Wu Zhao, the only woman to rule as emperor of China, has long fascinated historians. But quite how much material we can use to explore her unique genius is amply on display in this highly revealing study of the way in which her agents commandeered what seems to have been all available representations of female power on her behalf so as to shape her public image to her advantage. The author has produced a work both wide ranging and thorough, yet in his introduction he is at pains to stress that his study concerns but one aspect of the multifaceted career of this extraordinary political genius. His list on p. 16, for example, of the male authority figures whom she also seems to have put to use gives some idea of the work that still remains to be done to describe the propaganda achievements of her regime – and even this leaves out the important bodhisattva Mañjuśrī, a figure later touched upon in connection with Wu Zhao's promotion of Wutaishan as an international Buddhist centre. The establishment of the bodhisattva's cult in this locality was in fact one of the most important religious legacies of her reign, for though one or two female emperors did reign in more or less Chinese fashion – and without the use of titles preserved for males – over one later polity, the Western Liao, the innovations discussed in this volume for better or worse did not find many echoes under subsequent dynasties.

This, however, does not in the least detract from the fascination of the many figures discussed in this book. Its twelve main chapters are divided into four groups: the first three discuss goddesses of antiquity, the next four dynastic mothers and exemplary mothers, while the third grouping treats three female Daoist divinities. The fourth section contains two chapters on Buddhist devis and goddesses. As is explained upon the introduction of these four categories, they roughly follow the main phases of her long political career – first, as the consort of an emperor, then as an empress dowager and mother to sons descended from Laozi, the reputed ancestor of the Tang imperial line, and finally as a Buddhist monarch ruling in her own right. Twelve makes for a good number, and one hesitates to disturb this scheme by adding a thirteenth figure that rather spoils this readily intelligible progression. But as the convenient tabulation of her successive titles on p. xvii shows, between 674 and 683 she was known as Tianhou, the Celestial Empress, a designation she assumed at the same point that her imperial spouse became known as the Tianhuang. The chorus of horror from later traditional historians naturally accuses her of absurd pretensions, rather overlooking (as I have explained elsewhere) the politically astute divine overtones of her husband's title, which they seem to accept without demur.

But although those historians do not offer any explanation of the overtones of her own title – either through distaste, or more probably through lack of a firm grasp of the ideological environment of the seventh century – the expression “Celestial Empress” also has specifically Daoist connections of great interest. In the opening narrative of the very old Daoist scripture known as the Lingbao wufu xu, a source explored by Stephen Bokenkamp in Journal of the American Oriental Society 106.1 (1986), pp. 65–77, the Celestial Empress is named as the consort of Shangdi, God on High, the deity at the very apex of the traditionally accepted pantheon. It is from the subterranean palace of this mighty goddess that a numinous text is retrieved and shown to Confucius, thus as it were affirming the acceptability of this figure to all currents within the Chinese tradition. The scripture that carries this story was certainly well known not only through sixth-century hagiography, as in the Daoxue zhuan of Ma Shu, but also in the seventh century, since it is cited in a Daoist encyclopaedia composed during the lifetime of the Celestial Empress and Emperor to be, the Shangqing dao leishi xiang of Wang Xuanhe.

It is worth noting, however, that in this case, as similarly in most if not all of the twelve others, no surviving document issued at the time makes as far as I am aware explicit connection between the “Celestial Empress” espoused to the ruler of the Tang dynasty and her subterranean yet celestial divine counterpart. Occasionally one feels that the author has affirmed clear imperial agency in establishing linkages of this sort where to the best of my knowledge there is no unambiguous proof that such was the case. Did the Emperor of the Sui dynasty really order an Indian monk to modify a Buddhist translation to include the prophecy of the future appearance of a Buddhist divinity in China as ruler of the Sui dynasty (p. 15)? This may be one rather probable inference from the existence of that prophecy in the translated text today, but I know of no copy of the emperor's order, and it is conceivable, even if less likely, that a foreign monk, or perhaps one of his Chinese assistants, seeking to make secure the patronage that had been offered for his translation efforts, could have inserted something bound to please his royal employer entirely on his own initiative.

In like fashion we find that the allegation, reported on p. 9, that the appearance due to seismic activity of a small new mountain in 686 was an omen of undue female power was in due course refuted by the revelation from a local official that he had been informed by an Indian monk that this was the very peak on which the Buddha had preached, miraculously and most auspiciously transported to China (Quan Tang wen 722.14b–15a). Was this bold resolution of the problem by the local official prompted by briefing from the centre, or did he think of it himself? Unfortunately few regimes in world history have been good enough to preserve details of the workings of their propaganda departments for the benefit of later researchers, let alone disclose them at the time, and China decidedly does not fall into that category. The appearance of good omens from far away might well represent the desire of provincial officials to gain some credit at court, though in the case of the awkward intrusion of the mountain this does not look likely to have been the case. But we can only infer, and no more.

The great benefit of Harry Rothschild's work, therefore, is that it shows larger patterns of the consistent manipulation of imagery over a considerable period of time in such a way as to suggest that behind them lay for the most part a single and very remarkable intelligence. The amount of material that has been drawn upon to achieve this result is impressive, even if as with all scholars working with intractable sources from long ago one or two inaccuracies, perhaps due to the carelessness of others, appear to have slipped through – I suspect that the Yixinfang attributed to Sui times on p. 160 is actually the later Japanese compilation Ishimpō, but this is of no great consequence. I should also register my unease at the use of the word “faiths” to describe Buddhism, Confucianism and Daoism – as has been pointed out by Maruyama Masao and Mori Mikisaburō, faith does seem to have been an element in eighteenth-century Japanese Confucianism, but that phenomenon seems to be far from typical, and possibly due to influence from a particular strand in Japanese Buddhism. But it is hard to find a nomenclature that precisely fits the circumstances of the seventh century: a rendering of these traditions simply as “teachings” could equally lay one open to the charge of under-translation. And such problems will not detain the reader who is ready to appreciate the real strengths of this book. It is heartening indeed to show how diligence and good academic sense can reveal facets of the achievements of a figure long vilified, especially one whose skill in shoring up her own position in the teeth of a hostile world still stimulates many resonances in the world that we ourselves now inhabit.