ABSTRACT
When Arnold Toynbee published A Study of History (1956), the reviews were both many and varied. Scholars of all sorts of subjects contributed the reviews, from Indology and Sinology to American history. Most of the reviews ran more or less like this: this work is a real contribution. In my field, it is not very good, but on every other front it is brilliant. Everyone enjoys truly multi-disciplinary conferences, for those are the venues in which scholars in other disciplines reveal their disagreements, and the logic actuating their positions. Otherwise, it is our tendency to defer to the authority, rather than to understand the logic and evidence, of scholars in other fields. The ‘Bible and radiocarbon dating project’ has exposed us all to the limitations of our various fields-radiocarbon, archaeology, dendrochronology, history. Indeed, the next conference we convene should focus specifically not on the positive contributions we can make, but on the limits of our knowledge, with regard to some particular subject. A recurring theme among the participants in this volume has been the Drang nach more sophisticated analysis. Radiocarbon experts are already calibrating their equipment against one another (Fourth International Radiocarbon Intercomparison). Archaeologists are admitting that ceramic periods are not all precisely contemporaneous-a major concession in a community that has regularly equated most pottery horizons with a layer of destruction. Indeed, the Iron I, as
David Ilan remarked, starts in the 13th century BCE. The Late Bronze II, on the other hand, ends in the later 12th century BCE. The two overlap by a century or so. The same is true of every ceramic horizon: each is a rolling horizon, passing from one site to the next, and passing from one site before it passes from the next. And yet the power of the words, ‘Iron IA here, Iron IA there’, is such as to induce scholars to think of a fixed and limited period of time. It begins ca. 1200 BCE, and ends ca. 1100 or 1050. This is bankrupt as a principle of chronology. As Israel Finkelstein intuits (Chapter 3, this volume), pottery types can slide in time one against the other, without violating common sense. The same holds for other sorts of argument to simultaneity, as opposed to rolling contemporaneity. And simultaneity is the biggest issue in our current scholarship about chronology, because archaeological chronology is inevitably only rough. Architectural forms, types of inscriptions, cuisines, trends in artistic motifs on pottery or on glyptic-all these things move wave-like across space, meaning that they reach different locales, and even neighboring locales, in different degrees at different times. Local innovation or conservatism, foreign contact, wealth, ideological freighting —all these things affect the degree and speed of appropriation, differentially by place. We cannot assume that forms such as the bit hilani suddenly were adopted all over the Near East, and in fact there is textual evidence that dates the form’s introduction to Assyria to the 8th century BCE, under Tiglath-Pileser III (Clay Tablet r 18: Tadmor 1994: 172), whereas its appearance in central and northern Syria, at least, clearly dates at latest to the mid-9th century, and perhaps earlier, with antecedents in the Late Bronze Age. But even were we to think we have a clear date for the beginning of a phenomenon, that is no argument against its having begun earlier elsewhere. Innovations spread, or they can, like calculus, be invented independently based on earlier forerunners. Deciding that Newton, that universal genius, must have been the first to develop the calculus, or Darwin the first to develop the theory of natural selection, does considerable violence to the history of science, and slights Leibniz and Watson into the bargain. More supple approaches are needed. One, adumbrated by Amihai Mazar et al. (Chapter 13, this volume) in connection with the evidence from Tel Rehov, is to separate the pottery from the architecture (one needs to do the same with chronological data in Egypt and Israel-separating synchronisms from regnal lengths, for example). Architecture comes from the life of a layer, but ceramic partly from heirlooms and mainly from its final phase. We also need to distinguish different sorts of time-not the famous divisions of the Annales school so much as archaeological time, radiocarbon time, and historical time. We need to distinguish different kinds of history, as well: narrative and its language, the intention behind the narrative, the reconstruction behind the intention, and, something very different, reality. For history is never the truth, and is at its best a version of a truth-manque, about the past. Sue Sherratt said (Chapter 9, this volume) that getting dates, and we may add strata or pots, right, just tempts you to create pseudo-history, like the Dorian invasion of Greece, like the reification of dates of assemblages. We need to sort our variables. We must not be seduced by the words we use into equating a ceramic assemblage with a delimited time, or a carbon date with the life of a layer. We must be especially careful in combining data about various sites into one large contemporaneous database, given that dates come from various points in the life of the layers. And, in relating to text, one must be careful in the extreme, because text, even more than archaeology (so far as archaeologists will concede this at all), requires qualitative evaluation. What makes such evaluation difficult is that texts cannot be subjected to meaningful quantitative evaluation. Aspects of language can be evaluated quantitatively. So we can subject elements of texts to tests that are not dissimilar from those to which we subject archaeological data. Yet, the standards of the field are such that what emerges in journals can be of very low quality-the publication of hunches in many cases, or even of delusions. The field has no watchdogs, no
