Misleading Statistics About the Synoptic Puzzle
(avg. read time: 12–24 mins.)
The Synoptic Puzzle, being concerned with both the similarities and differences among the Synoptic Gospels, has a massive scope. To attempt to give students a handle on it, professors and the textbooks they use can often resort to statistics. However, the kinds of statistics popularly utilized give only an illusion of a handle because of their vague and overall misleading characteristics. Today, I will be looking at some New Testament and Synoptic Gospel introductions, examining the cited statistics, and explaining why they are unhelpful.
I had initially planned to make this a paid subscriber exclusive as a preview draft for an article I planned to submit for publication. However, due to time constraints and budget constraints (since I do not live anywhere near an academic library), I was not able to write a draft at this time that I would be willing to submit to a journal. I do not know when or if I will get around to revising this for such a submission, but it would need to wait for if I ever get another job in or related to academia, the prospects of which are looking bleaker by the day. But for now, a perk I will include for my paid subscribers is my gargantuan group of synoptic tables whence I derive the figures I use, without the accompanying commentary. It is a lot, but if you want to see the work for how I got my raw data, it is there for you after the footnotes. Throughout, I am using the NA28 as my base text, not because it is necessarily the best, but simply because of its popularity.
Statistics About Mark and the Other Gospels
The heading for this section is vague because scholars tend to use statistics that skew high for how close Mark is to the other Gospels, but they state this high degree of similarity in distinct and not necessarily overlapping ways. As the first example, N. T. Wright and Michael Bird write in their introduction, “Large swaths of Mark are paralleled in Luke, so that around half of Luke’s (much longer) gospel is ‘Markan’ material, while Matthew replicates 90 per cent of Mark, though likewise containing a good deal of non-Markan material.”1 What they mention as the only unique material is Mark 7:33–36; 8:22–36 [sic.];2 and 14:51–52.3
Now, you may be wondering what it means that half of Luke is “Markan” material and what it means that Matthew “replicates” 90% of Mark. Are the statistics determined according to verses or words? If it is the latter, these figures are simply wrong, as I will illustrate later. If it is the former, these figures are not only anachronistic—given that verse divisions were introduced much, much later—but they are also obfuscating.4 The figures lump together verses that have few to no parallels, but which may be part of parallel units/pericopes, with verses that are verbatim similar (by which I mean both the same words and the same word order) or close to that standard of similarity. No source is immediately attached to these figures, but the nearest one is Bird’s The Gospel of the Lord.5 In that book, Bird provides two different figures, one is drawn from B. F. Westcott’s Introduction to the Study of the Gospels, which says that the “concordances” (which Bird describes as “shared material”) are 93% for Mark to 7% peculiarities, whereas for Matthew the figures are 58% and 42% respectively, and for Luke the figures are 41% and 59% respectively.6 Westcott, in turn, drew these figures from William Stroud’s A New Greek Harmony of the Gospels, but that figure is also not based on either verses or words. Rather, it is based on “lines” according to a particular edition of presentation.7 As long as there is a parallel to be made, he counts that as a concordance. By his own admission, “In the foregoing Summary, the Lines are taken from Bagster’s English Hexapla, 4to. London, 1841. The numbers refer to entire paragraphs, and not to their fractional parts, and are therefore to be regarded only as near approximations to the real quantities; but under a more minute analysis, the number of Peculiarities would be greater, and that of Concordances less, than is here represented.”8 As for Bird, he later says, “Matthew possesses about 85% of Mark and Luke in the vicinity of 65% of Mark.”9 No source is given for these statistics, so I am not sure how he arrived at this determination. The 90% figure in his book with Wright appears to be a rounding up to nearly split the difference between the one from this book and what he got from Stroud via Westcott. The 65% figure is an even greater divergence, and it is unclear how he got there.
