場所、影像、詞書編纂：金尼閣的教育背景如何影響《西儒耳目資》

全文

(1)國立臺灣師範大學國際與僑教學院國際漢學研究所碩士論文 Graduate Institute of International Sinology College of International Studies and Education for Oversea Chinese. National Taiwan Normal University Master Thesis. 場所、影像、詞書編纂：金尼閣的教育背景如何影響《西儒耳目資》 Loci, Image, and Lexicography: How Trigault's Learning Influenced the Siju Ulmoçu. 簡宏逸 Hung-yi Chien. 指導教授： Dr. Henning Klöter 潘鳳娟博士. 中華民國 100 年 2 月 Febrary 2011.

(2)

(3) 國立臺灣師範大學學位論文授權書本授權書所授權之論文為授權人在國立臺灣師範大學國際與僑教學院國際漢學研究所九十九學年度第一學期取得碩士學位之論文。論文題目：場所、影像、詞書編纂：金尼閣的教育背景如何影響《西儒耳目資》. 指導教授：Dr. Henning Klöter、潘鳳娟博士授權事項：一、授權人同意非專屬無償授權本校無償授權本校將上列論文全文資料以微縮、光碟、數位無償授權本校化或其他方式進行重製作為典藏之用。本校在上述範圍內得再授權第三人進行重製。二、授權人. 同意. 非專屬無償授權本校及國家圖書館無償授權本校及國家圖書館將前條典藏之資料無償授權本校及國家圖書館 □不同意收錄於資料庫，並以電子形式透過單機、網際網路、無線網路或其他傳輸. 方式，提供讀者基於個人非營利性質之線上檢索、瀏覽、下載、傳輸、列印等利用。本校得將上述權利再授權于第三者。. 三、論文全文電子檔上載網路公開時間：【第二項勾選同意者，以下須擇一勾選】 □ 即時公開. 自 102 年 1 月 1 日始公開授權人姓名：學. （請親筆正楷簽名）. 號： 696210275. 註：1. 本授權書須列印並簽署兩份，一份裝訂於紙本論文書名頁，一份繳至圖書館辦理離校手續 2. 授權事項未勾選者，分別視同「同意」與「即時公開」中. 華. 民. 國. 1 0 0. 年. 2. 月. 1 7. 日.

(4)

(5)

(6)

(7) ABSTRACT Placing the Ulmoçu back to its cultural-historical context, this thesis investigates how Trigault’s educational background offers the framework to deal with the Chinese script. This study finds that the art of memory, which Trigault had studied in school like other Jesuits, provided a framework of the lexicography. Trigault is clearly influenced by the Combinatory Art, which became the phonetic wheels and the tabulations of Chinese syllables that Trigault devised for language analysis. Trigault is also influenced by alphabetization, the method to arrange words in its alphabetical order. For Trigault, these intellectual heritages were his toolkit in tackling the exotic Chinese script. Moreover, I also justify Trigault and his contemporary Jesuits’ belief about Chinese characters, namely, the ideographic and universal myth. Pre-occupied with this idea, Trigault and the early modern European travelers’ observed the international intelligibility of the Chinese script and develop the idea that “the Chinese hieroglyphs” are also ideographic. In Trigault’s lexicography, this idea permitted him to place Chinese characters in the places denoted by romanized syllables. Finally, the approach taken in this thesis explores a new horizon in the study on the Ulmoçu and underscores how a missionary’s intellectual background contributes to his analysis of foreign language.. Keywords: Xiru Ermuzi, Nicolas Trigault, S.J., Jesuit Education, the art of memory, the ideographic myth, the phonetic wheels..

(8)

(9) 摘要. 本研究根據《西儒耳目資》出版的歷史文化背景，重新思考編者金尼閣（Nicolas Trigault, 1577-1628）的教育背景如何形成這本字典的架構，並對傳教士之教育背景對語言分析的影響提出新的見解。本研究發現金尼閣和其他耶穌會傳教士在養成過程中所接觸的記憶術和字典編纂有相同的功能，可以提供金尼閣編纂《西儒耳目資》的基本架構。文藝復興時期所流行的結合術（the Combinatory Art）則提供金尼閣分析漢語音系的架構。接著金尼閣將歐洲既有的字母排列法（alphabetization）修改成獨特的漢語拼音排列法。除此之外，金尼閣和他同時期的歐洲人一樣，都把漢字是為表意圖像，並且將漢字聯想成東方的埃及象形文字。這樣的想法讓金尼閣把漢字放入由羅馬字母組合、標記而成的空間中，成為《西儒耳目資》成型過程中的關鍵要素之一。. 關鍵詞：西儒耳目資、金尼閣、耶穌會教育、記憶術、象形文字迷思、音韻活圖。.

(10)

(11) TABLE OF CONTENTS. LIST OF FIGURES ................................................................................................................. iii LIST OF TABLES ....................................................................................................................iv LIST OF ABBREVIATIONS ....................................................................................................v CHAPTER 1 BACKGROUND AND NEW QUESTIONS ......................................................1 1.1 Nicolas Trigault: Missionary and Linguist ..................................................................1 1.2 The Art of Memory ......................................................................................................4 1.3 Siju Ulmoçu in the Sino-European Cultural Encounter ............................................. 11 1.4 The New Questions....................................................................................................13 CHAPTER 2 TRIGAULT’S INTELLECTUAL BACKGROUND AND ITS INFLUENCE.17 2.1 The art of memory in the Jesuit education.................................................................18 2.2 Images and Loci .........................................................................................................23 2.3 Manipulating the Alphabets.......................................................................................28 2.4 The Combinatory Art.................................................................................................31 2.5 The Transformation of the Art of Memory................................................................36 CHAPTER 3 CHINESE AS AN IDEOGRAPHIC SCRIPT IN TRIGAULT’S EYES ...........45 3.1 Trigault’s Understanding of Chinese Characters.......................................................46 3.2 A Mentalité about Chinese Characters in Sinosphere................................................56 3.3 The European Imagination of Ideographic Scripts ....................................................61 i.

(12) 3.4 Allegoric Interpretations of Egyptian and Chinese....................................................63 CHAPTER 4 THE LEXICOGRAPHY OF THE SIJU ULMOÇU ..........................................71 4.1 Romanization: The Crucial Component of the Siju Ulmoçu .....................................72 4.2 From the Combinatory Art to Language Description ................................................75 4.3 Combinations and Tabulations of Syllables ..............................................................80 4.4 Trigault’s Alphabetical Order in Liè Yīnyùn Pǔ .......................................................85 4.5 Liè Biānzhèng Pǔ, An Index to Local Dictionaries ...................................................93 CHAPTER 5 CONCLUSION..................................................................................................97 BIBLIOGRAPHY..................................................................................................................101 APPENDIX 1 TRIGAULT’S NATIONALITY AND NATIVE LANGUAGE.....................107 APPENDIX 2 THE TABLE OF CONTENTS OF THE SIJU ULMOÇU ............................ 111. ii.

(13) LIST OF FIGURES. Figure 2.1 Instances to memorize numbers with corresponding images in Romberch’s Congestorium Artificiose Memorie.................................................................................27 Figure 2.2 A Lullian Wheel.....................................................................................................32 Figure 2.3 A portion of the combination chart from Figure 2.2. Source: gallica.bnf.fr / Bibliothèque municipale de Lyon Part-Dieu...................................................................33 Figure 2.4 The diagram of Publicius for the memory of Latin syllables.. ..............................40 Figure 2.5 Fünffacher Denckring der Teutschen Sprache.. ....................................................42 Figure 3.1 The List of Appendices of Piānhǎi Lèibiān 篇海類編. .......................................54 Figure 4.1 The Universal Phonetic Wheel ..............................................................................77 Figure 4.2 The Sinitic Phonetic Wheel ...................................................................................79 Figure 4.3 The General Table of Phonetic Coordinate ...........................................................83 Figure 4.4 The Complete Table of Phonetic Coordinate.........................................................83 Figure 4.5 Trigault’s rearrangement of the Roman alphabet ..................................................86 Figure 4.6 A Comparison of the alphabets in Yuǎnxi Qíqì Túshuō 遠西奇器圖說 (left) and the Siju Ulmoçu 西儒耳目資 (right).............................................................................88 Figure 4.7 The List of Finals in Liè Yīnyùn Pǔ Yùnmǔ Mùlù 列音韻譜目錄 (Partial). .......90 Figure 4.8 The 16th leaf of the 15th volume of Hóngwǔ Zhèngyùn 洪武正韻.....................95. iii.

