Back to issue


This article considers the emergence of an evangelical endorsement of the Two-Source Hypothesis as a solution to the Synoptic Problem in the first half of the twentieth century. Conservative scholars such as B. B. Warfield, Geerhardus Vos, A. T. Robertson, and W. Graham Scroggie considered the hypothesis, and its concomitant Q document, to be amenable to evangelical sensibilities. Specifically, the article details how the scholars considered the Two-Source Hypothesis to be a scientific conclusion, and one that presented an early source for the life of Jesus with a high Christology.
“The two-source theory has been appropriately dethroned from the status of being an ‘assured result of scholarship.’ Nevertheless, properly nuanced, it remains the best general explanation of the data.”1

The Synoptic Problem is the term used to describe the relationships between the Gospels of Matthew, Mark, and Luke. Why are those gospels so similar in order and wording in places, and yet different in others? I devoted my doctoral thesis to exploring how early orthodox Protestant and evangelical scholars answered the Synoptic Problem over the past almost-five centuries.2 While one can hardly summarize the findings of years of that research here, it suffices to say that there has never been an orthodox or evangelical solution to the Synoptic Problem. Over the centuries, conservative authors preferred, at first, the Independence Hypothesis,3 and then the Augustinian Hypothesis,4 before the Two-Source Hypothesis5 had been proposed. However, as Carson and Moo indicate, it is clear that, though it is no longer considered a proven fact, the Two-Source Hypothesis remains the dominant solution preferred by evangelical scholars of today. This is somewhat surprising, given the fact that source criticism of the gospels was “spawned – not from a mere exegetical vantage point, as was later the case with redaction criticism – but from a desire to identify the historic Jesus. Thus, H. J. Holtzmann was compelled to conclude his book with a chapter on the life of Jesus viewed from ‘Source A,’ truncated from Mark.”6 How is it that German historical-critical scholarship regarding gospel origins came to be mainstream in English-speaking evangelical circles by the middle of the twentieth century? It was through the work of American and British evangelical leaders (some of whom had trained in Germany) interacting with (and often rejecting) German scholarship that the Two-Source Hypothesis gained popularity in churches and seminaries, as is seen below. These scholars saw in the theory a rigorous scientific explanation of the biblical data that offered a very early source for the life of Jesus that demonstrates a high Christology.

1. The Princeton School and the Synoptic Problem: Using the Two-Source
Hypothesis to Combat the Quest for the Historical Jesus

As most readers of this journal will know, Princeton was known as a bastion of conservative Presbyterian theology during the late nineteenth and early twentieth centuries. Professors such as Charles Hodge, Archibald Alexander, B. B. Warfield, A. A. Hodge, C. W. Hodge, J. Gresham Machen and Geerhardus Vos “provided intellectual foundations for defending the faith” in their scholarship in what became known as the “Old Princeton School.”7 While most of the faculty at Princeton appear not to have addressed the Synoptic Problem during that time, the approach of two Old Princetonians, B. B. Warfield and Geerhardus Vos, is considered here.

2. B. B. Warfield (1851–1921): The “Lion of Princeton” and the Two-Source Hypothesis

Benjamin Breckenridge Warfield was born in Kentucky into a wealthy family, which enabled him to attend university in Princeton, Edinburgh, Heidelberg, and Leipzig. When he returned to the United States for good in 1878, he accepted a position at Western Theological Seminary in Pennsylvania. He remained at Western until 1886, when he returned to Princeton as professor, a position he held until his death, and served as one of most formative figures in early twentieth century evangelical thought.8

Warfield never wrote a book, article, or chapter specifically dedicated to the Synoptic Problem, though he did offer his opinions on the matter on several occasions. It is impossible to say with certainty what Warfield’s exact solution to the Synoptic Problem was, but three of his publications indicate a tentative endorsement of the Two-Source Hypothesis.9 In his chapter entitled, “The Primitive Jesus,” in The Lord of Glory,10 Warfield sought to show that the picture of Jesus in the gospels was a consistent one from the very beginnings of the church. Though he certainly would not accept all of the findings of modern critical scholarship, Warfield was glad to claim that the “hypothetical sources which the several schools of criticism reconstruct for our Synoptics” each contain a clear portrait of a “supernatural Christ.”11 He noted that the theory most “in vogue” was the Two-Source Hypothesis, and without indicating his own opinion of it, worked from the presumption of that hypothesis in his arguments. The first source was Mark, or a primitive version that contained practically all of that gospel.12 If the synoptics were based on this primitive Mark, and even if it were assumed to contain only the triple tradition, it would still portray Jesus as supernatural.13 He would still be called the Christ (8:29 and 14:61–62), would still be implied to be a king (15:2, 32), Son of David (10:47–48), Lord (11:3; 12:35) and Son of God (1:11; 5:7; 9:7; 12:6–7). Moreover, there would still be details concerning his betrayal and suffering (14:20) as well as his mocking, scourging, and death (10:33). Finally, there would still be mention of his resurrection (10:34), ascension (14:62), and his return with power and glory (8:28; 13:26).14

Warfield next moved to the other source of the Two-Source Hypothesis, the hypothetical logia15 document, singling out Harnack’s reconstruction.16 Again, Warfield decided to work with its bare minimum contents, the double tradition, in which are found intimations of Jesus’s messiahship (Matt 11:3 = Luke 7:19; Matt 8:8 = Luke 7:6), his control over the destinies of people (Matt 7:21 = Luke 6:46), as well as allusions to the titles “Son of God” (Matt 4:3, 6 = Luke 4:3, 9) and the “Son of Man” (Matt 11:19 = Luke 7:34; Matt 8:20 = Luke 9:58; Matt 11:27 = Luke 10:22; Matt 16:48 = Luke 12:47). As in primitive Mark, a minimal logia still mentions a Jesus who faces betrayal and death (Matt 16:28 = Luke 12:47) and ultimately resurrection (Matt 12:40 = Luke 11:30).17 Warfield rejected critical assumptions that the evangelists created a biased image of Jesus.18 Especially absurd to Warfield was the attempt to sift through all the mythical and “high claims” of the evangelists, searching “as if for hid treasure,” for the “real Jesus.”19 Here, Warfield singled out Schmiedel’s entry in the Encyclopaedia Biblica, a work he would address at length in 1913 (see below). Ultimately, Warfield concluded that the evidence from the earliest written sources presented the same Jesus as the one found in the gospels, and attempts to draw stark distinctions had failed.20

