Harmonization and Synoptic Similarity

Some scribes made harmonizing alterations deliberately. In most cases, one cannot discern nor safely speculate about the intentionality of a given variant, but occasionally the obvious nature or significant extent of the borrowing from a parallel text strongly suggests a harmonizing impetus.

See Also: Scribal Harmonization in the Synoptic Gospels, NTTSD 60 (Leiden: Brill, 2019).

By Cambry G. Pardee
Visiting Assistant Professor of Religion
Pepperdine University, London
United Kingdom
July 2019

In the gospels section of my introductory course on the New Testament, I expect students to read the four canonical gospels in their entirety. We begin with the Gospel of Mark, followed by Matthew, Luke, and finally John. Inevitably, by the time my students read the Gospel of Luke, their attention to the reading has begun to flag, and it is not unusual to hear one or two complaints about reading “the same” story over and over again. Of course, those familiar with the composition of the Synoptic Gospels understand that these texts are not “the same,” and in fact, each evangelist brings something new to the table—not only new material but also a unique literary genius. It is worth noting, however, that the common impression of lay readers is that the first three gospels are in some sense “the same.” The numerous, sometimes subtle, differences between the first three gospels where they share the same source material are easily overlooked against the backdrop of synoptic similarity.

Among those who copied or studied the Gospels in the first few centuries after their composition (2nd–4th centuries), the unique quality of each book, as well as the particular points where they diverge from one another, would have been more apparent. In fact, the manuscripts from this period testify to the fact that scribes copying a gospel from an exemplar would sometimes alter their newly produced text, deliberately or otherwise, in ways that obfuscated differences and, at least superficially, enhanced the overall sense of similitude between the Gospels. This type of textual activity is referred to as “harmonization” or “assimilation” since the scribe is suspected of bringing competing narratives into harmony by removing incongruities.

The following example of assimilation is illustrative of this scribal activity. The religious elite of Jesus’ day were often offended by Jesus’ radical choice of table companions. According to the Gospel of Mark, on one occasion the scribes and Pharisees asked Jesus’ disciples, “Why does he eat with the tax collectors and sinners?” (Mark 2:16). In the same episode in the Gospel of Luke, Jesus’ opponents asked him directly, “Why do you eat and drink with the tax collectors and sinners?” (Luke 5:30). In an alteration typical of the harmonizing impulse, the scribe of the fourth-century manuscript P88 has added the Lukan words “and drink” to his text of Mark: “Why does he eat and drink with tax collectors and sinners?” This variant first appeared, whether in this manuscript or an earlier one, as a result of the influence of the Gospel of Luke upon the memory of a scribe copying Mark. The reading became quite common later as scribes perpetuated the Lukan reading in the Markan manuscript tradition. This elementary example is representative of the majority of harmonizing variants in the second, third, and fourth centuries.

Deliberate Harmonization

Some scribes made harmonizing alterations deliberately. In most cases, one cannot discern nor safely speculate about the intentionality of a given variant, but occasionally the obvious nature or significant extent of the borrowing from a parallel text strongly suggests a harmonizing impetus.

In Luke 7:35, for example, Jesus says to his opponents, “Nevertheless, wisdom is vindicated by all her children.” In the parallel text in Matthew 11:19, Jesus says similarly, “Yet wisdom is vindicated by her deeds.” In numerous manuscripts of Matthew, the Matthean sentence has been altered to reflect the Lukan form: Luke’s children has replaced Matthew’s deeds. This reading appears frequently in the manuscripts and, given the unusual and unexpected Lukan term children, it clearly reflects the influence of the parallel text.[1] In Codex Vaticanus, the text of Matthew was copied faithfully by the scribe, but a corrector later erased the word deeds and supplied the Lukan term children. In contrast to the numerous other instances of the same variant, it is evident that the substitution in this manuscript was made deliberately as a correction that brings Matthew into agreement with Luke. Transparently deliberate alterations for the purpose of harmonization appear infrequently in the manuscripts, suggesting that most scribes aimed to transcribe faithfully rather than to eliminate disparities.

Some intentional harmonizing variants reveal an exegetical bent in the scribe or reader. Moments before his death on the cross in Matthew’s passion narrative, Jesus calls out, “My God, my God, why have you forsaken me?” (Matt 27:46). His exclamation is mistaken by bystanders as an appeal to Elijah. It is reported that one of those nearby went quickly, “got a sponge, filled it with sour wine, put it on a stick, and gave it to him to drink” (Matt 27:48). Others rebuked the man and said, “Wait, let us see whether Elijah will come to save him” (Matt 27:49). In the Gospel of Matthew, these words are followed directly by Jesus’ death: “Then Jesus cried again with a loud voice and breathed his last” (Matt 27:50). Many manuscripts, including Vaticanus and Sinaiticus, testify to a substantial addition to this scene. Between the objection to the raising of the sponge (v. 49) and the death of Jesus (v. 50), some scribes have added the following words: “And another, taking a spear, pierced his side and water and blood came out.” The evident source of this sentence is John 19:34 where the evangelist records, “Instead [of breaking Jesus’ legs], one of the soldiers pierced his side with a spear, and at once blood and water came out.”

Such an extensive addition cannot have entered the textual tradition originally apart from the deliberate inclusion of these words by a scribe. Some contemporary scholars have suggested that this reading originated as a reader’s note in the margin and that upon subsequent reproduction the variant was absorbed into the main text of the gospel.[2] The juxtaposed imagery of the wine-soaked sponge raised on a stick and the spear thrust into Jesus’ side producing blood and water is poetically and theologically evocative. The originator of the variant was drawing imaginative connections across literary boundaries. This type of deliberate, exegetical harmonization is rare, but the few variants of this sort reveal that some scribes were perfectly aware of how the differences between the gospels complemented or conflicted with one another. Scribes could be creative readers as well as simple copyists and were capable of making insightful intertextual connections between the books they were transcribing.

