Studies in Religion / Sciences Religieuses

Revue canadienne / A Canadian Journal

Volume 28 Number 1 / 1999

Terence L. Donaldson is Professor of New Testament and Biblical Languages, College of Emmanuel & St. Chad, 1337 College Drive, Saskatoon, SK S7N 0W6 (e-mail: donaldson

``For Herod had arrested John'' (Matt. 14:3): Making sense of an unresolved flashback


1 Introduction

In both Matthew (14:3-12) and Mark (6:17-29) the death of John the Baptist is presented as a flashback, an event that had already happened at some point prior to the narrative present. The flashback is triggered by Herod's response to the spreading fame of Jesus; when news of Jesus' activity reached the ruler's ears, he declared that it must be John the Baptist, raised from the dead (Matt. 16:1-2; Mark 6:14-16).

The activity of yet another crowd-drawing prophet might have come as surprising news for Herod; for first-time readers of the gospels, however, the surprising news in these few verses is the death of John. While we have been informed of his arrest (Matt. 4:12; Mark 1:14), we have heard nothing of his death. And so the retrospective account of John's death, which follows immediately after Herod's troubled speculation, serves an essential explanatory purpose, filling us in on the fate of this important character.

In Mark's Gospel, once John's death and burial have been recounted, the narrative returns cleanly to the present. The disciples had been sent out on a mission of preaching and healing (Mark 6:7-13); it is precisely this activity that brings Jesus to Herod's attention (6:14a), precipitating the flashback; once John's story has been brought to its gory end, the narration returns to the present with the account of the return of the disciples (6:30).

In Matthew, however, the flashback remains puzzlingly incomplete. As in Mark, the account of John's death is followed by Jesus' withdrawal to a wilderness place. What prompts this withdrawal, however, is not the return of his own disciples but the report of John's death brought to him by the disciples of John: ``[ John's] disciples came and took the body and buried it; then they went and told Jesus. Now when Jesus heard this, he withdrew from there in a boat to a deserted place by himself'' (Matt. 14:12-13). The opening of the scene clearly establishes a jump backwards in time: ``For Herod had arrested John'' (14:3). But Matthew does not return to the narrative present at the end of the scene; readers are given no instructions to reset their watches to local time. The next narrated events follow in temporal sequence not with Herod's statement but with John's death. The narrative continues on from the death of John as if no temporal jump backwards had happened at all. A few scholars have attempted to avoid this conclusion, arguing that what Jesus ``heard'' (see v. 13) was Herod's interest in him (e.g., Cope 1976; Stock 1994: 243-47). But the grammatical structure found in v. 13 (a participial form of akouo [to hear] without a direct object) is used regularly in Matthew (and elsewhere) to refer to something in the immediately preceding context; another example appears later in the same verse (v. 13b; see also Matt. 2:3; 8:10; 17:6; 19:25; etc.). What Jesus ``heard'' was clearly the report brought to him by the disciples of John (v. 12).1 And so the flashback initiated at Matt. 14:3 fails to return to the narrative present (see Davies and Allison 1991: 463; Hagner 1995: 417).

Traditional methods of interpretation (i.e., source and redaction criticism) have been able to provide a plausible account of how this peculiar Matthean wrinkle in time came to be. Because Matthew has already recounted the sending out of the disciples in Chapter 10, he found it necessary to omit Mark's account of their departure and return in Chapter 14, where except for this omission he follows Mark's sequence quite closely. Matthew's redactional activity here is well known, and need not be described in detail. Traditional methods of interpretation, then, lead to this result: in the process of carrying out his redactional purposes, Matthew inadvertently created a narrative blunder that he failed to rectify. His redactional purposes made it necessary for him to omit Mark's account of the disciples' return, leaving the departure to the wilderness as the next narrated event. But by stitching the two, newly adjacent units together without taking due account of the flashback currently underway, he produced a jarring disruption in the pattern of his narrative fabric as a whole.

But questions of origin are not the only questions that we might want to address to the text. Historical methods, hobbled by the assumption that the only pertinent questions are those that can be answered by their own canons of interpretation, provide us with very little guidance as to what to do with such a redactional seam once we have recognized it. What effect does it have on the flow of the narrative? How does it alter the experience of reading? What sense might we make of it as readers of the gospel narrative? Such readerly questions remain, even when the historical methods have been fully applied. These are the questions I would like to address in this paper.

