Japanese Monks and Chinese Books: Glimpses of Buddhist Sinology in Early Tokugawa Japan

: In the17th and 18th centuries, just as English scholars were reading and writing about their heritage in the continental prestige language of Latin, so too were Japanese members of the Buddhist clergy researching and publishing about the Chinese language heritage of their own religious tradition, drawing both on new printed books, often imported from China, and on much earlier manuscripts and printed texts preserved in their own country. The importation and reprinting of the canon by ¯Obaku monks and the subsequent ﬂowering of Zen scholarship is already well-known, but we should consider the efforts of Shingon monks in commenting on the heritage they received from China eight centuries earlier, and even the activities of Nichiren monks, who took steps to promote the legacy of Chinese Tiantai Buddhism. Critical reﬂection on the Buddhist tradition may not have emerged in Japan until the 18th century, but it did so in the context of a world of scholarship concerning an imported classical language that certainly stood comparison with that of the contemporary Anglophone world.


Introduction
The business of scholarship, especially if directed towards the past, has not generally been highly regarded beyond the narrow circle of its practitioners in the United Kingdom. "Always scribble, scribble, scribble! Eh! Mr. Gibbon!" William Henry, the first Duke of Gloucester (1743Gloucester ( -1805 supposedly jeered at one of the most outstanding of our number (Kemp 1999, p. 32). Fewer still, in Britain and beyond, have recognised that in China the patient work of generations of scholars did much to raise academic standards over the years in dealing with the heritage of the Chinese past (van der Loon 1961, p. 30). In Japan, moreover, while the work of earlier scholars within the Confucian tradition was recognised to some degree, the knowledge gained from Chinese sources of the Buddhist monks of the early Tokugawa period, even if many of us use it today, was much less celebrated. Even so, a comprehensive survey of their achievements would fill a book, if not several books, so the following remarks claim to do no more than illustrate some of the dimensions of the work of these normally less-than-fully-appreciated pioneers. Several of the figures mentioned here were important for the development of Japanese Buddhism, but this is not the angle from which their work will be examined, which relates solely to their use of Chinese books.
For reasons that will shortly become clear, the heroic age for such figures essentially began in the 17th century, by which point scholarly monks had a remarkablylarge quantity of Chinese materials upon which to draw. Historians of Chinese influence on Japan generally demarcate a first wave of imported culture peaking round about 800 AD, which prompted the importation of a large number of Chinese manuscripts, or copies of Chinese manuscripts, some of which survive to this day, though after the ninth century this traffic became much more sporadic (Sugimoto and Swain 1989, pp. 1-42). Importantly, however, those writing religious history saw another episode in the narrative of Sino-Japanese Buddhist contacts that gained indubitable significance from the mid-thirteenth century onward due to the unsettled conditions caused by the rise of the Mongols, again with consequences for Japan's Chinese studies, whereas those concerned with trade in general only see a new phase beginning in the fifteenth century (Bowring 2005, pp. 318-19;Sugimoto and Swain 1989, pp. 148-52). All scholars agreed, however, that the start of the 17th century saw the beginning of another very high peak of Chinese cultural influence, once again visible most prevalently in the importing or republication of very large numbers of Chinese books.
This new peak was particularly visible in the bibliographic record due to two consecutive developments. First, China in the 16th century saw the emergence of a print culture based on woodblock that, in terms of volume and resilience, eclipsed anything that had come before, demonstrating levels of production that were certainly capable of standing comparison with those that Europe achieved at the same time in the wake of the introduction of Gutenberg's technology (McDermott 2005). Secondly, the late 16th-century Japanese involvement with the Asian mainland, followed by the introduction of a long period of peace under the Tokugawa regime, ushered in an explosion of Japanese publishing based on the same technology, which had existed for some time but had been hampered by the limited development of a commercial market (Kornicki 1998). The scale of Buddhist publication made possible by these advances, which also made the skill of block cutting more widely known in East Asia, may be seen in one of the first English-language biographies of a 17th century Japanese Buddhist monk, devoted to the Zen master Tetsugen Dōkō鐡眼道 光 (1630-1682). One of his major undertakings was the printing of the entire series of Buddhist canonical texts, a project that involved the creation of over 60,000 woodblocks so that the series, as it eventually turned out, could be kept in print for a quarter of a millennium, producing nearly 2000 full copies (Baroni 2006, pp. 52-53, 147-58). However, Tetsugen's efforts were substantially assisted by his Chinese master, Yinyuan隠元 (1592-1673), who had arrived in Japan to escape Manchu domination after the fall of the Ming dynasty, carrying with him an entire set of the contemporary Chinese canon, which included his own writings, that could serve as a model for Tetsugen's work (Wu 2015, pp. 142, 147). It is against this background that the following paper should be read.

Zen Scholars and Zen Scholarship, Great and Small
Both Tetsugen and Yinyuan were major figures in the Zen or Chan tradition, the form of East Asian Buddhism that has long attracted the attention of Anglophone researchers. Although they were redoubtable promoters of the Buddha's word, they were busy men rather than cloistered academics. By contrast, their younger contemporary Mujaku Dōchū無著 道忠 (1653-1744), while initially a participant in the exchanges prompted by the importation of Yinyuan's new, distinctively Chinese style of practice, was fortunate to live a long life after leaving his administrative responsibilities to his monastic colleagues. This provided him with a budget to acquire important Chinese reference works, and so he was able to continue further in researching the textual legacy of the Zen tradition that proved extraordinarily inspirational to his 20th century successors, both in Japan and beyond. As a result, his scholarship has been well-described twice in English, first with reference to his encyclopaedic writings, and secondly with regard to his critiques of the Chan Buddhism of the Ming period, that became familiar in Japan during his lifetime due to both Yinyuan and the many imported Chinese works that became readily available, often through Japanese reprints (App 1987;Jorgensen 2007). 1 In these studies one may read of the many manuscript works he bequeathed to posterity, since his earlier administrative duties prompted the publication of only one treatise on monastic regulations during the course of his working career. However, the remaining 476 compositions he left behind at the time of his death at the age of 91, have been mined ever since by researchers, and two of his lexicographic compilations were so indispensable that they were eventually printed in the 20th century (Jorgensen 2007, pp. 26-27). His erudition, based upon a lifetime of diligent reading, was utterly intimidating: the bibliography of just one of his works of lexicography lists 479 Buddhist sources and 352 sources from Chinese secular literature (App 1987, p. 161). It is a relief to find that he was no mere pedant, and that his polemical comments on the shortcomings of famous Chinese masters did not preclude occasional ventures also into writing Chinese poetry. However, while his achievements compel admiration, he can hardly be regarded as typical of the scholar-monks of his time and place, since he was by any standards historical or contemporary plainly exceptional. As Jorgensen (2007, p. 38) points out, however, Mujaku Dōchū was certainly not alone in his pursuit of Zen historical studies, and he specifically brings forward as an example of another Zen scholar-monk from the same period Tokugan Yōson嚴養存 (1632-1703), and a work of his published as an appendix to a reprinted Chinese text in 1690.