In that same context, he gives the same list of texts that are unique to Mark with the correct citation this time. But here he diverges from other lists in his underselling of unique texts in Mark. He is only giving what he classifies as units, but M. Eugene Boring lists in his introduction other cases of unparalleled material: 1:33; 2:2, 19, 27; 3:20–21; 5:5; 6:22, 27, 31, 52; 7:3–5, 9, 24–26, 33–36; 8:22–26; 9:10, 15–16, 21, 23–24; 10:10, 24, 50; 11:16; 12:29, 32–34; 14:51–52; 15:8, 44–45.10 I have cleaned this up from the version in his book, where he accidentally cites some texts more than once, but it still disagrees with his count of forty-five unparalleled verses. This list also diverges from Craig Evans’s contribution to The Synoptic Problem: Four Views, where he lists: 1:1; 2:27; 3:20–21; 4:26–29; 7:2–4, 32–37; 8:22–26; 9:29, 48–49; 13:33–37; 14:51–52.11 As a matter of fact, here are the verses that have no absolute verbal similarity (simply using the same words the same number of times as a parallel, regardless of order) with a text in Matthew or Luke, with any weighted similarities (i.e., using synonyms or different forms of the same word) appearing in brackets:
Mark 1:1, 33, 43;
3:11 [the only rough similarities being the uses of πνεύματα and ἀκάθαρτα], 20–21, 30;
4:26–29;
5:5 [the only rough similarity being the use of κράζων], 20, 26 [the only rough similarities being the uses of ἰατρῶν and μηδὲν], 32 [the only rough similarity being the use of ἰδεῖν];
6:23, 52;
7:2–4, 9, 12–13, 26, 32, 34–36;
8:13, 22–26;
9:10, 15–16 [the only rough similarity being the use of ἠσπάζοντο αὐτόν in v. 15], 20–21 [the only rough similarities being the uses of καὶ, συνεσπάραξεν, and τὸν πατέρα], 23–24, 30 [the only similarity being a different case of referring to Galilee], 48;
10:12;
11:12, 16;
12:33–34 [the only rough similarity being the use of ἀπεκρίθη in v. 34];
14:51–52, 56–57 [the only rough similarities being the uses of πολλοὶ, ἐψευδομαρτύρουν, and λέγοντες], 59;
15:8 [the only rough similarities being the uses of ὁ ὄχλος and αἰτεῖσθαι];
16:12–13, 20 [the only rough similarity being the use of the common word δὲ]
One could also add to this list 4:23, but it should be noted that there are two words of similarity with a similar saying in Matt 13:43, although these are not strictly speaking structural parallels. The same could be said of Mark 16:9, where there are multiple weighted similarities with structural parallels, but no absolute similarities (the closest similarities are with Luke 8:2, simply because both texts describe the same background information about Mary Magdalene). 2:2 comes close, but for our purposes it is not included because of incidental similarities in two uses of καὶ. I am not sure why 2:27 is in either list, unless they meant only to reference part of it without offering specification. In other cases where other lists have noted verses by mistake, a few words have been missed here and there. Thus, we have at least fifty-six verses without any absolute verbal correspondence and as many as fifty-eight. Of course, I am aware that scholars tend to discount the ending of Mark (as I have addressed here and here), and we have also not taken into account the five other verses that are considered later additions by the consensus of textual critics in 7:16; 9:44, 46; 11:26; and 15:28. Among texts that the consensus would accept, the total is fifty-three or fifty-four.
