Two Nineteenth-Century Copies of Joseph Prémare’s Notitia linguæ sinicæ

Abstract

French Jesuit Joseph Prémare, a missionary to Qing Dynasty China, had completed by the end of 1728 the draft of a book entitled Notitia linguæ sinicæ, intended to assist aspiring Catholic missionaries from Europe in learning the Chinese language. One of the original manuscripts sent from Canton to Paris, now held in the Bibliothæque nationale de France in Paris, was rediscovered in the 1810s by the French sinologist Jean-Pierre Abel-Rémusat, and became the source for some manuscript copies, made mostly before the work’s eventual publication in 1831. This paper examines two of these manuscript copies dated between 1825 and 1830, held respectively in the Archive of the School of Oriental and African Studies in London and the the Bayerische Staatsbibliothek in Munich, both important in their own ways, and both somewhat misunderstood and largely neglected by recent studies.

French Jesuit Joseph PrÉmare, a missionary to Qing Dynasty China, had completed by the end of 1728 the draft of a book entitled Notitia linguæ sinicæ, intended to assist aspiring Catholic missionaries from Europe in learning the Chinese language. While the first original manuscript sent from Canton to Paris is believed to be lost, there are two extant original manuscripts largely written in the hand of the author, one complete, the other not. The former, then held in the Royal Library (Bibliothèque du Roi) in Paris, was ‘rediscovered’ in the 1810s, and became the source for some manuscript copies in this first wave of limited circulation of handwritten copies, mostly made before the work’s eventual publication in 1831. There has been some academic interest in various aspects of this magnum opus of the renowned French sinologist, including its original manuscripts, later copies, and published editions.

This paper examines two of the manuscript copies made in the nineteenth century, both important in their own ways, and both somewhat misunderstood and largely neglected by recent studies. These two copies of Notitia, different in a number of ways, are potentially rewarding objects of investigation for several reasons. Firstly, both are copies of the copy made by the French sinologist Jean-Pierre Abel-Rémusat which was, in turn, copied from the original manuscript in the Royal Library. Secondly, the eventual publication of this important work in 1831 was based on one of these two copies, not the original manuscript or the Abel-Rémusat copy. While the original manuscript in the Royal Library and these two copies survive in their entirety, we cannot study the intermediate Abel-Rémusat copy because its whereabout is unknown.

The author of Notitia, Joseph Henri Marie de Prémare, was born in Cherbourg in France on 17 July 1666 and had entered the Society of Jesus on 17 September 1683. On 7 March 1698, at the relatively mature age of thirty-one, he left La Rochelle, the seaport on the Bay of Biscay, for China. As revealed in a letter of his dated 17 February 1699, he crossed the Atlantic Ocean, passed the Cape of Good Hope, Achen, Malacca, ‘two or three deserted islands’, and ‘finally, on October 6, after around seven months, arrived in China and anchored in front of Sancian’.¹ Shangchaun Island, named from the Portuguese São João (‘Saint John’), was the former Portuguese base on the southern tip of China until they relocated to Macao (Macau) with the blessing of the Ming Court in 1557. More importantly for Prémare, it was the final resting place of the Jesuit missionary St Francis Xavier, who died there on 3 December 1552, before he could enter the mainland of China. So, as one of his companions said on the journey, they started their ‘apostolate where St Francis Xavier finished his’.² Prémare was to remain in China for over thirty years, labouring mostly in the province of Jiangxi until the Christian faith was proscribed severely by Emperor Yongzheng in 1724, when he and others were first exiled to Canton and then expelled to Macao. He died in the Portuguese colony of Macao in September 1736, never setting foot again on the Chinese mainland.

As Prémare himself recalled years later, in a work written in elegant classical Chinese, when he first landed he did not know a word of Chinese, and had reverted to learning the language like a child: when someone ‘pointed to the sky and said “sky”, pointed to the earth and said “earth”, pointed to a human and said “human”’, he gladly followed.³ Since then, however, he had plunged into years of intensive study of the Chinese language, often ‘skipping meals and forgetting sleep, reciting and reading incessantly’.⁴

It may seem that to preach effectively to the Chinese, all Prémare had to do was to learn simple colloquial Chinese. But that was not the case, as he and his fellow Jesuits Joachim Bouvet and Jean-François Foucquet were all followers of the short-lived Figurist movement in China. They believed, as Knud Lundbæk has noted, that ‘the oldest Chinese literature . . . [was] not Chinese, but rather prophetical works belonging to the Judeo-Christian tradition, speaking not only of the true God, but also about the Messiah. If that was explained to the Chinese they would all become Christians’.⁵ To realize the objective of Figurism, then, would require delving deeply into the ancient Chinese texts, which would require intimate knowledge of, besides the day-to-day language, what Prémare termed ‘sinica oratione in nobiliori librorum’, ‘Chinese speech in noble books’, or Classical Chinese, rendered in the modern Pīnyīn romanization system as wényánwén 文言文.

Prémare laboured diligently in this direction and wrote works in Chinese ranging from Liùshū shíyì 六書實義, which revealed the ‘Christian’ mysteries hidden in the six types of Chinese characters, to Rújiào shíyì 儒教實義, which cited relevant ‘Christian’ passages from classical texts from the Confucian school, to Rú jiāoxìn 儒交信, which promoted Christian concepts by way of a more accessible novelette on the friendship between several fictional Chinese intellectuals. The efforts not only continued the tradition of accommodating Chinese culture in Christianity—started by Jesuit forefathers such as Mateo Ricci—but went further, attempting to dig out what was believed to be hidden in Chinese classical texts and culture that was compatible with the Christian faith, to promote the accommodation of Christianity in Chinese culture.

Whether these Figurist efforts were fruitful or not is a fascinating topic beyond the scope of this paper. But, in the process, Prémare became highly knowledgeable about the Chinese language, both in its day-to-day version and its more literary form. He was, indeed, as Abel-Rémusat put it, the leading grammarian and philologist in the Chinese mission field at the time, followed by Antoine Gaubil, while ‘Couplet, Noël, Parrenin … Amiot and Cibot … did not equal Prémare in the profound knowledge of the Chinese language’.⁶ In other words, he would have been the best qualified European to write a book on the Chinese language.

Qualification must be matched with motivation to produce concrete results. Notitia linguæ sinicæ, written in Latin and Chinese, was primarily aimed at aspiring European Catholic missionaries to the Chinese mission field. Prémare wrote it in Latin instead of French mainly because he wished to reach out beyond his compatriots. It was completed in his years of exile in Canton, when he could no longer engage directly in evangelizing to the Chinese people, probably as an indirect form of missionary work. Hence, the motivation, like that of most of his earlier works, was again in disseminating the Christian faith. He once wrote to the French Orientalist Étienne Fourmont that:

The hidden and ultimate end to which I dedicate . . . all my . . . writings, is to make sure, if I can, that everyone knows that Christianity is as old as the world, and that the God-man was most certainly known by the person(s) who invented the hieroglyphs of China, and composed the ancient Chinese classics; here, my dear, is the only motive which supported and energized me for more than thirty years in my studies.⁷

Earlier sinologists like Fourmont, Rémusat, and Henri Cordier have left us with a large number of primary and secondary sources on Prémare and his Notitia. Since then, there has been continued interest among scholars, in the East and the West, in various aspects of Prémare’s works, often in the context of Figurism or, when dealing with the Notitia, on bibliography or linguistics.

