Where the world comes to study the Bible

John 5.2 One More Time: A Response to Andreas Köstenberger

Related Media

June 15, 2007

Andreas Köstenberger has recently responded to my essay, “John 5.2 Once Again,” posted on bible.org. His response can be found at the Biblical Foundations website (http://www.biblicalfoundations.org/?p=123). I appreciate his interaction with my article on “John 5,2 and the Date of the Fourth Gospel” (Biblica 71 [1990] 177-205). Indeed, I applaud him for interacting with the data and making a case for the present tense of εἰμι being used to refer only to past time. He is absolutely right that this kind of argument is needed to counter the arguments I made in my Biblica article. I only wish that more scholars were like him and were willing to interact with the data rather than simply write off a view that one disagrees with.

I’m on Patmos right now, photographing manuscripts at the Monastery of St. John the Apostle and must necessarily keep my comments brief. I also do not have access to a library here (ironically enough), so I have to speak from memory about certain issues. But I wish to underscore two or three points in Dr. Köstenberger’s blog.

Response to the Grammatical Arguments

First, his argument attempts to paint me as thinking that the present tense is the only basis on which I believe that John was written before 70. He said, “the tense form of a single Greek verb in John’s Gospel is hardly able to bear the heavy weight Wallace puts on it in proving the date of composition of the entire Gospel.” And, “I do not expect that many students of John’s Gospel will believe that the date of John’s Gospel can be established by a disputed use of one single verb tense in the Gospel” (italics added). But that is not a careful reading of my article in Biblica. In the final paragraph (of the prepublication draft—all that I have access to right now) I say:

By way of concluding remarks, what we have attempted in this paper is the resuscitation of an old argument which had been left for dead several decades ago. As I mentioned in the introduction, if it could be demonstrated that the arguments against the prima facie evidence in John 5:2 were wrong, such a demonstration might well reveal—on this one strand of evidence—the inherent improbability that John's gospel was published after AD 70. I will be the first, however, to admit that the arguments advanced in this paper are not air-tight. Nevertheless, I hope that we have shown that John 5:2 deserves greater consideration in future discussions over the date of this gospel.

Here I have to register a strong complaint: I clearly admitted that I was not basing my belief that John’s gospel was a pre-70 document exclusively on ἐστιν in John 5.2. At most, what I said was that “if it could be demonstrated… [a point I said was not conclusive]… this one strand of evidence… [would show] the inherent improbability that John’s gospel was published after AD 70.” Immediately after that, I conceded, “I will be the first, however, to admit that the arguments advanced in this paper are not air-tight.” Often I write in such a way as to focus on an issue that has been overlooked or simply dismissed, when I believe it should be given more weight. To represent me as saying that I considered the matter closed is not accurate. Indeed, my desire has been (for such essays) to open a dialogue, to get others to think about the implications, and to wrestle with the possibility that the evidence I am laying out may point to a larger issue, which could possibly bring about (with many other considerations, of course) a paradigm shift in how a particular text is viewed. I am sure that Dr. Köstenberger did not intend to misrepresent me, but this he did just the same.

What he has done is to assume that I was presenting (what I believed were) air-tight arguments and that my belief of John’s early date was solely based on the verb in this one verse. By so doing, all he needed to do was to show that my argument was not conclusive, and thus, he could dismiss my conclusions. The comments to his blog confirmed this: “I thought it was inappropriate [for Wallace] to build an argument on one verse”; “Wallace’s singular reliance on a debatable use of a single verb tense…” Yet Dr. Köstenberger did not correct the impression that these readers had, suggesting that he believed that ἐστιν in John 5.2 was the only thing that I’ve ever considered regarding the date of John! Even in the Biblica article I listed other arguments (and not a few scholars) for a pre-70 date for John. I would urge the reader to examine that article for the whole presentation.

Second, he said, “I continue to believe that we must be careful not to dismiss too quickly the possibility that a present tense form of eimi may be used in biblical narrative within a temporal scope that includes the past.” I have never disagreed with this viewpoint (except for instantaneous present indicatives, virtually all of them would include past time to some degree), but it hardly proves his point. What is needed is a present tense that refers to the past and only to the past. So, no matter how many “extending-from-past presents” he finds in Koine Greek, none of these overthrows the prima facie meaning of ἐστιν in John 5.2. Obviously, if John meant that the pool of Bethesda was still standing, to say so would necessarily imply that it had been built previously. Köstenberger’s argument from John 10.8 thus is irrelevant to the problem in John 5.2.