But even this undersells the differences, as there are many other verses—which we will stick with for now since that seems to be the preferred metric—that have a low degree of absolute or weighted similarity compared to Matthew or Luke. If we define “a low degree of similarity” as being less than 50% of the word count in a verse being shared with either Matthew or Luke, that characterization applies to the following verses (including those already cited):
1:1, 12 (A), 15, 21, 26 (A), 29 (A), 33, 35, 37–38, 43, 45
2:1–3, 4 (A), 6, 12 (A), 13, 27
3:8–11, 13 (A), 14, 17 (A), 20–21, 22–24 (A), 25, 26 (A), 28–29 (A), 30
4:2 (A), 10 (A), 13, 21 (A), 26–29, 30 (A), 31 (A), 33–34, 35 (A), 36, 37 (A), 38
5:2 (A), 3–5, 8–10 (A), 16, 18–19 (A), 20–21, 22–24 (A), 26–27, 29–30 (A), 31–33, 36 (A), 38–39 (A), 40–41, 43
6:2 (A), 5–6, 7 (A), 9 (A), 12 (A), 13, 15, 16 (A), 19–25, 27, 30 (A), 31, 33–34, 38–40, 43 (A), 48 (A), 51 (A), 52, 54 (A), 56
7:1 (A), 2–5, 9, 12–13, 15, 17, 19, 22–26, 29–37
8:1, 3, 4 (A), 7, 10–11 (A), 13–14, 18, 19 (A), 20–26
9:3, 6 (A), 8 (A), 10, 12–13 (A), 14–16, 20–27, 29–30, 33–35, 45, 48–49, 50 (A)
10:5, 10, 12, 16 (A), 24, 30, 32, 35, 36 (A), 39, 46 (A), 49–50
11:4–6, 11–13, 14 (A), 16, 18–19, 20 (A), 21, 23 (A), 24–25
12:5 (A), 28 (A), 29, 32–34, 35 (A), 41, 42 (A)
13:1 (A), 3 (A), 10 (A), 33–36, 37 (A)
14:33 (A), 51–52, 56–59, 64 (Α). 69 (Α)
15:1 (A), 7 (A), 8, 12 (A), 19 (A), 23–24 (A), 25, 41, 42 (A), 44, 45 (A)
16:3, 4–5 (A), 8–9, 10–11 (A), 12–20
We have also noted another text that could be added to this list in 4:23. In any case, it would definitely qualify for the (A) designation to signify that its absolute verbal similarity with a possible parallel from Matthew is under 50%, although the weighted similarity would be a different story. 1:2 parallels a different text than anything in the parallel pericope, and so it will not count here. Otherwise, we have 177 verses that have less than 50% verbal similarity of any kind (or 167 to those who reject Mark’s ending) with Matthew or Luke. An additional eighty-three or eighty-four verses (or eighty-two or eighty-one to those who reject Mark’s ending) have words that would push the weighted verbal similarity past 50%, but the absolute verbal similarity is still below 50% with Matthew or Luke. With a total of 260 or 261 out of 673 verses (counting the ending but not the other five), this description fits 38.6% or 38.8% of all verses in Mark. Of course, if we were to look for verses that had less than 50% absolute verbal similarity compared to only Matthew, only Luke, or both Matthew and Luke (i.e., the similarity threshold must be met in relation to both at once), the number would be higher.
As such, I have a hard time making sense of what it means for Matthew to “possess” 85% of Mark and for the same to be said of Luke to the tune of 65% of Mark. The reader might wonder how that can be since the only unique material he has identified from Mark constitutes eleven verses (less than 2% of Mark’s total number of verses). The numbers do not add up. If you take my more expansive list of those with zero absolute verbal correspondences and add others that are only paralleled by Luke, you would still need to ignore only a selection of others with low correspondence, including those that may have only incidental similarities (though it should be remembered that Bird also does not think Mark ends at 16:20). But I cannot read Bird’s mind to determine what is included and what is not.
Raymond Brown similarly stated, “Mark has 661 verses … Matt has 1,068, and Luke has 1,149. Eighty percent of Mark’s vv. are reproduced in Matt and 65 percent in Luke.”12 The latter figure is identical to Bird’s from his own book while the former is lower from both volumes Bird wrote. I was not able to access his source to see how the figures were calculated, but given what we have seen already, it is clearly overstatement to say that this many verses are “reproduced” in Matthew or Luke.
Bruce Metzger states, “The substance of 606 of the 661 verses of Mark reappears in somewhat shortened form in Matthew, and about 350 of the 661 verses of Mark reappear in Luke. Stated in another way, of the 1,068 verses of Matthew, about 500 contain the substance of 606 verses of Mark, while rather more than half of Mark’s material (about 350 out of 661 verses) is embodied in Luke’s 1,149 verses.”13 As with the aforementioned claims, it is difficult to know what is meant by the descriptions “contain the substance” and “embody” when it comes to verses (again, an anachronistic division). If Metzger was simply referring to a large degree of similarity, we have already seen that his figures are not accurate (and this is to say nothing of additional differences in Matthew and Luke’s versions). Of course, it is not clear where they came from, as he gives no source. These are either his own estimates, or he has adopted them from somewhere else.
