Tuesday, 15 October 2019

The Joy of Teaching the Apostolic Fathers

As someone who labours largely in canonical literature, it is nice to bring my head up now and again and teach other material. Right now in fact I'm teaching one of my dream courses: a graduate seminar on the Apostolic Fathers. We're working our way through the entirety of that corpus, and hopefully will also find time to touch upon Justin Martyr's First Apology and the Gospel of Thomas: works not typically reckoned among the Apostolic Fathers, but close enough temporally as to be of pedagogical relevance. As I teach this course, I am finding one unexpected benefit: the fact that none of my students consider these texts to be part of their biblical canon allows us to ask questions openly which some might find awkward when they are addressed towards canonical work.

Let us consider a concrete example. Last week we covered the Shepherd of Hermas. For those unfamiliar, you should go and read it; if you still haven't read it, then I'll give the relevant Coles' Notes here. The Shepherd centres around a series of visions by a Roman Christian named Hermas, active in perhaps the last quarter of the first century or the first half of the second. Students have noted that many of these visions are reminiscent of those in the Revelation of John. Now, if I asked students to explain what is going on with the visions recorded by John of Patmos, many would no doubt feel uncomfortable considering the possibility that they were the result of physiological or psychological realities, and perhaps even more they would feel uncomfortable considering the possibility that the visionary framework is entirely fictional. Much of that reticence disappears when they come to a non-canonical work, and students will much more openly consider possibilities with regard to the Shepherd of Hermas that they might find at first blasphemous with regard to the Revelation of John. That allows us to more openly consider such possibilities. And what that also means is that what when they do turn to canonical work they are all the better equipped to explore such possibilities there. Work with non-canonical texts often opens up a critical distance that many struggle to find when studying canonical texts. Because, after all, the aim of historical investigation is neither to "prove the bible wrong" nor to "prove it right," but rather to more modestly figure out what was happening a couple millennia ago so as to produce the materials available to us today.

Thursday, 12 September 2019

Of Star Trek and Reality

When I was in high school, I stumbled upon a book called The Star Trek Chronology: A History of the Future. It was 1994; Star Trek: The Next Generation had just finished its seven-year run; Trek was in the air; and as someone greatly interested in history this book intrigued me. It probably would not take heavy psychoanalysis to plot a line from picking up that book in 1994 and beginning in 2007 the work that led to my soon-to-be-completed monograph, Rethinking the Dates of the New Testament.

Over the years, I lost track of my old copy of The Star Trek Chronology. I recently was at a used bookstore in my hometown, located just a couple blocks from the store where I bought the book in the first place, and seeing a copy picked it up. I would like to quote from the introduction. "We choose, in this volume, to treat Star Trek's invented universe as if it were both complete and internally consistent. In effect, we are pretending that the Star Trek saga has unfolded according to a master plan, and there is a logical, consistent timeline in those episodes, even though we (and you) know very well that this is not entirely true." It is a fascinating vanity, and one that (again perhaps not coincidentally) closely resembles my own general approach to history, which proceeds on the understanding that our data issue from a complete and internally consistent reality. The difference of course is that the authors of this book knew full that their "data" did not issue from such a world. There is no logical, consistent timeline, because there were no actual events.

But things are different with history. There is a consistent timeline (one wants to be less sanguine about supposing the existence of a "master plan"). There were actual events; they happened in space and time; and in principle (although not always in practice, due to the limitations of data) we can determine the relationship between these events. And as such "consistency" becomes a significant test, one which forces us to consider the degree to which insights drawn from careful attentive to the data of the past can be brought together into a coherent, consistent understanding of said past. This is in large part what Collingwood means by the historical imagination. Sometimes, as Collingwood notes, such an understanding leads us to recognize that this or that ancient source is mistaken, where intentionally or otherwise. For instance, I have no doubt that Philo encountered a group of Egyptian Jewish sectarians whom he described as the Therapeutae, and I have no doubt that they in many respects closely resembled the Essenes. I have no doubt on this matter because Philo tells me that this is the case, and on the one hand there is no reason to think that he is lying and on the other he is someone who I can reasonably expect to know about Judaism in Egypt. I also have no doubt that Eusebius read Philo's account; he says that he did, and he knows its details quite well. But I also have no doubt that he is entirely wrong when he concludes that these Therapeutae were in fact Christians. No doubt, as late Second Temple Jewish groups the Essenes, the Therapeutae, and the early Christians all bore a "family resemblance," such that descriptions of any one of these would resemble to some extent the other two; and I would argue that Eusebius, in his zeal to find any evidence for early Christians, mistook resemblance for identity. I make this judgment even though I actually cannot prove Eusebius to be wrong; it is possible that the Therapeutae were Christians; there is nothing that excludes the possibility; but I frankly find an error on Eusebius' part to be a more compelling explanation. The data permits a coherent, consistent understanding of the past, and that understanding I would argue entails the judgment that one of my authorities erred. This is possible because the data issues from a complete and internally consistent reality, even though the data themselves are incomplete and inconsistent.