Such questions fall into the domain of the more recently developed approaches of narrative criticism and reader response criticism. In these approaches, the locus of meaning has shifted from the historical (the intention of the author vis-à-vis the intended readers, as it can be inferred from a reconstruction of the whole process by which the text came to be) to the narratological (the elements and structures of the story as it is narrated in the text)2 and the experiential (the experience of reading, as an actual reader experiences the story under the guidance of the narrator; see especially Bible and Culture Collective 1995: 20-69; Iser 1978; McKnight 1985). My intention, then, is to examine Matthew's incompleted flashback within the interpretive framework provided by these approaches.

2 Flashbacks and readers

To ask what sense a reader might be able to make of this textual element inevitably raises the antecedent question: which reader? For current study of the gospels is populated with a whole array of readers, who are assigned a correspondingly broad spectrum of roles in the production of meaning. Rather than developing and defending any particular theory of reading, however, I prefer to begin with the problem itself, and to make use of those concepts and tools that seem to offer some constructive usefulness.

Let us begin, then, by asking what options might be open to a reader of Matthew who arrives at the point of this narrative disjunction. One possibility, of course, is to fail to observe it at all. The episodic nature of the Synoptic Gospels, coupled with the natural psychological tendency for expectation to shape perception, often leads to precisely this result, as can be confirmed empirically by looking at the commentaries or by questioning people in the pew. Another possibility would be simply to recognize it as a narrative flaw and then to ignore it. Again the episodic nature of the Synoptic narratives is pertinent; the vagueness of Matthew's temporal indicators means that the precise chronological sequence of the episodes plays a minimal role in the overall unfolding of the story. The Synoptic Gospels are often read as a series of only loosely connected episodes. Consequently, readers' awareness of this temporal dislocation may not have any real impact on their construal of the story. Moreover, awareness of secondary flaws does not necessarily detract from an aesthetic experience.3 This flawed connection might be ignored as easily as the recurring ``then'' (e.g., 11:20; 12:22, 38) and ``at that time'' (e.g., 11:25; 12:1; 14:1) that serve to link so many of the other Matthean episodes with what precedes.

But are these the only options -- remaining oblivious to the temporal disjunction or disregarding it as a negligibly significant flaw? Here I suggest that the narrative critical concept of the ``implied'' or ``ideal'' reader might be useful, at least for heuristic purposes. Kingsbury defines the implied reader as ``an imaginary person who is to be responding to the text at every point with whatever emotion, understanding or knowledge the text ideally calls for'' (Kingsbury 1988: 38). At first glance, this concept appears to be of limited value. How can one talk of a reader who follows perfectly the guidance of the narrator in an instance when the narrator seems to be so clumsy and absent minded? Iser's more active reader, who actualizes the potential of the text by filling in its gaps and indeterminacies, might seem to offer much more promise. But it is precisely this trusting reliance on the narrator as exhibited by the ideal reader that I would like to capitalize on here.

My suggestion, then, is that as an experiment in reading we look at this aspect of Matthew's Gospel through the eyes of the implied reader. What sense might be made of this section of Matthew by such a reader, one who trusts the narrator's guidance implicitly and constructs the meaning of the story on the basis of the unfolding narration? The answer is readily apparent. Such a docile reader would believe what the narrator says at 14:13 (``Now when Jesus heard [about the death of John], he withdrew from there''), and would assume that the subsequent event, the feeding of the multitude, also took place prior to the event involving Herod. That is, the reader would assume that the flashback initiated at 14:3 continues past 14:13, and would read on in the expectation that the completion of the flashback was yet to come. It will be my contention that such a search for closure produces a fresh reading experience that makes the experiment worthwhile.

Before proceeding further with this reading, however, it will be useful to say a little more about the phenomenon of flashbacks in general. The requisite conceptual apparatus has been provided in large measure by Gérard Genette, in Chapter 1 of his Narrative Discourse: An Essay in Method (1980). Genette treats what in popular parlance is termed a flashback as one category of what he calls anachronies, instances where the order of narration differs from the actual chronological order of the events of the story. A retrospective anachrony, or flashback, he calls an analepsis (in contrast to a prolepsis, which anticipates a future element). The extent of an analepsis is the length of time occupied by the events that it comprises, while its reach is the length of the backwards leap in time. Those that reach back to a point prior to the temporal limits of the story are external analepses; internal analepses, by contrast, refer to an event lying within the time frame of the story as it has already been narrated. In these terms, our passage of interest is an internal analepsis, whose reach (since John the Baptist is still alive in Chapter 11) stretches over the indeterminate but probably brief period of time covered by the events of Chapters 11-13.

