Centuries of Darkness

A challenge to the conventional chronology of Old World archaeology.

Peter James in collaboration with I. J. Thorpe, Nikos Kokkinos, Robert Morkot and John Frankish.
Forward by Colin Renfrew

We ... began an in-depth investigation of the archaeological chronology of the entire ancient Mediterranean and Near East. Everything we found confirmed our suspicion that the original spanner in the works was the Egyptian time-scale, and that the 'centuries of darkness' inserted into the histories of so many areas between 1200 and 700 BC were largely illusory.

Initial questions and conclusions were then circulated in the form of a discussion paper, published in "Studies in Ancient Chronology" volume 1. The responses we received from scholars in fields ranging from Egyptology to astronomy were immensely encouraging and the expansion of the project towards an eventual book followed naturally.

We were now also confident that we had fingered a genuine solution to the widespread problems. In the meantime a steady stream of new papers was spontaneously appearing in the archaeological literature, in which the framework of ancient Mediterranean chronology was beginning to be laid bare. The feeling is now in the air that it is time to return to basics and re-examine fundamental assumptions. To mention only two examples: in 1987 a special international conference was held at Gothenburg in Sweden under the title of "High, Middle or Low?" with the aim of resolving the long-standing uncertainties in the Middle Bronze Age chronology of the Near East and Aegean; the second concerns the latest issue of the "Bulletin of the American Schools of Orienal Research" (Spring 1990) which was entirley devoted to a debate on a major question of Biblical archaeology - which levels of the ancient cities of Palestine belong to the time of King Solomon, Israel's most famous monarch? Neither of these prestigious ventures came to a definite conclusion.

But how can there still be such a degree of uncertainty? After all, scientific methods of dating, such as the radiocarbon technique, which should have resolved the problems have now been avalable for a generation. Desite this, take-up of the new methods has been surprisingly slow; all too often a dozen or so radiocarbon dates are included in an archaeological site report merely as scientific window dressing. This attitude is clearly reflected in a regrettably common practice: when a radiocarbon date agrees with the expectations of the excavator it appears in the main text of the site report; if it is slightly discrepant it is relegated to a footnote; if it seriously conflicts it is left out altogether.

Lack of understanding of the method by many archaeologists has led to the submission of large numbers of samples of little or no value in dating the contexts from which they come. There have also been problems caused by inconsistent treatment of samples by different laboratories. As the senior radiocarbon scientist Professor Ingrid Olsson frankly concluded at the Gothenburg conference: "Honestly, I would say that I feel that most of the dates from the actual Bronze Age are dubious. The manner in which they have been made ... forces me to be critical."

Where there have been enough good-quality radiocarbon dates available, for exmaple in tracing the spread of agriculture across Europe, the technique has been of immense value. In the Near East and Aegean, however, the lack of systematic sampling means that radiocarbon is still too blunt a tool to resolve the perennial controversies of Bronze to Iron Age chronology. (Relevant radiocarbon dates are generally discussed here in the notes to individual areas.) It needs to be stressed that the youngest dates from a given context or cultural phase are really the most significant. Old, residual material can always be present to supply misleading dates for a context; the yonger dates will more accurately reflect the time when the deposit was formed and when most of its assemblage was made. Simply averaging the results for a phase or context, as it often done, will obviously produce a false impression of anitquity. On the other hand, we are able to note for many areas an increasing number of radiocarbon dates which, though currently treated as "anomalous", are consistent with our theory; but they fail to be decisive because of the general problems affecting the method and its application. Sadly, for the later part of the period under review in this book, radiocarbon may never be able to provide meaningful answers (see Appendix I).