This paper focuses discussion on two specific manuscript copies of Notitia linguæ sinicæ, currently held respectively in two major libraries in Europe and both rarely reviewed and sometimes misunderstood by recent studies. In this paper, an original manuscript is defined as one that was drafted either directly by the author or by a copyist(s) authorized and supervised by him; while a manuscript copy is defined as one that was created later by a third person(s), typically not linked or known to the original author. In treating in some depth these two manuscript copies kept in an academic library in Great Britain and a public library in Germany, it is hoped that a deeper understanding of some currently unanswered questions may be found on this subject.

Before coming to address the manuscript copies, some background information is necessary on how they came about. In a letter dated 20 October 1728 to Fourmont, whom Prémare then considered his friend and supporter in Europe, he said: ‘I am sending you a fairly long work on knowledge of the Chinese language’ for ‘all missionaries and all those curious people of whatever nation’, hopefully to be published in Europe.⁸ Fourmont later recalled that he became worried when he received this letter in Paris in September 1729, for he had just completed a draft of a Chinese grammar himself. Would Prémare’s work make his own superfluous, since the Jesuit had been in China twenty-five years, and ‘had a perfect grasp of the language’?⁹ Fourmont’s friend, the Abbé Jean-Paul Bignon, then in charge of the Royal Library of Paris, proposed a wise solution—Fourmont could deposit his manuscript at the Library until Prémare’s book arrive, then the two could be compared. This was indeed what took place: the manuscript arrived on 11 January 1730, the comparison was made in a letter signed by Fourmont on 4 February and witnessed in another signed by Bignon shortly after, on 10 February.¹⁰

The story surrounding this first original manuscript of Prémare’s work is a long one and beyond our scope. Suffice it to say that Fourmont criticized it severely and probably had a hand in obstructing its publication (though, to be fair to Fourmont, there were also technical obstacles inherent in the book because of the large number of Chinese characters required); while he himself issued not one, but two, works on Chinese grammar shortly after Prémare’s death, in 1737 and then in 1742, prompting some to accuse him of possible plagiarism.¹¹ Fourmont also kept his copy of the Notitia very much to himself and, as far as we know, this first original manuscript has since been lost.

Fortunately, there was at least another original manuscript, slightly different from that described by Fourmont, which was forgotten in the Royal Library of Paris until the 1810s, when it was ‘found in the cabinet of oriental manuscripts, despite the then curator’ by Jean-Pierre Abel-Rémusat,¹² who had been appointed in 1814 to a new chair in Chinese and Manchurian languages and literature at the Collège de France. The Royal Library had been renamed the Bibliothèque nationale de France since the First French Republic of 1792, and this original manuscript is now held there under the mark MSS Orient. Chinois 9259. Soon after its rediscovery, a ‘very exact copy from this original was made’ by Abel-Rémusat, who added to it an index he had composed.¹³ He also introduced it to the students who were studying Chinese under him, and some of them made (partial) copies from his, starting a limited circulation of the Notitia by way of handwritten copies.

There is another original manuscript of the Notitia now in the British Library under the mark O/C ADD. 11707. This manuscript, ‘purchased at Klaproth’s sale at Paris, April 1840’, is incomplete but important as it contains pages from a chapter which was also in Fourmont’s copy but absent in the original manuscript in the Bibliothèque nationale de France.¹⁴ However, this manuscript has little to do with the nineteenth-century copies, so will only be referred to occasionally in this paper.

In his Élémens de la Grammaire Chinoise, Abel-Rémusat praised Prémare’s work profusely, and lamented that: ‘this is what will always obstruct the publication of this precious work, that it contains little less than twelve thousand examples and fifty thousand Chinese characters’, making it prohibitively expensive to produce in Europe at that time.¹⁵ As noted in his letter dated 14 September 1824 and addressed to Abel-Rémusat, this statement prompted the Irish nobleman Lord Kingsborough to ask for a manuscript copy of the Notitia and ‘to arrange with Dr … [Morrison] the publication of it’.¹⁶ The arrangement was generous, as the Anglo-Chinese College of Malacca, found by Dr Robert Morrison in 1818, would receive a donation of £1,500 with the sole condition that its interest income from the initial years should cover the expenses of the publication of this work.

Stanislas Aignan Julien, one of Abel-Rémusat’s more promising students who was later to assume his mentor’s post at the Collège de France, was commissioned in 1825 to make a copy to be used for the publication of the Notitia in Asia. It was based on Abel-Rémusat’s copy, not the original manuscript in the Bibliothèque nationale, and followed his teacher’s copy closely, including what Lundbæk termed Abel-Rémusat’s ‘mistaken purism’, which was to reverse the direction of the Chinese text, writing it ‘the Chinese way’, from right to left.¹⁷ While this was technically ‘proper’, it also made reading of the bilingual text awkward, as the reader had to switch direction each time the text changed from Latin to Chinese, and vice versa. It should be noted that this was not Julien’s customary practice. When he published his Meng Tseu in 1824, the Chinese and the Western texts were both written in the same direction, from left to right.¹⁸

This copy completed by Julien in August 1825 is indisputedly the most important of the nineteenth-century copies of the Notitia, as it was the bridge linking it to the book’s next stage of circulation as a printed edition. Here, we shall call it ‘Julien Copy A’, the suffix ‘A’ necessitated by the existence of at least two known later copies made by Julien, one, the British Library copy O/C ADD. 11708, also purchased at the Klaproth sale in 1840, the other previously noted by Cordier as owned by Marquis d’Hervey de Saint-Denys, now in private hands and recently on the market through a reputable antiquarian book dealer.¹⁹ We can refer to these, respectively, as Julien Copy B and Julien Copy C.