What is more germane to our particular question is the distinction Genette makes between partial and complete analepses. A partial analepsis is of the kind represented by Mark's account of John's death, where the analepsis is terminated by a corresponding jump forward in time. Here the purpose is usually to provide the reader with some necessary piece of information; such explanatory temporal digressions generally do not interfere with the unfolding story line. A complete analepsis, by contrast, is one whose extent equals its reach, that is, where the narrative continues on from the beginning of the analepsis until it rejoins the temporal point of departure. In these terms, we can say that as a result of some apparently clumsy handling of his source material, the author of the First Gospel has turned the partial analepsis of Mark into what by the middle of Chapter 14 the reader will perceive as either a partial analepsis with a longer extent (anticipating a jump forward to the time of Herod's concern), or a complete analepsis (anticipating a full narration of the intervening period and a rejoining of the original narrative time).

Extended analepses of this sort open up possibilities of more sophisticated narrative techniques. Narratives that begin in medias res, like the novel The English Patient or the film Forrest Gump, are good examples. At the same time, however, since the smooth merging of two temporal lanes of traffic is not easily accomplished, complete analepses ``present an obvious risk of redundancy or collision'' (Genette 1980: 50). Genette discusses several ways of overcoming the awkwardness inherent whenever there is a need to steer the narrative vehicle back into the main lane of traffic. An author can attempt to camouflage the seam, or, alternatively, can acknowledge it fully, finding a way ``to extract from this awkwardness a sort of playful charm'' (1980: 64). Another approach, which will turn out to be more relevant for the particular case under discussion, is to ignore the point of merger entirely, and to allow the secondary temporal sequence to overtake the first without marking the closure of the analepsis in any explicit way. In his description of an example from Proust, Genette makes a comment that is doubly interesting for our purposes, in that he seems to allow for the possibility that the absence of an explicitly acknowledged point of merger is due to an oversight on the part of the author. He writes: ``Whether trick, oversight, or offhandedness, the narrative thus avoids acknowledging its own footprints'' (1980: 66; italics original).

Such narrative techniques were not unknown in antiquity. Genette points out that the Iliad begins in medias res (1980: 35). One might also mention Josephus, who, in weaving together the strands of his complicated narrative, on several occasions develops the kind of ``complete analepses'' described above. In Ant. 15:202-208, for example, Josephus leaves off his account of Herod's post-Actium relationship with Ocatavian to pick up the story of Mariamme, who had been left at Masada when Herod had first departed to seek Octavian's favour. The story of Herod's subsequent estrangement from Mariamme (209-46) overlaps considerably with the period already recounted in 198-201 (see also War 1:431-44; Ant. 18:127-204).

Biblical narrative, however, tends to be less sophisticated in this regard. Flashbacks are generally short and contained (partial analepses, in Genette's categories), and are used for simple explanatory purposes (e.g., 2 Sam. 4:4; Esth. 2:6-7; Luke 2:36-37; Acts 8:9-11). Only rarely do we find analepses of any significant (textual) length (see Jdt. 8:2-8; 1 Macc. 8:1-16). The kind of more extended -- even completing -- analepsis which we are setting out to look for in Matthew's narrative is not the sort of thing that any of Matthew's original readers would have been inclined to anticipate or recognize. This means that our implied reader, who takes the narrator's guidance at face value and assumes that the flashback begun at Matt. 14:3 continues beyond v. 13, is not to be seen in first-century terms. I am not assuming that the reading strategy proposed here is one that would have been envisaged by the actual author, nor that it would necessarily have been adopted by any first century reader. The implied reader that I am proposing here, whose reading of the text is characterized by what might be called ``deliberate naïveté,'' is perhaps inferred more than implied, that is, is implied by the text only for interpreters who are familiar with more complex narrative techniques than were generally in use in the first century.

I propose such a reader, however, not to develop any new theoretical category, but to see what sense might be made of this curious feature of Matthew's Gospel. Let us return to the text, then, and begin to read this section through the eyes of our hypothetical reader.