Three years later, in 1910, Warfield published the entry for “Jesus Christ” in The New Schaff-Herzog Encyclopedia of Religious Knowledge,21 which later became a chapter in Warfield’s Christology and Criticism.22 Warfield offered many of the same arguments made in “The Primitive Jesus” with some minor adaptations. First, rather than mentioning the Two-Source Hypothesis Synoptic specifically, Warfield referred repeatedly to the one source used by all three synoptic writers and the other source Matthew and Luke had in common. Second, Warfield appeared intentionally to be vague regarding the nature of these sources by refusing to state whether they were written or oral, instead referring to their common “narratives” instead of common “documents.” Third, because this was an encyclopaedia entry, Warfield wrote in more general terms with fewer comparisons of synoptic data. Fourth, Warfield seemed to gently dismiss the Independence Hypothesis with the following statement:

If the three Synoptic Gospels do not give three independent testimonies to the facts which they record, they give what is, perhaps, better—three independent witnesses to the trustworthiness of the narrative, which they all incorporate into their own as resting on autoptic testimony and thoroughly deserving of credit.23

Instead of arguing, as many advocates of the Independence Hypothesis had done, that the differences in the synoptic gospels proved their independence, Warfield argued that they the differences demonstrated the Synoptics were independently based on the same narrative. Using similar arguments as those in the 1907 article, Warfield posited a narrative source and a sayings source behind Matthew and Luke, and added that the trustworthiness of these sources was guaranteed by Luke’s pledge to consult authentic eyewitness testimony (Luke 1:1–4).24

In 1913, Warfield continued an apologetic tone in his article, “Concerning Schmiedel’s ‘Pillar Passages,’”25 a work meant to answer the most extreme claims of Schmiedel in his 1901 entry in Encyclopedia Biblica.26 Warfield began by grouping Schmiedel with Reimarus27 and Wrede28 in “the quest for the Historical Jesus.”29 Warfield criticized Schmiedel’s desire for scholars to return to the “pre-Tübingen position of criticism” that did not appeal to source criticism. Interestingly, Warfield advocated the opposite approach, noting that F. C. Baur30 had “laid down the reasonable rule” that criticism of the sources must come before criticism of the gospels.31 Warfield faulted Schmiedel for wanting to regress to the approach of Strauss with its “unreasoned scepticism.”32 Schmiedel sought to recover the Jesus obscured by legend and faith, and his method for doing so was particularly bothersome to Warfield. Schmiedel argued that he could find the authentic Jesus by comparing those places in the gospels where one evangelist changed details provided by others in order to enhance the view of Jesus. Schmiedel was able to find nine of these passages and termed them “pillars” because they were the foundation of the true reconstruction of the historical Jesus.33 Warfield argued that Schmiedel erred in his admission, on the one hand, of a common source behind the Synoptics, and on the other, his lack of acknowledgement that the common source, Mark, was composed at a time very close to the events described.34 Likewise, Schmiedel failed to appreciate the even earlier construction of the logia. Warfield criticised Schmiedel for failing to look beyond Matthew and Luke to these sources:

If we are to break up the Gospels into their sources and appeal rather to these sources than to the Gospels … we do not lose but profit by the process. Instead of three witnesses of about the seventh decade of the century we have now in view quite a number of witnesses, all earlier than the seventh decade of the century, some of them perhaps very much earlier.35

Thus, Warfield used the Two-Source Hypothesis to counter the radical skepticism inherent in much of the search for the Jesus of history.

Finally, it is appropriate to consider one further article in which Warfield briefly mentioned the Synoptic Problem.36 In 1914, a year after he urged biblical critics to look at the older sources behind the Synoptics, in a footnote Warfield criticized Theodor Keim’s37 assumption of Matthean priority and added:

And in general no form of criticism is more uncertain than that now so diligently prosecuted which seeks to explain the several forms of narratives in the Synoptics as modifications one of another.38

This single quote would later be used multiple times by Robert Thomas, evangelical opponent of redaction criticism to show Warfield’s rejection of the method.39 While it is obvious that Warfield did not refrain from applying source critical methods to the gospels, he appeared to be wary of source criticism that focused on the editorial activities of the evangelists. Warfield was followed by his Princeton colleague Geerhardus Vos in using the Two-Source Hypothesis to defend the gospels against the “Quest for the Historical Jesus” just a few years later.

3. Geerhardus Vos: Using the Two-Source Hypothesis against Bousset

While not attaining the high profile that Warfield enjoyed at Princeton, Professor Geerhardus Vos was perhaps just as influential in Reformed biblical theology. Vos was born in the Netherlands to a German family, and he moved with the family to Grand Rapids, Michigan for his father to accept a position as pastor of a Reformed Church in 1881. Vos was fluent in Dutch, German, and English, allowing him to move freely between the Reformed institutions of America and Germany. He began his theological studies first at the Theological School of the Christian Reformed Church in Grand Rapids, then moving to Princeton, and on to Germany where he studied at Berlin and Strassburg, ultimately receiving a PhD in Semitics in 1888. From Germany, he returned to the Theological School as professor for five years, finally returning to Princeton as the Chair of Biblical Theology. He retired from Princeton in 1932 after 39 years and several publications.40