Third, he makes a better case in some respects from John 19.40 (“as is [ἐστίν] the custom of the Jews”). He argues that Jewish burial customs did change sometime during the second half of the first century, a point which he believes would invalidate my argument. But he does not address the key datum that would bolster his argument: when exactly did the Jewish burial customs on this particular point change? Would it have been in the 60s—the date that I believe John’s Gospel was written? Or is Köstenberger assuming a 90s date and thus providing evidence that by this time the Jewish burial customs had definitely changed? If so, his point becomes irrelevant for the date that I hold for John. Further, he argues along the lines of possibility here rather than certainty. And this raises a methodological point when it comes to grammar: If we want to see a particular syntactical category (such as historical present) being used in a certain passage, then to avoid the charge of reading the text through a theological lens, we need to provide clear examples of such usage elsewhere. To argue from mere possibility on one side as that which defeats probably on the other is logically fallacious. Specifically, to see John 19.40 as speaking of a past time if John were writing in the 60s, and to see the verb here as a historical present (when it is hardly treated that way in the majority of grammars or translations) is to beg the question. What Dr. Köstenberger would need to do is to provide clear, unambiguous evidence of εἰμι used as a historical present in Koine Greek. That is something that he has not done. Otherwise, a proper grammatical method—and not just the date of John—is at stake.

Broader Arguments for an Early Date of John’s Gospel

Allow me to offer some other observations about the date of John’s Gospel that may be pertinent for the overall discussion. This is intended to make explicit some of the reasons why I think that John’s Gospel was written in the 60s (and to offer a corrective that I am basing my views solely on a verb tense in one verse). Again, these comments will have to be brief—bare outlines of points, if you will—since I do not have access to my library while I am on Patmos.

First, the discovery of P52 has had some bearing on the date of the Fourth Gospel. Although Brent Nongbri recently argued that P52 is irrelevant for the dating of the Gospel of John, he is basing his views on what is possible, but not on what is probable.1 The likelihood that this fragment really belongs to the first half of the second century—and most likely to the first quarter of the second century—gives parameters as to when John’s Gospel could have been written. This tiny scrap of papyrus nailed the coffin shut on F. C. Baur’s dialectic view that John’s Gospel was not written until after 160, and probably after 170. As one wag put it, “The ink on the Gospel must have been barely dry before the scribe of this papyrus began to copy it.”

Now, to be sure, Köstenberger and others who date John in the 90s would have no objection to this point. But there is a larger issue in view here: The fact that many scholars dated John in the early decades of the second century until P52 was discovered shows the fallacy of depending on the majority of scholars for an argument. As William Lane was fond of saying, “An ounce of evidence is worth a pound of presumption.” P52 provides an ounce of evidence that does indeed bear the weight of a first-century date for the original of the Fourth Gospel. Köstenberger appeals to the majority of scholars (not as an argument in itself, but as one that at least shows that collective opinions in this matter must be taken seriously). What he does not note is that the majority of scholars also hold to a post-70 date for Matthew and Luke, a factor that plays heavily into this whole discussion, as we will see below.

Second, until the Dead Sea Scrolls were discovered, scholars tended to view John’s Gospel as thoroughly Hellenized in its outlook. And that, of course, pushed the date of John later, rather than earlier. This, too, was in a largely subconscious way dependent on Baur. But the Dead Sea Scrolls demonstrated that the mindset seen in John is Jewish. Hence, the date of John now became almost up for grabs. What I am a bit puzzled by is that even with the discovery of P52 and the Dead Sea Scrolls—two finds that should significantly impact the date of John—most scholars have hardly budged on their date for this Gospel.

Third, there is the patristic testimony that says that John’s Gospel was the last one to be written. Now, if Matthew and Luke were written in the 70s or 80s, and if the patristic testimony is accurate about John, then the matter is closed: John cannot have been written in the 60s. But if either of these points can be challenged, then the date of John is still open. Köstenberger, as we noted above, enlists the majority on his behalf. He says, “Not that the majority is necessarily always right in biblical studies, but it must be said that there are good reasons why the virtual consensus of Johannine scholars holds to a post-AD 70 date.” What he does not note here is the majority also holds to a post-70 date for Matthew and Luke. But that, too, has been strongly challenged by J. A. T. Robinson, Colin Hemer, and many others. One of the problems with citing majority opinion on one matter is that the presuppositions and beliefs of a given author may be out of sync with the majority on several of the supporting arguments. Thus, if Köstenberger (or another author) holds to a pre-70 date for Matthew and/or Luke, then to enlist the majority on the date of John is to ignore part of the basis for the majority’s opinion. It simply will not do to cite the majority opinion when many of the supporting arguments are at variance with one’s views. Rather than claiming that there must be sound, critical reasons for holding to a view, one ends up being less than critical in the appeal to the majority.