Wednesday, 11 September 2019

The Seduction of Artifice

There is a wonderful scene early in Agatha Christie's Lord Edgeware Dies where Lady Edgeware, upon learning that her husband has been murdered, throws herself into hysterical grief. Now, Hercule Poirot and Inspector Japp both immediately recognize that she's acting. Japp--who is already convinced that Lady Edgeware is the murderer--says "See, clearly she's guilty!" Poirot says, "No, I have interacted with Lady Edgeware before today. She is a professional actress. Her entire way of being in the world is to figure out what role others expect her to play at any given time, and playing it to the point of excess. Her grief is artificial, as I know that she hated her husband, but since everything that she does is artificial this tells us nothing about whether she committed the murder." (I am here paraphrasing: these are not direct quotes). There is, I think, an instructive lesson here for historians.

Very often I hear it said that this or that aspect of the gospels or of Acts is a "literary device" or "literary construct," and as such we can conclude that whatever event it describes never happened. When one makes such an argument one is thinking much more like poor Inspector Japp than like the astute M. Poirot. The nature of texts is such that everything in them is a literary construct; the mere observation of that fact is meaningless for the historian's purposes. What must be shown is that the aims behind this particular literary construction are such as to obviate its usefulness for those who want to reconstruct past events; that is to say, Japp must not merely suppose that Lady Edgeware's artificial grief is intended to hide her guilt but rather must show that this is the case. Now, in many cases one can show this. Returning to Dame Christie, her novels are all literary constructions, and one would be ill-advised in thinking that they describe events that actually happened. There was no Lord Edgeware who was murdered c. 1933; he had no American actress wife who flew into artificial hysterics at the news; a conceited Belgian detective did not solve the case. But we know this because we know that Christie aimed at fiction and not history; her literary constructs were all in service of that aim. By contrast, a true crime story might well use many of the same devices as she does to describe events that are well documented. (Think of such movies for instance as The Black Dahlia and Zodiac, and the point becomes evident). There is in all literature artifice; indeed, on one level all of literature is artifice; but the presence of literary artifice is not in and of itself an index of historical untruth.

Monday, 19 August 2019

Method on the Orient Express

I recently posted on Facebook that I have taken to suggesting to doctoral students that they read Agatha Christie's Murder on the Orient Express, as I consider the way that she depicts Poirot solving the case to exemplify the way in which historians should go about doing the work of history. There was some interest expressed in further explication of this statement, so here it is. (Warning in advance: the antepenultimate and ultimate paragraphs contain spoilers. Best not to read if you have not read Murder on the Orient Express or seen any of its various adaptations).

As many readers will no doubt know, the Lonergan Research Institute houses the Lonergan Archives. A big part of what we do is respond to electronic requests for scans and the like, so when I took over as Executive Director last year I spent some time familiarizing myself with the contents of the archives. I was surprised to discover the number of murder mysteries among Lonergan's books. Most of the murder mysteries found in the archives were published in the sixties and seventies. This is significant, because it is during this period that he fell more fully under the influence of the philosopher of history, R.G. Collingwood (my friend and colleague Jordan Ryan has looked into when Lonergan really began to read Collingwood heavily: IIRC, there's only one reference to Collingwood in Insight in 1957, but by Method in 1972 Lonergan is devoting whole sections to C.'s thought). Why is Collingwood significant in regard to the murder mysteries? "Historian as detective" is a significant theme in Collingwood's work, to the point that he develops his own miniature detective fiction to illustrate certain of his arguments. As far as I can tell, Lonergan's interest in murder mysteries increased in parallel with his reading of Collingwood, and given Collingwood's own interest in detective fiction this doesn't seem to me entirely coincidental. I think it most likely that Lonergan's own interest in murder mysteries received significant impetus from Collingwood's references to and use of detective fiction. (It should also be noted that the mid-60s is also the time that Lonergan was convalescing from his bout with lung cancer, and thus perhaps quite plausibly engaged in more leisure reading than normal; still, even if we grant that as a factor, the fact that he turned specifically to murder mysteries remains of interest). This receives partial confirmation in the work of Ben F. Meyer, Lonergan's devotee, whose own writings on how to do history (which date from the 70s through early 90s) are both heavily influenced by Collingwood and also employ the tropes of detective fiction to illustrate key points.