However, Mujaku Dōchū always stayed close to his textual and lexicographic interests, whereas Tokugan Yōson's engagement with debates over the history of Chan lineages in China lured him into territory that was much more treacherous for anyone to venture into without a good library and disciplined reading habits. As I have shown elsewhere (Barrett 2020a, pp. 168-69), Yōson's assertions about Chinese source materials may have been taken seriously in early 20th century Japanese scholarship, but they do not in the least stand up to scrutiny in an age more fully endowed with library resources and especially with readily available reference works, both pre-modern and modern. Mujaku Dōchū's great weight of learning would make him a welcome adornment to any university department even today; Tokugan Yōson would struggle to get tenure.
Yet he was a man very much in tune with the temper of his own times. We know little about him, but we do know that he was interested in the cult of Xu Fu 徐福, and visited a Japanese site associated with it in 1679 (Ng 2014, p. 166). The legend of Xu Fu goes back as far as the account of the First Emperor in the Shi ji史記 (Records of the Grand Historian): a man of this name is said to have been dispatched by the Emperor with an entourage of young boys and girls to search for the Islands of the Immortals in the Eastern Sea, but to have returned asking for more resources, after which he sailed east never to be seen again (Watson 1993, pp. 49, 58, 61). In medieval Japan it had come to be believed that Xu Fu had actually landed in the Japanese archipelago with his companions and populated the islands, and to the Sinophile Japanese of the 17th century it was a more than welcome idea that they were in fact the descendants of these immigrants. This convenient legend had the effect of collapsing the longstanding tensions in Japan between continental and insular culture (Pollack 1986) so that devotees of Chinese learning might feel no sense of distance from the source of their texts.
Zen monks of the late 17th century were, however, especially deeply involved in the study of China because of contemporary events such as those that brought Yinyuan and his books into their lives. Even before this, within their tradition links with China, including trade links with some of their monasteries, had always been important, so opportunities to study Chinese, including vernacular Chinese, had always been greater than in other early modern Japanese schools of Buddhism. To look for the true standing of Sinology among the early Tokugawa clergy it is important to look also at other schools, such as those founded almost nine centuries before this time by the great establishers of Tang Buddhist civilisation in Japan, Saichō最澄 (767-822) and . Both these men, as the creators of the Japanese Tendai and Shingon schools respectively, founded traditions that were in every sense 'established', often closely linked with the court and the imperial institution and based simultaneously on monastic institutions that were in their own way similarly durable. This was certainly how the Tendai school persisted in the early Tokugawa period, when one of its leaders, holder in his time of the title of official head of the school, or Tendai zasu天台座主, had been born as one of the many sons of the emperor Gomizuno-o 後水尾 天皇 (reigned 1611-1629). This man was not simply an eminent dignitary within his own ecclesiastical hierarchy, but also a scholar who produced a ground-breaking research aid so useful that it was reprinted in the 20thcentury and is still worth being aware of even today.

Guardians of the Heian Heritage
The Sōdenhaiin僧傳排韵 (Monks' Biographies Arranged by Rhyme) of 1680 takes 48 collections of biographies devoted to the Buddhist clergy of China and distributes them according to the names of their subjects as sorted into the 200 or so rhymes used in the composition of Tang verse, and recorded in then well-known tables preserving the no longer current phonetic distinctions between the rhymes in tabulated form. Because these tables were memorized in East Asia by anyone wishing to write Classical Chinese poetry, they could be used as a quick reference system for Chinese characters instead of the 'radical and stroke count' system of Chinese dictionaries more familiar in the West. This work, authored by the ordained prince, Gyōjo尭恕 (1640-1685), has been briefly described in English (McMullen 1975, pp. 191-92); as he notes, it was so useful as a fully referenced biographical dictionary that it was reprinted in typeset form in 1911 in the series Dai Nihon BukkyōZensho大日本仏教全書 (Complete Writings on Japanese Buddhism), volumes 99 and 100, with newly compiled indexes according to the now normal Japanese kana order (since by then, in late Meiji Japan, the urge to compose poetry in Chinese, though not absent, had waned appreciably and a knowledge of Tang rhyme schemes was no longer so widespread). More information on the career and other publications of this prince-abbot may be found in an article in Japanese that appeared a couple of years before McMullen's summary, from which we learn that initially its 108 fascicles were printed up from woodblock in only a handful of copies for friends, though soon requests for multiple exemplars started to come in (Murayama 1973, p. 517). The other publications for which he is known were mainly digests of the best-known texts in the East Asian Buddhist canon, for as Tetsugen's new printed copies were added to the less commonly encountered earlier printings of the canon that had entered Japan in times past, the need to find shortcuts to the mastery of such a wealth of written sources had become more acute (Murayama 1973, pp. 517-18). The production of these shortcuts, encouraged by a newly widespread print culture, may be compared to European phenomena such as the 1737 Bible concordance of Alexander Cruden (1699-1770), a work which in principle gave access to the whereabouts of every single word used in the King James translation of the Bible.