Julien Copy A was sent by Abel-Rémusat to Kingsborough, who then passed it to Morrison in February 1826. Morrison ‘embarked for China’ on 1 May 1826 via the Orwell, which called on Java and Singapore in August and landed at Macao on 19 September 1826.²⁰ He transferred the manuscript to the Anglo-Chinese College in nearby Malacca, when he was in Singapore. Work started soon after, but the College reported in 1829 that the project ‘has been unavoidably delayed’ because of, among other things, ‘the great number of Chinese characters required to be engraved’.²¹ Finally, in a letter dated 7 March 1831 from Samuel Kidd, the principal of the College, it was reported that ‘the Latin Chinese Grammar was finished in January last; it has proved a very tedious undertaking & contains about 300 pages Quarto’.²²

Our main interest is not the printed edition but Julien Copy A, which was lost from sight until it was revealed in Cordier’s Bibliotheca Sinica that a copy of the Notitia made by Julien, a large folio bound in leather, was ‘in the library of the learned Mr. Alexander Wylie’ in Shanghai. This was clearly Julien Copy A, as on its first page is written that ‘this Grammar was presented to the Anglo-Chinese College, through the President Dr. Morrison by a Gentleman’, and on the last page (p. 386), it was indicated that this was ‘manuscripto ex scripsit Stanislas Julien’.²³

For some reason, this important manuscript copy has been generally neglected by recent studies on Prémare and the Notitia, albeit repeating to a greater or lesser extent Cordier’s bibliographic information on it. Two recent journal articles by the same writer from 2013 and 2019 concerning Prémare’s work both contain a table entitled ‘A list of different extant editions of the Notitia’, purportedly listing all known manuscripts and published editions of the book.²⁴ The earlier table lists two original manuscripts from the British Library and the Bibliothèque nationale and Julien Copy B from the British Library, all noted above. The later table was updated to include one more manuscript from the Bavarian State Library in Munich, which will be discussed later. But both tables omitted Julien Copy A.

That the lists of extant copies of the Notitia do not include Julien Copy C in private hands is expected, for, after all, it would not be known unless one happened to be monitoring the antiquarian book market closely. But that it omitted Julien Copy A is most unusual, since it is well known and arguably the most important of the nineteenth-century copies, being the one used to produce the first published edition in 1831. One must conclude that it has been mistakenly assumed that its whereabouts are unknown. Julien Copy A is, in fact, well preserved in the Library of the School of Oriental and African Studies in London, with the call-mark CWML MSS 300. A careful examination of this copy of the Notitia would yield rich rewards in terms of both provenance and contents, above and beyond what was noted in Bibliotheca Sinica.

On the first unnumbered page of this copy to carry writing is the handwritten note by Morrison, mentioned by Cordier, declaring that:

This Grammar was presented to the Anglo-Chinese College through the President Dr. Morrison, by “A Gentleman”. The President engages on the behalf of the College to have this Grammar Printed from Funds provided by the Gentleman aforesaid. London, Language Institution 27 Bartletts Buildings, Feb. 15, 1826.

The ‘Language Institution, in Aid of the Propagation of Christianity’, was set up in 1826 by supporters of foreign missions, during Morrison’s one and only furlough back in England, as it was felt that while there were mission societies sending out missionaries to many foreign countries and Bible societies printing the Scriptures for them, there were ‘no effectual means … for imparting a knowledge of the languages spoken in those countries’. The Right Hon. Lord Bexley became its president, and the Institution was housed in Bartlett’s Buildings, ‘a convenient and centrical situation’. From the start, Dr Morrison was engaged to teach Chinese, and Mr Townley the Bengalee language, and the two, as indicated in the Institution’s first report, ‘may well be considered as the authors and founders of the Institution itself’.²⁵

On the second page, also unnumbered, there are other notes. The one in the middle of the page, in the hand of Morrison, indicated that ‘this Grammar is the Property of the Anglochinese (sic) College Malacca. Feb 15. 1826’, also noted by Cordier and providing no new information. But the one at the bottom, in a different and heavier hand, states, ‘Given to A. Wylie by Rev. E. J. Eitel of the London Missionary Society, Hong Kong, 1870’. This gives important information on provenance, but Cordier chose to ignore it. Based on this inscription, we may finally understand how Julien Copy A, owned by the Anglo-Chinese College, got into the hands of Wylie. In 1843, the College was relocated by its then principal, Dr James Legge, to the newly founded British colony of Hong Kong. But it was ‘closed at the end of the year 1856 owing to the results not justifying its continuance’.²⁶ By that time, Dr Morrison and his son John Robert Morrison had both passed away, so the properties of the College were taken over by its other main sponsor, the London Missionary Society. In 1870, the Revd Ernst Johann Eitel was attached to the Society, based in Hong Kong.

Wylie had met Legge for the first time in London in 1846, when the latter was in England for a visit, seeking ‘some guidance in prosecuting the study of Chinese’, having already acquired tolerable proficiency in the same with only a copy of the (Malacca edition of) Notitia ‘picked up, the winter before, at an old-book stall’.²⁷ It is, therefore, not surprising that he would be interested in having a manuscript copy of Prémare’s work. It is, however, debatable as to whether Eitel had the right to transfer this copy, which belonged to the former College, to Wylie, who at that time had also joined the London Missionary Society.

At the back of the first unnumbered page was another handwritten note not mentioned by Cordier, reading, undated, ‘From A. Wylie to his dear friend W. Lockhart Esq.’ William Lockhart was a medical missionary to China belonging to the same mission society and had worked in various parts of China, while Wylie was head of the London Mission Press in Shanghai. Once, in December 1854, the two took considerable risks in venturing out alone to negotiate with the Chinese triads for a peaceful resolution in a conflict between the latter, the Qing Imperial army and the French naval forces in Shanghai under Admiral Laguerre.²⁸ We know Wylie had sold his substantial collection of Chinese books, ‘including c. 100 Ming edns and a collection of c. 900 19^th-cent. Protestant missionary publications’, to the Bodleian Library in 1882,²⁹ but apparently this manuscript copy was not among them. Instead it was given to his friend and co-worker, who had become a director of the London Missionary Society, who, in turn, later donated all his books to the library of the Society.

In 1966 and then 1972, the London Missionary Society merged respectively with two other mission societies and formed the Council for World Mission, reorganized in 1977. According to archival records in the Library of the School of Oriental and African Studies, from 1973 onwards the Council for World Mission transferred all its records, files, publications and manuscripts to the School on indefinite loan. Among the items transferred were, of course, Julien Copy A.

In short, a closer examination of the handwritten notes on the first two unnumbered pages of Julien Copy A has yielded additional insights into the provenance of the manuscript above and beyond what was already known through Cordier, enabling its ownership history to be mapped from its creation in 1825 to the present day.

Following the two unnumbered pages noted above is the body of Julien Copy A. Prémare’s Notitia is generally divided into Introduction (Introductio), Part One (Pars Prima) and Part Two (Pars Secunda), each divided into chapter (caput), and then subdivided into article, paragraph and number (articulus, paragraphus, numerus) as appropriate. The pagination of the three main parts in this copy are as follows: Introduction, pp. 1–51, Part One, pp. 52–213 (p. 214 blank), and Part Two, pp. 215–386. This is then followed by twenty-four pages of Index, not numbered.