(14) LIST OF TABLES Table 3.1 The place where jīn 金 “gold” situates in dictionaries arranged by thematic and phonetic schema..............................................................................................................51 Table 4.1 The letters of a concentric wheel on the Universal Phonetic Wheel.......................77 Table 4.2 The result of Trigault’s alphabetization of finals ....................................................91. iv.

(15) LIST OF ABBREVIATIONS. HWZY. Hóngwǔ Zhèng Yùn 洪武正韻 (The Standard Rimes of Emperor Hongwu). WYPH. Wǔyīn Piānhǎi 五音篇海. YHXB. Gǔjīn Yùnhuì Jǔyào Xiǎobǔ 古今韻會舉要小補 (A Minor Supplement to the Extracts of the Ancient and Contemporary Rime Conglomerate). ZYHP. Zhèngyùn Hǎipiān 正韻海篇. v.

(16) i.

(17) CHAPTER 1 BACKGROUND AND NEW QUESTIONS. 1.1Nicolas Trigault: Missionary and Linguist In the history of the Sino-European cultural encounter, the French-speaking Flemish Jesuit Nicolas Trigault (Jīn Nígé 金尼閣 1577–1628)1 played an important role in the early seventeenth century. Arriving at Macau in 1610, Trigault, a former teacher of humanities from Flanders, had already preached for three years in India. He traveled between Nanjing and Beijing for the business of the Society of Jesus and learned the Chinese language from Alfonso Vagnoni (Gāo Yīzhì or Wáng Fēngsù 高一志/王豐肅 1566–1640) and Lazzaro Cattaneo (Guō Jūjìng 郭居靜 1560–1640). The latter became Trigault’s confessor.2 In 1612, Trigault returned to Europe under Nicholas Longobardi’s (Lóng Huámín 龍華民 1559–1654) order to request the establishment of a new province in China and the Pope’s permission of the accommodation policies for Chinese converts.3 During his voyage to Rome, he translated and edited Matteo Ricci’s (Lì Mǎdòu 利瑪竇 1552–1610) journal and letters from Italian to Latin, and the product titled De Christiana Expeditione apud Sinas was later published in Augsburg in 1615. It became the firsthand account of China in early seventeenth century Europe. 1. Trigault’s nationality and native language are discussed in Appendix 1.. 2. Anne-Marie S. Logan and Liam M. Brockey, “Nicolas Trigault, SJ: A Portrait by Peter Paul Rubens,” Metropolitan Museum Journal 38(2003): 157–167. 3. Trigault, as an agent sent by the Jesuit China Mission, requested granting permission to wearing hats in Mass and using Chinese in liturgy. The dressing code accommodated the Chinese custom that bare head is disrespectful and was eventually permitted by the Holy See, whereas the petition of using Chinese in liturgy has never been approved. The history of Jesuit’s accommodation policy and its implication in the Jesuit Mission in China has been discussed in George Harold Dunne, Generation of Giants: The Story of the Jesuits in China in the Last Decades of the Ming Dynasty (Notre Dame, Indiana: University of Notre Dame Press, 1962), 162–182. 1.

(18) Trigault’s procurer journey was a round trip. Between 1615 and 1618, he recruited new members, collected books, and raised funds for the Jesuit Mission in China. When Trigault reached Macau again in 1620, he brought not only 7,000 titles (the actual number is in doubt), but also several learned missionaries who later played significant roles in the Sino-European cultural encounter. Among these new missionaries, Johann Terrenz Schreck (Dèng Yùhán 鄧玉函; 1576–1630) specialized in physics and medicine, Johann Adam Schall von Bell (Tāng Ruòwàng 湯若望; 1592–1666) in astronomy, and Giacomo Rho (Luó Yǎgǔ 羅雅谷; 1592–1638) in mathematics.4 In contrast, Trigault was not an expert in any scientific discipline, but he had excellent achievement in literature and language studies. During Trigault’s second term in China, he compiled a dictionary which attracted modern linguists’ and historians’ attention, the Siju Ulmoçu 西儒耳目資.5 The Siju Ulmoçu is a significant milestone in the histories of Chinese lexicography, missionary linguistics, and Sino-European cultural encounters. It is a product under the convergence of European and Chinese intellectual traditions. The chief compiler,. 費賴之馮承鈞. 4. For Trigault’s general biography, see Louis Pfister , Zàihuá Yēsūhuìshì Lièzhuàn Jí Shūmù , trans. Chéngjū Féng (Beijing: Zhōnghuá Shūjú , 1995), 115–125. For his mission and journey in Europe, see Luke Clossey, Salvation and Globalization in the Early Jesuit Missions (New York: Cambridge University Press, 2008). Standaert discussed the Renaissance culture brought by Trigault and his fellow Jesuit. See Nicolas Standaert, “The Transmission of Renaissance Culture in Seventeenth-Century China,” Renaissance Studies 17, no. 3(2003). The number of the books that Trigault brought to China is in doubt. Fang Hao defends the number of seven thousand titles. Hao Fang , “Běitáng Túshūguǎn Cángshū Zhì ,” in Fāng Háo Liùshí [The Collected Works of Maurus Fang Hao Revised and Edited by the Zìdìnggǎo Author on His Sixtieth Birthday] (Taipei City: Xuéshēng Shūjú, 1969).. 在華耶穌會士列傳及書目方豪方豪六十自定稿. 中華書局. 北堂圖書館藏書志. 5. The title romanized in Pinyin is “Xirú Ěrmùzī,” usually abbreviated in XREMZ in the literature. However, since this book has its own romanization system, I would adopt Trigault’s system for the title as Siju Ulmoçu, but diacritics are removed for typographic clearness. The abbreviation is Ulmoçu, which was used in its time. This thesis refers to the three-volume facsimile reprint in Taipei in 1977, which is based on an older facsimile copy made by Peking University and the National Peiping Library (Taipei: Tiānyī (today’s National Library of China) in 1933. Nicolas Trigault, Xirú Ěrmùzī Chūbǎnshè , 1977[1626]).For the table of contents of the Ulmoçu, see Appendix 2.. 西儒耳目資. 天一出版社. 2.

(19) Trigault and the cooperators, Wáng Zhēng 王徵 (1571–1644) and Hán Yún 韓雲 (?-?), were fully aware of this significant status and hoped for a good response from the Chinese audience. However, following its publication in 1626, the Ulmoçu did not achieve the popularity that its compliers and sponsors expected, and its reviews in the following decades were mixed. After the seventeenth century, no traditional Chinese phonologist mentioned the Ulmoçu again. The importance of the Ulmoçu was not discovered until Luó Chángpéi’s 羅常培 in-depth introduction to modern linguists in 1930. Luó compared Ricci’s and Trigault’s romanization systems in order to trace the historical development of missionary romanization systems. With respect to the terminology of the Ulmoçu, Luó found the European origins of some linguistic terms and the romanization system itself, which served his phonetic reconstruction of the Chinese transcribed in the Ulmoçu. Finally, Luó discovered Chinese intellectuals’ negative and positive responses to the Ulmoçu to examine its influence in the field of traditional Chinese phonology.6 This thesis intends not to re-examine the romanization system or the historical phonology of the Ulmoçu, which had been well studied since Luó’s thesis. Rather, I would consider the Ulmoçu within its socio-cultural context, that direct cultural encounters between Europe and Asia were enabled and intensified by European navigators’ trade-driven expeditions and missionaries’ religion-driven enthusiasms. This alternative scope focuses on some less-addressed issues about the Ulmoçu, such as Trigault’s intellectual heritages. Considering this context of cultural encounter, the influence from European intellectual tradition naturally becomes a major topic for further investigation. In fact, the lexicography of the Ulmoçu is a less addressed field in Chángpéi Luó 羅常培, “Yēsūhuìshì Zài Yīnyǜnxüé Shàng De Gòngxiàn 耶穌會士在音韻學上的貢獻 [The Jesuit contributions to Chinese Phonology],” 1, no. 3(1930). 6. 3.

