Critical Readings on Tang China [Volume 3] 9004380191, 9789004380196

Citation preview

Critical Readings on Tang China Volume 3

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Critical Readings on Tang China volume 3

Edited by

Paul W. Kroll

LEIDEN | BOSTON

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Library of Congress Cataloging-in-Publication Data is available online at http://catalog.loc.gov Library of Congress Control Number: 2018962592

Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface isbn 978-90-04-28113-4 (hardback, set) isbn 978-90-04-28169-1 (hardback, vol. 1) isbn 978-90-04-28168-4 (hardback, vol. 2) isbn 978-90-04-28167-7 (hardback, vol. 3) isbn 978-90-04-28166-0 (hardback, vol. 4) isbn 978-90-04-38015-8 (e-book, vol. 1) isbn 978-90-04-38016-5 (e-book, vol. 2) isbn 978-90-04-38019-6 (e-book, vol. 3) isbn 978-90-04-38020-2 (e-book, vol. 4) Copyright 2019 by Koninklijke Brill NV, Leiden, The Netherlands. Koninklijke Brill NV incorporates the imprints Brill, Brill Hes & De Graaf, Brill Nijhoff, Brill Rodopi, Brill Sense, Hotei Publishing, mentis Verlag, Verlag Ferdinand Schöningh and Wilhelm Fink Verlag. All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission from the publisher. Authorization to photocopy items for internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA. Fees are subject to change. This book is printed on acid-free paper and produced in a sustainable manner.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Contents Volume 1 Tang Emperors’ Accession Dates and Reign Titles (nianhao 年號) xi General Introduction 1 Paul W. Kroll

History—Political, Intellectual, and Military 1 Wen Ta-ya: The First Recorder of T’ang History 11 Woodbridge Bingham 2 The Rise to Power of the T’ang Dynasty: A Reassessment 17 Howard J. Wechsler 3 The T’ang Imperial Family 41 Denis Twitchett 4 Canonical Scholarship 100 David McMullen 5 Neo-Confucianism and Neo-Legalism in T’ang Intellectual Life, 755–805 163 Edwin G. Pulleyblank 6 The Structure of T’ang Selection 214 P. A. Herbert 7 Decree Examinations in T’ang China 237 P. A. Herbert

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

vi

CONTENTS

8 The Bureaucratic Apparatus [of T’ang Historians] 267 Denis Twitchett 9 Bureaucracy and Cosmology: The Ritual Code of T’ang China 295 David McMullen 10

Wei Cheng’s Thought [esp. Regarding Government] 346 Howard J. Wechsler

11

Imperial Power and the Ruling Class [under Empress Wu] 367 Richard W. L. Guisso

12

The Chou Dynasty [of Empress Wu] 404 Richard W. L. Guisso

13

The Career of Yang Kuei-fei 455 Howard S. Levy

14

The Flight from the Capital and the Death of Precious Consort Yang 482 Paul W. Kroll

15

Foreign Policy 503 P. A. Herbert

16

The An Lu-shan Rebellion and the Origins of Chronic Militarism in Late T’ang China 518 Edwin G. Pulleyblank

Volume 2 Literature and Cultural History 17

T’ang Literati: A Composite Biography 545 Hans H. Frankel

18

Transparencies: Reading the T’ang Lyric 566 Stephen Owen

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

vii

CONTENTS

19

The Significance of the fu in the History of T’ang Poetry 584 Paul W. Kroll

20 An Offering to the Prince: Wang Bo’s Apology for Poetry 597 Ding Xiang Warner 21

Tamed Kite and Stranded Fish: Interference and Apology in Lu Chao-lin’s fu 636 Paul W. Kroll

22

A Re-evaluation of Chen Ziang’s “Manifesto of a Poetic Reform” 666 Timothy Wai Keung Chan

23

On Li Po 694 Elling O. Eide

24 Li Po’s Letters in Pursuit of Political Patronage 731 Victor H. Mair 25 Li Bai’s “Rhapsody on the Hall of Light”: A Singular Vision of Cosmic Order 762 Nicholas Morrow Williams 26 Tu Fu 825 Stephen Owen 27

Tu Fu’s Social Conscience: Compassion and Topicality in his Poetry 894 Shan Chou

28 Poems in Their Place: Collections and Canons in Early Chinese Literature 936 Pauline Yu 29 Heyue yingling ji and the Attributes of High Tang Poetry 967 Paul W. Kroll 30 The Formation of the Tang Estate Poem 1001 Stephen Owen

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

viii

CONTENTS

Volume 3 31

Lexical Landscapes and Textual Mountains in the High T’ang 1021 Paul W. Kroll

32 Historical and Literary Theory in the Mid-Eighth Century 1060 David McMullen 33 The Manuscript Legacy of the Tang: The Case of Literature 1096 Stephen Owen 34 Literary Collections in Tang Dynasty China 1126 Christopher M. B. Nugent 35 A Study of the Jinglong Wenguan Ji 1171 Jia Jinhua 36 The Old-Style fu of Han Yu 1204 David R. Knechtges 37

Another Go at the Mao Ying chuan 1230 Elling O. Eide

38 The Old Men [of the Early Ninth Century] 1235 Stephen Owen 39 The Inscription of Emotion in Mid-Tang Collegial Letters 1281 Anna M. Shields 40 Yüan Chen and ‘The Story of Ying-ying’ 1326 James R. Hightower 41

Nostalgia and History in Mid-Ninth-Century Verse: Cheng Yü’s Poem on “The Chin-yang Gate” 1360 Paul W. Kroll

42 Remembering Kaiyuan and Tianbao: The Construction of Mosaic Memory in Medieval Historical Miscellanies 1441 Manling Luo

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

ix

CONTENTS

43 The Dancing Horses of T’ang 1476 Paul W. Kroll 44 Falconry in T’ang Times 1504 Edward H. Schafer 45 Public Values in Calligraphy and Orthography in the Tang Dynasty 1540 Amy McNair

Volume 4 Religion 46 The Role of Buddhist Monasteries in T’ang Society 1559 Kenneth K. S. Ch’en 47 Buddhism and Education in T’ang Times 1580 Erik Zürcher 48 Imperial Patronage in the Formation of T’ang Buddhism 1622 Stanley Weinstein 49 Stūpa, Sūtra, Śarīra in China, c. 656–706 CE 1663 T. H. Barrett 50 The Birth of a Patriarch: The Biography of Hui-neng 1714 Philip B. Yampolsky 51

Metropolitan Chan: Imperial Patronage and the Chan Style 1743 John McRae

52 Time after Time: Taoist Apocalyptic History and the Founding of the T’ang Dynasty 1776 Stephen R. Bokenkamp 53 Taoist Ordination Ranks in the Tun-huang Manuscripts 1805 Kristofer M. Schipper

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

x

CONTENTS

54 Taoism in the T’ien-pao Era, 742–56 1829 T. H. Barrett 55 Li Po’s Transcendent Diction 1839 Paul W. Kroll 56 Immortality Can be Studied 1875 Jan De Meyer 57

The Worshippers of Mount Hua 1915 Glen Dudbridge Index of Personal Names 1949

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T’ang Paul W. Kroll The sinological study of what has been called “mountains and the cultures of landscape” began in earnest nearly ninety years ago with Édouard Chavannes’ monumental monograph on T’ai Shan.1 During the years since then, while important scholarly studies of other discrete peaks have appeared,2 the related but broader issue of traditional China’s accommodation with the natural world—from literary and political, as well as religious angles—has inspired a proliferating number of books and articles.3 Several scholarly symposia have Source: “Lexical Landscapes and Textual Mountains in the High T’ang,” T’oung Pao 84 (1998): 62–101. 1  Le T’ai chan: essai de monographie d’un culte chinois (Paris: Ernest Leroux, 1910). 2  Such as Michel Soymié’s “Le Lo-feou shan: étude de geographie religieuse,” Bulletin de l’École française d’Extrême-Orient 54 (1956), 1–132, and Edward H. Schafer’s Mao Shan in T’ang Times, rev. 2nd edn. (Boulder: Society for the Study of Chinese Religions, 1989). 3  Major landmarks include Aoki Masaru’s 1934 article, “Shinajin no shizenkan” 支那人の自然 観, rpt. in Aoki Masaru zenshū 靑木正兒全集, 6 vols. (Tokyo: Shunjūsha, 1969–75), 2: 552–91; Richard B. Mather, “The Landscape Buddhism of the Fifth-Century Poet Hsieh Ling-yün,” Journal of Asian Studies 18.1 (1958), 67–79; J. D. Frodsham, “The Origins of Chinese Nature Poetry,” Asia Major 2nd ser., 8.1 (1960), 68–104; Obi Kōichi 小尾郊一, Chūgoku bungaku ni arawareta shizen to shizenkan, chūsei bungaku o chūshin to shiru 中国文学に現われた 自然と自然觀, 中世文学を中心としる (Tokyo: Iwanami, 1963); Paul Demiéville, “La montagne dans l’art littéraire chinois,” France-Asie 183 (1965), 7–32, rpt. in Demiéville, Choix d’études sinologiques (1921–1970) (Leiden: Brill 1973), 364–89; Wang Kuo-ying 王國瓔, Chungkuo shan-shui-shih yen-chiu 中國山水詩硏究 (Taipei: Lien-ching, 1986); Tokura Hidemi 戶倉英美, Shijintachi no jikū: Kanpu kara Tōshi e 詩人たちの時空 : 漢賦から唐詩へ (Tokyo: Heibonsha, 1988); and, most recently, Donald Holzman, Landscape Appreciation in Ancient and Early Medieval China: The Birth of Landscape Poetry (Taipei: College of Humanities and Social Sciences, National Tsing Hua University, 1996). Jean-Pierre Diény’s studies on the symbolism of various natural phenomena and images should also be mentioned here; e.g., “Pour un lexique de l’imagination littéraire en Chine: le symbolisme du soleil,” Bulletin de l’École française d’Extrême-Orient 69 (1981), 119–52. Of the numerous works focused on particular flora and fauna, one of the more interesting recent offerings is Martin Kern’s Zum Topos “Zimtbaum” in der chinesischen Literatur: Rhetorische Funktion und poetische Eigenwert des Naturbildes kuei (Stuttgart: Franz Steiner, 1994).

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_033

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1022

Kroll

been convened to investigate various aspects of the topic4 and have added their measure to the expanding bibliography. The journal Cahiers d’Extrême-Asie devotes considerable space to the subject of sacred geography.5 It sometimes seems that one can scarcely pick up a sinological journal these days without finding at least one article focused on some Chinese mountain—a frequency of study beginning to rival the perennial topics of Wang Wei’s “nature poetry” or Liu Tsung-yüan’s “landscape essays.” I too have contributed items to this outcropping of mountain and landscape scholarship. But one begins to wonder where it is all leading. Of course we should be, and are, grateful for each new fact, all new data, on any subject or place that helps us see more clearly into the culture of traditional China. When you’ve seen one Chinese mountain, you haven’t seen them all. But their outlines often seem to blur and fade into one another. Perhaps it is time to re-examine some of our assumptions about the literary values of Chinese landscapes. Now, a student of the T’ang has a particularly blurry lens to unfog, because it is such a well-studied and ever-attractive era. We must first, therefore, do our best to skirt the reductive categories and smooth clichés that have come to determine our responses, to prescribe our views. Hence we shall not here admit the term “nature poetry”—an unreflecting tag of no critical value. To be sure, we may acknowledge the fact that Chinese poetry has since its beginnings—like virtually all traditional poetry anywhere—drawn deeply on the natural world as a primary fund of imagery; but that is an obvious and, in truth, unremarkable fact. Squeeze this idea and you will notice how little remains in your grasp. Nor shall we warrant, on any but the most vulgar level, a convincing distinction between “religious” and “literary” texts—as though the educated elite of the T’ang conceived of themselves and their writings in terms of the disciplinary boundaries of our university curricula. Instead we accept as given, from earliest times, the continual merging and mutual affinity between the power of the word and the presence of the numinous. There may be more to say about these matters. To shake off the more persistent fleas is not easy. But there are two areas of concern that I wish to stress, before turning to the heart of this article. We may begin with a quatrain by Li Po 李白 (701–762?), known to us all: 4  Including the one for which an earlier version of this paper was prepared, at the University of California, Santa Barbara, in January 1993. 5  One of the better examples is James Robson’s ambitious article, “The Polymorphous Space of the Southern Marchmount (Nanyue): An Introduction to Nanyue’s Religious History and Preliminary Notes on Buddhist-Daoist Interaction,” Cahiers d’Extrême-Asie 8 (1995), 221–64.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1023

You ask for what purpose do I roost in the cyan hills; Smiling, I don’t reply, my heart freely at ease. Drifting waters with peach blossoms go off inscrutably: It is another heaven and earth, not amidst the human realm.6 We often hear that this little poem demonstrates Li Po’s disdain for the world or the worldly, that it reflects his love of Nature, that it reveals the apparent or hopeful aloofness and playfulness of his outlandish temperament—platitudes and banalities, even if correct. Perceptive readers notice the allusion in line three to the Peach Blossom Font of T’ao Ch’ien 陶潛 (365–427) and the suggestion in line four of the Taoist grotto-heavens. We have even seen an analysis that finds the implied “dialogue” of the poem’s title (“Question and Answer in the Mountains” 山中問答, as given in late texts) to be no true dialogue at all, simply because the persona does not verbalize his response.7 Yet, if only we are alive to the historical echoes of earlier poetry—as Li Po surely was—we shall catch an even more important hint. That is the question-and-answer exchange of T’ao Hung-ching 陶弘景 (456–536) and Ch’i Kao Ti 齊高帝 (r. 479–83). When the Ch’i emperor summoned to court the influential Taoist adept and scholar from his retreat on sacred Mao Shan 茅山, the ruler is said to have sent an edict with the supercilious query, “What is there to be had amongst the mountains? (山中何所有). T’ao Hung-ching, declining to leave his sanctuary just then, replied with the following quatrain: “What is there to be had amongst the mountains?” Plenty of white clouds atop the ridge. Here alone may one be freely cheerful and content— Impossible to send this in hand to you milord.8 6  Li Po chi chiao-chu 李白集校注, ed. Ch’ü T’ui-yüan 翟銳園 and Chu Chin-ch’eng 朱金城 (Shanghai: Shang-hai ku-chi, 1980; hereafter cited as lpccc), 19.1095; Ch’üan Tang shih 全唐詩 (Peking: Chung-hua, 1960; hereafter cts), 178.1813. 7  Stephen Owen, The Great Age of Chinese Poetry: The High Tang (New Haven: Yale Univ. Press, 1981), 136. This surprisingly flat-footed reading ignores the poet’s verbalization—i.e., the poem itself. And of course a “question and answer” is hardly a “dialogue” in any meaningful— much less, literal—sense, to begin with. Worse yet is that Owen bases his interpretation of the poem so heavily on this late and variant title, despite the fact that all the earliest editions of Li Po’s works, as also recensions of the mid-eighth-century Ho-yüeh ying-ling chi 河嶽英靈集 in which the poem was first anthologized, record the title as “Reply to a Worldly Fellow’s Question” 答俗人問, clearly enough denoting that there is a real response here. 8  Hsien-Ch’in Han Wei Chin Nan-pei-ch’ao shih 先秦漢魏晋南北朝詩, ed. Lu Ch’in-li 逑欽立 (Peking: Chung-hua, 1983; hereafter cited as Lu, Hsien-Ch’in), 1814; Mao Shan chih 茅山志

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1024

Kroll

The point in Li Po’s poem is not that the persona doesn’t reply but that he replies wordlessly: the scene replies for him, everything that can be said in words having already been said two and a half centuries earlier by T’ao Hung-ching when asked the same question.9 But we cannot overlook the fact that it must obviously be with carefully chosen words and images that Li Po the poet— as opposed to the smilingly unresponsive persona—portrays the scene itself. And those words point here not to a literal view of the mountain landscape but rather to a pair of literary images. The primacy of the word—not the eye—and of its accumulated resonances through time is the first item to be noted when rethinking the landscapes of T’ang poetry. This will occupy us at greater length in what follows. A second factor is the religious potency of the living world, taken so much for granted by the men of T’ang that, unless we find it specially lit up in text or rite, we are apt to make the false assumption that the landscape was for most of them—as for most of us now—a merely scenic backdrop to an urban-oriented life. But, for all the undeniable civilization of T’ang culture, these were people who were daily in a much more familiar relationship with the physical world than we can easily imitate. It requires an effort of imagination and understanding to insert ourselves conceptually into such a world, in the same way we must consciously force ourselves to realize the terrific difference between living in a print (or, increasingly, post-print) culture as opposed to a manuscript culture. The difference is so enormous that, no matter how hard we try, we can’t program ourselves to keep a grip on the alternate image for more than a moment: the reality of it is too alien for a prolonged embrace. Likewise, the simple fact that even in Ch’ang-an a nobleman or commoner of the eighth century would see from his urban residence every cloudless night the star-splashed sky, with a clarity and immediacy granted us only upon removal from the light-pollution and gaseous haze of our cities and suburbs, or that the moon in its nightly change of dress and varying light was not then just an occasionally noticed (hy 304), 28.2b; P. W. Kroll, Meng Hao-jan (Boston: G. K. Hall, 1981), 101. See T’ai-p’ing kuang-chi 太平廣記 (Taipei: Ku-hsin, 1976), 202.414a, quoting the early 7th-century T’an-sou 談藪 of Yang Sung-chieh 楊松玠, for the anecdotal context of the exchange. 9  Another possible echo from the past is identified by the earliest commentator on Li Po’s works, Yang Ch’i-hsien 楊齊賢 (fl. 1190): “Someone inquired of Chu-ko Liang’s ambitions. Clasping his knees, Liang smiled but did not reply.” For the original passage, see San-kuo chih 三國志 (Peking: Chung-hua, 1973), 35.911, quoting the Wei lüeh 魏略. The Shu connection between the poet and the fabled recluse and strategist is also relevant here. See Fen-lei pu-chu Li T’ai-po shih 分類補註李太白詩 (Ssu-k’u ch’üan-shu), 19.3a. Yang Ch’i-hsien’s comments are often pertinent, but unfortunately—as in this case—they are not always included in the more modern editions of Li Po’s works, even those purporting to be comprehensive in scope.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1025

but rather an inescapable and influential phenomenon—such facts are not insignificant. They are fundamentally determinative of one’s connections with the surrounding world. As noted at the outset of this article, much scholarly work in recent years has gone into studies of China’s “sacred geography.” So much so that it has now become a cliché to say that these “most prominent features of the Chinese landscape are not merely natural, but numinous, objects.”10 Still, it is good to remind ourselves of how deep-seated this notion was, both in the political and the religious sphere (which amounts to the same thing in this context). Recall, for instance, Li Lung-chi’s 李隆基 (pht. Hsüan Tsung 玄宗, r. 712–56) extreme and continuing concern for honoring and acknowledging the powers of his realm’s supreme mountains and rivers, evident in the new titles and enfeoffments bestowed during his reign on each of the Five Marchmounts (wu yüeh 五嶽: Mounts T’ai 泰, Heng 衡, Hua 華, Heng 恆, and Sung 嵩), Four Conduits (ssu tu 四瀆: the rivers Ho 河, Kiang 江, Huai 淮, and Chi 濟), Four Holdweights (ssu chen 四鎮: Mounts Kuei-chi 會稽, I 诉, I-wu-lü 醫無閭, and Huo 霍), Four Seas, and other assorted alpine and fluvial divinities.11 The imperially ordered sacrifices and propitiatory rites to these spirits were performed by highranking envoys of the throne, often princes of the blood.12 Hsüan Tsung’s own successful accomplishment of the feng 封 and shan 禪 observances at Mount T’ai in December 725 is, of course, well known. But what is interesting in all this is the impression one receives, when viewing the physical area of China in this perspective: it is a land anchored and watered not by geographical forces, but by the sway of a host of more or less local, more or less potent spirit-lords. 10  P. W. Kroll, “Verses From on High: The Ascent of T’ai Shan,” T’oung Pao 69 (1983), 223; rev. version in The Vitality of the Lyric Voice: Shih Poetry from the Late Han to the T’ang, ed. Lin and Owen (Princeton: Princeton Univ. Press, 1986), 167. Cf. Gaston Bachelard who noted the priority of the terrestrial (as opposed to the celestial) as an image of immensity and awe; hence, “La moindre colline, pour qui prend ses rêves dans la nature, est inspirée.” See his provocative La terre et les rêveries de la volonté (Paris: Librairie José Corti, 1948), 378–402, esp. 379–80 and (for the quote) 384. 11  See Ts’e-fu yüan-kuei 册府元龜 (Peking: Chung-hua, 1980), 33.7a–23b, for some of the imperial edicts relating to Hsüan Tsung’s activities in this area; also T’ang hui-yao 唐會要 (Taipei: Shih-chieh, 1960), 47.834, and T’ang ta chao-ling chi 唐大詔令集 (Taipei: Tingwen, 1978), 74.418–19. Kroll, “Verses From on High,” 236–37 n53 (184–85 n53) summarizes some of this information. 12  See, e.g., Ts’e -fu yüan-kuei, 33.13a (edict dated K’ai-yüan 25/x/8, i.e., 4 Nov. 737), 33.15b–16a (edict dated T’ien-pao 3/iv/3, i.e., 19 May 744), 33.21b–22a (edict dated T’ien-pao 10/iii/16, i.e., 16 April 751); “Ling ssu Cheng-wang Hsi-yen fen chi wu-yüeh ch’ih,” T’ang ta chao-ling chi, 74.418.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1026

Kroll

One might argue that by the T’ang this view was becoming somewhat obsolete, if it was not anachronistic; that the imperial rites referred to above were, for instance, turning largely into formal exercises. One might argue this, but the facts seem otherwise. It is a rather more complicated matter than wishing or declaring some vague—but, to modern minds, comforting—advance of “rationalist” tendencies. Jean Levi’s work in particular has made clear that the chief conceptual role of state officials in medieval times was as representatives of the state cult, the Confucian tao, centered on the “Son of Heaven”—not as political placemen.13 We have become so used to speaking of T’ang government functionaries as bureaucrats (with the unavoidable connotations, for us, of the corridors of power, the steps of the Capitol) that we look right past their defining activities as religious emissaries. Like the Taoist priests, they aimed to exert control over the sundry local gods—the gods of the profane—of any region, who were worshipped through “irregular cults” (yin-tz’u 淫祠); the wellattested duty to “transform by instruction” (chiao-hua 敎化) was no less than a replacing of provincial beliefs with education in and adherence to Confucian teachings, effected through the establishment of schools—and temples. Indeed, the functional activities of the warranted Confucian official and those of the ordained Taoist priest share so many similarities in medieval times that “peut-être fonctionnaires impériaux et maîtres du Tao ne sont-ils que les deux faces de la prêtrise en Chine.”14 In this regard, we may better appreciate the map of T’ang sacred geography we obtain when we overlay the imperially recognized Marchmounts, Conduits, etc., with the contemporary Taoist scheme of the ten major and thirty-six lesser grotto-heavens, and seventy-two “favored lands” ( fu-ti 福地), codified in Hsüan Tsung’s time by the great prelate Ssu-ma Ch’eng-chen 司馬 承禎 (647–735)15—to say nothing of the Buddhist network. We can even see 13  See especially his “Les fonctionnaires et le divin: luttes de pouvoir entre divinités et administrateurs dans les contes des Six Dynasties et des Tang,” Cahiers d’Extrême-Asie 2 (1986), 81–106, and Les fonctionnaires divins (Paris: Seuil, 1989). 14  Levi, “Les fonctionnaires et le divin,” 106. See especially K. M. Schipper, “Taoist Ritual and Local Cults of the T’ang Dynasty,” in Tantric and Taoist Studies, in Honour of R. A. Stein, vol. 3 (Mélanges chinois et bouddhiques 22; Brussels: Institut Belge des Hautes Études Chinoises, 1985), 812–34, for examples of encounters between local cults and the “pure” theology of the Tao. The best discussions of the centrality of Taoism in Chinese culture are Schipper’s Le corps taoïste: corps physique—corps social (Paris: Fayard, 1982) and, on a smaller scale, Anna Seidel’s Taoismus, die inoffizielle Hochreligion Chinas (Tokyo: Deutsche Gesellschaft für Natur- und Völkerkunde Ostasiens, 1990). 15  For the full layout, see Ssu-ma’s “T’ien-ti kung-fu t’u” 天地宮府圖, preserved in the 11thcentury encyclopedia Yün-chi ch’i-ch’ien 雲笈七籤 (hy 1026), ch. 27. This formed the

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1027

in this period a subsumption of the imperial plan by the Taoist one. When Hsüan Tsung, convinced by Ssu-ma Ch’eng-chen that the true lords of the Five Marchmounts were in fact Realized Ones (chen-jen 眞人) of the Shang-ch’ing 上淸 pantheon (in contrast to the mere “gods of mountain and grove” 山林之神 that the emperor had been fostering), come down to discharge their responsibilities in the mortal world from seats in the “grotto archives” (tung-fu 洞府) of the great mountains, the monarch decreed the establishment of shrines on these peaks, to be erected according to Ssu-ma Ch’eng-chen’s specifications, called “Bethels of the Realized Lords” (chen-chün tz’u 眞君祠) in honor of these prepotent divinities.16 But the landscape was not, of course, simply a matter of terrestrial geography. This is another point we usually slip over too unthinkingly, unfeelingly. So it merits reminding, for instance, that the traditional notion of the fen-yeh 分野 or “apportioned countrysides,” by which the various geographic and political divisions of early China were each paired with a corresponding sky sector and subject to the latter’s accompanying astral influences, was still very much alive in T’ang times. The Taoist view, let us recall, could see earthly landscapes as representations and, in some respects, containers of celestial phenomena—just as the human body could itself be regarded as a microcosm, embodying somatic and psychic identity with mountains, rivers, valleys, sun, moon, and stars. And to the Buddhists, notably the monk I-hsing 一行 (683–727), who had an important role in the spread of the Esoteric School, goes the honor of perfecting a fully worked-out system that added to the old fen-yeh template a finer correlation with the twenty-eight “lunar lodgings” (hsiu 宿) that had earlier been imported from Indian astronomy. But I only mention these matters—ignoring many others (such as the growing interest in private gardens, the development basis for the now better-known arrangement in the Tung-t’ien fu-ti yüeh tu ming-shan chi 洞天福地嶽瀆名山記 (hy 599) of Tu Kuang-t’ing 杜光庭 (850–933). In his preface Ssu-ma Ch’eng-chen gives a most interesting description of the formation of the sacred abodes and the taking up of residence in them by the various divinities. There are numerous pre-T’ang formulations of Taoist sacred geography; a convenient summation of the earthly locales is that in the sixth-century compendium Wu-shang pi-yao 無上祕要 (hy 1130), 4.5b–10b. Note also the early list of twenty-eight mountains that Ko Hung 葛洪 (283–344) gives as being places “conducive to concentrated thought and the compounding of medicaments for transcendence”; see Pao-p’u Tzu nei-p’ien chiao-shih 抱釙子內篇 校釋, ed. Wang Ming, rev. edn. (Peking: Chung-hua, 1988), 4.85. 16  See Chen hsi 眞系 (by Li Po 李渤), preface dated 805), as preserved in Yün-chi ch’i-ch’ien, 5.15b. The Chen hsi biography of Ssu-ma Ch’eng-chen is the basic source for his biography in the Chiu T’ang shu 舊唐書 (Peking: Chung-hua, 1975), 192.5127–29, where this incident is also recorded.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1028

Kroll

of historical maps, the spread of Buddhist maṇḍalas, the continuing fascination with “container gardens”)—on the fly, as examples of broad-scale ideas common, indeed commonplace, to eighth-century writers but which we must struggle to keep in front of our eyes, or at least not too far back in our minds, lest they slide imperceptibly out of the picture and thereby allow us to displace the T’ang landscape inadvertently with our own. What all this leads to is that, despite the well-documented dulcifying of the landscape effected during the Nan-pei-ch’ao period, despite the gentler appreciation of a more benign natural environment that came with the continuing cultural assimilation of the Chiang-nan region, we are yet involved with a landscape of “natural supernaturalism,”17 even if the deeper fastnesses are no longer to be fully dreaded as “zones of sacred horror.”18 Inadequate as these remarks may be, they shall have to serve for preliminaries. Almost every area of T’ang life is somehow relevant to this topic, and it is easy to find forking paths. But it is time now to consider some texts. For the moment, we shall let Li Po be our touchstone and shall remain earthbound. Look first at two poems on the waterfall at Mount Lu 廬山. The better known of the two is the quatrain, in heptasyllabic lines: Sunlight illumines the Incense Burner, quickening a purple haze; Far off I see the Sheet of Spray—a vertical waterway before me. Its flying flow descends straight down three thousand feet; I fancy that is the Silver Ho, dropped down from the Nine Heavens!19

17  The phrase is M. H. Abrams’, used as a defining concept in his famous study, Natural Supernaturalism: Tradition and Revolution in Romantic Literature (New York: Norton, 1971), a book that offers many interesting points of comparison for the present subject. 18  Demiéville, “La montagne dans l’art littéraire chinois,” 15 (372), describing the “classical” feeling for mountains as it persisted into the early medieval period. See Marjorie Hope Nicholson’s Mountain Gloom and Mountain Glory: The Development of the Aesthetics of the Infinite (1959; rpt. New York: Norton, 1963), for the story of the comparable transformation of mountain landscapes in European literature from the late-seventeenth through the eighteenth century. 19  “Wang Lu Shan p’u-pu, erh shou,” no. 2, lpccc, 21.1241; cts, 180.1837. The tenth-century Wen-yüan ying-hua 文苑英華 (Taipei: Hsin-wen-feng, 1979; hereafter wyyh), 164.4a prints a variant first couplet, in which the present opening line is displaced to line two and preceded by a new first verse that reads: “Mount Lu on high with the Starry Dipper is joined.” This seems to me too bald and flat an opening gambit; nor does the second line then follow in a satisfactory fashion.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1029

The final couplet has been a focus of critical acclaim, including special praise from Su Shih 蘇軾 (1036–1101),20 with its surprising conception of the waterfall as the upended Milky Way pouring down from the sky. It is not so often noticed how neatly the poet has prepared for this conceit from the very beginning of the poem. It is not the waterfall we see first, but Lu Shan’s northwest bluff,21 Incense Burner Peak 香爐峯, seen lit by the sun. As often, Li Po here plays on the semantics of a place-name, seeing Incense Burner as exactly what its name proclaims it to be. The first image the poet reveals to us, it is a true mountaincenser (a po-shan lu 博山爐) with a cap-cloud of heaven-hued mist. The cloudtrailings are the visible mark of the mountain’s energy (ch’i 氣). But then it is the color of the haze—this one detail—that fuses the visual imagery of the line into a notional unity, purple being the color of the celestial pole, of cosmic  “Sheet of Spray” may at first seem an overtranslation of p’u-pu 瀑布, the common “waterfall.” But it has the virtues of retaining both the original image behind this phrase (it is not a binom) and a hint of its phonetic resonance (M. C. *bau- pu-). This rendering also allows us to deal more readily with related phrases, such as 瀑布泉 , 瀑泉 , 瀑布水 etc., which must be distinguished from p’u-pu and which otherwise might be reduced to homologous “waterfalls” as well. Although it may be a commonplace in modern Mandarin, the term p’u-pu was not yet so in the eighth century. There is little evidence for its use prior to the fourth century, when Sun Ch’o 孫綽 (ca. 310–97) employs it in his “Yu T’ienT’ai Shan fu” (Rhapsody on Roaming in the T’ien-t’ai Mountains), a composition whose influence over later “ascent” literature was enormous. P’u-pu was also used sometimes, as in Li Po’s poems, as the proper name of the famous Lu Shan waterfall: in this case, merely adding a capital W to “waterfall” does not suffice, nor can the article “the” (even should it too boast a capital) bear this much semantic weight without seeming foolish. We must, therefore, keep the phrase intact and not truncate it to a single word—at least, not till after the T’ang. In any event, once one sees the phrase several times, “sheet of spray” soon loses any unusual coloring. 20   Tung-p’o chih-lin 東坡志林, ed. Wang Sung-ling (Peking: Chung-hua, 1981), 1.4. 21  There are four Incense Burner Peaks at different spots on the Lu Shan massif. The one referred to in all pre-T’ang references is that on the northwest side, associated with the Buddhist monk Hui-yüan. The pronouncement of Wang Ch’i 王琦 (1690–1774), the great Ch’ing editor and commentator of Li Po’s works, seemed to resolve any dispute about which Hsiang-lu feng was meant by Li Po. However, in her recently published edition of Li Po’s works, with chrono-biographical notes, An Ch’i 安旗 argues—as had some preCh’ing scholars—that the peak in question is the Incense Burner further south (though still “west” from the poet’s starting-point in line one). See Li Po Ch’üan-chi pien-nien chushih 李白全集編年注釋, 3 vols. (Chengtu: Pa-Shu shu-she, 1990), 49. While An Ch’i’s geographical proofs seem persuasive, they fly in the face of literary tradition that was already established in Li Po’s time. Note also that Hui-yüan’s biography in the sixth-century Kao seng chuan 高僧傳 locates a beautiful waterfall near the northwestern Incense Burner; see Kao seng chuan 6 (T 2059), 50: 358b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1030

Kroll

unity; here it establishes an initial prejudice for the otherworldly that is realized in the poem’s conclusion.22 Under the pressure of that first, controlling image, the coursing waterfall of the more straightforwardly presented lines two and three is turned into something more than its normal seeming—the revelation that strikes the poet in line four. What we have observed here, without (I hope) treading too heavily, is a small example of imagination and language molding a scene that might have been thought purely descriptive. Before I say anything more, we should put alongside this quatrain its longer companion, written in pentasyllabic lines:

4



8

To the west I scaled Incense Burner Peak, To the south viewed the waters of the Sheet of Spray— Vertical in flow for three hundred staves, Spurting through the strath several tens of leagues; Flashing like the onset of lightning in flight, Or dulled as though at the raising of a white rainbow. At first I feared that the Ho and Han had dropped down, Half spewed out from within the clouded heavens!



Look upward and observe the rolling might of its power— Oh, valiant—the exploit of the Shaper of Mutations! A wind from the sea blows on, unrelentingly; And the river’s moon shines, back into the void.



16

Out of that void a confused confluence shoots forth, Washing to left and right the walls of bice-blue. Flying pearls scatter in buoyant rose-pink clouds, As a flowing froth churns over vaulted rocks.



And it is mine to take delight in a mountain of renown! Confronting it, the mind is ever more free. Let there be no more talk of quaffing rose-gem liquor,

12

22  The symbolic significance of the color purple has been summarized by Manfred Porkert as “kosmische Ganzheit und Fülle, ungeschmälerte Macht, deshalb auch die wiedergewonne Einheit, die Rückkehr zum kosmischen Tao”; see Porkert, “Untersuchungen einiger philosophisch-wissenschaftlicher Grundbegriffe und Beziehungen im Chinesischen,” Zeitschrift der deutschen morgenländischen Gesellschaft 110.2 (1961), 439–40. On the atmospheric phenomenon imaged here, see Kroll, “Li Po’s Purple Haze,” Taoist Resources 7.2 (1997).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang 20



22

1031

For I am now to gain a face washed clean of dust! Now reconciled to sojourn where I please, Let me take leave forever of the human realm!23

Although not as well known as the previous work, this poem is an extraordinary composition that reveals its merits the closer one looks. We notice right off that most of the imagery contained in the quatrain reappears here—in somewhat expanded form—in the first octave. But this time the image of the fallen Milky Way (ll. 7–8) leads on to a further and fuller examination of the waterfall. A change of rhyme at this juncture allows for a brief pulling back, for an imperative command and exclamation to appreciate the work of the “Shaper of Mutations,” the great selecting and synthesizing principle of Nature24 (and let’s catch the peculiarly apposite use here of the adjective “rolling” 轉 in describing the virile surge of the waterfall), after which Li Po picks out two details for special attention. Anyone who has had experience of a waterfall knows exactly what line eleven refers to—there seems in the vicinity always to be a wind blowing, harder and blusterier than expected. Line twelve’s moonlight, reflected in and off the water (the waterfall here being seen as a vertical “river” 江 [cf. the quatrain’s suspended “waterway” 川]), seems to radiate an aura that is diffused back into the sky—a beautiful image that also recalls the lighting effects of lines five and six.25 The next rhyme-change is rhetorically bridged through the use of anadiplosis (Ch. ting-chen 頂針), the “void” 空 into which the moonlight melts at the close of line twelve being the identical emptiness at the beginning of line thirteen out of which seems to shoot the “confused confluence” 亂潀 that we follow in this stanza more closely in its rush down the dark cliffside. Its outflung droplets, the “flying pearls” of line fifteen, are dissipated into the surrounding iridian mists (as the reflected moonlight earlier was thrown back to the sky), while the current bounds on, foaming down its steep path. With the next rhyme-change, the ostensibly descriptive section of the poem has come to an end; so let us pause a moment to consider how Li Po has presented this rather extended description. Throughout these first three stanzas the directional imagery has alternated up and down the vertical plane, with 23  “Wang Lu Shan p’u-pu, erh shou,” no. 1, lpccc, 21.1238–39. cts, 180.1837. 24  See Edward H. Schafer, “The Idea of Created Nature in T’ang Literature,” Philosophy East and West 15.2 (1965), 153–60, for a cursory view of the concept. Gerard Manley Hopkins’ notions of “instress” and “inscape” include similarities worth pursuing. 25  These two lines are lauded highly by those scholars—including the commentators Hu Chen-heng 胡震亨 (fl. 1590) and Wang Ch’i—who prefer this poem to the quatrain.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1032

Kroll

some carefully inserted horizontal spreads (1 up, 2–4 down, 5–6 up, 7–8 down, 11 horizontal, 12 down/up, 13 down, 14 horizontal, 15–16 outward/up/down), imitating perfectly the “rolling might” of the waterfall: one’s eye is always in movement, drawn in different views down and up constantly, as variously repetitive as the streaming cascade itself. It would retard our discussion too much to examine all the spots where Li Po has reinforced sense and sight with sound, but two really must be mentioned. Lines nine and ten—the lines of exclamatory interjection—are identical in their tonal prosody, this being the only instance in the poem where the prosody is not varied between the lines of a couplet, as is usually expected; the rhythmic reiteration gives an emphatic punch to these lines.26 Also we note the unusual internal assonances in line thirteen (M. C. *kung tyung lwan- chung dzyek), a conscious confluence of sound that seems to suggest phonetically the “confused confluence” of the waters themselves. In the final stanza the poet delivers his judgment on the scene, declaring its effect on him and its influence over his future actions. Lu Shan’s Buddhist connections, from the time of Hui-yüan’s 慧遠 (334–416) famous monastic community, are historically more prominent than its Taoist links.27 In this context Li Po puts aside the Taoist elixirs (“rose-gem liquors”) favored by him at other sites for a Buddhist clearing away of dust: the waterfall has washed off his earthly grime, cleansing his mundane self (note the appropriateness of the literal imagery), releasing him now and, he vows, forever from the “human realm” 人間.28 Lastly, let us look back at one more technical excellence that contributes to the poem. That is the rhyme scheme. The first stanza, in eight lines, rhymes in deflected-tone, xAxAxAxA. The second stanza, a quatrain, rhymes in level-tone, BBxB. The third stanza, another quatrain, rhymes in deflected-tone, CCxC. The 26  Note that this occurs not only at a rhyme-change, but also underlines an alteration in the rhyme pattern—from the even-line rhyming of the opening octave to the AABA scheme of the quatrains following. Of course the strict prosodic guidelines that apply to “recentstyle” verse are not the issue here. The tendency of T’ang poets to play with conscious prosodic patterns for effect in “old-style” verse is well established. For Li Po in particular, see e.g., Elling O. Eide, “On Li Po,” in Perspectives on the T’ang, ed. Wright and Twitchett (New Haven: Yale Univ. Press, 1973), esp. 377–83; and Kroll, “Li Po’s Transcendent Diction,” Journal of the American Oriental Society 106.1 (1986), esp. 115–17. 27  On Hui-yüan and Lu Shan, see especially Erik Zürcher, The Buddhist Conquest of China: The Spread and Adaptation of Buddhism in Early Medieval China, 2 vols., rev. edn. (Leiden: Brill, 1972), 1: 208–11, 241. 28  The final image of the soon-to-be-abandoned “human [or mortal] realm” may remind us of the same closing to the “cyan hills” quatrain. The phrase jen-chien 人間 always has a pejorative connotation for Li Po.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1033

final stanza, a sestet (or, as it is probably better to see it, a quatrain with an attached couplet as coda, the suppletory nature of the couplet evident in the way its first word, 且, repeats in anaphoristic manner the first word of line twenty), rhymes in level-tone, DDxDxD. This untypical alternation of rhyming tonics is perhaps another means of suggesting here the same “rolling” movement we found in the poem’s flow of imagery.29 Now, we must resist the temptation to draw facile inferences from these two poems. It might seem obvious that the longer poem is a more “realistic” or “accurate” description. Certainly it is more circumstantial. In commenting on it six hundred years later, Wei Chü-an 韋居安 (fl. 1368), after calling it a composition truly incomparable for all time, adds, “But unless one has had the experience of actually looking upon the scene itself, one cannot appreciate sufficiently this poem’s perfection.”30 If we take this comment as more than an individual reader’s self-congratulatory attempt to assert special rights of ownership, we must accept that the full reach of the poem requires an act of perception beyond it. It is completed outside the words that form it. We need not invoke sages from Bologna or Paris to validate—and obfuscate—this idea: it is well—and clearly—expressed in traditional Chinese musings on the relationship between language and reality, from Confucius, Chuang Tzu, Lao Tzu, Hsün Tzu, and on. We know its influence in other areas, particularly in theories of landscape painting. But the question here is the old one of verisimilitude. How “real,” how faithful, is a verbal description as a representation of scene? Stated that simplistically, the question is probably pointless. However, let’s try to hold it in mind as we seek to complicate the issue. Li Po was not, of course, the first to write in verse about Mount Lu. But before his poems, composed most likely in 725 or 726,31 the waterfall enjoyed only 29  And here is a final curiosity, which I do not think is accidental: in every stanza the words that end non-rhyming verses are, as normal, in the tonic group opposite to that of the rhyming verses, with the sole exception of line seven’s 落 (*lak), “dropped”—a phonetic means to add an exclamation point to this word, which we have seen is the crucial verbal surprise in the Milky Way image. Now, I don’t believe that Li Po or any good poet tediously plots out such phonetic strategies as though they are a hidden code; it is rather that the particular sound of a word or the rhythmic and resonant shape of a phrase—that is, a native awareness of aural patterning—often determines a writer’s choice between lexical and grammatical options. See the appendix to this article for a full phonetic reconstruction of this poem into Middle Chinese, with a diagram of its tonal patterns. 30   Mei-chien shih-hua 梅澗詩話 (tscc), 1.3. 31  An, lpcc, 49, says 725, while Chan Ying 詹瑛, the other leading contemporary authority on Li Po, says 726 (see his Li Po shih-wen hsi-nien 李白詩文繫年 [Peking: Tso-chia, 1958], 5; also lpccc 21.1240–41). Huang Hsi-kuei 黃錫珪, however, thinks these poems

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1034

Kroll

passing mention in poetry. Yet, if we care to look, we may come upon diction and imagery from a few earlier poems that seem to prefigure some of Li Po’s phrasing—even some of his more memorable and, we would have thought, more “individualistic” lines. Thus in a “Rhapsody on Mount Lu” 廬山賦, written by Chih T’an-ti 支曇諦 (d. 411), a Central Asian monk and associate of Hui-yüan, the last couplet of the thirty lines that remain to us, preserved in the seventhcentury I-wen leichü 藝文類聚 where Li Po is sure to have seen it, reads: The Incense Burner ejects a cloud, in semblance of haze; The sweet wellspring showers foam, soaking us first off.32 Can one doubt that the opening image in Li Po’s Lu Shan quatrain (the purple haze quickened by Incense Burner) has its basic origin—though not its color—here, as also the juxtaposition of Incense Burner Peak and the waterfall? The point is not that there is an allusion here—there is not, at least not in the normal sense—but that a lexical loop has been activated, such that these lines from Chih T’an-ti’s fu have affected, in small part, how Li Po is disposed to view and represent the scene. One can locate similar sources of verbal resonance for Li Po in passages from two other Lu Shan poems, both also included in I-wen lei-chü. First, from Pao Chao 鮑昭 (?414–66): The tall crag cleaves half the sky; Long bluffs are sheer for a thousand leagues.33 which finds plausible echoes in line three of the quatrain (“straight down three thousand feet”) and in line eight of the longer poem (“half spewed out from within the clouded heavens”). Second, and more interestingly, from Chiang Yen 江淹 (444–505): Scarlet vapors fall upon embrangled thickets; White clouds rise into inscrutable faintness. were written in 756 during Li Po’s period of reclusion at Lu Shan; see his Li T’ai-po nien-p’u, fu Li T’ai-po pien-nien shih mu-lu 李太白年譜附李太白編年詩目錄 (1906; rpt., Peking: Tso-chia, 1958), 75. 32   I-wen lei-chü (Taipei: Wen-kuang, 1974; hereafter iwlc), 7.134; Ch’üan Chin wen 全晋文, 165.16b, in Ch’üan Shang-ku San-tai Ch’in Han San-kuo Liu-ch’ao wen 全上古三代秦漢三 國六朝文, ed. Yen K’o-chün (1893 edn.; hereafter cited as Yen). 33   i wlc 7.134; “Teng Lu Shan shih, erh shou,” no. 2, Lu, Hsien-Ch’in, 1282.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1035

Sitting here, I catch sight of a swelling rainbow; Huddled with lowered head, I behold a coursing star.34 Have these “scarlet vapors” 絳氣 become Li Po’s “purple haze” 紫煙? Is this “swelling rainbow” 蜿虹 the prototype of the rising white rainbow (1. 7, longer poem)? Most intriguing is the possibility (probability?) that Chiang Yen’s image of looking down from the mountain, on a meteor (a “coursing [or flowing] star” 流星) below him, may have helped trigger Li Po’s now-famous image of the waterfall as the Sky River sluicing earthward. Perhaps such understandings afford a glimpse into a corner of the poet’s creative craft. While we do not solve the mystery or grasp the magic of poetry by reconstructing the lode of texts and words possessed by a poet through his readings, I think we have that much more insight into Li Po—or any T’ang poet—for contemplating his mental storeroom, even if we cannot approach as thorough and meticulous an unpacking as J. L. Lowes did of Coleridge’s resources.35 (Among other pre-T’ang texts on Lu Shan, special note should be taken of the lengthy passage in Pao Chao’s well-known letter to his sister from the Ta-lei 大雷 riverside, which acts as something of an undertone in many later poems.36)

34   i wlc, 7.134; “Tsung kuan chün Chien-p’ing wang teng Lu Shan Hsiang-lu feng,” Lu, HsienCh’in 1557; also Wen hsüan 文選 (rpt., Taipei: Wen-chin, 1987; hereafter wh), 22.1058, an anthology that Li Po, like all T’ang poets, knew thoroughly. 35  See Lowes’ famous volume, The Road to Xanadu: A Study in the Ways of the Imagination (1927; rpt., Boston: Houghton Mifflin, 1964). Perhaps the best contribution to Li Po studies in this area is Juan Yen-yü’s 阮延瑜 chapter on the sources of Li Po’s poetic diction, in his Li Po shih-lun 李白詩論 (Taipei: Kuo-li pien-i kuan, 1986), 183–225; this, however, ignores Buddhist and Taoist influences. Many perceptive comments pertinent to this subject are made throughout Ko Ching-ch’un’s 葛景春 collection of essays, Li Po ssu-hsiang i-shu t’an-li 李白思想藝術探驪 (Cheng-chou: Chung-chou ku-chi, 1991), the most remarkable book on Li Po’s poetry to appear in the last decade. 36   Pao Ts’an-chün chi-chu 鮑參軍集注, ed. Ch’ien Chung-lien (Shanghai: Shanghai ku-chi, 1980), 2.84–85. The very fragmented translation of this passage offered—oddly enough, in verse form—by Kang-i Sun Chang in her Six Dynasties Poetry (Princeton: Princeton Univ. Press, 1986), 89, is too flawed to be recommended as anything but a partial paraphrase. Her remarks on this passage as being an accurate description of the sunset scene viewed by Pao Chao bypass the fact that it would have been impossible for Pao Chao to see from his stated vantage point all that he mentions—the description is rather a lyrical and imaginative tour de force. Chang’s other comments in this chapter on the “verisimilitude” of Pao Chao’s “microscopic,” “realistic” imagery, which she pushes very hard, are, in my opinion, overtaxed.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1036