Besides page numbers, Julien Copy A also has more detailed passage numbers marked on the left and right margins of each opening, from no. 1 on p. 1 to no. 510 on p. 385. While both original manuscripts in the Bibliothèque nationale de France and the British Library have more than one pagination series, neither has any passage number in the margins. Fourmont had also not mentioned the existence of passage numbers in the original manuscript which Prémare sent to him. Since Julien copied from Abel-Rémusat’s copy, it could be safely concluded that these references were added by Abel-Rémusat. The helpful passage numbers were not reproduced in the edition published by Malacca in 1831, possibly due to the additional technical challenge of adding numbers in the margins.

The Introduction contains three chapters, on the Chinese books (libris sinicis), the Chinese characters (characteribus sinicis), and nine extended lists of Chinese characters organized not alphabetically by their romanized pronunciation, but according to their (end) rhyme (secundum rhythmos), useful for writing or appreciating Chinese poetry. Part One deals with the day-to-day language (lingua vulgari) and contains two chapters with articles and paragraphs, each of the latter typically concerns the usage of one Chinese character. Part Two discusses the literary Chinese style in ‘noble books’ (sinica oratione in nobiliori librorum), and contains five chapters, the first four subdivided into articles, paragraphs and even numbers, each again typically explains the usage of a character; and Chapter Five, a collection of elegant phrases (elegantium locutionum colletio) of different lengths. The text ends abruptly with a paragraph entitled ‘selected phrases with five characters’ (selectiores phrases quinque litterarum), which lacks any entry, rendering the copy incomplete.

This empty paragraph was not unique to Julien Copy A or the Abel-Rémusat copy on which it was based, as the original manuscript held by the Bibliothèque nationale de France likewise contains an empty last paragraph for Chapter Five of Part Two, on p. 232. A comparison could not be made with the original manuscript in the British Library, as it has only Part One, missing both the Introduction and Part Two. Although the first original manuscript sent to Fourmont is believed to be lost, Fourmont did provide a detailed description of it in his Grammatica duplex, and this source confirmed that this last paragraph in that manuscript also did not have any entry. Prémare for some reason neglected to at least cross out the title of this empty paragraph on the last page of the manuscripts, giving the critical Fourmont another occasion to declare: ‘your book is not perfect’.³⁰

At any rate, on the bottom left-hand corner of this last page of the text of Julien Copy A are two notes in Latin by Julien. The first one, signed by the copyist, confirmed that ‘this Chinese language grammar manuscript was copied by Stanislas Julien’, and the second one, a boxed annotation below in the same hand, indicated that the copy was completed on 25 August 1825, ‘after 600 hours of assiduous labour’.³¹

As Cordier had noted, the Chinese characters and phrases inside the text were generally written horizontally from right to left, while the Latin texts were written from left to right, making it necessary to switch reading direction every time the passage involves a change in language. This is different from the treatment of the Chinese texts in the two original manuscripts held in Paris and in London and is, as has been already observed, a practice adopted by Abel-Rémusat when he copied from the original in the Royal Library. However, many extended quotations in Chinese were written in square blocks from right to left and vertically from top to bottom, as is customary in books in Chinese. This treatment is consistent between Julien Copy A and the two original manuscripts.

From time to time, in the margins of the pages of Julien Copy A, corrected Latin or Chinese texts were written in ink by the copyist, indicating that the manuscript had been carefully proof-read at least once by Julien. Abel-Rémusat stated that the Malacca printed version of the Notita was ‘a representation, faithful to the minutia, of the primitive manuscript and the transcriptions … without the slightest alteration, without the smallest addition’.³² This assessment was largely valid, but not entirely so. There were also occasional markings in pencil in a different hand in the margins, usually additions or corrections of Chinese texts, likely that of the editors or proof-readers at the Anglo-Chinese College. For example, on p. 177 or passage no. 237, the first two double Chinese characters for the phrase ‘yão yão paì paì 搖搖擺擺(yáoyáo băibăi, to swing, or to walk with a swing)’,³³ omitted in Julien Copy A, were added in pencil in the margin, and hence the printed edition of 1831, on p. 121. As these two characters were not missing in the original manuscript in the Bibliothèque nationale de France, it could be deduced that the omission was either by Abel-Rémusat or Julien. We will return to this point later in the paper.

An extensive index followed the last page of the text, done in a different hand. Two other annotations in French provide additional information on the Index. A note signed by Abel-Rémusat (initialled ‘JP. A-R.’) at the top of the second unnumbered page before the text indicated that it was a ‘manuscript of the Notitia of P. Prémare, with ten packets of cards, containing the Index’.³⁴ This seems initially puzzling as the manuscript contains at the end an extensive index written in unnumbered two-column pages, but no packet of index cards. The answer to this puzzle is at the bottom left corner of page [24], the last page of the Index, which indicated that ‘the Index (was) transcribed from cards by Richard Boswell, March 20, 1826’.³⁵ Cordier did not mention the first note by Abel-Rémusat explicitly, though he did so implicitly by inserting the word ‘of Rémusat’ in parenthesis directly after the word ‘cards’ in Boswell’s note above.³⁶

Boswell, incidentally, was the ‘Assistant Secretary’ of the Language Institution, who ‘resides on the premises, and conducts the general business of the Society’.³⁷ He was apparently among the first thirteen students who had attended Morrison’s gratis three-month course in Chinese. Judging from the fact that most entries in the Index contain Chinese characters besides their romanized forms, Boswell had advanced considerably in his study of the language when he was commissioned to transcribe the Index. Abel-Rémusat’s note confirmed that when received in London, Julien Copy A had only the body text and ten decks of cards, which were then converted into index pages attached to the end, before the manuscript copy was bound. We do not know what happened to the packets of index cards although Morrison might well have kept them for his own reference.

Besides Julien Copy A discussed above, there is another manuscript copy of the Notitia which is often neglected and, most recently, also misidentified. This copy, entitled ‘P. Henrici a Prémare Notitia linguæ Sinicæ, 1830’, is now held in the Bayerische Staatsbibliothek in Munich, under the call-mark Cod. Sin. 1, [Guangzhou] 1830. This manuscript was bound, unusually, in Chinese-style blue-cloth wrapped boards, not in the typical Western calf or vellum, as was the case with all the other extant original manuscripts or copies of the Notitia extant from the eighteenth and nineteenth centuries. The seal of the Royal Library of Munich (Bibliotheca Regia Monacensis) is found on some pages of this copy, confirming that it was added into the library’s collection before 1919, when that library was renamed. The manuscript itself does not contain any reference to the Chinese city Guangzhou or its earlier equivalent ‘Canton’, but this city name is part of the catalogue description of the item.

Bibliotheca Sinica’s entry on this copy consists of two lines of title and two lines of text which indicate that it ‘contains only what was printed in Malacca; without interest’.³⁸ Perhaps this assessment from Cordier provides the best explanation as to why the manuscript has been largely neglected by most studies, past or present, despite the fact that, of all the extant manuscripts of the Notitia, this is the only one that has been fully digitized and is easily available online, thanks to the impressive digitization efforts of the Bayerische Staatsbibliothek. But, in fact, this copy has some unique features and is worthy of further study. It will be referred to here as the German Copy, though at a later stage a new label will be suggested.