It would appear that the only publication in which Vos addressed the Synoptic Problem was a lengthy article he composed for The Princeton Theological Review in 1915. Like Warfield a decade before, Vos used his pen to combat the influence of skeptical gospel criticism, but his target was the work of William Bousset, who had the year before published his puissant Kyrios Christos.41 As Bousset had used the Two-Source Hypothesis to argue that the doctrine of Jesus’s lordship was a late first-century development, Vos used the Two-Source Hypothesis to argue exactly the opposite. At Strassburg, Vos had been a student of Holtzmann, who first exposed him to the Two-Source Hypothesis,42 but Vos used Harnack’s reconstruction of Q to refute Bousset.43

In his article, “The Continuity of the Kyrios Title in the New Testament,”44 Vos showed that the lordship of Jesus had been proclaimed from the earliest NT times. Bousset had observed that the objective title Kyrios was only applied to Jesus once in Mark and nowhere in the logia, and in the vocative form appeared once in each source. To Bousset, the paucity of the occurrences of the title “Lord” in the earliest gospel tradition compared to its frequent use in Matthean and Lukan non-logia contexts implied that the title applied to Jesus was the development of a later tradition.45 Using evidence from the proposed logia, Vos attempted to disprove Bousset’s conclusions. First, Vos disagreed with Bousset’s definition of “titular” form of Kyrios, which Bousset did not find in the logia, as falsely disallowing clear titular occurrences of the word. Vos appealed to Matt 24:43–51 = Luke 12:39–46, where Kyrios was used in parabolic form, but clearly implied a “corresponding relationship between Jesus and the disciple.”46 He also cited Matt 10:24–25, where Kyrios was similarly used in a parable, and where Harnack had concluded that Kyrios was original to the logia.47 Moving to the vocative form Kyrie, Vos argued that Bousset’s claim, that the double Kyrie, Kyrie in Luke 6:46 was evidence of a later cultic use, was invalidated if the Lukan passage be admitted as coming from the logia.48 Vos then further developed his argument by offering a discussion of the logia. Because the document consisted primarily of a list of sayings by Jesus, there would be little occasion within it to include appeals to Jesus as Kyrie or Kyrios. Given its nature, the more conclusive proof from the logia should come from the parabolic forms where Jesus indirectly taught about himself.49

Again, using the Two-Source Hypothesis as a buttress for his arguments, Vos moved to the other source document, Mark. In Mark 2:28, the Son of Man is called Kyrios of the Sabbath, an instance where Bousset too easily dismissed the notion that any kind of sovereignty was meant.50 Bousset came to the same false conclusion with regard to Mark 12:35–37 by refusing to allow that the messiah could be both Kyrios and son of David, the very point that Jesus was trying to make.51 The one occurrence of Kyrios in Mark which Bousset considered a proper title, Mark 11:3, he still incorrectly minimized because of his negative presuppositions concerning “the supernatural in the consciousness of Jesus.”52 Vos also saw a fallacy in Bousset’s acknowledgement that the Syro-Phoenician woman (Mk 7:28) referred to Jesus as Kyrie but rejected the occurrence as evidence for the early use of the title in Palestine because the woman was a foreigner. In fact, Vos argued, Mark’s inclusion of the title was evidence of her great faith, offering her as an example to the Palestinians of the time.53 From that point, Vos argued that if it be admitted that Bousset’s disqualification of the many occurrences of Kyrios and Kyrie was faulty, then Bousset’s argument was rendered unnecessary. Regardless of how many times Matthew and Luke used the title outside of the logia source, the fact that the title was in their sources proved that the development of the title Kyrios as applied to Jesus was not strictly a later development.54

Vos’ article seemed to portend a change among evangelical advocates of the Two-Source Hypothesis in arguing from that hypothesis against its more radical proponents. Evangelical scholars came to be comfortable with the notion of logia, or Q, because they considered it to reflect a high view of Jesus. This idea was taken even further a few years later by the evangelical scholar A. T. Robertson.

4. A. T. Robertson (1863–1934): Evangelicalism’s Strongest
Advocate for the Two-Source Hypothesis

Archibald Thomas Robertson was born in Virginia and raised in North Carolina, where he later attended Wake Forest University, before moving to Louisville, Kentucky to attend the Southern Baptist Theological Seminary (SBTS). After receiving the Th.M. in 1888, he remained at SBTS, working first as a teaching assistant, and then as a professor from 1892 until his death.55 The volume of Robertson’s writing is phenomenal. He is perhaps best known for his Greek grammars—A Short Grammar of the Greek New Testament (New York: George H. Doran, 1908) and Grammar of the Greek New Testament in the Light of Historical Research (London: Hodder & Stoughton, 1919)—and his six-volume Word Pictures in the New Testament (Grand Rapids: Baker, 1930–33).

More than any evangelical scholar before 1950, Robertson was an optimistic and wholehearted advocate of the Two-Source Hypothesis, confident in the ability of the hypothesis to solve the Synoptic Problem and provide a clearer picture of Jesus. It appears that before 1905, Robertson was not convinced by any particular solution to the Synoptic Problem. In 1905, Robertson was one of four NT professors from various theological institutions interviewed by The Biblical World and asked questions relating to biblical matters. One question was, “What is your theory of the relation of the synoptic gospels to one another?” Robertson replied, “The oral, documentary, and mutual dependence theories all have an element of truth in them, though neither by itself can explain all the phenomena.”56 However, over the next four years his tone toward the several solutions changed.