Kroll

Intrinsically satisfying as Quellenforschung can be, we cannot be trapped by it here. It shows, though, another danger of assuming too much about the “realism” of a landscape description. At least in shih poetry there is rarely scope for attempting anything approaching a detailed examination of a scene. Thus the so-called “nature poetry” of Wang Wei 王維 (701–61), for example, consists largely of generic imagery, founded on a relatively restricted vocabulary, that presents little in the way of visual particularity.37 Other T’ang poets are more exact in this regard than Wang Wei, but for all of them the imperatives of shih composition are a push toward the suggestive arrangement of a selected few images rather than a punctilious depiction of scene. I like Kenneth Burke’s formulation, which can readily be applied to Chinese poetry: One cannot long discuss imagery… without sliding into symbolism. The poet’s images are organized with relation to one another by reason of their symbolic kinships. We shift from the image of an object to its symbolism as soon as we consider it, not in itself alone, but as a function in a texture of relationships.38 These “symbolic relationships” are what hold a poem together—if it works— and allow the words to speak with more than dictionary meanings. The skill with which the “texture of relationships” is woven, a texture that also implicates both the warp of tradition and the woof of individuality (I had almost said the hum of the past and the howl of the present), is what we continually judge as readers. Here we may quote Liu Hsieh 劉勰 (ca. 465–522), who has something relevant to say on every literary topic; at the moment the pertinent passage is from his chapter on “The Guises of Physical Things” (物色): Though the guises of physical things are diverse, in hewing one’s phrases there is need of succinctness, so that the flavor rises lightly, wafted aloft, and the feelings are ever renewed, blithely burgeoning. Since long ago, poets of one age have trod heel-to-toe with those of other times, never failing to join ranks and intermingle in effecting their own variations, to accept or recast [the words of their predecessors] in working out their own achievements. Those by whom the guises of physical things are used to the utmost while there remains no end to the feelings [suggested by those things] are those who understand with total comprehension. And so it is that mountain forests and riverbank loam really are the tacit treasury of literary thought! But if one’s wording is too abbreviated, there 37  For more on this, see Kroll, Meng Hao-jan, 99–100. 38   Attitudes toward History (1937; 2nd edn., Los Altos, Calif.: Hermes, 1959), 281–82. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1037

is something wanting; if one’s saying is too detailed, there is more than need be. Yet the reason why Ch’ü P’ing (i.e., Ch’ü Yüan) could see so penetratingly into the feelings behind the feng 風 and sao 騷 is very likely due to the assistance of the rivers and the mountains!39 This goes to the heart of our subject, embracing the triple concerns of literary economy, the use of precursors, and the inspiration of landscape. To return to Li Po and round out his views in poetry of the Lu Shan waterfall, we need only refer first to a couplet in his “Lu Mountain Ballad, Sent to ‘Emptyboat’ Lu, Aurigal Attendant,” so ably discussed by Elling Eide.40 The couplet reads: The Silver Ho is upended, vertical, across three joists of stone; Incense Burner and Sheet of Spray gaze on each other afar.41 Here is the Milky Way waterfall again, again paired with Hsiang-lu feng. The two go together once more in a couplet from the poem “To Leave Behind at Parting from Several Gentlemen in Chin-ling” in which Li Po speaks of Mount Lu as his destination, adding: There, Incense Burner’s purple haze is snuffed out, As Sheet of Spray drops down from the realm of Grand Clarity.42 The heavenly purple haze reappears, here to be extinguished in the cascade’s downward rush from the pure limits of the sky.43 It is evident that, once the 39   Wen-hsin tiao-lung chiao-ch’üan 文心雕龍斠詮, ed. Li Yüeh-kang 李曰剛 (Taipei: Kuo-li pien-i kuan, 1982), 43.1903. The “Wu-se” chapter is no. 46 in the standard sequence but Li Yüeh-kang, following Fan Wen-lan’s 范文瀾 suggestion that it was originally placed elsewhere, shifts the chapter to no. 43 of the fifty that make up the work; for Fan’s views, see his Wen-hsin tiao-lung chu 文心雕龍注 (rpt., Taipei: K’ai-ming, 1958), 10.2a. Li’s extremely thorough edition, along with that of the mainland scholar Chou Chen-fu 周振甫, Wenhsin tiao-lung chu-shih 文心雕龍注釋 (rpt., Taipei: Li-jen, 1984) (see 46.846 for the cited passage), should now rival Fan’s as the edition of choice for scholars. 40  “On Li Po,” 379–87. 41  “Lu Shan yao, chi Lu shih-yü Hsü-chou,” lpccc, 14.863; cts, 173.1773. This poem is dated to 760 by An Ch’i (p. 1580) and Chan Ying (p. 143), but to 756 by Huang Hsi-kuei (p. 77). 42  “Liu-pieh Chin-ling chu-kung,” lpccc, 15.926; cts, 174.1784. Dated to 750 by both An Ch’i (p. 914) and Chan Ying (p. 76), but to 756 by Huang Hsi-kuei (p. 75). 43  “Grand Clarity” (T’ai-ch’ing 太淸) is the celestial zone that commences at a height of forty li above the earth, according to Ko Hung (see Pao-p’u Tzu nei-p’ien chiao-shih, 15.275). In the developed Taoist cosmology of medieval times it was surmounted in ascending order Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1038

Kroll

poet hit on these images, they became symbols for him—the inescapable figures of what Lu Shan signified. If we accept the dating of An Ch’i or Chan Ying, we find that these images remained potent for Li Po for more than a quarter of a century. This picture appeals. Huang Hsi-kuei’s dating, however, would cluster all these poems together, written in one and the same year—in which case we have the poet ringing a close succession of changes on the original idea. But so far we have been talking only of Li Po’s shih poetry. If we look elsewhere in his works, we can discover another very interesting text to put beside those quoted above. This is a prose piece entitled “Autumn at Ching-t’ing, a Preface to See Off my Paternal Nephew, Tuan, Journeying to Mount Lu” 秋于 敬亭送從姪耑遊廬山序, written sometime in the mid-750s.44 Like most of Li Po’s prose writings, it is all but unknown. I quote it in its entirety, because it is charming in its sentiments and deserves to be read. But our attention should zero in on the lines about Mount Lu, roughly halfway into the piece; these will sound familiar. When I was young, an adult set me to reciting the Rhapsody of Master Vacuous 子虛賦, and I took it to my heart in admiration. When I was grown, I traveled south to Yün-meng 雲夢 and looked out upon the valiant sight of the Seven Marshes 七澤 [just as in the poem].45 Ten years then slipped tumbling by, as I lived reclusive with wine in An-lu 安陸.46 Some time later, when my paternal uncle in Chia-hsing 嘉興 was going back west, upon demotion to Ch’ang-sha 長沙, and I paid my respects to him at a farewell party held within a grove, you, Tuan, were then just a stripling, cavorting in high spirits by his side. And here you are now quite mature, richly bearing about you an impressive air. As for me, I have been in decline for a good while, but seeing you brings comfort to my heart, and the sadness that comes over me in speaking of long-ago things is changed from tears to smiles. by the regions of “Grand Culmination” (T’ai-chi 太極), “Highest Clarity” (Shang-ch’ing 上淸), and “Jade Clarity” (Yü-ch’ing 玉淸). However, “Grand Clarity” is often used in medieval poetry as a general term for high heaven. 44  An Ch’i (p. 1968) says 757, Chan Ying (p. 94) says 753, Huang Hsi-kuei (p. 93) says 754. 45  See “Tzu-hsü fu,” wh, 7.349; tr. David R. Knechtges, Wen xuan, or Selections of Refined Literature, vol. 2 (Princeton: Princeton Univ. Press, 1987), 55. 46  From about 727 to 735 An-lu, in northern Hupeh, was Li Po’s “home” with his first wife, née Hsü 許 (but “wine” in the translation is no typo for “wife”; the phrase is 酒隱). These were relatively quiet years, of a sort scarcely to be known by him again. See especially Maeno Naoaki 前野直彬, “Anriku no Ri Haku” 安陸の李白, Chūgoku koten kenkyū 中国古典研究 16 (1969), 9–22.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1039

You have just now acquainted me with your [plans for a] distant excursion, on which you’ll scale Incense Burner to the west. There where long mountains are creased athwart and the Nine Rivers draw rolling back,47 the Sheet of Spray drops down from the heavens, half contending in its flow with the Silver Ho: a leaping rainbow, darting lightning, a confluence that shoots forth into a myriad straths—it is a singular anomaly of all space and eternity! And on the summit there is a square lake and stone well, impossible to get a look at.48 I envy you this trip, as I [can merely] pet a crane and whistle long, regretting that the cinnabar ichor is not yet completed and the white dragon’s arrival is late.49 Should the man of Ch’in show the lash, I would soon take to the waters of peach blossoms.50 Alone I carry still a longheld vow, but I am embarassed to return to the mountains of renown.51 Once the appointed time is fulfilled, we’ll go hand-in-hand through the Five Marchmounts. 47  The reference is to Kiukiang 九江 (also known in T’ang times as Hsün-yang 潯陽), north of Mount Lu, where the Yangtze was traditionally said to break into nine branches. 48  These sites and the red-scaled fish that frolic in them were part of local legends about Mount Lu, according to Hui-yüan who noted in his “Yu Lu Shan chi” 遊廬山記 that the rustics of the area could only sigh wordlessly about the waters on the Hsiang-lu peak. See the fragment quoted in Liu Chün’s 劉峻 (462–521) commentary to Shih-shuo hsin-yü, sec. 10, certainly known to Li Po: Shih-shuo hsin-yü chiao-chien 世說新語校箋, ed. Hsü Chen-o (Peking: Chung-hua, 1984), 2.314; tr. Richard B. Mather, Shih-shuo Hsin-yü: A New Account of Tales of the World (Minneapolis: Univ. of Minnesota Press, 1976), 288; also in Yen, Ch’üan Chin wen, 162.7b. 49  Li Po pictures himself as an adept in hiding: he only strokes but cannot ride off on the crane that is the transcendent’s vehicle, he amuses himself with the cosmic “whistling” 嘯 associated with Sun Teng 孫登, as he waits for his elixir to be finished in his alchemical crucible and looks for the arrival of a white dragon to bear him away into transcendence as one did for the retired official, Tou Tzu-ming 竇子明, on Mount Ling-yang 陵陽山. 50  I.e., if political troubles arise in the empire (fears—or remembrances [?]—of An Lu-shan?), he will slip away as did the inhabitants of Peach Blossom Font at the time of the disturbances in the Ch’in era. The “man of Ch’in’s” lash recalls Chia I’s 賈誼 (201–169 bc) description of Ch’in Shih Huang driving the world ahead of him with a long whip; “Kuo Ch’in lun,” Yen, Ch’üan Han wen 全漢文, 16.5b. 51  He still wishes to roam, free of the world, on the famous mountains such as Lu Shan but feels unworthy now to do so. The variant reading in wyyh, 721.5b (“… but am embarrassed not to have returned to the mountain of renown”), provides a nice example of how two apparently contradictory readings may, through the application of a little thought, be seen to be ultimately no different in meaning at all—in this case proving also that the wyyh variant is an editor’s attempt to clarify a line that he simply misunderstood.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1040

Kroll

My feelings hereby send you off afar; [adding] a poem could only detract from this!52 It is plain enough that Lu Shan, at this time, is a memory for Li Po, and that the words he had earlier used in celebrating it remained with him, to be summoned forth again—some phrases verbatim—in this prose evocation. The very consistency with which he reverts to the same images is testimony to their symbolic status. If it is true that entropy ineluctably advances with time and that memory therefore increases disorder, perhaps we can regard the cultivation of symbols—whether purposeful or not—as a means to stabilize flux. An alternative and less generous interpretation might chide such recurrence as a yielding to the thoughtless cliché—except that these clichés are of the poet’s own devising. Hence the worst offense in question here is probably verbal narcissism, a crime eminently chargeable against every poet. Our modernist passion for novelty notwithstanding, we may appreciate the skilful shifts that Li Po makes in employing his symbolic keys in different environments and sense something of the potency that these images must have held for him. We might say, in short, that Li Po’s experience and consciousness of Mount Lu was permanently fixed for him by certain of the lexical decisions he made in his first attempts to frame the place in words. Is Li Po’s vision of the mountain and the waterfall, so consistently imprinted on his mind once arrived at, of a different order from that of his contemporaries? We do not have the space to be exhaustive about this. So we shall content ourselves with examples from two other, important writers. The first is Chang Chiu-ling 張九齡 (678–740), one of the most influential poets—and especially during the 730s one of the most powerful officials—of the K’aiyüan (713–42) period. Hailing from northern Kwangtung, Chang Chiu-ling is the most conspicuous eighth-century example of a southern “creole” rising to importance and acceptance in the Han heartland. His poetry has been little studied, which is a pity, for there are some rare beauties hidden in it and some intriguing features. Chang was self-conscious about and (this was unusual) proud of his southern provenance. Many of his most engaging works are those in which he celebrates the heretofore unappreciated (by northerners) glories of his native region and attempts to effect a reorientation of traditional geographic prejudice. His marvelous “Rhapsody on the Lichee” 荔枝賦 is a case

52  “Ch’iu yü Ching-t’ing, sung ts’ung-chih Tuan yu Lu Shan hsü,” lpccc, 27.1566; Ch’üan Tang wen 全唐文 (Taipei: Ta-t’ung, 1979; hereafter ctw), 349.6b–7a. Ching-t’ing is Mount Ching-t’ing, north of Hsüan-ch’eng 宣城 in southeastern Anhwei.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1041

in point.53 In other compositions set in the south, he sometimes reverses the standard responses to objects in the non-Han landscape. Thus in several poems we find the usually mournful cries of the gibbon surprisingly driving away the morose thoughts of the southern poet: for him they are welcome sounds of familiar companions.54 Chang Chiu-ling was not, in other words, shy to chase game upon his own guidance in the fields of poetry, when he deemed the pursuit warranted (similar in this respect to Li Po, an outlander from a different region). Now, Lu Shan was not in the mid-eighth century a particularly “southern” locale but, keenly sensitive as Chang Chiu-ling was to southern preferences about beauty in landscape, it seems natural that he too saw the lush zone of the waterfall as most worthy of his muse. Here are two of his poems devoted to the scene. The first is called “Gazing from the Mouth of the Lake, toward the Wellspring of Mount Lu’s Sheet of Spray” 湖口望廬山瀑布泉 (the lake referred to is of course P’eng-li 彭蠡, modern P’o-yang 鄱陽): Through ten thousand staves the flooding wellspring drops, Distance beyond distance, half perfused in purple. Flying headlong, down past trees of different sorts; 4 Spewed and scattered, coming out from the layered clouds. Sunlight illumines what seems a rainbow iris; The sky is clear, but wind and rain are heard. This numinous mountain is full of impressive guises; 8 Its air, its water, both with full-favoring auras.55 A lü-shih, this tight little poem succeeds wonderfully in thrusting us into the cascade—even though (remember the title) the poet is gazing at it from some distance. If we analyze the imagery phrase-by-phrase and line-by-line, we shall find it similar in places to the progression of imagery in the first sixteen lines of Li Po’s longer poem on the waterfall. Chang Chiu-ling, however, has here

53  “Li-chih fu,” Ch’ü-chiang Chang hsien-sheng wen-chi 曲江張先生文集 (sptk; hereafter cited as ccchs), 1.11a–12b; ctw, 283.1b–3b. 54  Some of the material in the preceding sentences is taken over from my entry on Chang in The Indiana Companion to Traditional Chinese Literature, ed. Nienhauser (Bloomington: Indiana Univ. Press, 1986), 207–9. 55  “Hu-k’ou wang Lu Shan p’u-pu ch’üan,” ccchs, 4.8a; cts, 48.590; wyyh, 264.3b. In lines two and three I follow the reading in ccchs and wyyh; in line four I follow that of wyyh (I do not think Chang would have used the word 落 twice in four lines in a lü-shih).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1042

Kroll

chosen to work within the constraints of regulated verse,56 and one must admire what he has done: the dynamism of the parallel lines three and four, for instance, and the effective tension built into the parallel lines five and six. We cannot but notice the empurpled hue, the rainbow, the wind, that we’ve already met in Li Po, although Chang Chiu-ling’s fashioning of each of these images is different from Li Po’s. Since we read Li Po’s poems first, and since he is the more renowned of the two poets, we are probably inclined to grant him the palm of priority, if there is any question here of influence. It is fairly certain that Chang Chiu-ling wrote this poem (and the next) in the spring of 727 when he passed southward through the area on his way to take up a post in Hungchou 洪州 (off the southwest extension of Lake P’eng-li).57 If An Ch’i or Chan Ying are correct in their dating of Li Po’s poems to 725 or 726, Chang Chiu-ling composed after him. However, as Li Po was scarcely known at that time and his poems were not yet circulating to any degree, there are no grounds to suppose that Chang Chiu-ling had seen the younger man’s verses on the waterfall. Of course, if Huang Hsi-kuei is correct in dating Li Po’s poems to 756, we have an entirely different picture, for Li Po can surely be assumed to have read by then the writings of Chang Chiu-ling. But I think we are probably safest in imputing no direct influence at work, in either direction. It is most likely that both poets arrived at their images independently, their fancy doubtless informed equally by the works of fourth- and fifth-century forerunners (as mentioned earlier) that were included in such standard compendia as Wen hsüan and I-wen lei-chü.58 The trick was to refocillate the inheritances from the past in the context of one’s own heedful contributions to the fund: “tradition and the individual talent” can thus be recognized for a literary inflection of Confucius’ dictum, wen ku erh chih hsin 溫故而知新, or as similar to the controlled invention of a first-rate jazz musician or of a performer of Indian ragas. 56  Some sinologists, following the practice of E. H. Schafer, now refer to all lü-shih as “double quatrains.” One has to regard this as a misnomer. A “double quatrain” ought properly to refer either to (a) a two-stanza poem made up of a pair of quatrains with different rhymes—which can never be the case with a lü-shih, or, possibly, (b) the unusual case of a lü-shih in which the tonal patterns of the first four lines are matched exactly by those of the corresponding final four lines—this exact “doubling” being contrary to the standard mirror-imaging of the two quatrains’ tonal sequences. The present poem by Chang Chiuling is, in fact, an example of (b). 57  Yang Ch’eng-tsu 楊承祖, T’ang Chang Tzu-shou hsien-sheng Chiu-ling nien-p’u 唐張子壽 先生九齡年譜 (Taipei: Shang-wu, 1980), 52. 58  Cf. this couplet from Meng Hao-jan’s “On Lake P’eng-li, Gazing Afar at Mount Lu”: “Now Incense Burner rises up in sunlight,/Its cascading waters spurting into a rainbow!” Kroll, Meng Hao-jan, 73.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1043

Here now is the second of Chang Chiu-ling’s poems, twice as long as the first, and titled “Looking Up at the Waters of the Sheet of Spray upon Going Into Mount Lu” 入廬山仰望瀑布水:

4



8



12



16

From the steep crest there is a wellspring suspended; With deafening din it comes out from the nibs of the haze. One cannot tell what is the season here or the year; One only sees there is neither dusk nor dawn. Flashing, flaring—it drops down the bice-blue scarp; Fresh and fair—glistening in the white light of day. Its spewing flow wets the moving clouds, And the swashing froth startles flying birds. A roaring of thunder—what frenzied fury! An arrow speeding—into the covert depths. Long ago, I heard of it as a mist below the mountain, And now it is there, beyond the wooded tor. In the nature of things there are anomalous outbursts; In the source of the Latent how could there be muddled promptings? Wordlessly, it placed this here and went off; Flux and transformation—who is able to understand them?59

This is the waterfall in a different light. The first ten lines of the poem show it as a natural phenomenon only—there is, this time, no hint of anything supramundane about it. But it is, nonetheless, a most unusual feature of the landscape, more strikingly conspicuous when before the poet’s eyes than he had previously imagined it. As such, it is an example of the “anomalous outbursts” (kuei-chi 詭激; note the recharged aptness here of the etymon of 激: water under pressure) occasionally produced among the myriad phenomena of the world by the generative, fecund principle of Earth (k’un-yüan 坤元60). Surely, wonders the poet, such things are not brought into being for no reason? But the purpose of the vast forces that engender them and transmute them is unspoken and is hardly fathomable by us.

59  “Ju Lu Shan, yang wang p’u-pu shui,” ccchs, 4.8b; cts, 47.573–74. 60   K’un or, as here, k’un-yüan, is the principle of latency and receptivity, symbolized by Earth, bearer of the fruits of union with the complementary principle of potency and creativity, ch’ien 乾 or ch’ien-yüan 乾元, symbolized by Heaven; the dynamism behind this ancient matching might best be indicated for us, remembering our Greek mythology, as k’un = Gaia, ch’ien = Uranus.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1044

Kroll

This is a surprising and evocative view of the natural landscape—or, rather, of an uncommon piece of that landscape. It is a view that Chang Chiu-ling reflects on elsewhere, when confronting a natural scene that astonishes him. Look, for instance, at this poem on “The Chen-yang Gorge” 湞陽峽, located on North River 北江 in central Kwangtung, about halfway between Chang’s native place of Ch’ü-chiang 曲江 and Canton: The advancing boat sidles by the crags of Yüeh, As, coved and comely, the streams of Yüeh deepen. The waters, turned dull, are first to feel autumn’s chill; 4 The mountains, unclouded, can be dim even in daytime. Their layered forests interleave all the five colors, And the facing cliffs rear upward for a thousand spans! I regret such things are produced so far off and out of the way— 8 Who can know the mind of the Shaper of Mutations?61 No student of T’ang literature, upon coming to the final couplet of this poem, can help but recall the concluding speculations of Liu Tsung-yüan 柳宗元 (773–819) in his famous “Record of the Little Hill of Stone Citadel” 小石城山記.62 Exiled to the south, the unhappy Liu puzzles that the Shaper of Creatures (an alternate name for Chang’s “Shaper of Mutations”) should dispose incomparably beautiful sites in the barbarian south, where there are no cultivated persons to appreciate them. Liu Tsung-yüan’s well-known meditation is, we can see now, clearly indebted to Chang Chiu-ling’s remarks of some seventy years earlier. (There is no question that Liu Tsung-yüan was acquainted with Chang’s works.) But we digress. It is Chang Chiu-ling’s work that involves us. In the poem above, as in the preceding one, Chang moves easily from exterior landscape to interior reflection (the regular meshing of 景 and 情), affirming—at least in these two compositions—that there must be more than a merely sensual purpose to the material form and placement of “anomalies,” even though that purpose be beyond our ken: 61  “Chen-yang hsia,” ccchs, 4.11a; cts, 48.590. For the traditional lore of the Chen-yang Gorge, see Shui ching chu shu 水經注疏 (Yangchow: Chiang-su ku-chi, 1981), 38.3186. This carefully printed edition of the text, with the valuable sub-commentary of the late-Ch’ing scholars Yang Shou-ching 楊守敬 and Hsiung Hui-chen 熊會貞, is a real boon. 62   Liu Ho-tung chi 柳河東集 (Taipei: Shih-chieh, 1963), 30.317–18; ctw, 581.16a/b. This essay has been frequently translated, but the rendering of E. H. Schafer, in his The Vermilion Bird: T’ang Images of the South (Berkeley: Univ. of California Press, 1967), 117, remains unsurpassed.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1045

A presence that disturbs me with the joy Of elevated thoughts; a sense sublime Of something far more deeply interfused … but whose dwelling, while “in the light of setting suns,/And the round ocean and the living air,/And the blue sky,” does not open itself to “the mind of man.” The contrast with Li Po’s poems on the Lu Shan waterfall is obvious. And there is, of course, no reason to expect a coincidence of views, even if literary tradition often impels a degree of congruence in matters of diction and imagery. For a final look at the now-familiar waterfall of Mount Lu, we turn to Li Hua 李華 (chin-shih 735, d. 769?), who wrote a “Rhapsody on Gazing at the Wellspring of Spray” 望瀑泉賦 during his years of voluntary reclusion after the An Lu-shan rebellion63—hence, after both Li Po and Chang Chiu-ling, and most of the other High T’ang poets who had visited Lu Shan, had written their poems on the place. We can expect this, a fu, to license a more expansive treatment of the subject than we have encountered so far. The poet sees the scene at dawn, from a boat on Lake P’eng-li:64 With no clouds in the dawnglow—no waves on the waterway, My boat is set adrift upon the deep blue of the void. Mount Lu, there, Floats up on top of the layered lake— 4 Upborne in pinnacled walls reaching to the sky. An intense luminosity in hue of thickened kohl, And, oh, the craggy stillness of its sylvan ranges! In the inscrutable faintness shaped by some divinity, 8 There leaps a curvetting wellspring upon the mountain’s spine. Its unique flow brightly glistens against the verdant joists, Whooshing halcyon-blue, a thousand fathoms—of hanging silk! A jade cord is let down from the endless sky, 12 The Silver Ho is fallen in amplest almifluence. Pounding wind and thunder—sifted frost and snow! 63  Li Hua had been forced to collaborate with the rebel government upon its seizure of Ch’ang-an in 756, a shame that seems to have haunted him till his death. 64  Varying indentations mirror varying verse-lengths in the original text; there are no rhymechanges, hence no stanza breaks, in the excerpt given here; and the rhythmic particle hsi 兮 is represented by a dash. (In translating shih-poems, where hsi does not appear, I use a dash for other purposes, but with fu I use it to “translate” only hsi.) “Lead-in” lines of twoor three-character phrases are not counted in the line numbering.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1046

Kroll

Piercing the layered clouds, downward it shoots forth: A white dragon drinking head-downward from the level lake, 16 Or it is like the initial sundering of Heaven from Earth …65 This should be enough to give the flavor of the piece. More indulgent and farreaching in its diction than our shih-poems, this section of Li Hua’s fu is a rich feast of imagery and metaphor. The poet seeks out as many suggestive ways as possible to picture the waterfall, and some of his phrasings are truly remarkable: one is struck particularly by the thousand fathoms of suspended brightblue silk (reminding one of the textile imagery inherent in the phrase “sheet of spray”), and by the sidereal cord of jade dangling from the sky, the sifted frost and snow, and especially by the marvelous upside-down white dragon. These are mixed with Li Hua’s versions of the fallen Milky Way, the stiff wind, and the close-clinging clouds. The vividness of it all might put one in mind of the time of creation, when Yin and Yang first split apart. Here the waterfall has become the motive for verbal virtuosity, acquiring in the process fresh aspects to its poetic reality. We have to do, always, with two fundamentals: the world and the text, the world and the word. On the most literal (and also perhaps the most abstract) level, we have to do with the world as text. The notion of the landscape— cosmic or terrestrial—as a grand text for decipherment is classically embodied in the myth of Fu-hsi inventing the eight trigrams of the I ching through observation of the figures (象) and exempla (法) of heaven and earth, and the markings (文) of birds and beasts. If this doesn’t quite compare with “In principio erat Verbum,” we need only look to the medieval Taoist idea of the eternal scriptures that were revealed and “spoken” in characters of blazing light at the very beginning of time, before the creation of the physical world.66 To speak in cosmogonic terms, and equally in terms of cognitive science, Word creates World. If we descend into the stream of mortal, historical time, language—no, not just language, but artfully organized language—becomes the means whereby 65   c tw, 314.10a. 66  Much more could—and should—be said about this, but it would at this stage take us too far from our focus on the T’ang. Interested readers may begin by consulting the discussion in Isabelle Robinet, Méditation taoïste (Paris: Dervy-livres, 1979), 29–44. The English translation of this book, Taoist Meditation: The Mao-shan Tradition of Great Purity, tr. Pas and Girardot (Albany: State Univ. of New York Press, 1993) is often less than accurate and should be consulted cautiously.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1047

to perpetuate a mental moment into the future or an eternity of futures. I wish to be very clear on this point, for it is another that may separate us subtly but fatally from the men of T’ang. We live in an age, indeed a global culture, that is superlatively, grandly contradictory—in both senses of the word. And however much some of us may wish it otherwise, our world is one that, with all its riot of communication media and technology, no longer requires poetry or art as an essential need. Not that some citizens of this culture do not support and appreciate the arts, or take pleasure in them and “know” a great deal about them—but most, even of these, could live without such ornaments of existence, if they had to do so. Sixty years ago Paul Valéry defined such people as the true bourgeois, not insensitive to art, “not at all the man without eyes or ears of whom we are told; he is merely the man who is not tormented by ‘what exists only in the forgetting of what exists,’ who is not harassed by a mad desire to live as though the luxuries of the mind were a necessity of life itself.”67 But for the educated elite of medieval China, self-conscious inheritors and bearers of a text-centered culture already over a millennium and a half old—or twice that old, according to tradition—it was precisely the command and continual exercise of the arts of language, the knowing manipulation of lexicon, grammar, and rhythm, that represented the indispensable and defining core of life. For the hyperliterate minority who produced the works we study, it was everything else that was peripheral. That is to say, the cardinal view was of a world seen through words—perhaps, a world seen as words. (And no wonder, then, that exile to the more backward precincts of the empire, promising a dearth, at best, of similarly grounded companions, was such a fearsome punishment: the sense of isolation must have been inconceivably withering.) So what is it that we touch when handling the images and scenes of T’ang landscapes? We work with our sympathetic (we hope accurate) understandings, our imaginative (we hope honest) recreations of the visions of T’ang writers. The pivotal word being “visions,” for these writings are themselves imaginative recreations, not meticulous descriptions. And here is where the importance of tradition inserts itself, the poets themselves being conditioned and stimulated, to varying degrees, by the verbalized visions of previous writers. We deal with haze over the mountain, mist from the waterfall, with visions—emotional, personal, spiritual, intangible—kindled and quickened by the landscapes but no less by thoughts and crafted deployments of the words in which other visions of landscapes have been expressed or might yet be expressed. We see not with the eyes of sight but through the imposition of 67  From “The Necessity of Poetry,” in Paul Valéry, The Art of Poetry, tr. Folliot (rpt., Princeton: Princeton Univ. Press, 1985), 219.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1048

Kroll

the eye of the mind. Farther this way lies the dissolution of the skandhas, if not the deconstruction of the text. Yet when we return to particular poems about the landscape, we can be quite certain that most of these considerations will dissolve steadily as we read, like the salt rimming a margarita glass. We so want to believe that what is described in a text is what was there, before the writer’s eyes. Much or some of it might have been. But listen to Wallace Stevens, a poet who was preeminently concerned with the exchange between noumenon and phenomenon: Poetry is the imagination of life. A poem is a particular of life thought of for so long that one’s thought has become an inseparable part of it or a particular of life so intensely felt that the feeling has entered into it. When, therefore, we say that the world is a compact of real things so like the unreal things of the imagination that they are indistinguishable from one another and when, by way of illustration, we cite, say, the blue sky, we can be sure that the thing cited is always something that, whether by thinking or feeling, has become a part of our vital experience of life, even though we are not aware of it. It is easy to suppose that few people realize on that occasion, which comes to all of us, when we look at the blue sky for the first time, that is to say: not merely see it, but look at it and experience it and for the first time have a sense that we live in the center of a physical poetry, a geography that would be intolerable except for the non-geography that exists there—few people realize that they are looking at the world of their own thoughts and the world of their own feelings.68 Remember Liu Hsieh: “Those by whom the guises of physical things are used to the utmost, while there remains no end to the feelings suggested by those things, are those with total comprehension.” Thus, in the reality of literature the Delectable Mountains are as solid as Mount Lu, the starry pathways of Highest Clarity as tangible as the fissures of the Grand Canyon. An uncommon text is needed to reinforce the mood. If our subject had to do with American literature, the choice would be Herman Melville’s dedication for his psychological novel, Pierre: or, The Ambiguities. Inscribed in three grateful and glorifying paragraphs to Mount Greylock in the Berkshires, it is the only book dedication I know of that is addressed to a mountain. Equally unusual, for the High T’ang period, would be a text in which a mountain itself is made to speak. There is such a text. It was written, as one might have guessed, by 68  From “The Figure of the Youth as a Virile Poet,” in The Necessary Angel: Essays on Reality and the Imagination (New York: Vintage, 1951), 65–66.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1049

Li Po. This is his “Missive on Behalf of Mount Shou, in Answer to the Despatch of Junior Repositor Meng” 代壽山答孟少府移文書. Here Li Po speaks in the persona of Shou Shan, one of the lesser mountains in the An-lu area. Sometime during his An-lu years (ca. 727–35), Li Po kept a retreat on this mountain. The “Missive” is his response to a local official (shao-fu is an alternate designation for a wei 尉, “Pacificator,” at the district [縣] level) who had apparently written a chiding—one supposes, facetious—despatch (移文) in which he derogated Shou Shan as a mountain not sublime enough to shelter recluses who should more fitly be offering themselves for official service. The first two-thirds of the answering “Missive” is Mount Shou’s defense of itself and its self-definition. This is the part that interests us. The final third of the composition turns to the mountain’s introduction of its kindred spirit and denizen, Li Po, featuring its extolment of his untrammeled virtues; this portion of the letter, in which Li Po himself is the focus, has been translated by Victor Mair.69 In the first twothirds of this extremely witty, and at least half-serious, piece, Mount Shou (as recorded by Li Po) tries to clarify the proper relationship of mountain to state. Being literate and, as will be seen, obviously versed in the classics, the mountain writes in good parallel prose but with a much more pleasing and flexible bent of mind than its only possible rival, K’ung Chih-kuei’s 孔稚珪 (447–501) “Despatch from North Mountain” 北山移文.70 This is what it has to say: The Lesser Mount of Longevity 小壽山 in Huai-nan commissions with deference a pair of golden-garbed cranes from its eastern peak to bear a flying-cloud missive [wrapped] in damask to His Majesty, Lord Meng of Wei-yang,71 reading: I, your servant, enfold the pneumas of the Great Ball-of-Earth,72 produced in the midst of Whelming Waste.73 Linking the terrains allotted

69  See his “Li Po’s Letters of Political Patronage,” Harvard Journal of Asiatic Studies 44 (1984), 141–42. 70  For which, see the translation of James Robert Hightower, “Some Characteristics of Parallel Prose,” in Studia Serica Bernhard Karlgren Dedicata (Copenhagen: Ejnar Munksgaard, 1959), 70–74. 71  Wei-yang 維揚 is the archaic designation of Yangchow 揚州, presumably Lord Meng’s native place. 72   大塊 = all the earth, “the great glebe (i.e., clodded soil of earth),” as a symbol of Nature itself. Cf. Chuang Tzu chi-shih 莊子集釋, ed. Kuo Ch’ing-fan (rpt., Peking: Chung-hua, 1961; hereafter ctcs), 2.22, 6.110, 6.119. 73   洪荒, the era of universal chaos, when the earth was a vast undifferentiated desert.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1050

Kroll

under the stations Wing and Axletree,74 I dominate the far contours of the realms of Ching 荊 and Heng 衡.75 Distended, diffuse, for a myriad ages, reaching so remotely to the Starry Ho; propped against Heaven’s rainbow as I knot my peaks together, I lean on the Dipper and Polestar in laying out my palisades. Quite adept at gathering and drawing in auroral clouds and rain, or giving secluded residence to numinous transcendents, I generate pearls luminous as that of the Marklord Sui,76 produce gems bright-lit as that of Mister Pien.77 Exhausting all the beauty of space and eternity, I deplete the wonders of the Shaper of Mutations. Right along with K’un-lun 崑崙, I am its high peer in act, as with Lang-feng 閬風 I touch bounds.78 So how could I bear to be arrayed with Mounts Wu 巫, Lu 廬, [T’ien-]t’ai 天台, and Huo 霍 of the mortal realm?79 I sat yesterday with the mountain-dweller Li Po, looking respectfully at your despatch, in which your servant is dispraised for his many wonders, in which your servant is vilipended for his special qualities, while you speak fulsomely of the beauty of the Three Mountains and Five Marchmounts.80 You refer to me as just a little mountain without name (無名) and without potency (無德) enough to be esteemed thereby. But, musing on these words [of yours], how grandly, how exceedingly preposterous they are! How is it, sir, that you have not heard?—“Without 74  “Wing” 翼 and “Axletree” 軫 are the two lunar lodgings—made up of stars in Crater and Corvus, respectively—that together controlled the fate of the geographical area of the old state of Ch’u (i.e., Hunan and southward). See E. H. Schafer, Pacing the Void: T’ang Approaches to the Stars (Berkely: Univ. of California Press, 1977), 76–77. 75  I.e., Ching-chou and Heng-chou; or, according to another view, Mounts Ching and Heng. 76  Marklord Sui 隋侯 once healed a wounded snake, who later repaid him with a luminous pearl from the Kiang. See Huai-nan Hung-lieh chi-chieh 淮南鴻烈集解, ed. Liu Wen-tien (Peking: Chung-hua, 1989), 6.198, commentary of Kao Yu 高誘 (fl. 212). 77  Pien Ho 卞和 presented a gemstone, still in its original matrix, to two successive rulers who could not recognize its worth and had, each, one of his legs sawn off in punishment; a third ruler finally had the stone brought out and polished. See ibid., paired there, likewise, with Marklord Sui. Various sources identify the three rulers differently: cf. Kao Yu’s commentary; Han Fei Tzu 韓非子(Pai-tzu ch’üan-shu), sec. 13, 4.4a; and Hou Han shu 後漢書 (Peking: Chung-hua, 1965), 80B.2633, Li Hsien’s 李賢 (651–84) commentary. 78  Cosmic Mount K’un-lun in the west; Lang-feng (“Wind-on-the-Fells”) is one of its three heights. 79  Four great and fabled mountains within China, which, however, hardly bear comparison with K’un-lun, Shou Shan’s peer. Note again Li Po’s dismissive use of the term 人間. 80  The “Three Mountains” 三山 are the paradise isles of P’eng-lai 蓬萊, Fang-chang 方丈, and Ying-chou 瀛洲, in the eastern sea.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1051

name, it is Inceptrix of Heaven and Earth;/With name, it is Mother of the Myriad Things”!81 Suppose that you were to order an ascent for the feng-sacrifice, the incense-rites, and libations; how is that anything but a mocking of the Great Tao? Even if one could [in so doing] expend his people and waste creatures, butchering in the shambles to deliver to sacrifice, to scorch and root out plant and tree,82 or to incise and carve in metal and stone so that tableau and testament be indited—still it is not much of an honor! What’s more, there is Student Chuang himself, the Unconstrained One, always in possession of elaborate arguments, who had it that the merest quail is not envious of the p’eng-bird, and that an autumn hair can be equal to T’ai Shan.83 To speak from this premise, what difference is there between the “little” and the “great”? Again, to place blame on my several hills for hoarding treasures useful to the state or secluding men worthy for state service—which then causes our sovereign to post placards by wayside or burn out whole mountains84 in a search and quest that is ultimately unsuccessful—is not intelligent talk. You see, when an august king ascends to the pinnacle of power, auspicious phenomena are brought forth in all their splendor: grapes and halcyon feathers are brought in as tribute,85 the Ho diagram and Lo document come out as responsive signs,86 Heaven’s mainstays are adjusted and men of worth are gathered in,87 the chasms of the moon are 81  Quoting from the first chapter of the Lao Tzu. 82  I.e., to prepare the space for the sacrificial altar. 83  See ctcs, 1.14, 2.39. 84  Sun Hui 孫惠, in the early third century, sought to protect himself from court intrigue by hiding out in a mountain retreat. The Prince of Tung-hai 東海王 forced him out of retirement by posting “wanted” placards for him. Chin shu 晋書 (Peking: Chung-hua, 1974), 71.1883. A century earlier Juan Yü 阮瑀 sought to avoid a summons to court from Ts’ao Ts’ao 曹操 by the same means, but the latter had Juan’s mountain burnt out, in order to acquire him. San-kuo chih, 21.600, P’ei Sung-chih’s 裴松之 (372–451) commentary, quoting the third-century Wen-shih chuan 文士傳. The prototype of burning out a mountain to force a recluse into official service is Chin Wen Kung’s 晋文公 seventh-century bc firing of Mien Shan 綿山 to find Chieh-tzu T’ui 介子推; this legend, recorded in the Hsin hsü 新序 (Han Wei ts’ung-shu), 7.14b–15a, of Liu Hsiang 劉向 (77–6 bc) ends with T’ui refusing to come out of the conflagration and being burned alive. The unhappy ending is not part of the story in the Tso chuan and Shih chi accounts of Chieh-tzu T’ui. 85  I.e., exotic items from Central Asia and the distant south. 86  Ho-t’u 河圖 and Lo-shu 洛書, as tokens of a Sage’s advent. 87  “Heaven’s mainstays” 天網, which bind together the network of stars in the sky, and, therefore, the Son of Heaven’s lines of authority on earth.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1052

Kroll

depleted to present those best suited for official duties.88 Heaven does not then hide away its treasures, Earth does not hoard its prodigies; the prevailing wind then overawes the hundred-fold barbarians, and that is a springtime that nurtures the myriad things. There is nothing then standing outside the Royal Way, so how could it be that the finest men of worth or the rarest gems would lie hidden and unrevealed in inaccessible caves? This business of “posting placards by wayside” and “burning out whole mountains” is, then, but a matter of the Kingly One’s moral potency not being ample enough [to attract outstanding men to him naturally]. Long ago the great worth of T’ai-kung 太公 took refuge by the waters of the Wei River 渭川 and the illustrious potency of Fu Yüeh 伏說 was closed away amid the cliffs of Yü 虞 and Kuo 虢, eventually able to take form in enigmatic omen or be sensed in dreaming vision.89 These were obscure concurrences90 of the Way of Heaven, and certainly not belaborings in quest and pursuit. As a result, the one cast aside his fishing-line and advanced to the nation’s standard, the other gave over pounding earth and became State Minister—the one assisting King Wen of the Chou, the other commending Wu Ting of the Shang. To speak of it in short: how can the mountains, for their part, be at fault? We know, after all, that the cliffs and caves are realms for nurturing men of worth, that the groves and springs are not spots for keeping treasures hidden. In such case, how indeed could your servant’s hills shrug aside state and homefold?91 Thus, even the mountains are political entities, but very much on their own terms! And only Li Po could think of speaking as a mountain (speaking for 88  “Chasms of the moon” 月竁, both the cold lunar haunts and—on earth—the lands of the Far West. 89  T’ai-kung is Lü Shang 呂尙 (originally Chiang Tzu-ya 姜子牙), found fishing on the north bank of the Wei River by Wen Wang 文王 of Chou, after the latter had obtained this omen on going out to hunt: “What is caught will not be dragon or wyverne, neither tiger nor bear; what is caught will be the prop of an overlord king.” Shih chi 史記 (Peking: Chunghua, 1972), 32.1477–78. Lü Shang became, of course, Wen Wang’s chief adviser, with the title T’ai-kung. Fu Yüeh, who became the prime minister of Wu Ting during the Shang, was said to have been discovered among the convict-laborers pounding earth to make a road through the mountainous territory of Yü and Kuo, where Fu had been living in reclusion. Before finding him, Wu Ting had had a dream in which he saw himself obtaining the support of a sage named Yüeh. Shih chi, 3.102. 90   闇合, or “in-the-dark agreements,” i.e., fated coincidence. 91  “Tai Shou Shan ta Meng Shao-fu i-wen shu,” lpccc, 26.1521–24; ctw, 348.16b–18a. An Ch’i (p. 1851) and Chan Ying (p. 9) date the piece to 727; Huang Hsi-kuei (p. 92) dates it to 731. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1053

himself). On other occasions he could speak with T’ai-po Shan 太白山, a mountain that shared his byname and therefore could be regarded as an exalted double of himself.92 The “cultures of landscape” are truly manifold. One is tempted at last to say that the landscape is not so much a topic in T’ang literature as it is, in Stevens’ terms, the “non-geography” that exists amidst the poetry of physical geography. “A poet’s words,” as he says elsewhere, “are of things that do not exist without the words.”93 But the last word should be reserved for a T’ang writer. Let us close with a fu by Wu Yün 吳筠 (d. 778), a poet and Taoist adept who was Li Po’s contemporary. Wu Yün has lately emerged from the borderlands of literary history, thanks to the attention of E. H. Schafer who presented in two long articles most of Wu Yün’s shih-poems on Taoist themes.94 Wu Yün’s fu, however, remain untranslated, and it is with one of these, his “Rhapsody on Roosting in the Cliffs” 巖棲賦 that I wish to end. It will give us another and powerful vision of “the assistance of rivers and mountains” spoken of by Liu Hsieh. The title of the piece is taken from the opening line of the preface to Hsieh Ling-yün’s 謝靈運 (385–433) famous “Rhapsody on Dwelling in the Mountains” 山居賦, in which he says, “Of old, to dwell nesting in a cave was called ‘roosting in the cliffs’; to dwell under ridge-pole and roof in the mountains was called ‘dwelling in the mountains.’ ”95 Here is a full exercise in the transformative virtues of the landscape. Wu Yün locates or creates in the natural world a transcendent immanence. Hence, instead of breaking free of this earth into a higher plane of being (which he is perfectly capable of doing in other compositions), he rejoices to lose himself—that is, his “self”—in the mountainscape. Or, to put it the other way round, as he does, he finds in the end his whole self (“become[s] self-possessed”), harmonized in balanced and tenuous quietude with the rarefied non-materiality of a non-geography where he can hide his light (“sheathe my essence”), engulfed in the dark backwaters or “eddies of reclusion.”