The German Copy has a total of 197 pages, and the ink pagination by the copyist of the three main parts are: Introduction, pp. 1–14, Part One, pp. 15– 93, p. 94 being blank, and Part Two, pp. 95–197, followed by an unnumbered blank page. There is also a set of pages not bound with the body text, consisting of two pages which listed the contents, with pencilled page numbers, pp. 199–200; and an extensive index, with two unnumbered title-pages and pencilled numbers for pp. 203–226, the index entries. It appears that the pagination of the unattached Table of Contents and Index pages were added later to create a consistent set of page numbers as a continuation from the text.

While the body text of Julian Copy A consists of 386 pages, that of the German Copy consists only of 197 pages, or roughly half the former. The main reason is that the texts in the latter copy were more tightly written—it contains both more words per line and more lines per page. But there are also occasional omissions in the German Copy. In the Introduction, some Chinese characters were omitted, leaving only the romanized forms. But the most notable omissions are the lists of Chinese characters grouped by rhyme, leaving only a line for each list. For example: the first list for the end rhyme ‘a’ in Julien Copy A contained close to eighty characters rhyming with ‘a’, ‘ya’ and ‘oua’. This was reduced in the German Copy to one short line (‘1. a 2. ya 3. oua’) without any character. Chapter Three of the Introduction took up thirty-eight pages, from pp. 19–51, in Julien Copy A, but only three pages in the German Copy, from pp. 12–14. The copyist noted simply: ‘it does not seem good to me to copy’ these mere lists.³⁹

In Part One, aside from the differences in pagination, Julien Copy A and the German Copy have basically the same contents for the two chapters. But at the end of Chapter Two (the end of Part One for Julien Copy A) the copyist of the German Copy added several short paragraphs on p. 93 under the title ‘Chapter Three (Caput Tertium)’. These describe briefly its contents, noting that ‘Father Prémare said that there are three chapters in this part’, and then listed the titles of the two articles in the third chapter, namely ‘a variety of refined modes of expressions in ordered gathering’ and ‘some dialogues or accounts in polite writing’, but provided no specific contents of either article.⁴⁰ Finally, the copyist observed inconclusively about the two absent articles of Chapter Three, ‘I do not know whether it was the Parisian Professor (Abel-Rémusat) who had removed them or rather Father Prémare had neglected to compose them’.⁴¹

It appears that the copyist did not have access to any manuscript containing the contents of Chapter Three—the article titles mentioned by him are exactly as described by Fourmont in Grammatica Duplex.⁴² Furthermore, it turned out that both of his conjectures were invalid. Abel-Rémusat did not remove Chapter Three of Part One from his copy of the original discovered in the Royal Library—that latter manuscript only has two chapters to this part. Nor did Prémare neglect to compose Chapter Three, as is confirmed by the fact that the partial original manuscript now held in the British Library does contain Chapter Three of Part One, albeit it is incomplete, with only eight paragraphs to the first article and nothing for the second.⁴³

In Part Two, consisting of five chapters, the contents of the German Copy and Julien Copy A are again largely the same. Occasionally the Chinese characters were omitted in the former. For example: in the first page of Chapter Three (p. 128) of the German Copy, ‘kou ven (gŭwén, or classical Chinese)’, ‘y chi et chu (Yì, Shī and Shū, or the ancient Book of Change, Book of Poetry and Book of Documents)’, and ‘chan-hai-king (Shānhăi jīng, or the Classic of Mountains and Seas)’ were only noted in romanized forms, without their equivalent Chinese characters, which are古文,易,詩,書 and 山 海經 respectively, unlike in Julian Copy A. Also, at the end of Article Six of Chapter Four in Julien Copy A, there is a bilingual version of a short treatise entitled De Deo et attributis divinis (‘Of God and Divine Attributes’); in the German Copy, there is only the Latin version, with four blank pages (pp. 167–170), presumably to be used for transcribing the Chinese version at a later date, suggesting that the copyist had limited time when making the copy.

Evidence from the text confirms that both the German Copy and Julien Copy A were based on Abel-Rémusat’s copy. Like Julien Copy A, the German Copy also has passage numbers marked in the left and right margins of each opening. These numbers go from no. 1 on p. 1 to no. 510 on p. 196 in the German Copy, and they all appear next to the same passages as those in Julien Copy A. The direction of the Chinese texts when intermingled with the Latin texts in the German Copy also went from right to left. In the original manuscript then held by the Royal Library in Paris, Prémare crossed out a paragraph regarding the Chinese novel yo kiao li 玉 嬌梨 (Yù jiāo lí, translated and published later by Abel-Rémusat as Iu-kiaoli ou Les Deux Cousines), ⁴⁴ though it was still legible. Abel-Rémusat added an asterisk in front of the crossed-out passage, removed the latter from the text and added it back in as a footnote, in effect ‘reviving’ the passage. This same treatment appeared in Julien Copy A and in the German Copy (passage no. 62 in both copies). These features provide strong textual evidence that, like Julien Copy A, the German Copy was also made from the Abel-Rémusat Copy, not from the original manuscript in the Royal Library.

A check on p. 74 or passage no. 237 in the German Copy reveals that all four Chinese characters for the phrase ‘yão yão paì paì’, referred to above, are present. This is convincing proof that the Abel-Rémusat copy was correct, and the first two characters missing in Julien Copy A were indeed left out only by Julien.

Finally, the German Copy contains a set of pages for a table of contents and an index, not bound with the body text but placed at the end of it. The two pages dedicated to the Table of Contents list the chapters, articles, paragraphs and even numbers of Introduction, Part One and Part Two. This helpful list of contents is missing in Julien Copy A and, consequently, also the first edition issued in Malacca in 1831. The Index has a cover page with a long descriptive title noting that it is an index of ‘all things worthy of note, which occur in Prémare’,⁴⁵ followed by entries in romanized forms and sometimes with their equivalent Chinese characters, listed alphabetically, with the related page numbers and then passage numbers in parentheses. There is also an index in Julien Copy A, not titled, and it also contains numerous entries in romanized forms, almost all with the equivalent Chinese characters, alphabetically arranged, again with reference to both the page numbers and passage numbers. The published edition in 1831 has an index based on Julien Copy A but has only page numbers.

