The third edition of Graham Connah’s African Civilizations, published almost thirty years after its first incarnation, is a substantially bigger book than the previous editions, primarily because the author made the bold decision to expand his coverage to the entirety of the African continent, whereas previous editions confined themselves to tropical Africa. While the geographical coverage now encompasses the whole of Africa, the organization of the book remains basically unchanged from earlier editions. An introductory chapter provides a “Context” for the descriptive chapters that follow, which are structured geographically, progressing roughly from north to south. As with the previous editions, the volume ends with a chapter that asks, “What are the common denominators?”

With the addition of new chapters, including one on the civilizations of ancient Egypt and the Mediterranean littoral of North Africa and one on the stone-walled settlements and Zulu capitals dating to the last five hundred years in the Highveld of South Africa, some pruning had to take place elsewhere in the book to meet, one assumes, the dictates of the publisher. This has led to a reduction in the theoretical discussion of the processes of state formation and urbanization, as well as judicious editing of the substantive chapters. However, Connah has made every effort to bring each of those chapters up to date with the results of recent research. As with previous editions, the end product is a remarkably comprehensive book on what is known of the archaeology of African civilizations, mostly prior to substantial European intervention. The emphasis on description is arguably both a blessing and a curse: a blessing because the book is a rich single-volume compendium of archaeological evidence, and a curse because the reader is assailed with data that seem to be included for the sake of comprehensive coverage but are not generally harnessed to any discussion of broader issues. The book is peppered, for example, with sentences listing such things as crops grown in different regions and types of artifacts and raw materials imported and exported, while numerous plans of cities and settlements seem to have been included merely because they exist.

Each of the ten substantive chapters is approached in the same way: after a couple of introductory paragraphs, there are sections titled “Geographical location and environmental factors,” “Sources of information” (itself subdivided into “Historical evidence” and “Archaeological evidence,” with the latter the longest part of each chapter), “Subsistence economy,” “Technology,” “Social system,” “Population pressures” (a title that is perhaps most telling in terms of interpretation), “Ideology,” and “External trade,” followed finally by a brief conclusion. This consistent structure was no doubt helpful for the author in marshalling his evidence and it also permits relatively easy comparisons between regions. However, the repetition of subjects in each chapter also makes the chapters seem repetitive. Moreover, as one proceeds chapter by chapter, one has soon grasped Connah’s interpretation of the process of state formation so that the final chapter on the “common denominators” offers few surprises. Indeed, what is particularly striking is that, aside from a somewhat backhanded acknowledgment of the “important idea of heterarchy”—a concept that Connah in the end dismisses as based on “a lack of evidence for hierarchy rather than on the presence of positive evidence for heterarchy” (353)—the author’s conclusions, and indeed much of the wording with which they are expressed, are virtually unchanged from the first edition.

According to Connah, control of highly productive land within the context of increasing population pressures was the “crucial common factor underlying the emergence of African elites” (351). Added to this basic ingredient was external long-distance trade, an “intensifier” of social changes that had begun with control of limited resources, such as copper, salt, iron, and ivory, which already circulated in “extensive internal trading networks” (350, emphasis in original). A further important ingredient was “religious ideology” that “legitimized and reinforced” the social hierarchy (352). Connah himself bemoans this “somewhat mechanistic” explanation for the origins of social complexity with a cri de coeur that “surely the reality is more complex than that” (353)—which itself is followed by a surprising suggestion that perhaps Jared Diamond has the best explanation in Guns, Germs and Steel (Vintage, 1998).

In summary, it is remarkable that a single author is still capable of mastering such a mass of archaeological evidence on African civilizations; one suspects that nobody will be in any hurry to follow in Graham Connah’s distinguished footsteps. However, one cannot entirely escape the impression that this book, with its “mechanistic” approach to state formation and its emphasis on data rather than theory and method, is a product of a bygone era, though that is not necessarily a bad thing.