Though he had hinted at his newfound confidence in the Two-Source Hypothesis as early as 1909,57 Robertson made his full endorsement of the Two-Source Hypothesis known in 1911 with statements in his commentary on Matthew’s Gospel.58 He repeated his endorsement that same year in John, the Loyal: Studies in the Ministry of the Baptist.59 In 1915, Robertson appealed to evidence from Q that the animosity between the Pharisees and Jesus was documented in the “earliest strata of the Gospel narratives.”60 In his Luke the Historian In Light of Research,61 Robertson again advocated the Two-Source Hypothesis, claiming that it had been “practically demonstrated” that Mark was used by Matthew and Luke, and that the “oral theory” was insufficient.62 Likewise, he was confident that Matthew and Luke had used a common Q document because of the existence of collections of sayings of Jesus at the time, a fact confirmed by the scraps of logia found at Oxyrhynchus.63 Robertson believed that Mark was written after Q, and perhaps the evangelist Mark had made use of the document.64 In 1922, in A Harmony of the Four Gospels,65 Robertson extolled the Two-Source Hypothesis as a product of biblical criticism “that is likely to stand the test of time” and further, that the theory “seems to be proven.” He opined that it was “plain as a pikestaff” that Matthew and Luke used Mark’s contents and order, and recommended the works of Sanday, Hawkins, and Harnack on Q.66

In his 1924 work The Christ of the Logia Robertson used the Two-Source Hypothesis to argue for the trustworthiness of the gospels.67 The book was a compilation of articles Robertson had contributed to various periodicals, with the title coming from the first essay. Robertson rejected the cleft supposed by critics between the portraits of Jesus in the Synoptics, John, and Paul’s writings. The problem with those who sought “the historical Jesus” was that they failed to face the facts demonstrated in the gospels. However, a critic’s “real attitude” toward Jesus was irrelevant, because the correct approach involved “rigid scientific research into facts.”68 Robertson boldly stated that the Two-Source Hypothesis was “one certain result of Synoptic criticism.”69 Of the two sources used by Matthew and Luke, Q was earliest, perhaps composed in Jesus’s lifetime.70 Therefore, if a scholar desired to find the earliest and simplest Jesus material, the correct place to look was Q.71 The critic must remember that the full extent of Q can never be known, and the portion present in the synoptics is merely “a torso.”72 Robertson validated B. H. Streeter’s conclusion73 that, since about two-thirds of Mark is common to Matthew and Luke, it can be reasonably assumed that Matthew and Luke reproduce about two-thirds of Q.74 Even with only a portion of Q’s contents available, Robertson was sure that the character of Jesus was not diminished in the remainder. Much as Vos had done, Robertson used Harnack’s delineation of Q as his basis to prove the supernatural portrait present in the earliest source. However, instead of using Harnack to combat Bousset, Robertson used Harnack to refute Harnack.

Robertson considered it obvious, from Harnack’s section on the temptations, that Jesus was called “the Son of God” by Satan, an occurrence which Harnack himself felt referred back to the voice from heaven at Jesus’s baptism.75 Harnack also admitted several other instances of the “Son of Man” terminology into his Q.76 Robertson argued that if Jesus was called the Son of God and the Son of Man in Q, it was clear he was called Messiah as well. Harnack’s Q contained important indications of Jesus’s messiahship—the mention that in prison John heard of the “works of Christ,” messianic phraseology of Jesus as “the coming one,” Jesus’s power to give his disciples power to judge Israel, the appeal of “Lord, Lord” to Jesus (language used in the LXX for God)—though Harnack tried to remove the theological and supernatural elements from Jesus’s self-consciousness.77 To Robertson, however, the data in Q would not allow a merely human Jesus. He argued, “The facts in Q are open and simple and beyond reasonable dispute,”78 and though the search for the historical Jesus was “laudable,” it had not “gotten rid of the theological Christ.”79

Robertson moved next to the other common source of Matthew and Luke, Mark’s gospel, which offered a similar portrayal of Jesus as Q. Mark gave clear indications of Christ’s divine nature in two accounts which were picked up by Matthew and Luke. The first, in Mark 2:7–10, demonstrated the divine power to forgive sins. The second account, Mark 9:7, recounted the voice from heaven after the Transfiguration which called Jesus “son of God.”80 Even in the more cryptic portions at the end of the gospel, Mark’s mention of the centurion who called Jesus “son of God” (15:39), as well as the empty tomb of Mark’s shorter ending (16:1–8), must have implied incredible characteristics of Jesus.81

After his brief chapter on Mark, Robertson appealed to the Two-Source Hypothesis once more, in the beginning of his chapter on Matthew, where he detailed his understanding of the chronology of the synoptics. First, the apostle Matthew composed the logia in Aramaic, followed by Mark’s gospel which was written under the guidance of Peter. Robertson did not clarify whether he placed Luke or Matthew third. He admitted that the author of the Greek Matthew was unknown, but posited that it was reasonable to reckon the apostle Matthew took up his logia and Peter’s (Mark’s) gospel and “blended” them into the canonical gospel of Matthew.82

While Robertson used argumentation similar to that of Warfield and Vos, his unqualified acceptance of the Two-Source Hypothesis as “a certain result” of biblical criticism meant that the ultimate strength of his approach rested upon the validity of his assumption. Because he knew that most of the skeptical critics involved in the search for the Jesus of history accepted the Two-Source Hypothesis, his confidence in the certainty of his solution to the Synoptic Problem was not a liability. However, Robertson’s fulsome endorsement of the Two-Source Hypothesis was a surprise to some evangelicals. In 1938, R. C. Foster of Cincinnati Bible College wondered how a “scholar with the conservative reputation of A. T. Robertson” could adopt the “radical Two-Source Theory.”83 Likewise, in 1958 Merrill C. Tenney remarked with surprise that “even such conservative writers as A. T. Robertson and W. Graham Scroggie in Britain have espoused the Two-document theory.”84 Scroggie’s work is considered next.