The two indices of the manuscript copies are not entirely identical. For instance, the German Copy has 104 entries under the letter ‘C’ and fifty-four entries under ‘V’, but Julien Copy A has 107 entries under ‘C’ and fifty-two under ‘V’. Their alphabetical order also differs in terms of placement of entries under ‘J’ and ‘Y’. The German Copy adopted the order ‘I-J-Y-K’ but Julien Copy A used the order ‘I-Y-J-K’. It seems that the alphabetical order of Julien Copy A (done by Boswell) is more logical: putting entries under ‘Y’, or ‘ix’ in Latin, after those under ‘I’; then putting entries under ‘J’, non-existent in the Latin alphabets, directly before those under ‘K’, since in both English and in French, ‘J’ precedes ‘K’. Finally, as already noted, the entries in the German Copy often tended to drop the Chinese characters next to the romanized forms of the entries.

Other than these minor discrepancies, almost all the entries of the two indices are identical. For example, all the first entries on p. 1 of the index of Julian Copy A are exactly the same as the first entries in three columns on pp. 203–5 of the German Copy, except the latter split one entry into two, switched the order (by mistake) of two entries, and at times omitted Chinese characters and used a vertical stroke to represent repeated ones in adjacent entries. These indicate that, despite the slight differences, the Index of the German Copy was copied from an index based on the same packets of index cards originally done by Abel-Rémusat, not an index developed independently by its copyist.

The 2019 article by Lĭ Zhēn mentioned above (n. 24) is, so far as I am aware, the only recent academic study dealing exclusively with the manuscript held by the Bayerische Staatsbibliothek. The conclusion of that article is that ‘the copyist of this manuscript copy was the French sinologist Julien, and he had completed his copying work in Paris on October 7, 1825’, and, furthermore, when copying, Julien had added ‘a table of contents and an index composed by himself’.⁴⁶ The conclusions were based on a small annotation in Latin at the end of the Index pages of the German Copy, which Lĭ Zhēn transcribed as ‘Explicit Julien. Hunc indicem ordine disposui ac scripsi. St. Julien. Parisiis 7a die 8bris 1825’ and translated as ‘Julien hereby stopped writing. This index was arranged and written by me. Julien, October 7, 1825, Paris’. Furthermore, since the handwriting in the body text of the German Copy appears very similar to that of its Index pages, ‘it could basically be deduced that the manuscript copy is also completed by Julien himself, and the text should be finished even before the index, which was done on October 7, 1825’.⁴⁷

These conclusions are likely to be incorrect. Firstly, as noted earlier, the almost identical entries of the indices of the German Copy and Julien Copy A (which was based on Abel-Rémusat’s index cards) would easily refute any suggestion that the index of the former was ‘composed by’ its copyist, who-ever he was. But, more importantly, the sole support for linking the German Copy to Julien—the notation at the end of its Index pages—has probably been misread.

Julien Copy A was definitely made by Julien, an exercise finished on 25 August 1825, after ‘600 hours’ of labour. The German Copy was dated 1830 on its cover, but also has the date ‘October 7, 1825’ on the last page of its separate sheets for the Table of Contents and Index. As they were copied around the same time and were both based on Abel-Rémusat’s copy, if they were both by Julien then the handwriting would be identical or at least remarkably similar. Figure 1 shows on the left the annotation at the end of the body-text pages of Julien Copy A, and on the right that at the end of the index pages of the German Copy. It is obvious that the copyist who wrote the Index of the German Copy was not Julien, as his handwriting style and especially the signature on the third line is very different from that in the Julien Copy A, including the signature also on the third line.

The closing annotation on the Index pages of the German Copy has clearly been misinterpreted. The second word on the third row is indeed ‘Julien’, but the second word on the first row is not ‘Julien’ but ‘Index’, making the first line ‘Explicit Index’.⁴⁸ So the first row can be translated as: “Here ends the Index”. Some readers may be puzzled by this reading, as the first letter of the word “Index” and that of the word “Julien” clearly appear to be the same, as may be seen in the excerpt in Figure 1. It should be noted that this copyist often (though not always) wrote majuscule “I” and “J” exactly the same way, as in the “I” in “Infinitivum” and the “J” in the Romanized form “Jo”, both of which appear on the same page of the Index in the German Copy.⁴⁹

The remaining three lines could indeed be translated as ‘I have ordered this index and written it. Stanislas Julien. Paris, 7 October 1825’ (‘Hunc indicem ordine disposui ac scripsi. St. Julien. Parisiis 7a die 8bris 1825’). Since the copyist of the German Copy, whom we already know was not Julien, merely copied from another index, there was no need to order anything before copying. But this additional step would indeed have been necessary when Julien was working from packets of loose cards written by Abel-Rémusat. In short, this annotation could easily mean that the Index in the German Copy was made on a date unstated from another index which was, in turn, ‘ordered’ and ‘written’ by Julien in October 1825. The later copyist was simply so faithful to the original copy of the Index done by Julien that he kept the first-person verbs used by Julien.

At this point it would be instructive to compare the handwriting of the text of the German Copy and Julian Copy A to ensure that they were indeed done by different copyists. It is true that the same copyist’s manuscripts can differ somewhat depending on many factors, including time, place, purpose, what was being copied, and the copyist’s maturity and grasp of the language(s), etc. Fortunately, because the two copies were made around the same time (between 1825 and 1830) in Paris from the same master copy, they are likely to be rather similar if indeed executed by the same person. The present author is not a palaeographer, and the following comparisons are meant to be supplementary to the other analyses here, supporting their conclusions rather than being standalone proofs.

Figure 2 shows Latin words from these two manuscripts placed side by side, all typically drawn from comparable parts of the text. It is clear from the Latin that the handwriting of the two manuscripts is quite different in style and extremely unlikely to have been done by the same copyist, especially given that they were supposedly completed one after the other in 1825 in the space of months. In the first excerpt, ‘Notitia(e) linguæ Sinicæ Pars prima’, note in particular the N and the two occurrences of t in the first word, the g in the second word, and the S in the third word in the first line, and both the P and the p in the second line. In ‘Omnes reducuntur ad’, note the O, the terminal s of the first word and the two occurrences of d in the second and third words. In ‘atque haec tria’, note the two occurrences of t, the q in the first word and the t in the third; in ‘articulis alphabeticus’, the p and especially the two occurrences of terminal s; in ‘et significat’, the s, f and t; in ‘Dum stupidus’, the initial D and the final s; and, finally, the M and ss in the word ‘Missionnarii’.

Figure 3 shows some Chinese characters or phrases, again drawn from the texts of the two copies. One does not have to know the language to see that the texts were copied by two different people. (Note the Chinese texts in the first column of Figure 3 go from left to right, consistent with the direction of the pinyin romanization and English texts, while those in the second and third columns taken from the manuscript copies go from right to left). The character jīng 經 (‘classics/latitude’) in the first row and the character năo 腦 (‘brain’) in the second row share in their upper right corner the same component, itself a Chinese character, chuān 川 (‘river’), as illustrated in the third row (though this last character on its own was not used in the manuscript). Note that Julien and the copyist of the German Copy executed the chuān component very differently, and this difference in treatment applied consistently to all Chinese characters containing this component. To some extent, the unique way in which the copyist of the German Copy wrote this component could be viewed as his ‘signature’ in the Chinese texts. Note also the different ways of writing the character xiào 笑 (‘laugh’) in the fourth row, the second and fourth characters jiàn 見 (‘see’) and miàn 面 (‘face’) (the two together, literally see face, means ‘meet’) in the fifth row, both the characters in the sixth row, but especially the character qīng 清(‘clear’), and, finally, all three characters shuō bùdé 説不得 (‘can’t say’) in the last row.