Now, the above is already an example of how one does history. I start with a known fact: Bernard Lonergan owned and presumably read a good number of murder mysteries that were published in the late sixties and into the seventies. This prompts a question: why? In order to answer this, I look at other relevant data, such as what else he is apparently reading around the same time. I note that there is evidence that he was reading and influenced by R.G. Collingwood's work in the philosophy of history at the time, and also that Collingwood used detective fiction to illustrate key points. I note also that his follower, Ben F. Meyer, picks up on Collingwood's use of detective fiction and employs a similar strategy in his own work on historical method. I conclude that it is likely that Collingwood's work prompted in Lonergan an increased interest in murder mysteries.

But that leaves unaddressed why I specifically think that Murder on the Orient Express provides an exemplar of historical method.

(Here spoilers follow. You have been warned).

In the novel, a man is killed on a train. The great Belgian detective Hercule Poirot (who appeared in Christie's first novel, The Mysterious Affair at Styles, and is the character who appears most frequently throughout her stories) happens to be onboard, and as the train will be stuck in the snow for some time he's called upon to investigate the murder. He interviews his fellow passengers as well as the staff, and eventually determines that they are all lying. Even the dead man was lying, living under an assumed identity. But Poirot does not throw his hands up and say that because they are all lying we cannot know the truth. No, he asks a further question: why are they lying? And from that, he infers that they were all involved in killing the man, who had years earlier committed and gotten away with an awful crime.

If I turn to historical Jesus studies, one of my primary areas of focus, I would make an immediate parallel. Throughout the various quests for the historical Jesus the question has been which if any data in the gospels was accurate. Which claims in the gospels can be affirmed? But the lesson from Poirot is that the more important question is not "Is it true?" but rather "Why is this claim made?" There are of course cases in which a claim is made precisely because it is more or less what happened; for instance, the gospels all report that Jesus died by way of crucifixion, and almost certainly they do so because that is the case. But one might go further, and argue cogently that the gospels claim that Jesus died by way of crucifixion because the early Christians felt that the manner of his death was of theological significance; after all, what interests us is in no small part why the evangelists chose to report Jesus' death in the first place. Or to use an example where the gospels do not as fully agree: in the Synoptics Jesus cleanses the temple at the end of his ministry, in John at the beginning. So the question: why do the Synoptics report this action at the end of Jesus' ministry and John at the end? I note that all the gospels agree that the cleansing occurred at Passover. I also note that the Synoptics only depict Jesus going up to Jerusalem once for Passover. I note that John depicts him going up to Jerusalem for several Passovers. An insight: given that the Synoptic Gospels only narrate one trip to Jerusalem for Passover, they have no choice but to situate a Passover cleansing in that particular trip; conversely, as John narrates multiple trips to Jerusalem at Passover during the ministry, he has the option of situating the cleansing at any of these. I thus advance an hypothesis: the Synoptic decision to situate the cleansing during Jesus' final week is a result of their prior decision to narrate but one trip to Jerusalem. But we still need to explain why John situates it at the beginning? And this is where it gets more interesting: there is no comparably compelling explanation on literary or one might add theological grounds as to why John situated the cleansing at the end. There is however a historical one: John situates the cleansing at the beginning of the ministry because he had reason to think that this is where it should be situated. Combined with the evidence that John is either an eyewitness or has access to eyewitness testimony, I further suggest that he had such reason because he either was present for the event or because he had access to the words of someone who was; and from this I am inclined to think that the cleansing occurred nearer the beginning of Jesus' ministry than the end.

Much like Poirot, I conclude that three of my sources are less than fully accurate. (I wouldn't say they are lying: I would say rather that they are constrained by their own narrative decisions). But the precise content of that inaccuracy is a significant clue to what actually happened. And that's why Mrs. Christie still rewards reading.

Saturday, 23 March 2019

Re-Visioning Ancient Israel

I've been working through parts of Neil Ormerod's Re-Visioning the Church with my graduate seminar on Early Christian Institutions. This has led to at least two developments in my thinking. One, that this book is too rich to only assign parts to students: you need to work through the whole thing. Two, a deepening of my conviction that there is need for a prequel, which uses the same basic Lonerganian approach to look at the history of ancient Israel and Judaism through to the early rabbinic era (ideally, such an account would continue through to modern Judaism, in order to combat the impression that the history of ancient Israel and Judaism leads most naturally into that of Christianity, but I personally lack the competence to produce such an account; the best I can do is take the account some distance into the rabbinic era).