3 The search for closure

We begin not with the flashback itself, but with the little narrative unit on which it depends -- the account of Herod's interest in Jesus (14:1-2). In contrast to Mark, Matthew's account is striking for its lack of connectedness, its almost gratuitous character (another result, the redaction critic would say, of Matthew's omissions of Markan material here). In Mark, Herod's interest is the direct result of preceding events; fanning out through the countryside, the disciples extend Jesus' ministry of preaching and healing, so that it inevitably comes to Herod's ears. While Matthew contains the vague comment that a report about Jesus' mighty works (14:2) reached Herod, this report is not directly precipitated by anything in the preceding narrative; indeed, in the immediately preceding section of the narrative, accounts of Jesus' wonderworking activity are noticeably scarce. The bulk of chapter 13 is given over to parables, and the subsequent short narrative of Jesus' visit to his hometown (13:54-58) stresses instead the absence of mighty works (v. 58). In contrast to Mark, then, readers of Matthew are left wondering why any mention is made of Herod's puzzlement at all. To be sure, it provides the hook on which the flashback is hung. But at the level of the narration, the flashback simply functions to explain why Herod thought of Jesus as he did. The flashback is just the subordinate clause in the narrative grammar; the main clause is the account of Herod's interest in Jesus, and it should make narrative sense on its own. Apart from its role in setting up the flashback, then, what narrative function is served by this detail about Herod? Of what cause-and-effect sequence is it a part?

In Mark, Herod's interest can be understood primarily as effect, i.e., as the result of preceding events. While a possible connection with subsequent events is not eliminated, the presence of this detail can be sufficiently accounted for by linking it primarily with what precedes. The fact that news of Jesus' activity reached Herod can be read simply as an indication of how energetically the disciples took up their commission. In Matthew, however, the detail is introduced without warning, so that the reader is likely to deal with it as a cause, i.e., to expect that its significance will become clear as the narrative unfolds. Since there seems to be no immediate reason to mention Herod's interest in Jesus, the reader expects something to be made of it in what follows. As the flashback begins, then, the reader is in an anticipatory frame of mind, expecting that Herod's interest in Jesus will have some impact on subsequent events.

This expectation, of course, will be shaped to a significant extent by the content of the flashback. Matthew's readers have already been instructed to be suspicious of Herod the elder and his family (Chapter 2). These suspicions will certainly be confirmed by the younger Herod's execution of John. As a result, since Herod has identified Jesus with John, the reader will inevitably fear that Jesus is now in danger of experiencing the same fate as John. In other words, upon reaching 14:12 our reader will have come to expect that Herod will attempt to do with Jesus what he has already done with John, and -- derivatively, but significantly -- that this is why Herod has been mentioned. It is this expectation that will guide our reader as he or she attempts to bring closure to the analepsis that seems to continue past 14:13. That is, from this point on our reader will be on the lookout for the reappearance of Herod the tetrarch, an event that would also signal the end of the flashback and the return to the narrative present. Such an event would by itself bring closure to both unresolved elements -- the otherwise gratuitous mention of Herod's interest in Jesus, and the flashback itself.

Our experiment in reading begins in a confirmatory way. While Chapter 13, as we have seen, contains little in the way of mighty works that would have accounted for Herod's comment about Jesus, the episodes immediately following the retrospective account of John's death are quite different. This material -- that our obedient reader would take to be part of an extended flashback -- is replete with accounts of Jesus' mighty works: the feeding of the multitude (14:14-21); the walking on the sea (14:22-33); and, in particular, widespread healing activity in Genessaret, part of Herod's own territory (14:34-36). Our hypothetical reader could easily take this as confirmation that the narrator's guidance at 14:3 is to be taken at face value: after John's death, Jesus carried out a series of mighty works that eventually brought him to the attention of Herod himself. That is, because of its content, our reader is encouraged to see the remainder of Chapter 14 as part of an extended flashback, and thus to expect the closure of the flashback to come with the reappearance of Herod at some subsequent point in the narrative.

Despite this auspicious beginning, however, the experiment in reading is doomed to failure. Herod himself does not reappear in the narrative at all. The only hint of his presence is the reference in 22:16 to the Herodians who accompany the Pharisees in an attempt to trap Jesus. While this action of Herod's followers is fully in keeping with what the reader expects of Herod himself, by this point it is clear to even the most docile and trusting reader that the flashback has not been explicitly resolved. Jesus' departure from Galilee for Judea (19:1) represents the last possible point at which the temporal lanes might merge. By the time the Herodians are confronting Jesus in Jerusalem, it is clear that the narrative time clock has advanced well beyond the point reached by 14:1.