In short, as illustrated by the Latin and the Chinese texts in the copies, the German Copy could not have been written by Julien, as concluded erroneously in the 2019 Chinese article, and must be by a third copyist. But who might he be? In search of the true copyist of the German Copy, the remainder of this paper will take a two-step approach, mirroring the discovery process which took place when the investigation was conducted. First, historical evidence will be presented to support a good ‘educated guess’ that he might well be the German orientalist Karl Friedrich Neumann. Then a comparison of the handwriting of the German Copy and other known manuscripts by Neumann will be made to support this conjecture.

Though not widely known, Neumann had studied Chinese under Abel-Rémusat in 1828, before he made his trip out to China a year later. Heinrich Julius Klaproth stated in Briefe über den Fortgang der Asiatischen Studien in Paris, issued in 1830, that:

Remusat’s lectures are also attended by the learned Professor Neumann from Munich, who, with great versatility and a clear mind, possesses a very thorough knowledge of Armenian and other Asian languages, which he intends to apply to a desirable work on the polities of the people of this part of the world.⁵⁰

In a book published in 1831, Neumann also admitted that Abel-Rémusat, ‘that celebrated professor of Paris, was kind enough to instruct the author in the rudiments of the Chinese language’, and that he felt ‘indebted to the strictly grammatical precepts of that learned gentleman’ for his progress ‘in the study of so very difficult an idiom’.⁵¹ His gratitude did not prevent him from criticizing his former teacher a few years later, when he declared that Abel-Rémusat’s Èlemens de la grammaire chinoise ‘is a mere excerpt’ of the great work by the Jesuit Prémare, the Notitia, citing many examples of places where the former had borrowed from the latter.⁵² (Although, as an aside, it should be noted that Abel-Rémusat did praise Prémare repeatedly in his work, and freely admitted his intellectual debts to the Jesuit father, if not quite in accordance with the strict citation requirements of today.)

More importantly for our discussion here, even though Neumann did not stay long in Paris, he had access to Abel-Rémusat’s copy of the Notitia, as revealed in an undated letter from Klaproth, which noted that:

Like most of his students, Mr. Rémusat also gave him [Neumann] the copy of the Notitia linguæ sinicæ he had made himself and which had been cleaned of spelling and editing errors, so that he could use it and copy it as he wished.⁵³

In other words, Neumann certainly had a copy of Abel-Rémusat’s Notitia before he went to London in the spring of 1829, after a whole year in Paris.

While in England, Neumann secured an opportunity for free passage to Asia in exchange for French lessons given to D. I. Ward, the captain of the Sir David Scott, a ship belonging to the East India Company. He then went back to Germany for funding to buy Chinese books in China. While his efforts failed in Munich, he managed to get some money in Berlin from the Royal Library (Königlich Bibliothek) of Prussia.⁵⁴ He apparently also took on private commission(s). For example, he wrote to Klaproth on 3 November 1829 and informed him of his intention to ‘sail from England to Canton at the end of April’, and would be ‘happy to handle all (Klaproth’s) orders (for Chinese books)’, which he could deliver in 1831.⁵⁵

Neumann left England on 17 April 1830 and arrived in Macao on 7 September. He then went to Canton in October and stayed there in the foreign traders’ quarters as the guest of the British merchant Lancelot Dent for three months. Neumann was also introduced to Morrison through the latter’s life-long friend and supporter Sir George Thomas Staunton. In a letter dated 8 November 1830, Morrison informed Staunton of his understanding that Neumann was in China ‘to purchase books for the King’s Library’ and noted that he had ‘rendered every civility’ in his power to the German scholar. The two got along well, as Morrison said that Professor Neuman’s ‘general knowledge and good sense made him an agreeable visitor in our family (in Macao), frequently before I came up to Canton’.⁵⁶ In the same letter there was another piece of news of interest. Morrison told Staunton that ‘Pere Premare’s Notitia Linguæ Sinicæ is just issuing from the College. It is in quarto. I will send a copy as soon as possible’.⁵⁷

In those days, foreigners were forbidden by the Qing court to either learn Chinese or acquire Chinese books, but Neumann had managed to do both discreetly while in Canton, with the help of a Chinese teacher. He bought many books ‘in the large bookstores located within the city, which foreigners who were restricted to some streets in the suburbs were not allowed to visit’. Then, finally, as Hartmut Walravens reported:

more than 12,000 volumes, covering all branches of Chinese knowledge, and quite a few rarities from the Celestial Empire were ready to be transported to Europe, which, however, again were forbidden due to an export ban for books. The Chinese customs officials were, however, persuaded by the usual arguments to declare the books as paper.⁵⁸

One could easily imagine that while Neumann was dealing with the book traders, he could have had one of them bind his manuscript of the Notitia. This would explain why the German Copy has a blue-cloth Chinese binding instead of the typical Western leather binding of all the other manuscripts of the same work, as well as why the title-page of this copy would include the year ‘1830’, for that would indeed be the year when it was bound, though not the year when it was copied, which should be late 1828 or, at the latest, early 1829. The index pages were clearly copied from a copy made by Julien in 1825. While we cannot be sure when exactly Neumann made his copy of the index, it is reasonable to assume that it was probably completed while Neumann was in Paris in 1828 and 1829, but, together with the table of contents he himself composed possibly also around the same time, intentionally left unbound for ease of reference when studying the contents.

Neumann departed China with his treasures in early February 1831 and arrived in London on 24 May 1831. He stayed in England for a while and was able to secure a copy of the printed edition of the Notitia fresh from the press. A document entitled ‘Memorandum Respecting Prémare’s Notitia linguæ Sinicæ’, held by the headquarters of the London Missionary Society, noted that ninety-four copies of the Notitia had arrived in London on 19 August 1831, the Directors fixed the price at £1 1s, and Professor Neumann picked up a copy at the Society’s office on 20 September, as he was ‘still in London on his way to Germany’. It appears that he was probably the first outside buyer of this work, for twenty copies were sent out only on 6 October to ‘Parrbury [sic, Parbury?] & Allen for sale’, as directed by the Board of the Society.⁵⁹

In September Neumann went to Berlin and delivered 2,400 Chinese books purchased on behalf of the Royal Library of Prussia. Efforts to negotiate for the sale of other books bought at his own expense were unsuccessful there, so he returned to his hometown of Munich and struck a deal with the remaining items, including at least 3,500 books (although there seems to be some disagreement among scholars as to the exact number of volumes transfer red).⁶⁰ The deal with the Royal Library of Munich was not in cash but was in exchange for his appointment as curator of the Chinese collection of the library and also Professor of Armenian and Chinese Languages (Professor der armenischen und chinesischen Sprache) at the University of Munich in 1833. Given Neumann’s love of his subject and the desire of the Royal Library to own an excellent collection of Chinese texts, this was no doubt a much better win-win arrangement than a straightforward sale.