Among the scholars who supported Gyōjo in his efforts to advertise his biographical masterpiece we find one name in a postscript which was certainly not that of a Tendai monk, but of a leader of the Shingon tradition. This tradition looked back through Kūkai to forms of Buddhism that had not long been introduced to China from India, so in looking at Tokugawa Buddhist Sinology we should not ignore the development at the same time of a newly visible Japanese interest in Indology as well, a trend that was not, as we shall see, confined to Buddhist Indology, nor simply to those in the Shingon tradition (Okitsu 2005). However, Kūkai was self-consciously a master also of Chinese learning, well-read for example in literary thought, on the basis of which he eventually compiled a handbook that now preserves much information from China not otherwise transmitted (Bodman 2020). However, even before he went to study in China in 804, he took pains to display his erudition in secular Chinese learning by composing a treatise on a theme already well established in China, comparing Buddhism with Confucianism and Daoism, the Sangōshiiki三教指帰, or Indications of the Goals of the Three Teachings (Hakeda 1972, pp. 101-39). It has long been known that both Saichō and Kūkai in their continental travels obtained copies of another work in this genre composed in China at some time in the middle of the eighth century and brought copies back to Japan, and in the early 21st century it was discovered that this work still survived there in manuscript. This work, Sanjiaobuqilu n三教不齊論, "Essay on The Inequality of the Three Teachings", by one Yao Bian姚 䛒 , has now been studied in some detail by a team led by Fujii Jun 藤井淳, since it has become clear both that it influenced some redrafting carried out by Kūkai on his own work after his return to Japan, and that it would also have been at least in principle available to all later commentators on his writings (Fujii 2016). 2 It has also been known for some time that a 12thcentury or later modified Chinese rewriting of Yao's work was printed in Japan under his name but under a new title, Sanjiaoyouliezhuan三教優劣傳, "Account of the Relative Value of the Three Teachings", in 1650 (Makita 1962).

Commentary Ancient and Modern
This is worth knowing, since Kūkai's writings were densely allusive, and so attracted repeated explication by learned followers of the Shingon tradition in later centuries, including the Shingon signatory to the Sōdenhaiin, Unshō運敞 (1614-1693). There is no readier illustration of the Sinology of a well-informed early Tokugawa monk than a reading of how Unshō handled the task of commenting on one of the most challenging texts within his tradition, and specifically that part of it that dealt with Daoism, since secular sources relating to Confucianism were generally well transmitted and well understood in 17th century Japan, whereas Daoist materials had only arrived in Japan piecemeal and much more sporadically, and were much more of a test of research skills. Of course Unshō was not always dependent on his own initiative in such matters, since clearly there was much in earlier manuscript commentary that he could readily put to good use, but as we shall see, he was living in an age during which scholarship was plainly starting to be undertaken on a somewhat different basis than hitherto, and an examination of the details of his work does allow for that transition to be studied from concrete examples, not just from summaries of existing academic research.
This process has now become more transparent than before, not simply because of the recovery of Yao Bian's work in its original form, but also because scholars in Japan have in recent years added to our knowledge of the medieval commentary tradition. The main repository for standard modern typeset editions of works on the Sangōshiiki is the fortieth volume of the pre-war Shingonshūzensho真言宗全書 (Complete Writings of the Shingon School). This includes not only Unshō and a younger contemporary, Tsūgen通玄 (1656-1731), but also full or partial earlier readings by the literary figure Fujiwara no Atsumitsu (Tonkō) 藤原敦光 (1063-1144), and by two anonymous writers, based on manuscripts transcribed in (though not necessarily composed in) 1220 and (as it would seem) 1372 (Takaoka 1935). (However, there seems to be a problem I have not been able to resolve with the 1372 date, either with the original scribe or with the reprint: the former seems to add a non-existent year within the reign name to the cyclical indication given, while the table of contents of the latter gives a date in 1239 or 1240. Therefore, or simplicity of reference, the date 1372 is used here to indicate this work.) However, Toyama City Public Library 富山市立図書館 has now published online a further commentary by the late 12th century figure Kakumyō覚明, entitledSangōshiikichū三教指歸注 (Notes on the Indications of the Goals of the Three Teachings) on the basis of a woodblock edition of 1620, and Fujii, using earlier research by another scholar in this field, has put online an edition of a 1133 manuscript commentary originally produced by Jōan成安 in 1088, the Sangōshiikichūshū三教指歸注集 (Collected Notes on the Indications of the Goals of the Three Teachings).
These early works are at a glance very different from the commentaries of the 17th and 18th centuries, wherein the influence of an age of print makes itself conspicuous. The sources of allusions are identified relatively succinctly, and while the sources cited may have been consulted directly, in the case of lesser-known works one suspects that the information may have been located at second hand in encyclopedias, though since this will have included at least one Chinese encyclopedia that no longer survives, establishing the point beyond question is unfortunately not entirely easy (Barrett 1980, p. 169). One aspect of the later age of print is a greater striving for encyclopedic detail in accounting for allusions, perhaps not simply because there were more books to read but also because of an increased sense of competition imparted by commercial factors in publishing. After all, while a 17th century British reader might well have been persuaded to part with his or her money for a book entitled The Compleat Angler, any such work more modestly claiming to represent 'a few tips on angling' might have placed itself at a commercial disadvantage. Izaak Walton's fishing classic of 1653 was not, however, based on book learning, so a more exact parallel might be a tome like Robert Burton's Anatomy of Melancholy of 1621, a massive compilation drawing on all kinds of sources ancient and modern, including some information on China drawn from Matteo Ricci, though here the striving for comprehensiveness seems to have come more from the author himself rather than from publishing pressures.