92  See Kroll, “Li Po’s Transcendent Diction,” 113–17. 93  “The Noble Rider and the Sound of Words,” in The Necessary Angel, 32. 94  “Wu Yün’s ‘Cantos on Pacing the Void,’ ” Harvard Journal of Asiatic Studies 41 (1981), 377– 415, and “Wu Yün’s Stanzas on Saunters to Sylphdom,’ ” Monumenta Serica 35 (1981–83), 309–45. Schafer’s first article includes a brief sketch of Wu Yün’s life, a few points of which are inaccurate. Wu Yün is traditionally said to have been responsible for Li Po’s summons to court in 742, but see Yü Hsien-hao’s 郁賢皓 1981 article, “Wu Yün chien Li Po shuo pien-i” 吳荺荐李白說辨疑, rpt. in his Li Po ts’ung-k’ao 李白叢考 (Sian: Shensi jen-min, 1983), 65–78, for a scrupulous examination of the data pertaining to this and other events in Wu’s life. 95   Sung shu 宋書 (Peking: Chung-hua, 1974), 67.1754; Yen, Ch’üan Sung wen 全宋文 31.1a. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1054

Kroll

Rhapsody on Roosting in the Cliffs96

Sensible of the instructions descended from the Mystic Saint,97 My awakening now draws close and the wish to be known dissipates.98 Words may be cast off as one becomes attached to silence; 4 One’s person is meet to curl away, no need to be slackened.99 I am partial to the grove where the pippit makes its nest, Lighting on a single bough and still with room to spare.100 What pleases my inbred nature is impossible to oppose; 8 Taking refuge in this mountain, here I have built a hut.101 Indeed, “roosting in indolence,” I am well contented;102 This is exactly the holy ground of the “footloose and free”!103

Observe here The winding and lofty tors, cross-angled and lifted valleys; 12 The welling and gushing springs, thick-set and enveloping trees. Behind, exaltedly upborne, in their twisting and looping round; Before, erratically arrayed, upthrusting and huddled low. I pursue the summer coolness of shaded straths, 16 Rest in the winter warmth of sun-lit scarps.104 I admire sturdy composure in pine-tree and bamboo-culm, Enjoy the especial fragrances of mum and thoroughwort. Attenuated pipings purify the ear,105 96  “Yen-ch’i fu,” Tsung-hsüan hsien-sheng wen-chi 宗玄先生文集 (hy 1045), 1.1a–2a; ctw, 925.3b–4a; wyyh, 98.2a–3a. 97  The “Mystic Saint” 玄聖 is Lao Tzu. 98  With his “awakening” 悟 to reality, the poet’s “wish to be known” 名 in the world dissolves. 99  That is, to “curl away” 卷 (or be rolled up) in obscurity, rather than stretch out or “be slackened” 舒 in the sight of others. 100  Cf. ctcs, 1.13: “The pippit 鹪鵪 nests in the deep grove, needing no more than a single bough.” For a full-scale effusion on this theme, see Chang Hua’s 張華 (232–300) “Chiaoliao fu,” in Yen, Ch’üan Chin wen, 58.3a/b, the preface of which begins, “The pippit is a little bird. Born amidst the wormwood and weeds, growing up beneath the hedgerows, hovering and perching in the realm of the ordinary, its system for keeping alive the living (see ctcs, 6.115) is complete.” 101  Lexical overtones from T’ao Ch’ien are strong in this couplet. 102  Cf. Shih 138, ll. 1–2: “By a slat-wood gate,/One can roost in indolence.” 103  “Footloose and free” is hsiao-yao 逍遙 from chapter 1 of Chuang Tzu. 104  Both lines of the couplet furnish a nice example of yin-yang complementarity. 105  The “attenuated pipings” 虛籟 are the finely-tuned (to normal ears, indaudible) strains of the “pipes of heaven” 天籟; ctcs, 2.22–24. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang

1055

20

And languid clouds bedazzle the eye. I depend on the seagoing crane to advise me of nightfall,106 Trust in the painted stork to let me know of sunup.107 Apprehensions are subdued in a want of excitement, 24 And spirit is calmed in a lessening of desires.108

28



32



At this I sing of “fulfilling my joy” like the man of the Odes,109 Chant of the “estimable retreat” in the great Change.110 Afar I drift out of the troubling hazards of the vulgar, Dissolving and sapping the profit and loss of acclaim. Treading the open route of the ultramundane,111 Is truly to be in avoidance of dismay and misgiving. Having taken up the darkness, to put shadow in repose, Hereon I proceed no longer but instead blot out my tracks. Though I make no plan at all for within this sphere, I have in fact a lasting program for beyond the world.112

106  Cf ll. 9–10 from the second of Pao Chao’s two poems called “Autumn Night”: “In the fair light of dawn I see the clouded peaks;/In the breezy night, hear the seagoing cranes.” Lu, Hsien-Ch’in, 1308. 107  Cf. from the “Great Summons” 大招: “Painted storks and swan-geese flock in the morning,/Mingling with adjutant storks, with gray cranes.” Ch’u tz’u pu-chu 楚辭補注 (Peking: Chung-hua, 1983) 10.224. 108  Cf. “Show the plain and embrace the unhewn,/Diminish the personal and lessen desires.” Lao Tzu, chapter 19. 109  “Fulfilling my joy” is k’ao p’an 考槃, the first two words and (therefore) traditional title of Shih 56, a song celebrating the happiness of a hermit. The Mao interpretation, as standardized for the T’ang period by K’ung Ying-ta 孔穎達 (574–648), glosses 考槃 as 成樂, which is clearly how Wu Yün uses it here; Mao Shih cheng-i 毛詩正義, 3.53c, in Shih-sanching chu-shu, fu chiao-k’an chi 十三經注疏附校勘記 (Peking: Chung-hua, 1979; hereafter ssccs). 110  See Chou I cheng-i 周易正義 (ssccs), 4.36b, commentary of Han K’ang-po 韓康伯 (d. ca. 385) and subcommentary of K’ung Ying-ta anent the phrase 嘉遯 in the fifth line of hexagram no. 33 (遯), signifying a praiseworthy retreat that is owing to the rectification of one’s will. 111  “The ultramundane” is fang-wai 方外. See esp. ctcs, 6.121 for Confucius’ comment on those who “roam beyond the mundane.” 112  Cf ll. 79–80 from Lu Chi’s 陸機 (261–303) “Rhapsody on Sighing Over the Departed” (“T’an shih fu,” wh, 16.727): “Quintessence drifts off, spirit founders,/And suddenly I am beyond the world.” But, in contrast to Lu Chi, Wu Yün plans to escape the mortal realm. wyyh reads “… a good program for getting along in the world” (… 世途之良策). Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1056

Kroll

What is discarded by others is what I bind up, 36 Hence I dwell in restraint with perpetual aptness. 40

Behold the unseen, thereby gathering in vision; Listen to the soundless, thereby dismissing hearing.113 For congenial blending, one is not restricted to fine-flavored wine; For glee-making, what need is there of strings and paulownia-wood? I burn the purest incense and refine my ch’i;114 Undoing book-tags of jade, I strike against ignorance.115 Engaged to banish hindrances in “splendid enhancement,”116

113  Reading 黜聰 with wyyh and ctw, to preserve the rhyme with 桐 in line 40; the hy text reads 逃默 “escaping to silence,” in which 默 would be an off-rhyme with 適 of line 36. Cf. ll. 175–76 of “Far Roaming” (“Yüan yu,” Ch’u tz’u pu-chu, 5.174): “As I beheld the flickering instant, there was nothing to be seen—/Giving ear to the humming hush, there was nothing to be heard”; see Kroll, “On ‘Far Roaming,’ ” Journal of the American Oriental Society 116.4 (1996), 663, 669, also noting ctcs, 11.173: “Behold nothing; hear nothing. Hold spirit close and so be still.” Cf. also line 15 from Lu Chi’s “Rhapsody on Literature” (“Wen fu,” wh, 17.763): “To begin with, one gathers in vision, pulls back from listening.” 114  Cf. “I resent that the purest incense does not linger long,” from Hsieh Ling-yün’s “Shan chü fu,” Ch’üan Sung wen, 31.4b. The poet “refines his ch’i” 鍊氣, in hope of spiritualizing his corporeal form and effecting translation to Higher Realms. 115  The jade book-tags 玉檢 are on the scroll-wraps of the few texts that the poet reads in his reclusion—evidently scriptures of the Perfected (see line 44). The “ignorance” (meng 蒙), or “ignorants,” that he combats thereby recalls the “childish ignorance” of I ching hexagram no. 4, which moreover is formed from the trigram kan (mountain) above the trigram k’an (water), the traditional judgment (t’uan 彖) on which interprets the image as “a perilous place below the mountain; it being perilous, one stops”; Chou I cheng-i (ssccs), 1.8b. The allusion is thus more than minimally suggestive. 116  Cf. ctcs, 12.198, describing the “spiritual one(s)” 神人: “The higher spirits are borne up by light, as their physical form is snuffed into nothing: this we call ‘splendid enhancement’ (chao k’uang 照曠). They bring their destiny to its ultimate point, taking their true being to the utmost. In the delight of heaven and earth, the myriad concerns [of the world] disintegrate into nothing, and all things return to their true being: this we call ‘coalescing in darkness’ (hun ming 混冥).” Also relevant here is a passage (again) from Hsieh Ling-yün, using the phrase chao k’uang in the concluding lines of a poem, “Fu ch’un chu” (wh, 26.1240), where he celebrates a newly gained determination to repudiate official service and indulge in “far roaming”: “This long-held intent is ever more discovered and declared;/As the myriad concerns all waste and wither away./Now that my heart has enfolded ‘splendid enhancement,’/Seeing material things as extraneous, I am but a [sleeping] dragon or [inconspicuous] inchworm.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Lexical Landscapes and Textual Mountains in the High T ’ ang 44



1057

I hope to draw near the Perfected in empathetic connection.117 Trapping the active images contained in the Great Void, I ensnare the non-emptiness of the Wonderfully Useful.118

For audience at court there is too little time; 48 For the culling of herbs there is plenty of leisure. My physical frame remains supplied by “expelling and receiving,”119 While purposeful mind is shattered in “escorting and welcoming.”120 I recognize that the Tao is without rise or fall, 52 Although phenomena have their presence and decay. Hence I draw from the root of life, making life persistent; Embody transformation’s ideal, no longer transforming.121 Serene and sedate, sundered from the dust, 56 Who is there to be neighbor to me? 117  The “Perfected” 眞 are the high entities whose home is in the empyreal reaches of starry space. The phrase “draw near” suffices for both the hy reading yen 延 (taken causatively) and the wyyh and ctw reading chin 近 (whether causative or active). “Empathetic connection” is kan-t’ung 感通, the resonant relationship holding between spiritual beings and humans of like attitudes. In T’ang writings both religious and secular it indicates the contact established between the two realms through a spirit’s visitation in response to a human’s devout behavior. 118   h y gives 筌 (“to trap”) and 覈 (“to examine”) as the opening word of ll. 45 and 46, respectively; wyyh and ctw read 鑒 (“to mirror”) and 覆 (“to cover”). Considerations of synonymy suggest that the correct pairing should be either “to mirror//to examine” or “to trap//to cover.” I have emended with the latter reading. In each line the five-word phrase functioning as object (太虛之有象//妙用之非空) is a neatly fashioned paradox on the concept of presence-in-absence, absence-in-presence. 119  “Expelling and receiving” 吐納 refers to breath-control practices of channeling one’s ch’i, during which one expels stale breaths and ingests new “vital” breaths. See, inter alia, ctcs, 15.137, describing the “adepts of guiding and conducting [ch’i]” who are men that “nourish their physical form.” 120  Reading 逬 (“shattered”) with wyyh and ctw, in place of hy’s 屛 (“set aside”). “Escorting and welcoming” 將迓 refers to accepting equably the coming and going that forms the flux of existence, as elaborated in the succeeding couplet. Cf ctcs, 6.115: “One who kills off living, never dies; one who keeps alive the living, is never born. As to phenomena, there is nothing he does not escort (將), nothing he does not accept (迎), nothing he does not bring to ruin, nothing he does not bring to fulfillment.” 121  He has attained to the adept’s equivalent of “the still point of the turning world.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1058

Kroll

With my tracks reaching far, friends and familiars spread ever wider; With mundane machinations forgotten, birds and beasts may be tamed. My consonance agrees in no respect with the present age, 60 As my heart inclines persistently toward the men of olden times. I exalt the imprints left behind, deeply distant, by Ch’ao and Yu;122 Celebrate the suasive incorruption, mutely hushed, of Hsi and Nung.123 I take for tutors Huang and Lao with their arcane mysteries,124 64 Look to Sung and Ch’iao as mates in the realization of the Way.125 Abashed at my lack of merit in keeping hold of material things, It is right to “advance in solitude the goodness in my own person.”126 Simply to gain the best fortune, to become self-possessed, 68 I venture now to sheathe my essence in the eddies of reclusion. Appendix Li Po’s 望廬山瀑布二首,其一 Middle Chinese reconstruction, tonal pattern (O = level, X = deflected), and rhyme scheme: 西

登

香

爐

峯

sei

těng

hyang

lu

byong

南

見

瀑

布

水

nam

ken-

bau-

pu-

shwi:

挂

流

三

百

丈

kwǎi-

lyou

sam

păk

dyang

OOOOO OXXXX

rA

XOOXO

122  The ancient worthies, Ch’ao[-fu] 巢父 and [Hsü] Yu 許由, both of whom chose to remain aloof from and uncontaminated by society. 123  That is, [Fu-]hsi 伏羲 and [Shen-]nung 神農, the culture-heroes who, in the mists of prehistory, first instructed the people in the ways of hunting, fishing, and pasturage (Fu-hsi), and husbandry and basic pharmacology (Shen-nung). For ll. 61–62 wyyh and ctw read “I exalt the imprints left behind by [Hsü] Yu and the [Four] Grayheads (i.e., the recluses of Mount Shang; see Han shu 漢書 [Peking: Chung-hua, 1975), 72.3056), “Celebrate the suasive incorruption of [Fu-]hsi and [Shen-]nung.” 124  The Yellow Thearch and Lao Tzu. 125  That is, [Ch’ih-]sung [Tzu] 赤松子 and [Wang-tzu] Ch’iao 王子喬, whose fame as a pair of transcendents traces back at least to the ‘Yüan yu,” ll. 23–24, 54, 61–74; see Kroll, “On ‘Far Roaming.’ ”. 126  Cf. Meng Tzu, 7A/9: “Destitute, they (i.e., the men of old) advanced in solitude the goodness in their own person; successful, they advanced concurrently the goodness of the whole world.” Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1059

Lexical Landscapes and Textual Mountains in the High T ’ ang 4

8

12

16

20

22

噴

壑

數

十

里

p’en

hak

sryu:

shěp

li:

欻

如

飛

電

來

hwět

nho

pěi

den-

lai

隱

若

白

虹

起

OXXXX XOOXO

iěn:

nhak

bǎk

ghung

k’i:

初

驚

河

漢

落

tr’yo

kyǎng

gha

han-

lak

半

灑

雲

天

裡

pan-

sei:

ywěn

t’en

li:

仰

觀

勢

轉

雄

ngyǎng:

kwan

shei-

tywen:

yung

壯

哉

造

化

功

XXXOX

tsai

ts’au-

hwǎ-

kung

海

風

吹

不

斷

hai:

pyung

ch’wiě

pyeu:

dwan-

江

月

照

還

空

kaung

ngywǎt

cheu-

ghwǎn

k’ung

空

中

亂

潀

射

k’ung

tyung

lwan-

chung

dzyek

左

右

洗

靑

壁

you-

sei:

ts’yeng

pek

飛

珠

散

輕

霞

pěi

chu

san-

k’yeng

ghǎ

流

沫

沸

穹

石

lyou

mat

pěi-

k’yung

zhek

而

我

樂

名

山

rA

OOOXX

tryang-

tsa:

rA

XXOOX

rA

XOXXO

rB

XOXXO

rB

XOOXX OXXOO

rB

OOXOX

rC

XXXOX

rC

OOXOO

nhi

nga:

lak

myeng

srǎn

對

之

心

益

閑

twai-

chi

syěm

iek

ghǎn

無

論

漱

瓊

液

myu

lywin-

sou-

gyweng

yek

且

得

洗

塵

顏

ts’ya:

těk

sei:

dyin

ngǎn

且

諧

宿

所

好

OXXOX

rC

OXXOO

rD

XOOXO

rD

OXXOX

ts’ya:

ghǎi

syuk

sryo:

ghau-

永

願

辭

人

間

ywǎng:

nghwǎn-

zi

nhin

kǎn

XXXOO

rD

XOXXX XXOOO

rD

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical and Literary Theory in the Mid-Eighth Century David McMullen The outbreak of the An Lu-shan Rebellion in the late autumn of 755 is recognized as a turning point in the history of T’ang thought. The sudden disruption of over a century of internal peace stimulated the interest of intellectuals in social and political problems and forced them to reexamine their own tradition. A fresh critical spirit grew up. New emphasis was given to history and literature as disciplines relevant to the contemporary situation. Most modern accounts suggest, however, that this reawakening was a gradual process and portray the two decades after the rebellion as a time of delayed rather than immediate reaction, a period of preparation for the fuller flowering of intellectual life that occurred during the late eighth and early ninth centuries.1 Only recently has it been pointed out that a lot of thinking and writing went on among those who actually lived through the rebellion.2 For the purposes of this chapter, I have studied five of the leading prose writers who were active during the rebellion period—Li Hua (c. 710–c. 767),3 Hsiao

Source: “Historical and Literary Theory in the Mid-Eighth Century,” in Arthur F. Wright and Denis Twitchett (eds.), Perspectives on the T’ang, New Haven: Yale University Press, 1973, 307–347. 1  E.g., Aoki Masaru, Shina bungaku shisōshi (Tokyo, 1943), p. 98; Lo Ken-tse, Chung-kuo wenhsüeh p’i-p’ing shih (Shanghai, 1957), vol. 1, pp. 122–27; Ch’ien Tung-fu, T’ang Sung ku-wen yüntung (Shanghai, 1962), pp. 11–14. All these accounts treat this period as one of preparation for the ninth-century ku-wen movement. 2  E. G. Pulleyblank, “Neo-Confucianism and Neo-Legalism in T’ang Intellectual Life, 755–805,” in The Confucian Persuasion, ed. A. F. Wright (Stanford, Calif., 1960), pp. 84–88. My debt to this article will be apparent in the following pages. 3  Li Hua, biographies in cts (Po-na edition) 190C.1a–b and hts (Po-na edition) 203.1a–b; prose works in ctw (Taipei, 1965 reprint) 315.1a–321.19a and T’ang wen shih-i (Taipei, 1962 reprint) 19.13a–b; poems in CTShih (Peking, 1960), vol. 3, pp. 1585–90. Li’s approximate dates may be inferred from his hts biography, his collected works, and the fact that he was about the same age as Hsiao Ying-shih, having studied with him in the T’ai-hsüeh before coming of age; see T’ang wen-ts’ui (sptk edition) 15B.5b. Li was alive but sick in 766; see ctw 315.10a and note 138 below.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_034

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1061

Ying-shih (706–58),4 Yüan Chieh (719–72),5 Tu-ku Chi (725–77)6 and Yen Chench’ing (709–84).7 Before analyzing their remarks on history and literature, I have sketched their family backgrounds and described briefly the intellectual world in which they grew up. This may yield some understanding of their motives in writing, and perhaps also of their thought. These five men were, their writings tell us, closely connected with one another. They knew each other at various times in their lives and often dedicated works to one another. They had much in common in background and attitudes, since all five came from cadet branches of long-established aristocratic clans that had intermarried in T’ang times. Hsiao Ying-shih gave an informal account of his own origins in a long letter to Wei Shu, a leading official historian, written before the rebellion. Hsiao was descended from one of the princes of the Liang royal house. His ancestors had served the Sui and T’ang in some numbers and with considerable distinction, but since the late seventh century, he had had no forebear of rank.8 This combination of an illustrious pedigree with immediate ancestors of no great eminence was typical of the intellectuals in the group. Less is known of Li Hua’s immediate descent. But from remarks made in commemorative works for members of his clan and from independent genealogical sources, it is clear that he was from the Li clan of Chao-chün. This was 4  Hsiao Ying-shih, biographies in cts 190C.1b–2a and hts 202.12a–13b; prose works in ctw 322.1a–323.24a; poems in CTShih, vol. 3, pp. 1591–98, and vol. 12, p. 9970. T’ang Cheng-pi puts Hsiao’s death at 768; see Chung-kuo wen-hsüeh-chia ta tz’u-tien (Taipei, 1962 reprint), vol. 1, p. 292. This date, which is accepted by Pulleyblank (“Neo-Confucianism and Neo-Legalism,” p. 86), is almost certainly wrong. Li Hua, Hsiao’s lifelong friend, offered him a sacrificial graveside prayer in 760 (ctw 321.16b–17a). He also grouped him with Yüan Te-hsiu, who died in 754 (ctw 383.21b) and Liu Hsün, who died c. 761 (hts 132.3a), in his “Essay on the Three Sages” (ctw 317.3b–7a). cts 202.14b moreover says that Hsiao died “at the start of Ch’ien-yüan (c. 758).” 5  Yüan Chieh, biography in hts 143.2a–5a; prose works in ctw 380.1a–383.27b; poems in CTShih, vol. 4, pp. 2690–717, vol. 12, p. 10052, and Ch’üan T’ang shih-i, in CTShih, vol. 12, pp. 10214–17; on the identification of these poems, see Ōta Shōjirō, “Kaiyō sen chō kō,” Rekishi chiri 86, no. 2 (1955): 31–54. I accept the dating argued by Sun Wang, Yüan Tzu-shan nien-p’u (Shanghai, 1962), p. 1. 6  Tu-ku Chi, biography in cts 162.1b–3a; prose works in ctw 384.1a–393.26a and T’ang wen shih-i 22.7b; poems in CTShih, vol. 4, pp. 2760–79. 7  Yen Chen-ch’ing, biographies in cts 128.5a–9b and hts 153.5a–9b; prose works in ctw 336.1a–344.25a and T’ang wen shih-i 19.22b–20.1a; poems in CTShih, vol. 3, pp. 1582–84, vol. 11, pp. 8880–86. 8  c tw 323. 11a–b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1062

McMullen

the nonimperial Li family, distinct from the Lung-hsi Li clan, the T’ang royal house, from which they gained a certain amount of reflected luster.9 Yüan Chieh was descended from the ruling house of the Toba Wei dynasty.10 Collateral members of his clan, like the historian and canonical scholar Yüan Hsing-ch’ung, had distinguished themselves in the early eighth century, but Yüan’s own grandfather and father had not held high office, and he spoke of “two generations of impoverishment.”11 Such claims of deprivation in early life by writers of this period should be treated with reserve, since poverty, after all, was highly honorable and part of the topos for the virtuous scholar. But in Yüan’s case the facts about illustrious remoter origins and immediate forebears of no great distinction are unassailable. Tu-ku Chi came from what must have been the most illustrious Turkic clan not to have claimed full imperial honors, one which had provided consorts for the emperors of Sui and T’ang. His writings for relatives prove that he was very conscious of his origins, but again, his father had not held high rank.12 Finally Yen Chen-ch’ing came from a Chinese gentry family with the strongest traditions of scholarship and bureaucratic service, dating back to the end of the Han period. He too was interested in his clan history and wrote on the subject.13 Families with such a glorious but remote history and a more obscure record in recent times might well be expected to be conservative. It is not surprising to find them interested in those periods of the past when their own clans had flourished. Hsiao Ying-shih, for example, showed special interest in the Liang period, writing several works on it;14 and Yüan Hsing-ch’ung, a collateral cousin of Yüan Chieh active in the K’ai-yüan period (713–42), wrote a history 9  ctw 388. 11b; Yüan-ho hsing tsuan (1802 edition) 1.1a; Li Chao, Kuo-shih pu (Shanghai, 1957), p. 20. Li Hua composed commemorative texts for both clans, e.g., ctw 321.5a–7a and 11b–12a for the Chao-chün clan, ibid., 321.9b–11b for the Lung-hsi clan. 10   h ts 143.2a and ctw 344.16a. 11   c tw 381.12b. 12  For remarks on the Tu-ku family in Sui and early T’ang times, see Woodbridge Bingham, The Fall of Sui and Rise of T’ang (Baltimore, 1941), p. 5. The emperor T’ai-tsung also took a concubine from the clan; see Chao Lin, Yin-hua lu (Shanghai, 1957), p. 69. For Tu-ku Chi’s immediate ancestry and family, see Ts’en Chung-mien, T’ang chi chih-i, in T’ang jen hangti lu (Peking, 1962), p. 361; also ctw 522.3a. 13   c tw 337.10a–b. 14   h ts 202.12b lists a “Genealogy for the Liang Hsiao Clan” and an essay on the question of dynastic legitimacy as it involved the Liang, “Essay on the Fact that the Liang Did Not Abdicate for the Ch’en.” See also Pulleyblank, “Neo-Confucianism and Neo-Legalism,” p. 86, and Hiraoka Takeo, Keisho no dentō (Tokyo, 1951), pp. 107–10. Hsiao’s interest in the

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1063

of the Wei dynasty.15 But these men, though of aristocratic descent, were no longer courtiers, nor did they have significant local power. They depended for their own advancement on the examination system, the civil service, and the bureaucratic state. Though they often had an ambivalent attitude toward those in power, they were deeply committed to the T’ang and to the idea of a great, united, and long-lived dynasty. Their loyalty to the T’ang modified their clan ties, leading them, as we shall see, to prefer the Han empire to the Period of Disunion, when their ancestors had been preeminent. The record of the five men in the examination system testifies to their cohesion as a group. Two examiners were responsible for admitting all five to their degrees. Hsiao Ying-shih, Li Hua, Yen Chen-ch’ing, and also Yüan Chieh’s cousin and teacher Yüan Te-hsiu all passed under Sun Ti in the late K’ai-yüan period.16 Tu-ku Chi and Yüan Chieh passed in late T’ien-pao (742–56), during the brief interlude between the dictatorship of Li Lin-fu and the outbreak of war. Their examiner was Yang Chün, who in his selection reportedly took the advice of Hsiao Ying-shih, ten or more of whose disciples were successful.17 It is highly likely that Yüan Chieh, whose cousin Hsiao admired,18 and Tu-ku Chi, whom Li Hua thought highly of,19 were helped on the way by Hsiao Ying-shih.20 Four of the five men were teachers and had disciples following them both before and after the rebellion. Hsiao Ying-shih, whose pupils were so successful in the examinations of the late T’ien-pao, was particularly influential in this role. When some of them gathered to feast him in 754, they gratefully likened him to Confucius himself: “The subtle words of the First Teacher came to an Liang was not consistent with his sweeping dismissal of the entire Period of Disunion, for which see below. 15  The Wei tien, in 30 chüan, by Yüan Hsing-ch’ung is a chronicle style history for the Later Wei rather than a compendium; see cts 102.9b and hts 48.5b. 16  Yüan Te-hsiu, Yüan Chieh’s cousin, graduated chin-shih in 733, see Hsü Sung, comp., Tengk’o chi-k’ao (Nan-ching shu-yüan ts’ung-shu edition) 8.3a; Yen Chen-ch’ing in 734 (ibid., 8.4a); Li Hua, Hsiao Ying-shih, and also the historian Liu Fang (see note 86) in 735 (ibid., 8.11a–12b). Yen Chen-ch’ing (ctw 337.12a–b) and Hsiao Ying-shih (ctw 323.9a) paid tribute to Sun Ti. 17  Yüan Chieh gained his chin-shih in 754 (Teng-k’o chi-k’ao 9.27b). Tu-ku Chi was successful in a Taoist degree examination in the same year (ibid., 9.28a). For Hsiao Ying-shih’s part in the selection of graduates, see ctw 317.6b and CTShih, vol. 3, p. 1594. 18   h ts 202.12a; CTShih, vol. 3, p. 1595. 19   c tw 522.4a. 20  For the friendship between Yen Chen-ch’ing, Li Hua, and Hsiao Ying-shih before the rebellion, see ctw 317.7a. Some of this circle had met before coming of age, at the T’ai-hsüeh; see T’ang wen-ts’ui 15B.5b, hts 202.13b and 151.4b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1064

McMullen

end over a thousand years ago. Only after the advent of our Master has their immeasurable beauty been restored.”21 In reply Hsiao dwelt on the great responsibilities of the teacher in the four branches of instruction traditional to Confucianism—virtuous conduct, ability in speaking, administrative service, and literature and scholarship.22 He evidently made a great impression on his pupils, for they remembered him years after his death.23 Li Hua was involved in educational writing. His letter to two maternal granddaughters about to be married is the only surviving example of the chia–hsün (“family injunctions”) type of literature written by any of the five men studied here.24 A preface to the educational handbook Meng-ch’iu is also ascribed to him.25 He too might have been a teacher had he not been captured by the An Lu-shan rebels and forced to collaborate. Though a prospective pupil later came from afar to visit him, he did not feel qualified to give instruction.26 Yüan Chieh’s cousin Yüan Te-hsiu had had disciples, Li Hua and Hsiao Yingshih among them;27 and Yüan Chieh himself followed suit. When still a young man he called his pupils “little ones” in the manner of the Analects.28 After the rebellion, as governor of Tao-chou in the far south, he too attracted pupils from a distance.29 Tu-ku Chi as a governor in the southeast after the rebellion was also interested in education and personally appointed a lecturer to his prefectural school. A visitor there testified to the way in which Tu-ku “governed by virtuous conduct, learning, and literature, so that within a year Confucianism had spread widely.”30 Their roles as public exemplars and teachers may help to explain the highly serious and moral positions these men took on most issues and the dry tone of much of their writings. This may also have been a secondary factor legitimizing 21   c tw 395.5a. 22   C TShih, vol. 3, p. 1594; Analects (hyisis edition) 11.3. 23   h ts 203.14a. Several of Hsiao’s disciples composed an “Essay on the Collected Works of Hsiao Ying-shih,” since lost; see also ctw 395.3b. Hsiao’s disciples gave him the posthumous name of Wen-yüan hsien-sheng (hts 202.13b). He was also invited to go to Japan or Korea (ctw 395.5a). 24   c tw 315.3a–4b. 25   T’ang wen shih-i 19.13a–b. 26   c tw 315.15a. 27   h ts 202.12a and 194.1b. Li Hua gave Yüan Te-hsiu the posthumous name of Wen-hsing hsien-sheng. 28   c tw 383.4b; Analects 3.24, 7.24. 29   c tw 381.19b; cts 185B.14a. 30   c tw 518.24a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1065

their intellectual conservatism, since the Confucian Classics tended to advocate a conservative outlook and led students to look back to a better past. One area in which these five intellectuals differed was religion. Under Hsüan-tsung there had been a state-sponsored revival of intellectual Taoism,31 and for a time in the T’ien-pao period knowledge of Taoism considerably helped a man’s examination and career prospects. This and a family tradition of interest in Huang-Lao may account for the Taoist orientation of much of Yüan Chieh’s early thinking.32 Toward Buddhism, in his one recorded encounter with it, Yüan expressed only a good-humored sense of mystification.33 Tu-ku Chi’s early interest in Taoism seems to parallel that of Yüan Chieh. He was successful in a Taoist decree examination of 753.34 After the rebellion, when they both held governorships in the south, Taoism offered Yüan and Tu-ku psychological solace in times of stress.35 But, unlike Yüan, Tu-ku had also become interested in Buddhism, which he studied with dedication. He clearly derived comfort from it, for he told his friends to do likewise,36 and he was familiar with the history of the Buddhist monasteries he visited and with certain Buddhist texts and their doctrines.37 Buddhism seems to have been a diversion, or perhaps a consolation, to Hsiao Ying-shih, who wrote that he had fond memories of “sitting together with good friends, being served with tea from time to time, criticizing the sages of antiquity, and discussing Buddhist scriptures.”38 He is known to have read and studied sutras with his lifelong friend Li Hua.39 Li, at least in later life, had an active interest in Buddhism, which led him to compose commemorative texts for foreign Tantric priests40 and to be connected with the eighth and ninth partriarchs of the T’ien-tai sect.41 Although his

31  Hsiao Kung-ch’üan, Chung-kuo cheng-chih ssu-hsiang shih (Taipei, 1954 reprint), pp. 423 and 439n1–3. 32   c tw 383.10b, 11b, and 18a. 33   c tw 383.24a. 34   Teng-k’o chi-k’ao 9.28a. 35   c tw 386.11a and CTShih, vol. 4, p. 2704, both instances relating to times when they were prefectural governors threatened with heavy tax demands on exhausted populations. 36   C TShih, vol. 4, p. 2771. 37  E.g., ctw 389.20a and 24b. 38   c tw 323.11b. 39   c ts 190C.1a; hts 203.1b. 40  Chou I-liang, “Tantrism in China,” hjas 8 (1945): 250–51. 41   f ttc 188b, 189a, in T (Tokyo, 1924–32), vol. 49.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1066

McMullen

thought elsewhere suggests skepticism,42 he accepted the claims of Buddhist thaumaturgy. Li seems to have looked to Buddhism also as a means of escape from an oppressive personal situation, from the guilt he felt over his collaboration. The long poems he wrote toward the end of his life reflecting on his predicament end with expressions of dedication to Buddhism.43 Tu-ku Chi testified to his colleague’s devotion to the faith when he wrote of him in his final years that “he looked upon rank and salary and his own body as if they were merely his remains in the soil.”44 Yen Chen-ch’ing’s religious interests were possibly the most unusual of the group. He was a believer in the Taoist pantheon and in the miracles and apparitions of its tradition.45 But he also had extensive social contacts with Buddhism and knew the history of the monasteries he visited.46 Such diversity in religious interests among so homogeneous a group need cause no surprise. It is well known that in T’ang China there was, within predictable limits, no interference in private beliefs. Buddhism and Taoism generally affected only the private areas of a lay follower’s conduct and provided little in the way of active social values. It is true that Yüan Chieh, before taking an examination in 747, had written an essay suggesting that Taoism had once offered a social ideal, but he did so probably with an eye to the emperor’s interest in the subject and before he had had administrative experience.47 In areas where society was involved, such as historical and literary theory, members of the group in their maturity all turned to the Confucian tradition, endorsing it with a unanimity that stands in contrast to the diversity of their interests in religion. The scholarly careers of the five men were greatly affected by the rebellion. Late T’ang tradition was to look back on the reign of Hsüan-tsung, the age in which they started their careers and which shaped their expectations, as one

42   c tw 317.10a. Li was not as systematic a skeptic as was, for example, Lü Ts’ai (ctw 160.10a– 15b) at the start of the dynasty. 43   C TShih, vol. 3, pp. 1587, 1589. 44   c tw 388.11b. 45   c tw 338.25a. 46  E.g., ctw 339.2a, 2b. The work professing nonbelief in Buddhism ascribed to Yen in ctw 337.18b–19a which opens, “I do not believe in the Buddhist Law, but like staying in Buddhist monastaries…,” is very reasonably rejected as spurious by the commentator in Yen Lu-kung chi (sppy edition) 5.5b–6a. Ou-yang Hsiu strongly disapproved of Yen’s Buddhist and Taoist leanings; see Ou-yang Yung-shu chi (Kuo-hsüeh chi-pen ts’ung-shu edition), ch. 16, p. 3. 47   c tw 383.12a; see also Hsiao Kung-ch’üan, Chung-kuo cheng-chih ssu-hsiang shih, p. 425.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1067

of the great periods of the dynasty and indeed of all time.48 But even in the two decades after the outbreak of rebellion, before distance in time had lent them charm, intellectuals felt nostalgia for the years of security and prosperity. Thus Yen Chen-ch’ing wrote of the good government of the Chen-kuan (627–50) and K’ai-yüan periods,49 and Yüan Chieh reminisced about the “great peace” in which he had grown up.50 Li Hua recalled it as a time when “the world was at peace, and the gentleman (chün tzu) was able to take his leisure with learning, so that in the field of letters men of great ability abounded.”51 In history and literature, the particular concerns of this paper, the volume of production in the first half of the eighth century had been considerable. Many of the intellectuals of the time had led secure working lives in the academies, the imperial library, or the history office and had had exceptionally long careers in scholarship, particularly in the field of history. Thus Wu Ching, who died in 749, had held posts as a historian for nearly thirty years52 and had had a hand in some twenty-two works.53 Wei Shu, who died in 757, had had a scholarly life spanning forty years, twenty of which were spent as an official historian working on almost all the types of historical writing then practiced.54 Yet these are only two examples of a type of scholar active in Hsüan-tsung’s reign, and there were many others like them.55 The New T’ang History’s bibliography indicates the range of their work: official histories, chronicle form histories, informal historical works, collections of anecdotes and biographies, works on institutions, ritual, law, bibliography, genealogy, and geography.56 Nor should it be thought that this activity was all purely mechanical and uncritical or that the state was the only initiator of large-scale scholarly projects. Liu Chih-chi (661–721) is perhaps the best-known critic of state-sponsored scholarship in all Chinese history, and his Shih-t’ung is one of the most 48  Chang Yen-yüan, Li-tai ming-hua chi, translated by W. R. B. Acker, Some T’ang and PreT’ang Texts on Chinese Painting (Leiden, 1965), p. 203; see also ctw 555.18a. 49   c tw 336.13a–b. 50   C TShih, vol. 4, pp. 2704–05. 51   c tw 315.9a. 52  Wu Ching, biographies in cts 102.12a–b and hts 132.3b–5a. 53  These are listed in I-wen chih erh-shih chung tsung-ho yin-te (Peking, 1933), p. 35. 54  Wei Shu, biographies in cts 102.13a–14a and hts 132.5a–6a; works listed in I-wen chih erhshih chung tsung-ho yin-te, p. 172. 55  E.g., Hsü Chien, biographies in cts 102.8a–9a and hts 199.2a–3a, works listed in I-wen chih erh-shih chung tsung-ho yin-te, p. 188; Yüan Hsing-ch’ung, known as Yüan Tan, biographies in cts 102.9a–12a and hts 200.2a–3b, works listed in I-wen chih erh-shih chung tsung-ho yin-te, p. 62; T’ang Ying, for whose works see hts 58.5b. 56   h ts 58.1a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1068

McMullen

systematic works of intellect to have come to us from the first half of the eighth century.57 But there were other scholars showing independence and a critical sense. Even an official historian like Wu Ching could disapprove of the imperially sanctioned histories of earlier dynasties and revise them for his own purposes.58 There was criticism of the authorized canonical commentaries,59 at least one lively bibliographical controversy on the Classics,60 and debate on the best form for writing histories.61 Hsiao Ying-shih took part in this dispute not long before the rebellion, advocating, as we shall see, the chronicle form, and planning a long chronicle-style history himself. Literature had flourished in the pre-rebellion period as never before. Some ten thousand poems survive from the first half of the century. Early in Hsüantsung’s reign the regulated styles of verse had commanded most prestige, but in late K’ai-yüan and T’ien-pao freer forms were also popular. Themes varied from the fantasy journeys of Li Po to the frontier campaign verse of Ts’en Shen and Kao Shih, from the quietist nature poems of Wang Wei to the poetry of social criticism that Yüan Chieh had begun to write as a young man.62 This was an age of great self-confidence, when the compiler of a comprehensive anthology of verse just before the rebellion could claim in a preface addressed to the emperor that “after K’ai-yüan 15 (727), both tonal regulations and vigorous style attained perfection for the first time.”63 Literary criticism had also flourished, though it could not rival historical criticism in producing any long or systematic treatises. The ability to write verse was taken for granted among the educated, but for the technically more exacting forms, like the chüeh-chü or lü-shih, handbooks of rules and examples were written. Indeed this prescriptive criticism was the kind most typical of

57  See E. G. Pulleyblank, “Chinese Historical Criticism: Liu Chih-chi and Ssu-ma Kuang,” in Historians of China and Japan (London, 1961), pp. 135–51, and William Hung, “A T’ang Historiographer’s Letter of Resignation,” hjas 29 (1969): 5–52. 58   c ts 102.12b. 59   c tw 272.2a, and K. P. Kramers, “Conservatism and the Transmission of the Confucian Canon: A T’ang Scholar’s Complaint,” Journal of the Oriental Society 2 (1955): 118–32. 60  William Hung, “A Bibliographical Controversy at the T’ang Court, ad 719,” hjas 20 (1957): 74–134. 61  Liu Chih-chi’s comments on this controversy are given in Shih-t’ung t’ung-shih (sppy edition) 2.1a–3a; see also below at note 129. 62  For this fourfold characterization compare Liu Ta-chieh, Chung-kuo wen-hsüeh fa-chan shih (Shanghai, 1949), pp. 333–80. For Yüan’s “social realist” verse see CTShih, vol. 4, pp. 2696–98 and 2703. 63   c tw 436.20b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1069

the late seventh and early eighth centuries,64 when, as writers of the rebellion period remind us, the regulated styles attained their final form.65 Throughout Hsüan-tsung’s reign anthologies were compiled, and their prefaces, setting out the criteria for the selection of the poems they contained, were an important vehicle for criticism. Critical remarks are also to be found scattered in the verse of the period. This was a kind of “work-shop criticism,” to borrow T. S. Eliot’s phrase,66 and though random and unsystematic, sometimes offers clues to the attitudes of writers who did not otherwise commit themselves.67 The security and wealth of the intellectual world in which the five men grew up is also suggested by its libraries; this was a period of great collections. The imperial library itself had an establishment of a director, two vice-directors, and some seventeen staff;68 before the rebellion both Yen Chen-ch’ing69 and Hsiao Ying-shihg70 held posts in it. The imperial collection contained over fifty thousand chüan.71 And there were large private libraries as well—Wu Ching, the historian, had a collection large enough to need its own catalog,72 and Wei Shu one of some two thousand chüan.73 There were also many valuable collections of paintings.74 The sack of Ch’ang-an in 756 destroyed forever the security that had fostered this learning and wealth. Many intellectuals fled south to the lower Yangtse or attempted to join the loyalist forces at the temporary capital in the west. But a significant number, like Li Hua and Wei Shu the historian, were forced to 64   h ts 60.15a lists five works of prescriptive criticism likely to have been written between early T’ang and the An Lu-shan Rebellion. Some of these were incorporated in Kūkai’s Bunkyō hifuron and preserved in Japan ; see Bunkyō hifuron, annotated and indexed by Konishi Jinichi (Tokyo, 1953). 65  This judgment was made very early, for example by Tu-ku Chi (ctw 388.1a). Cf. note 64; and the modern critic Wang Yün-hsi, in “Shih Ho-yüeh ying-ling chi hsü lun sheng T’ang shih-ko,” T’ang shih yen-chiu lun-wen chi (Peking, 1959), p. 28, who gives the same judgment. 66  T. S. Eliot, “The Frontiers of Criticism,” in On Poetry and Poets (London, 1957), pp. 106–07. 67  See the two articles by Itō Masafumi, “Sei Tō shijin to zendai no shijin—jō,” Chūgoku bungaku hō 8 (1958) : 93–135, and “Sei Tō shijin to zendai no shijin—ge,” Chūgoku bungaku hō 10 (1959): 17–51. 68  R. des Rotours, Traité des fonctionnaires (Leiden, 1947), pp. 204–08. 69   c tw 514.10a. 70   h ts 202.12b. 71   c ts 46.1b; cf. hts 57.1a–b. 72   c ts 102.2b. 73   c ts 102.13a. 74  Acker, Some T’ang and Pre-T’ang Texts on Chinese Painting, pp. 204–05, 207–08.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1070

McMullen

collaborate with the rebels.75 The cause of intellectuals was further discredited when Fang Kuan, one of the intellectual leaders before the rebellion, offered the emperor disastrously bookish advice during the military crisis of 756.76 During the next decade or so, power belonged with military men, and intellectuals were, by comparison with pre-rebellion days, in the wilderness. All the five men under review were much affected by the war. Tu-ku Chi was a junior official of a county near the capital when he was forced to flee to the southeast. He took up local office there, and this service led to his reappointment to the region in later life. Yüan Chieh, though a chin-shih of 754, held no office at the outbreak of war. He fled to the Yangtse Valley, and his official career started only in late 757 when he was summoned north to give advice to the emperor. He later claimed that he had originally wanted “some leisurely appointment in an office to do with letters,”77 but he was soon involved in mustering troops in Honan. Almost all his later serving career was spent in the south, sometimes under dangerous and isolated conditions. He was finally summoned to Ch’ang-an in 772 but died before reappointment. Of all the group Hsiao Ying-shih had advanced his scholarly career furthest by the outbreak of war. He had established himself as a teacher, had held an appointment in the imperial library, sought the patronage of Wei Shu the historian, and had written a number of works on history. He might have continued to work on larger projects if the rebellion had not denied him the opportunity. When the war broke out, he fled to Hsiang-yang in Honan and then took up minor office with a military governor in the southeast. In 758, however, he died in Honan while trying to move his father’s coffin south for reburial. Li Hua fared worst of the group, for he was taken by the rebels while trying to rescue his mother and was forced to collaborate. He was later pardoned and in 758 summoned to court. But he felt that his service with the rebels permanently disqualified him from office and, in his own words, “made bold to disobey the order of the court and follow my private wishes,”78 by remaining in retirement in the lower Yangtse area.

75   t ctc (Peking, 1956), ch. 220, p. 7042, mentions a figure of three hundred officials. Apart from Wei Shu, Li Hua, Chao Hua, and Shao Chen were among the intellectuals forced to collaborate; see CTShih, vol. 3, p. 1588. Also, for Ch’u Kuang-hsi, see Hsin Wen-fang, T’ang ts’ai-tzu chuan (Shanghai, 1957), p. 18. 76   t ctc, ch. 219, p. 7004. For this episode and its background, see Pulleyblank, “NeoConfucianism and Neo-Legalism,” pp. 98–99. 77   c tw 381.12b. 78   C TShih, vol. 3, pp. 1587–88.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1071

Yen Chen-ch’ing was governor of P’ing-yüan in Hopei just before the rebellion, and in his leisure he had been working on a large dictionary project there.79 When An Lu-shan swept down from Yu-chou toward Loyang, he was forced to marshal all available military resources and organize resistance behind the rebel fines. Later he held numerous commissionerships and posts in the metropolitan bureaucracy. He was able to resume his scholarly activities only in 773, when a governorship in the southeast gave him the leisure to do so.80 The catastrophe of 755 and its sequel therefore meant that none of the five men enjoyed secure tenure in the metropolitan bureaucracy or worked in the imperial library as their predecessors had done in the days of peace. Instead, those who took up office suffered all the vicissitudes of bureaucratic service away from the capital, at the mercy of a temporizing court, and of increasingly insecure provincial conditions. Those who retired to the south, like Li Hua, and later on two occasions Yüan Chieh, paid for their security by their isolation. They managed to keep up only occasional contact with one another. We know only that Li Hua and Tu-ku Chi met in the southeast after the rebellion,81 and that Yen Chen-ch’ing communicated with Yüan Chieh82 and enlisted the help of Hsiao Ying-shih’s son when he resumed work on his dictionary project in 773.83 Yet it would be a mistake to attribute the pattern of their careers solely to the rebellion. Isolated service in the remote south had been a possibility in days of peace as well. Retirement, too, was a path often taken before the rebellion. And though the capital had been sacked in 756, was evacuated before the Tibetans in 763,84 and was to prove more and more difficult to administer,85 the former institutions continued to operate. The Old T’ang History claims that at the time of the rebellion, “the old books were almost all lost or dispersed,”86 but when the capital was recovered, prompt efforts were made to retrieve them.87 The history office was reestablished, and though there was a drop in the number 79   c tw 514.21a–b. 80   c tw 339.6b. 81   c tw 388.11b. 82   c tw 380.7a. Yüan Chieh’s “Eulogy for the Revival of the Great T’ang,” composed in 761, was written out by Yen Chen-ch’ing and carved in rock at Wu-ch’i, Hunan, in 771; see Sun Wang, Yüan Tzu-shan nien-p’u, p. 99. 83   c tw 339.7a; cf. cts 48.8b, biography of Ch’üan Kao. Ch’üan was known to both Yen and Li Hua after the rebellion. 84   t ctc, ch. 223, p. 7151. 85   c tw 384.21a–b. 86   c ts 46.1b. 87   h ts 57.2a; cts 149.1b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1072

McMullen

of large projects, work went on. At least one historian, a fellow graduate of the intellectuals under review and a colleague of the pre-rebellion historian Wei Shu, managed a long career there.88 Private libraries too were apparently soon built up again.89 Yüan Chieh and Yen Chen-ch’ing had collections with them in the south in the 770s,90 but it may be that, though forced to flee before the rebels, they had never lost their original libraries. But official service in dangerous conditions, or retirement and sometimes poverty, did mean that no member of the group was able to settle down to any work of major length for the two decades after the rebellion. Instead their later collections consist of shorter length san-wen (“miscellaneous prose”) pieces, memorials, biographical and commemorative works, inscriptions, essays, letters, prefaces, casual notes, and verse. The composition of their collected works is important to a student of their ideas. The mid-T’ang tradition of sanwen writing was, as Ch’en Yin-k’o noted,91 a highly stereotyped one. The genre a writer used often imposed narrow restrictions on the content of his writing. Most of the extant remarks on literature and history by the five men occur in short essays and prefaces, as the T’ang tradition of san-wen writing demanded. But what is new in these writings, when compared to early eighth-century sanwen, is their sharper critical spirit and stronger convictions. The five men felt keenly about the climate of the late T’ien-pao and the post-rebellion world. It was this that led them to reexplore and update traditional ideas on history and literature. Finally, it should be stressed that although these five men have the largest collections extant from their period, they do not represent completely the intellectual developments of their time. Others were writing sometimes in quite different ways, advocating, for example, legalist or quasi-legalist solutions to political problems or adhering to a more aesthetic approach toward literature.