This second conviction came out of a recent class discussion, in which we considered how the early church was not a creation ex nihilo but rather a new development within a history that goes back to the Canaanite Bronze Age. Leaving aside intellectually fraught efforts to argue that ancient Israel is an epic fiction manufactured in the Persian or Greek era--an effort that will forever stumble over the Merneptah stele--we can reasonably envision a basic chain of progression. What became Christianity first appears within the synagogues of Galilee and Judea; it soon spreads to those of the Diaspora, and both there and in the Land it sets up communities structured along the lines of Jewish precedents. But these synagogues and others Jewish precedents have a history. In the Land, they began as the city-gate, which stretches back to Bronze Age antecedents, a space in which the elders of the town assembled to debate and discuss matters of great concern to themselves (Ruth 4 offers an exemplary account of the sort of matters likely considered in that space, and one suspects that Genesis 23 does as well); and although the synagogues of the diaspora were largely modeled after the institutional forms of the Greco-Roman associations they nonetheless took on many of the same roles within the overseas Jewish communities as did the synagogues of home. It is altogether possible to write a history of both the synagogue and the church that begins with the Bronze Age Canaanite city-state, asking how various stages of development represented transformations necessary to better meet the needs of developing communities.

Of course, the deleterious effects of group bias would also need to be considered: sometimes--and we can argue about the relative frequency--transformations occurred not to better meet the needs of developing communities, but rather to allow particular groups (typically, the males of affluent lineages) to dominate others. Indeed, we would have to consider the degree to which the city-gate was rooted in such bias from the start; certainly, one could very easily build a narrative in which it originated precisely to further and sustain affluent male dominance. In any case however, ancient Israel probably in part emerged as a revolt against such domination, and there is I think evidence to suggest that it took measures to institute and sustain a degree (but only a degree) of egalitarianism. But as inevitably happens, revolution against inequality eventually gave way to new inequalities: certainly in the form of the monarchies, but probably yet earlier. This was probably the case especially for women, as revolutions have a remarkable capacity to insufficiently apprehend that women might desire the same freedom from oppression as do men. Consider that probably the most brutal description of violence against a woman is found within Judges 19, which leads directly to the Benjamite civil war in Judges 20 and in Judges 21 the kidnapping and forced marriage of four hundred young women who had just seen their families killed by the rest of Israel. Whether these accounts describe actual courses of events (and I am more sanguine than many about that possibility), they probably should be taken as an indication that the men of the Israelite settlements valued women less as persons and more as resources. The Marxist and feminist traditions excel in considering such sad realities, for probably no intellectual traditions have more fully explored how group bias--whether it be towards the affluent or towards the male--destroys the integrity of communities and the well-being of individual lives. (One of my students' criticisms of Ormerod was that he did not take much account of power; I am probably to blame for giving them that impression, as I did not have them read his discussions of bias. Again, that speaks to how Re-Visioning is not a book into which one can just dip).

Now, some students suggested that Jesus' life and ministry and salvific significance were such that there was some sort of rupture in the historical continuum. From a historical perspective, I would fully grant that Jesus represents one of those creative personalities who periodically emerges to address the urgent needs of their times; the emergence of such creative personalities might indeed be taken as signs of divine grace operating in history. And indeed, the needs of Jesus' time were urgent. The Jewish revolt was still a few decades away, but the Land was already suffering from many of the conditions that led to that event, most notably I would argue the economic depredations suffered by the local population under Roman rule. Jewish society in the Land was breaking down under these depredations, as lifeways were disturbed by predatory foreign rule. Leaders emerged, promising deliverance from these conditions. The real question for me as a historian is why from this particular situation this one particular leader, Jesus of Nazareth, began a movement that eventually took over the world. Theologically, one might articulate this in terms of God's gracious solution to the problem of human evil, but I would argue that historically this question can only be answered by the fullest understanding of the problems to which Jesus was responding and the conditions under which he operated, and that moreover the theological account would only be deepened by fully developing the historical.

Tuesday, 19 March 2019

How Old is the Hebrew Bible?