Despite the unusual nature of his expertise, he was apparently a rather popular lecturer among the students. But his politics was to cause him trouble in later years. He was sympathetic towards liberal political ideas and was at the ‘Preliminary Parliament’ (Vorparlament) which met in Frankfurt 1848, ‘mainly as an observer’.⁶¹ This led to subsequent investigation and eventually it cost him his university position in 1852. He then became a freelance writer, based in Berlin from 1863, and continued to publish historical works on Asia and America.⁶² He had a stroke in 1867 and died in Berlin on 17 March 1870.

Given Neumann’s long association with the Bayerische Staatsbibliothek, it is not surprising that many of his manuscripts and documents would pass to the library upon his death. The library holds a number of manuscripts identified as his, some of which have been digitized, including several volumes of his ‘Memorials’, his hand-written book catalogues, several manuscripts of his works, his diary kept during his visit to Paris and London in 1840, etc. It would be reasonable to expect that his copy of the Notitia would also have been transferred to the Library sometime after his passing. At any rate, the existence of these manuscripts in Neumann’s own hand made comparison with the German Copy possible.

Figure 4 gathers excerpts from the German Copy, here labelled as ‘Neumann’s Notitia’, and from three known manuscripts in Neumann’s own hand, namely his diary from 1840 and two catalogues drafted by him in the 1860s. Note that unlike in Figures 2 and 3, it is no longer possible to compare exactly the same texts; these excerpts are merely such as to allow com parison of letters or numbers rather than identical words or phrases. Despite the different words and sometimes different languages between the texts in the German Copy of the Notitia and Neumann’s other manuscripts, one can still quite confidently ascertain that they exhibit very similar handwriting styles, suggesting that the former was indeed done by Neumann. Note the dates and especially the year 1830 in the first two rows, the capital N in the third row, the two occurrences of d and the t in the Latin text ‘quando indigent’ in the left-hand column and the French text ‘le droit de faire’ on the right in the fourth row; also the terminal s of the two Latin words ‘articulis alphabeticus’ on the left and the two French words ‘les provinces’ on the right, and, last but not least, the distinctive way in which the Chinese character jīng 經 was written, in particular, the component chuān 川 in the upper right-hand corner of the character.

In short, the handwriting evidence, in conjunction with historical and textual analyses, strongly supports the conclusion that the German Copy could not have been written by Julien but was executed by Neumann. Therefore, it could be referred to as the ‘Neumann Copy’ instead.

Joseph Prémare, the Jesuit father from France, completed his manuscript for Notitia linguæ Sinicæ in Latin and Chinese in Canton in 1728 and sent it back to Paris in the same year in hopes of having it published and circulated there. This work by the man who was at the time arguably the most proficient European in the Chinese language took a novel approach to learning the language, namely:

It gives the rules separately … justifying each assertion with numerous examples; or, to put it better, it brings out the rules being proposed through the comparison of texts taken from the best authors, brought together and explained with great care.⁶³

In other words, it is a grammar book light on grammar, and heavy on examples, effective perhaps in aiding the acquisition of the language, but requiring a huge number of Chinese characters, so much so that its publication in Europe in the eighteenth or nineteenth centuries would have been prohibitively costly due to the need to cast many Chinese fonts.

The Jesuit father’s dream of circulating the work in Europe remained unfulfilled for nearly a century, until one of the original manuscripts was rediscovered in the Royal Library of Paris in the early nineteenth century by the French sinologist Jean-Pierre Abel-Rémusat. Abel-Rémusat made a careful copy of this manuscript and his students made copies from his copy, starting the first, albeit limited, wave of circulation for this important European work on the Chinese language, by way of handwritten copies between the 1810s and the 1840s. This was soon followed by three successive waves of wider circulation in the form of printed copies—its first Latin edition published in Malacca in 1831, an English translation issued in Canton in 1847, and a second Latin edition reprinted in Hong Kong in 1893.

This paper has reviewed the historical background leading to the need for and creation of a small number of copies of Prémare’s Notitia in Paris in the first part of the nineteenth century, after the original manuscript was rediscovered. It then examined closely two of these copies, both generally forgotten and misunderstood in different ways by current research. Firstly, the copy held in the Library of the School of Oriental and African Studies in London, CWML MSS 300, was examined. This copy, completed by Stanislas Julien in August 1825 in Paris and referred to in this paper as Julien Copy A, was the manuscript used to produce the first printed edition of the Notitia by the Anglo-Chinese College in Malacca in 1831. It has been rarely discussed in recent studies, despite its importance, presumably because its whereabouts have not been widely known. A closer examination of the copy yielded new clues on provenance as well as rich bibliographic and textual information, useful in itself and as the basis for comparison with other copies.

Secondly, the copy held by the Bayerische Staatsbibliothek in Munich, Cod. Sin. 1, [Guangzhou] 1830, was discussed. This copy, referred to here initially as the ‘German Copy’, has again been rarely noticed and, most recently, misidentified as another copy made by Julien. Through a closer examination of historical documents, textual evidence and handwriting styles of Julien Copy A, the German Copy and several other manuscripts in the library in Munich, it is confirmed that the German Copy was not by Julien but is, rather, in the hand of the German Orientalist Karl Friedrich Neumann. This copy of the Notitia was also based on the copy by Abel-Rémusat, and was most probably completed between 1828 and 1829 in Paris.

It is interesting to note that while all original manuscripts of the Notitia travelled from East to West—sent by the author or his friends from Canton/Macau back to Paris in the first part of the eighteenth century—the two French copies examined in this paper both travelled from West to East in the first part of the nineteenth century (although they also returned to and remained in the West sooner or later). Julien Copy A was the important bridge between the first and successive waves of the book’s circulation. The German Copy, or ‘Neumann Copy’, is a key example of the book’s first wave of circulation in manuscript in the early nineteenth century. The former was used primarily as an exemplar for printing and the latter was used solely as a study copy by its learned owner, and both are, in themselves, manuscript copies of historical interest. Furthermore, as they do contain some unique features, comparisons between Julien Copy A and the Neumann Copy yield some insights on, and better understanding of, their common master, the Abel-Rémusat Copy, which is, so far as we know, sadly lost.

The author would like to express his sincere gratitude to the two anonymous reviewers for their careful reading of this paper and their insightful comments, which have helped to improve significantly both the quality and presentation of the paper.

Footnotes