88  Liu Fang, biographies in cts 149.14a–b and hts 132.8a. 89  K. T. Wu, “Libraries and Book Collecting in China before the Invention of Printing,” T’ienhsia Monthly 5, no. 3 (October, 1937): 259–60, lists large private collections in the T’ang. In the post-rebellion period, Li Pi, who died in 789, and Su Pien, brother of the author of the T’ang hui-yao, both had large collections. Acker, Some T’ang and Pre-T’ang Texts on Chinese Painting, pp. 205–06, notes that collecting and trading in pictures and manuscripts went on after the rebellion. 90  Yüan was visited during his final retirement in the years 769–72 at Wu-ch’i in Hunan by Wang Yung, a fellow graduate, who remarked that “there were many books in the house”; see ctw 356.18b. Yen Chen-ch’ing needed a comprehensive library of classics to compile his dictionary; see ctw 339.6b. 91  Ch’en Yin-k’o, Yüan Po shih chien-cheng-kao (Shanghai, 1958), pp. 2–3.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1073

Yet it is not entirely fortuitous that the works of these five men should have been relatively well preserved, while others have been lost. Although unselective forces have destroyed large numbers of works from this period, preservation has not been wholly impartial. A somewhat sinister force is discernible in the selection process, namely, the power of the editor silently to reject what is not in his view sufficiently orthodox, exemplary, or consonant with the Confucian reading of history, and to keep what is. It is no accident, therefore, that those mid-eighth-century figures whom Confucian historiography condemns, whether intellectuals or not, have almost no writings preserved. The villains of the era—Yang Kuo-chung, Li Fu-kuo, and others—were not illiterate; it was just that their works stood little chance of survival. In less clear cases the same forces must also have operated. Thus Fang Kuan, the intellectual who advised the emperor so disastrously in the campaign of 756, was severely censured by the histories; virtually none of his writing survives today. Yet he must have been a productive and stimulating thinker in his own time. Even selective editing has not quite managed to excise the fact that he had friends and admirers among men who were judged respectable—Tu Fu, Li Hua, and Tu-ku Chi, for example.92 The same forces that made the survival of works by Yang Kuo-chung or Li Fu-kuo unlikely, and that lessened the chances for the works of men like Fang Kuan, worked the other way for the writings of the men under review. They were all loyalists, or at least if they strayed, they showed due penitence. None was closely identified with any party or individual whom history subsequently condemned. Two, in particular, had moments of greatness—Yüan Chieh, when as governor of Tao-chou he repelled successive Man incursions and protected his populace from the depredations of visiting tax collectors,93 and Yen Chench’ing, when in 784, after three decades of tireless service under four emperors, he was finally martyred to the loyalist cause.94 Historians were also interested in these men because of their connection with the emergence of exclusive Confucianism in the ninth century. The vital role here was played by Han Yü. Han had a link with Tu-ku Chi,95 who in turn 92  See Pulleyblank, “Neo-Confucianism and Neo-Legalism,” pp. 99 and 330n96, 99. Also ctw 522.4a, 317.3b, and 318.7a; and CTShih, vol. 4, p. 2478. In ctw 588.1a–2a, Liu Tsung-yüan praised Fang Kuan very highly and noted that Li Hua had composed an inscription for him that was not erected because of the rebellion. See also Li Chao, Kuo-shih pu, pp. 18, 22, and 49. 93  Sun Wang, Yüan Tzu-shan nien-p’u, pp. 66–67, 75. 94   t ctc, ch. 229, p. 7393 and ch. 231, p. 7443, and the sources cited in note 7. 95   c ts 160.1a; Pulleyblank, “Neo-Confucianism and Neo-Legalism,” p. 95.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1074

McMullen

had a strong connection with Li Hua and Hsiao Ying-shih.96 Han Yü and other ninth-century figures also commended Yüan Chieh.97 Thus a line of transmission was established. In the early Sung, when Han Yü was virtually canonized, the interest in his intellectual forebears intensified. From then on the chances of having their works preserved further increased. Their biographies in the official histories were also affected by Sung Neo-Confucian interest in them. Hsiao Ying-shih and Li Hua were given expanded biographies in the New T’ang History, and Tu-ku Chi and Yüan Chieh, who had been omitted in the older history, were now included. Only Yen Chen-ch’ing had so high a standing in the older history that his biography needed no expansion in the new.98

Historical Criticism

The writings of the five intellectuals under review show three main approaches to history. There were, first, general reviews of history as an impersonal process susceptible to periodization or demonstrating certain long-term trends. Second there was historical criticism in the specific sense of criticism of existing histories and the procedures for writing them. Finally, the five men analyzed the roles of certain figures in history, critically evaluating in moral terms. The general reviews of history by Hsiao Ying-shih, Li Hua, and other contemporaries open by referring to a concept of long standing in the Confucian historical and literary tradition, that of wen. Like other Confucian concepts, this idea was rooted in laconic statements in the Classics, the Tso chüan, the Analects, and the I ching.99 It had been elaborated in the Han period in the Li chi, in Tung Chung-shu’s Ch’un-ch’iu fan-lu, and in the Mao shih ta–hsü, where it had attained essentially its full range as an idea.100 Ever since Hsiao T’ung had alluded to it in his preface to the Wen-ksüan,101 writers had had the habit of starting their prefaces and introductions to works of history and literature alike by referring to it. Thus in early T’ang it is to be found in introductions by 96   c tw 388.11b, 518.6a. 97  For Han Yü’s commendation of Yüan Chieh, see ctw 555.3a; for that of Lü Wen, ibid., 628.11b; and that of P’ei Ching, ibid., 764.17a–b. 98  For Hsiao Ying-shih and Li Hua, see hts 202.12a and 203.11a; for Tu-ku Chi and Yüan Chieh, see ibid., 162.1b and 143.2a; for Yen Chen-ch’ing see note 46 above. 99   Tso chuan, in Ch’un-ch’iu ching chuan (hyisis edition), Hsiang 25, p. 307; Analects 3.14, 6.18; I ching (sptk edition), hexagram 22 ( fen), 3.2a–b. 100   Li chi (sptk edition) 11.6a–b, 17.6a; Ch’un-ch’iu fan-lu (sptk edition) 7.3b; Mao shih (sptk edition) 1.1b–2a. 101  J. R. Hightower, “The Wen-hsüan and Genre Theory,” hjas 20 (1957): 518.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1075

the historians Li Yen-shou102 and Wei Cheng103 and in prefaces by the poets Yang Chiung104 and Wang Po.105 In K’ai-yüan it appears in the writing of Han Hsiu,106 Chang Yüeh,107 and Liu Chih-chi.108 Wen meant primarily pattern, the rhythmic repetition of a shape. The I ching had spoken of two kinds, t’ien-wen—the pattern of the heavenly bodies and the pattern of the topography of China herself109—and jen-wen—human pattern, which denoted the three media of music, ritual, and the written word. Wen was hypostatized, that is, held to exist of itself, not merely to be the attribute of anything patterned. The concept of wen shares features with the early medieval European idea of musical harmony—the music of the universe, musica mundana, is parallel to t’ien-wen, while the music of man, musica humana, and instrumental music, musica instrumentalis, correspond to jen-wen.110 Both notions stress in different ways—the Chinese humanistically and the European theistically—man’s intimate connections with the universe. The ways in which the five intellectuals approached this concept fall into a polarity. At one extreme wen was considered as an objective phenomenon in the natural world or in society, to be observed and analyzed. Irregularities in heavenly pattern, according to this well-known theory, indicated irregularities in the human world below, and specifically in the moral conduct of government. Jen-wen was also an index of the condition of the state and, especially in the case of the written word, of the workings of the government. At the other extreme, wen was treated as a component in the traditional psychology of literary creation. This was an expressionistic concept: wen was the patterning of the inner feelings of man, mainly in literature, but also in music and ritual. Writers in the eighth century were conscious of the breadth of the idea. Thus Hsiao Ying-shih, referring at the start of a preface to t’ien-wen and jen-wen, the latter especially in the sense of the written word, remarked: “How perfect is wen! It is a great unifying principle to which heaven and man unite in responding, and through which the purpose of names and schedules is preserved.”111 102   Pei shih (Po-na edition) 83.1a. 103   Sui shu (Po-na edition) 76.1a. 104   c tw 191.9b. 105   c tw 182.15b. 106   c tw 295.4b. 107   c tw 225.17b. 108   Shih-t’ung t’ung-shih 6.12a, 6.16b. 109  For the clearest expression of this idea by a contemporary of the men under review, see ctw 443.16a. 110  E. de Bruyne, Études d’ésthetique médiévale (Bruges, 1946), vol. 1, p. 12. 111   c tw 322.15a–b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1076

McMullen

Such an allusion to the wen concept at the start of a preface was conventional, but some of the men under review explored the idea more fully. They considered jen-wen, in the sense of ritual, music, or the written word, as an indication of the condition of the state in any period, which offered them a measure for analyzing the historical process itself.112 Li Hua made this his theme in his “Essay on Substance and Pattern,”113 one of the most important works by any of the group. Li opened his essay by discussing an extension of the wen concept as a phenomenon in society. Wen, pattern, he explained, was antithetical to the value chih, substance. Chih was to be identified with austerity, tending to coarseness and then crudeness. Wen here had the extended meanings of refinement and extravagance, leading to falsehood. In this sense it took on its most extreme meaning, a negative one in contrast to its generally positive value. A balance between wen and chih was ideal, as Confucius had said.114 But if this could not be effected, then an excess of substance was preferable to an excess of pattern. The latter state was a dangerous one: “If when extravagance prevails, an attempt at reform is made, stability will be sought for but not secured.” Li was here alluding to the essential premise of all Chinese historical writing, that though the cycles and trends of history might be impersonal and mechanistic, yet men, and specifically the ruler, could intervene and arrest these processes by acting on a moral level. As long as the situation had not reached an extreme, Li implied, it was within the power of the ruler to redress the balance between wen and chih. Li went on to give historical examples of the way in which an imbalance toward chih could provide a basis for reform, while an imbalance toward wen made recovery difficult. His main example of chih was from the early Han. At the start of the dynasty, the royal house had been threatened by rebellion from the relatives of the empress Kao, and, a little later, in 154 bc, from the princes of Wu and Ch’u. This period, he argued, had been a time of austerity and frugality, when the harsh and complex laws of the Ch’in and the burdensome ritual of the later Chou had been abolished and simplicity had returned to administration. 112  This idea is explored earliest in Ch’un-ch’iu fan-lu 7.3b; see Feng Yu-lan, A History of Chinese Philosophy (Princeton, n.j., 1953), vol. 2, pp. 58–71. 113   c tw 317.1a–b. Chou shu (Po-na edition) 38.7a and Pei shih 64.20a mention an essay with a similar title, “Essay on Pattern and Substance,” by Liu Ch’iu. This is no longer extant, but to judge from the brief résumé in his biographies, it dealt only with literature; see Lo Kentse, Chung-kuo wen-hsüeh p’i-p’ing shih, vol. 1, p. 251. Li’s essay is datable on the evidence of ctw 388.15b to between 752 and 762. 114   Analects 6.18.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1077

It was precisely this frugality that had enabled the Han house to overcome the threats it faced. An example of the opposite situation was to be found in the later Hsia dynasty, when rulers had tampered with the perfect and sufficient decrees of Yü. As a result, wen had succeeded chih, and the dynasty had easily been overthrown. Li’s third example was that of the Chou, which after the reign of King K’ang (traditionally 1078–53 bc) had allowed its ritual and its administration to grow in complexity, making eventual disaster inevitable. Finally Li indicted his own times for an excess of wen over chih. In doing so he again made it clear that wen, though susceptible of control by the ruler, was an impersonal factor of great breadth. Punishments had become too involved, ritual too complex, divination and superstition rampant. The learned world was divided, a hundred schools existed where one would have been enough: “The empire is inundated like a lake, or flows along like a great flood. And even if the divine Yü were born again, neither he nor anyone else would be able to save it.” Unity and restraint must be restored where now there was plurality and extravagance. The idea of an orthodoxy, which was to appeal so strongly in the ninth century, is evident here: “I am of the humble opinion that any further search to bring about good order should start with the study and application of the Classics and histories. Works such as the Tso-chüan, Kuo-yü, Erh-ya, Hsüntzu, and Mencius are support for the Classics…. The theories of the remaining hundred philosophers … should be preserved but not used.” Of the other intellectuals in the group, Tu-ku Chi is said to have expounded wen and chih as they applied to early history.115 Yen Chen-ch’ing also alluded to the wen-chih polarity as it applied to one of his chief concerns, ritual. He was, like Li Hua, criticizing current practice. “They have done away with the chih of the ancients and set value on extravagant ornamentation,” he wrote; “substance is closer to antiquity, pattern to the present day…. When wen deteriorates, it should be rescued by chih.”116 His use of the concept, though an incidental one, is similar to Li’s. The wen-chih polarity was not, however, the only way in which these intellectuals viewed history. One of the fullest of the extant reviews, by Hsiao Yingshih, written before the rebellion, started by alluding to the wen idea, but did not make it central, as Li had done in his essay. Rather, Hsiao used it throughout only in its positive sense: patterned, cultured, pertaining to the written word.117 115   c tw 522.4a. 116   c tw 336.14b–16b; Yen’s knowledge of ritual was recognized when, in 779, he was appointed to a ritual commissionership. 117   c tw 322.15a–19a. There is a difference between the views Hsiao expressed here on behalf of Ch’en Cheng-ch’ing and those he put forward for his own projected history, described

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1078

McMullen

A striking feature of Hsiao’s review is that he commends exactly those periods in history Li had praised and condemns those Li had criticized. A common historical perspective may therefore be abstracted from the writings of the two men. Hsiao’s review is the more explicit, however, and supplies the details. Hsiao saw history not as the repetition of a strictly dynastic process, but as a thousand-year cycle involving first prosperity and good government and then disorder. There had been three great periods in history, the age of Yao and Shun in antiquity, the Han, and the T’ang itself. The reigns of Yao and Shun, described in the Book of History, had been praised by Confucius himself, Yao for his wen-ssu, culture and thought, and Shun for his wen-te, culture and virtue.118 During their reigns, the moral record had been perfect. Deterioration had set in with the Hsia, and even the early Chou had been short of perfection. Thereafter, “there was decline, usurpation, and disorder until the tyrannical Ch’in.” Ritual and music had declined, vertical alliances were linked, horizontal ones joined. But, Hsiao quoted the Analects, “Heaven does not let the cause of ssu-wen, culture, perish.”119 The Han arose, and there was a restoration of the ritual and laws of the period of Shun. The Han’s great achievements, reviving a simple ritual and the calendar, putting down the revolts of the Wu and Ch’u princes, and expanding its influence abroad, made it equal to the exemplary period of antiquity. After the Han, the record again was one of failure. China was sadly divided. “It was a matter of splitting into four or cracking into five, of success in the morning and defeat in the evening.” The Sui dynasty, lasting two generations, precisely the duration of the Ch’in, had prepared the way for the T’ang, and this parallelism was clear evidence of Heaven’s purpose. The Han and the T’ang were thus fully comparable. “What need is there to speak of Eastern Chin, Later Wei, Liang, Ch’en, Chou, or Ch’i?” Reviewing the T’ang itself, and particularly the period through which they lived, Hsiao Ying-shih and the other members of the group tended to have two standards. Their intense loyalty to the dynasty infuses many of their public writings, memorials, inscriptions, and so on, but their private writings bear more closely the stamp of their own experience. Before the rebellion they spoke of a rise in the level of culture and learning, leading to unseemly competition for office. Thus in 753 Yüan Chieh, who then still had to pass an examination, remarked of Hsüan-tsung’s reign that its ritual and music had reached below. Hsiao himself was clearly prepared to pay more attention to the Period of Disunion than he suggests here. 118   Shang shu (sptk edition) 1.1a, 2.6a. 119   Analects 9.5.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1079

as far as the outlying barbarians and that “even the slave classes can recite the works of the Duke of Chou and Confucius, or discourse on the way of T’ao T’ang and Yü Hsia.”120 Under such conditions, he continued, the scramble for office was degrading. Hsiao Ying-shih also protested before the rebellion about the unseemly competition for posts.121 There is an echo in these remarks of Li Hua’s plea for a more restricted and more austere learned world. Wen had exceeded chih. After the rebellion, as we have seen, the five men felt nostalgia for the age of peace, their regret at its passing given edge by resentment against those who had hastened its end.122 The evidence of their prose works proves that they were well aware of the changes in power structure which were taking place, both at court and in the increasingly independent provinces,123 but there is little to suggest that they had concrete, technical proposals for reform.124 Those who were active as civil servants—Tu-ku Chi, Yüan Chieh, and Yen Chen-ch’ing—restricted their roles to education, ritual, and public morality. They appear in fact to be true Confucians in their emphasis on the wen aspects of statecraft rather than on technical, institutional, or financial problems, on the moral role of the individual rather than on his special skills as an administrator. The five intellectuals found the wen concept and the wen-chih polarity useful not only in analyzing history in broad terms, but also in the second kind of history they practiced, the discussion of historical writing itself. Histories, like anything written, were wen, both in the objective sense of being a written record of the past and in the expressionistic sense as historians’ judgments on events. Viewed as such they had solemn functions indeed: “For effecting the transformation of the world, nothing is of greater importance than wen. The great institution of wen is the dynastic history, which has the office of giving praise and blame, reproof and exhortation, of separating the light from the dark.”125 120   c tw 383.19a–b. 121   c tw 323.8a. 122  E.g., ctw 336.14a, 315.9a. 123  E.g., ctw 380.8b, 336.12a. 124  A possible exception to this was Yen Chen-ch’ing, who was said to have rediscovered in discussion the idea of the salt monopoly while governor of P’ing-yüan in Hopei, in 756; see ibid., 514.18b and D. C. Twitchett, Financial Administration under the T’ang Dynasty (Cambridge, 1963), p. 263n13. Yet even here Yen did not act promptly on the idea, for Ti-wu Chi is said to have stolen it and memorialized it to Su-tsung before him. 125   c tw 316.5a, datable to 748. See William Hung, “The T’ang Bureau of Historiography before 708,” hjas 23 (1960–61): 95.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1080

McMullen

The writing of full-length histories needed settled conditions. Only Hsiao Ying-shih, the member of the group most active before the rebellion, showed an interest in the technical problems of history writing. Hsiao’s ambition before the rebellion had been to compile a history of the period from the start of Han until the end of Sui, a Li-tai t’ung-tien in one hundred chüan.126 His main objection to existing accounts was formal. He held that the proper organization for a history was the chronicle, with entries made chronologically and by region. The annals-biography model introduced by the Han historians was, he claimed, a presumption and a departure from the practice of antiquity as enshrined in the Book of History and the Ch’un-ch’iu. Ssu-ma Ch’ien and Pan Ku were therefore to be condemned. The structure of their works was repetitious, and they used inconsistent language to describe the same event in different sections of their works. The resulting lack of precision defeated the main purpose of historical writing, and would, Hsiao said, “be inadequate to promote the basic principles of government.” Later historians had followed the Han models, Hsiao complained, but had lacked the talent of Ssu-ma Ch’ien or Pan Ku. However, the chronicle style had not by any means died out. Hsiao mentioned ten writers of chronicle histories for the Period of Disunion and implied that he knew of several more. “But to the end they were unable to blunt the speartip of the Han historians’ presumption or continue even by a thread the Lu style of discussing events.” It had been Hsiao’s ambition to correct this situation, but he had failed to secure a post at the history office and had withdrawn to the southeast some years before the rebellion. An observer later recalled that he had been writing busily,127 but we know that he had abandoned his project before the rebellion broke out.128 This dispute about the ideal form for histories was a lively and longstanding one in T’ang times. At least four chronicle histories had been written in the late K’ai-yüan and T’ien-pao periods.129 The catalog to the imperial library which Wei Shu, a friend and correspondent of Hsiao Ying-shih, had compiled listed no fewer than fifty-five chronicles.130 Yet Hsiao regrettably left no detailed 126   c tw 323.15a–b; see also Hiraoka Takeo, Keisho no dentō, pp. 101–04. 127   c tw 395.3b. 128   c tw 323.16a–b, 317.4b. 129  Yüan Hsing-ch’ung’s Wei tien (hts 58.5b); Ch’en Cheng-ch’ing’s Hsü Shang shu (for which see ctw 322.15a and hts 57.3b); Wu Ching’s T’ang Ch’un-ch’iu and Wei Shu’s work of the same name (hts 58.3b). 130  This figure, quoted from cts 46.17b, assumes that the cts I-wen chih derives from the Kuchin shu-lu of Wu Ching, itself a condensation of the catalog in 200 chüan compiled by Wei Shu and others and submitted in 721 (cts 46.1b); see cts 46.4a and Piet van der Loon, “On the Transmission of the Kuan tzu,” T’oung Pao 41, nos. 4–5 (1952), p. 368 and n. 3.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1081

reasons for his preference, other than the canonical status of his model, the Ch’un-ch’iu, and his insistence on the utmost brevity in narrative. Perhaps in implying that the chronicle, because it is closer to the Ch’un-ch’iu, is more efficient at dispensing praise and blame, he gives some clue to his position. Hsiao seems to have seen historical writing as fulfilling a narrowly moral task. The pien-nien, or chronicle, form perhaps performed this function better because it concentrated on showing men in action. It combined in one comprehensive yet concise narrative the essential matter of both the pen-chi and lieh-chuan sections of the conventional chi-chuan histories. Moreover, it avoided the specialist technical issues to which the treatise sections of conventional histories were devoted. If Hsiao thought along these lines, stressing moral action over technical information, then he was adopting, even before the rebellion, a position generally characteristic of the intellectuals under review. The third and most typical kind of history the five men wrote reflects precisely this greater interest in men than in processes or in technical knowledge. It involved the evaluation of historical figures and their moral roles. This was again a traditional activity; essays rating the respective merits of groups of figures may be found as far back as the early Six Dynasties.131 The appreciation, or tsan, essays of the early T’ang dynastic histories provided more recent models.132 As Hsiao Ying-shih suggests, criticism of this kind seems to have enjoyed a vogue in mid-T’ang times, even before the rebellion.133 The two best surviving examples of this form are essays by Yüan Chieh and Tu-ku Chi. Yüan’s Essay on Kuan Chung, written in 757, was an attempt to correct what its author thought an excessive respect for this figure among contemporaries.134 It is not difficult to imagine why Kuan Chung should have appealed to many T’ang thinkers and to see why Yüan should have been anxious to redress the balance.135 The Kuan tzu combined, as no other work did, an acceptable moral position with a frank discussion of the techniques of control and administration. Kuan Chung had been a very successful practitioner of 131  E.g., Ts’ao Chih, “Essay on the Respective Merits of the Two Han Founding Emperors,” in Yen K’o-chün, Ch’üan Wei wen (Canton, 1887–93) 18.1b. 132  E.g., Sui shu 75.21b, 76.20a. 133   c tw 323.13b. 134   c tw 382.9b–12a. 135  The Kuan tzu had had two commentaries written for it during the early eighth century; see W. Allyn Rickett, Kuan tzu, a Repository of Early Chinese Thought (Hong Kong, 1965), pp. 16–20. Li Han, a cousin of Li Hua, had written an essay very similar to Yüan’s, evaluating Chu-ko Liang’s claim to be the equal of Kuan Chung and Yüeh I; see ctw 431.2b–5a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1082

McMullen

his own ideas and had been commended, though in less than fulsome terms, by Confucius himself.136 But Yüan Chieh, perhaps most of all the group, analyzed political problems relentlessly in moral terms. There were parallels, he suggested, between Kuan Chung’s era and the situation in which China found herself after the rebellion—both were times when the central authority had lost power. But to Yüan it was wrong that Kuan Chung should be admired or that it should be thought that his example was relevant to the present situation. He had in fact failed morally because he had not devoted his energy to the weakening Chou dynasty; instead, he had dedicated his powers solely to the interests of the state of Ch’i. This failure, Yüan argued, had been a disservice to China. For had Kuan Chung put his able leadership behind the Chou feudal organization, then, despite their weakness, “the Chou emperors would never become slaves, nor would the states of the feudal lords have been destroyed, nor would Ch’in’s position in the empire have advanced as far as it did.” A second essay, which demonstrates even more clearly a wish to revise conventionally accepted evaluations of historical figures, is Tu-ku Chi’s Essay on Chi Cha of Wu.137 Chi Cha was an exemplar of the Ch’un-ch’iu period, primarily known for his ability to predict the fate of states by listening to their music. Like Kuan Chung, he seems to have been discussed in the mid-eighth century.138 Tu-ku Chi, however, was not interested in Chi Cha for the role in which he was best known, but rather for his conduct in refusing three times to accept the throne of the state of Wu.139 The principle of declining or abdicating a throne was, he showed, to be thought of as relative. The only overriding, absolute consideration involved was that the best possible ruler be on the throne. “Hence for the sake of a man’s wisdom one should disregard his age, for the sake of his right conduct one should disregard auguries against him, and for the sake of his ruler’s commands one should disregard the requirements of ritual in his case.” 136   Analects 14.16, 3.22. 137   c tw 389.10a–11a. 138  Yüan Te-hsiu, Yüan Chieh’s cousin, had written an essay on him, since lost, but quoted by title by Li Hua (ctw 320.16b). Li mentioned him and, probably in 766, visited and left an inscription in his honor at one of the temples that stood to him in the area of Jun-chou and Ch’ang-chou; see ctw 315.10a and the Sung dynasty geographical handbook Yü-ti chisheng (Taipei, 1962 reprint) 7.17b; this text is not included in Li’s works in ctw and CTShih. Tu-ku Chi held his final post, the governorship of Ch’ang-chou, in the area where Chi Cha had lived, and where his tomb still stood. 139  He had also been alluded to in this context by Ts’ao Chih; see Huang Chieh, ed., Ts’ao Tzuchien shih chu (Peking, 1957), p. 32. Li Hua’s inscription, cited above, read, in part, “Chi tzu declined the throne of Wu, but by his writing perpetuated the Kuo-feng [section of the Odes].”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1083

It was wrong, Tu-ku considered, when Chi Cha was the ablest of his family, for him to disobey his father and brothers and refuse the throne. He had “preserved his own integrity,” withdrawing in a way that was usually praiseworthy, but had neglected the demands of this particular situation. If he had accepted the throne, “he would certainly have been able to bring luster to the Chou and to hold hegemony over the Ching and Man barbarians. Thus the great heritage would have been secure and many difficulties would have been avoided.” The parallels with Yüan Chieh’s essay are evident. Yet Tu-ku is the more incisive, since he revised the implicit judgments on Chi Cha of the Ch’un-ch’iu, Tso chuan, and Shih-chi.140 Both essays discuss moral problems in history. Kuan Chung’s conduct is analyzed in terms of its consequences, as well as by the absolute requirement of loyalty to the ruling house. In Tu-ku’s essay the results of Chi Cha’s action are considered, but they are secondary to the insight that moral principles may be relative rather than absolute. Yet these essays also show a characteristic limitation. In the West such factors as the results of an action and the absolute or relative nature of moral precepts are recognized as general ideas. Ethics, the science of morals, which supplies these ideas, has been a subject in its own right, with its own body of theory. But here there is no comparable tradition of ethics to draw on, and the level of generalization is low. In Yüan’s essay the general principles have to be inferred from the particular case under review. In Tu-ku’s they are enunciated, but placed crudely at the start of the essay as normative rules, rather than as its final deduction. The approach of these essays to moral problems in history may therefore be characterized as personal and individual, in the sense of always requiring a specific figure as the principal object of inquiry. Without their specific examples before them, Tu-ku and Yüan would not have been able to argue as they did, for they could not have given their warning messages in generalized abstract form. This approach to historical analysis stands in contrast to the other main approach we have described, that of seeing history as an impersonal process susceptible to periodization, or of analysis in terms of wen and chih. The personal approach was evidently very natural to the intellectuals under review, since a similar concern for the moral role of the individual is to be seen in many of 140  The Ch’un-ch’iu itself only once mentions Chi Cha by name, stating merely that he paid a ritual visit to Lu; see Ch’un-ch’iu ching chuan, Hsiang 29, p. 326. The praise (pao) Tu-ku mentions is presumably that implicit in the Kung-yang chuan, Ku-liang chuan, and Tso chuan commentaries on this passage; see ibid., pp. 326–27. There is an account of the whole episode in Shih-chi (Po-na edition) 31.3b–13a; Ssu-ma Ch’ien’s opinion of Chi Cha is given in Shih-chi 31.20a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1084

McMullen

their random remarks on history, remote and recent. The same kind of moralism underlay the biographical and commemorative works they produced for their contemporaries with such great facility and in such great numbers; and the same approach was, we shall see, integral to their literary theory as well.

Literary Criticism

The high level of knowledge and keen sense of historical perspective that informed the historical criticism of these five men is present also in the second aspect of their thought which is the concern of this chapter, literary criticism. The two fields moreover shared some key concepts and had a few important differences, which throw interesting light on the way they looked at their literary heritage. Literary criticism may be divided broadly into three types: the prescriptive, which formulates rules for genres, prosodic models, and so on; the descriptive, which concentrates on the exposition of existing works of literature; and the theoretical, which discusses such questions as the nature of literature, its psychological origins, and its functions.141 Prescriptive criticism, the formulating of prosodic rules, was more typical of the early eighth century, when the regulated styles of verse were being perfected, than of the rebellion period. Only one of the five men under review, Yen Chen-ch’ing, seems to have been even marginally involved in this kind of criticism. When he resumed work on his rhyming dictionary, nearly twenty years after the outbreak of war, he had as a companion a monk who wrote a work of prescriptive criticism.142 Together, Yen, the monk, and others composed lienchü, one of the most technically exacting of verse forms.143 But other members of the group disapproved of excessive concern for the intricate requirements of the regulated styles and gave little hint of interest in this sort of criticism. Descriptive criticism, the exposition in detail of existing works of literature, is not a feature of T’ang critical writing. There was no parallel in literature to the detailed treatment of historical figures like Chi Cha or Kuan Chung. But in their prefaces, the men under review did pass epigrammatic comments on past literature and on each others’ works. These remarks, though laconic, are

141  George Watson, The Literary Critics (London, 1962), pp. 9–18. 142   c tw 339.7b; and Lo Ken-tse, Chung-kuo wen-hsüeh p’i-p’ing shih, vol. 2, pp. 39–45. 143   CTShih, vol. 11, pp. 8880 ff.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1085

useful, for they set an author in his literary historical context and suggest, for example, whether he was appreciated for moral or for aesthetic reasons.144 Finally, there is theoretical criticism, which is concerned with the nature and function of literature, the psychology of composition, and so on. This is perhaps the characteristic critical activity for the two decades after the rebellion and for the intellectuals under review. Here they reformulated some of the most striking insights of the Chinese literary critical tradition, focusing on the relation of literature to society and the psychological genesis of literature. The cornerstone for their position on the theoretical aspects of literature was again the versatile concept of wen. The breadth of this idea has already been suggested. First, it was an objective phenomenon in society, morally determined, to be noted as an index of the condition of the state. And second, it was a component in an expressionistic psychology, the patterning of the inner feelings through the media of music, ritual, or the written word. Wen in the first sense was the concept Li Hua had treated in his “Essay on Substance and Pattern.” But though wen could apply to ritual and music, as it did in this essay, it had always had special reference to the written word. The very term had a specific substantival meaning, translatable by “literature” in the wide sense, marking it off from ritual and music. And the most influential account, the Mao shih ta-hsü, shows that literature, most of all the three media to which the wen concept referred, was envisaged as reflecting the state of the government and the feelings of the people.145 The condition of the government and the populace, moreover, changed with time, and literature reflected these changes. In modern terms, this amounts to a primitive literary historicism: the nature of literature is related to the condition of the society that produced it. The intellectuals under review endorsed this traditional idea, though they never made very direct statements about it. They followed early T’ang writers in upholding a distinction between the literature of “those above,” that is the emperor, government, and themselves, and that of “those below,” meaning broadly the people. The former was thought of as an expression of the rulers’ moral desires, and as having a transforming effect on the society that received it. The latter was the spontaneous response of the people to government.146 It was naturally the latter, the literature of the governed, which best illustrated 144  E.g., ctw 388.11b and especially Yin Fan in Ho-yüeh ying-ling chi, in T’ang-jen hsüan T’ang shih (Peking, 1958), pp. 40–124. 145   Mao shih 1.1b–2a. 146   Sui shu 76.1a makes this distinction very clearly. An earlier source for the same idea is Pan Ku, Wen hsüan (sptk edition), 1.3a, quoted by Chu Tzu-ch’ing, Shih yen chih pien (Peking, 1956), p. 76.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1086

McMullen

the traditional concept of literature as socially and politically determined. But the literature of those who ruled, as an instrument for the moral transformation of society, was also thought of as closely related to the age in which it was produced. Yen Chen-ch’ing, for example, held that in high antiquity the emperors, that is “those above,” had illustrated their great virtue as rulers through their wen, compositions, and that these in turn had had a profound effect on society. “When Shun composed his song, prince and subject changed their demeanor. When the royal grace expired, reform of customs did not proceed. The flourishing or decline of government is in truth bound up with this.”147 Yen went on, rather brazenly, to introduce a brief survey of past literature by taking up the wen concept in its descriptive rather than substantival sense and restating the idea of a wen-chih polarity as a means of analyzing literature. A balance was ideal, but “age has followed age, and none has been able to find the mean.” Li Hua referred to the historicist idea in slightly clearer terms when he stated that two factors, the character of the author and the mood of the age, were determining forces in literature. “Literature has its origin in the author, but the joy or sorrow it expresses are bound up with the times…. It reflects rejoicing over kings Wen and Wu, and grief over kings Yu and Li.”148 The five men were especially interested in the figure of Chi Cha, who, as we have seen, had used the literature of “those below” to diagnose the condition of the states in the Ch’un-ch’iu period.149 They referred also to the kuo-feng principle, by which popular songs were to be collected and studied as indications of the attitude of the people to the government. This idea was beginning, in the mid-eighth century, to be taken seriously again on the creative level. Yüan Chieh, who led this development, wrote a series of imitation ballads and collected folk songs in the lower Huai Basin before the rebellion.150 This literary historicism, rudimentary though it appears, was unparalleled in the post-classical West until the eighteenth century, when Vico related Homer’s Iliad and Odyssey to the changing society that had produced them.151 The theory, however, was never worked out in detail. It is not clear whether the five writers took it at the most primitive level and thought literature good only to the extent that it reflected good government; or whether their historicism 147   c tw 337.10b–11a. 148   c tw 315.4b. The clearest statement of this idea comes slightly later, ibid., 518.2b, 3b. 149  See above, note 138. 150   CTShih, vol. 4, pp. 2696–98, 2703. Yüan also suggested that his own verse describing popular conditions should be presented to the central authority (ibid., vol. 4, pp. 2697, 2704). 151  Edmund Wilson, The Triple Thinkers (London, 1952), pp. 244–45.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1087

was more developed, and they thought that only a good political age allows the production of good literature—in other words, that literature has features of its own, to be judged independently of its social base, but that a good social base is prerequisite for literature of any excellence. Confronted with this vagueness, it may be helpful for us to compare the remarks the five writers made on past literature with those they made on history itself. It was a long-established practice in prefaces to collected works or anthologies to review past literature and to set the author in his literary historical context. Reviews of this kind are to be found in prefaces by Li Hua,152 who also recorded Hsiao Ying-shih’s position,153 Yen Chen-ch’ing,154 and Tu-ku Chi.155 Their scheme for literature was indeed parallel to their scheme for history itself. Just as they considered the canonical age one of social and moral near-perfection, so they insisted that canonical literature was uniquely authoritative and that the process of decline started thereafter. “Confucius’s writings were transmitted by Yen and Shang. When they had died, K’ung Chi and Meng K’o wrote, and may be said to have handed on the six canons. But Ch’ü P’ing and Sung Yü were mournful and distressed, and extravagant beyond recovery. The way of the six canons lapsed.”156 None of the men under review provided an elaboration of this well-known judgment or described in detail why the canons had such great authority. We may infer that they appreciated the Odes at least partly for literary reasons because so much of the technical critical vocabulary they used, words like feng-ya and pi-hsing, stemmed from them. But it is clear that Li Hua, who wrote most about this, was more concerned with the connection between the Odes and the society that produced them than with their technical quality as literature: “When the kings of old neglected the mandate of Heaven, the Grand Master expounded poems for them to observe the feelings of the people. Emphatically these were not about ‘pines and snow on distant peaks,’ or ‘cloud and moonlight over limpid rivers.’ ”157 Li saw some of the Odes as reflections of a perfect administration and society, and others of a period when the ancient rulers “neglected the mandate of Heaven,” that is, of a government that fell short of perfection but of a populace that still produced exemplary poems of criticism. 152   c tw 315.4b–5a. 153   c tw 315.8a–b; another slightly different review by Hsiao Ying-shih is included in his hts biography (hts 202.13b). 154   c tw 337.10b. 155   c tw 388.1a. 156   c tw 315.4b. 157   c tw 315.14a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1088

McMullen

Moral factors in the society of early China, implied Li, were essential to the excellence of the Odes. Ch’ü Yuan and Sung Yü, the principal authors of the Ch’u tz’u tradition, marked the start of the age of decline. Though their writing was, in Hsiao Yingshih’s words, “very virile and robust,” it “could not be counted as canonical.”158 It was, as Li Hua remarked in the quotation above, “mournful and distressed,” in contrast to the Odes, which Confucius had commended for being “mournful but not distressed.”159 The later Chou was, we saw in Li’s and Hsiao’s reviews of history, a period of general decline. History and literature ran parallel. The Ch’in was barren as far as literature was concerned and was not discussed. But the Han was another matter. We have seen how Li Hua had characterized it as an age when chih had exceeded wen, and Hsiao Ying-shih considered it one of the three great ages of history. We would expect, then, that its literature should reflect its general excellence, and we do indeed find the intellectuals under review praising Han authors. But their commendation is typically less enthusiastic for Han writers than for the Ch’u tz’u authors, as if they represent a further stage in a continuous process of decline from the canonical age. Thus Hsiao Ying-shih, having stressed that their writing “does not approach the Feng and the Ya [sections of the Odes],” singled out Han writers and commended them briefly on technical points, while Tu-ku Chi was said to have imitated the Han writers Tung Chung-shu (c. 179–c. 104 bc) and Yang Hsiung (53 bc–ad 18).160 After the Han dynasty, the five men might be expected to run into some sort of conflict if they held to a primitive historicist theory of literature. For it would not seem reasonable for them to dismiss all the literature of the post-Han and pre-T’ang period as contemptuously as Hsiao Ying-shih had dismissed its history. They all knew that major technical developments had taken place since the Han and that the verse of the Wei and Chin, disastrous periods politically, was particularly vigorous. Here then literature and history must have seemed to part company, in a way that would challenge any simple historicist theory. In fact, the writers under review seem to have dodged this problem by considering the Wei and early Chin as an extension, from the point of view of periodization, of the Han. They may have seen political unity—the existence, however precarious, of one nominal government for all China under the Chin until the loss of the north in 317—as their justification for this position. At any rate they commended the Wei and Chin in general terms and praised the poets 158   c tw 315.8a. 159   Analects 3.20. 160  For Hsiao Ying-shih, see ctw 315.8a–b; for Tu-ku Chi, see cts 160.1a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1089

of the late Han, like Wang Ts’an (died 217); of the Wei, like Ts’ao Chih (died 232); and of the early Chin, like Chang Hua (died 300), Tso Ssu (died 306), and P’an Ni (died 310). Significantly perhaps, the last names to receive commendation were Liu K’un, who died in 317, and Kan Pao, who was also writing just at the time when the northern homeland was lost to barbarian invaders and China became divided.161 In their treatment of the Period of Disunion itself, the five men consistently dismissed its literature as emphatically as they had its history. Hsiao Ying-shih, breaking off after mentioning Kan Pao, remarked, “Beyond these the remainder are far behind and without renown.”162 And he noted elsewhere that he “never let his attention be detained by any literature since Wei and Chin.”163 Yen Chen-ch’ing censured the verse of the Liang and Ch’i and the kung-t’i, or Palace style, for its triviality.164 Only Tu-ku Chi implicitly admitted the positive technical developments of the Period of Disunion. He did this grudgingly, merely pointing out that mid-T’ang verse owed its special excellence to its fusion of ancient qualities with more recent tonal rules. His review of five-word verse, “passing over a thousand years,” jumped from Han and Wei to the seventh century.165 Since these writers viewed the T’ang as a great dynasty, and the seventh century as a time of order and prosperity, we would expect them to praise the literature of that time as well. But here they seem to have operated the same sort of double standard that they used in interpreting history itself. Before the rebellion Yüan Chieh and Hsiao Ying-shih had noted the high level of literature Hsüan-tsung’s reign had brought about, and Li Hua too recalled that this had been a time when “in the field of letters, men of great ability abounded.” But after the rebellion the members of the group were on the whole reluctant to praise any early T’ang literature. Perhaps this reluctance to bestow approval on seventh- and early eighthcentury literature is to be explained partly by the idea that there was a time lag between the establishment of the T’ang and the creation of a T’ang style of writing to reflect the improved conditions. This explanation had been used by early T’ang historians reviewing the literature of the Sui period.166 Yüan Chieh 161   c tw 315.8b; hts 202.13b. The one exception to this scheme is Hsiao’s praise for the Liang dynasty historian P’ei Tzu-yeh as being a “good writer” (hts 202.13b). 162   c tw 315.8b. 163   c tw 323.12a. 164   c tw 337.11a. 165   c tw 388.1a.–b. 166   Sui shu 76.2a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1090

McMullen

hinted at it when he remarked after the rebellion, “Writers of recent generations have adhered to inherited practice.”167 But whatever their reason for so negative an attitude toward early T’ang literature, the men under review show striking agreement in singling out one figure in the seventh century for commendation. This was Ch’en Tzu-ang, the official and writer of the empress Wu’s reign. Shortly after his death, a contemporary had claimed that through him, “wen and chih had undergone a change.”168 This praise was reiterated by Hsiao Ying-shih,169 Tu-ku Chi,170 Yen Chen-ch’ing,171 and, implicitly, Yüan Chieh,172 besides a number of other eighth-century writers.173 This impressive consensus highlights a paradox in the attitudes these men had toward literature. Up to this point in their reviews they had treated it as a general phenomenon running more or less parallel to history itself. But now this broad characterization is abandoned, and a single, exceptional author is selected for commendation in contrast to his age. The key to this paradox is almost certainly to be found in the two ways these writers looked at history and, indeed, at the wen concept itself. History, it will be recalled, was both considered as an impersonal, cyclic process and also analyzed in terms of specific individuals involved in it and in terms of their actions. These two approaches to history seem to have been paralleled exactly in literary criticism. There was the broad historical review of literature, when the writing of successive ages was implicitly related to its historical background. And there was the more detailed consideration of individual authors, often in contrast to the age that had produced them. But this more detailed approach occurred only in prefaces, which were usually reserved, by convention, for the collected works of the recently deceased. The parallels in historical writing are the biographies and other tributes to dead colleagues that these intellectuals wrote in such quantity. The desire to comment on the collections of their contemporaries and on their own creative experience led these men to pay attention to an aspect of Confucian literary theory distinct from the question of literature’s social 167  This idea was more clearly expressed slightly later (ctw 518.3b). 168   c tw 238.4a. 169   c tw 315.8b. 170   c tw 388.12a. 171   c tw 337.11a. 172   c tw 381.15b; CTShih, vol. 4, p. 2711. Yüan Chieh’s remarks echo Ch’en Tzu-ang’s and are probably modeled on them. 173  Li Yang-ping, writing on Li Po (ctw 437.13b); Li Chou (ibid., 443.16a); Liang Su (ibid., 522.6b). Li Po modeled his ku-feng series of poems (CTShih, vol. 3, p. 1670) on Ch’en’s kanyü shih (ibid., vol. 2, p. 889). Also Liang Su (ctw 522.6b) and Chao Tan (ibid., 732.1a).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1091

origins and effect on society. They turned to reexplore a second area in which the Chinese tradition had shown great insight, that of the psychological genesis of literature. Their position is the expressionistic one mentioned earlier in connection with the wen concept. Tu-ku Chi formulated their position in greatest detail: “When emotion overflows it finds expression in words. When words overflow they find form as wen. The feelings move within and take form in sounds. This is an aspect of wen which is still intangible. They [i.e. words] are made to blaze forth as song or hymn, spread abroad as deed or act. This is the aspect of wen which is fully apparent.” Li Hua wrote similarly, “Emotion that is declared forth is called words; words that are ornamentally wrought are termed wen.”174 Literature to members of the group was thus the patterning, wen, of the inner feelings, first in the medium of sound, and then in writing itself. The terminology here stemmed from brief statements in the Classics, the Book of History, the Tso chuan, and the I ching.175 These had been brought together in the influential Mao shih ta-hsü, to which, in effect, Li and Tu-ku were alluding. The Mao shih ta-hsü had said, “Poetry is concerned with the movement of emotion. When in the mind, it is emotion; when expressed as words, it is poetry.” This was a well-known graphic pun; the character shih, “poem,” consisted of the speech radical and the word chih, “emotion,” which in turn was divisible into the heart radical and a verb meaning to go. Poetry, and indeed all literature, wen, in the passages by Tu-ku and Li cited above, was therefore the verbal expression of movement in the heart or mind, or of emotion. The word chih, “emotion,” or “earnest thought” in Legge’s translation,176 which figured in these early definitions of poetry, had in the Analects and other early sources a strongly positive moral connotation. Tu-ku Chi, Li Hua, and other T’ang writers preferred it and the definitions of poetry and literature in which it occurred to rival terminology from a later definition of the shih. In the Chin period, the critic Lu Chi (261–303) had said, “Poetry traces feelings gracefully.”177 It was felt that “tracing feelings” (yüan ch’ing) in the context of poetry had less force than “expressing emotion” (yen chih). Lu Chi’s definition 174   c tw 388.2b, 315.4b. The word tsu, strictly meaning to fill up, is here translated freely as to overflow. 175   Shang shu 1.11a; Tso chuan, Hsiang 25, p. 307; I ching, hexagram 2 (k’un), 1.6b–7b. Wilson, The Triple Thinkers, p. 254, has a remarkably similar definition of lyric poetry, “A lyric gives us nothing but a pattern imposed on the expression of feeling.” 176  James Legge, The Chinese Classics, vol. 3 (Hong Kong, 1865), p. 48, and vol. 4 (Hong Kong, 1871), prolegomena, p. 34. 177  Lu Chi, Wen fu, in Wen hsüan 17.6a; cf. Achilles Fang, “Rhymeprose on Literature—the Wen-fu of Lu Chi (ad 261–303),” hjas 14 (1951): 536.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1092

McMullen

was almost certainly linked in the minds of Tu-ku Chi and others with the excesses of the verse of the Period of Disunion, and with the aesthetic approach to poetry associated with the regulated styles. Tu-ku noted of the latter that they “perfected the task of tracing feelings gracefully” but insisted that they were “very far removed from the Ya [section of the Odes].”178 He and Li Hua preferred the canonical terminology, therefore, because of its strongly moral implications. Their position, since it implied that writing which results from the true expression of the inner feelings would be good, is optimistic about human nature, and therefore truly Confucian. This expressionistic account of the origins of literature is to be seen as a concomitant of the individual and personal approach of these writers to the authors whose collected works they prefaced. For, as in the West, an expressionistic emphasis in criticism went with an interest in the character and biography of the author. It seems to have been assumed that good people would produce excellent writing.179 This assumption justified the many fulsome tributes these men paid one another in prefaces whose formal function was to introduce the written works rather than their authors. It also lay behind their belief that literature was a legitimate means of self-advancement, of, in Li Hua’s words, “establishing one’s person and promoting one’s name.”180 The literature of the gentleman (chün tzu) was thus, Tu-ku Chi observed, a fitting bequest to posterity, and a permissible way to win everlasting fame.181 The five writers under review all insisted on what amounts to the essence of their position, that literature had to be explicitly monitory, didactic, or exhortatory. Even Yüan Chieh, the most adventurous san-wen writer of the group, insisted on this, in a preface to his own collection: “My purpose has always been to encourage loyalty and filial piety, to urge fairness and uprightness, to induce charity and compassion, and to lead people to preserve integrity and 178  See Chu Tzu-ch’ing, Shih yen chih pien, pp. 1–42, for a discussion of the distinction between shih yen chih and shih yüan ch’ing, especially in the Period of Disunion. In early T’ang times, the yüan-ch’ing definition of the shih was disapproved of by Wang Po (ctw 182.15b); in mid-T’ang times implicitly by Jui T’ing-chang (ibid., 356.23b), Shang Heng (ibid., 394.20a), Tu-ku Chi (ibid., 388.1a) and Li Hua (ibid., 315.4b, 9b and 13b). 179  Confucius had in fact made a distinction between the man and his words; see Analects 15.23. This is quoted wryly by Wei Cheng (Sui shu 76.2b), conceding that Sui Yang-ti, though an evil man, wrote good literature. Such a position is not to be found in the writers under review. Li Hua, for example, wrote in a preface (ctw 315.7a), “When you see his writings, you will know his life,” and Tu-ku Chi (ibid., 388.2a), “When you see his compositions, you will know his values.” 180   c tw 315.4b. 181   c tw 388.4b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1093

accept their lot in life.”182 This same austerity of purpose also informed their demands for style. Simplicity and plainness were required both before the rebellion and, with greater insistence, after it. Tu-ku Chi spoke metaphorically of this requirement in a well-known passage. To indulge obsessively in ornate or intricate stylistic devices was “like having an orchid as a boat and a kingfisher feather as an oar. One may toy with them on dry land, but they cannot be used for crossing rivers.”183 Yüan Chieh,184 Hsiao Ying-shih,185 Li Hua,186 and Tu-ku Chi condemned writing that merely demonstrated powers of description or technical dexterity. Tu-ku spoke against those who “take the eight faults [to be avoided] and four tones as if they were manacles and observe them as if they were honoring laws or commands.”187 Their seriousness of purpose led these writers to disdain one of the major literary developments of their day, the growing tradition of fiction. Perhaps their attitude was influenced by their interest in historical criticism, and they felt, as Liu Chih-chi had, that tales of the exotic and supernatural (later called ch’uan-ch’i) were not to be classified with the works of serious historians.188 They may also have wished to observe the traditional Confucian injunction against interest in “extraordinary things, feats of strength, disorder, and spiritual beings.”189 But it is interesting that a theoretical justification for recording exotic or supernatural occurrences, the sort of material basic to many stories of the period, was available to them. This apology is used by Ku K’uang, a late contemporary of the group. Exotic happenings were evidence of disturbance in the pattern of heaven, t’ien-wen and were therefore to be collected and studied just as jen-wen was in the human sphere.190 But this idea is never mentioned by any of the men under review. Probably if they had any justification for fiction of any kind, it would have relied on the concept of indirect protest, of giving admonition through fable or allegory. The fable occurred often in Chuang tzu and other pre-Ch’in texts and therefore commanded literary

182   c tw 381.15b. 183   c tw 388.11b. 184   c tw 381.16b. 185   CTShih, vol. 3, p. 1594. 186   c tw 315.4b, 7a. 187   c tw 388.12a. 188   Shih-t’ung t’ung-shih 18.3b. 189   Analects 7.21, quoted by Aoki Masaru, Shina bungaku shisō shi, p. 17. 190   c tw 528.13b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1094

McMullen

prestige. Satirical animal fables by Li Hua are the nearest any of the group came to writing fiction.191 Conclusion This chapter has explored the historical and literary ideas of a small group of intellectuals active over the An Lu-shan Rebellion period. These men came from aristocratic backgrounds and lived in a time of decline for aristocratic power and of sudden military crisis. Their intellectual outlook was conservative. It could be said that there were few new elements in their thought, for most of their theoretical ideas may be traced back through early T’ang to roots in the Han dynasty or earlier. But they wrote more freely and personally than their predecessors about their literary experience, and more critically about the past. One result was that the wen concept in its full range, which lay behind much of their historical and literary theory, was as prominent in their writings as in any since Han times. We have seen that wen in the writings of the men under review could refer to a phenomenon in the universe, a value in society, literature itself, or a value in literature. A modern critic may perhaps feel disappointment that this versatility was not more analytically explored. We should be on our guard, however, against making anachronistic demands of these writers. Their main aims in writing were to express disapproval of the contemporary political world and of current literary practice and to praise one another for their efforts to put things right. In effect they were invoking the authority of traditional ideas rather than subjecting them to critical analysis. They would not have felt the need to provide a more systematic description of wen. The wen concept was, nonetheless, the main Confucian theoretical idea about which they wrote, and it would be difficult to illustrate any other philosophical term from their collected works in such detail. This is first and foremost an indication of the importance of literature and history to these men. Since their identity depended so much on these activities, it was natural that they should have discussed them as they did. But secondly, and more tentatively, their liking for wen reflects the character of the concept itself and the intellectual milieu to which it appealed. We have seen that wen was either substantival, meaning culture or the written word, or a hypostatized descriptive 191   c tw 316.20a, 318.5b. There are also some brief examples of historical fiction, set in the pre-Ch’in period, by Li Hua (ibid., 318.1a–4a) and Yüan Chieh (ibid., 380.1a–7a). These are likely to have been acceptable for similar reasons.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Historical And Literary Theory In The Mid-eighth Century

1095

term, meaning the patterned, cultured, or refined. In none of these senses did it present the analytical problems that arose in the more metaphysical or psychological ideas of Confucianism, or in the increasingly important common ground between Confucianism and Buddhism. Wen did not, moreover, unlike much Confucian psychological terminology, have to be reconciled to other closely connected but conflicting terms. Its moral value was clear. It had a descriptive, objective, almost tangible quality. Its prominence in the writings of the men under review thus suggests the corollary that other, more involved Confucian ideas were less immediate to them. They could endorse the idea of wen and make keen observations on literature and history. But they did not yet feel the need to discuss the complex and introspective philosophical ideas that exercised Confucians so much from the late eighth century on. To this extent it is right to describe their writings and the post-rebellion decades through which they lived as preparatory for the expansion of intellectual life that followed. Despite these reservations, the account of wen abstracted from the writings of the men under review is detailed enough to provide a specific example of a well-known generalization often applied to the Confucian speculative tradition. It was, like its medieval European counterpart, predisposed to make elaborate correlations between man’s immediate experience and the universe and between the moral and physical worlds. The concepts by which it did this were flexible. The intellectuals discussed here referred to wen in different senses according to context and quoted as their authority the brief and scattered canonical sayings from which these senses derived. But we know that they were aware that in the background all senses of the term were related. It was indeed as “a great unifying principle” that Hsiao Ying-shih extolled the perfection of wen.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy of the Tang: The Case of Literature Stephen Owen Tang dynasty literature can never be fully separated from the question of how we happen to have what we have. Changes in taste and canon are important, but those changes are constrained by what survives, as much as they earlier shaped what has survived. Our “Tang” does not look like the choices and preferences of any particular moment in the Tang dynasty itself; nor do we have the luxury of looking at Tang literature as a whole and making up our own minds. The closer one looks, the more one finds that the survival of Tang literature is a conjunction of historical accident and the period taste of the ninth and tenth centuries.1 Our knowledge of the Tang comes filtered through the peculiarities of manuscript circulation and the turbulence of a single era, the near century between the Huang Chao Rebellion and the final assemblage of the largest regional libraries in the Song capital at Bianjing (modern Kaifeng) that took place right after the conquest of the Southern Tang in 976. So many texts have survived that we are perhaps disinclined to question the processes by which they survived; however, there are so many cases in which writers we now consider important survived by a tenuous but traceable thread that it is worthwhile reexamining how some things were lost and some preserved. If we often find that Tang critical judgments do not match the writer’s existing works, those judgments may have been based on versions no longer extant. Source: “The Manuscript Legacy of the Tang: The Case of Literature,” Harvard Journal of Asiatic Studies 67 (2007): 295–326. 1  The “period taste of the ninth and tenth centuries” was, of course various and is difficult to define in simple terms. Nevertheless, we can see its traces. For example, a clear preference for regulated verse poets of the late eighth century is evident in the extant anthologies and in critical works such as Zhang Wei’s 張為 Shiren zhuke tu 詩人主客圖, from around the turn of the tenth century. The works of ninth-century poets have been preserved rather well, often in independent editions rather than in editions reconstituted from anthology sources. At the same time men in this period showed little interest in many poets of the “High Tang,” who had been important in their own day and would be much admired in later ages. The poems of High Tang poets such as Chang Jian 常建, Cui Hao 崔顥, Qiwu Qian 綦毋潛, Cui Guofu 崔國輔, and Zu Yong 祖詠 have survived only in modest selections, preserved primarily in anthology sources.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_035

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1097

Modern Chinese scholarship meticulously cites early bibliographies and existing editions with their lineages; but it rarely touches on the difficult question of the manuscript tradition as it survived in the early Northern Song. I like the term flotsam and jetsam, the remainder of a shipwreck; it is an apt metaphor. Following Huang Chao’s occupation of Chang’an early in 881, the city was repeatedly sacked—by Huang Chao’s followers, by the city mob, and later by the loyalist forces that retook the city. The great imperial library, along with large private and temple holdings, was gone in a city left in ruins. The wars before and after this event devastated much of China. This period, lasting until the situation gradually stabilized in the mid-tenth century, wreaked havoc on the manuscript legacy of the Tang. So many manuscripts had been in circulation throughout China that a vast amount survived; the problem lay in what works survived and the form in which they survived. Texts that were popular, texts that people copied or paid to have copied, survived; but late ninth- and tenth-century taste did not necessarily favor what later bibliophiles and curators of culture saw as the “literary heritage.” The works of even popular authors often survived only as selections. The loss of works, once preserved complete only in the imperial library and perhaps in a few great collections, was stunning. Losing works was sometimes merely losing track of works, not knowing where the best and most complete versions could be found. But if those versions were not found before they rotted away, burned in a library fire, or were reused on the verso as ledgers by the pragmatic descendents of bibliophiles, they too were lost.