Similarly, to suggest that competition may have made for bigger books in Japan is not to suggest that contemporary learned monks were swayed by any commercial considerations there, but just that a secular culture of big books could also have affected their monasteries: it was now possible to write books stuffed with erudition in a new way. Another factor that seems to have progressively come into play globally was an increase in the scholarly sense that quotations in standard reference works were not always accurate, and that checking them against a full text was always worthwhile-the attitude summed up by the Oxford scholar Martin Routh (1755-1854), who is alleged to have told a young don seeking the benefit of his long accumulated wisdom "You will find it a very good practice always to verify your references, sir!", though I have never myself got round to verifying this quotation. The age of print further meant that the books required for checking such references were more abundantly at hand, and as we shall see in the Japanese case more detail is often given as to where to locate quotations within these sources, even if citing exact pagination of a specific edition lay as yet in the future. However, it is important to understand that many of the ancient sources of Tang date or even earlier that had been available as complete works or in early citations to commentators of the manuscript age were not necessarily available in either fashion in the age of print. Unfortunately commentators like Unshō or Tsūgen, on finding these citations in the works of their predecessors, did not always as a matter of course make it clear if they had managed to verify them by other means or no, a problem that has some bearing on whether the recently retrieved Tang text on the Three Teachings by Yao was known in 17th and 18th century Japan or not.
Unshō, for example, attributes several quotations in his commentary to a compilation, apparently made in the late fourth century, of 'key sources on nourishing life', the Yang-shengyaoji養生要集 (Collected Essentials on Nourishing Life), that had certainly disappeared as an integral work in China after the Tang. How long had it survived in Japan? It apparently did so either as an integral text, or as excerpted at some length in other intermediate materials, well into the 10th century. For some of his quotations of this source, perhaps, could have been taken from the Ishimpō医心方 (Prescriptions at the Heart of Medicine), a large and still surviving Japanese medical compendium of 982 that seems to have drawn directly from the Chinese text at that stage in its Japanese transmission (Barrett 1980, p.175, n. 28). However, though this work was printed in the 19th century and circulated widely in manuscript by that point, in the 17th century it was much rarer. On the other hand, one citation of the Yangshengyaoji by the Shingon monk that certainly survives also in Chinese compendia, and theoretically therefore could have been taken by him from such a source, is also listed as having been in Kakumyō's medieval period commentary, and though I have not been able to check it myself, this source was certainly consulted by Unshō according to the preface to his own work; indeed the passage entered the commentarial tradition even earlier than that, since it is also in the yet older manuscript of 1133 of Jōan's annotations, which for that matter are mentioned in preface as well, along with those of Fujiwara Tonkō. 3 How this ancient fragment got into the tradition of commentary in the first place is another question, but these earlier citations provide the most obvious answer as to where Unshō found the passage and its attribution.
A similar situation confronted him with regard to the Shenxianzhuan神仙傳, the collection of hagiographical notices translated by Robert Ford Campany as Traditions of Divine Transcendents (Campany 2002). This fourth century Chinese collection was already being cited, perhaps indirectly, by 831, and had certainly been imported into Japan as an integral work by 891. 4 It had even proved popular enough to inspire a Japanese imitation in the late eleventh century. 5 However, over subsequent centuries in both China and Japan the transmission of the text in its complete form does not seem to have been sustained, though in China in particular a large amount of its content was preserved in quotation. At least one of the repositories of all this quoted material in China then became very widely available in the 16th century, when the Chinese publishing industry achieved a level of commercial development not seen in Japan (as we have noted) until slightly later. Specifically, the printing in the mid-16th century of the Taiping guangji太平廣記 (Extended Records from the Taiping Period), the vast compilation under imperial auspices of pre-Song prose tales put together in 978 to 981, brought an impressive number of early sources back into circulation, and plainly had an immediate impact on late Ming writers-and even on early Tokugawa writers too. 6 However, for some late 16thcentury publishers it further allowed them to piece together passages from lost books so as to launch once more one of their more problematic forms of product, namely substantial if not necessarily complete versions of works long unknown to a wide readership. 7 Among such reconstructions we must count the 1592 publication in a major series of early works, of a Shenxianzhuan that waslargely derived from the new printing of this one compendious late 10th century source. 8 Now the earlier Shenxianzhuan of the age of manuscripts, which may well have been known directly to Kūkai, was certainly drawn upon to explicate his remarks on the Daoist section of his Sangōshiiki from at least the 11th century (Sangōshiikichūshū,p. 4,col.12,28). However, as I have already noted elsewhere, Unshō's treatment of phrases in Kūkai's text relating to this source reveals some interesting features of his working methods. 9 In Jōan's commentary a total of 144 characters are devoted to the immortal Huang Chuping黄初平 and 80 in the commentary of 1372, on the basis of the Shenxianzhuan, but Unshō omits all mention of him, though he could have enlarged considerably on their notes (Sangōshiikichūshū, p. 4, col. 28;Takaoka 1935, XL, p. 132b). Looking back earlier in the text, however, one finds that before reaching the point in Kūkai'scomposition where his predecessors had referred to Huang's details, he had tried to check in his copy of the Shenxianzhuan on the biography of Dong Weinian董威輦 or Jing京, but had not found it there, so perhaps the experience had discouraged him. 10 As Campany shows, in this case Dong's biography is simply among those that never entered the Taiping guangji, so Unshō was plainly thwarted in his hope of extending his reading in this case (Campany 2002, pp. 537-38). By contrast, however, where Jōan only gives a brief snippet from the Shenxianzhuan concerning the Isles of the Immortals, Unshō weighs in with a lengthy passage from that source concerning Liu An劉安, king of Huainan淮 南. 11 Here he certainly had the relevant biography in the printed work in front of him (Campany 2002, pp. 442-47).