I've been reading through Hendel and Joosten's How Old is the Hebrew Bible?, and thought I'd post some thoughts on the book. Now, a caveat as I do so: this book is largely concerned with arguing that historical linguistics offers an important set of data when it comes to dating the Hebrew bible, and it should be noted that I am very far from being a historical linguist. But I do know a thing or two about how one goes about thinking about the date of biblical texts, and it's on that basis that I write here.

Hendel and Joosten's basic hypothesis is that if a text is predominantly written in Classical Biblical Hebrew (CBH), then it likely was composed when Classical Biblical Hebrew was predominant; if a text is predominantly written in Transitional Biblical Hebrew, then it likely was composed when Transitional Biblical Hebrew (TBH) was predominant; and if a text is predominantly written in Late Biblical Hebrew, then it likely was composed when Late Biblical Hebrew (LBH) was predominant. It is difficult to argue with the basic supposition underlying this hypothesis, namely that all things being equal (a term that they themselves use, quite rightly), there is good reason to think that a text likely dates from the period in which its language seems most fully to have flourished. In terms that I use, this is a form of contextualization, and one that is almost unusable in New Testament studies due to the much shorter span in which that corpus was written. In Hebrew Bible however, it can potentially be used to good effect.

By comparing the biblical literature with extant extra-biblical inscriptions, Hendel and Joosten suggest that CBH flourished in the eighth through sixth centuries BCE, TBH in the sixth, and LBH from the fifth onward (note the overlap between CBH and TBH; this is probably inevitable, given that TBH is transitional). They also rightly note that as we do not know how long before the earliest extant inscriptions CBH emerged, we cannot exclude the possibility that texts written in this "chronolect" predate the eighth century. Hendel and Joosten argue that Genesis through 2 Kings is written predominantly in CBH, Ezekiel, Lamentations, Jeremiah, Deutero-Isaiah, Job, Jonah, Haggai, and Zechariah 1-8 in TBH, Chronicles, Ezra-Nehemiah, Esther, Ecclesiastes, and Daniel in LBH. This leads to the other plank of their argument, regarding what they call "consilience." Consilience in their usage is virtually synonymous with what I call "convergence," a term that I prefer only because its meaning is more intuitively obvious to most readers; that terminological difference however does not even rise to the level of a quibble. By these terms we all mean simply that the preferred date for a given text is that upon which the greatest amount of data converges. So, without going into the details of Hendel and Joosten's argument here, one might note that those texts which they conclude are written in CBH are concerned largely with life in pre-monarchic and monarchic Israel (i.e. the late sixth century and earlier); those which they conclude are written in TBH are concerned largely with life in the last days of the Judean monarchy through the exile (i.e. the sixth century); and those which they conclude are written in LBH are concerned largely with life in the Persian and Greek periods (i.e. the fifth through second centuries). In other words, there is significant convergence between the temporal "home" of the language utilized by these texts and the periods in which they are interested.

If the arguments from historical linguistics are granted, then this is a strong cumulative case for a broad periodization of the Hebrew Bible, although certainly other forms of work would be necessary to find more precise dates for any given biblical text (Hendel and Joosten are of course quite aware that what they do in this book is just one part of a larger strategy towards establishing such dates; they simply want to make clear that it is an indispensable part). Having said this, I do want to register one caveat and one criticism. The caveat returns to the beginning of this post, and reminds the reader that I am not a historical linguist and thus my capacity to evaluate their empirical argumentation is limited; my conditional statement about the affirmation of their arguments from historical linguistics is thus rooted in the limitations of my own expertise, rather than any difficulties that I detect in their work. The one criticism that I would register has to do with user-friendliness. Even a simple chart summarizing their findings would make this book far more readable. Instead, one has to flip from chapter to chapter to find summary statements embedded in (typically, but not invariably) introductions and conclusions. That criticism however is secondary to the strength of the book, which is that even if one or more of their particular empirical arguments should fall it remains a model of how to go about thinking about the dates of biblical literature.

Sunday, 10 February 2019

1 Clement v. Tradition

Biblical scholars have often like to cast ourselves as critics of received tradition. Given that self-conception, it is fascinating to note just how frequently we tend to unreflectively repeat the received traditions of our scholarly heritage. A sterling example of such repetition is the standard date given for 1 Clement (not a biblical text per se, but certainly biblical adjacent), namely c. 95. This date was first proposed in 1633, by Patrick Young, Royal Librarian to Charles I, who was the first western scholar in the post-medieval world to have access to a(n incomplete) copy of 1 Clement, and was given the not-insignificant imprimateur of J.B. Lightfoot in the latter's magisterial study of the Apostolic Fathers. In 1633, we were two centuries away from the advent of modern historical consciousness, and almost two-and-a-half away from having a complete text of 1 Clement. Neither of these realities shows the date to be wrong, of course: persons operating before the nineteenth-century's "Copernican Revolution" in historical thinking were altogether intelligent and quite capable of reaching conclusions that we now on a more solid basis know to be correct, and the portions of 1 Clement most significant for purposes of establishing its date were already available to Young. What is more significant is that the two primary bases for the standard c. 95 date are now known to be almost certainly false.