5. W. Graham Scroggie (1877–1958): The British Preacher for the Two-Source Hypothesis

W. Graham Scroggie was born in Great Malvern, England, into a devoted Baptist family. He attended C. H. Spurgeon’s Pastors College, and immediately began a career in ministry, first at various churches in England, and then at Charlotte Chapel in Edinburgh for several years (1916–1933). In 1927, Scroggie received the Doctor of Divinity degree from the University of Edinburgh, and in 1938 he became pastor of the Metropolitan Tabernacle (popularly called Spurgeon’s Tabernacle), where he helped lead the church through WWII. He retired from the Tabernacle in 1944, but he continued to lecture at the Pastor’s College in London, and to travel and preach throughout the world. Scroggie was a featured speaker on several occasions at the annual Keswick Convention, a large gathering of evangelical Christians in Cumbria.85 In addition to his busy preaching schedule, Scroggie was a prolific writer, authoring Scripture Union, a daily bible study guide, as well as educational material for The Sunday School Times. He published over 20 books, including commentaries on all four gospels and the massive 680-page A Guide to the Gospels.86 It was in A Guide that Scroggie offered his thorough consideration of the Synoptic Problem.

Scroggie did not propose to break new ground on the issue, but in A Guide he offered, from an evangelical perspective, a summary of what he considered to be the accepted opinion of most scholars. His statements on the issue came in four separate sections of the book, first in the section entitled “The Synoptic Problem” and then in discussion of each synoptic gospel. Though his guide was written on the popular level, Scroggie was confident in stating, “[t]hat there is such a [synoptic] problem is a fact, and everyone who is interested in the Gospels should know something about it.”87 He encouraged his readers to avoid the two potential perils that accompany the Synoptic Problem: either indifference to the gospels’ origins or preoccupation with their origins and a lack of appreciation for the gospels themselves.88 As evidence that it was typical for the biblical authors to make use of earlier documents, he cited Paul’s inclusion of “snatches of song from an early Church hymn book” and summaries of written creedal statements (1 Cor 11:23–25; Eph 5:14, 19; Col 1:13–2089; 1 Tim 3:16, 6:15–16).90 He quickly dismissed two alternatives to the hypothesis he intended to advocate. The “oral tradition hypothesis” was based on the idea of early “catechetical schools,” which Scroggie believed existed, but oral tradition could not completely solve the Synoptic Problem. He also described the “Mutual Use Hypothesis,” by which he meant a strict dependency hypothesis with no other written sources. He dismissed the six possible permutations of this hypothesis because it had few advocates.91 Scroggie then moved on to his preferred solution, the Two-Source Hypothesis, and cautioned his readers that Q was “a theory and not a certainty.” He was more confident, however, in the priority of Mark.92

While Scroggie was clearly acquainted with many works of critical scholarship on the Synoptic Problem, he mentioned that he had compared seven different authors on the contents of Q. These scholars were: Harnack, Holtzmann, Wellhausen,93 Wendt,94 Hawkins,95 Stanton,96 and Redlich.97 Curiously, Scroggie later listed the contents of Q according to Streeter,98 though he did not include Streeter in the initial seven scholars under consideration. Scroggie decided that, when four of the seven agreed on the verses found in Q, those verses could be considered part of the foundation for exploration into the contents of Q. From his comparison, Scroggie determined that four or more of the scholars agreed on a total of 237 verses in Luke. Scroggie compared a few of these Lukan pericopae with their Matthean counterparts to show the likelihood of a common source, invoking Papias’s mention of the logia as evidence of the existence of a Q-like document.99 Scroggie offered further evidence that sayings documents were common in the early church by mentioning a fragment of papyrus from Oxyrhynchus with sayings very similar to those found in the Sermon on the Mount.100 He was presumably referring to P.Oxy. 654 (which was still a relatively new discovery at the time), though those logia are now known to be quotations from the Gospel of Thomas.101

Scroggie then provided a table with twenty-six parallel pericopae between Luke and Matthew, which he considered a “syllabus of Q.”102 Although he had earlier cautioned his readers that Q was simply a theory, he did not refrain from confidently endorsing it. After giving the syllabus, he encouraged his readers to write out the passages side by side and compare the results of these non-Marcan parallels. He remarked that, though the texts were written by different men, at different times, at different places, and for different purposes, the conviction will not be escaped that these passages come from a common source, the lost document Q.103

Aware of the sensitivities of his evangelical readership, Scroggie also dealt with the implications for inspiration such an explanation of gospel origins might have. He laid out what he considered to be an appropriate approach to framing a theory of inspiration. It was fatal, explained Scroggie, to form a theory of inspiration “and then attempt to explain the Scriptures in the light of it.”104 The correct course would be to “let a doctrine of inspiration arise from the facts” drawn from the Bible. He offered three “facts” that should inform one’s conception of inspiration. First, the individuality (style, mode of expression, arrangement of material) of each evangelist is preserved. Second, the accounts reveal a “great variety of report,” with none necessarily giving “the exact words” throughout. Third, the evangelists did not receive supernatural information directly from God which they could otherwise obtain by their own investigations, as described in Luke’s preface.105 However, it could be stated without equivocation, opined Scroggie, that the evangelists were guided by the Holy Spirit in the selection of material to fit their individual designs.106

Concerning the order of the gospels, Mark was first because it was closest in form to the apostolic oral gospel.107 Q preceded Mark, and Mark may have made use of it, but his primary source of information was the apostle Peter in Rome.108 Scroggie considered Luke’s gospel to have been written during Paul’s imprisonment in Caesarea (58–60 CE), thus Matthew’s gospel, which came earlier, would receive a date in the middle fifties with a provenance of Jerusalem.109 Matthew’s sources were Q, “a Manual of Messianic Prophecy,” Mark, oral tradition, various records now lost (Scroggie referenced Luke 1:1), and Matthew’s own recollections.110 Luke received his information from Q, Mark, Jesus’s mother, and “information derived from the court of Herod, Paul and his associates, and Philip of Caesarea.”111 Thus, Scroggie outlined in brief his particular version of the Two-Source Hypothesis, complete with extra sources beyond Q and Mark.