These reactions may further be contrasted by what he does when he feels himself bibliographically on firmer ground, which can be seen in the case of a third Chinese work of Daoist inspiration. The origins of the Han Wudineizhuan漢武帝内傳 (The Inner Tale of Emperor Wu of the Han), the legend of the emperor's encounter with Xiwang Mu 西王 母, Queen Mother of the West, are by no means clear, but its transmission over the past millennium has been textually relatively stable and straightforward, at least in China, quite unlike either those of the Yangshengyaoji or the Shenxianzhuan. 12 Its text has furthermore been published as part of a pioneering monograph by Kristopher Marinus Schipper (1934Schipper ( -2021 in an edited nineteenth century form that allows for comparisons to be made between the best version in the Daoist Canon and that contained in the Taiping guangji, the source as it would appear of a separate printing in the same series of 1592 as that which included the first reconstituted Shenxianzhuan. 13 The Han Wudineizhuan was certainly present as a separate work in Japan in 891 (Ceugniet 2000, p. 172, n. 18), but seems to have required reintroduction from China by Unshō's time.
The story of the meeting between the Han emperor and the goddess was evidently well known to Kūkai, since he refers to it twice (Hakeda 1972, pp. 115, 117). These brief allusions were further expanded in some detail by Jōan from whatever text of the Han Wudineizhuan he had available to him (Sangōshiikichūshū, p. 2, col. 13-p. 3, col. 4). Though his account is abbreviated, comparison with the edition provided by Schipper makes it quite clear that he is drawing on a text much closer to that of the Daoist Canon than to the version contained in the Taiping guangji, and as a 12th century manuscript one hopes that its readings will be of value to future editors of this work. By contrast Unshō, while clearly aware of Jōan's version, does adopt some variants from the Taiping guangji tradition, and also restores some abbreviations imposed by Jōan, again using material apparently drawing on the Taiping guangji or a source dependent on it. 14 A full edition of all the Han Wudineizhuan quotations made by Shingon commentators would be necessary to prove conclusively that Unshō in this instance deployed printed books to supplement the evidence he had gained from earlier manuscripts, but that at least must constitute a good working hypothesis.
In the case of Tsūgen, however, matters are much clearer, since he seems to be much closer to the aforementioned Martin Routh in his approach. The Shenxianzhuan biography that Unshō could not trace he verifies by reference to excerpts in early Chinese encyclopedias, even telling us in which fascicle of each he has located the relevant passages, while he also makes it clear that his succinct account of Wudi and Xiwang Mu relies on Taiping guangji, fascicle 48. 15 A poem at the end of his commentary tells us that he had written it for younger persons, striving for concision and clarity over mere verbiage. Without a preface announcing his intentions one cannot tell whom he may have been thinking of chiding here, though one obvious candidate naturally comes to mind. I have already suggested that it would be wrong to impute merely commercial motives to Unshō; rather, his only fault seems to have been that he was something of a Kūkai 'completist' who could not bear to relinquish any effort in trying to squeeze every last element of meaning from his founding father's writings. We have already seen that Unshō mentions several predecessors from the manuscript tradition in his preface, but the way in which he describes them is worth noting too. Thus Kakumyō is commended for bringing together the annotations of his two predecessors, but at the same time he is also criticized for being less than exhaustive: 雖集 而大成，尚未爲詳悉, a criticism that Unshō certainly seems to have been at pains to avoid (Takaoka 1935, vol. XL, p. 153b, col. 5).
What I think a consideration of these examples of Unshō's approach to the task of writing commentary establishes is that he was an exegete with the resources of an early modern book world at his disposal, but without the constraints of academic method that developed later, as that world matured. This surely should be kept in mind in assessing his references in his preface to the Sanjiaobuqilun. Yao's essay was certainly known to the earlier manuscript tradition, even if it is not always clear whether commentators are drawing directly on his text or referring back to citations in the writings of their predecessors, sometimes including cases in which there is in fact possible evidence in favor of the former option. A lengthy series of excerpts by Tonkō, for example, could have been the source of a briefer, overlapping one in the commentary of 1239. 16 Laterhowever the commentary of 1239 draws on another passage from a different part of the text which cannot be found in the published, surviving part of Tonkō's work-though since it is the equivalent section on Daoism that is missing this may not be very significant. 17 By contrast Unshō mentions the Sanjiaobuqilun, but does not cite it even though he certainly had Tonkō's earlier commentary in front of him, and appears unaware of its authorship. 18 No one to my knowledge in the manuscript tradition associated with the Sangōshiiki uses the expanded rewrite, the Sanjiaoyouliezhuan, but here again Unshō lists it, presumably on the basis of its 1650 Japanese printing, since he assigns the authorship by Yao to the Song period, but explicitly at least makes no use of it either. However, between his mention of these two works he inserts also a reference to yet another work on the 'Three Teachings, a Bianliangsanjiaolun辯量三教論 (Essay Discriminating and Weighing Up Between the Three Teachings) by Fayun法雲 (Takaoka 1935, vol. XL, p. 153a). Unshō had certainly never seen this source, since although it is listed a couple of times in Buddhist bibliography of the seventh century, thereafter there is no trace of its existence whatsoever. Though he might have mentioned one other rather later Buddhist work on the Three Teachings, the likelihood is that these three titles were all the writings he was aware of that on the evidence of their titles discussed them from a Buddhist viewpoint. 19 He was simply being a 'completist', enumerating all relevant materials known to him whether he had read them or not, so there is no reason to suppose that he had a copy of Yao's Tang work in front of him at all.