But while the project inevitably fails, it nevertheless has an effect on the experience of reading that makes the attempt worthwhile. Herod himself might not reappear, but other elements in the passage do, or at least are echoed in ways that our expectant reader would find highly suggestive. My point, then, is that although the reader's expectations are ultimately disconfirmed, the desire to look for closure serves to draw attention to several elements in the subsequent narrative and to increase their impact on the overall reading experience.

While our reader is on the lookout primarily for the reappearance of Herod, the context and content of Herod's first appearance suggest several subsidiary elements that would be associated with Herod in the reader's mind. Included here are references to Jesus' spreading fame, speculations about his identity, similarities with John the Baptist, and (less forcefully) opposition to Jesus on the part of ruling authorities in general. The presence of any of these elements would cause our reader's ears to prick up, alert to the possibility that the narrative is heading back into the thematic territory from which the flashback took its point of departure.

The first such passage appears two chapters later. In 16:13 Jesus himself poses a question that resonates with several of these themes. ``Who do people say that the Son of Man is?'' he asks, touching at the same time both on the question of his identity and (since the question presumes that people are talking about him) on the fact of his spreading fame. Moving on in the text, our reader's interest, already aroused by Jesus' question, will be dramatically heightened by the disciples' response: ``Some say John the Baptist.'' Since we have already heard Herod ``say'' this precise thing (14:2), our reader will be inclined to read this episode with 14:1-12 as its immediate background. Indeed, the connection between the two passages is more than simply verbal, in that it is only the death of John, reported in the earlier passage, that makes the later identification of the two understandable; no one would have thought to identify Jesus with John when the latter was still alive. A reader actively seeking to fill this particular gap in the narrative (see Iser 1978: 173-203), then, will inevitably recognize the connections between the two passages, and thus will read on in anticipation that the extended flashback might be about to come to an end.

This expectation receives some confirmation as the episode unfolds. Jesus' declaration that he will undergo great suffering at the hands of the politically powerful and be killed (16:21) confirms suspicions that he will meet the same fate as John. But there are disconfirming elements as well. There is no mention of Herod among the list of Jesus' opponents; his suffering and death will be caused by ``the elders and chief priests and scribes.'' Further, the location of these events is not Herod's Galilee, but Jerusalem, to which Jesus ``must go.'' Neither detail is incompatible with the possibility of Herod's involvement, of course, but already the anticipated events are assuming a shape different than what had been expected at 14:13.

This eclipsing of Herod in Jesus' own description of his coming suffering might serve to direct the reader's attention to another possibly disconfirming element in the episode. In their reply to Jesus' question (v. 14), the disciples say that some (hoi men) identify Jesus with John. While the identification of Jesus with John necessarily calls to mind the earlier passage where Herod makes this connection (14:1-2), in this passage the identification is made not by ``some'' but by a single individual, Herod himself. There is nothing in the earlier passage to suggest that the opinion was held at that time by anyone else, and no reason to believe that the opinion originated with anyone other than Herod. Since the disciples' report in 16:14 seems to reflect a more widespread opinion, our reader might be inclined to wonder whether the narrative has not already moved beyond the temporal point reached with Herod's musings in 14:2 (the opinion being spread by Herod's servants, for example [see 14:2]). That is, our reader might begin to suspect that the flashback will not reach closure in any explicit way, but that the two temporal lanes have already been rejoined in an unmarked merger somewhere between 14:13 and 16:13. Even so, at this point it would be no more than a suspicion, and there would still be unresolved issues -- the reason for the unprompted appearance of Herod, for example -- that would drive our reader on.

Despite these disconfirming elements, however, this reading project has already had a positive result, in that it has served to heighten our reader's engagement with a passage that on any reading of the gospel has an important part to play in the narrative. As Kingsbury and others have observed, with its formulaic ``from that time on Jesus began to'' and its introduction of the theme of Jesus' suffering, 16:21 marks an important transitional moment in the discoursing of the story (Kingsbury 1975: 1-25, 1988: 40-93). Just as the first appearance of this temporal phrase at 4:17 serves to introduce Jesus' public ministry (4:17-16:20), so the second appearance serves to introduce the climactic final section (16:21-28:20) in which Jesus' story reaches its conclusion with his death and resurrection. Because of the desire to bring closure to the flashback, our reader arrives at this pivotal verse in a state of full alert, ready to receive and to engage with just the type of theme that is highlighted in the passage. Unlike Peter, our reader is ready to contemplate the possibility of Jesus' suffering and death (see 16:22). Our reader may also be ready to identify and to enjoy the irony in Jesus' anticipation of his own resurrection -- that what Herod postulates about John to make sense of Jesus (i.e., his having been raised from the dead) will in the end turn out to be true of Jesus himself (16:21).