Reflexive Harmonization

The reader’s note and the correction described above are strong examples of alterations that were made deliberately and which reflect details from gospels other than the one being copied. In the vast majority of cases, however, harmonization did not occur deliberately but rather reflexively. That is to say, as the scribe copied a gospel from a physical exemplar, the text itself recalled alternative versions of the material latent in the scribe’s memory. Occasionally, the version in the scribe’s memory, what Philip Comfort calls the scribe’s “cognitive exemplar,” ended up becoming the text of the new copy of the gospel rather than the text of the actual exemplar.[3] In these cases, harmonizing variants are not evidence of what motivated the scribe but of what influenced him.

Several clear examples of reflexive harmonization can be found in the various versions of the beatitudes. Jesus’ statement in Luke 6:20, “blessed are the poor,” has been augmented in numerous manuscripts to reflect the parallel in Matthew 5:3, “blessed are the poor in spirit.” Similarly, Jesus’ blessing in Matthew 5:4, “blessed are those who mourn,” has been amplified with the adverb “now,” which is twice repeated in Luke 6:21: “blessed are you who are hungry now” and “blessed are you who weep now.” Although the beatitudes are the site of frequent harmonization, no scribe has consistently or thoroughly adapted the text from one gospel to the alternative version in another gospel. Alterations to the beatitudes are isolated and unpredictable. This state of affairs, which is typical of the vast majority of harmonizing variants and reflexive harmonization in general, suggests that the changes do not reflect a deliberate effort to reduce discrepancies between the gospels, but rather the prevailing influence of dominant versions of gospel material.

It is apparent from the examples of harmonization in the beatitudes and from the larger body of evidence that Jesus’ words were particularly prone to alteration under the influence of parallel material. Nearly sixty percent of the harmonizing variants from the 2nd–4th centuries can be found in Jesus’ words rather than in the narrative portions of the gospels or in the speech of other individuals.[4] To some, the notion that Jesus’ words were altered might be unsettling. Did not the early scribes respect the words of Jesus? Their willingness to alter dominical sayings could suggest that Jesus’ words were not held sacrosanct. In fact, just the opposite is true. Jesus’ words were altered because they were revered. The teachings of Jesus were read and copied regularly and in many forms. They were recited from memory in oral performances, gathered in collections of Jesus’ aphorisms, and read aloud from written gospel stories, both canonical and non-canonical. The comparatively high rate at which Jesus’ words are subject to harmonization in the manuscripts demonstrates that any one version of Jesus’ teachings could activate and react with memories of a number of other renditions of Jesus’ sayings. The very popularity of Jesus’ words led to the diversity with which they were transmitted.

Jesus’ words were not the only major context for reflexive harmonization; pithy, poignant, or otherwise remarkable statements easily retained in the memory were also altered harmonistically. A clear example of reflexive harmonization in such a context can be found in Mark 8:29. In response to Jesus’ question about his identity, Peter states, “You are the messiah.” In Matthew 16:16, Peter’s reply is longer: “You are the messiah, the son of the living God.” In the fourth century, the scribe of Codex Sinaiticus was among the earliest to alter the text of Mark under the influence of the extended statement in Matthew. He has written: “You are the messiah, the son of God.” The addition is not a perfect replication of Matthew’s phrase, indicating that it was not created deliberately to reconcile competing versions of the same statement but clearly owes its origin to the version of Peter’s confession recorded in Matthew’s Gospel. The variant represents the memory’s response to the stimulus of the exemplar—a textual-memorial reflex. The scribe read “you are the messiah” and automatically recalled and supplied “the rest” of the sentence, or at least part of it: “the son of God.”

As in the previous example, it was often the text itself which suggested and recalled competing versions of material being copied. In the scene of the transfiguration, Mark records the words spoken from the cloud: “This is my son, the beloved; listen to him” (Mark 9:7). Some scribes reading this sentence as they copied it were automatically reminded of the longer version of this proclamation in Matthew 17:5: “This is my son, the beloved; with him I am well pleased; listen to him.” These scribes then altered their text of Mark by including the Matthean phrase. Their intent was not thoroughgoing assimilation, as evidenced by the fact that the transfiguration episodes in manuscripts with this variant remain plainly Markan. Rather, the variant demonstrates how the text at hand could recall an absent “text” in the scribe’s memory and, eventually, cause the creation of a variant reading.

Closing Remarks

Within the discipline of textual criticism, harmonization has traditionally been described as a practice designed to remove discrepancies between the gospels and to bring them into greater alignment. This, in fact, is not what the scribes accomplished. No manuscript, and therefore no scribe to whom we have access, demonstrates such an agenda in anything approaching a systematic way. In most cases, harmonizing variants are not an indication of the scribes’ dissatisfaction with variations in the written gospels, but of their detailed familiarity with multiple iterations of the gospels. The scribes were steeped in gospel material and never copied one book in isolation from the diverse oral-literary gospel tradition in which they listened, read, and worked. Indeed, copying the text regularly became the impetus for the creation of new readings reflecting the influence of parallel material. Harmonization, then, is not evidence of a scribal motive to remove differences between the gospels but rather of the significant effect the diversity of the gospel tradition had, and continues to have, upon careful readers of the gospels.