D. A. Carson and Douglas Moo attempt to be more granular by observing:
Mark is considerably shorter than both Matthew and Luke: 11,025 words as against 18,293 and 19,376, respectively. It is not Mark’s relative brevity per se that provides evidence of Markan priority (it cannot be demonstrated that the shorter is necessarily the earlier), but its brevity taken in conjunction with its close relationship to Luke, and especially to Matthew. Over 97 percent of Mark’s words have a parallel in Matthew; over 88 percent in Luke.14
These statistics are drawn from Robert Stein, who draws them from Joseph Tyson and Thomas R. W. Longstaff.15 These statistics are irreconcilable with what Wright, Bird, Brown, and Metzger have claimed. But like those claims, this is broadly stated. The totals that Tyson and Longstaff arrive at are based on adding up word totals of pericopes that parallel, excluding the ending of Mark, regardless of the degree of parallel. Thus, unparalleled verses that are part of parallel pericopes are lumped into this total. When they state matters more precisely in the chapter on Mark, they tabulate that Mark has 4,406 unique words (40% of their total), which you would think would be excepted from “words [that] have a parallel.” They then note that Mark has 3,512 words in “continuity”16 with Matthew (31.9% of their total), 920 words that have “identity”17 with Matthew (8.3% of their total), and 925 words that have “equivalency”18 with Matthew (8.4% of their total). My own tabulations that I will present below are more generous with similarities than this, but they are closer than the summary that has been presented. Likewise, in terms of Mark’s verbal agreements with Luke, they give figures of 2,135 words in continuity (19.4%), 738 words that have “identity” (6.7%), and 670 words that have equivalency (6.1%). The complete agreements between Matthew, Mark, and Luke are stated to be 1,140 words in continuity (10.3%) and 592 words in identity (5.4%).19 By any measure, Carson, Moo, and Stein have overstated the point in a misleading fashion because they have refused the path of precision. The same applies to Boring when he claims, using different sources, “Of Mark’s 11,242 words, ca. 94 percent are found in either Matthew or Luke, ca. 72 percent in both.”20
But these scholars at least had the right idea of trying to go by number of words. This metric also has its weaknesses, since it depends on the base text, but the approximation will at least be more valuable than claims about “verses” that obfuscate how different such units of texts in parallel pericopes can be. Based on my own work for my Gospel synopsis commentary, which uses the 28th edition of the Nestle-Aland Novum Testamentum Graece as the base text (simply because it is so widely used among scholars, not because it is by any means immaculate), I have tried to tabulate the similarities between the Gospels. I counted every case of absolute verbal similarity, or identical words between texts used the same number of times (regardless of order in which they appear), as 1, even if they appear to have been incidental similarities, just to try to get the highest score of absolute similarity. For words that were different forms of the same word used in multiple texts, I counted those similarities with a score of .75. For words that were synonymous, I counted those similarities with a score of .5. I do not know how best to score similarities based on word order variation, and so when word order varied, I simply marked those cases with a < sign to signify that the actual degree of similarity is less than the simple ratio provided. With these parameters, setting aside John for now, and including Mark 16:9–20, here is how similar the texts are only where they parallel each other (rather than using total word count for each book) in absolute terms (including only when the same words are used the same number of times) and in weighted terms to include the other instances of similarity noted above:
That is still plenty of similar text to consider, but there is no need to exaggerate it with claims of high degree of overlap.