In order to get a more complete sense of the effect of our project on the experience of reading, however, we need to move on a little further with our reader. The next episode in the narrative is the account of the transfiguration, and here our reader encounters more material that resonates with 14:1-12. The material of interest is found in the discussion between Jesus and his disciples on the way down the mountain, a discussion having to do with the identity and fate of Elijah. In response to a question about the end-time appearance of Elijah, Jesus declares that ``Elijah has already come,'' that ``they did to him whatever they pleased,'' and that ``so also the Son of Man is about to suffer at their hands'' (17:11-12). In contrast to Mark, the narrator in Matthew clarifies these enigmatic statements, indicating that Jesus was really speaking about John the Baptist.

Because of the references to John, to suffering, and to their having done to him ``whatever they pleased,'' our reader will once again be inclined to read the passage against the background of 14:1-12. Two aspects of the post-transfiguration passage are of particular interest when read in this way. The first is the explicit connection between John and Jesus in their suffering (``so also [houtos kai] the Son of Man is about to suffer''). The attempt to make narrative sense of the reference to Herod in Chapter 14 has led our reader to suspect just such a connection between John and Jesus, and the passage containing the next reference to John (16:13-21) has tended to confirm these suspicions. But to this point the connection has remained implicit. In 17:13, however, the connection is confirmed by the explicit authoritative voice of the narrator: in saying that the Son of Man was to suffer in the same way as ``Elijah,'' Jesus ``was speaking to them about John the Baptist''; thus Jesus and John are to suffer the same fate. Indeed, this explicit linking of John and Jesus in 17:13 has an important part to play in the gospel as a whole. While throughout the gospel Jesus and John are presented as parallel -- albeit unequal -- figures (Davies and Allison 1991: 289-90), this is the only point at which the parallelism is made explicit. Our reader's encounter with this verse, then, serves as a clarification of instances of parallelism already encountered (cf. 3:2 and 4:17; see also 11:16-19) and as preparation for further instances yet to come (cf. 21:26 and 21:46; 14:5 and 21:46).

The second element of interest is the use of the generalizing ``they'' and ``their'': they did to John whatever they pleased; so will the Son of Man suffer at their hands. Because of information already received, the reader knows that at one level those who were responsible for the death of John and those who will be responsible for the death of Jesus are different -- Herod in the first case, the ``elders and chief priests and scribes'' in the second (16:21). But Jesus' way of speaking here tends to eliminate any significant distinction between the two and to identify them; it is one and the same group (``they'') that are ultimately responsible for the deaths both of John and of Jesus.

The result of this identification is to bring a kind of closure after all to the seemingly gratuitous appearance of Herod. Because of the nature of his first appearance in Chapter 14, our reader expects Herod to reappear. As we have noted, this expectation is frustrated -- at least at the literal level. But Jesus' use of the generalizing third person plural in 17:11-12 allows the reader to conclude that Herod does reappear in another guise, in that Herod is part of a larger group opposed both to Jesus and to John. This reappearance, in the form of the ``elders and chief priests and scribes'' whom Jesus identifies as those who will be responsible for his own death (16:21), does not bring the flashback to an end, of course. But it does provide the reader with the resources -- once the reader has realized that the flashback will not achieve closure in any explicit way -- to bring closure to the other element left open by 14:1-2, namely, the role of Herod. Herod's appearance need not be seen as gratuitous and anomalous, even though he does not reappear in the story as initially expected. Since Jesus combines into a single group both Herod, who caused John's death, and the religious leaders who are to orchestrate his own, there is a necessary connection between the appearance of Herod and later appearances of the ``elders and chief priests and scribes,'' whom Jesus has identified (16:21) as the ones at whose hands the Son of Man will suffer (17:12). Any unfinished business with respect to Herod's appearance will be taken care of in the various appearances of these groups in connection with Jesus' own passion.4