The first of these bases begins from 1 Clement 1.1, which speaks about "sudden and repeated misfortunes" recently suffered by the Roman church. Young confidently concluded that this is a reference to the Domitianic persecution, and thus argued that the letter must date to the very end of Domitian's reign (81-96). There are two problems here, which E.T. Merrill brought convincingly to our attention in 1924. First, the "Domitianic persecution" seems to have been limited to the execution of the emperor's cousin, Flavius Clemens, and the exile of Clemens' wife (also the emperor's niece), Flavia Domitilla (III: both her mother and grandmother appear to have borne the same name), who may or may not have been Christians. Second, 1.1 probably does not refer to a persecution at all. More likely, it is just a typical rhetorical move common to the sort of letter that 1 Clement represents. In other words, 1.1 probably doesn't reference a persecution that didn't happen. 1 Clement 1.1 probably contains no data usable for chronological purposes.

The second basis for the c. 95 date is the dual supposition that Clement must have written this letter while he was monarchical bishop of Rome, and that on the basis of ancient succession lists his episcopacy can be dated to the 90s. There are again (at least) two problems here. First, as there probably wasn't a monarchical bishop in Rome until the second century, Clement probably never held such a position. Certainly, 1 Clement does not seem to envision a monarchical bishop in Rome. Now, I don't necessarily think that the succession lists are without historical value, and think it entirely plausible if not probable that figures such as Linus, Anacletus, and Clement (the second, third, and fourth "popes" respectively) held some sort of privileged position that was in retrospect recognized as standing in succession with the later monarchical bishops. But that brings us to the second problem with this basis: it is not clear that Clement was associated with the letter because of his episcopal office. Here Shepherd of Hermas Vis. 2.4.3 (8.3) is potentially relevant, as it says that a certain man named Clement was responsible for sending letters abroad from the Roman church. This coheres remarkably well with the evidence furnished by Dionysius of Corinth, who c. 170 wrote that the Corinthian church still regularly read the letter that the Roman church sent to them through Clement. We might rather suspect that Clement became associated with the letter not because he was bishop when it was written, but rather because he was a sort of chief scribe for the Roman church. In such a case, the possibility that the letter predates Clement's "episcopacy" is very real. And this is before we even address the matter of authenticity, for if the association with Clement is spurious then there would be no reason to connect it with the particulars of his biography.

Now, of course, the c. 95 date could be correct, but if so it would be correct for reasons other than those which were most instrumental in establishing that date in the first place. Young and Lightfoot would have been correct by accident. Until alternative and compelling arguments for this date are put forth, it would be best if we caught up with 1924 and stop treating it as a given.

Thursday, 7 February 2019

Context and Transmission

In On the Reliability of the Old Testament, Kenneth Kitchen argues that many of the details found in Genesis are most fully at home in the Middle Bronze Age Levant and broader Near East. Here of course he is building upon an older stream of scholarship, one which our scholarly memories connect most fully with William Albright. Regardless of its pedigree or veracity, this sort of argumentation is largely what I would group broadly under the category of contextualization. In order to tease out how contextualization works in practice, especially in relation to a text such as Genesis where the dates that scholars have suggested for its composition range over a matter of centuries, let us suppose that Kitchen's argument is correct in at least the broad outlines. Let us also stress that this supposition is entirely for purposes of argumentation: no judgment is rendered here regarding the veracity of Kitchen’s argument. We are merely asking what follows if Kitchen is correct in arguing that the setting-in-life of much of the material in Genesis is to be found within the Middle Bronze.

If that argument is correct, we would then have to ask how Genesis came to include so much material that is most fully at home in the Middle Bronze Age. Under such conditions, the likelihood that this material originated much later—into the Iron II or even as late as the Persian or Hellenistic periods—would be exceedingly slim, demanding affirmation of such a high number of coincidences as to beggar the intellect. Far more likely would be that the bulk of this material originated in the Middle Bronze and was transmitted to the later times in which they are more fully documented. The extent to which this material was transmitted in a form that resembled our Genesis would remain a matter of investigation (and given the state of the evidence, probably also conjecture), but any adequate hypothesis would require transmission processes that yield a reasonably high rate of preservation. The state of the data might well be such that we cannot now reconstruct those processes with any degree of precision, and we might not be able to do much more than state that they almost were operative, but the alternative "coincidence theory" would be so significantly improbable as to be functionally excluded. Arguments that would deny a priori that such transmission is possible would have to yield to evidence which makes such transmission necessary a posteriori.