Building the Song Library

The regional states that rose from the ruins of Tang China must have built their own book collections out of provincial libraries and perhaps out of private and temple holdings as well. We know next to nothing about provincial libraries in the Tang, except that in the eyes of the new Song dynasty the libraries of Jingnan 荊南, the Latter Shu 後蜀, and the Southern Tang 南唐 were especially prized—even though these collections were no larger than those of some great private bibliophiles in the Tang. The Dunhuang trove was but one temple library in an independent Han Chinese principality, and it suggests not only the richness of the material that once existed but also something of the vagaries of library holdings.2 It is true that Dunhuang has added immeasurably to our 2  The Dunhuang trove, sealed around 1035, contains texts dating back to the fifth century; most of the manuscripts, however, are from the ninth and tenth centuries. Extensive reproduction

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1098

Owen

understanding of Tang literature; but it is useful to pose the alternative question: what if all we had of Tang literature was what had been found in the Dunhuang library? Gao Shi 高適 (ca. 700–765) would clearly be the most prominent Tang poet, followed by Wang Fanzhi 王梵志 (the Brahmacārin Wang).3 It is true that the interests of a temple library were different from those of the library of a regional state with imperial pretensions, but the regional state libraries were built out of provincial libraries, collections that had addressed local needs and had not been intended to represent the Tang literary heritage—until local dynasts came to see themselves as conservators of the Tang heritage. At the beginning of his Panizzi Lectures, “The Lost Books of Medieval China,” Glen Dudbridge gives an excellent overview of how the Song imperial library was built from the modest library of a regional state with somewhere between 10,000 and 13,000 scrolls, to which the Song added 13,000 scrolls that it took from the Shu Kingdom, the uncertain holdings of Jingnan and the other conquered states, and 20,000 scrolls from the Southern Tang. Dudbridge estimates holdings of the imperial library to have been 40,000 scrolls (excluding duplicates), which were used in the great compilation projects of the 970s and 980s.4 of manuscripts, especially of large works, depends on large scriptoria. Buddhist venues had a great advantage in preserving manuscripts in that every great monastery complex had a scriptorium. The central government had a scriptorium. We can assume that provincial and prefectural governments had scriptoria commensurate with their practical needs; however, the huge manuscript compilations and relatively large collections evident in the legacy of the imperial library catalog required a large scriptorium for preservation and dissemination. Provincial libraries probably lacked the scribal resources to reproduce the famous large compendia whose loss we now lament. That the mid-seventh-century court compilation Wenguan cilin 文館詞林 in 1,000 scrolls was preserved partially in Japan but not in China is probably the result of imperial foreign policy, which tapped the imperial scriptorium for purposes of cultural diplomacy. Prefectural and provincial libraries had neither the motive nor the resources to reproduce such texts. In Wenyuan yinghua we often do not find Tang works of then recognized cultural importance; at the same time we do find copious representation of examination poems and fu, as well as standard bureaucratic prose forms. This seems to represent the motives of provincial libraries, which were to help students preparing for the examination and to provide models for civil servants to participate in political life. 3  Wang Fanzhi, by legend a seventh-century lay Buddhist devotee, was the name to which several, rather different sets of popular poems were attached in the Dunhuang materials. 4  Glen Dudbridge, The Lost Books of Medieval China (London: The British Library, 2000), p. 2. I use the term “scroll” here for juan 卷. When I am certain I am dealing with printed works, I use the romanized term “juan.” Some late tenth-century books may have been printed, but the vast majority is in manuscript. The manuscript books may have been in scrolls, but many were probably in the increasingly popular folded scroll, often called “butterfly books,” the precursors of print format.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1099

Sources differ on the sizes of the regional libraries, but the aggregate number (including duplicates) of the Song imperial library holdings after acquiring the regional libraries is given as 80,000.5 A half century later the Chongwen zongmu 崇文總目 (General catalog for the Institute for the Veneration of Literature), compiled between 1034 and 1042, had only just over 30,000, including many large works by Song authors, which would have been presented after the completion of Wenyuan yinghua 文苑英華 (The flower of the park of letters). These 30,000 scrolls include the 3,500 scrolls of the so-called “four big books” (si da shu 四大書): Wenyuan yinghua, Taiping guangji 太平廣記, Taiping yulan 太平 御覽, and Cefu yuangui 冊府元龜. We face the perennial problem of Chinese bibliography: how do we account for the difference in numbers? One possible explanation might be the palace fire of 1015, in which many books perished. The best explanation, however, would seem to be that Dudbridge’s estimate is wrong due to one simple fact: libraries have mostly the same books.6 The larger numbers we see in the initial formation of the Song library represent an “inventory,” not a catalog of distinct titles as we find in the later Chongwen zongmu. The conquering Song armies did not come with a resident bibliographer familiar with the collection in Bianjing who chose needed titles (there is no indication that the library in Bianjing had a catalog at the time); they packed up 5  See Xu Zizhi tongjian changbian 續資治通鑒長編 (1957; rpt., Beijing: Zhonghua shuju, 1999), p. 422, and Shishi leiyuan 實事類苑 (Huangchao leiyuan 皇朝類苑), Siku quanshu edition [hereafter skqs], 31.5b. Dudbridge’s estimate of 40,000 different works, less duplicates, may be based on the assumption that the imperial library kept only one set of duplicates; it seems more likely that it kept everything it had, distributed in the three branches of the library. In addition to the sources cited by Dudbridge, there are other sources that claim the Southern Tang library was larger than 20,000 scrolls. An estimate of over 20,000 scrolls comes from the Xu Zizhi tongjian changbian, p. 422. The Shishi leiyuan, skqs, 2.7a, puts the Southern Tang library at 30,000. In Jingding Jiankang zhi 景定建康志, skqs, 33.1b, Zhou Yinghe 周應合 claims that there were 60,000 scrolls taken from the Southern Tang library, but this is a thirteenth-century source, and the figure seems inflated. 6  The numbers of works recovered during the Song dynasty from the libraries of the regional states more likely come from inventories rather than from bibliographical lists of distinct works. It is, moreover, quite likely that these regional libraries themselves contained duplicates. The history of Chinese textual conservation in the age of manuscript culture has many cases when violent upheavals drastically reduced very large collections of books in imperial libraries. Actual loss was due to the fact that many books in such libraries were rare books— often the only copy. Subsequent attempts to rebuild an imperial library were done primarily by acquiring texts that were in general circulation. When the government asked, as it often did, for donations of rarer books, it assembled such books in a central place, thus inadvertently making the literary heritage vulnerable to loss, as occurred in the 1015 fire in the Song imperial libraries.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1100

Owen

the books of the conquered kingdoms, sent them off to Bianjing, and celebrated the size of their acquisitions. In other words, if one adds Bianjing’s 10,000 scrolls, Chengdu’s 13,000, and Jinling’s 20,000, one gets 20,000 scrolls of distinct titles, plus several thousand scrolls that were in other regional collections but not in the Southern Tang collection. Equally significant in adding titles to the new Song library was the search for rare books, undertaken by Taizong’s edict of 984.7 This yielded only 1,228 scrolls that were not in the existing imperial collection.8 Dudbridge eloquently describes the sense of loss underlying this enterprise, comparing early Song book possessions with older catalogs, with loss as a perennial trope in Chinese bibliography. Let us then generously estimate the imperial library of the 980s as holding about 25,000 scrolls of distinct titles.9 Taking a major library fire into consideration, this figure, along with large early Song contributions, would give us roughly the 35,000 scrolls of distinct titles that we find half a century later in the Chongwen zongmu. Let us add another fact. Most of the books in the imperial libraries were not literature. In the “Bibliography” of the Sui shu just over 7 percent of the items belonged to the literature section ( jibu 集部). The Xin Tang shu “Bibliography” fares better with just under 23 percent in the jibu or 5,750 scrolls (about the same percentage as in the 721 catalog). We must, however, keep in mind that a significant proportion of those works are pre-Tang, and the sizes of pre-An Lushan collections were large and apparently copied from earlier bibliographies. To know the actual holdings of the library, the Chongwen zongmu is by far the most reliable source.10 The Chongwen zongmu lists, by my count, approximately 2,000 scrolls of Tang individual literary collections (full collections, poetry collections, fu collections, and prose collections in various 7  The compilation of the 721 catalog is discussed in Denis Twitchett, The Writing of Official History Under the T’ang (Cambridge: Cambridge University Press, 1992), pp. 231–32. This was the book list used in the “Jingji zhi” 經籍志 of the Jiu Tang shu 舊唐書. In other words, when the Song library was being constituted there was apparently no extant bibliography for literary writings of the preceding two and a half centuries. 8  Xu Zizhi tongjian changbian, p. 178. 9  If this seems too modest a figure, we should keep in mind that after the destruction caused by the Huang Chao Rebellion, the Tang court was able to rebuild the collection to only somewhat over 20,000 scrolls, which in the subsequent troubles shrank to only 18,000 scrolls. Jiu Tang shu (Beijing: Zhonghua shuju, 1975), 46.1962. 10  The history of the Chongwen zongmu itself is complicated. It was once an annotated bibliography in sixty scrolls, and we dearly wish we still had all the annotations, some of which have, however, been supplied from other sources. Unfortunately these are largely not in the jibu. The essential booklist, however, seems basically intact, which is how we use it here.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1101

forms). The anthologies are also a significant source, but there are many large anthologies where we cannot tell if it is a Song or Tang compilation. Let us then presume in the jibu between 3,000 and 4,000 scrolls of Tang individual collections and anthologies compiled before the 980s; that is, at most 16 percent of the collection. If we add the literary collections stretching back to the middle of the Liang (material that would also be included in Wenyuan yinghua), that figure will not significantly increase. This rough figure of three to four thousand scrolls of Tang literary writing in the imperial library in the 980s is meant to serve as the background of a question I will pose later: what was the Wenyuan yinghua in its one thousand scrolls? One thing should be obvious from the outset: an “anthology” of a thousand scrolls represented a considerable proportion of the literary holdings in the Song imperial library of the 980s. In fact, “anthology” is the wrong word; from the way it was apparently compiled, Wenyuan yinghua was what might be called the “epitome” of the largest collection of Tang and Five Dynasties literary manuscripts of the time. Just as Song Taizong lamented the loss of books when he compared his library to the old catalogs, we lament the loss of the Tang books whose names are preserved in Song catalogs. A substantial portion of those books, evenly represented, may still exist, with the individual literary works that comprised them scattered among the various categories of Wenyuan yinghua. Let me provide some essential background for thinking of manuscript culture as it was received in the early Song; then give a few pieces of anecdotal evidence for the state of textual preservation of Tang literature in the first part of the Northern Song to suggest the nature of many of the texts in circulation; and finally return to Wenyuan yinghua, to consider it not simply as the gold-mine of Tang literature that it is, but as an “epitome” of the jibu in the imperial library.

Manuscripts of Literary Works

The transmission of Tang literature was quite different from many other kinds of textual transmission. The government oversaw the textual integrity of the Confucian Classics, the Histories, and the institutional compendia, as the Tang had done. Texts of the Classics were sometimes imperfect, but such was the nature of learning in the Classics that scholars were disposed to note variants as errors and to compare texts. The Buddhist and Daoist churches oversaw their respective canons with varying degrees of textual care. Serious scholarship was done on the Wen xuan 文選 and Chuci 楚辭, but before the Song there was no scholarship on Tang literature. In the Tang there were initial editions of a

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1102

Owen

writer’s works; anthologies, even serious ones (though these are now lost); and a few commentaries such as Zhang Tingfang’s 張庭芳 mid-eighth-century annotation of Li Jiao’s 李嶠 (646–715) Baiyong 百詠 (though we suppose that this was for purposes of compositional pedagogy rather than to explicate Li Jiao’s poems). Among anthologies, Gu Tao’s 顧陶 mid-ninth-century Tangshi leixuan 唐詩類選 suggests a scholarly scrupulousness in refusing to anthologize a poet before a full edition of his works was in circulation (although this is the only case in the Tang where fair judgment was linked to the possession of a complete collection); this anthology survived into the Song, but only its prefaces survive today. There was, however, no actual Tang literary scholarship on Tang literature, nor was there a general sense of constraints on the exact reproduction of texts. Because many texts in the Wenyuan yinghua were preserved in other venues, we know the textual problems of that compendium in its present form; we do not have grounds for similar worries about texts of the stories in Taiping guangji because there are few extant complete texts against which to measure them. In her important article “Book Culture and Textual Transmission in Song China,” Susan Cherniack takes us to the last part of the eleventh century, when readers of a scholarly bent expressed a strong nostalgia tor the manuscript and claims of its superiority to printed books, both for reading and accuracy.11 This was indeed an age of literary scholarship, to a degree that had no parallel in the Tang. A meticulous scholar of that later era, carefully collating his edition with others, could indeed boast of the superiority of his manuscript to printed editions. This was not, however, the state of the manuscript legacy of Tang literature a century earlier in the 980s, the time of Wenyuan yinghua’s compilation. The texts of some authors seem to have been widely circulated. Ouyang Xiu 歐陽修 (1007–1072) may have been deeply concerned about the textual state of Han Yu’s 韓愈 (768–824) works, and many people complained that his collection was “not circulated”; but Han Yu’s works did indeed circulate widely, albeit in textually various editions. The situation with many other writers is far less clear. When I discuss Wenyuan yinghua, I will be considering the holdings of the imperial library; but it is important to recognize that Tang and Five Dynasties manuscripts and copies of them were spread all over China—in provincial libraries, temple libraries, and private libraries. In a number of cases versions later entered general circulation that were far more complete than what we find in the imperial library, as represented by the Chongwen zongmu or as we infer from Wenyuan yinghua. Some more popular works may have been in many such libraries, but “popular” here is defined by ninth- through 11   h jas 54.1 (1994): 5–125.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1103

early eleventh-century taste, which was far from the canon of Tang literature that we now know. Much literary material in general circulation probably looked like the Dunhuang manuscripts, where the predominant form of literary work was a “personal anthology” of poetry; that is, a selection of poems by different poets that does not seem to have been based on a known anthology. Less “popular” works survived in odd places in the Northern Song; one chanced to find them by accident or heard about them. Unless one happened to go to a place where a manuscript was or heard a friend mention it, that manuscript was unknown. Northern Song literary scholars were, by and large, civil servants. Bound to their assignments, they could not, like the great Renaissance humanists, tour monastery libraries hunting out lost codices. The dispersal of texts influenced not only the preservation of works but also the preparation of editions. A scholar might consult a number of editions in preparing a critical edition, but no one could do a comprehensive comparison of existing manuscripts. Our evidence is that these manuscripts differed considerably both in what they included and in variants. A rule of thumb is that the more texts there were, the more the variants. Manuscripts were everywhere for those who looked. This is evident in the case of Kong Yanzhi 孔延之, a Song scholar with access to regional manuscripts and inscribed texts. When posted to Yue, Kong compiled Guiji duoying zongji 會稽掇英總集, with a preface dated to 1078 and beginning with a characteristic Northern Song lament for lost books. This anthology celebrates the region, but it also includes many Tang texts with no other known source, giving us reason to believe that Kong Yanzhi had access to texts representing the interests of a regional library or local bibliophiles, along with inscriptional sources. We may not trust all the ascriptions, but we see much evidence of local survival of manuscripts that were elsewhere unknown.

Some Accounts of Literary Collections

In an exemplary account of the preparation of a critical edition, Song Minqiu 宋敏求 (1019–1079), one of the most famous Northern Song scholars, tells us a great deal about the manuscript legacy standing behind those Song editions that are the basis for our modern editions. Song Minqiu prepared his edition of Meng Jiao 孟郊 (751–814) from four primary versions and an unspecified number of other sources. Of the primary versions, one was in five scrolls consisting of 124 pieces; one was in ten scrolls with 331 pieces; one was in five scrolls with 340 pieces; and the last was in two scrolls with 180 pieces. The first lesson we may draw is that the number of scrolls bears only a highly variable relation

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1104

Owen

to the actual size of the collection—between twenty-five and ninety poems per scroll. Song Minqiu continues in his colophon: “The rest are not editorially organized and just record pieces randomly. Every one is different. I have now made a general gathering of what was left out and have gotten rid of the duplications and those whose style seems not like Meng’s. I have 511 pieces.”12 This is eloquent testimony to the variation in versions of Meng Jiao’s poetry in the second half of the eleventh century. None of these versions is marked as an “anthology.” Not only do the numbers vary, but so do the contents, giving Song Minqiu a composite number of 511 pieces. The largest single collection available to Song Minqiu was only two-thirds of the aggregate number of poems he recovered by comparing versions. The phrase that should give us particular pause—if I have interpreted it correctly (and it is peculiar)— is “those whose style seems not like his” 若體制不類者.13 He has apparently taken it upon himself to exclude works on stylistic grounds; and if our current collection of Meng Jiao’s poetry sounds remarkably homogenous, we cannot say if this is due to Meng Jiao himself or to Song’s editorial intervention.14 The lesson is loud and clear: we do not have Tang literature; we have Tang literature as it circulated in ninth- through tenth-century manuscripts, as those manuscripts survived in the Song, and as Song editors preserved them. When Wang Zhu 王洙. compiled the 1039 edition of Du Fu’s poetry that became the basis of all subsequent editions, he used nine versions ranging from one to twenty scrolls. Song Minqiu was a remarkably diligent editor, managing to find four fairly large editions. Again the Chinese is not clear, but the quoted section beginning “The rest are” (zi yu 自餘) suggests informal manuscripts, perhaps readers’ personal selections of a poet’s work. I like the term the “little collection” (xiaoji 12  Hua Chenzhi 華忱之 and Yu Xuecai 喻學才, Meng Jiao shiji jiaozhu 孟郊詩集校注 (Beijing: Renmin wenxue chubanshe, 1995), p. 694. “What was left out” presumably refers to pieces attributed to Meng Jiao that were not in the major manuscripts. 13   Tizhi 體制 (體製) was the term used to describe an authorial style or a period style, so that the alternative interpretation of bulei 不類 as “not good” does not fit. Compare Yan Yu, writing of a poem that he thinks was falsely incorporated into Tao Yuanming’s collection: “Its style and tone are not like [Tao] Yuanming’s” 其體制氣象與淵明不類. Wu Wenzhi 吳文治, ed., Song shihua quanbian 宋詩話全編 (Nanjing: Jiangsu guji chubanshe, 1998), p. 8732. 14  Song Minqiu is now best known for an anecdote recounted by Su Shi that immortalizes Song as the exemplary bad editor of poetry. According to Su Shi, Song objected to Du Fu’s line “A white gull sinks into flooding vastness” 白鷗没浩蕩 on the grounds that a gull could not “sink,” and therefore changed mo 沒 to bo 波 (“White gull, waves vastly flooding”). Dongpo quanji, skqs, 100.2a–2b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1105

小集), the term for an “anthology” of a single writer’s work. Song Minqiu’s note

on his editorial work, however, calls into question the usefulness of that term.15 If the largest single manuscript of Meng Jiao’s poetry is only two-thirds of the aggregate collection from the various manuscripts (excluding those poems that did not “sound like” Meng Jiao), then all we had were “anthologies” (xiaoji). From the evidence the xiaoji was the norm of circulation for literary manuscripts. Within xiaoji there is immense variation, from large collections, like Song Minqiu’s version with 340 poems, to very small collections, represented by Song Minqiu’s “the rest.” Some xiaoji were formal anthologies of an author’s works, prepared for the purpose of circulation. Fan Huang’s 樊晃 late eighthcentury anthology of Du Fu’s poetry is perhaps the outstanding example of a large anthology widely circulated, with 290 of Du Fu’s poems. Other such anthologies were made simply by copying personal favorites, and these might in turn be copied by others, making them impossible to distinguish from a formal “anthology.” Song Minqiu’s list of texts used in the collation of Meng Jiao’s poetry, each with the number of scrolls and poems included, is a healthy antidote to the usual form of traditional Chinese bibliography, giving “the works of so-andso” in a certain number of scrolls ( juan, or, as we enter the world of butterfly books and printed texts, “fascicles”). We have cases (as we will see with Li Jiao’s Baiyong) in which the size of the scroll (and probably the size of the manuscript hand) might be varied to produce a satisfying number of scrolls/ juan.16 Every piece of evidence suggests that the reality was far closer to what Song Minqiu describes—until Song Minqiu’s edition of Meng Jiao became the 15  I have somewhat twisted the use of this term. Song bibliographies use this term to describe what they see as purposeful anthologies, and they use it only when they have a more complete edition. The “personal anthology” is more properly a chaoben 鈔本. Eventually chao 鈔 or 抄 came to mean simply “copy by hand,” but it originally meant “to select certain pieces to copy.” Other terms are used for making a complete copy in the Tang. Such anthologies of a writer’s work would be copied (chao) in turn, augmented or diminished according to the inclinations of the copyist. 16  The Xin Tang shu 新唐書 bibliography proudly lists the collected works of Li Jiao in 50 scrolls, 20 scrolls larger than the collection as given in the High Tang catalog copied into the Jiu Tang shu. Then we note that Li Jiao’s Baiyong, actually numbering 120 poems, occupies 12 scrolls. If this number is not a scribal error, these were either very small scrolls or scrolls with very large characters, even if we factor in Zhang Tingfang’s commentary. This suggests that the librarians—much like students adjusting margins and font size to produce the requisite number of pages—sometimes used the format of scrolls to produce collections of a size that gave the appearance that there had been no loss.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1106

Owen

authoritative one, as it did, along with Northern Song editions of other writers. The works of an author circulated in different sizes with different works included (and varying degrees of overlap) and sometimes very different texts. Scrolls had normative physical sizes (though they were not uniform), but the number of works they contained in manuscript depended on the size of the handwriting (which should be evident from Song Minqiu’s descriptions of the texts he used). At the low end, particularly important for many authors, were those short collections of favorites, which in some cases included all that remained of an author’s work and in other cases might be a valuable source for otherwise missing texts (though Song Minqiu’s sources seem to have given him some poems by Meng Jiao’s late ninth-century admirer Nie Yizhong 聶夷中 as Meng Jiao’s own). I use the term xiaoji as a convenience; in some cases—Han Yu, Liu Zongyuan 柳宗元 (773–819), and Bai Juyi 白居易 (772–846)—we have the works much as they were made or constituted after the author’s death; in most cases, however, all we had were xiaoji of varying sizes, often combined or supplemented in the Song. One more point should be made: so far as we can tell, the early Song imperial library judged its completeness on titles (and in the case of Kaiyuan and earlier works, matching the number of scrolls with the bibliography from Xuanzong’s reign). If Song Minqiu made a composite edition, we can assume that the imperial library eventually got a copy, but there is no evidence that the imperial librarians actively sought more complete editions of Tang literary works dating after the 721 catalog. Meng Jiao, though much neglected after the eleventh century, was a popular poet in the late ninth and tenth centuries. Song Minqiu does not speak of any difficulty in finding several versions of Meng’s works. Other cases of compiling a more complete edition of an author’s works often involved some sort of search for manuscripts, with varying degrees of effort and frustration. If the predominance of partial versions (“every one is different,” as Song Minqiu says) is the primary complication in assessing the manuscript legacy of Tang literature, an important secondary complication is the issue of the specialized collection. The dominant model of Chinese bibliography from the Song to the present has been the “[complete] literary works” (wenji 文集). Writings on history, classical scholarship, technical treatises, informal compositions, and, later, vernacular literature were placed outside the wenji; but the wenji was supposed to represent all of the author’s works in classical literary genres. The problem was that an author’s literary works might circulate in sub-collections that were not originally intended to become part of the wenji or poetry collection (shiji 詩集) as it was compiled by the author or first editor; sometimes works in these sub-collections were explicitly meant to be excluded. There was no term for this kind of work in traditional bibliography,

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1107

and they were simply included as another work by the author in the wenji or shiji sections. Such specialized collections became popular in the second half of the eighth century and were widespread in the ninth and tenth centuries. Some of these were banquet collections, “exchange collections” (changhe ji 唱和集), and other kinds of group collections, which were integrated easily into the wenji of individual authors; but other sub-collections present a variety of issues. Song editors like Song Minqiu wanted a wenji or a shiji as the “complete works” to represent an author comprehensively. Out of the flotsam and jetsam of the manuscript tradition they sometimes found a specialized subcollection but no primary collection, and sometimes, a primary collection without the specialized sub-collections. The case of Li Shen 李紳 (772–846) is instructive. We have only small, fragmentary remains of his regular poetry collection; what we have instead is a sub-collection, Zhui xiyou 追昔遊 (Recollecting past travels), in three scrolls, with a preface dated to 838. This collection has a chronological sequence of the poems that Li Shen should have written on his experiences between 820 and 836, but composed retrospectively between 836 and 838. When Song and later editors had diverse sources, they sometimes integrated them into a single collection, using one of the then popular classificatory systems, such as genres or themes. In Li Shen’s case the surviving poems from the original poetry collection were simply added on after Recollecting Past Travels. This accident of textual preservation gives us a highly skewed picture of the poet. Li Shen, the poet who wrote the ballad version of Yingying zhuan 鶯鶯傳 (which survives in only a few passages) and who was a lively participant in the poetic experiment of the Yuanhe 元和 generation (806–820), appears to us primarily as an old man nostalgically reviewing the previous decades of his life between 820 and 836. The case of Wen Tingyun’s 溫庭筠 (?–866) poetry collection, whose date of compilation is unknown, is a good example of what would be lost if we had only the primary collection or only material from one of the secondary subcollections. The Xin Tang shu “Bibliography” lists a number of collections under Wen’s name, including a “poetry collection” in five juan (perhaps from a Tang source; this collection is not in Chongwen zongmu, although two other specialized collections are). Our current version of Wen Tingyun’s poetry is in nine juan. If we consider juan 3 to 7, we have five juan in the form of a remarkably standard mid-ninth-century “poetry collection,” beginning with some conventional yuefu, which are followed by a collection of primarily occasional regulated verse in the predominant style of the day. This is fine work, and includes a few of Wen Tingyun’s best-known poems, but it is nothing out of the ordinary. The first two juan of the nine-juan version, however, consist of brilliant and

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1108

Owen

daring yuefu in the style of Li He 李賀 (790–816), like nothing else in the five following juan; these are probably from one of the specialized collections. The eighth juan is yet again different, with sometimes bizarre, witty poems bearing no resemblance to either the blander “poetry collection” or the Li He-style yuefu. We know with some certainty that the poems in the eighth juan come from a specialized anthology (zongji), Duan Chengshi’s 段成式 Hanshang tijin ji 漢上題襟集, a collection representing a group of friends in Xiangyang between 857 and 859. Here we find a poem on a fight between two singing girls comically done in the most ornate poetic language. The ninth juan seems to consist of poems recovered from other sources, the ubiquitous addendum that reminds us how collections continuously grew with the discovery of new texts. The remarkable diversity of Wen Tingyun’s poetry is apparently dependent on the fortuitous survival of different sources. A generic economy was at work in textual circulation in the ninth and tenth centuries. Had only one of the components of Wen Tingyun’s poetry collection survived—as was probably the norm—we would have had a one-sided view of Wen Tingyun as a poet (just as we now can have only a one-sided view of Li Shen). We see similar economies in collections kept explicitly apart. The Han Wo 韓偓 (842–923?) of the Han Hanlin ji 韓翰林集 was a follower of Du Fu during the last days of the Tang, a dynastic loyalist who described the disintegration of the empire with a dignity and pathos that is still poorly appreciated. The Han Wo of the specialized collection, the Xianglian ji 香奩集, wrote erotic vignettes, such as a description of a woman taking a bath. Each of these two collections excludes works of the kind that appear in the other collection. The two collections survived separately with the Xianglian ji preface apologizing for such poetry As Anna Shields has discussed, the Yan shi 豔詩 of Yuan Zhen 元稹 (779–831) were excluded from Yuan’s Yuanshi Chongqing ji 元氏長慶集, but they clearly circulated independently and were preserved in part in the mid-tenth-century anthology Caidiao ji 才調集.17 Most literary manuscripts were not what we call a complete wenji or a shiji. The independent circulation of sub-collections is important in understanding Tang literature, particularly from the ninth century on, precisely because of the vagaries of the manuscript legacy. Sometimes a poet is represented by a bland “poetry collection”; sometimes a poet is represented by a single series of poems or a set of works on unusual topics in an unusual style. Such differences, often used to characterize poets, may be no more than the accident of what survived in the manuscript tradition. The poet Li Shangyin 李商隱 17  Anna M. Shields, “Defining Experience: The ‘Poems of Seductive Allure’ (yanshi) of the Mid-Tang Poet Yuan Zhen (779–831),” jaos 122.1 (Jan.-March 2002): 61–78.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1109

(ca. 813–ca. 858) was first known through a xiaoji restricted to only one kind of poetry; his image as a poet was completely transformed by the devoted quest to rebuild his poetry collection. Song Minqiu appears again in Zeng Gong’s 曾鞏 (1019–1083) Bao Rong shiji mulu xu 鮑溶詩集目錄序 (Preface to the index to Bao Rong’s Poetry Collection), where a bibliographical error opens up a can of worms with editions of literary works in the imperial library:18 Bao Rong’s Poetry Collection in six juan. Among the Historiography Office books the former title was “Bao Fang’s Collection” in five juan. The Chongwen zongmu is the same in its account in individual collections. The Drafter Song Minqiu told me that poems in this collection that also appear in the [Tang] Wencui 唐文粹 (Essence of letters) and Tang shi leixuan 唐詩類選 (Tang poems selected by categories) are all given as Bao Rong’s works; moreover, Fang’s “Za gan” 雜感 (Miscellaneous stirrings), though his most prominent poem, is not in this collection, by which we know the collection is not Fang’s work. I investigated this with the Essence of Letters and Tang Poems Selected by Categories, along with “Miscellaneous Stirrings,” and what Minqiu said is correct. I also got a “Bao Rong Collection” from the library of Ouyang Xiu, Participant in Determining Government Matters, which is the same as this collection. Only then did I understand for sure that this was indeed [Bao] Rong’s collection. The book in the Historiography Office is in five juan and has about two hundred poems in all. Ouyang Xiu’s book has no juan division and just over a hundred poems, thirty-three of which are not in the Historiography Office version. I have now appended these as a separate juan, giving them the collective title Bao Rong’s Poetry Collection in six juan. This preface is a mine of issues in manuscript culture, which come to the surface only because of the problem of the correct name of the author: is this the collection of Bao Fang or Bao Rong, both of whom were known poets? Had Zeng Gong’s curiosity not been stirred by Song Minqiu’s observation regarding the error in the name of the author, he would not have checked the Historiography Office version of the text against Ouyang Xiu’s version. Comparison, however, revealed that Ouyang Xiu’s version had only about sixty-seven poems in common with the Historiography Office version; about one-third of Ouyang Xiu’s edition was not in the Historiography Office version 18   Zeng Gong ji 曾鞏集 (Beijing: Zhonghua shuju, 1984), p. 192.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1110

Owen

(which is the imperial library version). The purpose of Zeng Gong’s “Preface” was to announce a new, more complete edition, with a final scroll of addenda from Ouyang Xiu’s version, making now a six-scroll edition of Bao Rong’s poetry, with the name corrected. In simple terms, the imperial library had a different xiaoji from that in the collection of Ouyang Xiu’s. In all probability Bao Rong wrote many more poems than were found in either collection. Zeng Gong took the fortuitous discovery as an opportunity to supplement the missing poems. The imperial library was not looking for Bao Rong’s (or Bao Fang’s) poetry collection because the librarians thought this was what they already had. If, however, by some magical power one had been able to recover and compare all the versions of Bao Rong’s poetry circulating in China at that time, I strongly suspect that the collection would have grown many times over. In a pair of letters to one Han Ji, Liu Kai 柳開 (947–1000) tells of having once found a copy of the poems of mid-Tang poet Lu Tong 盧仝 for sale in a bookstore and purchased it for a hundred cash.19 His brother gave the manuscript to a friend, Zheng Yan. After Zheng’s death, the manuscript disappeared, and Liu Kai was unable to retrieve it; Liu then describes searching fruitlessly for a copy of Lu Tong’s poems everywhere in his travels. Liu Kai has heard from a friend that Han Ji has a copy of Lu Tong’s poems and wonders if this could be a copy of the same edition he lost. He asks to borrow the manuscript. The second letter deals largely with other matters, but closes by noting that he was returning the manuscript with the comment that he copied it out, and found that it was a different version of Lu Tong’s poems. Until the age when literature became widely available in print, different manuscripts often meant different selections of texts, and certainly different variants.20 The great saga of recovering a collection from scattered sources is Yang Yi’s 楊億 (974–1020) account of rebuilding the collection of Li Shangyin’s poetry.21 In the Zhidao reign (995–997) of Song Taizong, Yang tells us that he initially got a copy of Li Shangyin’s poems with somewhat over a hundred pieces (that is, a xiaoji). He was quite taken with these but, speaking from his later perspective, they “did not get the deeply engaging quality [of his poetry]” 未得其深趣. Later, in the Xianping reign (998–1003) of Zhenzong, he speaks of a general search for surviving Tang poems; he does not speak of specifically finding 19   Quan Song wen 全宋文 (Chengdu: Ba Shu shushe, 1988–), pp. 634–36. 20  The rareness of the collection is partially confirmed by the absence of any Lu Tong poems in Wenyuan yinghua and the inclusion of only one set in Tang wencui. 21   Shishi leiyuan, skqs, 34.12a–b. Cited in Wan Man 萬曼, Tangji xulu 唐集敘錄 (Beijing: Zhonghua shuju, 1980), pp. 283–84.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1111

another edition or anthology, but in the process he built his Li Shangyin collection to 282 poems. He had heard that at the end of the Tang there were many editions of Li Shangyin’s poetry in the lower Yangzi region, and a friend found one of these, bringing the Li Shangyin poetry collection to over four hundred poems (somewhat over two-thirds of the size of the current collection). Here is the case of an editor who did undertake something like a national search for manuscripts and whose perseverance and political power made his attempt to recover a collection relatively effective. The result was a collection over four times the size of the one with which he began (eventually becoming, in its current form, a collection six times the size of the initial xiaoji). The case of Li Shangyin’s poetry, where we know something of the history of the collection’s reconstitution and the changing responses, well demonstrates how the move from a readily available xiaoji to a fuller collection could completely change a poet’s reputation. Toward the end of the Tang or perhaps early in the Five Dynasties period, Li Fu 李涪 denounced Li Shangyin’s poetry in the following terms: “not a single word he wrote contributes to the state, nor is there even a slender thought encouraging virtue; [his poetry] is a mere display of literary craft.”22 Such a judgment, so incomprehensible to the modern reader of Li Shangyin, was almost certainly based on a xiaoji that selected precisely the erotic pieces and rhetorical tours de force that sustained Li Fu’s judgment. This was the kind of Li Shangyin’s poetry that was popular in the tenth century and in general circulation. The same bias is evident in the selection of Li’s poems in the mid-tenth-century anthology Caidiao ji 才調集 and in the poems in Wenyuan yinghua, representing the collection in the imperial library (though the selections in Caidiao ji and Wenyuan yinghua have remarkably little overlap). This was probably the kind of poetry represented in Yang Yi’s initial xiaoji. We can imagine Yang Yi’s surprise when he saw still more poems by Li Shangyin. Continuing the story of the collection, Yang Yi cites and concurs with a friend’s praise of the poem “Jia Yi.” This suggests that “Jia Yi,” a poem that offers unambiguously ethical judgment on history, was added by Yang Yi in the final stage of compiling the collection (it was not one of the Li Shangyin poems imitated in Yang Yi’s earlier Xikun chouchang ji 西崑酬唱集) and that its qualities differed from those he valued in the initial xiaoji. In short, without Yang Yi’s continuous effort over many years, Li Shangyin would have held a very different place in the history of Tang poetry. There are many other stories of loss and recovery, such as Zhang Jì’s 張洎 (933–996) account of his efforts between 946 and 965 to rebuild the poetry 22  Li Fu, Kan wu 刊誤, in Song Zuogui 宋左圭, Baichuan xuehai 百川學海 (photorpt., Kyoto: Chūbun, 1979), p. 519.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1112

Owen

collection of the mid-Tang poet Zhang Jí 張籍: “Hardly one in ten of his works survived.” The theme of loss was a common bibliographical trope, but with a strong basis in historical fact. What we see are the accidents of loss and recovery, the general difference in works included and textual variants among different manuscripts, the centrality of a devoted editor in the process (allowing that most Tang writers did not find such editors), and the way in which these conjunctions of chance shaped individual reputations and the Tang literary heritage. We should not overlook the correlative problem of the ever-growing collection. Although some scholars, like Song Minqiu, were interested in less wellknown Tang writers, Song dynasty scholars devoted most of their attention to a few famous poets, like Li Bai, Du Fu, Du Mu, and Li Shangyin. The consequence of such attention was that the “collected poems” of these poets kept growing. Because the original Du Mu collection was well defined (based on the poet’s own radical selection of his work), the growth of the poetry collection was in addenda, which, we now recognize, contain numerous false attributions. We may suspect the same is true of the collections of the other famous poets. Poems often circulated under various names; and if one judged the authenticity of a work by Song Minqiu’s criterion of “being like the style” of a poet, the way was open for the inclusion of pieces by epigones, as Nie Yizhong’s poems entered the collection of Meng Jiao.

What is Wenyuan yinghua?