Kamakura Buddhism and Its Curatorship of Earlier Tradition
One may object that, in the case of the Shingon monks whose work has just been described, the example of the Japanese founder of their tradition and of his writings gave them a particular interest in mastering the secular sources that he had drawn upon. However, the Chinese roots of all Japanese Buddhist schools, even those that had arisen from the efforts of Japanese founders during the ferment of the Kamakura period, were always there, and in the 17th century they seem to have come to the fore to a greater extent than in earlier times. A good example of this would be the Nichiren日蓮 monk SōsanGensei艸山元政 (1623-1668), who has been introduced to an Anglophone readership in the appealing poetry translations of Burton Watson (1925Watson ( -2017, from which it is fully evident that he wrote poetry as fluently in Chinese as a young John Milton (1608-1674) was then writing poetry in Latin. The main impression derived from the biographical sketch that precedes Watson's translations is of a quiet life, limited by ill health and primarily dedicated to the filial support of the poet's elderly mother. However, it is made quite clear that China also had an immediate impact on him, in the person of the refugee Chinese scholar Chen Yuanbin陳元贇 (1587-1671) (Watson 1983, p. xxiii). Chen in particular introduced him to the literary ideals of Yuan Hongdao袁宏道 (1568-1610), which greatly influenced his own writing both in kanbun Chinese and in Japanese (Chou 1988, pp. 44-48). Watson also notes in passing that his monastery had an outstanding collection of Chinese materials, and that the dynamism of the publishing industry in the Kyoto of his day meant that he had no difficulty in getting his works and those of his associates published there (Watson 1983, p. xxiv, n. 10;p. 99, n. 5). In fact, many of Gensei's publications were works of Buddhist scholarship, with the concerns of his school very much predominant; in a description of the evolution of Nichiren interpretations of the Lotus Sutra over time, his name occupies a prominent position. Moreover, he is thought to exemplify a movement of the period in exegesis back from its Kamakura starting point to consider its Chinese antecedents (Motai 1965, p. 644).
However, Gensei has at least one more academic achievement to his credit that has generally been overlooked. He was responsible for the printing in 1661 of the first edition of a Tang Buddhist work by a lay follower of Tiantai Buddhism text, which survived intact into the 20th century and otherwise would only be known from some fragments of a Southern Song edition: the Shandingzhiguan定止觀 (Abbreviated and Edited Great Cessation and Contemplation), by Liang Su梁肅 (752-793). This is not immediately apparent today, at least online, because in CBETA the portion of the text giving Gensei's editorial comments has become detached from the work in question and published separately; even the typeset edition of the supplement to the Canon on which CBETA is based eliminates one or two helpful details such as the name of the Kyoto publisher that Gensei used. Others in his school who picked up the ad fontes mood of the times, going back to the earlier historical levels of their heritage, expended greater efforts, but the republication of Chinese texts that have subsequently become rare or disappeared is an aspect of the Tokugawa scholarly achievement that should not be underestimated.
One follower in the Nichiren tradition whose work has proved of value to 20th century Anglophone scholarship on Chinese Buddhism was Nyokai如海 (?-1711). A study mentioning his name by Okitsu has already been cited above, since in 1695 he seems to have been the first independent modern publisher of the JinQishilun金七十論 (Suvarn . asaptatiśāstra), a work of non-Buddhist Sāṅkhya thought, translated in the sixth century and included thereafter in all editions of the canon (currently, T. 2137), but without attracting much attention at all. Okitsu quotes Nyokai's appended argument for his independent edition as affirming that this non-Buddhist philosophy was certainly superior to that of Confucius or Laozi, even if not on a par with Buddhism; he also notes a succession of studies of this work in later Tokugawa times that seem to have stemmed from this publication, though it may be that in promoting this proto-Indological venture Nyokai was simply being mindful of the commercial interests of his publisher (Okitsu 2005, p. 34). However, Nyokai certainly did not skimp on his own Chinese learning: his detailed biography of the great Chinese systematiser of the tradition upon which both Japanese Tiantaiand Nichiren drew, Zhiyi天 台智顗 (538-597), Zui Tendai Chishadaishikinenrokushōge隋天台大師智者記年録詳解 (Detailed Explanations of the Chronological Record of Zhiyi, Great Master of Tiantai), completed in 1688, is commended for its documentary thoroughness by Leon Hurvitz (Hurvitz 1962, p. 100). Perhaps even more significantly, the outline biography ofZhiyi upon which this more thoroughly documented work comments, Zui Tendai Chishadaishikinenroku (Chronological Record of Zhiyi, Great Master of Tiantai), begins with something perhaps more routine nowadays but surely very unusual for his time: a critical survey of the accuracy of all existing biographies, including an account of their interrelationships that extends to remarks on biographies that may once have been influential but were subsequently lost (Tendai 1987, pp. 378-80). This survey is not quite comprehensive: one late eighth century Chinese monk mentions Du Zhenglun杜正倫 (?-658) as a writer on Zhiyi, and his work seems to have disappeared thereafter (Wenyuanyinghua 863.5a3) and to have escaped Nyokai's attention. However, I know of no earlier East Asian biographer of anyone, Buddhist or not, who even tries to adopt such a modern source-critical approach: even such a model scholar as Zhu Xi 朱熹 (1130-1200), for example, may dispute specific points in editing a biography, but does not seem to attempt any systematic overview of his sources (Ma 1957, pp. 435-45).
Finally, lest it be thought that those in the Pure Land tradition did not participate in this upsurge of scholarly activity, we should note that the importation of the late Ming canon and the creation of Tetsugen's new edition was as much to the benefit of Pure Land Buddhists as it was to adherents of Zen, who because of Ming Chan's Pure Land tendencies were actually somewhat inclined to look this gift horse in the mouth. However, it was a Japanese Pure Land Buddhist, Ninchō忍澂 (1645-1711) who discoveredthat Tetsugen's edition was textually not as reliable as a thirteenth century Korean edition still preserved in carefully edited Japanese typeset editions of the 20th century that are used today (Mizuno 1982, pp. 182-83).
Before concluding, since I am conscious of having used the word 'modern' in various places, and of having referred to British scholars of the 17th and 18th centuries in passing, it may be helpful to introduce one such writer briefly, in order to explore possible comparisons a little further. I would hope that the examples of learned Japanese clerics mentioned so far may have persuaded the reader that these individuals and their responses to Chineselanguage materials do merit further study, if only for their heroic levels of engagement with extraordinary quantities of Chinese text. Beyond that however, might they not also have a place in world history? How might they compare in their working methods as scholars, despite the differences in their subject matter, with a British contemporary? In order not to tax the reader too much with any consideration of some mightier hero of Anglophone intellectual achievement, I have chosen figure generally remembered today as an unscholarly fool, though in fact his enduring success in at least one case raises the entire issue of how to assess minds across cultures that were plainly more sophisticated in some ways than those of most of their forebears, whilst still seeming to us in other ways very much part of an age that is now quite alien to us.

