecessors.
Minor agreements
Direct: Matthew and Luke share wording because Luke used Matthew; Mark sometimes preferred Luke's version, leaving Matt-Luke wording aligned against him.
Luke's order
Luke reorganized Matthew. Mark followed Matthew's order where he overlapped with it, sometimes departing to follow Luke.
Sermon on the Mount vs Plain
Luke abridged Matthew's sermon. Mark omitted the sermon material altogether in his abridgment.
Oral tradition
Double tradition (~235 verses)
Matthew and Luke drew on a shared, partly oral, partly written tradition of Jesus's sayings. 'Q' is shorthand for this shared tradition, not a single text.
Minor agreements
Attributable to the fluidity of oral tradition and to overlapping written collections that no longer survive.
Luke's order
Luke arranged the tradition for his own narrative purposes. Order was less fixed in oral transmission than in a written source.
Sermon on the Mount vs Plain
Each evangelist composed a sermon from sayings that circulated in multiple forms. There was no single 'Sermon Q.'

Timeline of the question

The Synoptic Problem has been studied continuously since the late eighteenth century. Q entered the conversation in 1838 and dominated it after 1863.

Patristic and early modern
Modern academic debate
400 CE
Augustine, De Consensu Evangelistarum
Augustine treats Matthew as first, Mark as Matthew's epitomizer, Luke as third. This is the traditional patristic order, preserved by the modern Two-Gospel hypothesis.
0% along range
1789 CE
Griesbach's hypothesis
Johann Jakob Griesbach proposes that Matthew was first, Luke second, and Mark abridged both. The modern Two-Gospel hypothesis revives this proposal.
86% along range
1838 CE
Christian Hermann Weisse proposes Markan priority and a lost sayings source
Weisse and Christian Gottlob Wilke independently argue for Markan priority. Weisse introduces a sayings source ('Logia') alongside Mark, the seed of the Q hypothesis.
89% along range
1863 CE
Holtzmann formalizes the Two-Source hypothesis
Heinrich Julius Holtzmann's Die synoptischen Evangelien gives the Two-Source model its classic form. Markan priority plus a sayings source becomes the dominant academic view.
90% along range
1908 CE
Bernhard Weiss reconstructs Q
Weiss publishes Die Quellen der synoptischen Überlieferung. The first major attempt to reconstruct Q's content from the double tradition.
93% along range
1924 CE
Streeter's Four Gospels
B. H. Streeter formulates the Two-Source model into a Four-Source model (Mark, Q, M, L). The standard English-language version of the academic mainstream.
94% along range
1945 CE
Gospel of Thomas discovered at Nag Hammadi
A sayings-only gospel of about 114 logia. Its existence proves that sayings collections of the Q type did circulate; this strengthens the Two-Source case but does not by itself confirm Q.
95% along range
1955 CE
Farrer's 'On Dispensing with Q'
Austin Farrer proposes that Luke read Matthew directly. The modern challenge to Q begins.
96% along range
1964 CE
Farmer's Synoptic Problem
William R. Farmer revives the Griesbach hypothesis in modern form. The Two-Gospel hypothesis enters the academic conversation.
96% along range
1989 CE
Goulder's Luke: A New Paradigm
Michael Goulder develops the Farrer hypothesis in a full two-volume Luke commentary. The Farrer-Goulder position becomes the leading alternative to Q.
98% along range
2000 CE
Critical Edition of Q published
Robinson, Hoffmann, and Kloppenborg publish a full reconstruction of Q's text. The Two-Source position offers a reconstructed document for the first time at scholarly book length.
99% along range
2002 CE
Goodacre's Case Against Q
Mark Goodacre presents the Farrer hypothesis as a positive case rather than a negative critique. The leading modern statement of the no-Q position.
99% along range
2003 CE
Dunn's Jesus Remembered
James Dunn argues that oral tradition explains the Matt-Luke overlap, treating the Q hypothesis as too literary. The oral-tradition position gains modern academic standing.
99% along range
2021 CE
Eve's Solving the Synoptic Puzzle
Eric Eve restates the Farrer hypothesis with refinements. The Q debate continues to be active in publication.
100% along range

The Gospel of Thomas and what it does and does not show

The Gospel of Thomas was found in 1945 at Nag Hammadi in Upper Egypt, in a jar of Coptic manuscripts buried in the fourth century. The Thomas text contains 114 sayings of Jesus, with almost no narrative framework. The sayings are arranged loosely, sometimes thematically, with the formula 'Jesus said' opening most entries. The text has no birth narrative, no passion, no resurrection, no journey to Jerusalem. It is a pure sayings collection.

For the Two-Source position, Thomas is a major confirmation. The objection that sayings collections like Q are an inventor's fiction loses force once one has a real example. The Q hypothesis predicted that early Christian groups produced sayings collections; Thomas shows they did. Kloppenborg has argued in Excavating Q that Thomas's structure (sayings introduced by 'Jesus said,' grouped by topic, no narrative) is the closest surviving parallel to what Q is reconstructed to have been.