Statistics About Q
I have already made clear elsewhere that I am a Q skeptic. The scholars I cite here are not. While Wright and Bird’s textbook does not give any statistical indication of how much material is addressed by hypothesizing Q, Bird’s other book puts the figure at “Approximately 220 to 235 verses are shared between Matthew and Luke and not present in Mark.”21 This is the same estimate Brown gives.22 Donald Hagner simply says, “Matthew and Luke have a large amount of material in common, known as the double tradition, consisting of more than 230 verses not found in Mark.”23 Metzger skews to the upper end of the aforementioned range in saying, “Matthew and Luke have each about 235 verses in common, comprising chiefly discourse material, which are not in Mark.”24 Carson and Moo outstrip the others in their estimate of “approximately 250 verses common to Matthew and Luke that are not found in Mark.”25
Despite the inconsistencies of these various measures, one feature they all have in common is that none of them cite where these figures came from. They also have an acute lack of specificity about them. I do not mean the fact that they are supposed to be approximations. I mean that none of them tell us if the number of verses derives from Matthew or Luke. The figures are not the cumulative total from both. Since verses are inconsistent divisions—sometimes they match the length of a sentence, but this is often not the case—the versifications of parallel texts are not always the same. Here, I will reiterate and expand on what I wrote on this matter in response to ReligionforBreakfast’s video on the Gospels. Here are the instances where the number of verses is different between parallel texts in Matthew and Luke:
Matt 3:7–10 // Luke 3:7–9
Matt 4:1–11 // Luke 4:1–1326
Matt 5:1–12* [5:5, 7–10 have 0 verbal similarity] // Luke 6:20–26* [6:24–26 have 0 verbal similarity]
Matt 5:25–26 // Luke 12:57–59*
Matt 5:31–32* [5:31 has 0 verbal similarity] // Luke 16:18*
Matt 5:38–42* [5:38, 41 have 0 verbal similarity] // Luke 6:29–30
Matt 5:43–48* [5:43, 47 have 0 verbal similarity] // Luke 6:27–28, 32–36* [6:34 has 0 verbal similarity]
Matt 6:9–13* // Luke 11:1–4* [11:1 has 0 verbal similarity]
Matt 6:19–21* [6:19 has 0 absolute verbal similarity, but one word is incidentally similar in a weighted fashion] // Luke 12:33–34*
Matt 6:22–23 // Luke 11:34–36
Matt 6:25–34 // Luke 12:22–32
Matt 7:3–5 // Luke 6:41–42
Matt 7:15–20* [7:15 has 0 verbal similarity] // Luke 6:43–44*
Matt 7:24–27* // Luke 6:47–49*
Matt 8:5–10, 13 // Luke 7:1–10* [7:5 has 0 verbal similarity]
Matt 8:18–22 [8:18 has 0 verbal similarity] // Luke 9:57–62* [9:61–62 have 0 verbal similarity]
Matt 9:37–38 // Luke 10:2
Matt 10:37–38* // Luke 14:25–27* [14:25 has 0 verbal similarity]
Matt 11:7–11, 14–19 [11:14 has 0 verbal similarity] // Luke 7:24–35 [7:29–30 have 0 verbal similarity]
Matt 11:12–13* // Luke 16:16
Matt 11:20–24 [11:20 has 0 verbal similarity] // Luke 10:12–15
Matt 11:25–27 // Luke 10:21–22
Matt 12:34–35 // Luke 6:45
Matt 13:33 // Luke 13:20–21 [13:20 has 0 absolute verbal similarity]
Matt 16:1–4* [16:1, 4 have 0 verbal similarity] // Luke 12:54–56*
Matt 22:1–14* [22:5–6, 10–14 have 0 verbal similarity] // Luke 14:15–24* [14:19, 22, 24 have 0 verbal similarity]
Matt 23:13, 15–36* [23:16, 19, 24, 28, 34 have 0 verbal similarity] // Luke 11:39–52*
Matt 23:37–39 // Luke 13:34–35
Matt 24:37–41, 43–44 // Luke 17:26–36; 12:39–40* [17:28–29, 32–33 have 0 verbal similarity]
The texts with a * next to them signify cases of less than 50% absolute verbal agreement between the text and its parallel. And if we only count these verses without the other cases of parallels, we have a total of 166 verses for Matthew and 147 verses for Luke (17:36 would not count in most totals). It does not work to simply say they share X number of verses because one or both of them is not going to come to that total. Moreover, we can see that there are cases where the larger units may have parallels, but individual verse divisions have no verbal parallels, including twenty-eight verses in Matthew with 0 verbal similarity to parallel pericopes in Luke (twenty-nine if we include Matt 6:19) and eighteen verses in Luke with 0 verbal similarity to parallel pericopes in Matthew (nineteen if we include Luke 13:20). This is not even to account for verses where there are only incidental similarities to parallels or an otherwise low degree of similarity, but we will simply focus on these cases for simplicity. That leaves us with 137 or 138 verses in Matthew that have any degree of absolute verbal similarity with a parallel pericope in Luke and 128 or 129 verses in Luke that have any degree of absolute verbal similarity with a parallel pericope in Matthew.