Early Modern Scholarship in Comparative Perspective
In some respects, Britain under the Tudor and Stuart rulers took a narrower view of the cosmos than inhabitants of Tokugawa Japan. To be precise, their world was believed to have been created not countless kalpas ago but at around 6 pm on 22 October 4004 BC, at least according to the well-known calculations of Bishop James Ussher (1581-1656). Only a rather small minority would support such a chronology today. However, it would be wrong for that reason to dismiss Ussher's scholarship in toto as not worth consulting any more. He may have based his chronology in this case on unsafe assumptions about his main source, the Bible, but he was a very widely read and methodologically careful man, a reader not only of Latin and Greek but also of other languages too, including Arabic, and a moderate in an age of intense political violence in Ireland and Britain who retained to the end the respect of both sides. In at least one other case, moreover, his scholarship remains completely valid to this day.
Ussher, like the Japanese, lived in a world in which the manuscript culture of earlier times had given way to a vigorous print culture that was still gathering up the heritage of the past and communicating it through the new technology to a wider readership. The task of scholars of his generation was to sort the wheat from the chaff, and to examine critically sources within the Christian tradition that were not always what they seemed to be. These included sources that reached back in time to very soon after the age of the New Testament, and in particular the letters of Ignatius of Antioch, who lived in the early second century AD. More than a dozen documents under his name had been in circulation before Ussher's time, and prior to the rise of Protestantism their authenticity had not been challenged. Ussher, however, in 1644 published a treatise in Latin pointing out that only seven of them withstood critical scrutiny, and adduced as part of his case two manuscripts in which a mere seven of the letters had been translated from Latin into Greek in 13th century England. One of these manuscripts is still in the library of Gonville and Caius College, Cambridge; the other was then in the possession of the cleric Richard Montagu (1577-1641). These seven he put forward as the genuine core within the corpus. 20 The details of Ussher's manuscript discoveries and their consequences for our understanding of the 13th century, and of the 17th century and later reactions to his Latin treatise, need not concern us here. 21 What is worth pointing out is that the Montagu manuscript was lost soon after he took full collation notes from it, making his work extremely valuable for the information he preserved on Greek sources in medieval England, and that the conclusions of his 17th century scholarship on Ignatius was subsequently confirmed by later historians, and have still remained worthy of mention in the 21st century. 22 At the same time as Ussher was making his breakthrough however, a print culture had come into being off the eastern coast of the Eurasian landmass that was marked by a very similar effort at examining and publishing an ancient continental religious heritage, yet that scholarly effort, with one or two recent exceptions mentioned above, has been so far largely overlooked in the Anglophone world, even though much that was achieved in those days is likewise of continued importance for contemporary research, as we have noted. Perhaps part of the reason for the neglect is that the clerics of Japan were no more writing in Japanese than Ussher was writing on Ignatius in English: both ends of the landmass were working throughout in a learned language that was not their own. In the Tokugawa Japanese world of learning, Chinese books had very much the same sort of role that books in Latin had in 17th century England, though with the advantage that there was no need to learn Chinese as a spoken language.
However, did the similarity go beyond materials and methods? If Ussher assumed the veracity of the Bible, did not his Japanese contemporaries assume the veracity of Buddhist texts? Of course, Confucians in China, and those in Japan who followed them, insisted that they were nonsense, but this was just a matter of partisanship: little had been done to demonstrate rather than simply affirm even that they were not necessarily the words of the Buddha. Ussher's success came with weighing up the authenticity of post-canonical writing; in Japan, the only person mentioned above who attempted something similar was Yōson, and his efforts turn out to have been unreliable. Do we see a failure of critical spirit in Japan, despite the advances in the organization of very high levels of erudition? If so, was this due to larger problems in the outlook of Japanese Buddhists, or mere chance? Consider, for example, what might have happened if Chen Yuanbin had introduced Gensei not to the latest in Ming literary thought but to someone like Jiao Hong 焦竑 (1540-1620), who (whatever his limitations) came closer to foreshadowing the evidential scholarship (kaozheng考證) style of the Qing era (Yü 2016, pp. 321-54)? Or might progress in East and West been more a matter of time spent collectively over the years in reading rather than any differential in intelligent responses to what that wide reading revealed?In addition, 17th-century Britain too experienced the benefit of a refugee scholar, Isaac Casaubon (1559-1614), even if it turns out that Richard Montagu was amongst those who sadly chose not to make him welcome. However, England had already been subject to the improving continental influence of Erasmus (1466-1536), and if we look for critical innovation in the study of Buddhism in Japan, it is certainly there later, however one wishes to present it, in the work of Tominaga Nakamoto 富永中本 (1715-1746) (Barrett 2016).
The Japanese case is intriguing. In China too scholarship became more precise, but there a suspicion of Western influence via the first Catholic missions has long complicated the picture, especially perhaps because 18th-century Chinese scholars sometimes were prepared to show a surprising level of acknowledgment of the virtues of Western methods, at least as a means of spurring on the efforts of their Chinese peers (Sela 2018). However, early Tokugawa Buddhists were working solely on their own heritage in an intellectual sphere remote from the knowledge brought by Catholic missions or later by Dutch contacts.