Wenyuan yinghua is not only the single most important source in the preservation of Tang literature; it also provides a window on the manuscript legacy of the Tang as it existed in the late tenth century.23 With the Taiping guangji completed in 978 and the great encyclopedia Taiping yulan nearing completion, in 982 Song Taizong commissioned Li Fang 李昉 and many of his previous editorial associates to “examine the literary collections of preceding eras and select what is essential in them” 閱前代文集撮其精要.24 Li Fang and many of the original editorial board moved on to other tasks and were succeeded by others. In January 987 the work, in a thousand juan, was presented to the throne. The emperor approved. He had more pressing tasks than to check the 23  For a thorough study of the history of Wenyuan yinghua, see Ling Chaodong 凌朝棟, Wenyuan yinghua yanjiu 文苑英華研究 (Shanghai: Shanghai guji chubanshe, 2005). 24  Unlike Taiping guangji and Taiping yulan, Wenyuan yinghua, at least in its current version, does not come with a list of the books consulted.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1113

compilation against the books in the imperial library. It soon became apparent, however, that the work had textual problems, and it was recollated in the first decade of the eleventh century. This reedited version was apparently lost in the great palace fire of 1015 that burned the Treasury of the Left, the Imperial Repository (Bige 祕閣), and the repository of books in the Academy for Venerating Literature (Chongwen yuan 崇文院). Zhenzong was reported to have said (perhaps worrying about the Treasury), “What a pity that all that had been amassed in the two preceding reigns is almost entirely gone in a single morning!”25 After that date the imperial library, the collection from which Wenyuan yinghua had been compiled, was not quite the same. Wenyuan yinghua was not printed until 1201–1204, over two centuries after it was compiled in manuscript form. For this edition the retired Zhou Bida 周必大 (1126–1204), acting on imperial commission, did another collation. Corrections for this edition were also made by Peng Shuxia 彭叔夏 who in 1204 published a work on textual problems, the Wenyuan yinghua bianzheng 文苑英華辨證, which survives and is eloquent testimony to the standard problems of manuscripts and to the cases in which Wenyuan yinghua had the textually better version. We do not know where the Southern Song got its manuscript copy of the work; it would probably not have been the version in the Northern Song imperial library and the quality of the transcription is uncertain. The Song printed edition eventually came to be disseminated again in manuscript copies, which introduced new errors. Probably from one of these manuscript copies we have the Ming edition of 1566–1567 that was printed in Fujian, which survives in toto (130 juan of the Song printed edition also survive). There are fragments of a 1516 Korean printed edition, and a group of Ming manuscript editions in varying degrees of completeness. Adequate collation remains to be done.26 25   Xu Zizhi tongjian, p. 716. 26  Since so many collections were reconstituted from Wenyuan yinghua, it is tempting to use it as the source against which to measure the texts of such collections. Many of these collections were first made before our current version of Wenyuan yinghua and may possibly represent a better stage in the transmission of the manuscript. In his Wenyuan yinghua bianzheng (Analysis and presentation of the evidence in Wenyuan yinghua), prepared in conjunction with Zhou Bida’s Southern Song edition, Peng Shuxia gives long lists of cases where the Wenyuan yinghua version is superior to the texts then in circulation. A remarkable amount of laborious textual scholarship has gone into editions of works that have clearly been reconstituted almost entirely from Wenyuan yinghua, while there has. to my knowledge, been no attempt to collate the Ming printed edition with the Ming manuscript copies, or with Gao Sisun’s 高似孫 twelfth-century Wenyuan yinghua zuanyao 文苑英華纂要, a collection of excerpts, done before the work was edited at the beginning

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1114

Owen

Although Zhou Bida claimed that Wenyuan yinghua’s purpose was to preserve texts at a time when it was hard to find the works of many Tang authors, the initial imperial commission, to “select what is essential,” was quite clear. Zhou Bida was trying to explain the peculiarity of the product, rather than reiterating its initial mandate. The Wenyuan yinghua was implicitly intended as a continuation of the Wen xuan.27 The work, however, was no “anthology” in the usual sense; from the point of view of any canonical selection of Tang literature, both earlier and later, its omissions are striking and many of its inclusions perplexing. The editorial board was made up of an impressive array of court officials, who conceivably divided the tasks among themselves for personally taking charge of their respective sections. The actual work, however, looks like the product of a group of scribes set loose in the imperial library with a skeleton of rubrics. One constraint seems to have been not to copy too many examples of one person’s work under one category.28 A second apparent constraint—and a significant one for reading back from the “anthology” to the sources—was the need to omit works for which a rubric was lacking.29 Over the centuries scholars have mined Wenyuan yinghua to rebuild lost Tang collections, alternately praising its riches and lamenting the state of its text (which in many cases may be the very text from which our current versions of collections originated). Wenyuan yinghua and, to a lesser degree, the much smaller one-hundred-scroll Tang wencui 唐文 粹, compiled by Yao Xuan 姚鉉 (968–1020) in 1012 before the library fire (but presented to the throne in of the thirteenth century. Gao Sisun’s work, however, was collated with existing literary collections, and in my sample collation of a section with the corresponding juan in the Song printing of Zhou Bida’s edition, every variant appears with Peng Shuxia’s note that this is the reading in the author’s independent collection. 27  In the original version there was surely a preface and a presentation memorial, which would have made this explicit. 28  As large as the sections on parting poems or verse letters are, Wenyuan yinghua includes a much smaller proportion of such pieces from independently surviving collections than less common topics. 29  We must explain Wenyuan yinghua’s extensive representation of Chen Zi’ang’s 陳子昂 (659–700) poems, but also the complete exclusion of his famous series Ganyu 感遇, along with the complete exclusion of Li Bai’s Gufeng 古風. It would be tempting to conclude that the editors wanted to avoid poems that were supposed to be “critical” of the ruler, but we find no such avoidance in selecting from Bai Juyi’s Xin Yuefu 新樂府 (New yuefu). The exclusion is better explained by a predetermined set of categories and subcategories in which there was no obvious place for such poems in the “ancient style.” Ling Chaodong also suggests that the categories were the reason for excluding Chen Zi’ang’s Ganyu, in Wenyuan yinghua yanjiu, pp. 42–43. Such omissions were remedied in [Tang] Wencui a few decades later.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1115

1020, by his son), are the sources for much of our knowledge of Tang literature.30 Many authors we now consider important are preserved primarily in Wenyuan yinghua and Tang wencui texts, with only a few additions from other sources. In many cases the number of surviving works is relatively small (say, under thirty pieces), which suggests that in most of those cases the editors did not have even a xiaoji. The editors did, however, have access to a range of Tang anthologies far beyond what we currently possess, and such works probably come from anthology sources. Although Wenyuan yinghua is by far the more important compendium, Tang wencui is useful because Yao Xuan sought important texts overlooked by Wenyuan yinghua. Many other collections survived independently, but these were by very popular authors or fortuitous survivals. We are so accustomed to the large modern compendia of Tang poetry and prose, as well as to fine editions and commentaries, that we often forget the sources from which they were built. It often turns out that a distinguished history of editions going back to the Southern Song ultimately derives in large part from gleanings from Wenyuan yinghua, supplemented by Tang wencui.31

30  The relationship between the Tang wencui and the imperial library is a vexed question. The preface merely says that he “thoroughly perused the various collections” 遍閱群 集, but not where those collections were. The Junzhai dushu zhi 郡齋讀書志, skqs, 4 xia.35b, claims that Yao Xuan had a large library with many rare editions; this may indeed have been the case, but the Junzhai dushu zhi brings this up in connection with the first fifty juan, which he later expanded to the current one hundred. Yao Xuan may have had a very good library indeed, but we should seriously doubt that he had the range of collections from which the Tang wencui provides the unique source for particular pieces. There are so many collections that have been reconstituted almost entirely from Wenyuan yinghua supplemented by Tang wencui that it is hard to believe that Yao Xuan did not compile at least his expansion of the original fifty juan from the imperial library before the fire. As is well known, Tang wencui chose “old style” prose and poetry; in some ways it was positioning itself against Wenyuan yinghua, but the selections often overlap. 31  We might consider the case of the famous eighth-century prose writer Xiao Yingshi 蕭穎士 (716–768). He has 26 surviving prose pieces, 23 of which are preserved in Wenyuan yinghua, with the remaining pieces all from Tang wencui. Xiao Yingshi’s contemporary Li Hua 李華 (715–774) commented in the preface that a ten-scroll collection was in circulation, but that often the titles were preserved while all or part of the text itself was missing. The Jiu Tang shu bibliography and all the Song bibliographies give the collection in ten scrolls; the Chongwen zongmu and the major Southern Song bibliographies give us confidence that a ten-scroll edition existed. In Tang ji xulu Wan Man seems to profess ignorance about who made the current selection. This “selection” was evidently made by the Wenyuan yinghua editors and by Yao Xuan, perhaps from the ten-scroll version, and the modern one-juan version was compiled from those sources.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1116

Owen

My interest here is somewhat different; I want to look at the Wenyuan ying­ hua as a snapshot of the textual remains in the imperial library in the 980s, a snapshot in many ways superior to the Chongwen zongmu, which is the only credible library catalog we have from the Northern Song.32 About half a century, including a great deal of scholarly work, new manuscript discoveries, and a major library fire, separates Wenyuan yinghua from Chongwen zongmu. Li Shangyin’s poetry collection, for example, appears in the latter, but had not been fully pieced together when Wenyuan yinghua was compiled. By looking where particular works were preserved and considering the processes by which collections were formed, we can better understand how the record can be skewed. The standard bibliographical practice of counting juan in old bibliographies is only of tangential use. I have not yet compared all the Tang literary collections with Wenyuan yinghua, looking at their sources and bibliographical histories, but I have detected some patterns in my study of the majority of such collections. My first observation is that we are far more indebted to Wenyuan yinghua for Tang prose than for Tang poetry—although we are heavily indebted to Wenyuan yinghua for Tang poetry. This is to be expected; apart from the prose of Han Yu and Liu Zongyuan, poetry collections circulated much more widely than individual prose collections. In a remarkable number of cases, prose collections have obviously been almost entirely reconstituted from Wenyuan yinghua.33 In both poetry and prose we commonly find two extremes: either the collection comes predominantly from Wenyuan yinghua (often supplemented by Tang wencui), or only a few pieces come from Wenyuan yinghua. In the first case we can sometimes see from the size of the collection as given in Chongwen zongmu that the recompilation probably occurred after the library fire.34 In 32  After the fire of 1015, a separate version of the imperial library was established outside the palace gates, the Outer Academy for Revering Literature 崇文外院. This was probably for copies. Although making second and even third copies of works in the imperial library was common practice, the known losses in the 1015 fire suggest that at least prior to this in the Song copies were not always made. See Xu Zizhi tongjian, p. 717. 33  This is especially true of fu. It is no exaggeration to say that without Wenyuan yinghua and the “old style” fu from Tang wencui, our corpus of Tang fu would be remarkably small. Most Tang fu are regulated fu, and it seems likely that we possess these in the current numbers because they served as models for preparing for the examination from fu anthologies and collections preserved in regional libraries. 34  If the pieces of a collection are entirely or almost entirely included in known anthology sources (of which Wenyuan yinghua is the main source in almost every such case), then that collection has been reconstituted from anthologies. This means that the anthology sources are the oldest textual sources. When, in such cases, one sees a note on a variant

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1117

the second case it is reasonable to assume that the pieces included in Wenyuan yinghua came from anthologies and that the imperial library did not have the individual collection, which was found later.35 The sequence of texts in poetry collections by lesser-known writers often remains the same from edition to edition, and that sequence often seems to reflect the history of the collection’s formation. We find a number of cases in which the larger part of the collection derives from Wenyuan yinghua, followed by a long sequence of poems or prose pieces that survive only in the collection as it later circulated. This probably indicates the situation that we saw in Zeng Gong’s preface to the Bao Rong collection; the core of the collection was reconstituted by using Wenyuan yinghua, Tang wencui, and other surviving anthologies; then a xiaoji was found and missing texts were copied in at the end. Another common sequence is finding Wenyuan yinghua selections scattered throughout the collection, with a sequence of poems entirely from Wenyuan yinghua at the end. This suggests that the collection survived independently and that eventually an editor compared the contents with pieces included in Wenyuan yinghua and at the end added the Wenyuan yinghua pieces that were not in the collection. Both cases provide evidence of the predominance of xiaoji. In some cases (a few of which are well documented, like the poetry collections of Meng Jiao and Li Shangyin discussed above) a larger version of the literary collection was found or compiled after the completion of Wenyuan ying­hua. In these instances the imperial library of the 980s probably had a xiaoji or its editors included works from anthology sources, unaware that much more complete versions existed elsewhere. Let us examine a particular case. As mentioned above, in 984 Song Taizong issued an edict to recover as many works as possible from the 721 Tang imperial catalog. Dudbridge speculates that the emperor evidently felt satisfied with his late eighth- and ninth-century holdings—not realizing that the version of a literary collection he had in his library might well have been a xiaoji, which could quadruple or sextuple in size if compared with other versions. One collection in the printed edition of Wenyuan yinghua, that “the collection reads” ( ji zuo 集作) thus and so, this is most likely not the original collection but the product of Song editing. Song editing is sometimes on the mark, but often the original Wenyuan yinghua version should be preferred. 35  If Wenyuan yinghua includes only 19 of Chu Guangxi’s 儲光羲 (ca. 706–ca. 763) 226 extant poems, it is highly unlikely that this is due to editorial taste. Rather, it seems clear that a more complete version of the Chu Guangxi collection was recovered later. We see it in the 5-juan edition in Chao Gongwu’s 晁公武 Junzhai dushu zhi and Chen Zhensun’s 陳振孫 Zhizhai shulu jieti 直齋書錄解題. The original Tang version of the collection was in 70 juan.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1118

Owen

probably recovered (or fortuitously surviving in the existing imperial collection) was that of Li Jiao 李嶠 (645–714), a major poet and literary courtier of Empress Wu’s reign. The Jiu Tang shu “Jingji zhi” (Bibliography), using the 721 catalog, lists this collection as having thirty scrolls. Li Jiao has 209 poems in the Quan Tang shi 全唐詩.36 Of those 209 poems, 120 represent his famous yongwu 詠物 collection, the Baiyong 百詠, which circulated independently (there is a complete manuscript in Japan with the commentary of Zhang Tingfang and a partial manuscript among the Dunhuang findings).37 Only 43 of the Baiyong are included in Wenyuan yinghua. Sixtythree of the remaining 89 poems were preserved in Wenyuan yinghua.38 Many of the other extant Li Jiao poems come from the known Tang anthologies. In this case we have good reason to believe that the imperial library of the 980s possessed the complete collection of Li Jiao’s poems. The reason for this assumption is the inclusion in Wenyuan yinghua of group court compositions, done to imperial command, in which Li Jiao’s poem is consistently given first. This is how the poems would have appeared in Li Jiao’s collected works. The Li Jiao collection disappeared in the Chongwen zongmu, perhaps lost in the 1015 fire. A notice of the collection reappeared in the Xin Tang shu “Bibliography,” listed as having an incredible fifty scrolls, with the Baiyong (given with its alternative title Zayong 雜詠) listed in twelve scrolls (this must have been in enormous characters, considering that such a manuscript would have only ten poems per scroll; even with Zhang Tingfang’s commentary, twenty scrolls is extravagant). Clearly, the collection as we have it was reconstituted from Wenyuan yinghua and other anthology sources; not only do we owe most of Li 36  I use this rather than numbers augmented by Japanese and Dunhuang finds, because it represents the process of compiling a literary collection from those materials that survived in the Song. 37  For a facsimile of the Japanese manuscript, see Hu Zhi’ang 胡志昂, ed., Ricang guchao Li Jiao yongwushi zhu 日藏古抄李嶠詠物詩注 (Shanghai: Shanghai guji chubanshe, 1998). 38  Counting is a less straightforward task than it might appear. I have done my primary count from Hiraoka Takeo 平崗武夫, Tōdai no shihen 唐代の詩篇 (Kyoto: Jinbun kagaku kenkyūjo, 1964–1965), and Hiraoka Takeo, Tōdai no sanbun sakuhin 唐代の散文 索引 (Kyoto: Jinbun kagaku kenkyūjo, 1960). I have checked that against the Wenyuan yinghua index. In one case, that of Bai Juyi’s prose, the discrepancy is very large (Tōdai no sanbun sakuhin clearly did not give the majority of Bai’s prose works). In many cases there are small discrepancies (the Kyoto series, for example, counts individual works within sets, whereas the Wenyuan yinghua index counts sets as one work). These differences are, however, not statistically significant. To keep proportions of Wenyuan yinghua selections to total number of works consistent, I have followed the numbers in the Kyoto series, with the exception of Bai Juyi’s prose.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1119

Jiao’s surviving poems (apart from the Baiyong) to Wenyuan yinghua, but also almost all of his 157 prose works are drawn from Wenyuan yinghua. Let us return to the 43 Baiyong poems included in Wenyuan yinghua, out of an original 120. Here we have a known original number of source poems and a known proportion of those poems—about a third—selected for Wenyuan yinghua. We also know that the Baiyong circulated independently in two scrolls, which, with 60 eight-line poems per scroll, is within the broad parameters of a normal sized scroll. If we multiply the 63 other Li Jiao poems in Wenyuan yinghua by 3, we get about 190 poems, or roughly three more scrolls. If we then multiply the 157 prose works by 3, we get 470 works. Since fewer prose works occupy a scroll, this could easily fill twenty-five scrolls. In short, if we consider the Li Jiao selections in Wenyuan yinghua to be about a third of the collection, the full collection would in turn be about the size of the thirty scrolls given in the Jiu Tang shu “Bibliography.” Let me offer the hypothesis that the editors of Wenyuan yinghua quite literally and somewhat mechanically took the emperor’s mandate to “examine the literary collections of preceding eras and select what is essential ( jingyao).” They were not making a true anthology but a survey and epitome of the jibu in the imperial library. They divided each collection and anthology into “three grades” 三品 (top, middle, and bottom) and selected roughly what they considered the top third from each collection (working within the constraints described above, for example, avoiding too many instances of one type of writing for a particular author). This procedure has interesting consequences that remain significant for scholars of the Tang. Not only were the editors of Wenyuan yinghua not making an anthology of the best Tang literature; they were not even making a selection from individual collections proportional to their sense of the importance of the author (as in the anthology model established by Yuan Haowen’s 元好問 Zhongzhou ji 中州集 of 1250). It was not that they believed the best of Li Jiao was better than the middle of Li Bai; they felt that their charge from the emperor was to give the best of each collection. Large selections from insignificant writers appear because the editors happened to have that collection and were doing their job as they understood it. In the end they produced a thousand scrolls, about a quarter to a third of the jibu in the imperial library. One of the common misconceptions about Wenyuan yinghua is that it favors parallel prose and regulated poetry. This misconception is in part because Wenyuan yinghua includes so much regulated fu, parallel prose, court poetry, and examination poetry; it is also, in part, because Tang wencui so aggressively defined itself as a fugu anthology, selecting “old style” fu and poetry. If there are many examination poems and examination fu, as well as model political

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1120

Owen

documents, in Wenyuan yinghua, we need only think of the interests of regional libraries in the Tang, preparing students to take the examination and participate in political life. The texts in those libraries were, no doubt, the foundation of the collections in Bianjing, Jingnan, Shu, and the Southern Tang. In fact, the Wenyuan yinghua seems to have had no biases whatsoever (apart from the rubrics it includes and those it omits). Many of the Tang prose writers in the guwen lineage, both before and after Han Yu, survive primarily through selections in Wenyuan yinghua, rather than in Tang wencui. As the emperor had ordered, the Wenyuan yinghua editors represented what the library had, perhaps choosing individual works by personal preference; but they had no apparent biases either for or against the “ancient style.” We would wish that the editors had included a list of works used, as was done in Taiping yulan and Taiping guangji. Such a list does not appear in the current version, but considering the absence and presumable loss of a preface or presentation memorial in the current version, it is quite possible that such a list might have existed at some earlier stage of the text’s circulation. If the hypothesis is valid, however, Wenyuan yinghua represents the Tang and Five Dynasties manuscript legacy, not as it was scattered all over China, but in the collections of Bianjing and the regional kingdoms, augmented by targeted searches for missing titles from the Kaiyuan reign and earlier. In connection with my hypothesis let me bring up one of Wenyuan yinghua’s most notorious “sins”: the inclusion of the same poem twice under different rubrics. This is taken as a mark of the carelessness of the editors—and on some level this judgment may be correct. Such duplications were noted in the printed version of the early thirteenth century but the duplicated texts were expunged. If we presume an editorial board, changing over time, working with copyists, the reasons for this “sin” become quite clear. Charged with giving the best of every work in the jibu, the editors also took selections from the anthologies—and by far the largest number of anthologies were of poetry. Categorization of texts under a given rubric is sensitive to titles, and any scholar of Tang poetry knows that, however unstable the texts of poems may be, titles are the parts of poems most subject to change. Li Bai’s famous “Jiang jin jiu” 將進酒 (Bring in the ale), appears twice in Wenyuan yinghua (195.9a and 336.2b). The second time it appears, however, it appears in the “song” (ge­ xing 歌行) category under the tide “Xi kong zun jiu” 惜空尊酒. This is also the tide under which the poem appears in the Dunhuang manuscripts.39 In this case the same text was surely copied from two different sources; in one source 39  Xu Jun 徐俊, Dunhuang shiji canjuan jikao 敦煌詩集殘卷輯考 (Beijing: Zhonghua shuju, 2000), p. 74.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1121

the title “Jiang jin jiu” led to it being copied in the yuefu section; in the other source the title “Xi kong zun jiu” led to it being copied in the gexing section. If the same scribe had been working with the same source text, this probably would not have happened. This suggests rather two scribes or editors working with different source texts. Possibly these were different versions of Li Bai’s collection; but the most likely scenario is that one version was copied from the collection, and the other from an anthology. I must strongly emphasize that the history of each collection is different. When a collection consists almost entirely of pieces from Wenyuan yinghua and Tang wencui, we have a clear indication that our current version of the collection was compiled from those sources; when we have a now sizable collection represented by only a handful of pieces in these two large early Song “anthologies,” we also have a clear case of what was not in the early Song imperial library. The “one-third” hypothesis is a way to understand the history of the formation of a collection; pure “one-third” representations of collections are the minority, but they represent a minority so large as to be statistically significant. As we will see in the cases of Li Bai and Du Fu, what at first glance seem exceptions to the “one-third” hypothesis often turn out to be consistent with what we know of the collection in the early Song. New manuscripts remained to be found and many of the old manuscripts in the library were to go up in flames. In a few anomalous cases we have some confidence in the imperial library’s possession of a widely disseminated collection, but the representation does not fit the “one-third” hypothesis. Of over 2,800 poems by Bai Juyi, only 287 were selected for the Wenyuan yinghua. It may be that the editors either did not want to overrepresent him or had a xiaoji; they included 425 of his 750 prose pieces.40 Other collections with relatively wide circulation, however, often give us roughly a third of the author’s corpus. The prose collections of Han Yu and Liu Zongyuan were in circulation. Han Yu’s prose is represented in Wenyuan yinghua by 103 pieces out of an extant 336; Liu Zongyuan’s prose by 184 pieces out of an extant 443. We can infer that a relatively full collection by Zhang Yue 張說 (667–731) was available to the Wenyuan yinghua compilers by the same criterion we find in Li Jiao: the presence of court compositions to imperial command in 40  Zhou Bida in his colophon claimed that Wenyuan yinghua included all of Bai Juyi. The high representation of certain prose genres in Bai Juyi may be due to the fact they were pedagogic models, likely circulating as independent works in provincial libraries. This seems the case with the large number of model examination answers and “judgments” (panwen 判文).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1122

Owen

which Zhang Yue’s name comes first. Wenyuan yinghua includes 128 of Zhang Yue’s 358 extant poems, with the vast majority of the poems not included in Wenyuan yinghua having no known source other than Zhang Yue’s collection as it has been preserved. This both confirms the rule of a third and suggests that Zhang Yue’s current collection of poetry is roughly the same as that available to the Wenyuan yinghua compilers. The case of Zhang Yue’s prose is rather different. Here Wenyuan yinghua has 142 prose pieces out of a surviving 212, with many of the other surviving works drawn from Tang wencui. This strongly suggests that the prose works of the original collection were lost and were reconstituted from Wenyuan yinghua, Tang wencui, and the numerous anthologies of public prose listed in the Chongwen zongmu. If we multiply the 142 prose pieces in Wenyuan yinghua by 3 and add the surviving poems, we have numbers roughly consistent with the thirty-scroll collection given in the Jiu Tang shu “Bibliography.” The Zhang Yue and Li Jiao cases suggest that the poetry and prose editors alike had wenji that included both poetry and prose; but in a number of cases we find strange disproportions. Liu Zongyuan’s prose is represented by about a third of the extant works as expected, but Wenyuan yinghua includes only one poem (out of 161 surviving). It seems likely that the editors did not have Liu Zongyuan’s poetry. The inclusion of only 45 of Han Yu’s 419 poems suggests that the library had a xiaoji. These and other cases indicate that prose collections and poetry collections often circulated separately. The rule of a third recurs so often in Wenyuan yinghua that it cannot be easily dismissed. Wenyuan yinghua includes 66 of Wang Changling’s 王昌齡 (?–ca. 756) 182 poems in Quan Tang shi; 37 of Li Qi’s 李頎 (?–757?) 123 poems; 91 of Meng Haoran’s 孟浩然 (689–740) 266 (or 218 by the Tang count). It includes 32 of Shen Yazhi’s 沈亞之 (jinshi 815) 90 prose pieces. The pattern recurs regardless of the importance of the author. When the rule breaks down, the reason is often evident. Wang Wei’s 157 poems in Wenyuan yinghua represent a high proportion of the 382 extant poems, but the number remains roughly within the rule of one in three. In this case the somewhat higher representation is easily explained: Wang Wei was well represented in the existing Tang poetry anthologies; we would expect him to have been included as well in the much larger Tang anthologies, now lost but still available in the Northern Song. Insofar as those anthologies were also represented in Wenyuan yinghua by a third, this would potentially boost Wang Wei’s numbers.41 Wenyuan yinghua includes only 242 of Li Bai’s 41   Wenyuan yinghua includes only four pieces of Wang Wei’s 65 prose works, probably indicating that the library had only the poetry collection.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1123

1,003 poems listed in the Quan Tang shi; however, we know that in the Xianping reign (998–1003) Yue Shi 樂史 combined two versions of Li Bai’s collection and came up with 776 pieces; if we add in the 13 prose pieces in Wenyuan yinghua, we have a number remarkably consistent with a third of the two versions of the collection. The case of Du Fu is quite similar to that of Li Bai. Roughly 1,400 poems are now attributed to Du Fu, but the Wenyuan yinghua includes only about 240. The history of Du Fu’s collections is complex, and the “complete works” continued to grow in the Song. When Su Shunqin 蘇舜欽 (1008–1048) combined two versions of the collection, he came up with about 800 poems, to which he added texts from another version to produce a bit over 880 poems. Although the versions he used were not those in the imperial library in the 980s, the number is such that we can easily infer that 240 was about a third of what was probably available from a good single version.42 Meng Jiao is represented in Wenyuan yinghua by only 50 poems, but this number is consistent with a third of the size of two of the four editions that Song Minqiu used in the last part of the eleventh century. Wenyuan yinghua includes only 52 poems by Li Shangyin, which would lead us to suppose that the imperial library had a xiaoji with 150 to 160 poems. We know that Yang Yi eventually built the corpus of Li Shangyin’s poetry to over 400 poems about two decades after the compilation of Wenyuan yinghua; however, Yang Yi began with a xiaoji of over 100 poems, and by the turn of the eleventh century he had increased the collection to 282 poems, an addition of about 170 poems. This number is remarkably close to the size of our hypothetical xiaoji in the imperial library.43 Scholars of the Tang who have looked at old bibliographies like the Chongwen zongmu will lament how much of Tang literature is now lost; the trope of loss that Dudbridge describes lives on. If my hypothesis is correct, however, a third of those books do survive, disassembled and hidden beneath the categories of Wenyuan yinghua. A good part of the jibu of the imperial library of the 980s, 42  The fact that almost 40 percent of Du Fu’s current collection was not included in a composite of two large editions circulating in the mid-eleventh century, combined with the knowledge that there was a premium on discovering hitherto unknown Du Fu poems, should give us some unease when we delve into the complete poems as we have them. 43  One might reasonably object that there should be a significant overlap between poems in Yang Yi’s initial xiaoji and our hypothetical xiaoji in the imperial library. There is, however, remarkably little overlap between the forty Li Shangyin poems in the mid-tenth-century Shu anthology Caidiao ji 才調集 and the Wenyuan yinghua selection, suggesting the circulation of xiaoji representing very different tastes.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1124

Owen

the last trove of Tang and Five Dynasties manuscripts gathered from Bianjing, Jinling, Jiangling, and Chengdu, are still extant. The nature of the manuscript tradition and the problems inherent in transmitting a collection like Wenyuan yinghua do not necessarily make it the best textual source, but we should never forget that for many titles and works it is the oldest textual source.44 The centuries that followed belonged to scholars like Song Minqiu, who carefully conserved and tried to make single collections out of many sources. Though grateful for their efforts, we should always remember how those sources came to them, usually in bits and pieces; and in knowing that, we will understand how tenuous our link is to the Tang author’s collection as it once existed in the Tang. Our understanding of Tang literature came through a filter of destruction and dispersal, in which material was preserved by good fortune. The probability of preservation was in direct proportion to the number of manuscripts in circulation and thus to the popularity of authors and works at a particular moment in history. Song Taizong’s edict to recover works in the old catalog helped preserve some rare complete collections from before the Tianbao reign of the Tang. Some of those survived into the mid-eleventh century and some did not. Apart from this act of imperial will, Song scholars found what had been copied. When we speak of “popularity,” we are referring primarily to late ninth- and tenth-century taste. To understand why we have what we have, it is important to understand the competing values in that particular period. Han Yu’s immense prestige as a prose writer no doubt led to the recovery of his complete poetry collection. If we look at the 45 poems by Han Yu preserved in Wenyuan yinghua, we presume a xiaoji of about 135 poems in the imperial library. Virtually all of the Wenyuan yinghua’s selections represent what now seem an odd and unrepresentative “take” on Han Yu’s poetry; it includes primarily his social regulated verses and rhetorically inventive yongwu. We must allow the possibility that the skewed selection represents the choices of the poetry editors of Wenyuan yinghua; however, the alternative possibility is that this 44  The problem with using Wenyuan yinghua as the “oldest” source is that Wenyuan yinghua itself has been edited; moreover, the greater part of the compilation survives only in the Ming edition, probably a product of one of the manuscript copies made from the edited Song printed edition. One important source here to check the probabilities of textual variance is the Wenyuan yinghua zuanyao 文苑英華纂要 from the twelfth century (before it was edited). This book survives only in a Yuan edition, whose pages are sometimes incomplete. This is a selection of passages. Comparison with the current edition of Wenyuan yinghua shows that the texts are substantially the same, though there are a number of variants.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Manuscript Legacy Of The Tang: The Case Of Literature

1125

represents the normative choices of the copyist who compiled the xiaoji, reading Han Yu’s poetry through the literary interests of the late ninth and tenth centuries. Here and in other cases the discovery of a more complete collection shows us how radically skewed the earlier versions were. We should then ask ourselves if the same might not be true of that large majority of literary collections where we may have supplements from other xiaoji but clearly do not have the full collection as it once existed. If we cannot reconcile contemporary Tang statements on a writer with the works that survive, as is often the case, we have some rather clear evidence in the line of transmission why this might be so. Each case must be understood individually; nevertheless, by and large our “Tang” has been filtered not through a tradition of scholarly preservation, but through acts of partial copying largely determined by period taste (at the same time acknowledging that the period taste of the late ninth and tenth centuries was various and complex). This history of loss and recovery is as much an essential part of the history of Tang literature as the works we discuss with such intense regard. Even though we cannot repair the losses, if we clearly understand just how the historical record is skewed, in our accounts we can acknowledge how and where the record is probably biased.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections in Tang Dynasty China Christopher M. B. Nugent With* the substantial amount of Tang dynasty poetry available to us today, from surviving Song printed editions to computerized databases, it is easy to overlook the vast divide that separates us from this literature as it was produced and circulated in the Tang itself. Poetry in the Tang existed in many forms, from songs on the lips of singing girls to poems meticulously copied out and stored on scrolls of paper. Yet the singing girls are long dead and, with the exception of a small number of poetic works from the finds at Dunhuang, the scrolls are lost.1 We have almost no access to poetry as it existed and was experienced in the Tang itself. And this is no trivial matter. Every Tang poem that we read today was by definition composed and circulated during the Tang before its long and circuitous path to the present.2 If we do not understand how Tang people composed, experienced, and circulated their poetry, then we are missing something very fundamental about the literature in its original contexts. In this article I examine one of the key means of preserving and circulating written poetry during the Tang: literary collections consisting of the works of a single author. Literary collections were not new in the Tang, but in the course of the dynasty they reached a level of popularity that was previously Source: “Literary Collections in Tang Dynasty China,” T’oung Pao 93 (2007): 1–52. *  I would like to thank Wilt L. Idema and Sarah M. Allen for their careful reading and useful criticisms of drafts of this article. I presented a portion of this study at the workshop “The Early Development of Print Culture in China” (Fairbank Center for East Asian Research, Harvard University, Cambridge, Mass., April 20, 2005) and benefited greatly from the many comments by the participants. I would also like to thank the anonymous referees for T’oung Pao for a number of helpful suggestions, editorial and otherwise. 1  For a comprehensive treatment of the poetic texts found at Dunhuang, see Xu Jun 徐俊, Dunhuang shiji canjuan jikao 敦煌詩集殘卷輯考 (Beijing: Zhonghua shuju, 2000). The physical evidence provided by limited Dunhuang poetic manuscripts provides a different, if equally important, perspective on Tang poetic production and circulation. They do not, however, tell us much about Tang literary collections per se, as the finds include no complete collections of this sort. I will examine some of the important poetic manuscripts from Dunhuang in a forthcoming article focusing on textual variation in multiple copies of a number of Tang poems. 2  For an extensive discussion of the history of the Tang literary collections that exist today (in printed, not original manuscript form), see Wan Man 萬曼, Tangji xulu 唐集敘錄 (Taipei: Mingwen shuju, 1982).

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_036

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1127

unknown.3 Thanks to the wide availability of paper and an increasing understanding of the importance of such collections in establishing a writer’s reputation, both in the present and in the future, the compilation of individual literary collections became the norm rather than the exception.4 I will argue here that individual literary collections, especially by the Mid- and Late Tang, were read widely and constituted one of the primary means of access to the works of poets from earlier in the dynasty. Perhaps more importantly, these collections, along with the Tang anthologies for which they were the source texts, ultimately formed the basis for Song printed editions, the primary avenue through which Tang poetry survived into later periods. Inspired by the increasing volume of excavated manuscripts that differ, sometimes significantly, from the received textual tradition, scholars working on literature from what is commonly called “early” China have lately begun to focus more attention on the specific contexts of literary production and circulation. How exactly, they are asking, “were texts composed, preserved, transmitted, and received?”5 Here I will devote the same kind of attention to the modes of textual preservation, circulation, and reception in the Tang period. This topic has been almost completely neglected in English-language scholarship and is only beginning to attract the attention of scholars in mainland China and Taiwan. While the work of the latter is of great value, there has been a tendency in it to focus too heavily on the examples of Bai Juyi 白居易 (772–846) and his contemporary Yuan Zhen 元積 (779–831).6 Both poets wrote 3  Fan Zhilin 范之麟, “Tangdai shige de liuchuan (shang)” 唐代詩歌的流傳(上), in Tangdai wenxue 唐代文學, 5 (April, 1984): 278. It has been estimated elsewhere that there were probably no less than a thousand such collections compiled during the Tang. See Shang Xuefeng 尙學鋒 et al., Zhongguo gudian wenxue jieshou shi 中國古典文學接受史 (Ji’nan: Shandong jiaoyu chubanshe, 2000), p. 245. Note also that many, though not all, literary collections contained works other than poetry: prose writings were often present as well. In some cases the prefaces to these collections would specify the various genres included. Most, however, employed a more general term such as “poetry and prose writings” (shibi 詩筆). My discussion in this essay focuses specifically on poetry, as this was also the focus of most collectors and the main content of many collections. At the same time, many of the issues I will address apply to the circulation and preservation of prose writings as well. 4  Fan Zhilin, part i, p. 278. 5  Martin Kern, ed., Text and Ritual in Early China (Seattle: University of Washington Press, 2005), p. ix. 6  For examples in Chinese secondary scholarship see Fan Zhilin, who relies extensively on the writings of Bai Juyi and Yuan Zhen, and Shang Xuefeng et al., who do so for their speculations on the spread of printing in the Tang (p. 245). For an example in English see Denis Twitchett, Printing and Publishing in Medieval China (New York: Frederic C. Beil, 1983), p. 17.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1128

Nugent

extensively on the circulation and preservation of their literary works and are thus appealing subjects for study. Yet their seemingly obsessive focus on such matters also implies exceptionality. The example of these two writers looms large and I will not ignore it; still, by including in my discussion a number of other writers and commentators I hope to provide a more balanced and broadly valid perspective. The greater part of this study examines the full process of compiling a literary collection in the Tang, from collecting the poems to editing and copying them into a formal collection. I focus on the Tang prefaces to these collections, not as literary works in their own right (though they are also that), but as sources of evidence for the particulars of how Tang collections were created. These prefaces reveal a world in which the survival of literature was recognized to be distinctly tenuous. The costs and effort required for textual reproduction in an age before print meant that any given person was unlikely to have multiple copies of a given work; poems were seen as being in constant danger of falling out of circulation and disappearing for good. Moreover, at each stage of the process of creating collections their content was altered. This happened both on the macro level of selecting some works for inclusion while rejecting others, and on the micro level of changing individual poems for a variety of reasons. I will conclude by briefly discussing some differences between the attitudes of Tang collectors and compilers toward their work and the approach of those who would undertake similar endeavors in the Song. The Tang sources I examine reveal a literary world in which the fluidity of poems and of their textual manifestations was accepted as the norm. This is a very different world from the one that was produced by the expansion of print technology and the advent of a more scholastic attitude toward poetry in later periods. By looking at the contexts in which Tang poetry was actually composed and first circulated, preserved, and received, our understanding of what Tang poetry was and is will stand on firmer ground.

Gathering the Texts

Accounts of the first stages of compiling a literary collection—that is, gathering together the materials that will ultimately form the basis for that collection—often emphasize the difficulties faced by the compiler. Through their depictions of loss, the prefaces portray the tenuousness of poems as physical objects always subject to the hazards of social upheaval, intentional destruction, and simple neglect. This concern is especially prominent in writings from the Early and High Tang. Poets of the seventh and eighth centuries showed

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1129

far less interest in maintaining their own collections than did their successors in later periods, and the compilers of their works are thus particularly focused on this issue of loss. As Lü Cai 呂才 notes in his preface to the collected works of the Sui and Early Tang poet Wang Ji 王績 (586–644), “Many of the poems and rhapsodies he composed are scattered and lost. With the collecting and searching as yet unfinished, I have for the time being compiled [his works] into five juan” 所著詩賦，並多散逸，鳩訪未畢，且緝成五卷.7 According to Lü’s account, Wang Ji often wrote in a state of spontaneous drunkenness, his poems being recorded and circulated only through the efforts of “those who enjoyed them” (haoshizhe 好事者).8 The implication is that without the efforts of these “fans,” even more of Wang Ji’s works would have disappeared. Wang Jin 王縉 (d. 781) expresses a similar anxiety about the works of his elder brother, the High Tang poet Wang Wei 王維 (701–761). In his “Memorial on Presenting Wang Wei’s Collection” 進王維集表, of 764, Wang Jin laments that, “Some [of his works] were scattered among his friends, some were left in bamboo trunks. I recently searched for them and still worry about those that are scattered and lost” 或散朋友之上，或留篋笥之中°，尙慮零落.9 Wang Wei’s status as a poet quite famous in his own day and, moreover, a member of two of the Tang’s most important aristocratic families, must have ensured that there were a good number of his poems both in circulation and in the possession of members of his social circle. At the same time, the social upheaval of the An Lushan Rebellion, during which Wang Wei was accused of treason, clearly made locating copies

7  Lü Cai, “Donggaozi houxu” 東皋子後序, in Quan Tang wen 全唐文 (hereafter qtw) (Beijing: Zhonghua shuju, 1983), 160.1640. 8  q tw, 160.1639. 9  q tw, 370.3756–7. For a slightly different wording of the cited passage, see Zhao Diancheng 趙殿成, Wang Youcheng ji jianjiao 王右丞集箋校 (Shanghai: Shanghai guji chubanshe, 1998), p. 494. Note that Wang Wei and his brother were members of elite clans, their fathers being the Taiyuan Wangs 太原王 and their mothers the Boling Cuis 博陵崔. Fan Zhuanzheng 范傳正, who compiled a collection of works by Wang Wei’s contemporary Li Bai 李白 (701– 762) about fifty years after the latter’s death, describes Li Bai’s poems as being in a state similar to that of Wang Wei’s. In his “New Funeral Tablet for the Tang Reminder of the Left, Hanlin Academician, Master Li” 唐左拾遺翰林學士 李公新墓碑, Fan Zhuanzheng writes of the poems in the collection that, “Sometimes I obtained them from literati of the time, sometimes I got them from his ancestral clan members. I compiled the incomplete fragments in order to circulate the collection in the age” 或得之于時之文士，或得之于公之宗 族，編緝斷簡，以行于代. See Fan Zhuanzheng, “Tang zuoshiyi Hanlin xueshi Li gong xin mubei,” in Qu Tuiyuan 瞿蛻園 and Zhu Jincheng 朱金城, eds., Li Bai ji jianjiao 李白集箋校 (Shanghai: Shanghai guji chubanshe, 1998), “fulu” 2.1781–2.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1130

Nugent

of his various works a more difficult endeavor than it might otherwise have been.10 Some poets were so critical of their own literary output that they themselves endangered its survival, leading compilers to desperate measures likely to result in questionable attributions. Wang Shiyuan 王士源, the compiler of the first known collection of Meng Haoran’s 孟浩然 (689–740) poems, encountered circumstances substantially more difficult than those faced by Lü Cai and Wang Jin in their efforts. He writes: Whatever Haoran composed he would soon destroy or discard without editing or copying it down, often sighing to himself that the written text did not capture his intent. Since his wanderings and wayfarings were many, his pieces and drafts were scattered and lost. I have offered rewards for them, gathering them from parish and hamlet, but don’t have even half of them. 浩然凡所屬綴，就輒毀棄，無編錄，常自嘆爲文不逮意也。流落既多，篇草 散逸，鄕里搆採，不有其半。11

Unlike Wang Ji, who was merely neglectful of the material of his literary legacy, Meng Haoran took to heart the notion from the Yijing 易經 that “writing does not fully capture speech, and speech does not fully capture meaning” 書不 盡言，言不盡意, and destroyed much of what he created before it could be ordered or even recorded. Moreover, as Meng only held an official position for a very short time, had numerous friends, and spent much of his life traveling, his works were necessarily difficult to find in any one location.12 Extreme 10  When An Lushan and his forces seized the capitals in 756, Wang Wei was captured and forced to serve the usurping regime despite alleged attempts to poison himself and feign muteness. After the Tang armies recaptured the capitals, Wang Wei was held on charges of sedition. A loyalist poem he is said to have orally composed during his imprisonment by An Lushan was cited as a mitigating factor that led to his release, though, as Marsha Wagner points out, the intercession of his politically powerful brother was probably a far more important factor in his rehabilitation. See Marsha Wagner, Wang Wei (Boston: Twayne Publishers, 1981), pp. 53–4. 11  Wang Shiyuan, “Meng Haoran ji xu” 孟浩然集序, in Tong Peiji 佟培基, Meng Haoran shiji zhu 孟浩然詩集注 (Shanghai: Shanghai guji chubanshe, 2000), p. 433. This translation is based on the one by Paul W Kroll in “Wang Shih-yüan’s Preface to the Poems of Meng Hao-jan,” Monumenta Serica 34 (1979–80): 350. 12  Kroll notes that Meng’s “official rank was quite humble, only one step from the bottom of the thirty-rung bureaucratic ladder” and that he held it for less than a year. See Paul W. Kroll, Meng Hao-jan (Boston: Twayne, 1981), pp. 79–80. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1131

actions on the part of the poet inspired extreme actions by the compiler. Using a method that is unique in the descriptions of Tang poetry collection (though quite possibly not unique as actual practice), Wang Shiyuan “offered rewards” for poems ostensibly by Meng Haoran. It goes without saying that such a practice would provide great incentive to come up with poems in Meng’s style, and thus throw the legitimacy of many attributions into doubt. Yet even this methodologically flexible collection process and the imitations it likely inspired left Wang Shiyuan bemoaning how limited his discoveries ultimately remained. If Meng Haoran’s poems suffered at the hands of their self-critical creator, the writings of other poets fell victim to larger forces. Liu Yuxi’s 劉禹錫 (772–842) preface to Lu Xiang’s 盧象 collection simply states, “Seventy-three years after he died, his grandson Yuanfu presented [Lu Xiang’s] remaining drafts to me and begged for words to show with it. Having gone through chaos and turmoil, there were many pieces that had been scattered and lost. Those that survive today come to twelve juan …” 公下世後七十三年，其孫元符捧遺草來乞詞以 表之。嘗經亂離，多所散落，今之存者，十有二卷.13 Liu Yuxi thus makes explicit what Wang Jin only implied: rebellion and social upheaval could take a toll not only on people but on poetic works as well. Li Yangbing’s 李陽冰 preface of 762 to a collection of Li Bai’s works, the Grass Hut Collection of Hanlin Academician Li of the Tang 唐李翰林草堂集, describes encountering the poet on his death bed, surrounded by tens of thousands of rolls of writings in draft form, unordered and uncorrected. Li Yangbing writes, “Since the troubles in the Central Plains, [Li Bai] had fled [that] land for eight years. Of [the works] he had composed at that time, nine out of ten are lost. Of those that survive today, all of them were obtained from other people” 自中原有事，公避地 八年，當時著述，十喪其九。今所存者，皆得之他人焉.14 The “troubles” to which he refers are, of course, the An Lushan Rebellion, during which Li Bai, like Wang Wei, was accused of treason, in this case for his association with a prince who attempted to set up his own state.15 In his years of imprisonment and wandering between 755 and his death in 762, Li Bai’s works were scattered and eventually reassembled in part by the poet himself. Yet even his diligent efforts to reclaim his writings and pass them on to be ordered and prefaced resulted in a collection that represented but a fraction of his earlier oeuvre.

13  Liu Yuxi, “Tang gu shangshu zhuke yuanwailang Lu gong wenji xu” 唐故尙書主客員外 郎盧公文集序, in Liu Yuxi ji 劉禹錫集 (Beijing: Zhonghua shuju, 2000), 19.234. 14   Li Bai ji jianjiao, “fulu” 2.1790. 15  For a brief account of this incident and Li’s travels in the years that followed, see Stephen Owen, The Great Age of Chinese Poetry: The High T’ang (New Haven: Yale University Press, 1981), pp. 117–8. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1132

Nugent

The related case of Luo Binwang 駱賓王 (c. 640–684), one of the famed “Four Elites of the Early Tang” (Chu Tang sijie 初唐四傑),16 is interesting in that his works were both lost and recovered thanks to intrigue and decisions made at the highest levels of court. Xi Yunqing 郗雲卿, the compiler of Luo’s collection, writes: In the Wenming reign period (684) [Binwang] schemed together with Xu Jingye17 in Guangling to raise a righteous rebellion. The rebellion failed, leading him to flee. This then caused his literary collection to be completely scattered and lost. Later, the court of Emperor Zhongzong18 sent down an order to seek out Binwang’s poems and writings, commanding me, Yunqing, to collect them. The recorded ones that were missing at that time came to ten juan. This collection is made up of pieces stored in private homes and it is indeed worth circulating among those who enjoy [such things]. 文明中，與敬業於廣陵共謨起義兵。事既不捷，因致逃遁，遂致文集悉皆散 失。後中宗朝降敕搜訪賓王詩筆，令雲卿集焉。所載者，即當時之遺漏，凡 十卷。此集並是家藏者，亦足傳諸好事。19

Xi’s account implies that these works had apparently ceased to circulate. They were no longer in the public domain but rather stored away in private homes. This situation is not surprising given the political upheavals at the time. And just as Wang Shiyuan’s rewards provided incentive to “discover” that one had a number of Meng Haoran’s poems in one’s possession, so did Luo Binwang’s association with a failed rebellion surely make people more likely to either

16  For a full discussion of the Chu Tang sijie, see Stephen Owen, The Poetry of the Early T’ang (New Haven: Yale University Press, 1977), pp. 79–150. 17  Xu Jingye was also known by the surname Li 李. His grandfather, originally named Xu Shiji 徐世勣 (594–669), was a rebel from Shandong who eventually became a military leader for the Tang and was granted the surname Li. In 684 Xu Jingye raised a rebellion against Empress Wu that ultimately failed. The rebellion and Luo Binwang’s role in it are covered in Denis Twitchett, ed., The Cambridge History of China, vol. 3: Sui and T’ang China, 589–906, part 1 (Cambridge: Cambridge University Press, 1979), pp. 294–300. 18  Zhongzong ruled for most of January and February of 684 and for the period from 705–710 after the restoration of the Tang. See Paul W. Kroll, “The True Dates of the Reigns and Reign-Periods of the T’ang,” T’ang Studies 2 (1984): 25–6. 19  Xi Yunqing, “Luo Binwang wenji xu” 駱賓王文集序, from Luo Cheng ji 駱丞集 (Shanghai: Shangwu yinshuguan, 1937), p. 1.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1133

destroy their copies of his poems or at least keep quiet about them until the political winds changed, as they eventually did. In all these cases poems are portrayed as fragile things, and whether through neglect, the author’s own hand, or the impact of historical events, they were easily lost or destroyed. Underlying this anxiety over their fragility is a keen awareness of the physicality of literature in this period. Poems could be destroyed, scattered, and lost because they were essentially rare objects. That is, even in the case of the best known and socially best connected writers, the arduous process of textual reproduction in an age before print meant that there would always be limits on how many copies of any given work could exist. There were no print runs, and very few contexts in which such works of literature as poems could be purchased. It was thus a real possibility that all copies of a given work might disappear. Writers in the Mid- and Late Tang period typically took a more active role in maintaining a set of their literary works that could eventually be ordered into a formal collection; this may have been due, at least in part, to an awareness of the literary losses caused by such events as the An Lushan Rebellion.20 There are a number of examples of writers who passed on their collected writings to trusted friends when they feared that their writings or lives might be at risk. In his preface to Liu Zongyuan’s 柳宗元 (773–819) collection, Liu Yuxi recounts how Liu Zongyuan left him a letter, saying, “I’m ill-fated to die demoted, and so trouble an old friend with my remaining drafts” 我不幸卒以謫死，以遺草累 故人.21 The term “remaining drafts” (yicao 遺草) is similar to the caogao 草稿 that Li Yangbing relates having received from Li Bai. Both indicate works that are in need of revision and are being passed on in such a state out of a sense of urgency. Yuan Zhen’s “Letter Discussing Poetry, Sent to Letian” (“Xushi ji Letain shu” 敘詩寄樂天書, from 812) similarly describes how Yuan left a set of texts with his good friend Bai Juyi when he was demoted and exiled to Jiangling 江陵 for his political involvements.22 Unlike Liu Zongyuan, however, Yuan Zhen sets 20  Circulating one’s writings in the form of small collections often referred to as xingjuan 行卷 was also becoming an increasingly important means of attracting the attention of officials who might be helpful in advancing one’s career. See Fu Xuancong 傅璇琮, Tangdai keju yu wenxue 唐代科舉與文學 (Shaanxi: Shaanxi renmin chubanshe, 1986), and Cheng Qianfan 程千帆, Tangdai jinshi xingjuan he wenxue 唐代進士行卷和文學 (Shanghai: Shanghai guji chubanshe, 1980). 21  Liu Yuxi, “Tang gu shangshu libu yuanwailang Liujun wenji ji” 唐故尙書禮部員外郎柳 君文集紀, Liu Yuxi ji, 19.236–7. 22  Jiangling is in modern Hubei. For a detailed discussion of Yuan Zhen’s political career and opinions see Lily Hwa, “Yuan Chen (ad 779–831): The Poet-Statesman, His Political and Literary Career,” Ph.D. dissertation, University of Illinois, 1984. Hwa speculates that Yuan

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1134

Nugent

out the content of these materials in considerable detail, discusses his ideas on poetry and distinguishes ten specific forms (ti 體) among his poems, ranging from “poems of social criticism in the ancient style” (gufeng 古諷) to “poems of seductive allure in ancient and modern style” ( jingu yanti 今古艷體).23 He writes, “From when I was sixteen to this seventh year of the Yuanhe reign period, I’ve already accumulated over eight hundred poems… When I recently came to the capital, they happened to be in my boxes and trunks. I’ll deposit all of them with you when I leave for Tongzhou” 自十六時，至是元和七年，已 有詩八百餘首… 昨來京師，偶在筐篋，及通行，盡置足下.24 Though Yuan Zhen may have left his works with Bai Juyi out of fear that they would perish with him in exile, he had apparently prepared them well in advance, as evidenced by their being carefully ordered and categorized by type. It was probably this categorical schema that inspired Bai Juyi, the Tang’s most dedicated custodian of his own works, to come up with a similar system to classify them.25 Even in the Late Tang, when maintaining copies of one’s own works had become common practice, the task facing a compiler could be difficult. In his postface to the Late Tang poet Du Mu’s 杜牧 (803–852) Fanchuan Literary Collection (Fanchuan wenji 樊川文集), the poet’s nephew Pei Yanhan 裴延翰 relates that Du Mu had requested that he compile his works and write a preface for the resulting collection. Unfortunately, while feeling ill one day, Du Mu gathered thousands of pages of his writings and burned all but a small number of them.26 Not one to be daunted, some years later Pei Yanhan went about putting together the collection from other sources. “Even though I went to hidden and difficult-to-find places,” he wrote, “I did not think of many thousand li as being far. I insisted on having the works copied out and shown to me” 雖適僻 阻，不遠數千里，必獲寫示.27 Du Mu’s actions, though apparently the result of fever-induced illusion, were similar in effect to Meng Haoran’s more methodical attempts to destroy Zhen left his works with Bai Juyi partly out of fear that he would not be allowed to return to the capital. See p. 105. 23  See Anna M. Shields, “Defining Experience: The ‘Poems of Seductive Allure’ (yanshi) of the Mid-Tang Poet Yuan Zhen (779–831),” Journal of the American Oriental Society 122 (2002): 61–78. 24  Yuan Zhen, “Xushi ji Letian shu,” Yuan Zhen ji 元鎭集 (hereafter yzj) (Beijing: Zhonghua shuju, 1982), 30.352. 25  This is based on claims made by both Angela C. Y. Jung Palandri and Hwa. See Hwa, p. 121 and Palandri, Yüan Chen (Boston: Twayne Publishers, 1977), p. 64. 26  Pei Yanhan, “Fanchuan wenji houxu” 樊州文集後序, in Feng Jiwu 馮集梧, ed., Fanchuan shiji zhu 樊川詩集注 (Shanghai: Shanghai guji chubanshe, 1998), p. 5. 27  Pei Yanhan, p. 5.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1135

works that he found wanting. Both examples show that the dangers faced by poems were not limited to chaos and rebellion; simple authorial discontent and even temporary delusion were threats as well. Yet beyond its amusing twist, Pei Yanhan’s postface does give us some insights as to the state of a poet’s draft manuscripts. In describing the documents that Du Mu set aflame, Pei Yanhan hyperbolically states that there were “a hundred thousand pages (zhi 紙),” whereas collections were more frequently described in terms of either scrolls, juan 卷, or pieces, pian 篇. Even if Du Mu had not burned his works, the image of Pei Yanhan—or any other compilers—having to edit, order, and copy thousands of pages of text gives us an idea of just how difficult an undertaking putting together a collection could potentially be. A final example from the late Tang demonstrates the broad range of sources of which compilers would avail themselves in their pursuits. In his preface to the monk Guanxiu’s 貫休 (832–912) second collection, the Chan Moon Collection 禪月集,28 Tan Yu 曇域 gives a detailed description of the process he went through to compile his teacher’s works: The burial procedures were complete and the mourning finished. On their free days sometimes special worthies would come asking, sometimes guests from court would come seeking. Sometimes they would read one or two of our late teacher’s works and sometimes jot down a couple of lines… Everyone begged me, Tan Yu, to collect and edit the earlier and later poems, prose pieces, and encomiums that he had composed. Every day someone came to inquire [about the work] and I had no time to resist them. I thereupon sought out and examined draft manuscripts and people who had set [his works] to memory. There were around a thousand pieces. Then I had printing blocks carved of the whole set and entitled it the Chan Moon Collection. 葬事既周，哀制斯畢。暇日或勛賢見訪，或朝客見尋，或有念先師一篇兩 篇，或記參句五句・・・眾請曇域編集前後所製歌詩文贊。日有見問，不暇 枝梧。遂尋檢稿草，及暗記憶者，約一千首。乃雕刻成部，題號禪月集。29

Most accounts depict collection efforts as focusing on the location of written texts, with source materials described in distinctly physical terms: they might be “scattered and lost,” or “stored” (cang 藏) in “trunks” or “cases” (qie 28  Guanxiu’s first collection, the Xiyue ji 西岳集, was put together by the monk himself in 896 and prefaced by Wu Rong 吳融. 29  Tan Yu, “Chanyue ji xu” 禪月集序, qtw, 922.9604.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1136

Nugent

篋);30 authors also speak of “disordered drafts” that are “piled up.” Pei Yanhan

requested that texts be copied out or shown to him. Tan Yu’s account describes similar materials, noting that he sought out “draft manuscripts” (gaocao 稿草), like those passed on to their compilers by Li Bai and Liu Zongyuan. However, his mention of “those who had set his works to memory” or, more literally, “who silently remembered them” (anjiyizhe 暗記憶者), clearly speaks of a different type of source, indicating that Tan Yu sought out people who had memorized works by Guanxiu and could reproduce them. Whether he would have these people recite the pieces so that he could write them down, or have them write them out from memory themselves, is not clear. In either case, his preface offers strong evidence that at least some collection compilers made use of both written and oral sources when putting together collections. It is also the first mention in the Tang of printing blocks being carved for a literary collection, and thus gives us a glimpse of the diverse literary world of the Late Tang—a world in which oral culture, manuscript culture, and the beginnings of print culture coexisted and overlapped. Tan Yu’s extensive efforts to gather as many of his master’s works as possible returns us to the reality of how much he and other compilers must have missed. Monasteries were important sites of textual preservation in the Tang— and yet the works of such a prominent monk as Guanxiu, who was also a wellknown writer, artist, and calligrapher, were scattered and difficult to locate. One can only assume that those of ordinary scholar officials or of the more peripatetic poets, such as Meng Haoran, Li Bai and Du Fu, would have suffered even greater losses. By detailing what has survived, prefaces inevitably remind the reader of what has not. The clearest examples of this are a number of prefaces which note that the compiler was able to find only fragments of certain works, or ephemeral traces that remained to give a hint of what was missing. Thus, Dugu Ji’s 獨孤及 (725–777) preface to the works of Li Hua 李華 (d. 774) lists a long series of titles of works ranging from stele inscriptions to shi poems, and then notes, “All of them were lost because of the [An Lushan] rebellion. The names survived but the pieces themselves have disappeared” 並因亂

30  Trunks are the most common storage place quoted in prefaces. See Shen Ziming’s account of storing Li He’s collection, described by Du Mu in “Taichangsi feng lilang Li He geshiji xu” 太常寺奉禮郎李賀歌詩集序. Other examples include Wang Wei’s poems, about which Wang Jin writes, “… a few were left in bamboo trunks” 或留篋笥之中, and the works of Cen Shen 岑參 (715–770), discussed below, whose heir is said to have “possessed [Cen Shen]’s remaining writings and stored them in a trunk” 有公遺文貯之筐 篋. See Du Que 杜確, “Cen Jiazhou ji xu” 岑嘉州集序, qtw, 459.4692.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1137

失之，名存而篇亡.31 Li Hua himself expresses a strikingly similar note when, in his preface to the works of Xiao Yingshi 蕭穎士, he says of Xiao’s writings that have been lost, “While the titles of the pieces survive, sections and lines are missing. This is what in antiquity was called ‘to have the meaning but lack the words’ ” 其篇目雖存，章句遺落。古所謂有其義無其詞者也.32 The act of collection itself thus creates, ironically, the figure of loss: only when someone attempted to compile the entirety of an author’s works were the gaps discovered and given concrete expression in prefaces. The loss of some or even many of an author’s works did not become a fact until an attempt definitively to “set” a collection had been made. By setting the collection, the lacunae in the collection were also set.