(20) previous literature. As an attempt to integrate the Chinese and the European elements to create a unique lexicography, the Ulmoçu is not a translated version of any European original or model. Rather, it looks like its contemporary Chinese rime books, consisting of a theoretical section, several tables, lists of homophonous Chinese characters, and an index organized according to radicals of Chinese characters. However, some European elements, such as the phonetic wheels placed at the very beginning, the alphabetized list of characters, and the syllabary tabulation without “grade děng 等” divisions, are obviously non-Chinese. Thus, the Ulmoçu represents a mixed intellectual background, and this background must be understood through the analysis of the Renaissance culture and that contemporary knowledge to the Chinese language and script.. 1.2 The Art of Memory Recent studies on the Ulmoçu consider the influence from European intellectual traditions on Trigault’s lexicography. The studies of Wang, Tan, and Chien, all argue that Trigault’s phonetic wheels are influenced by the concentric wheels and the combinatory art in Lullism.7 Named after the Catalan Franciscan friar and inventor Ramon Lull (ca. 1232–1315),8 Lullism turns concentric wheels to generate the combinations from the seven God’s Dignities, designated by letters on each wheel, and uses a set of rules to interpret each combination. This mechanism was called ars 7. 王松木. 明代等韻家之反第十一屆國際暨第二十七屆全國聲《西儒耳目資》源流辨金尼閣的音有鳳初鳴年刊. Sung-mu Wang , “Míngdài Děngyùnjiā Zhī Fǎnqiè Gǎiliáng Fāngàn [Improvements of Fanqie Proposed by Phonologists in Ming Dynasty],” in Dì Shíyī Jiè Guójì Jì Dì Èrshíqī Jiè Quánguó Shēngyùnxué Xuéshù Yántǎohuì [The 11th International and the 27th National Conference of Chinese Phonology] (Fu Jen Catholic University2009); Huìyǐng Tán , “Xirú Èrmùzī Yuánliú Biànxī [On the Formation and Effect of An Audio-visual Aid to Western Scholars]” (Beijing: Foreign Language Teaching and Research Press, 2008); Hung-yi Chien, “Jīn Nígé De Yīnyùn Huótú [Trigault's Phonetic Wheels],” Yǒu Fèng Chū Míng Niánkān [Annual of Graduate School of Chinese Literature Soochow University] 5(2009).. 切改良方案韻學學術研討會析韻活圖. 譚慧穎. 8. Lull’s name has various spellings. For example, Ramon Llull, Raymond Lully, Raymond Lull, etc. Ramon Lull is the spelling used by Frances A. Yates. 4.

(21) combinatoria, or the Combinatory Art. The Lullian wheel resembled a logic machine, but it should rather be considered as a form of mnemotechnique, which helps the practitioners to memorize religious arguments.9 There is no evidence indicating that Trigault mastered Lullism or its Renaissance variations, but the design of the phonetic wheels suggests that there is a certain relation between Trigault’s lexicography and the Lullian art of memory in the European intellectual tradition.. The art of memory, or mnemonics, is a skill of memorization and recollection. It belongs to the discipline of rhetoric, one of the seven liberal arts in the Middle Ages.10 Rhetoric and Latin grammar were two essential parts of Jesuit missionary education. Latin was the lingua franca in the Church and rhetoric was the skill of preaching. However, in the greater context of the European intellectual tradition, both rhetoric and Latin grammar have more important implications for education prior to the Enlightenment. In short, Latin grammar was the key with which to access the seven liberal arts, and rhetoric was one of the three language arts, the trivium, among the seven liberal arts.11 Moreover, rhetoric is a synthetic discipline that requires not only oration but also erudition, which serves as the basis of an orator’s argumentation. Therefore, even though the influence from rhetoric is not as obvious as from Latin grammar, we should not under-evaluate the influence of rhetoric on missionary’s language studies.. The most famous mnemonic treatise written by China-based Jesuits is Ricci’s 9. Frances Amelia Yates, The Art of Memory (Chicago: University of Chicago Press, 1966).. 10. Vivien Law, The History of Linguistics in Europe from Plato to 1600 (Cambridge; New York: Cambridge University Press, 2003), 101. 11. Ibid. 5.

(22) Xīguó Jìfǎ 西國記法 (The Mnemonics of the Western Countries).12 Ricci presented the essential principles of mnemonics and raised many examples suitable for Chinese in this treatise, but what made this treatise legendary in the history of Chinese Christianity was that it helped Ricci in winning the local governor’s endorsement and in penetrating the cultural barrier of Chinese literati.13 However, it is also worth to consider the assistance of mnemonics in Ricci’s Chinese learning. In chapters four and six, Ricci analyzes the radicals of each Chinese character and interprets this analysis with his mnemonic image, revealing his endeavor in grammatology, the study of writing systems. Nevertheless, Ricci’s analytic approach in this mnemonic treatise also reflects the relation between Jesuits’ Chinese learning and their rhetoric training.. Our understanding of the rhetoric curricula of Jesuit education comes from two major sources. The first is Allan P. Farrell’s 1970 English translation of the Ratio Studiorum with an introduction and annotations. The second source is the designated textbook of rhetoric in the Ratio Studiorum, Cyprian Soarez’s De Arte Rhetorica. According to the Ratio Studiorum, we know that the Ratio Studiorum assigned Cicero’s oratorical works and Aristotle’s Rhetoric and Poetics to the daily lesson and prescribed that Cicero was the model of oration14. However, the work of Quintilian (ca. 35–ca. 100), a Roman rhetorician in the first century, was also possible to supplement the. 12. 西國記法. 利瑪竇中文著譯集. Matteo Ricci, “Xīguó Jìfǎ ,” in Lì Mǎdòu Zhōngwén Zheyì Jí , ed. Zhū Wéizhēng (Hong Kong: City University of Hong Kong Press, 2001[1595]), 181–214. 13. 朱維錚. 利瑪竇書信集. See Ricci’s letter to Father Claudio Acquaviva, the Superior General of the Society of Jesus. Matteo Ricci, Lì Mǎdòu Shūxìn Jí [Letters of Matteo Ricci], vol. 3 (Taipei: Kuangchi & Fu Jen Catholic University Press 1986), 230. 14. Jesuits, The Jesuit Ratio Studiorum of 1599, trans. Allan P. Farrell (Washington, D.C.: Conference of Major Superiors of Jesuits, 1970), 72–75. 6.

(23) rhetoric texts.15 In fact, most of the text about memory in Soarez’s De Arte Rhetorica comes from Quintilian’s Institutio Oratoria.16 I will discuss the details in Chapter 2.. For the investigation of Trigault’s intellectual background, I will first review the mnemonic literatures that he must have pursued in his rhetoric classes, and then the modern studies of the art of memory. The art of memory was incorporated in the classical rhetoric training. In On the Making of an Orator, Cicero told the story of Simonides of Ceos, who was regarded as the discoverer of the art of memory. Ricci cited the same story in Xīguó Jìfǎ,17 indicating the influence of Cicero’s oratorical works. In addition to Cicero, Aristotle’s and Quintilian’s rhetoric works are also considered to be the guides of the art of memory.18 However, the Latin text that is the most important to the art of memory is Rhetorica ad Herennium, which introduces the basic principles of mnemonics19 and lays out the five faculties required of an orator, including inventio (invention), dispositio (arrangement), elocutio (style), memoria (memory), and pronuntiatio (delivery). Although Rhetorica ad Herennium was mistakenly attributed to Cicero in the past, this mistake does not affect its status in the. 15. Robert Schwickerath, Jesuit Education; Its History and Principles Viewed in the Light of Modern Educational Problems (St. Louis: B. Herder, 1903), 120. 16. Quintilian, The Institutio Oratoria of Quintilian, trans. H. E. Butler (London: William Heinemann Ltd., 1968[1922]). See also Flynn’s note in Lawrence J. Flynn, “The De Arte Rhetorica (1568) by Cyprian Soarez: A Translation with Introduction and Notes” (University of Florida, 1955), 417. 17. Ricci, “Xīguó Jìfǎ. 西國記法,” 184.. 18. Quintilian’s advise is cited and discussed in Yates, The Art of Memory, 37–42. For the discussion about Aristotle’s mnemonics, see Yates, The Art of Memory, 51–55. 19. Marcus Tullius Cicero, “Rhetorica ad Herennium,” (London: William Heinemann, 1954),. 205–225. 7.