Without a source for the tabulations of these scholars, I cannot be entirely certain of what they include, but if I were to draw from my own work, we could also note these texts that have the same number of verses in each:
Matt 3:12 // Luke 3:17
Matt 6:24 // Luke 16:13
Matt 7:1 // Luke 6:37*
Matt 7:7–11 // Luke 11:9–13 [11:12 has 0 verbal similarity]
Matt 7:12* // Luke 6:31
Matt 7:13–14* [7:14 has 0 verbal similarity] // Luke 13:23–24* [13:23 has 0 verbal similarity]
Matt 7:21–23* [7:22 has 0 verbal similarity] // Luke 6:46; 13:26–27* [13:26 has 0 verbal similarity]
Matt 8:11–12 // Luke 13:28–29
Matt 10:26–33* // Luke 12:2–9*
Matt 10:34–36* // Luke 12:51–53*
Matt 10:39* // Luke 17:33*
Matt 11:1–6* // Luke 7:18–23* [7:20–21 have 0 verbal similarity]
Matt 12:40–42 // Luke 11:30–32
Matt 12:43–45 // Luke 11:24–26
Matt 15:14* // Luke 6:39b*
Matt 18:10–14* [18:10 has 0 verbal similarity] // Luke 15:3–7* [15:3, 6 have 0 verbal similarity]
Matt 24:45–51 // Luke 12:42–48 [12:47–48 have 0 verbal similarity]
Matt 25:14–30* [25:16–19, 30 have 0 verbal similarity; 25:23 has 0 absolute verbal similarity] // Luke 19:11–27* [19:11, 14–15, 27 have 0 verbal similarity]
This adds sixty-nine or seventy total verses to each tally. That does lead us to 235 total verses for Matthew (18:11 would not go towards most counts), but it leaves us with 217 verses for Luke. Moreover, we see again that eight more verses for Matthew have 0 verbal similarity with the parallel pericopes in Luke (nine if we count 25:23) and thirteen more verses for Luke have 0 verbal similarity with the parallel pericopes in Matthew. When we add these to the twenty-eight or twenty-nine from the previous tally in Matthew and to the eighteen or nineteen from the previous tally in Luke, thirty-six or thirty-eight verses in Matthew have 0 verbal similarity to Luke and thirty-one or thirty-two verses in Luke have 0 verbal similarity to Matthew. What remains after these verses are subtracted is 197 or 199 verses from Matthew that have any degree of absolute verbal similarity to Luke and 185 or 186 verses from Luke that have any degree of absolute verbal similarity to Matthew (including incidental ones). In each case, the lower number signifies having any absolute similarity, while the higher number signifies having any verbal similarity at all. And this is according to some guesswork about what would be included in the tallies cited earlier.
Stein attempts to use both kinds of statistics: “This material consists of approximately 235 verses. Expressed in more exact statistics, the relationship is as follows: Matthew has 4,290 words that have parallels in Luke but not in Mark, and Luke has 3,559 words that have parallels in Matthew but not in Mark.”27 We have already seen the problems with the verse statistic. For the word figures, he once again cites Tyson and Longstaff.28
Stein gets this figure by adding the parallel and partially parallel rows in Tyson and Longstaff’s tabulations. But again, he ignores the column next to it that identifies unique words in each source. For Matthew, 2,699 words (62.9% of their total) are unique in comparison to parallels in Luke.29 For Luke, 1,968 words (55.3% of their total) are unique in comparison to parallels in Matthew.30
I have already supplied my tabulations of overall similarities between Matthew and Luke. But if we shrink the scope to the aforementioned texts that many prefer to identify as being due to Q, we get the following figures:
Matthew to Luke: <1,678/3,853 (A)[~43.6%]; <1,943.5/3,853 (W)[~50.4%]
Luke to Matthew: <1,678/3,770 (A)[~44.5%]; <1,946.5/3,770 (W)[~51.6%]
Some differences in accounting are probably due to the fact that I included some texts where both Matthew and Luke overlapped with Mark but went beyond Mark, which would be a distinct section in Tyson and Longstaff’s work. But I would not have gotten to any 235-verse total otherwise for Matthew or Luke. In any case, I could not find how one could get to 250, per Carson and Moo. Again, there is still an interesting phenomenon of mixed similarities and differences to consider here—with most parallels skewing high or low in similarity—but there is no need for overegging its extent.