common standards, which is what distinguishes it from the natural sciences. But when, in Toynbeeesque fashion, scholars of other disciplines lean on Biblical or Near Eastern historians, they run the risk of choosing incompetent guides. This happens frequently in archaeological evaluations in particular, just as textual scholars have often chosen to rely on shaky archaeological interpretations, or, indeed, on no archaeology at all. The archaeologists of a generation ago found remains to identify with Biblical records. Scholars today attack what they think the Biblical material suggests, such as a Davidic empire stretching from the Brook of Egypt to the Euphrates. Late sources say such things, but 2 Samuel restricts David’s Israelite realm to the region between Dan and Beersheba, over and over again. Attacking the late assertions or the implications of our early texts, the deliberate insinuations, misses the point and misprises the nature of writing. If David had ruled to the Euphrates, the early text would certainly say so. Its reticence (2 Samuel 8.3: he ruled to the ‘river’) bespeaks a concern that such a claim would be risible, falsified by knowledge or by memory. No king, and no dynasty, can afford to invite extended ridicule. At the same time, the text invites us to make exactly that mistake, hopes that we will fall into the trap of believing the insinuation rather than the literal remark. In a word, royal propaganda involves spin. Heavy spin. Some textual analysis is reliable. Most is not. The same problem exists, albeit not in the same measure, in archaeology. We just do not have the same degree of collaboration, cross-checking, and non-competitive openness that we see in the radiocarbon community. Richard Feynmann once wrote that in the early days of physics, relations among scientists were ‘argumentative’. Probably, he was referring to the days before the spread of international scholarly journals and of international collaboration. ‘But in physics today’, he continues,
Historians and archaeologists need to establish a like degree of community, and an admission of uncertainty, and a real test against a complete history of relevant observations, and common standards of evidence. The latter was, of course, Robert Merton’s sine qua non for the development of science (Merton 1979). II
Bertrand Russell began his autobiography by explaining how he learned geometry at a precocious age from his brother. ‘The first thing you need to know’, said his brother, ‘is that parallel lines never meet’. ‘Why not?’, asked Bertrand. ‘That’s just the way things are’, said his brother, ‘it’s a definition’. ‘Prove it’, said Bertrand. And on and on. Finally, his brother put his foot down: ‘either you accept that parallel lines never meet, or we cannot go on’ (Russell 1967). Anyone who has dealt seriously with chronological or complex scholarly issues will recognize the import of this anecdote. One needs fixed points, axioms, in order to work out a systematic analysis. Where are our fixed points to come from? Because without a link to the texts, we have no history, no framework for embedding characters or developments into time, in a way that a
historian would find recognizable. We might date layers, in the future. But in historical archaeology, we also need to put the right people into them. The tendency in the past decades has been to lump historical in with prehistoric archaeology, a mistake whose dimensions deserve exploration —in no other field of scholarship do mainstream practitioners exclude, rather than evaluate, evidence. Among historical archaeologists, equipollent proportions deny the value of, which is to say, claim to ignore, textual data. The reasons for that error indict multiple constituencies, including formally inadequate philology in Biblical Studies. How to figure our way out of this deliberate ignoring of textual data, or, on the textual side, of archaeological data, is the real task of the birthing of this century. Maybe we could start with a few anecdotes, such as the time Yigael Yadin told a group of seminar-mates that an inconvenient text just needed to be emended, or the many times when Albright, according to G. Ernest Wright, swept inconvenient sherds from the sorting table. It might be good to know the history of the divide in cultures-text and artefact. At the least, William Dever should have memories of about a thousand such incidents, and we should be recording the history of every one of them. And, similarly, the absorption into the world of text of a goodly number of forgeries, most of them not exposed to date, speaks to the nature of the community of epigraphers and other sub-specialty experts in areas related to the reconstruction of the past. In the last two decades, it has become a fashion to compare Biblical King David to the mythical King Arthur. Neither ever lived, say some vocal Europeans. On the face of it, this position has some merit: not a single building, except perhaps a revetment south of the temple mount, is associated with this 10th-century ruler whose empire, some texts claim, stretched from the Nile to the Euphrates. Scholars have long pointed to similar gates built simultaneously at Megiddo, Hazor, and Gezer to attest building by David’s son, Solomon: the archaeology seemed to confirm a report in 1 Kings 9.16 attributing fortifications at these sites, among others, to Solomon. (We might also add a gate to this mix in the Negev, should we choose to identify Ein Hazeva or Tel Ira with Judahite Tamar.) But others have disputed the attribution of these gates. If their converse position holds, there is no remnant of David’s monarchy at all-either the kings were kings in a nutshell, or they were no kings at all. These are different positions. Either David and Solomon are mythical, or they were minor, local rulers. The first position is that of a group often dubbed ‘Minimalists’, the second that of adherents to ‘the low chronology’. As the controversy over the low chronology has shown, correlating texts to objects is ticklish business. It has never been sufficiently sophisticated, especially in Israelite history. One doesn’t just find a building in the ground and decide that it is the ‘stables of Solomon’, as P.L.O. Guy did at Megiddo in the 1920s, in identifying structures that were neither Solomonic nor, in my view, stables. You must first date the stratum-in historical archaeology, almost always a controversial issue-and then date and evaluate the reliability of the textual source, ever a matter of controversy. Then connections remain a matter for critical scrutiny. The temptation to easy correlation is the temptation of laziness, and of confirming one’s beliefs. Even so, a minimalizing approach was historically misconceived: like a spooked squid, it obscured the past with ink. It led us down the blind alley of arguing about David’s existence rather than about his activity. My brief is to evaluate neither the low chronology nor minimalism. It is to evaluate the likelihood that David reigned in Jerusalem and in its neighborhood. Was he Butch Cassidy and the Hole in the Wall Gang? If he wasn’t the maximum that the text can be taken to claim, say the naysayers, he never reigned. But David and Solomon existed. They even reigned over a unified Israel.