(24) discipline of classical rhetoric because of its clarity in mnemonic principles.20. I should clarify several closely related terms about the art of memory which are used in modern literature before discussing the art of memory in Chapter 2. First of all, memoria, the term from Rhetorica ad Herennium, specifically refers to the faculty in classical rhetoric. The author of Rhetorica ad Herennium distinguishes two kinds of memoria, the natural and the artificial.21 The artificial memory refers to the discipline that helps memoria in general, which might also be called either “mnemonics” or “the art of memory.” These terms are used interchangeably in this thesis, for “mnemonics” has an adjective form (mnemonic) to modify a noun (e.g., mnemonic system). However, when we discuss the artificial memory from which the Renaissance and Enlightenment philosophers derived their new methods, I restrict the usage to “the art of memory.” Thus, “the art of memory” has a broader sense than has “mnemonics.” Finally, individual techniques of mnemonics are specified by the term “mnemotechnique.”22. Academic study on the art of memory in English emerged quite late. Although we have few English titles on general mnemonics from the nineteenth century, these are either aids to learn foreign languages or a long bibliography of mnemonic masters’ works.23 Frances Yates’ The Art of Memory, which is now the classic in this subject, is. 20. Yates, The Art of Memory, 43.. 21. Cicero, “Rhetorica ad Herennium,” 207.. 22. The usage referred here is Carruthers’.. 23. Not to exhaust the bibliography, I have found the following books teaching how to improve artificial memory: Gregor von Feinaigle and John Millard, The New Art of Memory: Founded upon the Principles Taught by M. Gregor von Feinaigle (London: Printed for Sherwood, Neely and Jones, 1812); Richard Grey and Solomon Lowe, Dr. R. Grey’s Memoria Technica (Oxford: Printed for J. Vincent, 1838); A. E. Middleton, Memory Systems New and Old (New York: Fellows & Co., 1888); Edward Pick, On Memory and the Rational Means of Improving It (London: Trübner & Co., 1861). 8.

(25) the first specific title to comprehensively address the history of mnemonics. Yates frequently referenced Paolo Rossi’s Clavis Universalis,24 which was translated into English in 1992. Before Mary Carruthers’ research was published in 1990,25 Yates and Rossi were the sources of modern studies on the art of memory.. Yates’, Rossi’s, and Carruthers’ works are all important for my approach of the art of memory, but they contribute to different aspects. Yates’ and Rossi’s primary objective is to determine how the art of memory influenced philosophers during the Renaissance and Enlightenment in developing new methods with which to understand the world. Rossi analyzes the thoughts of Francis Bacon (1561–1626) and René Descartes (1596–1650), and Yates focuses on two Neoplatonists, Giordano Bruno (1548–1600) and Robert Fludd (1574–1637). To review the cultural context of the philosophers in question, it must also be noted that they studied the influential Lullism in the history of mnemonics and provided two comprehensive modern accounts of Lullism and the Combinatory Art. The Combinatory Art, which was introduced to Europeans by Lull, greatly influenced the Renaissance and the early modern philosophies, including the Universal Language proposals which pursued a real language that would be intelligible to all people speaking different languages. Carruthers, on the other hand, studied the medieval teaching and practice of mnemonics to unveil that the art of memory was a medieval mentalité that adapted to its times through history, influencing various ideas through the history.26 Moreover, Carruthers. 24. Paolo Rossi, Logic and the Art of Memory: The Quest for a Universal Language, trans. Stephen Clucas (London: Continuum, 2006). 25. Mary J. Carruthers, The Book of Memory: A Study of Memory in Medieval Culture (Cambridge [u.a.: Cambridge Univ. Press, 2008). 26. Ibid., 153. Mentalité is a notion introduced by French historiography. The study of mentalité emphasizes the way of thinking common to the people of a particular time. Mentalité is also a link 9.

(26) also refers to neuropsychology in her discussion of the psychological foundation of the basic mnemonic principles which were passed down from Ancient Greece. My present objective is similar to Yates’ and Rossi’s studies, which emphasize the derivational aspect of the art of memory; thus, I will refer to their studies on both informational and methodological aspects. Carruthers’ work is also informative with respect to our purposes. In addition, we can consider her insight in the medieval mentalité to study the history of cultural encounter.. Besides the major studies mentioned above, two additional works about Lullism merit mention. In his study of the Universal Language, Umberto Eco discusses Lull’s Combinatory Art in the Ars magna and the non-Christian heritage of Lullism. Eco also reviews the operation of Lull’s concentric wheels, which influenced Trigault’s phonetic wheel. Regarding primary literature on Lullism, I shall rely on Anthony Bonner’s English translation of Lull’s major works.. Finally, alphabetization, which Trigault used to arrange the romanizations of Chinese syllables, was also a mnemotechnique in the Middle Ages.27 Note that alphabetization should not be confused with romanization or literacy campaign. It is a method to alphabetically arrange things by their names. Alphabetization is only applicable to items written in alphabets. To analyze this issue, we shall refer to Lloyd W. Daly’s Contributions to a History of Alphabetization in Antiquity and the Middle Ages.28. between history and other human sciences. In the case of memory, Carruthers borrows psychology to study the art of memory, demonstrating interdisciplinary feature of mentalité study. For the definition and the development of mentalité in history research, especially in the French Annales School, see Jacques Le Goff, “Mentalities: A History of Ambiguities,” in Constructing the Past: Essays in Historical Methodology, ed. Jacques Le Goff and Pierre Nora (Cambridge: Cambridge University Press, 1985[1974]), 166–180. 27. Carruthers, The Book of Memory, 129.. 28. Lloyd W. Daly, Contributions to a History of Alphabetization in Antiquity and the Middle 10.

(27) Daly mentions various methods for the alphabetization of lexical items. Many of them are different from the full-alphabetization that is used in contemporary European dictionaries. In fact, as will be addressed in Chapter 6, Trigault’s alphabetization is a unique alphabetization, which orders the romanizations by not only roman letters but also their tones and the number of letters in a syllable.. 1.3 Siju Ulmoçu in the Sino-European Cultural Encounter To Trigault and his contemporary Europeans, the Chinese script is an odd writing system that is composed of thousands of characters standing for distinctive meanings and mutually intelligible across different dialects. This cross-linguistic intellectualability fascinated the European scholars in the sixteenth and seventeenth century, and they eventually developed the idea that the Chinese script has a nature distinctive from European scripts, but similar to the mythical hieroglyphs of ancient Egypt. Trigault was an important figure in the development of these ideas, and they were held by the public for centuries and became a conventional wisdom to this script. The modern linguist, John DeFrancis, conceptualized and criticized these ideas as several “myths” about Chinese in Chinese Language: Fact and Fantasy.29 Among the myths that DeFrancis criticized, “the ideographic myth” and “the universality myth” are directly related to Trigault.30 Since the Ulmoçu was a dictionary dealing with Chinese characters, Trigault’s understanding could influence its compilation. Then, I have to mention the sources which could become Trigault’s references.. Ages (Bruxelles: Latomus, 1967). 29. John DeFrancis, The Chinese Language: Fact and Fantasy (Honolulu: University of Hawaii Press, 1984). 30. Ibid., 134. 11.

