Shi Jinbo: Tangut Language and Manuscripts: An Introduction. (Languages of Asia Series.) xx, 547 pp. Leiden: Brill, 2020. €25. ISBN 978 0 04 41453 2.

sions with respect to the representation of tones and other structural features of Old Chinese phonology. Consider the different ways in which the author writes the Chinese medial *-r-. After fricatives and affricates, such as *sr-, *tsr-, *tshrand *dzr-, in William H. Baxter and Laurent Sagart (Old Chinese: A New Reconstruction, Oxford: Oxford University Press, 2014) and Schuessler (2007 and 2009), the medial is dropped, resulting in initials such as sh-, ch-, j-, and dj-, while in other cases it is retained. The consequences of this unfortunate decision are best shown in the reading for 瑟 *srit in Schuessler’s system, which must be spelt “shit” in Sampson’s system: the author tries to avoid this by turning to Baxter’s system in this single case and writing the character reading as “sprit”. Another example of inconsistency is aspiration, which is shown by the fact that the author uses tzand tsto write *tsand *tsh-, while retaining the letter h for other aspirated consonants. A further surprising decision is the spelling of what would be a glottal stop in all recent reconstruction systems. Here, the author uses, -mp, -nt, and -nk for what would be traditionally transcribed as *-mʔ, *-nʔ and *-ŋʔ and -c elsewhere. This collection of largely inconsistent and ad hoc decisions results in a transcription system full of ambiguities, with a symbol like h indicating both aspiration and retroflexion, s indicating aspiration, z occurring in both voiced and voiceless onsets, and p/t/k representing a glottal stop in addition to their values in traditional phonetic transcription systems. While this system is a nuisance for experts on Old Chinese phonology, given that it makes it extremely difficult and impractical to use the book as a reference or to look up words in the glossary, it is at best misleading for lay people who might easily take what looks like sounds in this transcription system at face value. Since at least Bernhard Karlgren (1950) and Wáng Lì (1980), editions of the Book of Odes with commentary, translations, and reconstructions have been considered the “royal discipline” of Sinology giving scholars the possibility of illustrating their individual treatment of Old Chinese with respect to both grammar and phonetics. Its future reception by experts in the field will show if this edition of the Odes can compete with its predecessors.