Conclusions
As a creative response to imported Chinese Buddhism, the scholasticism of early Tokugawa monks cannot be characterised as displaying religious creativity after the fashion of the great Japanese figures of the Kamakura period. Yet they did innovate, if only in matters of academic technique, an area where the break from the past is easily overlooked. However, it might be wrong to downplay the value of their efforts. This development has simply never been properly assessed, perhaps because it is only one facet in Sino-Japanese interactions, a field that itself still seems to count as a minor specialism in the study of East Asia. I may be completely wrong in suggesting that in fact the achievements of the scholars I have briefly mentioned, and of many more besides, are actually of substantial importance in world historical terms, particularly if we wish to understand how modernity might not be the outcome solely of the diffusion of Western learning but also (as many are now beginning to suspect) of simultaneous progressive developments elsewhere in the world. At least I hope to have persuaded the reader that such comparative investigations are worth pursuing in the future, wherever they may lead. This discovery also revealed that a work with the same title excerpted in a Dunhuang manuscript was an unrelated composition (Barrett 2020b). 3 The passage in question reads 一曰精二曰唾三曰涙四日涕五曰汗所以損人之者也and occurs in Unshō, Sangō shiiki chū 4, (Takaoka 1935, XL, p. 221a, col. 11-12); using the continuous pagination of the Shingonshū zensho volume rather than the individual pagination of its various texts. For a listing of this quotation as occurring in the second fascicle of Kakumyō's work, which I have been unable to access directly, see the listing of ancient works surviving as quoted in other texts provided online by Kyoto University, www.zinbun.kyoto-u.ac.jp/~takeda/edo_min/edo_bunka/syuitu/edono_kagaku_syuitu_03-14.html, accessed on 26 July 2020. For the same passage in the 1133 manuscript, see Sangō shiiki chūshū p. 6, col. 14, accessed 26 July 2020. Unshō's preface, p. 153b, cols. 4-5, lists his predecessors-we return to this document below. 4 (Penny 1996, p. 195), shows that it is cited-perhaps indirectly from some intermediate compilation-in the Japanese encyclopaedia Hifuryaku秘府略 of the former year; (Ceugniet 2000, p. 172, n. 18), lists it among the works of Daoist inspiration in the Nihonkoku genzaisho mokuroku日本国現在書目録 of the latter year. 5 (Ceugniet 2000, p. 172). The best study in a European language of this Japanese product, the Honchō Shinsenden 本朝神仙傳, is (Calzolari 1984). 6 The exact date of the first Ming edition of the Taiping guangji is unknown, though some details of the persons involved have been discovered: see (Zhang 2004, pp. 25-26). For one instance of the impact of this printing, see (Dudbridge, Glen 1983, p. 10); a Ri A Monogatari 李娃物語also seems to have resulted in Japan in the late sixteenth or early seventeenth centuries. 7 One such work of reconstitution that notoriously seems to have been carried out with scant regard for the authenticity of the material used is examined in (Thompson 1979). 8 Note (Campany 2002, p.122), and cf. the very useful table in (Penny 1996, pp. 207-9). The series of 1592 in which the reconstituted work first appeared was the Guang Han-Wei congshu廣漢魏叢書. 9 See p. 232 in particular (Barrett 2003, p. 232), though (as will have become apparent) the details given in n. 5 on p. 230 have now proved to be incorrect. 10 Unshō, Sangō shiiki chū 4, (Takaoka 1935, XL, pp. 210a-b), concluding 今本無載. 11 Jōan, p. 4, col. 13;Unshō, Sangō shiiki, 4, (Takaoka 1935), XL, p. 216a, col. 10-217a, col. 2; cf. also p. 229a, col. 2-4. Note that these examples do not exhaust possible passages of value in studying the use of the Shenxian zhuan in this commentary: see also pp. 208b-9a. 12 For some arguments for pushing its date of composition back from the sixth century (as favored in the monograph cited in the next note) into the fifth century, see (Barrett 2007, p. 488). Some would argue even for the late fourth century: cf. the summary of recent scholarship in (Knechtges and Chang 2010, pp. 350-51). 13 (Schipper 1965, p. 4); the critical edition contained in the Shoushange congshu 守山閣叢書 of 1844 is transcribed on 22 separate, reverse paginated pages following p. 144 of this study. 14 Unshō, Sangō shiiki chū 4, (Takaoka 1935), XL, pp. 213a-14a. Further extracts from Han Wudi neizhuan may be found on pp. 219a, 228b-29a. 15 Note the indications given in his interlineated references, Sangō shiiki kanchū 簡註, in (Takaoka 1935), vol. XL, p. 336b, col. 15;p. 338b, col. 15. 16 Compare these two works in (Takaoka 1935), vol. XL, p. 7b, col. 8 to p.8a, col.1, and p. 109a, col. 2-3, all taken from Sangō shiiki chūshū, p. 62 17 (Takaoka 1935), vol. XL, p. 22a cols. 3-4, and cf. Sangō shiiki chūshū, p. 65. 18 For the incorrect identification of the author here, see (Fujii 2016, p. viii). 19 A Yuan period work by Liu Mi 劉謐, the Sanjiao pingxin lun 三教平心論 (T2117) was certainly known in seventeenth century China, where it was printed as part of the Jiaxing Canon in 1658, but might not have been known in Japan when Unshō was writing. Records of lost books such as the Sanjiao quanheng 三教詮衡 by the seventh century medical writer Yang Shangshan 楊上 善 and the Sanjiao jiefen lun 三教解論 by the Daoist Sun Yizhong 孫夷中 may not have been known to Unshō, or may not have been considered relevant. 20 (Elrington 1864, p.106): "Libros enim duos nactus sum, alterum in publica collegii Gunwelli et Caii apud Cantabridgienses, alterum in privata D. Richardi Montacuti, Norwicensis nuper episcopi, bibliotheca repositum: qui Ignatianarum epistolarum interpretationem continebant, a vulgata nostra Latina divertissimam"; i.e., "I have come across two manuscript books, one kept in the college library of Gonville and Caius College, Cambridge, one in the private library of Richard Montagu, lately Bishop of Norwich, which contained a translation of the Ignatian letters very different from that of our usual Latin versions". 21 For a recent account, see (Lockadoo 2020). For ecclesiological and political reasons Ussher's work was soon attacked by no less a figure than John Milton (1608-1674); for a while in the nineteenth century further manuscript discoveries in Syriac appeared to cast doubt on his analysis, but as will be seen from the next note, by the late nineteenth century his deductions had been deemed irrefutable. 22 The full story of Ussher's manuscripts, and of other versions of the same genuine core discovered after 1644, is told in (Lightfoot 1889, pp. 73-109); for a recent reference to Ussher's key part in Ignatian scholarship, see (Ehrman 2003, pp. 206-11).