(28) Trigault mentioned four Chinese language references in the Ulmoçu and complained about the difficulty of using them. Two out of these four dictionaries are the popular dictionaries, which had been commonly circulated in the Chinese publication market in the sixteenth and seventeenth century. These popular dictionaries feature a gigantic collection of characters, usually up to sixty thousand characters or more, and they usually use hǎi 海 ‘sea; ocean’ in their titles to highlight this feature. In fact, not only Trigault mentioned these dictionaries; many Jesuits who visited China in the seventeenth century reported this kind of dictionary in their travelogues or treatises about China.31. Although these popular dictionaries primarily served as a language reference, their appendices usually offered informative knowledge about the Chinese language and script, and it was eventually adopted by Jesuit missionaries in their reports. For example, an unpublished manuscript written by Prosper Intorcetta (Yīn Duózé 殷鐸澤 1626–1696) and other Jesuits simply adopted the common theory to the Chinese script in these popular dictionaries.32 This manuscript offers an instance about how the Chinese tradition influenced the Jesuits on their understanding to this odd script. We will cite this manuscript frequently in the following chapters.. Besides the direct adoption, Trigault’s cultural background and the preexisting knowledge in his mind also influenced his understanding of the Chinese script. In the. 渡辺宏, “Rai Mei Senkyōshi Shiyō No Kanjisho Kaihen-ni Tsuite 来明宣教師使用の漢字書『海篇』について,” Tōyōbunko-sho-hō 東洋文庫書報 28(1997): 32–72. 31. Hiroshi Watanabe. 32. Knud Lundbæk, The Traditional History of the Chinese Script: From a Seventeenth Century Jesuit Manuscript (Århus, Denmark: Aarhus University Press, 1988). The cooperating Jesuits are Couplet and Rougemont, see Noël Golvers, "The Development of the “Confucius Sinarum philosophus' Reconsidered in the Light of New Material,” in Western Learning and Christianity in China: The Contribution and Impact of Johann Adam Schall von Bell, S.J. (1592–1666), ed. Roman Malek (Sankt Augustin: China-Zentrum & the Monumenta Serica Institute, 1998). 12.

(29) publication that made Trigault and Ricci become renowned around Europe, De Christiana Expeditione apud Sinas, Trigault introduced the Chinese script and compared it with the Egyptian hieroglyphs. By doing so, he incorporated an exotic writing system into the existing knowledge category of European intellectuals, who had already learned about Egyptian hieroglyphs from Horapollo’s Greek manuscript of the early fifteenth century.33 However, Horapollo discussed the hieroglyphs without citing concrete examples, which led to a misleading image of the hieroglyphs. Before Jean-François Champollion’s successful decipherment, early Egyptologists had believed for centuries that the hieroglyphs were ideographic, conveying meaning without interference by languages. According to DeFrancis, Trigault’s comparison, multiplied by the early-modern misconception of the Egyptian hieroglyphs, produced the myths about the Chinese script, and the popularity of Ricci-Trigault’s book amplified this influence.34 The above review shows that Trigault was a key-figure in disseminating the myths about the Chinese script in Europe, but it does not mean that Trigault is the first to establish the myth. In fact, we will find that the myths about Chinese had already circulated in Europe since the late sixteenth century.. 1.4 New Questions While Trigault brought the latest account about China to Europe, in China, his role became the transmitter of European culture. In the Ulmoçu, Trigault employed the Roman alphabet to represent and analyze Chinese phonology and introduced the terminology of European linguistics to his Chinese readers. Since Luó Chángpéi’s 33. Horapollo, The Hieroglyphics of Horapollo Nilous, trans. Alexander T. Cory (London: Pickering, 1840); Umberto Eco, Serendipities: Language and Lunacy, trans. William Weaver (New York: Harcourt Brace & Company, 1998). 34. DeFrancis, The Chinese Language: Fact and Fantasy, 134. 13.

(30) groundbreaking thesis, the etymology of Trigault’s terminology has been the mainstream of studying the European influence brought by the Ulmoçu. To mention the major works only, Chen Liang-chi’s 陳良吉 textual criticism significantly deepens our understanding of the etymology of Trigault’s terminology and its relation to similar terms in traditional Chinese phonology.35 Wang Sung-mu 王松木 focuses on historical linguistics and reconstructs the initials, finals, and tonal categories in the Ulmoçu. He also analyzes the cases of multiple pronunciations of particular characters, identifying the diachronic status of the Ulmoçu in the history of Chinese language36. Recently, Tán Huìyǐng 譚慧穎 focused on the correspondence between signs and sounds in the Ulmoçu, and she has compared Trigault’s romanization system with four orthographies of Romance languages as well as several Mandarin transcription systems preceding and following the Ulmoçu.37. Each of these treatises improve and expand our understanding of the Chinese described in the Ulmoçu since Luó’s initiative work in 1930, but they more or less repeat each other on certain issues, and they are also biased on the metalanguage of language description but neglect the influence of Trigault’s background on his lexicography. Therefore, even though we have understood Trigault’s metalanguage, we cannot explain how the diagrams, tables, and indices function in the Ulmoçu, and the sources of these devices are still left unaddressed.. 陳良吉. 35. Liang-chi Chen , “Eine funktionell-strukturelle und historisch-vergleichende Untersuchung des Xi Ru Er Mu Zi (1626, Hangzhou): eine vergleichende Studie zur traditionellen chinesischen Lexikographie” (Universität Trier, 1987). 36. 王松木. Sung-mu Wang, “Phonology of Late Ming Dynasty Mandarin Reflected in Hsi Ju Erh Mm Tzu” (National Chung Cheng University, 1994). 37. 譚慧穎. 《西儒耳目資》源流辨析” [On the Formation and. Tán , “Xirú Èrmùzī Yuánliú Biànxī Effect of An Audio-visual Aid to Western Scholars], 29–47. 14.

(31) As I have reviewed in the previous sections, the art of memory and the Renaissance conception to the Chinese script may both influenced Trigault’s lexicography, and the reviewed materials can help to solve my questions, I still need a historiographic guideline for my study. In this thesis, I will adopt the historian of linguistics Konrad Koerner’s three principles to ascertain “influence” serves as guidelines in this thesis. He argues that first, the most reliable sources are the direct citations made by the linguist in question. Although this information can only support a claim after conducting a critical comparison, direct citation is more favorable to historians than a hypothetical claim. Second, the biography of the linguist in question is helpful because we can look for probable influences from the linguist’s family and educational background. Finally, we may identify an influence from a particular linguist’s non-linguistic studies that are suggested in his or her biography.38 In the case of the Ulmoçu, I only find direct references to its contemporary Chinese dictionaries and none to any particular European treatise. Thus, I must consult Trigault’s major biography published by Dehaisnes in 1861.39 Although this biography was published more than two centuries after Trigault’s death, Dehaisnes refers to Trigault’s letters and documents, offering valuable information for researchers like Pfister, Dunne, Logan and Brockey, and Clossey. Furthermore, I also consider the classical rhetoric tradition, which Trigault learned and taught, as a significant source of influence on his lexicography. The intellectual atmosphere in the late Renaissance, especially European knowledge of non-alphabetical writing, should not be neglected.. 38. E. F. K. Koerner, “On the Problem of ‘Influence’ in Linguistic Historiography,” in Practicing Linguistic Historiography: Selected Essays (Amsterdam: John Benjamins, 1989), 40–41. 39. Chrétien César Auguste Dehaisnes, Vie du Père Nicolas Trigault de la Compagnie de Jésus (Paris: H. Casterman, 1861). 15.

(32) However, I do not exclude the possible influence from Chinese traditions. I shall critically compare both traditions in order to establish a sound argument concerning Trigault’s lexicography. With Koerner’s guidelines, the following new questions regarding the Ulmoçu will be investigated and answered in this thesis:. 1.. How did Trigault’s educational background, intellectual heritage, and mentalité influence his lexicography?. 2.. How did Trigault transfer the Lullian art of memory to Chinese lexicography?. The questions above will be address in Chapter 2.. 3.. How did Trigault’s background influence his understanding of Chinese characters, and how did this knowledge influence the Ulmoçu?. 4.. Did the Chinese philological tradition shape Trigault’s understanding of Chinese writing? If yes, how?. These questions will be addressed in Chapter 3. After the discussion over the influences, in Chapter 4, I shall answer. 5.. How these influences embodied in the Ulmoçu.. After answering these new questions, we will understand how Trigault developed his Chinese lexicography from the art of memory and his understanding to the Chinese script, and the results will explore and expand today’s understanding to the linguistic achievements of early modern missionaries.. 16.