6. Conclusion

While English-speaking evangelicals were reluctant to embrace Q until the twentieth century (the subject matter of this essay), they followed in a long line of biblical conservatives who were interested in the interrelationships between the gospels. Since the outset of the Protestant Reformation, there have been three primary reasons that biblical conservatives have attempted to answer the Synoptic Proble.112 First, important figures such as John Calvin and Martin Chemnitz offered solutions to the Synoptic Problem in the sixteenth century when constructing gospel harmonies. When one is trying to decide how to arrange events in the life of Jesus, and whether passages describe the same or different events, it is natural to wonder which gospel came first, and who might have adapted whom. The second impetus for investigation into the synoptic problem was the construction of a critical text of the New Testament. Thus, John Mill, J. A. Bengel, and Henry Alford, all textual critics, addressed the Synoptic Problem in their publications. In these two pursuits (the construction of gospel harmonies and critical texts), biblical conservatives were at the forefront. Lastly, with the advent of the Quest for the Historical Jesus, conservatives have been playing catch-up for over a century. This essay shows that there were likewise three primary reasons that Warfield, Vos, Robertson, and Scroggie (and the evangelicals who have followed them) were persuaded by the Two-Source Hypothesis. First, they assumed that the appeal to the Q document represented a scientific approach to solutions to the Synoptic Problem. This supposed scientific approach meant that, as Robertson stated, one’s faith (or lack thereof) was irrelevant to the investigation, meaning that it could occupy theological “neutral ground.” It also meant that older traditions regarding the origins of the gospels (particularly Matthean priority) were dismissed as irrelevant. Second, the fact that Q was considered a very early written collection of Jesus’s sayings meant that it was composed even closer to the time of Christ, perhaps even during his earthly ministry. Third, they were all convinced that Q contained a high Christology, so that it could be used to combat the most radical assertions of the Quest for the Historical Jesus. While the current author is not persuaded of the need for appeals to Q, preferring instead the Farrer Hypothesis (or Markan Priority without Q),113 it is clear that the hypothetical document has had its appeals to evangelicals because they saw in it an early written source, that demonstrated a high Christology, and which best explained the synoptic data.

[1] D. A. Carson and Douglas J. Moo, An Introduction to the New Testament, 2nd ed. (Grand Rapids: Zondervan), 95.

[2] Published as Michael Strickland, The Evangelicals and the Synoptic Problem, American University Series 336 (Berlin: Peter Lang Academic, 2014). I summarize some of the findings of that work below in the conclusion. I have adapted and expanded some of my research in this essay.

[3] The Independence Hypothesis holds that each evangelist composed his gospel without having seen the work of his predecessors. John Calvin was an early advocate, with more recent advocates being Louis Berkhof, Eta Linnemann, Robert Thomas and F. David Farnell. For greater detail on the works in which these authors advocated for the Independence Hypothesis, see Strickland, The Evangelicals and the Synoptic Problem, 193.

[4] The Augustinian Hypothesis, following comments made by Augustine (De Consensu Evangelistarum 1.2.4), holds that Matthew wrote first; Mark made use of Matthew; and then Luke made use of both Matthew and Mark. Martin Chemnitz was an early Lutheran scholar of the sixteenth century who advocated for the Augustinian Hypothesis. Most recently, the English evangelical scholar John Wenham argued for this hypothesis. For greater detail on the works in which these authors advocated for the Augustinian Hypothesis, see Strickland, The Evangelicals and the Synoptic Problem, 194.

[5] First advocated by C. H. Weisse (Die evangelische Geschichte kritisch und philosophisch bearbeitet [Leipzig, 1838]), the Two-Source Hypothesis gained notoriety through the work of H. J. Holtzmann in his Die Synoptische Evangelien: Ihr Ursprung iind geschichtliche Charakter (Leipzig: Wilhelm Engelrnann, 1863). The theory holds that Mark wrote first, and then Matthew and Luke both made use of Mark (the first shared source). Both Matthew and Luke also made use of a shared document, now lost, identified with the siglum Q, representing the German word for source, Quelle. This document contained at minimum the material of the double-tradition (material common to Matthew and Luke but not in Mark). The Four-Source hypothesis, or Oxford Hypothesis (popularized by the work of Oxford don, B. H. Streeter, The Four Gospels [London: Macmillan, 1924]), also posits a unique source for Matthew and another for Luke, but retains the standard elements of the Two-Source Hypothesis. The evangelical scholars who have advocated for this view are too many to list, but perhaps the best known modern scholar is Robert H. Stein.

[6] Andreas J. Köstenberger, L. Scott Kellum, Charles L Quarles, The Cradle, the Cross, and the Crown: An Introduction to the New Testament (Nashville: B&H Academic, 2009), 112n40.

[7] George M. Marsden, “Reformed and American,” in Reformed Theology in America: A History of Its Modern Development, ed. David. F. Wells (Grand Rapids: Eerdmans, 1985), 8.

[8] For an excellent description of Warfield’s life and legacy, see Kim Riddlebarger, The Lion of Princeton: Benjamin Breckinridge Warfield on Apologetics, Theological Method and Polemics (PhD diss., Fuller Theological Seminary, 1997).

[9] William Baird (History of New Testament Research: From Jonathan Edwards to Rudolf Bultmann, vol. 2 [Minneapolis: Fortress, 2003], 348) considers it clear that Warfield adopted a “modified form” of the Two-Source Hypothesis.

[10] Warfield, “The Primitive Jesus,” in The Lord of Glory: A Study of the Designations of Our Lord in the New Testament with Synoptic Problemecial Reference to His Deity (London: Hodder and Stoughton, 1907), 133–58.

[11] Ibid., 135.

[12] Ibid.

[13] Ibid., 136.

[14] Ibid., 137.