New scientific work in progress holds out ineresting prospects for absolute chronology. Recently, attempts have been made to date the volcanic explosion which devastated the Minoan colony on the Aegean island of Thera (towards the beginning of the Late Bronze Age) by tracing climatic effects in the tree-ring records from California and northern Europe and peaks of activity in the ice cores from Greenland. The difficulty with this is that it is impossible to be sure whether such effects always originate from volcanic eruptions, and if so, which volcano was responsible. As volcanologist David Pyle (1989,90) wrote concerning the Thera eruption:

Direction radiocarbon dating has so far yielded a large scatter of dates that can, at present, be interpreted according to one's prejudice. Indirect methods, (acidity peaks, tree rings) are beguiling, being potentially more precise, but at the same time highly ambiguous and should only be treated with the utmost caution.

The outcome for the absolute dating of Minoan civilisation thus remains uncertain. More definite results may come from the on-going development of a tree-ring sequence for ancient Anatolia (Turkey) and Greece. When complete, the Anatolian dendro-chronology will provide a more precise calibration for Near Eastern radiocabon dates. Further, if it can be firmly linked with local Bronze Age archaeology, we will also have an invaluable control on historical chronology, including that of Eygpt itself, because of the close connections which existed between the Hittite kings and the pharaohs.

In the meantime, radiocarbon dating is still of little help in providing answers to the conundrum of Dark Age chronology. In practice, we have to fall back on traditional methods, primarily pottery dating. Being virtually indestructible, pottery is found in vast quantities on ancient sites and constitutes the bread-and-butter of archaeologists. Basic typological sequences for the development of ceramic styles are well established (though the pigeonholing into minute phases by some experts can be excessive). Pottery thus enables the strata of a given site to be easily dated within a local sequence. Discoveries of imported pottery allow links to be made between the chronolgies of different cultures, while finds of key styles of pottery in those areas with written records allow the whole framework to be attached to historical dates.

Ancient history has often been compared to a mosaic, a patchwork built up from tiny scraps of evidence. A jigsaw puzzle is a much better metaphor, especially when dealing with chronology. For well-known periods (such as the time of the Roman Empire) the edge pieces of the puzzle, representing the dating framework, can be set down with confidence. But before about the 7th century BC the task is different. The edges of the puzzle, in this case the chronologies of ancient Egypt and Mesopotamia, are not as certain as they are usually thought to be. The major argument of this book is that the dates conventionally attributed to ancient Egyptian history are inflated by as much as two and a half centuries. Imagine, then, trying to complete a jigsaw where the sides are far too long. Frustratingly, many pieces will appear to fit into two places in the puzzle, while many "ghost pieces" will be needed to fill the space that is unaccounted for.

This is precisely the dilemma into which so many archaeologists have been forced, dating and redating artefacts backwards and forwards across the span of the Dark Age, in attempting to fit their evidence into a framework defined by Egyptian chronology. Stretching the sides of the time puzzle by raising the dates further would only make the problems more acute. The only remedy, as our investigation shows, whould seem to be to shorten the sides and compress the overall scheme.

The idea of a radical shift in the chronology of this period is not entirely new. At the turn of the century the classical scholar Cecil Thorr and Egyptologist Jens Lieblein stood firm against the newly established "high" Egyptian chronology, but their arguments for a lower dating fell on stony ground. The next challenge to the status quo came in the 1950s from Immanuel Velikovsky, the wayward polymath whose work outraged scientists in many fields other than ancient history. His model for a "revised chronology" based on a new series of links between Egyptian and Israelite history, proved to be disastrously extreme. Involving a reduction of Egyptian dates by a full eight centuries at one point, it produced a rash of new problems far more severe than those it hoped to solve. Sadly, while he pointed the way to a solution by challenging Egyptian chronology, Velikovsky understood little of archaeology and nothing of stratigraphy.

Rocking the boat of course has never been popular in any field of study. Torr went against the grain of contemporary trends, while Velikovsky was too much of an outsider. But the major problem with the attempts of these writers was that they were working as individuals, and realistically could never have tackled the vast range of material from the many disciplines embroiled in the argument.
Preface pp. xviii-xxi

© Peter James