Now, again, let me be clear: I am not here affirming Kitchen's arguments. I am simply teasing out the historiographical implications that would follow if he is indeed correct. And let me be equally clear that if one were to affirm that Genesis contains material from the Middle Bronze Age one would not necessarily need to affirm that its stories describe actual events from that time. Although the presence of material from the Middle Bronze would likely increase the probability that Abraham et. al. were historical figures whose lives to some degree resembled themselves described in Genesis (at the very least, the absence of such material would tend to militate against historicity), it could also be the case that such material is entirely in the service of fictional accounts. Once again however, this would have to be figured out on the basis of a posteriori investigations, not a priori suppositions.

Wednesday, 6 February 2019

On Dating the Old Testament

(A small apologia for my use of "Old Testament" rather than "Hebrew Bible." Working in a Catholic context, I am interested in not just the books shared in common by Judaism and Christian, but also in the so-called "Deuterocanon" or "Apocrypha." As really the only historic category that encompasses these 46 books is "Old Testament," I use this term, with the clear statement that supersessionism is neither implicit nor intellectually and morally acceptable).

I have been giving some more thought to the method of Old Testament chronology, specifically how to go about establishing when specific books were written. The starting point is what I have learned from working on NewTestament chronology (cf. previous posts, and eventually my forthcoming monograph with Baker, Rethinking the Dates of the New Testament), and to that end I would remind my devoted reader of the three basic categories of argumentation that I have recognized in the work of establishing the date of specific NT books: synchronization, authorial biography, and contextualization. Synchronization is the classic work of establishing the text’s temporal relationship to independently datable events or situations, including the composition of other texts. For instance, when Isaiah 1:1 tells us that Isaiah had a vision in the days of kings Uzziah, Jotham, Ahaz, and Hezekiah of Judah, we can be reasonably certain that the text dates no later than the reigns of these men. At least two factors tend to vitiate but do not necessarily obviate the use of synchronization in the work of Old Testament chronology: one, the relative dearth of independently datable events or situations in the Old Testament corpus; two, the fact that certain texts appear to have been composed over very lengthy periods of time. Both problems exist within the work of New Testament chronology, but are more acute in Old Testament (although the latter problem is perhaps overstated at times, and is often driven more by lingering Victorian commitments than by careful attention to the evidence). Indeed, without getting into the nitty gritty of the evidence, my guess is that there are several prominent books of the Old Testament for which arguments from synchronization will be almost entirely fruitless.

Authorial biography begins from what we know about the author independent of the text to ask where in her or his life the text is best situated. Nowhere in the biblical corpus is this of greater utility than the work of Pauline chronology, where it yields the most precise compositional dates of probably either testament (for instance, one must argue in spite of the best evidence if one wants to date Romans at any time other than late 56 or early 57). As with synchronization, authorial biography is of greatly limited utility to the work of establishing the date of Old Testament books, in large part because in very few cases do we know much about the authors independent of the texts in question. Indeed, off the top of my head, I cannot think of any Old Testament book for which arguments from authorial biography could be particularly fruitful. It is entirely conceivable to me that this entire category of argumentation would be absent from efforts to establish the dates of the Old Testament texts.

Within New Testament chronology, contextualization consists of establishing the text’s temporal relationship to the general course of early Christian development. James Crossley’s treatment of the date of Mark’s Gospel is a classic example here. Noting that Mark’s Gospel takes it for granted that Jesus was Torah observant whereas Matthew’s and Luke’s must address questions about whether he was, Crossley argues that Mark’s Gospel is most appropriately situated before the Gentile mission had forced early Christians to seriously confront the matter of Torah observance. In Old Testament studies, contextualization would comparably seek to establish the text's temporal relationship to the general course of developments in Israel and the broader Near Eastern milieu. A recent example might be Sandra Lynn Richter’s argument that the economics supposed by the Urdeuteronium (which she defines roughly as consisting of Deuteronomy 4.44–27.26) are most reflective of those which straddle the Iron I/Iron II transition (i.e. in more classical terms, the period of the Judges through the early monarchic period); if Prof. Richter is correct on this matter, then all other things being equal such a range becomes the best candidate for dating this material, as any other alternative will almost certainly be significantly less parsimonious. My expectation is that arguments from contextualization would be the most fruitful in the work of establishing the dates of particular Old Testament writings.