Fourth, one of the major reasons for holding to a post-70 date for John is the advanced Christology. This is certainly a weighty point, and not one I wish to dispute. John’s Christology is indeed more developed than the other Gospels’ Christology. But a question that must be addressed is whether it is more developed than, say, the Christology of Paul’s letters. I would argue that it is not more developed than Paul’s later writings, and those were written in the 60s. Another question has to do with whether the Synoptics are attempting to present Jesus from the same perspective as John. That is, are they attempting to instruct the church about who Jesus is in all the fullness of their theological convictions that had developed over the early decades of the church’s existence or are they trying to present Jesus in a more inductively historical narrative approach? Despite the fact that all the Synoptic Gospel writers were genuine believers and were thus heavily influenced by Jesus of Nazareth, I do think it is too much to argue that they had no concern for historical accuracy. John, on the other hand, uses a historical foundation from which to build a theological superstructure. His view of Jesus is more from the perspective of heaven itself, for he wishes to paint a picture of Jesus that tells the readers from the beginning what the disciples only discovered some time after the resurrection. Yes, he has historical concerns, but his Gospel shows much more of a theological agenda than the Synoptics do.

Fifth, another major consideration—and one that is hardly discussed in the literature—is the relation of two issues: (1) the literary relationship between John and the Synoptics; and (2) how this issue impacts the date of the Fourth Gospel. Two scholars in particular made out an excellent case that John’s Gospel was literarily independent of the Synoptics: Percival Gardner-Smith, in his 1938 doctoral thesis at Cambridge University, and C. H. Dodd in his brilliant work, Historical Tradition in the Fourth Gospel (1963) Today, this view has been strongly challenged, but it is still most likely the majority opinion. I, too, hold to Johannine independence. I won’t get into the details why, but I think the evidence is fairly strong that John had not seen the writings of the Synoptic writers. If this is the case, then an important principle is brought into focus: the bigger the gap between the Synoptics and John, the more likely it is that John had access to these other Gospels; the shorter the gap, the less likely it is that he had such access. When one considers the fact that there is extensive literary interdependence among the Synoptics, we need to begin to probe some historical issues. How did Luke and Matthew find out about Mark? When did they do so? Caird makes an excellent argument that Luke came across Mark late in the writing of his own Gospel. But Matthew used it from the beginning. Now if Mark was written before 70 (as most scholars would argue) and if Matthew and Luke were written before 70 (as an increasing number of scholars, though still a minority, would argue), then that implies that the dissemination of at least one of the Gospels occurred fairly quickly—within just a few years. But if John’s Gospel was written in the 90s, how is it possible that it shows no substantive evidence of acquaintance with Mark or Matthew or Luke? Even given a remote location for the writing of John, this hardly seems likely. Let me underscore this point in relation just to Mark: Even if Matthew and Luke were written as late as the 80s, how likely is it that a decade later John would still be unaware of any of the Synoptic Gospels? I do not off-hand know what Dr. Köstenberger believes about the dates of the Synoptic Gospels, but the earlier he dates them the more difficulty he would have holding that John is written in the 90s—if, that is, he thinks that John is literarily independent of the Synoptics. On independent grounds (i.e., not related to my dating for John) I have come to the tentative conclusion that Dodd was essentially right about Johannine independence. But I also have felt that the Synoptics were all pre-70 documents. If John is independent and if the Synoptics are all pre-70 (or even if just Mark is), it becomes increasingly difficult to maintain that John was written in the 90s.

The point of this short essay was to show that there are interlocking pieces of evidence that we all weigh when considering matters of historical reconstruction. It is not fair to dismiss someone’s views because of a faulty assumption that such views are based on only one piece of evidence. My point about John 5.2 is that this is a strong piece of evidence that I do not think has been adequately dealt with, but it is by no means my only argument for an early date of John’s Gospel.

All in all, I think there are good reasons for thinking that John’s Gospel was written before the destruction of Jerusalem. I have only presented a few of them here, but I wanted to offer a corrective on the notion that I only had one reason for holding to this view. Again, I am grateful to Dr. Köstenberger for his careful scholarship and for marshaling evidence that would contravene my view of the ἐστιν in John 5.2. I do not, however, think his arguments are substantial enough to overthrow the prima facie force of the present tense in this verse. He may well find some examples that do just this, and if so, I will need to revise the level of conviction that I have about how important this verse is to the date of John. But that is only part of the package on the date of this Gospel.


1 See Brent Nongbri, “The Use and Abuse of P52: Papyrological Pitfalls in the Dating of the Fourth Gospel,” HTR 98.1 (January 2005) 23-48. Nongbri argues that the date of this fragment is much later than 100-150, the date that most paleographers would assign to it.

Related Topics: Textual Criticism