(33) CHAPTER 2 TRIGAULT’S INTELLECTUAL BACKGROUND AND ITS INFLUENCE A recent turn in the study of the Ulmoçu is to consider the influence from Trigault’s intellectual background. The latest studies suggest that the art of memory may have influenced Trigault’s linguistic thinking,1 and his fellow missionaries’ citations and applications of the art of memory, such as Ricci’s Xīguó Jìfǎ. Thus, I would take the art of memory to understand Trigault’s intellectual background, and how this background influenced his linguistic thinking. However, the origins of Trigault’s linguistic thinking are vague. In the Ulmoçu, he simply claims that he had known the theory of language and script (yánzì zhī suǒyǐrán 言字之所以然, lit. ‘why language and script actually are what they are’) since his youth.2 Trigault probably heard Lambert Schenkel, who taught mnemonics at Douai in Trigault’s teenage years,3 but I have no evidence suggesting Trigault had learned Schenkel’s art at his hometown. In his brief autobiography extracted from l’Album novitiorum Domûs probalionis Tornacenis, he only mentioned the curricula he had completed in the Jesuit college,4 but the art of memory is not explicated here: I, Nicolas Trigault from Douai, was born on March 3rd, 1577…I studied in the college of the Society of Jesus in Douai, spending one year in the primary class, another 1. See Footnote 7 on Page 3.. 2. Ulmoçu I, 1a. 3. Schenkel’s career as a mnemonic master, see Middleton’s introduction and the genealogy compiled by his descendents living in the United States in the late nineteenth century. Middleton, Memory Systems New and Old, 15–16; Louisa Jane Shinkle Abbott and Charles L Abbott, The Shinkle Genealogy, Comprising the Descendants of Philipp Carl Schenckel, 1717–1897 (Cincinnati, OH: Press of Curts & Jennings, 1897), 7. 4. The history of the Jesuit college at Douai, which became the University of Douai, see Bernard Ward, “Douai,” in Catholic Encyclopedia, ed. Charles George Herbermann (New York: The Encyclopedia Press, 1913).. 17.

(34) one in the secondary class, two years in grammar, one year in humanities, two years in rhetoric, two years in philosophy. With these studies, I obtained the Master of Art at the end of September 1594.5 Among these curricula, rhetoric attracts our attention because memoria was one of the five faculties in the classical rhetoric, which had been incorporated in European pedagogy since the Middle Ages. As a graduate and a rhetoric teacher of the Jesuit school, Trigault must have had an experience with the art of memory as he learned and taught classical rhetoric. In fact, modern academic investigations in the art of memory suggest that this art formed the way of thinking in medieval and in Renaissance Europe, and its fundamental principles inspired new methods to knowledge organization and logical reasoning, such as the alphabetization and the Combinatory Art (see Section 1.2). Both techniques were adopted by Trigault and became the principles to organize Chinese characters in the Ulmoçu. In this chapter, I will review the history of the art of memory and its medieval derivations. I will also go beyond Trigault’s time to see how the art of memory could have inspired a new linguistic theory, focusing on the Fünffacher Denckring der Teutschen Sprache (Five-Folded Thinking Ring of the German Language) of the German poet Georg Philipp Harsdörffer (1607–1658).. 2.1 The art of memory in Jesuit education The art of memory had been a part of rhetoric since antiquity, and so was it in Jesuit education of the sixteenth century. Since the late sixteenth century, the Jesuit schools used two major texts in the rhetoric curriculum. One is the classical Rhetorica Ad Herennium, a classical rhetoric text from the Roman era, and the other text is De Arte Rhetorica (The Art of 5. “Ego Nicolaus Trigault Duacensis natus anno 1577, 3° martii...Studii in scholis societatis Jesu duaci in infima classe uno anno, in secunda semiliter, in syntaxi duobus annis, in humanitate uno, in rhetorica duobus, in philosophia duobus. Ibidem artium licentiatus factus feci in fine septemberis 1594.” Dehaisnes, Vie du Père Nicolas Trigault de la Compagnie de Jésus, 217.. 18.

(35) Rhetoric) of the Portuguese Jesuit Cyprian Soarez. The former had been an important rhetoric text taught since the Roman era, and the latter is a late Renaissance text, which had served as a bridge between Latin grammar and the classical Latin literature.6 Soarez followed the classical way to discuss the five faculties of rhetoric: invention (inventio), arrangement (dispositio), style (elocutio), memory (memoria), and delivery (pronuntiatio), but he did not intend to alter the classical doctrine. This conservative textbook had been a popular text for centuries. Originally published in 1568, it had been reprinted 134 times throughout Europe until the eighteenth century. In the Jesuit system, Soarez’s text was adopted by some Jesuit colleges since its first publication and eventually became an assigned material to the humanity class in the 1599 Ratio Studiorum.7 I am not sure what text Trigault had used as a student in his humanity class, but it was unlikely Soarez’s text, which primarily circulated in Spain and Portugal before it became an officially assigned material in the Ratio Studiorum in 1599,8 but Trigault had already completed his liberal education before it. However, since Trigault had taught humanities for nearly a decade in Flanders until 1607, he must have used Soarez’s text in his class under the regulation of the Ratio Studiorum. I may therefore start to discuss the art of memory in Trigault’s education background from Soarez’s De Arte Rhetorica. Moreover, as a teacher of rhetoric, it is hard to imagine that Trigault would stay with Soarez’s elementary text and have no experience with Rhetorica ad Herennium in his career. Therefore, I must also bring the classical Rhetorica ad Herennium into our review of Trigault’s intellectual background. Soarez discussed memory from Chapter 52 to 55 in Book Three.9 It is a relatively short 6. Flynn, “De Arte Rhetorica,” 367, 369.. 7. Jesuits, The Jesuit Ratio Studiorum of 1599, 79–84; Flynn, “De Arte Rhetorica”, 14–15, 21–41.. 8. Flynn, “De Arte Rhetorica,” 23.. 9. Ibid., 417–423. 19.

(36) treatise in comparison with other faculties of rhetoric in De Arte Rhetorica. Flynn points out that most parts of this mnemonic text are extractions from the Roman rhetorician Quintilian’s (ca.35-ca.100) memory treatise in the Institutio Oratoria. Soarez began the text with the story that Simonides, the discoverer of mnemonic in Ancient Greece, was able to identify the victims’ corpora buried under a collapsed hall according to their seats in that fatal banquet. This story points out the foundation of memory: places.10 He then elaborated on the importance of memory for an eloquent orator in the following chapter.11 In Chapter 54, Soarez turned to explain the theory of memory. The theory comprises two major elements: places and images. First, one needs to memorize a series of places in the mind. They must be expansible and clearly arranged in the proper order of the things to be memorized and should avoid alternation, making them firmly fixed in one’s mind. The places, for example, may be a large house or similar structures. Once the places are prepared, one can associate the images with the places in a reasonable order, so they may be easily recollected on demand. The images, on the other hand, must be impressive and distinguishable with the things to be memorized. When a thing is no longer needed, one may discard the associated image to preserve the memory place, which ought to be constantly used.12 Mary Carruthers called this system “the architectural mnemonics.”13 In the final chapter on memory, Soarez advised that one should carefully use the memory places to memorize important things, or they would be quickly exhausted. One may excel in this method only through exercise. Finally, he demonstrated the merits of excellent memory with several memory masters’ cases. For example, the Greek king Mithridates knew 22 languages spoken in his governed nations. 10. Chapter 52; Ibid., 417–418.. 11. Chapter 53; Ibid., 418–419.. 12. Chapter 54; Ibid., 419–421.. 13. Carruthers, The Book of Memory, 89. 20.

(37) Crassus, the Roman commander of Asia, spoke five Greek dialects. Cyrus, King of Persia, remembered every soldier’s name in his army.14 The method stated by Soarez in De Arte Rhetorica is also called “the method of loci,” which is the art of memory transmitted with rhetoric from Ancient Greece to Rome. Since rhetoric had been one of the three liberal arts (trivium) in the medieval European pedagogy, the method of utilizing loci had been deeply imprinted on every rhetoric students’ mind, becoming an important mentalité, or way of thinking, in the intellectual history of Europe.15 The Dominicans were the major promoters of this method. Because of their efforts in the late Middle Age, the art of memory had been disseminated into general European culture by the thirteenth century, especially in France and Italy.16 Therefore, it is not surprising to see European missionaries mentioning similar ideas in East Asia. Ricci’s Xīguó Jìfǎ., for example, cites many statements which are also quoted in Soarez’s text. However, our comparison between De Arte Rhetorica and Xīguó Jìfǎ shows that Ricci’s theory is much more detailed than Soarez’s. In fact, both Soarez’s and Ricci’s theories relied on the method of loci from three classical rhetoric texts: De Orator of Cicero, the Institutio Oratoria of Quintilian, and Rhetorica ad Herennium. Among these texts, Rhetorica ad Herennium had a similar status like Soarez’s text in the Jesuit rhetoric curriculum.17 Thus, if Trigault did not use Soarez’s text as a student, he probably used Rhetorica ad Herennium instead. In the history of the art of memory, and rhetoric as well, Rhetorica ad Herennium is the oldest surviving rhetoric text and plays an important role among the three mnemonic texts. Written by an anonymous rhetorician in early first century BCE, it had mistakenly been 14. Chapter 55; Flynn, “De Arte Rhetorica,” 421–423.. 15. Carruthers, The Book of Memory, 153.. 16. Ibid., 129. 17. Flynn, “De Arte Rhetorica,” 24.. 21.