[15] It was common in this period to use the term logia to refer to the Q document, based on Papias’ description of Matthew’s written account of sayings of Jesus (see Eusebius, Hist. eccl. 3.39.16). However, the German evangelical scholar Theodor Zahn, whose Einleitung in das neue Testament, 2 vols. (Leipzig, 1897, 1900) was soon translated into English (Theodor Zahn, Introduction to the New Testament, 2 vols, trans. John Moore Trout, et al. [Edinburgh: T&T Clark, 1909]), had already demonstrated that Papias’ logia could hardly have meant an Aramaic document with a Greek title, especially one that was never mentioned in any other church writings. See Zahn, Introduction, 2:603–4. According to Eta Linnemann, Zahn was the first to argue against equating Q with the logia. Eta Linnemann, “The Lost Gospel Of Q—Fact Or Fantasy?” TJ 17.1 (Spring 1996): 6. However, because the authors considered in this essay used the terms logia and Q interchangeably, I have adopted their practice here.

[16] Adolf von Harnack, Sprüche und Reden Jesu: die Zweite Quelle des Matthäus und Lukas (Leipzig: Hinrichs, 1907).

[17] Ibid., 140–41.

[18] Ibid., 143.

[19] Ibid., 146.

[20] Ibid., 157.

[21] Benjamin B. Warfield, “Jesus Christ,” The New Schaff-Herzog Encyclopedia of Religious Knowledge (New York: Funk & Wagnalls Company, 1910), 6: 150–76. Quotations are from this edition.

[22] Benjamin B. Warfield, Christology and Criticism, (Oxford: Oxford University Press, 1929), 149–77.

[23] Warfield, “Jesus Christ,” 152.

[24] Ibid.

[25] “Concerning Schmiedel’s ‘Pillar-Passages’,” The Princeton Theological Review 11 (1913): 195–269.

[26] Paul W. Schmiedel, “Gospels,” in Encyclopaedia Biblica, ed. Thomas K. Cheyne and John S. Black (London: A. & C. Black, 1901), columns 1765–1896.

[27] H. S. Reimarus (1694–1768) is often considered the Father of the Quest for the Historical Jesus. His writings on Jesus were first published posthumously by G. E. Lessing in 1774, without identifying the original author. See Gerd Theissen and Annette Merz, The Historical Jesus: A Comprehensive Guide (Minneapolis, Fortress Press: 1996), 2–3.

[28] William Wrede, Das Messiasgeheimnis in den Evangelien (Gottingen: Vandenhoeck & Ruprecht, 1901); ET The Messianic Secret (Cambridge: James Clarke, 1971).

[29] Warfield, “Concerning,” 195.

[30] F. C. Baur, Die christliche Gnosis (Tübingen: C. F. Osiander, 1935).

[31] Warfield, “Concerning,” 195.

[32] Ibid., 197.

[33] Ibid., 203–4. As Walter P. Weaver, The Historical Jesus in the Twentieth Century: 1900–1950 (Harrisburg, PA: Trinity Press International, 1999), 47, remarks, “These passages would guarantee that such a figure as Jesus existed, that some irreducible minimum could be known with certainty about him, and on this basis a wider picture could be constructed, provided that it did not contradict the pillar passages.”

[34] Warfield, “Concerning,” 240.

[35] Ibid. Later, Warfield suggested the sources predated Matthew and Luke by as much as twenty years. See p. 242.

[36] Benjamin B. Warfield, “Jesus’ Alleged Confession of Sin,” The Princeton Theological Review 12 (April 1914): 177–228.

[37] Theodor Keim, Die Geschichte Jesu von Nazara (The History of Jesus of Nazara), vol. 5 (1876–1881), 37.

[38] Warfield, “Jesus’ Alleged,” 196 n. 34.

[39] See Thomas’s “Introduction,” in The Jesus Crisis, ed. Robert Thomas and David Farnell (Grand Rapids: Kregel, 1998), 13–36. The Warfield quote appeared three times in the book, all by Thomas (pp. 14, 24, and 358). Thomas also used the Warfield quote in several other publications: Robert L. Thomas, ed., The Master’s Perspective on Contemporary Issues (Grand Rapids: Kregel, 1998), 221; “Historical Criticism and the Evangelical: Another View,” JETS 43.1 (March 2000): 98; and Evangelical Hermeneutics: The New Versus the Old (Grand Rapids: Kregel, 2002), 301.

[40] James T. Dennison, “Geerhardus Vos: Life in Two Worlds,” Kerux 14.2 (Sep 1999): 18–31.

[41] Wilhelm Bousset, Kyrios Christos: Geschichte des Christusglaubens von den Anfangen des Christentums bis Irenaeus (Gottingen: Vandenhoeck & Ruprecht, 1913).

[42] James T. Dennison, The Letters of Geerhardus Vos (Phillipsburg, NJ: P&R Publishing, 2006), 24.

[43] Adolf von Harnack, The Sayings of Jesus: The Second Source of St. Matthew and St. Luke, trans. J. Wilkinson (London: Williams & Norgate; New York: G. P. Putnam’s Son’s, 1907–1908).

[44] Geerhardus Vos, “The Continuity of the Kyrios Title in the New Testament,” Princeton Theological Review 13 (April 1915): 161–89.

[45] Vos, “Continuity,” 164–65.

[46] Ibid., 167.

[47] Ibid., 168–69.

[48] Ibid., 169–71. Also, considering Matt 8:8 = Lk 7:6, Vos argued, citing Harnack, that Bousset’s failure to include this occurrence of the vocative Kyrie in the logia was a mistake.

[49] Vos, “Continuity,” 172.

[50] Ibid., 174. Indeed, as Benedict Viviano, Matthew and His World: The Gospel of the Open Jewish Christians: Studies in Biblical Theology (Fribourg: Academic Press Fribourg, 2007), 114, observes, “The OT background would lie with humans as the crown of creation and the image of God (Gen 1 :26, 28), as well as in the parallelism between man and son of man that one finds in Ps 8:5–9, and in the dominion over creation (Gen 1:28–31; Jub. 2:14)…. The Christological rereading of the verse by Mark would affect not only Son of man but also kyrios. Jesus as Daniel’s Son of man would stand as the divine kyrios over the Sabbath.”