N. T. Wright and Michael F. Bird, The New Testament in Its World: An Introduction to the History, Literature, and Theology of the First Christians (London: SPCK; Grand Rapids: Eerdmans, 2019), 686.
This second verse should be v. 26.
Wright and Bird, New Testament., 687 n. 28.
My comments on the statistics from Wright and Bird also apply to the similar statistics one can find in Donald A. Hagner, The New Testament: A Historical and Theological Introduction (Grand Rapids: Baker Academic, 2012), 132; Stanley E. Porter and Bryan R. Dyer, “The Synoptic Problem: An Introduction to Its Key Terms, Concepts, Figures, and Hypotheses,” in The Synoptic Problem: Four Views (Grand Rapids: Baker Academic, 2016), 7.
Michael F. Bird, The Gospel of the Lord: How the Early Church Wrote the Story of Jesus (Grand Rapids; Cambridge: Eerdmans, 2014), 125–54.
Bird, Gospel, 128, adapting B. F. Westcott, Introduction to the Study of the Gospels, 8th ed. (London: Macmillan, 1895), 179.
William Stroud, A New Greek Harmony of the Gospels (London: Bagster and Sons, 1853), cxvii, 359.
Stroud, New Greek Harmony, 359.
Bird, Gospel, 160.
M. Eugene Boring, An Introduction to the New Testament: History, Literature, Theology (Louisville: Westminster John Knox, 2012), 476 n. 12.
Craig A. Evans, “The Two Source Hypothesis,” in The Synoptic Problem: Four Views, ed. Stanley E. Porter and Bryan R. Dyer (Grand Rapids: Baker Academic, 2016), 35.
Raymond Edward Brown, An Introduction to the New Testament (New Haven: Yale University Press, 1997), 111. Cf. Boring, Introduction, 475–76.
Bruce M. Metzger, The New Testament: Its Background, Growth, and Content, 3rd ed. (Nashville: Abingdon, 2003), 96–97.
D. A. Carson and Douglas Moo, An Introduction to the New Testament, 2nd ed. (Grand Rapids: Zondervan, 2005), 96.
Robert H. Stein, Studying the Synoptic Gospels: Origin and Interpretation, 2nd ed. (Grand Rapids: Baker Academic, 2001), 50, 52; Joseph B. Tyson and Thomas R. W. Longstaff, Synoptic Abstract, The Computer Bible 15 (Wooster, OH: College of Wooster, 1978), 169–71.
By this, they mean “identical words in continuous agreement.” Tyson and Longstaff, Synoptic Abstract, 11.
By this, they mean “identical words not in continuous agreement.” Ibid.
By this, they mean “the same root word with different grammatical inflections.” They thus do not have among these a category for synonyms like I do.
Ibid., 75.
Boring, Introduction, 475 n. 10.
Bird, Gospel, 162.
Brown, Introduction, 111.
Hagner, New Testament, 132. Cf. Porter and Dyer, “Synoptic Problem,” 8.
Metzger, New Testament, 97. Cf. Boring, Introduction, 476.
Carson and Moo, Introduction, 98.
There is a parallel for this story in Mark 1:12–13, but it is obviously significantly condensed.
Stein, Studying, 97.
Tyson and Longstaff, Synoptic Abstract, 169, 171.
Ibid., 169.
Ibid., 171.