Editing and Copying

The materials gathered or received by compilers33 were never considered to be finished products; rather, they were the raw materials out of which a collection could be made. Both the compilers and the writers themselves consistently refer to these sources as “uncorrected” (weixiu 未修), “neither ordered nor copied” (wu bianlu 無編錄), or as “drafts” (caogao, yicao). The implication of such terms was that the pieces would undergo intentional alteration in the process of being turned into finished collections—they would be fixed and improved. In her study of textual transmission in the Song, Susan Cherniack points out that “the sanction for textual change was from the beginning implicit in the role of the editor as one who transmits.”34 She demonstrates that even with the Confucian classics, “where the desire for textual perfection and the prohibition against change posed by traditional authority are at their strongest,”35 change and emendation were constant features of transmission and reproduction. If this was so with the classics, one can imagine it would be even truer of works 31  Dugu Ji, “Jianjiao shangshu libu yuanwailang Zhaojun Li gong zhongji xu” 檢校尙書 吏 部員外郎趙郡李公中集序, qtw, 388.3947. 32  Li Hua, “Yangzhou gongcao Xiao Yingshi wenji xu” 揚州功曹蕭穎士文集序, qtw, 315.3198. 33  Compilers, who played an active role in ordering and editing the contents of a collection, often in addition to composing a preface, must be distinguished from simple preface writers who were only responsible for a preface to head the collection. Although the two roles frequently overlapped, this was not always the case. 34  Susan Cherniack, “Book Culture and Textual Transmission in Sung China,” Harvard Journal of Asiatic Studies 54.1 (1994): 18–19. 35  Cherniack, p. 19.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1138

Nugent

that lacked any sort of canonical authority and tradition. A Qing-dynasty editor might have been disinclined to alter a line in one of Li Bai’s by then wellknown and oft-anthologized works, but it is doubtful that Li Yangbing or even Fan Zhuanzheng would have felt similar compunction in compiling the earliest collections of his poems. The first step in the editorial process—choosing which poems to include (and thus which to exclude)—followed the example set by Confucius himself in his legendary editing of the poems of the Shijing. The “Hereditary House of Confucius” (Kongzi shijia 孔子世家) from Sima Qian’s Shiji, a prominent source for the legend, reads as follows: “In ancient times there were over three thousand of the Poems. Then Confucius got rid of redundant ones and chose those that would propagate ritual and righteousness” 古者詩三千餘篇， 及至孔子 去其重， 取可施於禮義.36 Sima Qian’s account makes it clear that one element of Confucius’s motive in “culling the poems” (shanshi 删詩) was to give a moral import to the work by cutting out the licentious pieces and keeping only those that could be interpreted as setting appropriate ethical examples. Tang preface writers are not often explicit about the need to cut out certain poems (and certainly never imply that their subjects wrote licentious poems). However, the following passage suggests that such selectivity was considered a normal part of the compilation process and did not bear mentioning unless unusual circumstances applied. Sun Guangxian’s 孫光憲 (d. 968)37 preface to the Buddhist monk Qiji’s 齊己 ( fl. 881) poetry collection, the White Lotus Collection 白蓮集, states: As for all the draft copies of poems from the master’s life, he never had the spare time to cull them. All of a sudden he unexpectedly died and his disciple Xi Wenbing was given [the writings] that had been collected. He was then able to edit them into 810 pieces that he bound as ten juan and entitled The White Lotus Collection. 師平生詩稿，未遑刪汰，俄驚遷化。門人西文倂以所集見授，因得編就八百 一十篇，勒成一十卷，題曰白蓮集。38 36  Sima Qian 司馬遷, Shiji 史記 (Beijing: Zhonghua shuju, 1959), 47.1936–7. Note that this legend, though it had already come under attack by the Tang, was still an important cultural template. For a more detailed treatment of Confucius’s purported editorial roles, see Cherniack, pp. 15–17. 37  Sun Guangxian is best known as the author of the important Five Dynasties anecdote collection Beimeng suoyan 北夢瑣言. 38  Sun Guangxian, “Bailian ji xu” 白蓮集序, qtw, 900.9390–1.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1139

Here we see “culling” (shantai 刪汰) noted because Qiji died before he could perform this essential task himself. The responsibility then fell upon his disciple Xi Wenbing. Let us return for comparison to Guanxiu’s Chan Moon Collection, which must have been compiled by Tan Yu under similar circumstances. Guanxiu’s works at the time of his death can be put into three general groups. The first consisted of the works included in his self-compiled collection of 896, the Western Peak Collection (Xiyue ji 西岳集), which would have already gone through the culling process. The second was the works that he had written in the sixteen years since then and that were in his possession. There is no indication as to whether or not he had begun preparing these to be put into collection form. The third group was those works that Tan Yu describes hunting down from both written and oral sources after Guanxiu’s death. Given Sun Guangxian’s description of how Qiji’s collection—also the work of a prominent poet-monk edited by one of his disciples—was put together, it is very likely that Tan Yu culled poems from at least the second group of Guanxiu’s works, and definitely from the third. In other words, works originally gathered for a collection would not necessarily be included in that collection. They were merely raw source material that still needed to be picked through and culled. According to Sima Qian, Confucius’s criteria for exclusion were both practical and moral: he eliminated poems that either were redundant or did not propagate “ritual and righteousness” (liyi 禮義). The standards underlying Tang editorial practices were more varied. In the self-preface to his own collection, Sun Qiao 孫樵 (fl. 860–888) implies an aesthetic basis, stating that, of the two hundred pieces he looked over, “I gathered those worth looking at into thirtyfive sections and compiled them into ten juan. I stored them in a bamboo chest in order to pass them down to my sons and grandsons” 叢其可觀者，三十五 篇，編成十卷，藏諸篋笥，以貽子孫.39 For his part, Bai Juyi excludes some works from his collection of matching poems composed with Liu Yuxi because of their lack of formality, writing, “As for the other [poems], which followed a momentary whim or were the product of drink, and were generally produced orally, they are not counted here” 其餘乘興扶醉，率然口號者，不在此數.40 39  Sun Qiao, “Zixu” 自序, qtw, 794.8326. It should be noted that no poetic genres are included in Sun’s list of his own writings. There is also a problem with the use of the term pian 篇 here: it is typically a measure word for a piece of writing, yet thirty-five pieces seems a small number to fill ten juan. At the same time, as the pieces in this case were probably prose rather than poetry, some may have been quite long. 40  Bai Juyi, Bai Juyi ji jianjiao 白居易集箋校 (hereafter bjyjjj) (Shanghai: Shanghai guji chubanshe, 1988), “Liu Bai changheji jie” 劉白唱和集解, 69.3711.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1140

Nugent

In other contexts, however, Bai Juyi cannot bear to exclude certain pieces despite the informal circumstances of their composition and the knowledge that a later editor will likely remove them. His well-known “Letter to Yuan the Ninth [i.e. Yuan Zhen]” (Yu Yuanjiu shu 與元九書) of 815, which also served as the de facto preface for an early version of his primary collection, states: As for these remaining miscellaneous poems, some were coaxed out by a certain time or thing, they came forth as a laugh or a chant and were put together hastily, they are not what I would typically esteem. They were only taken up on occasions when relatives and friends were gathered or parting, to release sorrow or add to joy. In my arranging and ordering today, I haven’t been able to cut them out. At another time when someone collects and disseminates my writings for my sake, it will be fine to omit them. 其餘雜律詩，或誘於一時一物，發於一笑一吟，率然成章，非平生所尙者，但 以親朋合散之際，取其釋恨佐懽。今銓次之間，未能刪去，他時有爲我編集散 文者，略之可也。41

Bai Juyi here may be the exception that proves the rule. He feels that his reluctance to cut out certain poems goes against the norm and thus begs explanation. Moreover, he states that he would not object to someone else omitting the poems in a later version of the collection—of which there were many, though most were also compiled and edited by Bai himself. Other editorial standards employed in the Tang are more troubling from the standpoint of the construction of our understanding of literary history, in that compilers intentionally shaped the image of the writer to conform to their own views. The above-mentioned collection of Wang Ji’s poems in five juan, compiled and prefaced by Lü Cai prior to the poet’s death, was not the only collection of Wang’s poems put together in the Tang. In fact, as Ding Xiang Warner pointed out in her recent study of Wang Ji, “[u]ntil the 1980’s, no reader since the end of the Song dynasty had recorded seeing the original five-juan collection of Wang Ji’s works.”42 By the end of the Song Lü Cai’s collection had been eclipsed by another collection, compiled by one Lu Chun 陸淳 (d. 805?). Lu Chun’s collection was actually a redaction of Lü Cai’s original compilation, and its postface is appropriately titled “Postface to the Culled Collection of Master 41  Bai Juyi, “Yu Yuanjiu shu,” bjyjjj, 45.2795. 42  Ding Xiang Warner, A Wild Deer amid Soaring Phoenixes: The Opposition Poetics of Wang Ji (Honolulu: University of Hawai’i Press, 2003), p. 4.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1141

Donggao” (Shan Donggaozi ji houxu 刪東舉子集後序). Lu removed from Lü’s original collection any works that did not conform to his ideal of Wang Ji as a free spirit in the mold of Tao Qian 陶潛 (365–427) or Ruan Ji 阮籍 (210–263).43 Specifically, he “expunged those words of purposive intent and [so] made complete [Wang’s] ambition to be unbound” 祛彼有爲之詞，全其懸解之志.44 In doing so he reduced the contents of the collection by more than half and provided readers with a far less nuanced portrayal of Wang Ji’s works than they would have received from Lü Cai’s original collection. Most Tang editors appear to have been less extreme than Lu Chun in their standards (or at least less forthright), but it may well be that a similar process took place in innumerable other instances. It is doubtful that Wang Shiyuan accepted and paid for every poem presented to him as a work of Meng Haoran—and what better measure upon which to base his acceptance or rejection than whether or not a work sounded like Meng Haoran to him? The result would be a more uniform style that would build on itself and attract the attribution of more poems in that style to Meng Haoran. For other poets, political consideration surely came into play. When Li Bai and Wang Wei died, it had not been long since they had been charged with treason; Li Yangbing and Wang Jin would have been wise to consider any potentially controversial works among those mourned as “scattered and lost.” Whether for reasons of aesthetics, image, or politics, it is clear that individual literary collections in the Tang represented not the totality of an author’s available works, but a selection from among them. The process of reshaping a writer’s oeuvre did not come to an end once the culling of unwanted works was complete. A formal collection was finished only when a final fair copy had been made of the selected works, and this stage offered a last important opportunity for editorial changes.45 As Cherniack has demonstrated, even works with such strong prohibitions against alteration in the copying process as the classics were constantly amended and unintentionally miscopied. In a pre-print manuscript culture such as that of the Tang, 43  Lu Chun, “Shan Donggaozi ji houxu,” in Jin Ronghua 金榮華, ed., Wang Ji shiwenji jiaozhu 王績詩文集校注 (Taipei: Xinwenfeng chubangongsi yinhang, 1998), p. 388. See also Warner, pp. 4–5. 44  Lu Chun, p. 388. Note that “purposive intent,” translating youwei 有爲, is the antithesis of the Daoistic ideal of “non-action” (wuwei 無爲) and corresponds with Lu Chun’s association of Wang Ji with Zhuangzi earlier in the postface. 45  Editorial change here being considered distinct from other changes, whether intentional or otherwise, that would occur in the process of transmission after the collection was compiled.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1142

Nugent

every copy of a work represented a unique copying event and thus a new opportunity for variation to creep in. And unlike the classics, private collections of poetry were not governed by either official regulations or professional standards meant to reduce the number of such variants in the process of copying. Du Que’s 杜確 preface to Cen Shen’s 岑參 (715–770) collection is one of the earliest accounts we have of the process and it is typical in its lack of detail. Du writes, “I accepted the command to edit and arrange [the works] and thereupon ordered that a fair copy be made. I divided them into different categories and bound them into eight juan” 受命編次，因令繕錄，區分類聚，  勒成八卷.46 Elsewhere in the preface he notes that he received the works from Cen Shen’s heir, who had been storing them in trunks. In other words, we see here a very brief summary of the entire editorial process, from obtaining the pieces to editing them and having a copy made. We know that Du Que did not make a final fair copy himself, and he leaves open the important question of whom he ordered to do so. This information is, however, provided in many other prefaces, and it is clear that, typically, such copying was not done by professional scribes under anything resembling the strict conditions one might find in such official contexts as the Directorate of Education.47 In his “Self-Preface to the Black Silk Door Poems” 烏絲闌詩自序, Xu Hun 許渾(c. 787–854) describes finding himself with a surfeit of free time of which he took advantage to put together his own collection. He writes, “I edited and compiled my new and old [works] into five hundred pieces and set them on my table. I worked at leisure and took my time, as the aim was not fame … I wrote out this copy in my own hand” 編集新舊五百篇，置於几案間。聊用自適，非知之志 也・・・ 手寫此本.48 The impression here is one of informality. Putting together a finished copy of one’s literary collection was an activity of leisure, done late in life.49 Even Bai Juyi, known for his scrupulous attention to his various collections, left the job of copying at least sections of them to his young nephew. In the 829 preface to the collection of his matching poems with Liu Yuxi, he writes, “I thereupon ordered my young nephew Gui to edit, copy, and bind them into two juan. He then wrote out two copies. One I gave to Gui and one I presented to Mengde’s (i.e. Liu Yuxi) young son Lun. Each was ordered to store them away, and attach them to the collections in each 46  Du Que, “Cen Jiazhou ji xu,” qtw, 459.4692. 47  The exacting standards of which are described in Cherniack, pp. 58–61. Note that these standards still resulted in innumerable textual emendations. 48  Xu Hun, “Wusilanshi zixu,” qtw, 760.7903. 49  Xu Hun’s preface was written in 850 when he was about sixty-three.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1143

household” 因命小姪龜兒編錄，勒成兩卷。仍寫二本，一付龜兒，一授夢 得小兒崙朗，各令收藏，附兩家集.50 The significance of the editing and copying practice employed by Bai Juyi becomes clear when the full extent of his collection project is set forth. So far as we know, Xu Hun wrote out a single copy of what was a relatively short collection. Bai Juyi’s collection, on the other hand, was a conglomeration of numerous sub-collections. Multiple copies of the full collection were all carefully accounted for and distributed. In an 845 postface written about a year before his death, Bai Juyi gives a useful summary of the various copies of his collection and a sense of the meticulousness with which he managed them: I, Mr. Bai, previously composed the Changqing Collection in fifty juan. Yuan Weizhi [Zhen] wrote the preface. The Later Collection, in twenty juan, I prefaced myself. Today I again continue the Later Collection with five juan and write an account of it myself. The previous and later [parts] come to seventy-five juan, with poems and writings great and small, in all three thousand eight hundred and forty pieces. There are five copies of the collection: one is in the Sutra Cache Hall of the Donglin Temple on Mt. Lu; one is in the sutra cache at the Nanchan Temple in Suzhou; one is in the Vinaya Storehouse Building in the Pota Hall at the Shengshan Temple in the Eastern Capital; one I’ve given to my nephew Gui; and one I’ve given to my grandson Tan Getong. Each of the [latter two] is stored in their homes to be passed down to posterity. As for the ones copied out by people from Japan, Silla, various other countries, and the two capitals, they are not accounted for here. There is also the Collection of Continued Matching Poems of Yuan and Bai, seventeen juan in all; the Matching Poems of Liu and Bai in five juan; and the Sightseeing and Banqueting Downstream from Luoyang Collection in ten juan. These writings are all copied out from the larger collection and circulate separately today. Those not in the primary collection and circulating falsely under my name are all spurious. 白氏前著長慶集五十卷，元微之爲序。後集二十卷，自爲序。今又續後 集五卷，自爲記。前後七十五卷，詩筆大小凡三千八百四十首。集有 五本: 一本在廬山東林寺經藏院，一本在蘇州南禪寺經藏內，一本在東都聖善 寺缽塔院律庫樓，一本付姪龜郎，一本付外孫談閣童，各藏於家，傳於 後。其日本、新羅諸國及兩京人家傳寫者，不在此記。又有元白唱和因繼集

50  “Liu Bai Changhe ji jie” 劉白唱和集解, bjyjjj, 69.3711.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1144

Nugent 共十七卷，劉白唱和集五卷，洛下遊賞宴集十卷，其文盡在大集內錄出，別 行於時。若集內無而假名流傳者，皆謬爲耳。51

The comments on the various sub-collections are of particular interest, with Bai Juyi’s statement that they were “copied out” from the larger collection standing as a unique piece of information about Tang copying practices. First, smaller occasional collections must have been put together and re-copied to form one larger collection. But subsequent circulating versions of those occasional collections would be copied out of the larger and more general grouping. This indicates that Bai Juyi saw his primary literary collection as functioning as a sort of standard, a formal version of the various writings from his life. Moreover, as we know that his nephew had only made two copies of at least one of the constituent parts of this collection, the three or more copies of the complete collection each represented new productions. Not even taking into account the different versions copied and circulating in Japan and Korea or considered spurious, it is clear that a great number of texts were copied and recopied many times. And if Bai Juyi’s nephew continued to serve as the primary copyist, he was a busy man. According to the very rough estimate proposed by Stephen Owen, producing a single copy of Bai’s collection in its later stages would have required over thirty-two days of steady labor based on the relatively fast rate of twelve characters per minute, eight hours a day.52 The picture that emerges from Bai Juyi’s account is one of multiple reproductions, and thus opportunities for emendations and errors, before the completed collection was made available to the public. Bai Juyi’s writings suggest the maximum possible size of a collection and the extent of the editing and copying process. They do not, however, provide much detail on how copying might take place, especially if more than one person was involved. Wei Ai’s 韋藹 903 preface to the collection of his brother Wei Zhuang 韋莊 (c. 836–910) fills in some of these gaps. After noting that written versions of Wei Zhuang’s earlier poems had been destroyed and that the poems were preserved only in oral form (they had been set to memory and could be recited), Wei Ai goes on to describe the creation of his brother’s Washing Flowers Collection: “Taking advantage of free days, I would record things from among my brother’s drafts. From time to time I would also silently record from his chantings. I arranged [everything] into [? three characters missing] and titled it the ‘Washing Flowers Collection,’ sharing the intent of the one who lived 51  Bai Juyi, “Baishi Changqing ji houxu,” bjyjjj, “waiji” 3.3916–7. 52  Stephen Owen, “Butterflies Dangling in a Spider’s Web: The Literary World of the 830’s,” unpublished paper delivered at Princeton University, April 20, 2002, pp. 6–7.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1145

in Duling (i.e. Du Fu)” 藹便因閒日，錄兄之稿草中，或默記於吟詠者，次爲 口口口,目之曰浣花集，亦杜陵所居之義也.53 Wei Ai was thus working with both written and oral sources. Earlier in his preface he refers to some poems that survived only because they could be recited—the kousongzhe (口誦者); this is what he must be referring to here. As it was assumedly Wei Zhuang himself who remembered his own poems, and not other people (as was the case in the compilation of Guanxiu’s collection), Wei Ai could at least show his transcriptions to his brother and check whether corrections were needed. Still, the editing and copying process is again shown to be an informal one performed not by a professional scribe but by a family member. We can assume that Wei Ai and Bai Juyi’s nephew, if not Li Yangbing and Wang Jin, kept their intentional editorial emendations to a minimum. When a poet set out himself to edit his works, however, the changes could be more substantial. Guanxiu’s preface to his “Twenty-four Poems on Dwelling in the Mountains” is not concerned with a full literary collection per se, yet it bears examining as a remarkable statement of the flexibility of Tang editorial practices when a poet was dealing with his own works: In the fourth or fifth year of the Xiantong reign period, I composed twentyfour “Dwelling in the Mountains Poems” while in Zhongling. When I put down my brush, the drafts were snatched away by someone. Later on, some of them were spread around, written on the walls of dwellings and some were recited by people. Sometimes I would hear one or two of them and they all had mistakes in words and lines. In the xinchou year of the Qianfu reign period,54 while taking refuge at a mountain temple, I unexpectedly acquired a copy of [the poems]. Their style was rustic and common, their manner low and muddled. How could I let them be heard by sophisticates? One day I took out a brush and reworked them. Some I kept, some I cut out. I corrected some and added to some. They actually came to twenty-four pieces and were indeed quite brilliant. 愚咸通四五年中，於鍾陵作山居詩二十四章，放筆，被人將去。厥後或有散 書于屋壁，或吟詠於人口，一首兩首，時時聞之，皆多字句舛錯。洎乾符辛 丑歲，避寇於山寺，偶全獲其本，風調野俗，格力低濁，豈可聞於大雅君

53  Wei Ai 韋藹, “Wanhua ji xu” 浣花集序 in Nie Anfu 聶安福, ed., Wei Zhuang ji jianzhu 韋 莊集箋注 (Shanghai: Shanghai guji chubanshe, 2002) “fulu” 4.483–4. 54  I.e. 881, which is actually the first full year of the Zhonghe reign period.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1146

Nugent 子。一日抽毫改之，或留之、除之、修之、補之，卻成二十四首，亦斐然 也。55

Guanxiu does not describe what specific errors he found in the poems when they would come back to him over the years, but his reactions to their different versions contain important implications. Even without the original drafts, he must have believed that he remembered the poems well enough to recognize mistakes when he heard and saw them. More telling, however, is what Guanxiu does with the poems when he obtains a full copy of them after so many years. He does not merely restore them to what he recalls having originally written, but rather uses the opportunity to undertake a full-scale revision of the set. Portions and even whole poems are taken out, others are added and rewritten. His explicit goal was to improve them, feeling their style no longer appropriate for a sophisticated audience, and he appears to have been happy with the result. It remains unclear whether Guanxiu’s displeasure with the poems as he received them years after their original composition was due to changes that had taken place in the transmission process or came from the fact that his own literary standards had changed. If the latter was the case, then his willingness to rewrite them so thoroughly is all the more revealing. These were not works that he had kept stored in a trunk since his youth and decided to revise in his old age. They were poems that had already circulated widely in a number of media. Thus, Guanxiu’s rewording is not part of the more prolonged compositional process that became accepted in the Mid- and Late Tang, but is an unabashed altering of works that had long been publicly circulated, albeit without his blessing. Moreover, while Guanxiu does explain the revisions, he maintains at least some conceit that these are the same poems. The title remains unchanged, as does the number in the set, indicating that he was comfortable presenting these new versions as something other than completely new works. Indeed, it appears that revision, even if not as extreme as that practiced by Guanxiu, was the rule. And once again, Bai Juyi provides a confirming exception. In the preface to the collection of matching poems with Liu Yuxi cited earlier, Bai goes on to promise that “The words are all here and I have not corrected the writing” 語盡於此，亦不修書.56 The reasons for such a claim 55   Quan Tang Shi 全唐詩 (hereafter qts) (Beijing: Zhonghua shuju, 1960), 837.9428. I was alerted to this very important passage by Xiaofei Tian’s talk, “Possession and Loss: Tao Yuanming, Su Shi, and Acquiring A Mountain,” delivered at Harvard University on October 21, 2002. 56  Bai Juyi, “Yinji ji chongxu” 因繼集重序, bjyjjj, 69.3709.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1147

can only be guessed at: perhaps Bai Juyi meant to promise the reader a direct transcription of his spontaneous genius. In any case, his specific mention of a lack of correction clearly implies that common practice was otherwise. The picture that ultimately emerges from these accounts of the editing and copying process is one of substantial flexibility. Tang individual literary collections typically represented not all of an author’s works but a selection based on any number of criteria. Moreover, rather than being a record of a writer’s works as they were originally composed and even circulated, collections consisted of edited and corrected versions of those works. Earlier I noted the distance between modern readers and Tang poetry as Tang poets produced it; an important stage in the development of this distance was the significant changes that poems often underwent between their original composition and their enshrinement in collections, sometimes even while the author was still alive. As presentations of a literary history, Tang literary collections were both selective and mediated by the changing tastes of authors and the judgments of their compilers, editors, and copyists.

Roles of Collections in Tang Literary Culture

Whether compiled in response to imperial command or simply as an attempt to gather together the works of an author one admired, literary collections played two essential and interrelated roles in the literary culture of the Tang: preservation and circulation. Poems led precarious lives in this period; they existed not in multiple copies of printed books but rather as individually handcopied scrolls and sheets of paper. They might be written on walls at temples and post-stations or passed from mouth to mouth by singing girls in the entertainment quarters.57 All of these media were susceptible to the unruly forces of entropy. Sheets of paper were scattered and lost, walls were whitewashed, 57  For a brief account of the circulation of poetry in the Tang see Fan Zhilin, op cit. For a more detailed account see Christopher M. B. Nugent, “The Circulation of Poetry in Tang Dynasty China,” Ph.D. dissertation, Harvard University, 2002, pp. 27–79 and 100–133. Judith Zeitlin discusses poems written on walls during the Tang in “Disappearing Verses: Writing on Walls and Anxieties of Loss,” in Judith T. Zeitlin and Lydia H. Liu, eds., Writing and Materiality in China: Essays in Honor of Patrick Hanan (Cambridge, Mass.: Harvard University Asia Center, 2003), pp. 73–132. See also Wu Chengxue 吳承學, “Lun tibishi— jianji xiangguan de shige zhizuo yu chuanbo xingshi” 論題壁詩——兼即相關的詩歌 製作與傳播形式, Wenxue yichan 1994.4: 13. For a detailed examination of one aspect of orality in Tang poetic culture, see Ren Bantang 任半塘, Tang shengshi 唐聲詩 (Shanghai: Shanghai guji chubanshe, 1982).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1148

Nugent

and poems heard in a tavern were often quickly forgotten. The promise of the Shijing 詩桱 that song would make language last (歌永言) offered little solace in the face of the reality of constant loss so often lamented by compilers. The effort required for textual reproduction further made the circulation and dissemination of literary texts a slower and more laborious process than it would be in later periods. Prefaces to literary collections accordingly show an acute concern with preservation and circulation and a conviction that literary collections themselves are crucial to ensure that both occur. Central to the issue of loss is the physicality of poems in the Tang. When authors and compilers discuss literary works they often speak of them in a very material sense, not as disembodied works that would always exist somewhere (as we might think of much literature today),58 but as actual documents that could be permanently lost or destroyed. This notion of literary works as vulnerable physical objects is clear in Bai Juyi’s poem of 835, “Written on My Collection Cabinet” 題文集權: I cut cypress to make a cabinet for writings, The cabinet was sturdy and the cypress was strong. And whose collection is stored there? The inscription says “Bai Letian.” My life’s endeavor has been writing, Since I was young until my elder years. From earlier and later there are seventy juan, More or less three thousand pieces. I truly understand that in the end they’ll be scattered and lost, But cannot bear for them to be hastily tossed aside. I open [the cabinet] and lock it shut myself, Placing it in front of my study curtain. … 破柏作書櫃，櫃牢柏復堅。收貯誰家集，題云白樂天。 我生業文字，自幼及老年。前後七十卷，小大三千篇。 誠知終散失，未忍遽棄捐。自開自鎖閉，置在書帷前。59

…

58  For example, while I might worry about my copy of Hamlet being lost or destroyed for financial or sentimental reasons, I can be confident that there is little risk of all copies of Hamlet being destroyed. Hamlet exists in my consciousness as a work independent of its individual physical incarnations. 59  Bai Juyi, “Ti wenji gui” 題文集櫃, bjyjjj, 30.2072.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1149

Following the convention of prefaces, Bai Juyi quantifies his life’s literary output in terms of pieces and scrolls. He focuses not on worries that his works will be forgotten, but that they will be scattered and lost. His concern is very much with their physical existence as written documents. While conceding that the process of entropy cannot be forestalled forever, Bai is yet determined to resist it by keeping this copy of his works close by under lock and key. His aim is preservation. The poet here shows little concern with other people actually reading his poems. Bai Juyi knew that his works circulated widely in the world. Yuan Zhen reported finding them for sale in markets, written on pillars, and recited by school children, going as far as to claim that, “Ever since there has been poetry, there has never been such wide dissemination” 自篇章已來未有如是流 傳之廣者.60 Contemporaneous writings suggest that this may indeed not have been an empty boast.61 It is perhaps because of this that Bai’s concern with his collection throws into such sharp relief the contrast between a writer’s works as an abstract oeuvre and his collection as a material manifestation of that abstraction. Confident that he was being read, Bai Juyi’s energies centered on keeping specific copies of his works physically secure. Nowhere is this concern more obvious than in the descriptions he wrote of copies of his collection deposited at various monasteries. His “Account of the Donglin Temple Literary Collection of Mr. Bai” 東林寺白氏文集記, written in the same year as the poem on his collection cabinet, ends with this unusual request: From time to time the elders would ask for my literary collection so that it could also be placed in the sutra cache. But I would merely agree tacitly that I would do this some other day. Up to now, it has been over twenty years [since then]. Today, my earlier and later writings come, great and small, to two thousand nine hundred and sixty-four pieces, arranged into sixty juan. With the ordering complete, I am depositing it in the cache … 60  Yuan Zhen, “Baishi Changqing ji xu” 白氏長慶集序, bjyjjj, “fulu” 2.3973. 61  In a funerary inscription for Li Kan 李戡, Du Mu notes that Li had complained of Bai Juyi and Yuan Zhen’s poems that: “Of those who are not proper scholars and refined men, most have been ruined by them. They have spread among the common people and are written on screens and walls. Sons and fathers, daughters and mothers pass them along to each other by mouth. Their lewd words and indecent talk are like winter’s cold and summer’s heat. They enter into men’s flesh and bone and cannot be expelled” 非莊士 雅人, 多爲其所破壞。 流於民間，疏于屛’壁。子父女母， 交口教授。 淫言媟 語， 冬寒夏熟。 入人肌骨不可除去. See Du Mu, “Tang gu Pinglu jun jiedu xunguan Longxi Li fujun muzhiming” 唐故平盧軍節度巡官隴西李府君墓誌銘,” in Fanchuan wenji 樊川文集(Shanghai: Shanghai guji chubanshe, 1978), 8.136.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1150

Nugent

I further request that the elders of this temple and the monk in charge of the cache follow the example of the literary collection of Master Yuan62 and not lend it out to outsiders or let it pass the temple gates. That would be fortunate indeed! 時諸長老請佘文集，亦置經藏。唯然心許他日置之，迨茲餘二十年矣。今 佘前後所著文大小合二千九百六十四首，勒成六十卷。編次既畢，納于藏 中。。。仍請本寺長老及主藏僧依遠公文集例，不借外客，不出寺門，幸 甚。63

Bai Juyi wanted his collection protected, not passed around. He expected that interested parties would themselves take the initiative to come and see the collection if they wished to read or copy it. This same request was made the following year when he deposited an additional copy of the manuscript at the Shengshan Temple. In his “Account of the Shengshan Temple Literary Collection of Mr. Bai” 聖善寺白氏文集記, he requests that his collection “… not leave the hall and not be lent out to official guests. If there is someone who is interested they can just look at it [here]” 仍請不出院門，不借 官客，有好事者，任就觀之.64 We may take Bai Juyi at his word that he was, to some extent, simply responding to repeated requests from the temple elders to donate a copy of his collection to them. He was the most famous poet of the day, and such a gift would surely have been much prized. Moreover, Bai Juyi associated depositing copies of his collection at temples with the accumulation of Buddhist merit.65 At the same time, there is no doubt that he was well aware of the potential benefits, in terms of preservation of the collection, of keeping it stored in the sutra cache of a prominent temple. Such a method of storage certainly did not preclude circulation, since temples were popular destinations in the Tang and people were apparently welcome to read and copy as much of Bai’s collection as they saw fit when visiting temples that held copies. Yet Bai Juyi’s primary concern remained with keeping the deposited copies safe. 62  The monastery’s founder, Huiyuan 慧遠. 63  Bai Juyi, “Donglinsi Baishi wenji ji,” bjyjjj, 70.3768–9. 64  Bai Juyi, “Shengshansi Baishi wenji ji,” bjyjjj, 70.3770. 65  In two prefaces to copies of his collection that he deposited in temples, Bai Juyi explicitly claims that he is doing so to make up for his past misdeeds and establish the basis for his future salvation. This motivation, however, appears to apply only to depositing the copies at the respective temples, not to the actual compilation of the collection. See “Shengshansi Baishi wenji ji”, discussed immediately above, and “Suzhou Nanchanyuan Baishi wenji ji” 蘇州南禪院白氏文集記, bjyjjj, 71.3788.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1151

If Bai’s obsessive focus on preserving his collection as the ultimate representation of his life’s work was unusual, this was as a matter of degree, not of kind. Numerous other writers express a faith that compiling works into a formal collection with a preface will by itself serve to preserve them. In his preface to the works of Li Guan 李觀 (766–794), Lu Xisheng 陸系聲 writes, “Ever since the devastation of the Guangming rebellion, the world’s literary collections have mostly disappeared. I obtained Yuanbin’s66 writings by the River Han. Fearing that they would still be wiped out, I arranged them into three sections and discussed their meaning at the beginning” 自廣明喪喬亂，天下文 集略盡。予得 元 賓 文 於 漢 上 ， 惜 其 恐 復 磨 滅 ， 因 條 次 爲 三 編 ， 論 其 意 以 冠 於 首 .67 Lu Xisheng explicitly claims a connection between arranging and prefacing this set of Li Guan’s works and keeping them from harm: the act of collecting was by itself an act of preservation. Scattered poems were gathered together and recopied; and if some were eliminated in the process, those that remained were given a new life. Collections also preserved by contextualizing works. As part of a collection a poem gained the company of other works that might implicitly help to situate them both as points in the biography of the author and as aspects of his literary style. An isolated poem tells the reader only about itself; as part of a collection it is a piece in a larger construction. Prefaces contribute to this contextualization by giving biographical information about the writer; putting his literary output in historical, contemporary, and aesthetic contexts; and making arguments about what is valuable and interesting about the author’s works. Pauline Yu has written that prefaces trace “forms to their sources, works to their prototypes, and authors to their predecessors” and that they “verify the venerable genealogy of the genres included as well as the concrete historicity of the works themselves.”68 These two types of context, in turn, would preserve the poems when the dissemination process began anew. In essence, a collection tells the reader that these are no longer isolated poems discovered at random or uttered in a transitory moment, but that they are rather part of a body of work with specific and knowable historical and social connections. We see this context at work when Lu Xisheng makes a connection between “discussing their meaning” and keeping the poems from perishing. His preface gives the works additional weight. It implies that they were worthy of collection 66  Li Guan’s style name (zi 字). 67  Lu Xisheng, “Tang taizi jiaoshu Li Guan wenji xu” 唐太子校書李觀文集序, qtw, 813.8550. 68  Pauline Yu, “Poems in Their Place: Collections and Canons in Early Chinese Literature,” Harvard Journal of Asiatic Studies 50.1 (1990): 170.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1152

Nugent

and critical attention. In a second preface to the collection of Meng Haoran’s poems compiled by Wang Shiyuan that I will discuss in more detail below, a certain Wei Tao 韋縚 writes, “If these poems had not met Master Wang, then they would have been nothing but a few dozen sheets of old paper” 此詩若 不遇王君， 乃十數張故紙耳.69 Here and in many other cases we have seen, the preface is also a record of the tenuous past of the works that went into the collection. As an account intended thereafter to accompany the collection, it serves as a warning to future readers, informs them that what they are reading came perilously close to disappearing and that they should thus treat it with care. This context both explains the survival of the collection and implicitly entreats the reader to continue the ongoing process of preservation. Lu Xisheng’s portrayal of himself as saving Li Guan’s works from the brink of the oblivion into which many other works had fallen implies that Li Guan’s writings were not in wide circulation. Yet even works that had enjoyed substantial dissemination were seen as needing to be compiled into a collection and prefaced to preserve them. In his preface to the collected poems of the monk Fanggan 方干, written near the end of the dynasty, Wang Zan 王贊 writes: When I was young I obtained many dozen pieces of the master’s poems and was especially fond of them. At that time the master was still alive, but he lived far away and we were never able to meet. Later his reputation was increasingly given tribute and among those who composed poems there were many able to chant [his works]. But then the master died. This year I ran into Sun Yan of Yue’an in Jing. He had been close to the master from early on. He took out and showed me the “Biography of Master Xuanying [i.e. Fanggan]” that he had composed. He then said, “I went with [Fanggan’s] nephew, Yang He, to see the [master’s] disciple, the monk Juyuan, and gathered up his remaining poems. We obtained over three hundred and seventy pieces and separated them into ten juan.” He wanted me to write a preface for [the collection], hoping that together they would not perish. 予爲兒時，得生詩數十篇。心獨好之，生時尙存，地遠莫可相見，其後生名 愈藉，爲詩者多能諷之，而生歿矣。今年遇樂安孫郃於荊，早與生善，出示 所作玄英先生傳，且曰：「與其甥楊弇洎門僧居遠收掇其遺詩，得三百七十 餘篇，析爲十卷。」欲予爲之序，冀偕之不朽。70

69  Wei Tao, “Meng Haoran ji xu” 孟浩然集序, qtw, 307.3124. 70  Wang Zan, “Xuanying xiansheng shiji xu” 玄英先生詩集序, qtw, 865.9069–70.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1153

Unlike Lu Xisheng or Xi Yunqing, Wang Zan does not claim that the works in question were in any proximate danger and does not imply that it was he alone who saved them. At the same time, he directly ties his preface to the continued survival of the works in Fanggan’s collection, relating Sun Yan’s hope that by being together the collection and the preface would both “not perish” (buxiu 不朽). The collection might not have come into being were it not for the compilation efforts of Sun Yan and Yang He, but it would not live on and serve its intended purpose without the addition of his preface. Again we see a distinction between Fanggan’s literary legacy as an abstraction and as a concrete set of texts. The former appeared to be safe, as both his fame and his poems had spread widely, to the extent that many people could recite his works from memory. Yet Sun Yan, Yang He, Juyuan, and Wang Zan all understood that with the poet’s death, fame could quickly fade and poems remembered could be forgotten. By collecting the poems together and giving them a literary, biographical, and historical context, they could both preserve a physical set of texts and the larger literary oeuvre that it represented. Important as preservation was, it was typically not an end in and of itself. Works were preserved so that they could ultimately be disseminated both in the present (often described in prefaces as to “circulate in this time” 行於代) and into the future. In some contexts these two goals could come into conflict. Many of the precarious conditions in which poems could be found in the Tang were actually conducive to circulation. Poems written on post-station walls might not survive long in such a state, but they had a far greater chance of catching the attention of travelers than did poems stored away securely in a cabinet of strong cedar. Likewise, singing girls would surely move on to new songs as old ones went out of style, yet they also served as an important means of spreading those works that were in style for a time, no matter how briefly. In contrast, though the copy of Wang Wei’s collection locked up in the Imperial Library may have been physically secure, it was unlikely to gain a broad audience in such conditions.71 Most collections in the Tang were not, of course, compiled for storage in the Imperial Library, and prefaces clearly state that one of the primary goals of compilation was to provide a broad audience with access to the works. Xi Yanqing notes that the works of Luo Binwang that he found stored in private 71  As Fan Zhilin points out, “The primary responsibility of the Imperial Library was storing books, it was certainly not open to the outside. Only the emperor, high court officials, and related functionaries would have opportunities to read there.” Fan Zhilin, part i, p. 283. It is unclear whether these people would be allowed to make copies for personal use or for dissemination.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1154

Nugent

homes were “worth circulating among those who enjoy [such things].” That is, they were worth bringing back into the public realm. Tan Yu states that he put together a new collection of Guanxiu’s works specifically in response to popular demand. People wanted access to his master’s poetry, and a collection would bring a large number of such works together for people to enjoy and copy. The fact that Tan Yu then had printing blocks carved of the collection further supports the idea that the ultimate purpose of his labors was to circulate the works more widely. In a similar vein, Yan Zhenqing 顏真卿 (708/9–784/5) makes the connection between compiling a collection and circulating works explicit in his preface to the collected works of the High Tang writer and official Sun Ti 孫逖, writing, “I thereupon compiled and edited his literary collection into twenty juan, as shown below. I hope that people who enjoy [his works] will copy them out and chant them in order to pass them down to eternity. Indeed, what need is there for storing them on famous mountains and placing them in chambers of stone?” 乃編次公文集爲二十卷，列之於左。庶 乎好事者傳寫諷誦，以垂乎無窮，亦何必藏名山而納石室.72 Dissemination, in this case, is preservation. Protective storage is deemed unnecessary when one’s works reach the level of popularity Yan Zhenqing foresees for those of Sun Ti.73 The compilation of collections remains a key part of the overall circulation process as it gives people access to a substantial number of works at one time and location. Circulation in the present era seems to be the prime concern for Xi Yunqing and Tan Yu. Like Yan Zhenqing, many other compilers are more interested in passing works on to later generations and audiences. Wang Shiyuan laments that Meng Haoran’s deeds will go unrecorded by official historians as Meng never held a position in the bureaucracy, and cannot bear the thought that his “excellent rhymes will be from this point severed” 妙韻從此而絕. He writes, “Having no other important affairs, I am passing [Meng Haoran’s works] down for posterity so that hereafter all those within the seas wearing the robe and cap of officialdom and with the official’s tablet stuck in their belts will remember him when passing through Xiangyang, and look over his writings” 既無他 事，爲之傳次，遂使海內衣冠縉紳，經襄陽思睹其文.74 His earlier description of Meng Haoran’s carelessness toward his own literary legacy, and the 72  Yan Zhenqing, “Shangshu xingbu shilang zengshangshu youpuye Sun Ti wengong ji xu,” 尙書刑部侍郎贈尙書右僕射孫逖文公集序, qtw 337.3416. 73  Yan Zhenqing was proved correct in some sense. Sun Ti’s collection does not survive though many of his individual works do. 74  Wang Shiyuan, “Meng Haoran ji xu,” Meng Haoran shiji zhu, “fulu” 433; cf. Kroll, “Wang Shih-yuan’s Preface,” p. 356.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1155

lengths to which he had to go to save what was left of the poet’s output, give the reader a distinct sense of the tenuousness of poems left on their own. Yet clearly these poems rescued from falling into, at best, obscurity, and at worst, oblivion, were not meant to be hidden away for safe keeping. The larger intent of dissemination and circulation into the future underlies the act of preservation. It was not enough to know that the poems continued to exist. To remember Meng Haoran and read his poems, later generations had actually to have access to them as well. Dissemination of the poems in the form of a collection increased the chances that this would happen. Writers compiling their own works were also cognizant of the important role collections would play in securing them an audience beyond their own lifetimes. Thanks to his poetry’s widely attested contemporary popularity, Bai Juyi may have been unconcerned about increasing the circulation of his poems in his day. He was, however, well aware of the importance of future readers for his works, especially those pieces that did not enjoy much acclaim at present. In his letter to Yuan Zhen accompanying his very first collection, he writes: Coming to [my poems] of allegorical remonstrance, their ideas are stimulating and their speech unadorned. As for the [poems of] leisure and pleasure, their thoughts are calm and their words meandering. Because they are unadorned and meandering, it’s only fitting that people don’t cherish them. The only one living in the world with me today who cherishes them is you. However, a hundred thousand years hence, how could we know that there won’t appear someone like you who will cherish my poems? 至於諷諭者，意激而言質，閑適者，思澹而詞迂。以質合迂，宜人之不愛 也。今所愛者，並世而生，獨足下耳。然千百年後，安知復無如足下者出而 愛我詩哉。75