[51] Vos, “Continuity,” 176.

[52] Ibid., 177.

[53] Ibid., 180–81.

[54] Ibid., 185–89.

[55] Baird, 2:412–14.

[56] “An Interview with New Testament Scholars: W. F. Adeney, D. A. Hayes, A. T. Robertson and Frank C. Porter,” The Biblical World 26.3 (Sep 1905): 199–200.

[57] A. T. Robertson, Epochs in the Life of Paul: Study of Development in Paul’s Career (New York: Scribner’s, 1909), 87.

[58] A. T. Robertson, Commentary on the Gospel According to Matthew (New York: Macmillan, 1911), 11–22.

[59] A. T. Robertson, John, the Loyal: Studies in the Ministry of the Baptist (New York: Scribner’s, 1911), 61.

[60] A. T. Robertson, The Pharisees and Jesus: The Stone Lectures for 1915–16 (New York: Scribner’s, 1920) 62.

[61] A. T. Robertson, Luke the Historian in Light of Research (New York: Scribner’s, 1920), 61–72.

[62] Ibid., 66.

[63] For many years, until the positive identification of the Oxyrhynchus fragments (pOxy. 1, 654, and 655) with the Gospel of Thomas (after the Nag Hammadi discovery in 1945), scholars assumed that published findings of the fragments by B. P. Grenfell and A. S. Hunt (Logia lesou: Sayings of Our Lord, Egypt Exploration Fund [London: Frowde, 1897]) provided evidence that collections of Jesus’s sayings were common in the first century.

[64] Robertson, Luke the Historian, 70.

[65] Robertson, A Harmony of the Four Gospels for Students of the Life of Christ (New York: Harper & Row, 1922).

[66] Ibid., 255–56.

[67] A. T. Robertson, The Christ of the Logia (New York: George H. Doran, 1924).

[68] Ibid., 17.

[69] Ibid.

[70] Ibid., 18–19.

[71] Ibid., 20.

[72] Ibid., 23.

[73] B. H. Streeter, “The Original Extent of Q,” in William Sanday, ed. Oxford Studies in the Synoptic Problem (Oxford: Oxford University Press, 1911), 185–208.

[74] Robertson, The Christ, 24.

[75] Ibid., 29–30.

[76] Ibid., 31.

[77] Ibid., 33–34.

[78] Ibid., 37.

[79] Ibid., 38–39.

[80] Ibid., 48–49.

[81] Ibid., 52–53.

[82] Ibid., 54–55.

[83] R. C. Foster, Studies in the Life of Christ (Grand Rapids: Baker Books, 1938), 59.

[84] Merrill C. Tenney, “Reversals of New Testament Criticism,” in Carl F. H. Henry, ed., Revelation and the Bible: Contemporary Evangelical Thought (Grand Rapids: Baker Books, 1958), 353–67, at p. 355.

[85] David. L. Larsen, The Company of the Preachers, vol. 2 (Grand Rapids: Kregel, 1998), 710–11.

[86] W. Graham Scroggie, A Guide to the Gospels (London: Pickering & Inglis, 1948). Quotations are from the reprint edition (Grand Rapids: Kregel, 1995).

[87] Scroggie, Guide, 83.

[88] Ibid.

[89] Ibid., 84. Presumably, Scroggie meant Col 1:15–20.

[90] Ibid.

[91] Ibid., 85.

[92] Ibid., 87–89.

[93] Julius Wellhausen, Einleitung in die drei ersten Evangelien (Berlin: Reimer, 1905).

[94] H. H. Wendt, Die Lehre Jesu, 2 vols. (Gottingen: Vandenhoeck & Ruprecht, 1886, 1890).

[95] J. C. Hawkins, Horae Synopticae: Contributions to the Study of the Synoptic Problem, 2nd ed. (Oxford: Clarendon, 1909; repr. 1968).

[96] V. H. Stanton, The Gospels as Historical Documents: Vol. 2, The Synoptic Gospels (Cambridge: Cambridge University Press, 1909).

[97] E. B. Redlich, Student’s Introduction to the Synoptic Gospels (London: Longmans, 1936). See Scroggie, Guide, 89.

[98] Scroggie, Guide, 92.

[99] Ibid., 91.

[100] Ibid., 92.

[101] See John S. Kloppenborg, Excavating Q: The History and Setting of the Sayings Gospel (Minneapolis: Fortress, 2000), 338–39, who quotes Benjamin W. Bacon (“Logia,” in A Dictionary of Christ and the Gospels, ed. James Hastings [New York: Scribner’s Sons, 1908], 2:45), “The discovery of Grenfell and Hunt of papyri of the 2nd and 3rd century, in which sayings attributed to Jesus are agglutinated with no more of narrative framework than the bare words ‘Jesus saith’ (legei Iesous), proves that such compilations actually circulated, fulfilling a function to the Pirke Aboth, or the ‘Sayings of the Fathers’ in the contemporary and earlier Synagogue.” See also Edwin Yamauchi, “Logia,” ISBE 3:153.

[102] Scroggie, Guide, 92–93.

[103] Ibid., 93.

[104] Ibid., 141.

[105] Ibid.

[106] Ibid., 145.

[107] Ibid., 179.

[108] Ibid., 183.

[109] Ibid., 139.

[110] Ibid., 254.

[111] Ibid., 343.

[112] This represents one of the major findings of my doctoral research. See Strickland, The Evangelicals and the Synoptic Problem, 205–6.

[113] For a good introduction to the Farrer Hypothesis, see John C. Poirier and Jeffrey Peterson, eds., Marcan Priority without Q: Explorations in the Farrer Hypothesis, LNTS 455 (London: Bloomsbury T&T Clark, 2015).