(38) attributed to Cicero until the Renaissance.18 It gave detailed instructions on the construction of places and images and demonstrated the application of the method of loci with examples. Although I do not have to repeat the method of loci cited by Quintilian and then by Soarez, it is worth to note that Rhetorica ad Herennium compared wax tablets and papyrus with memory places, letters with images, writing with the arrangement of images, and reading with delivery.19 These analogies between mnemonic elements and documentation imply that the art of memory is a method of archiving. The difference is that documentation requires real stationary to achieve, but the art of memory operates purely psychologically without graphical aids in the real world. The psychological nature gives the art of memory a great freedom to create memory places by imagination, opening the grand avenue toward abstract place-making, such as numerical grids or alphabetic combinations. After discussing memory place, the author of Rhetorica ad Herennium turned to the making of memory images. Although I will reserve the detail of image-making to our discussion on Jesuit’s conception of Chinese writing, I need to mention two kinds of images, verba and res, distinguished in Rhetorica ad Herennium. Res (lit. ‘things’) means the matters or the contents in one’s oration, and verba (lit. ‘words’) means the verbatim of one’s oration.20 Rhetorica ad Herennium uses the image of a lawsuit to explain the memory of res, and a poetic passage to exemplify the verbatim memory. In both cases, this Roman rhetorician was constrained by his alphabetic writing system, so he recollected the res and the verba through phonetic heuristics. In the following sections, I will see how this alphabetic 18. Yates, The Art of Memory, 5. In Loeb Classical Library, Rhetorica ad Herennium is still placed in Cicero’s collection. Regarding this editorial arrangement, I retain the attribution to Cicero in the citations and the bibliographical entry of Rhetorica ad Herennium. 19. Rhetorica ad Herennium Book III Ch 18. Cicero, “Rhetorica ad Herennium,” 211.. 20. It is Cicero’s definition cited in Yates, The Art of Memory, 8.. 22.

(39) constrain was broken when the Renaissance Europeans encountered “ideographic” writing, such as Egyptian hieroglyphs and Chinese characters.21. 2.2 Images and Loci In the previous section, I have reviewed the method of loci, a mnemotechnique introduced in the classical rhetoric texts and deeply institutionalized in the medieval European pedagogy through the rhetoric curriculum, creating a mentalité in organizing things among generations of European literati that lasted for centuries. Trigault and his contemporary Jesuits living in the late Renaissance were all immersed in this atmosphere. The method of loci involves a series of carefully arranged places in which many images are anchored. Now I will discuss the place further, for its transformation and symbolization after antiquity had influenced the methods of categorization, such as the focus of this thesis, lexicography. As the ancient rhetoricians repeatedly advised, the preparation of memory place is crucial in practicing a successful memory system. Rhetorica ad Herennium not only specified how a memory place should be in size, in distance, and in illumination, but also recommended the method of place arrangement. The arrangement of the places should be at best in a series, so one may virtually walk back and forth around these places to recollect memory images. This method is not Roman rhetoricians’ invention, but Aristotle had already mentioned it in his On Memory and Reminiscence. Aristotle discussed the importance of arrangement in recollection. He argued that recollection is executed by eliciting a stimulus to trigger its associated stimuli. Thus, we may arrange these stimuli in the natural or one’s habitual order to help recollection. Aristotle then explained an instance of recollection. 21. The origin and influence of the ideographic myth will be discussed in Chapter 3.. 23.

(40) through geometric reasoning, giving the art of memory a rational foundation.22 Aristotle’s emphasis on the importance of arrangement was cited and studied in medieval memory treatises, such as in the commentaries to Aristotle by Albertus Magnus (ca. 1206–1280) and Thomas Aquinas (1225–1274). Aquinas also mentioned Aristotle’s argument on arrangement in Summa Theologica,23 incorporating the method of loci into the most important theological discourse in Christianity. Aquinas’ fellow friars, the Dominicans, were primary promoters of this mnemotechnique.24 No method is perfect. When Quintilian cited the method introduced in Rhetorica ad Herennium, he had pointed out the drawback of the place arrangement in the method of loci. He criticized that this method forces one’s mind to do speech and recollection simultaneously, loading too much burden on one’s mind, which inevitably interrupts one’s speech. The purpose of mnemonics is to facilitate oration rather than to interfere in it. Instead, Quintilian recommended a simpler method by memorizing the passage as it is written on its original medium, such as a wax tablet, and committing signs or marks only for really difficult parts.25 Quintilian’s revisionist approach moved the memory place from mental imagination to an ever-tangible medium, but he still respected the importance of arrangement. In the revised method, the arrangement is determined by the written order of the passage. However, methods aimed at easing one’s memory burden are not absent in the Rhetorica ad Herennium. It recommended marking every five places with a cue to remind the location. For example, the mark for the fifth place is a golden hand; the five fingers prompt Number 5 22. Aristotle, “On Memory and Reminiscence,” in The Treatise of Aristotle (London: Robert Wilks, 1808), 163–177. 23. Thomas Aquinas, Summa Theologica, Volume 3 (Part II, Second Section) (New York: Cosimo, 2007),. 24. Yates, The Art of Memory, 84–85.. 25. Institutio Oratoria Book 11.2 32–33. Quintilian, The Institutio Oratoria of Quintilian, 75.. 1396.. 24.

(41) in one’s mind. For the tenth place, the Roman rhetorician suggested a man called Decimus; his name sounds similar to Latin “one-tenth (decima).”26 The mark for the tenth place was altered to a cross in Christian’s memory treatises (Figure 2.1). We may observe instances of this alternative in the collection of medieval mnemonic texts of the German Dominican friar Johannes Romberch (or Johann Host von Romberch, ca. 1480–ca. 1533) and in Xīguó Jìfǎ of Ricci in the early and the late sixteenth century.27 These numerical indications of places make all places in a memory system inducible through simple calculation, easing one’s memory burden to maintain an imagined structure in mind. The implication of numerical indications is significant to the symbolic place-making. The numerical mnemotechnique employs a sign to denote a place, and meanwhile, the sign becomes a symbol referring to that place. This semiosis suggests an idea that it is possible to mark all places with a set of signs that possesses a pre-established, familiar arrangement. Thus, if we mark places with the Roman alphabet, then these places would be arranged in the alphabetical order. This method respects and preserves the principle of arrangement with an economic method. In fact, the author of Rhetorica ad Herennium had already thought of a similar method, the Greek shorthand writing, or “tachygraphy.” The origin of tachygraphy is unclear, but the first record of this writing is dated to the first century BCE. Cicero was also reported being trained with this method. Tiro, Cicero’s confidential secretary, reportedly devised signs for Latin prepositions and declensions.28 Tiro’s shorthand writing employs simple signs to represent other concepts, which involves semiosis, associating the signifier (shorthand signs) and the signified (Latin prepositions and declensions). Conventionalization 26. 27. Yates, The Art of Memory, 7.. Johann Host von Romberch, Congestorium Artificiose Memorie (Venetijs: Melchiorem Sessam, 1533), 36–37; Ricci, “Xīguó Jìfǎ ,” 186–187.. 西國記法. 28. Herwig Maehler, “Tachygraphy,” in The Oxford classical dictionary, ed. Simon Hornblower and Antony Spawforth (New York: Oxford University Press, 1996), 1468–1469.. 25.