The Farrer position responds that Thomas proves only the genre, not the specific document. Goodacre argues that the existence of one sayings collection in fourth-century Coptic does not establish that the Matt-Luke double tradition came from a comparable first-century document; it shows only that the form was possible. The Two-Gospel and oral-tradition positions read Thomas in their own ways. The Two-Gospel reading treats Thomas as later than the Synoptics and dependent on them. The oral-tradition reading treats it as evidence that sayings of Jesus circulated in multiple collections, of which Q is one possible candidate but not the only or the most likely.

Where the conversation sits now

The Two-Source hypothesis remains the working assumption in the majority of academic introductions and commentaries. Kloppenborg's Excavating Q, Tuckett's Q and the History of Early Christianity, and the Critical Edition of Q are the standard reference works for the position. Most graduate New Testament training in North America and Europe presents Q as the default Synoptic model.

The Farrer-Goulder hypothesis has gained ground over the last forty years, particularly through Goodacre's accessible defense in The Case Against Q. The position is no longer marginal; it is treated as a live alternative in current Synoptic scholarship. The Two-Gospel hypothesis has receded from the prominence it held in the 1970s and 1980s but maintains a small body of defenders, primarily through the Research Team led by Farmer and continued by Peabody, Cope, and McNicol. The oral-tradition position cuts across the others and continues to attract scholars (Dunn, Bauckham, Mournet) who treat the document-versus-document framing of the question as too narrow.

Each position has to account for the same data and has its specific challenges. The Two-Source position has to explain the minor agreements and the absence of any direct evidence for Q. The Farrer position has to explain why Luke restructured Matthew so freely if he had Matthew in front of him. The Two-Gospel position has to explain why Mark would abridge two existing Gospels into the rougher, shorter text we have. The oral-tradition position has to explain the high verbal agreement at points where Matt and Luke align nearly word for word.

Reading the Synoptic Problem with the question open means reading Matthew, Mark, and Luke knowing that the relationship between them is contested at the academic level and that the contest has been live for over two centuries. Q may or may not have existed as a document. Some form of shared tradition between Matthew and Luke certainly did. The relationship between the three Gospels is real, the data are stable, and the four positions are responses to that data rather than starting points for it.

Sources

Primary sources
  • The Gospel of Mark, Matthew, Luke (NA28; SBL Greek New Testament)
  • The Gospel of Thomas, Coptic and Greek fragments (Nag Hammadi Codex II.2; P. Oxy. 1, 654, 655)
  • Augustine, De Consensu Evangelistarum, books 1-4 (CCSL; PL 34)
  • Papias of Hierapolis, fragments preserved in Eusebius, Ecclesiastical History 3.39
  • Irenaeus, Adversus Haereses 3.1.1 (on the four Gospels)
  • Eusebius, Ecclesiastical History 3.24 (on the order of the Gospels)
Modern scholarship cited
  • Johann Jakob Griesbach, Commentatio qua Marci Evangelium totum e Matthaei et Lucae commentariis decerptum esse monstratur (1789)
  • Christian Hermann Weisse, Die evangelische Geschichte (Leipzig, 1838)
  • Heinrich Julius Holtzmann, Die synoptischen Evangelien (Leipzig, 1863)
  • Bernhard Weiss, Die Quellen der synoptischen Überlieferung (Leipzig, 1908)
  • B. H. Streeter, The Four Gospels: A Study of Origins (Macmillan, 1924)
  • Austin Farrer, 'On Dispensing with Q,' in D. E. Nineham, ed., Studies in the Gospels (Blackwell, 1955)
  • Joseph A. Fitzmyer, 'The Priority of Mark and the Q Source in Luke,' in Jesus and Man's Hope vol. 1 (1970)
  • William R. Farmer, The Synoptic Problem: A Critical Analysis (Macmillan, 1964; 2nd ed. Mercer, 1976)
  • Werner Kelber, The Oral and the Written Gospel (Fortress, 1983)
  • Michael Goulder, Luke: A New Paradigm, 2 vols. (Sheffield Academic, 1989)
  • Graham Stanton, A Gospel for a New People: Studies in Matthew (T&T Clark, 1992)
  • Christopher M. Tuckett, Q and the History of Early Christianity (T&T Clark, 1996)
  • Dale C. Allison, The Jesus Tradition in Q (Trinity Press International, 1997)
  • Allan J. McNicol, David L. Dungan, and David B. Peabody, eds., Beyond the Q Impasse: Luke's Use of Matthew (Trinity Press International, 1996)
  • David L. Dungan, A History of the Synoptic Problem (Doubleday, 1999)
  • John S. Kloppenborg, Excavating Q (Fortress, 2000)
  • James M. Robinson, Paul Hoffmann, and John S. Kloppenborg, The Critical Edition of Q (Fortress, 2000)
  • Mark Goodacre, The Synoptic Problem: A Way Through the Maze (T&T Clark, 2001)
  • Mark Goodacre, The Case Against Q (Trinity Press International, 2002)
  • David B. Peabody, Allan J. McNicol, and Lamar Cope, One Gospel from Two: Mark's Use of Matthew and Luke (Trinity Press International, 2002)
  • James D. G. Dunn, Jesus Remembered (Eerdmans, 2003)
  • Terence C. Mournet, Oral Tradition and Literary Dependency (Mohr Siebeck, 2005)
  • Richard Bauckham, Jesus and the Eyewitnesses (Eerdmans, 2006)
  • Eric Eve, Solving the Synoptic Puzzle (Cascade, 2021)