This book is the first English translation of the Shifa manuscript in the fourth volume of the Tsinghua Bamboo Slips published in 2013. The original manuscript has no title: the current Chinese title, which literally means “methods of divination”, is given by the Tsinghua editors. As the subtitle of the book shows, the Shifa manuscript bears witness to a non-canonical tradition of stalk divination with an intricate pattern of stated and unstated rules. It also testifies once again to the ancient practice of representing hexagram or trigram lines with numbers rather than broken and unbroken symbols. Cook and Zhao's book will be the first stop for Western readers who have little or no knowledge of Chinese paleography but who are still interested in how archaeology has changed our understanding of divination in early China.

The translation is preceded by a long introduction that places the Shifa in the context of early Chinese occult practices: the canonical I Ching, other I Ching-related manuscripts, the Yin Yang Wuxing scheme, divination records in the Zuozhuan and the Guoyu, the Chu Silk Manuscript, excavated hemerological almanacs, divination records in Chu bamboo slips, and so on. The main difference between the Shifa and the canonical I Ching is that in the Shifa, we do not have 64 hexagrams with fixed names and line statements, but a more practical system of trigrams organized in 2 x 2 matrices whose meanings are determined in the very same context. The interpretation of these trigram matrices is governed sometimes by explicit rules stated in the reference information provided in the second half of the manuscript and other times by unstated rules that must be inferred from the trigram examples themselves. Although several rules can be recognized by keeping track of recurrent verbal patterns, the manuscript on the whole seems to lack any consistency in following these rules. They are not so much “rules” as hermeneutical guidelines that the diviner can consult when trying to figure out the meaning of a trigram set. In fact, we cannot be sure how many guidelines are actually employed in the manuscript and how the diviner may have chosen from them in a particular case.

Cook and Zhao address many such uncertainties well in the “Interpretation” part of each section, but the translation itself occasionally gives the reader too much confidence in the intelligibility of Shifa's language. More than once the translation glosses over potential controversies about the reading of a specific character or phrase. To give just one example, Cook and Zhao read the second character in Section 2.6 (strips 11 and 12) as fu “inverted” and base their whole interpretation of the section (pp. 80–81) on the idea of the upper left Dui being the visual inversion of the bottom left Xun. As they explain in the “Interpretation” of 2.6, this reading requires that we disregard the appearance of the extraordinary numbers 4, 5, and 9 in Dui and take the trigram as consisting of traditional Yin and Yang symbols. Yet several scholars have read the same character as ding, the name of the fourth heavenly stem, rendering the whole sentence as “The number of ding appears”. This reading seems equally plausible, not only because number 4 does indeed appear in Dui (and rarely elsewhere), but also because the text uses the word shu “number” explicitly. Cook and Zhao may disagree with this line of thinking, but they should at least acknowledge its existence and explain why they prefer the visual interpretation to the numerical one. This does not mean that they must produce a much longer translation with critical apparatus targeted only at the scholarly community. A few brief notes explaining alternative readings will suffice for a translation that introduces the Shifa to a general readership.

Another problem of Cook and Zhao's presentation of the manuscript is the confusing terminologies they use for describing the spatial layout of sections in the mini preface at the beginning of chapter 3 (pp. 53–70). To clarify the codicological complexity of the Shifa, they use a variety of terms, such as “division”, “section”, “sector”, “register”, “set”, and “pair”, to explain how textual units are distributed on the bamboo slips, but unfortunately the terms are used somewhat haphazardly. The same term is sometimes applied to different levels of division without a consistent principle of designation. It becomes disorienting after a while without an illustration of the entire manuscript divided into numbered sections. This kind of terminological confusion might be the reason for a few mistakes in the authors’ description of sections 14–18 on slips 24–31 (pp. 57–60). In this part, the correspondence between “register” (codicological unit), “section” (textual unit), and their illustrations is totally mixed up. Illustration 3.4 gives the text of Section 14 (as opposed to 14 and 15) divided into two registers, but Cook and Zhao mistake the second half of Section 14 for Section 15. Illustration 3.5 gives the text of Sections 15–18 in three registers, not 16–18 in two. In other words, on slips 24–31 there are a total of five registers of more or less equal dimensions, not four registers graduated in length. Section 14 spans the first two registers, Section 15 the third, Sections 16–17 the fourth, and Section 18 the fifth. These codicological details are not irrelevant, for nowhere else do we see such meticulous attention paid to textual alignment in early scribal culture. Whoever was responsible for copying the Shifa took great care with controlling the overall visual effect of the manuscript, sometimes even at the cost of inserting superfluous grammatical particles. It is desirable that the translators should preserve the codicological rigour with a better organized description of textual layout.

Despite these problems, the book is still a readable, engaging study of the Shifa and is indispensable for anyone interested in early Chinese divination. The translators have made an important contribution by integrating this new piece of archaeological evidence into our current state of knowledge.