(42) is also expectable if shorthand writers share a system. However, the author of Rhetorica ad Herennium certainly opposed symbolic methods like shorthand signs for two reasons. First, he believed that an impressive image could be most effective to elicit its creator’s memory but less to stimulate others. Second, he believed it was impossible to signify each concept with a sign because the number of concepts is infinite, making the exhaustible signs quickly face shortage.29 Medieval Europeans’ solution to the shortage of signs was to employ more foreign characters for memory.30 A similar solution was also employed by Giordano Bruno.31. 29. Cicero, “Rhetorica ad Herennium,” 221–223.. 30. Carruthers reports that a manuscript in Morgan Library lists Hebrew, Greek, Latin, Scythian, and Runic alphabets. Carruthers argued that the only explanation to list these foreign alphabets is mnemonics. Carruthers, The Book of Memory, 138. 31. Yates, The Art of Memory, 208.. 26.

(43) Figure 2.1 Instances to memorize numbers with corresponding images in Romberch’s Congestorium Artificiose Memorie Note that ten, twenty, and thirty, all employ a cross, indicating they are multiples of ten. Multiples of five, such as are indicated with a hand, has five fingers. Source: gallica.bnf.fr / Bibliothèque nationale de France. 27.

(44) 2.3 Manipulating the Alphabets Although the classical text of the art of memory opposes the use of symbolic mnemotechniques, it could not stop people devising symbolic methods to help memory. In fact, in court and monastery, where oration was primarily exercised in medieval Europe, varieties of mnemotechniques did exist. These alternative systems were more digitalized, allowing recollection through computational inference, and had widely disseminated prior to the fourteenth century.32 In order to demonstrate that mnemotechniques derived from the method of loci, I will review several methods developed in the medieval Europe, including the use of numerical grid, alphabetization, and finally arrive the Combinatory Art (ars combinatoria), the most significant influence on the Ulmoçu. The numerical grid was the method that Hugh of St. Victor devised to memorize the Psalms. Dated around 1130 CE, this elementary mnemonic design divided the verses of Psalms into shorter passages and placed them in 150 numbered compartments, making a memory grid. The number assigned to each compartment became the mnemonic cue to stimulate the indicated passage of the Psalm in one’s mind. The numbers, in Aristotle’s term, have a natural order and allow the practitioner to search the passage on demand through calculation.33 Hugh’s mnemotechnique was not a radical reformation since it observed the advices from the ancient rhetoricians, but it still demonstrated the efficiency of symbolized places in recollection. Moreover, it is worth to note that the grid system represented not only a mnemotechnique, but also an indexing system in medieval Europe. The indexing system led the emergence of archiving systems around 1200 CE.34 The case of Hugh supports that. 32. Carruthers, The Book of Memory, 99.. 33. Ibid., 101–103.. 34. Ibid., 129.. 28.

(45) mnemonics and archiving share a similar rationale. The ultimate goals of an archiving system, mnemonic system, and lexicography are identical; they all seek an efficient approach to retrieve information quickly and accurately. Thus, it is reasonable to develop a coding and filing system from an existing mnemotechnique. Now let us turn to alphabetization, the method to arrange words according to their alphabetical order. This method is not only used in the art of memory but also in lexicography up to our time. Although the classical mnemonics warn that using the alphabet to mark memory place quickly faces shortage of place markers, it was a common device to arrange things. The alphabets provided a pre-existing order before setting up one’s artificial memory system because it was elementary and well known in the medieval European education. A person who needed an artificial memory aid was supposed to be literate and knew the order of the alphabets. Since the alphabets were also a script to represent language, its order was definitely an obvious arrangement to sort words in dictionaries or glossaries. However, we should not take alphabetic arrangement as a common, granted technique in any community using alphabets. Daly’s survey shows that although alphabetization was possible to be discovered as soon as the Greek adopt the Phoenician script, the Greek had not used alphabetization 500 hundred years after the adoption. Daly believed alphabetization was not economic if the data were not large enough to reach an economic scale; otherwise, alphabetization would take great effort for preparation, but reward little improvement in consulting data. The collection in the library of Alexandria was worth to apply alphabetization to organize and catalogue because it was large enough. This adoption made Alexandria become the center to disseminate this technique in antiquity.35 A certain regression of civilization in the Early Middle Ages made the medieval 35. Daly, Contributions to a History of Alphabetization in Antiquity and the Middle Ages, 93-–95.. 29.

(46) European have to revive this old method in the eleventh century. In Daly’s survey, it was not until the mid-eleventh century, the Elementarium doctrine erudimentum of Papias, completed around 1053, gave the first description of the alphabetization principle. Since there was nearly no use of alphabetization in Latin works in late antiquity, it could be considered as a lost technique in the Western Europe up to Papias’ time, even though people had still practiced similar arrangement elsewhere.36 Papias used the first three letters to arrange the Latin lexemes but ignored doubled consonants and aspiration in determining the order. In Papias’ description to the alphabetization, he used “subdivision (subdivisionis)” to call a set of words sharing the first three letters. Therefore, for example, the words with aba- initial form a subdivision out of the vocabulary in Papias’ lexicography, and the words with Abdinitial form another.37 In terms of symbolic place-making, Papias’ “subdivisions” are the places denoted by the initial letters, and these places are arranged in alphabetic order. There were other alphabetization schemas in the high Middle Ages, and they divided their “subdivisions” in different ways. For example, Hugutio of Pisa (late 12th century) determined word order only by the first letter,38 and Richard Fishacre (late 13th century) by the first two vowels, creating a syllable-basis alphabetization. The number of subdivisions is the crucial design here. Generally, a great number of subdivisions permit a fewer number of lexemes in each subdivision, saving time to look for words within it.39 In the Ulmoçu, I find Trigault had a more sophisticated schema to prepare subdivisions, and I will discuss it in Section 4.4.. 36. Ibid., 71–72.. 37. Lloyd W. Daly and B. A. Daly, “Some Techniques in Mediaeval Latin Lexicography,” Speculum 39, no. 2(1964): 233–234. 38. Ibid.: 235.. 39. Carruthers, The Book of Memory, 150–151.. 30.

(47) 2.4 The Combinatory Art In lexicography, a compiler can arrange entries by their first letters. This is an inductive process because the places of each lexeme had pre-determined by its spelling. A compiler’s task is to embody this pre-determined arrangement. However, there was also a deductive process that created alphabetically denoted places through letter combinations, and the combinations were interpreted to derive their denoting contents. It is Lullism introduced by Franciscan friar Ramon Lull. Under the influence of Lullism, the mechanism of letter combination was then called the Combinatory Art, influencing many intellectual innovations in the sixteenth and seventeenth centuries. Ramon Lull was born in today’s Mallorca, Spain, an island shared by Muslim, Jewish, and Christian communities in the thirteenth century. His major contribution to the history of thought was the proposal of an art of memory based on combination and reasoning. While the architectural mnemonics had been using stationary images and places to help memory, Ramon Lull’s art of memory, or Lullian mnemonics, took an active approach to formulate the memory skill. Lull introduced a series of letter notations to signify a set of holy attributes, or God’s Divine Names in Lull’s term, and defined the principles to interpret the combinations of these holy attributes. This operation was best demonstrated by a famous figure shown in Figure 2.2. This figure is made of three concentric wheels on which the letters denoting the Divine Names were inscribed. Interestingly, the peripheral wheel in this figure is stationary, and the two inner wheels rotate against it to form letter alignment, generating combinations of the Divine Names. If we read these combinations according to Lull’s method, we can render a set of arguments answering the designated theological problems.. 31.

(48) Figure 2.2 A Lullian Wheel Each Letter on the concentric wheels represents a God’s Divine Name. Source: gallica.bnf.fr / Bibliothèque municipale de Lyon Part-Dieu.. I will not discuss the metaphysics and theological foundations of Lull’s philosophy,40 which are out of the scope of this thesis. What I shall investigate is the combinatorial mechanism of Lullian reasoning. The Divine Names are the fundamental constituents in. 40. English translations of Lull’s major works: Ramon Llull, “Doctor illuminatus: a Ramon Llull reader,” ed. Anthony Bonner (Princeton, N.J.: Princeton University Press, 1993). The discussion on the philosophical background and the theological argumentation of Lullism, see Yates, The Art of Memory, 173–198; Umberto Eco, The Search for the Perfect Language, The making of Europe (Oxford: Blackwell, 1995), 53–72.. 32.