At least by 19:1, then, and probably by 17:13, it will be clear to our reader that the flashback will not reach closure in any explicit way. Our reader will have to conclude either that the author has made a mistake, or (trusting the narrator to the end!) that the temporal lines have already merged quietly somewhere between 14:13 and 16:13. But this does not mean that our experiment in reading has been without value. For despite its ultimate failure this reading strategy has had significant penultimate effects: setting up anticipations that shape the experience of reading in distinct and effective ways, privileging certain episodes, forging new connections, and bringing key Matthean themes forcefully to the attention of the reader. While such a reading strategy does nothing to alter the facts of Matthew's narrative blunder, it nevertheless justifies itself in the richer appreciation of the story that it generates.


Cope has attempted to avoid this conclusion by arguing that the combination of gar (v. 3) and de (v. 13) is a customary way of marking the beginning and end of an explanatory aside, citing as parallels Mark 7:3-4; Gal. 6:2-6 and Josephus Ant. 18:116-19 (which coincidentally deals with the death of John). But this argument does not withstand analysis. An investigation of similar explanatory flashbacks in related literature reveals that while gar often appears at the outset, de is by no means the common way of bringing the flashback to a close. In addition to de (Acts 20:16-17), one finds oun (John 9:22), kai (2 Sam. 3:22; Jdt. 4:3; 8:2-8), kai nun (Josh. 2:10-11; Mark 6:17-30), and te (Acts 21:29-30); the passage cited by Cope is the only example of de in Josephus (for counter-examples, see War 1.431-44; Ant. 15.202-208; 18.127-225; Life 179-84). And in every case, it is the context itself, rather than any grammatical signal, that determines the end of the flashback.
The literature is rapidly growing, but the approach has developed in dialogue with such works as the following: Booth 1961; Chatman 1978; Genette 1980; Stevick 1964. For New Testament studies, see, e.g., Culpepper 1983; Kingsbury 1988; Powell 1990.
My 15 year-old son, who loves the Star Wars movies, also loves discovering and pointing out flaws in the filming. This only seems to add to his enjoyment.
The three groups are mentioned at 26:57 and 27:41. Different pairs of the three terms appear in 21:15; 26:3, 47; 27:1, 3, 12, 20. The parallelism between Herod and these religious leaders will be further confirmed in the fact that the chief priests and Pharisees are reluctant to arrest Jesus because they feared the crowds, who held Jesus to be a prophet (21:46); Herod was reluctant to execute John for a similar reason, expressed in almost identical language (14:5).



Bible and Culture Collective 1995 The Postmodern Bible. New Haven: Yale University Press.

Booth, Wayne C. 1961 The Rhetoric of Fiction. Chicago: University of Chicago Press.

Chatman, Seymour 1978 Story and Discourse: Narrative Structure in Fiction and Film. Ithaca, NY: Cornell University Press.

Cope, Lamar 1976 ``The death of John the Baptist in the Gospel of Matthew, or, the case of the confusing conjunction.'' Catholic Biblical Quarterly 38: 515-19.

Culpepper, R. Alan 1983 Anatomy of the Fourth Gospel: A Study in Literary Design. Philadelphia: Fortress Press.

Davies, W. D. and Dale C. Allison 1991 A Critical and Exegetical Commentary on the Gospel according to Saint Matthew, vol. 2. International Critical Commentary. Edinburgh: T. & T. Clark.

Genette, Gérard 1980 Narrative Discourse: An Essay in Method. Translated by Jane E. Lewin, Foreword by Jonathan Culler. Ithaca, NY: Cornell University Press.

Hagner, Donald A. 1995 Matthew 14-28. Word Biblical Commentary. Dallas: Word.

Iser, Wolfgang 1978 The Act of Reading: A Theory of Aesthetic Response. Baltimore: Johns Hopkins University Press.

Kingsbury, Jack Dean 1975 Matthew: Structure, Christology, Kingdom.

1988 Matthew as Story. 2nd edition. Philadelphia: Fortress Press.

McKnight, Edgar V. 1985 The Bible and the Reader: An Introduction to Literary Criticism. Philadelphia: Fortress Press.

Powell, Mark Allen 1990 What is Narrative Criticism? Minneapolis: Fortress Press.

Stevick, Philip, ed. 1964 The Theory of the Novel. New York: Free Press.

Stock, Augustine 1994 The Method and Message of Matthew. Collegeville, MN: Michael Glazier.