Thursday, 3 January 2019

Revision and Dispute

All historical argumentation is probabilistic. This is also to say that any and all historical hypotheses are subject to revision or dispute. Hypotheses subject to revision are hypotheses whose probability sufficiently approaches 1.0 that we can treat them as virtually certain. Such hypotheses include the hypothesis that Germany invaded Poland in September of 1939, or that Jesus of Nazareth existed. Such hypotheses are virtually certain not necessarily because there are no conceivable alternatives, but in many (perhaps most) cases because all conceivable alternatives are sufficiently improbable that they can be excluded. Can I conceive of a world in which all the documentary and eyewitness evidence for Germany’s invasion of Poland in 1939 is falsified and it never took place? Perhaps. Is that alternative probable? Hardly. Nonetheless, in principle, even the most probable statement is subject to revision upon the emergence either of new evidence or new insights into old evidence. The recent resurgence in arguments for Jesus’ historical non-existence rested entirely upon the argument that there had emerged new insights into old evidence. The reason that these arguments fail is because those competent in the matter and fully familiar with the evidence recognized immediately that these were not new insights at all but almost without exception insights that had been advanced and rejected the better part of a century ago. For instance, much of the argumentation rested upon a literally Victorian-era understanding of the nature of ancient myth-making, which has long since and properly been abandoned as empirically unsustainable. There is a reason that one can count on two fingers the number of credentialed New Testament scholars who subscribe to the hypothesis that Jesus never existed: quite simply, competent familiarity with the data precludes affirmation of the hypothesis.

Hypotheses subject to dispute are different. These are hypotheses with which a competent person fully familiar with the evidence can reasonably disagree. For instance, it is a virtual certainty that Jesus died on a cross sometime around Passover, between the years 29 and 35. I would argue that among those years, 30 (the long-time majority opinion) remains the most likely. I can present a number of arguments in favour of that date. However, I recognize that a competent person fully familiar with the evidence could reasonably argue for any other date within that range. As with the case of virtually certain hypotheses subject to revision, new evidence or fresh insights into old evidence can alter the probabilities. For instance, the discovery of the Delphi inscription a century ago has significantly narrowed down the date for Paul’s meeting with Gallio as attested in Acts 18, such that what was once thought to have occurred sometime between the late-40s through the early-50s can now more precisely be said to have occurred sometime between July 1st of 51 and June 31st of 52. This range can now in fact be treated as a virtual certainty. What remains subject to dispute is when within this range the meeting took place. For instance, it can be and has been argued that the meeting is more likely to have taken place very early in that range, on the grounds that Paul’s opponents very conceivably seized upon the change in governor in order to gain a hearing for their charges; this is quite plausible, and yet can hardly be treated as a virtual certainty and is such that a competent person fully familiar with the evidence could reasonably disagree.

Both the year of the crucifixion and the timing of Paul’s appearance between Gallio demonstrate that the heuristic distinction between hypotheses subject to revision and hypotheses subject to dispute allows for a dynamic understanding of probability. In my own current work—i.e. my long-percolating study of and obsessive concern with the dates at which the New Testament texts—hypotheses subject to revision tend to yield relatively large ranges. It is beyond reasonable dispute that the Gospel of Mark was written sometime after Jesus’ death and before the first attestation to the gospel’s existence. This allows for a date anytime from c. 30 to c. 120, the latest likely time at which Papias writes about Mark’s compositional work. Dates outside this range are almost certainly non-starters, although the higher end is more likely subject to revision than the lower as the probability that Mark’s Gospel was written before Jesus’ death approaches zero (although in principle one could not rule out the possibility that parts of the text were written during his lifetime). Early 21st-century scholarship has tended to favour a date about midway through this range, with c. 65 to 75 probably representing something like a majority opinion, but the salient point here is that all debates regarding where in the range c. 30 to 120 to situate Mark’s Gospel have moved from matters of revision to matters of dispute, because we are now arguing about matters upon which competent persons fully familiar with the evidence can reasonably disagree. Awareness of this heuristic distinction between hypotheses subject to revision and hypotheses subject to dispute allow us to avoid wasting time and energy either disputing that which is beyond reasonable dispute or seeking virtual certainty when the data does not allow us to do so.