Note that preservation remains an important aspect of Bai Juyi’s goal. He was including these poems in the first phase of the collection on which he would continue to lavish great care. Yet access to the works is his main concern, rather 75  “Yu Yuanjiu shu,” bjyjjj, 45.2795. Amusingly, Bai Juyi includes himself as a possible future audience for his own writings. In his “Account of Mr. Bai’s Fragrant Mountain Temple Collection of [Works from] Luoyang,” he imagines himself coming back in a later life to read his own collection: “How can I know that in another life I won’t again travel to this temple, and gaze once more on these writings?” 安知我他生不復游是寺，復睹斯文. See “Xiangshansi Baishi Luozhong ji ji” 香山寺白氏洛中集記, bjyjjj, 71.3806.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1156

Nugent

than an afterthought. He wants his poems to get to others: it is for that reason that they must be preserved. Bai Juyi is also not explicitly focused on the material nature of his literary products in this passage. These poems are characterized by their content, not by where they are to be kept or the rules governing access to them. The notion of works of literature as material objects did, however, play a part in issues of circulation to a more specific future audience, namely one’s descendants. In some cases writers thought of the works they would be passing down in an abstract sense. Sikong Tu 司空圖 (837–908), for example, states that an important purpose of his collection is “no more than perhaps to startle awake my sons and grandsons” 庶警子孫耳.76 However, there are many cases in which the works to be passed down to posterity are described as a very material inheritance. As noted previously, Sun Qiao compiled his collection of pieces “worth looking at” with the expressed intent of passing it on to his sons and grandsons. He did not merely want them to know of his literary works, but to possess physically copies of them, stating that he “stored them in a bamboo chest” for this purpose. Actual copies of works in collections thus represented a bond between a writer and his descendants or other family members, a bond that could be valuable not only emotionally but socially as well. In his “Explanation of Mr. Liu’s Summary Collection,” Liu Yuxi describes a request from his son-in-law: The other day my son-in-law Master Cui of Boling reported to me: “When I was heading to the capital, eminent men often asked how many new writings of my father-in-law I had and wanted me to go get some. But I answered that I had none and shame instantly rose up in my face. Today I am again going west and hope to have something to end this shame.” Thus I picked out one of every four pieces to make my “collection summary” and give it to this lad. I don’t dare circulate it very far. 他日，子婿博陵崔生關言曰： 「某也曏游京師，偉人多問丈人新書幾何，且 欲取去。而某應曰無有，輒媿起於顏間。今當復西，期有以弭媿者。」由是 刪取四之一，爲集略，以貽此郎，非敢行乎遠也。77 76  Sikong Tu, “Zhongtiao wangguanggu xu” 中條王官谷序, qtw, 807.8488–9. Note that this text is elsewhere referred to as the “zixu” 自序 to the Yiming ji 一鳴集. See Wan Man, Tangji xulu, p. 335. There are also some small textual differences in the version cited by Wan Man. 77  Liu Yuxi, “Liu shijilüe shuo 劉詩集略說,” in Bian Xiaoxuan 卞孝萱, ed., Liu Yuxi ji 劉禹 錫集 (Beijing: Zhonghua shuju, 2000), 20.250–1.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1157

Liu Yuxi displays a reticence about wider circulation (though perhaps only as a move of self-deprecation) matched with a desire to pass on his works to close relations. The physicality of pieces is what is important here. Liu Yuxi’s sonin-law is expected to have actual texts of his writings, as opposed to merely a knowledge about them or the ability to recite them. Liu Yuxi, in turn, compiles this sub-collection to grant him that possession. The works are not passed on as abstract vehicles of moral suasion or literary models so much as physical proof of a connection with the author. Being Liu Yuxi’s son-in-law might have given the young man a right to claim a familial connection, but he apparently required actual copies of the writer’s recent works to be able to claim a literary inheritance as well.78 Some of the strongest evidence that authors envisioned collections as an important means by which to pass on copies of their works to descendants is found in examples of writers who worried that they would be unable to do so in the manner they had hoped. Yuan Zhen’s poem, “Sighing to Myself, I Send This to Letian” 自歎因寄樂天, laments that since “Heaven has left our families without male heirs, / to whom can we give our literary collections?” 天遺兩家 無嗣子，欲將文集與它誰.79 For Bai Juyi the answer to this question was his nephew and grandson. In “Written on My Collection Cabinet,” he expresses his worries about his collection eventually being scattered and lost, and concludes his poem with the words, “My only response is to give them to my daughter, / leave them to be passed on to my grandsons” 只應分付女，留與外孫傳.80 And as we have seen, it was with his grandson and nephew that he ultimately deposited two of the five copies of the final version of his collection, “to be passed down to posterity.” Indeed, this notion of a literary collection as inheritance is strong enough that there are examples of third parties going to great lengths to ensure that collections get passed on to rightful heirs. During the Taihe reign period (827– 837) Li Yisun 李貽孫 was serving as an official in Fujian when the son of the late Ouyang Zhan 歐陽詹 (c. 756–c. 798) brought a set of the latter’s works from Nan’an 南安 and “tearfully requested a preface” 泣拜請序. Li Yisun agreed but had not yet completed the piece when the son died. About two decades later, in the sixth year of the Dazhong reign (852), Li writes that he “was again made Surveillance Commissioner and ordered a search for [Ouyang’s] descendants. I thereupon found a grandson called Xie. I could not allow the transmission 78  It is also possible that the “eminent men” wanted to copy Liu Yuxi’s poems themselves. In this case, physical possession would be even more important. 79  Yuan Zhen, “Zitan yin ji Letian,” yzj, 22.247. 80  Bai Juyi, “Ti wenjigui,” bjyjjj, 30.2072.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1158

Nugent

of Mr. Ouyang’s writings to be thus severed. In composing this preface, I’ve also completed the wishes of [Ouyang’s] descendants” 又爲觀察使。令訪其 裔，因獲其孫曰澥。不可使歐陽氏之文遂絕其所傳也。爲題其序亦以卒後 嗣之願云.81 The transmission, chuan 傳, to which Li Yisun refers is clearly nei-

ther to the world at large nor simply to later times; it is a transmission of this specific set of works to Ouyang Zhan’s descendants. We again see a distinction between literary works in the abstract and a given collection of those works. Even if Li had spread Ouyang Zhan’s poems far and wide, it would not have resolved the issue of returning the set of works that he had been given to its rightful owners. Li Yisun himself does not, of course, use the word “owners,” yet the term is appropriate when discussing literary collections of the Mid-Tang and later. Bai Juyi and others thought of collections as property; they were tangible objects that could be passed down to selected recipients, just like land or other possessions. Stephen Owen has noted that questions of ownership in earlier periods are unusual and has discussed their increased prominence in the Mid-Tang.82 It is important to be clear about what sort of ownership is implied here. It is not, I think, akin to modern ideas of intellectual property, and the reasons why it is not so get to the heart of the question of works of literature as material objects that I have been discussing. Bai Juyi, for example, exerted more control over his literary legacy than any other Tang figure, going as far as to declare versions of his works that were not copied out from his “official” collections to be spurious. Thus he claimed complete authority over his literary oeuvre. Authority, however, is different from ownership. Bai did not assert that he should control the circulation of his works or profit from them in any way. Those works, as linguistic patterns, were not his possessions. As their creator he could determine what they were, but once he let them out into the world, people could circulate them as they saw fit. His collections as physical objects, however, were a different story. They were his to keep or distribute as he pleased. Moreover, he felt it within his rights to determine how they would be treated even after leaving his possession. Bai Juyi did not own his poems the way that a modern writer does. He did, however, own particular copies of them and exerted substantial effort to keep these safe. Collections were thus seen by their compilers as physical repositories of literary works that could keep them safe and thus serve as a means of circulating 81  Li Yisun 李貽孫, “Gu simen zhujiao Ouyang Zhan wenji xu” 故四門助教歐陽詹文集 序, qtw, 544.5514. 82  Stephen Owen, The End of the Chinese ‘Middle Ages’: Essays in Mid-Tang Literary Culture (Stanford: Stanford University Press, 1996), p. 24.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1159

those works in both abstract and material senses. But did they successfully play these roles in Tang literary culture? The available evidence strongly suggests that such was the case. There is no question that people in the Tang, at least by the Mid-Tang, did indeed read poetry in the form of collections. Starting some decades after the An Lushan rebellion, we begin to find a number of poem titles that refer to reading collections. The most basic are similar to the title of Meng Jiao’s 孟郊 (751–814) poem about reading the collection of his friend Zhang Bi 張碧: “Reading Zhang Bi’s Collection” 讀張碧集.83 Near the end of the dynasty, Meng Jiao’s own collection is similarly mentioned in the title of Guanxiu’s poem “Reading Meng Jiao’s Collection” 讀孟郊集.84 While Meng Jiao and Zhang Bi were contemporaries and friends, Guanxiu would have been reading Meng Jiao’s collection well after the latter’s death. An even greater temporal gap between author and reader is indicated by such works as Lu Guimeng’s “Reading the Collection of Reminder Chen” 讀陳拾遺集, referring to the Early Tang poet Chen Ziang 陳子昂 (659–700).85 Bai Juyi entitles one of his poems “I Read a Collection of Li Bai and Du Fu’s Poetry and Thereupon Wrote This at the End of the Scroll” 讀李杜詩集因題卷後.86 Judging by references in poems and poem titles, writers in the last century of the dynasty read collections with greater frequency than their predecessors. In addition to his poem on reading Meng Jiao’s collection, Guanxiu has several poems on reading other earlier poets’ collections as well, ranging from Du Fu (“Two Poems on Reading the Collection of Du of the Ministry of Works” 讀杜 工部集二首)87 to Liu Deren 劉得人 and Jia Dao 賈島 (779–843) (“Two Poems on Reading the Collections of Liu Deren and Jia Dao” 讀劉得人賈島集二首).88 Guanxiu’s fellow monk Qiji also writes of reading Jia Dao’s collection (“Reading Jia Dao’s Collection” 讀賈島集),89 as well as those of Li Bai (“Reading Li Bai’s Collection” 讀李白集)90 and Li He (“Reading a Collection of Li He’s Songs” 讀李賀歌集).91 Literary collections of contemporary writers were even used in educational contexts and were considered suitable models for young students. 83   q ts, 380.4261. 84   q ts, 829.9343. 85   q ts, 629.7219. 86   q ts, 438.4875. A more complete list of such titles, including many references to reading single scrolls not explicitly identified as collections, can be found in Xu Jun, Dunhuang shiji canjuan jikao, pp. 13–4. 87   q ts, 829.9339. 88   q ts, 829.9340. 89   q ts, 843.9525. 90   q ts, 847.9585. 91  Ibid.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1160

Nugent

In his preface to his “Poem Lamenting the jinshi Yan Zizhong” 傷進士嚴子重 詩, Pi Rixiu 皮曰休 (c. 834–c. 883) recalls, “When I was in the village school, I copied the collection of Secretary Du Mu in my books” 余爲在鄉校時，簡上 抄杜舍人牧之集.92 This marked increase in mentions of reading collections in the Mid- and especially Late Tang does strongly suggest that, as the dynasty progressed, collections were more and more often one of the formats through which readers encountered poetry. Moreover, there are indications that literary collections were an important source for the anthologies that would later be an important route for the transmission of Tang poetry into later periods.93 In the postface to his anthology composed in the 850’s, the Categorical Selection of Tang Poetry 唐詩類選, the Tang anthologist Gu Tao 顧陶 discusses poets whose works he chose not to include and gives his reasons for doing so. Following a list of over ten poets, from Linghu Chu 令狐楚 (766–837) to Liu Yuxi, Gu Tao justifies their exclusion by saying, “Their poems are still in the world, while they have just departed it, but their literary collections have yet to circulate. Even if there were a piece or some chantings that one could obtain from people, they could still not be called things that had been recorded” 詩猶在世，及稍淪謝，即文集 未行，縱有一篇一詠得於人者，亦未稱所錄.94 This passage highlights the fact that collections gave the works they contained a formal context that they might otherwise have lacked. It is not that Gu Tao had no access to poems by these authors, many of them quite prominent: he admits that obtaining their poems would not be difficult. The issue, rather, is the source; he trusts collections in a way that he does not trust other sources. Unlike poems circulating less formally, poems in a collection had, by definition, been “recorded” (lu 錄),

92   q ts, 614.7083. This is noted as well in Ji Yougong 言十有功 Tangshi jishi 唐詩糸己事 (Beijing: Zhonghua shuju, 1965), 66.994. 93  The question of the fate of collections in later dynasties, while important, is beyond the scope of my discussion here. For a useful examination of the Tang literary collections that survive today (many of which were compiled well after the Tang), see Wan Man, Tangji xulu. Xu Jun also notes that the Ming scholar Hu Zhenheng 胡震亨 made a list of individual literary collections listed in such works as the official Tang and Song histories and the Zizhi tongjian 資治通鑑. But as he points out, “The primary situation reflected by the bibliographies in [these works] was that of the writings stored by the government. Based only on these bibliographies and judgments about a number of books, it would be very difficult to gain a true understanding of the state of circulation of poetry collections among typical members of the scholar-official class of the time.” See Xu Jun, pp. 10–1 for further discussion. 94  Gu Tao, “Tangshi leixuan houxu” 唐詩類選後序, qtw, 765.7960.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1161

and this gave them a historical, biographical, and literary context that Gu Tao felt was necessary. Gu’s postface also implies that the term “ji” covered a broad and diverse range of textual sets in the Tang that included not only the single individual literary collections most often composed near the end of a writer’s life, or after his death, but significantly smaller sets of texts as well. Discussing Du Mu, Xu Hun, and others whose works he included in his anthology, he explains: All have poems and lines that spread orally. As they died only two or three years ago, I have yet to obtain their formal collections. If I get some of the writings from the end of their lives, I’ll make a separate scroll and attach it after the twenty juan. I hope this is not looked upon with animosity. If I waited until seeing the completed versions, I would definitely fail to succeed in my compiling. 並有詩句播在人口，身沒才二三年，亦正集未得。絕筆之文，若有所得，別 爲卷軸，附於二十卷之外，冀無見恨。若須待見全本，則撰集必無成功。95

There is a distinction here between collections in general, such as the wenji mentioned in the previous passage, and “formal collections” (zhengji 正集). The formal collections appear to be those compiled after the author’s death that would be intended as a more inclusive—or at least broadly representative— presentation of the writer’s works, i.e. the type of collections that has been our focus in this essay. As we have seen in the examples of Bai Juyi, Liu Yuxi, and others, smaller collections were frequently compiled and circulated while the authors were still alive. Such collections would be expected to have lacunae, under the assumption that the living authors continued to write after they had been compiled. It is those “last works” composed later in the authors’ lives, their juebi zhi wen 絕筆之文, that Gu Tao would only be able to acquire when their zhengji circulated. However, he makes it clear that, while the zhengji would be expected to be more inclusive, other types of collections were still acceptable as sources for his anthology. He is not excluding poems by Du Mu on aesthetic grounds, but is simply waiting until he can acquire them as part of a formal collection. This postface thus sheds light on what constituted a “collection.” We can see that the term “ji” covered a broad range of materials including, but not limited to, formal collections.96 95  Ibid. 96  The finds at Dunhuang provide even more direct evidence in the form of documents entitled “collection” that are far shorter than what we might expect from the content

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1162

Nugent

The diversity of materials labeled as ji in the Tang should not be taken to imply that larger formal collections did not circulate, as there is ample evidence that they did. Beyond Gu Tao’s stated desire to acquire zhengji, there are other examples of people in the Tang reading full collections. Recall that in 836 Bai Juyi wrote of depositing a copy of his collection as it then stood at Shengshan Temple in Luoyang with the following limitation: “I still request it not be lent out to official guests. If there is someone who is interested he can just look at it [here].” In the postface to the final version of his collection, written in 845, he also lists this as one of his designated collections, the implication being that he has updated it in the nine years since first depositing it at the temple. In fact, one of his friends was interested enough to “just look at it [there].” Bai Juyi’s contemporary, the writer Li Shen 李紳 (772–846), entitled a poem “On Bai Letian’s Literary Collection” 題白樂天文集 and included a prefatory note stating: “Letian stored his writings at the Eastern Capital Shengshan Temple, calling them ‘Mr. Bai’s Literary Collection.’ I have composed this poem to praise them” 樂天藏書東都聖善寺，號白氏文集，紳作詩以美之.97 Bai Juyi’s collection was protected by his admonitions that it not leave the temple. Other collections traveled more widely, and their fate reveals that collections were further altered after their compilation. While not a compiler of the collection proper, Wei Tao wrote a second preface for a collection of Meng Haoran’s poems that he had acquired and that included the preface by Wang Shiyuan discussed above. Wei Tao writes: In the Tianbao period, I unexpectedly obtained a collection of Haoran’s writings, and Shiyuan had, in fact, written the preface for it. The structure of the phrasing was so outstanding that, chanting them, I forgot my weariness. The writing was not in a single hand, and the paper and ink were thin and weak … If these poems had not met Master Wang, then they would have been nothing but a few dozen sheets of old paper … I have made a fair copy and added a table of contents. Moreover, to honor the pure talent of Shiyuan, I have dared to add my own description at the beginning of the scroll. I will respectfully present this volume to the Imperial Library, in hope that it will not perish for many ages and its fragrance will be transmitted without end.

descriptions found in Tang prefaces. As Xu Jun put it, “Obviously, there is a significant difference between the Dunhuang manuscript poetry collections and the traditional notion of a ‘poetry collection.’ ” Xu Jun, p. 9. 97   q ts, 483.5495.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1163

天寶中，忽獲浩然文集，乃士源撰爲之序傳。詞理卓絕，吟諷忘疲。書寫 不一， 紙墨薄弱・・・此詩若不遇王君，乃十數張故紙耳… 佘今繕寫，增 其條目， 復貴士源之清才，敢自述於卷首。謹將此本，送上秘府，庶久而不 泯，傳芳無窮.98

Wei Tao’s preface is dated to the third day of the first month of the ninth year of Tianbao (February 13, 750), which puts it some ten years after Meng Haoran’s death and five years or less after Wang Shiyuan’s compilation of his collection of Meng’s works. Assuming that Wang had made a fair copy of the collection when he originally compiled it, it is clear that the writings Wei Tao obtained were later-generation copies. Perhaps the most interesting comment he makes is that the writing was not in a single hand. Wei’s comment raises the possibility that this collection was not an exact copy of the original collection compiled by Wang Shiyuan, but rather portions of it and other poems by Meng Haoran cobbled together with Wang Shiyuan’s preface. Without the original collection itself, Wei Tao would have had no way of knowing how faithful his newly discovered copy was to Wang’s compilation; by making a fair copy of the collection and presenting it to the Imperial Library, he gave this new version a legitimacy it might have lacked had it stayed in the condition in which he found it. The tale of the different stages Meng Haoran’s poems went through bears revisiting. The poems that managed to survive his own destructive impulses were scattered among his friends and old haunts. By bringing them together, Wang Shiyuan intended to give them a new context that would help to preserve and circulate them. In this he was clearly successful. Yet the collection obtained by Wei Tao, fading and written in different hands, was almost unquestionably not a perfect copy of the one originally created by Wang Shiyuan. Following the latter’s footsteps, Wei in turn took further steps to save the poems and ensure that they would live on. In each case the preface writer took works he saw in danger of vanishing and preserved them. And each act of preservation obscured the questionable legitimacy of its sources; Wang surely traded his rewards for some spurious poems, and Wei says nothing about verifying the provenance of the collection he obtained. We have thus moved from scattered works, some of dubious origin, to a twice prefaced collection residing in the Imperial Library. Finally, neither Wang nor Wei seem troubled by the possibility that some of these works may not have been by Meng Haoran. Their compiling and copying 98   q tw, 307.3124. For Daniel Bryant’s translation, which is different from mine in some important ways, see Bryant, “The High T’ang Poet Meng Hao-jan: Studies in Biography and Textual History,” Ph.D. dissertation, University of British Columbia, 1977, pp. 368–9.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1164

Nugent

may have obscured the origins of their sources, but neither attempts to hide this fact. Both did give accounts of these sources in their prefaces, allowing us a rare glimpse of this aspect of the compilation process. Conclusions The collection of Meng Haoran and the many collections of Bai Juyi have figured prominently in our discussion. The contrast between the two sets of works speaks to the different methods of compilation of literary collections in the Tang and of the roles these collections played. We can be confident that Bai Juyi’s collection, secure in the Shengshan Temple sutra storehouse, was in roughly the same state Bai Juyi left it when it was later read by Li Shen. The version of Meng Haoran’s collection given to the Imperial Library by Wei Tao, in contrast, may have differed significantly from Wang Shiyuan’s original composition. Meng Haoran showed little interest in the material preservation of his literary legacy, frequently destroying written copies of his poems. It was left to one of his admirers, a man who had never even met him, to wander the countryside gathering the scattered remnants of that legacy. Bai Juyi, on the other hand, demonstrated a concern with his works that bordered on obsession. He kept multiple copies in different locations, updated them regularly, and even gave strict instructions on the conditions under which others would be allowed to read them. These two cases do not represent opposing poles per se—Bai Juyi may indeed be an extreme case, but the fact that Meng Haoran even had a literary collection compiled during the Tang differentiates him from innumerable writers who, along with their works, were more quickly forgotten—yet they demonstrate the great diversity of the process that produced and preserved Tang literary collections. Those who compiled literary collections in the Tang undertook this task with the intent to preserve and circulate the works they were compiling. If they were members of the writer’s family, they probably had the future reputation of their relative in mind. For self-compiled collections, authors had a similar set of goals, from preserving their property to passing their writings down to face the judgments of later generations. Yet we cannot allow these more practical goals to obscure the fact that, in many cases, people compiled collections of a given author’s works simply because they enjoyed them and thought that others would as well. Literary collections in the Tang were not meant as academic exercises in literary history the way they would often be in the Song. Something about how even the recent literary past was approached changed in the Song, when the collection of literary works simply for pleasure, while it

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1165

certainly continued, was increasingly eclipsed by academic endeavors to quantify, judge, and compare editions in a way that was not at all prominent in the Tang. A full comparison of Song attitudes toward collection compilation with those in the Tang is beyond the purview of this paper. But it is important to discuss at least briefly some of the roles literary collections seem not to have played to a significant degree in the Tang, but that would become more pronounced later. In her informative discussion of canons and collections, Pauline Yu has written that, “In the case of the collected works of an individual author… the primary aim will be comprehensiveness, the inclusion of every work that falls under the rubric that has been established.”99 As far as the post-Tang period is concerned, Yu may indeed be correct. However, the evidence I have presented suggests that compilers of collections in the Tang often did not aim at comprehensiveness, and if they did, they saw it as only one goal among many. Tang collections were typically not all-inclusive. In many cases numerous works were deemed unworthy of inclusion in collections either by the authors themselves or by other compilers. This was an accepted part of the process of turning a set of gathered poems into a formal collection. Writers and compilers hoped that, as Pi Rixiu said of his set of self-compiled works intended to serve as an introduction to his literary abilities, “those who peruse them will have nothing to ridicule” 覽者無誚矣.100 Scholars of literary history, as many of those who put together collections of Tang literature in later eras could be considered to be, might have seen collections as sets of data whose value was dependent largely on their comprehensiveness. Tang compilers took a more selective view. Even the dedicated Wang Shiyuan claims to have placed substantial limits on the time and effort he expended searching for Meng Haoran’s scattered works.

99  Yu, “Poems in Their Place,” p. 168. 100  Pi Rixiu, “Wensou xu” 文藪序, qtw, 796.8352–3. The full passage reads, “In the bingxu year of the Xiantong reign period (866), I ‘heaved the spear’ [i.e. took the civil service exam] but failed. I retreated to my prefecture and came to my villa, where I compiled and edited my writings, intending to present them to the appropriate officials. When I opened up my chest they were all bunched up, tangled as an overgrown marsh. Thus I named the book the ‘Tangled Marsh of Writings’… The ‘Genealogical Record of Master Pi’ is written at the end and shares the intent of the Grand Historian’s self-preface. Altogether there are two hundred pieces, making ten juan. Those who peruse them will have nothing to ridicule” 咸通丙戌中，日休射策不上第。返歸州來別墅，編次其文，復將貢 於有司。登篋叢萃繁如藪澤，因名其書曰文藪焉・・・皮子世錄，著之於 後，亦太史公自序之意也。; 凡二百篇，爲十卷，覽者無誚矣. Note that Pi Rixiu’s collection cited here is a xingjuan 行卷.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1166

Nugent

What is perhaps most telling is that there do not appear to be any Tang prefaces that entreat other people or later generations to add to a collection should they encounter works legitimately ascribed to the author but not included in the collection in question. There are no requests that others “carry on the work that I have started.” This fact is particularly interesting given that “lost” works are more typically described as “scattered and lost,” rather than lost for good or destroyed. While there was always the potential that additional works would be found (as they often were), the notion that they ought then to be added to a pre-existing collection is not addressed by these prefaces. If there were, in fact, new poems in the version of Meng Haoran’s collection discovered by Wei Tao, they were not inserted at Wang Shiyuan’s request. Collections were seen as “fixed,” as self-contained and complete documents, rather than as repositories that could be amended later. Tang collectors and compilers were not literary historians who undertook these activities as part of a search for scholarly truth. They were not devoted to gathering all available versions of a poem so that they could be compared and the “true” original could be discovered. In the Song, Ouyang Xiu 歐陽修 (1007–1072) and his friends would debate over the proper character to fill a lacuna in a Du Fu poem until the correct original could be found.101 Li Qingzhao 李清照 (1084–c. 1151) and her husband would collate any new book they purchased and make corrections to it on the basis of their research.102 Yet I have not found a single preface to a Tang collection that mentions the comparison of different versions of a poem to determine which one is more “authentic” or true to the author’s original intent. Nor have I found any extant preface to an anthology compiled in the Tang that discusses such issues. Gu Tao chose not to accept certain sources, but there is no indication that he compared different versions of a given work in the compilation of his anthology. Guanxiu expresses displeasure in changes in his poems through the course of unintended circulation, but he has no master copy with which to compare them. He is happy simply to change them as he sees fit. In general, one finds little concern with authenticity expressed in the prefaces. Wang Shiyuan apparently felt no need to justify his collection method. Though the offering of money in exchange for poems would seem likely to result in a great number of false attributions, he does not display any concern 101  See Ouyang Xiu 歐陽修, Ouyang Xiu quanji 歐陽修全集 (Beijing: Zhonghua shuju, 1986), “Shihua” 1036. 102  See Li Qingzhao 李清照, “Epigraph to the Records on Metal and Stone ( Jinshi lu 金石錄)”, in Wang Zhongwen 王仲聞, ed., Li Qingzhao ji jiaozhu 李清照集校注 (Taibei: Hanjing wenhua shiwu youxian gongsi, 1983), 3.176–92.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1167

over the matter. Lu Chun expunged numerous poems in his attempt to present an image of Wang Ji that he found more appropriate. Yet he never claims that, based on his textual research, the excluded poems appeared to be spurious. Rather than using all the available sources to determine the truth about Wang Ji’s literary temperament, Lu Chun approached the sources with his opinion already established and openly altered them to conform with that opinion. It is worth noting that both Wang and Lu are quite forthright about these activities. This fact alone tells us that they found no fault with them. Finally, there is a notable lack of competitiveness in evidence in prefaces. While most collectors may have made an effort to gather together as many of a given author’s works as they could, there are no boasts about any collection being the most comprehensive or the “definitive” collection. This is true even when there were multiple early collections for the same poet, as was the case with Guanxiu, Li Bai, and others. I would speculate that these differences between the Tang and Song may well be connected to the particular context and methods in and by which these works were reproduced and circulated. As far as we can tell, it was not a commercial context in the Tang. Individual poetry collections do not appear to have been commonly sold or printed, though there are claims that poems themselves would be.103 Just as Tang collectors were not literary historians, neither were they businessmen attempting to market a product. Absent a competitive commercial market, such as the one which came to exist for materials related to examinations preparation, novels, short fiction, etc. in the Song and later periods, we may conclude that Tang compilers felt no particular need to elevate their versions of the collections they put together above others. While dissemination may have been an important goal, the evidence strongly 103  Yuan Zhen, in his “Preface to Mr. Bai’s Changqing Collection” 白氏長慶集序 claims of his and Bai Juyi’s poems that “[a]s for making fair copies and printings and putting them out for sale in markets and swapping them for ale and tea, it is like this everywhere” 至於繕寫，模勒衒賣于市井，或持之以交酒茗者。處處皆是. See Yuan Zhen, “Baishi Changqing ji xu,” bjyjjj, “fulu” 2.3973. There is much debate about whether the term mole 模勒 truly means “printing” here. Wang Shiyuan, of course, offered “rewards” for poems as well. Yet neither of these cases indicate the sale of individual collections as complete documents. Note that the authors of Gudian wenxue jieshou shi surmise that “If there were no conditions conducive to publication, it is quite difficult to imagine that there would spring up so many individual literary and anthology collections during the Tang.” See Shang Xuefeng, et al., p. 245. This conclusion, however, is not supported by any specific argumentation. There seems to have been ample reason, especially within the relatively small subset of the educated reading public, to compile collections even if they would not be printed.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1168

Nugent

suggests that it was the dissemination of a given author’s works in general that was important, not that of a specific version of those works contained in one particular collection. Printing dominated the dissemination of written works in the Song and after. It was a commercial enterprise that both responded to and created its own economic logic. The example of Guanxiu’s collection, the first for which we know printing blocks were cut, is illustrative. Tan Yu portrays himself as having compiled Guanxiu’s collection in response to demand. People were coming to the monastery to read Guanxiu’s works and to copy them out themselves. One implication of this is that some collections by this time existed in a nexus of, if not consumerism—there is no evidence that people paid to copy out Guanxiu’s poems—at least supply and demand. Printing thus appears to be a further step in satisfying the demand for texts. It would allow much more rapid and efficient circulation of Guanxiu’s works than the copying of individual poems. It might also ensure that the collection circulated as a collection. The time involved in personally copying texts for one’s own use meant that people were unlikely to copy out an entire collection. Why take the time to copy the poems one does not like? The preface itself, with its mention of copying out a few lines (參句五句), implies as much. Printing would thus mean a difference in kind, not just degree, of circulation. People would still be likely to copy down individual poems that they enjoyed, but the presence of full printed collections meant that poems would be circulating in that significantly different context as well. Because of the generally non-commercial nature of literary circulation in the Tang and the lack of readily available multiple copies of the works in question, the idea of a best or most “accurate” version of a given work in the context of contemporary poetry was not important to most Tang literati. The lack of commercial printing meant that there were fewer versions of a given work available in the Tang. One did not have multiple editions available for comparative purposes. It is doubtful, for instance, that a single person would have different versions of a single writer’s full collection in the Tang. The time and effort required to produce such copies would have been prohibitive. In a literary environment in which poems took a number of different forms and existed in a number of different media, including oral song, the idea of an “original” version was not a particularly important one. This is not to say that people did not understand that there could be different versions of a given work, quite the contrary. They did not, however, find this troubling. And by being collected, the collected versions were, at least for a time, set. There were unquestionably variations, yet one can imagine that the full extent of the variations would not become as clear a problem until the proliferation of printed materials in

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Literary Collections In Tang Dynasty China

1169

the Song.104 It was only then, when the notion of a “literary marketplace” became the literal truth and when collecting became a specifically scholarly pursuit (and even obsession), that issues of comprehensiveness and authenticity would come to the fore. Bai Juyi and Yuan Zhen once again offer intriguing counterexamples that, in this case, foreshadow the changes to come in the Song and later periods. In many ways, Bai Juyi’s attitude toward his collection was more akin to the scholarly approaches taken toward Tang literature by Song literati than the comparatively casual, if diligent, attitudes exemplified by such Tang collection compilers as Wang Shiyuan. What was particularly unusual about Bai Juyi was that his object of study was his own literary output; he was a literary historian of himself. The careful counting of the numbers of pieces and the elaborate classificatory systems he used are simply not found in other Tang writings, with the possible exception of Yuan Zhen. Both poets also demonstrated an interest in the question of the authenticity of literary works that was unusual for their time. Yuan Zhen, in a note to one of his poems, claims that many poems circulating as his own were, in fact, fakes: The younger generation enjoys composing fake poems attributed to me; these circulate all over. Since I have arrived in Kuaiji someone has already written one hundred “Palace Verses” and “Miscellaneous Poems” in two juan. All of these are said to have been written by me. When I examined them myself, there was not a single one that was. 後輩好僞作予詩，傳流諸處。自到會稽，已有人寫宮詞百篇及雜詩兩卷，皆 云是予所撰。及手勘驗無一篇是者。105

Bai Juyi also ends the postface to his final collection by claiming that any omissions or false attributions are honest mistakes and not attempts at trickery on

104  It is worth noting in this context that printing clearly does not eliminate variation: it simply changes the type of variation. As each reproduction is not unique, there will be general uniformity within editions printed from the same blocks (though blocks were often altered). Yet any variation that appears in a printed edition will be much more widespread. Also, much as printing can give a sense of fixedness and uniformity, early printed editions were probably based on sources similar to those described by Tan Yu, with all the irregularity they imply. 105  Yuan Zhen, “Chou Letian yusi bujin jiawei liuyun zhi” 酬樂天餘思不盡加爲六韻之, yzj, 22.247.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1170

Nugent

his part. An intriguing implication of this is that perhaps Bai Juyi is admitting that he himself might not recognize a forgery!106 We can never definitively determine why these two poets, more than any other Tang author, showed such a concern with authenticity and control over specific versions of their works; but the idea of commercial culture may come into play here in a reverse of my discussion above. While there is no evidence that either Bai Juyi or Yuan Zhen directly made money from their poetry by selling it as a commodity, both describe their works being used by others to make a profit. Yuan Zhen tells of finding his and Bai Juyi’s poems for sale in a market and encountering a Korean merchant who would take them home and sell them to his country’s Chief Minister.107 Bai Juyi claims that knowledge of his poetry would increase a singing girl’s price.108 Thus, even if they did not personally participate in the commercial culture surrounding their literary works, both Bai Juyi and Yuan Zhen were aware that such a context existed, or at least imagined it did. Perhaps this knowledge, combined with a certain degree of natural meticulousness and self-regard, can begin to explain the exceptional motivations behind Bai Juyi’s and, to a lesser extent, Yuan Zhen’s compilations of their own literary collections. Individual literary collections were but one of the many avenues through which literary works in the Tang were preserved and circulated, but they were both a common way for people to obtain the works of poets from earlier in the dynasty and the partial basis for the anthologies that would become the source material for many later editions. Tang literature changed over time. Through the centuries the works produced in the Tang were edited, canonized, eliminated, and sometimes even created anew.109 Each successive generation of admirers and scholars has made its influence felt on what we now have as the received tradition. While it is never possible to bypass entirely the mediation this process implies, it is my hope that, by looking back at how people of the Tang themselves created and used one particularly important medium of literary preservation and transmission, we may gain a better understanding of what Tang literature was in the period of its genesis, the first step on the path that it has traveled down to us today.

106  “Baishi Changqing ji houxu,” bjyjjj, “waiji” 3.3917. 107  “Baishi Changqing ji xu,” bjyjjj, “fulu” 3973. 108  “Yu Yuanjiu shu,” bjyjjj, 45.2793. 109  Referring to the many spurious attributions and new discoveries made in the centuries since the Tang itself.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji Jia Jinhua 賈晉華 Abbreviations cqsf Tao Zongyi 陶宗儀 (1316–1403) (ed.), Tao Ting 陶珽 (re-ed.), Shuofu 說邪 (120 juan), as Chong ji Shuofu 重輯說郭 in Shuofu sanzhong 說郭三種 (Shanghai: Shanghai guji, 1988). gzj Ganzhu ji 紺珠集 (skqs). jlwgj Wu Pingyi 武平一, Jinglong wenguan ji 景龍文館記, fragments in gzj (skqs), ls (skqs), etc. jts Jiu Tangshu 舊唐書 (Beijing: Zhonghua shuju, 1975). ls Leishuo 類說 (skqs). qts Quan Tangshi 全唐詩 (Beijing: Zhonghua shuju, 1960). qtw Qun Tangwen 全唐文 (Taibei: Datong shuju, 1979). sf Tao Zongyi (ed.), Zhang Zongxiang 張宗祥 (collation), Shuofu (100 juan), in Shuofu sanzhong. skqs Siku quanshu 四庫全書. thy Tang huiyao 唐會要 (Taibei: Shijie shuju, 1982). tsjs Tangshi jishi 唐詩紀事 (Shanghai: Shanghai guji, 1965). xts Xin Tangshu 新唐書 (Beijing: Zhonghua shuju, 1975). wyyh Wenyuan yinghua 文苑英華 (Beijing: Zhonghua shuju, 1966). zztj Zizhi tongjian 資治通鑒 (Beijing: Zhonghua shuju, 1971).

The* Jinglong wenguan ji 景龍文館記 (A Record of the Jinglong Literary Institute), compiled by Wu Pingyi 武平一 (d. ca. 741) in the second decade of the eighth century, is a poetic anthology as well as a record of court literary activities in the Jinglong reign-period (707–710) of Emperor Zhongzong 中宗 (r. 705–710). Emperor Zhongzong transformed the national institute, Xiuwen guan 修文館, into a literary institute, to which he summoned nearly all the famous poets of the time. These poets were appointed as xueshi 學士, or Source: “A Study of the Jinglong wenguan ji,” Monumenta Serica 47 (1999): 209–36. © Monumenta Serica Institute, reprinted by permission of Taylor & Francis Ltd, www.tandfonline.com on behalf of Monumenta Serica Institute. *  I should like to thank James S. Glasscock and Roman Malek for their suggestions on draft versions of this article.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_037

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1172

Jia

academicians. They and other court poets formed an important literary group. However, since the Jinglong wenguan ji was lost for a long time, until now only a few scholars have noticed it. In this paper, I attempt to present a further study of this text. First, the text will be reconstructed to the extent possible from several early sources. Second, the literary activities of the Jinglong court and the relationship between political power and the development of literature will be described. Third, the poetic genres used by the Jinglong court poets will be discussed. Finally, the poems anthologized in this text and their styles and features will be analyzed. 1 The establishment of the positions of the academicians in the Jinglong reignperiod is recorded in several early sources. In the extant Jinglong wenguan ji we read: In the second year of the Jinglong reign-period (708), Emperor Zhongzong appointed to the Xiuwen guan four grand academicians modeled after the four seasons, eight academicians modeled after the eight solar terms, and twelve auxiliary academicians modeled after the twelve months.1 The biography of Li Shi 李適 in the Xin Tangshu 新唐書 records the following: Previously, in the second year of the Jinglong reign-period, Emperor Zhongzong established in the Xiuwen guan the positions of four grand academicians, eight academicians, and twelve auxiliary academicians, modeling them after the four seasons, the eight solar terms, and the twelve months. Therefore, [His Highness] appointed Li Qiao 李嶠 (ca. 645–714), Zong Chuke 宗楚客 (d. 710), Zhao Yanzhao 趙彥昭, and Wei Sili 韋嗣立 (654–719) as grand academicians; (Li) Shi (657–716), Liu Xian 劉憲 (d. 711), Cui Shi 崔湜 (671–713), Zheng Yin 鄭愔 (d. 710), Lu Cangyong 盧藏用 (d. ca. 713), Li Yi 李乂 (657–716), Cen Xi 岑羲 (d. 712), and Liu Zixuan 劉子玄 (661–721) as academicians; Xue Ji 薛稷 (649–713), Ma Huaisu 馬懷 素 (658–718), Song Zhiwen 宋之問 (ca. 656–712), Wu Pingyi, Du Shenyan 杜審言 (ca. 645–708), Shen Quanqi 沈佺期 (d. ca. 713), and Yan Chaoyin 閻朝隱 (d. 712) as auxiliary academicians. In addition, [his highness] summoned Xu Jian 徐堅 (d. 729), Wei Yuandan 韋元旦, Xu Yanbo 徐彥伯 1  j lwgj, fragments in gzj, 7.17b, and ls, 6.22b–23a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1173

(d. 714), and Liu Yunji 劉允濟 (d. 708) to fill the positions. Later, some other people were also chosen as academicians.2 The Xin Tangshu lists the names of four grand academicians, eight academicians, and eleven auxiliary academicians. In his Tang huiyao 唐會要, Wang Pu 王溥 (922–982) mentions one more auxiliary academician, namely, Su Ting 蘇頲 (670–727).3 Putting them together, we now have the full list of the first twenty-four academicians. In addition, the Xin Tangshu hints that some other people later were chosen as academicians. That must have happened when some academicians died or were banished from the court. For example, both Du Shenyan and Liu Yunji died soon after their appointments.4 I have found the names of three academicians who were appointed later, namely, Cui Riyong 崔 日用 (673–722),5 Li Jiongxiu 李迴秀 (d. 712),6 and Zhang Yue 張說 (667–731).7 According to the Tang huiyao, the establishment of the positions was on the twenty-second day of the fourth month in the second year of the Jinglong reignperiod (14 May 708).8 The last activity of the Jinglong academicians was in the fifth month of the fourth year of Jinglong (710),9 and Emperor Zhongzong died in the following month. Thus, Wu Pingyi, one of the academicians, might have compiled the Jinglong wenguan ji in the second decade of the eighth century. Several Song catalogues mention this text. The catalogue in the Xin Tangshu records: “Jinglong wenguan ji, in ten juan, compiled by Wu Pingyi.”10 The twelfth-century Junzhai dushu zhi 郡齋讀書志 records:

2  xts, 202.5748. The entry of Li Shi in tsjs (9.113, 127) copies this paragraph, but the editor of Shanghai guji adds the name of Wei Anshi 韋安石 as academician according to Tangyin guiqian 唐音癸簽. Wang Zhongyong 王仲鏞, in his Tangshi jishi jiaojian 唐詩紀事校 箋 (Chengdu: Ba Shu shushe, 1989, 9.211), argues that Wei Anshi should not have been an academician. I agree with Wang. 3  thy, 64.1114–15. 4  Du Shenyan died in the tenth month of that year. See Song Zhiwen, “Ji Du xueshi Shenyan wen” 祭杜學士審言文, qtw 241.16a–18a. The biography of Liu Yunji in xts (209.6641) records, “[Liu] was summoned to be academician of the Xiuwen guan. Having been banished for a long time, he was very happy [that he was back to the court.] He drank with his family for several days and died.” 5  xts, 121.4330. 6  xts, 99.3914. 7  jts, 97.3051. 8  thy, 64.1114–15. 9  See zztj, 209.6641. 10   x ts, 58.1485.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1174

Jia

Jinglong wenguan ji, in eight juan, compiled by Wu Zhen 武甄, whose byname was Pingyi, one of the academicians of the Xiuwen guan in the Tang dynasty. This book records the miscellaneous events of the institute since Emperor Zhongzong appointed the academicians. It also collects the poems and essays that the academicians wrote by imperial command or to one another. In addition, it reports the unrestrained feasts and improper familiarity between the emperor and the academicians, and the sudden death of the emperor. The last three juan are biographies of the academicians. Now two juan have been lost. Pingyi used his byname as his first name.11 From these records, we learn that the original Jinglong wenguan ji comprised three parts: records of the activities of the literary institute and the court, works of the academicians, and biographies of them. It originally had ten juan, but by the time of the Southern Song, only eight juan were extant. After the Song, even this eight-juan-edition was lost. The Japanese scholars paid attention to this important text quite early. Using the data given in the Quan Tang shihua 全唐詩話,12 Takagi Masakazu 高木正一 enumerates forty-one literary events of the Jinglong court, the names of twenty-three Jinglong academicians, and forty-three titles of their poems.13 Nishimura Tomiko 西村冨美子 lists fifty-four literary events by analyzing the sources of the Tangshi jishi, Zizhi tongjian, the biographies of Emperor Zhongzong in the Jiu Tangshu and Xin Tangshu, Wenyuan yinghua, and Quan Tangshi.14 However, among those events, three are duplicate,15 and two cannot be proved to be related with the institute.16 Thus, Nishimura’s list actually has forty-nine events in total. Andō Shunnoku 安東俊六 also studies some literary events of the Jinglong court and their background.17

11   Junzhai dushu zhi (Taibei: Guangwen shuju, 1968), 7.4b. 12   Quan Tang shihua is in fact an abridged version of tsjs. 13  Takagi Masakazu, “Keiryu no kyūtei shidan to hichigen risshi no keisei” 景龍の宮延詩壇 と 七言律詩の形成, Ritsumeikan bungaku 立命館文學 224 (1964): 45–81. 14  Nishimura Tomiko, “Sho Tō ki no ōsei shijin” 初唐期の應制詩人, Shitenōji joshidaigaku kiyō 四天王寺女子大學紀要 9 (1976): 119–139. 15  These are the events of visiting Jianfu Temple, visiting Linwei Pavilion, and visiting HotSpring Palace in Xinfeng District. 16  These are the events of visiting Princess Changning’s residence and Xiao Zhizhong’s residence. 17  Andō Shunnoku, “Keiryu kyūtei bungaku no sōsaku kiban” 景籠宮廷文學の創作基盤, Chūgoku bungaku ronshu 中國文學論集 3 (1972): 13–24.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1175

A Study of the Jinglong Wenguan Ji

In addition to the Japanese scholars’ studies, I have found one entry of Jinglong wenguan ji in the Taiping guangji 太平廣記,18 three in the commentary to Zizhi tongjian,19 twenty-four in the twelfth-century Ganzhu ji,20 fifteen in the twelfth-century Leishuo,21 three in the one hundred juan Shuofu,22 and ten in the one-hundred-twenty juan Shuofu.23 To sum up, we now have fiftynine literary events in total. I have also collected three hundred forty-three shi 詩 poems, five ci 詞 poems, one fu 賦 poem, and four prose prefaces, which are extant fragments of the Jinglong wenguan ji. Moreover, to the twenty-seven academicians I have added thirty-six names of other poets who participated in some of the literary activities and have some poems extant. They are: Emperor Zhongzong, Shangguan Wan’er 上官婉兒 (664–710), Xiao Zhizhong 蕭至忠 (d. 713), Yang Lian 楊廉, Xin Tifou 辛替否 (d. 742), Wang Jing 王景, Bi Qiantai 畢乾泰, Qu Zhan 麴瞻, Fan Chen 樊忱, Sun Quan 孫恮, Li Congyuan 李從遠 (d. ca. 711), Zhou Liyong 周利用, Li Heng 李恆, Zhang Jingyuan 張景源, Zhang Xi 張錫 (d. ca. 711), Xie Wan 解碗 (d. 718), Lu Jingchu 陸景初 (665–736), Li Jingbo 李景伯, Shao Sheng 邵昇, Wei Anshi 韋安石 (651–714), Dou Xijie 竇希玠, Li Xian 李咸, Zheng Nanjin 鄭南金, Yu Jingye 于經野, Lu Huaishen 盧懷慎, Li Rizhi 李日之, Empress Wei 韋後 (d. 710), Princess Changning 長寧公主 (d. 710), Princess Anle 安樂公主 (d. 710), Princess Taiping 太平公主 (d. 713), Li Chongmao 李重茂 (698–714), Dou Congyi 竇從一 (d. 713), Zong Jinqing 宗晉卿 (d. 710), Ming Xilie 明希獵, Tang Yuanzhe 唐遠恕, and one anonymous court actor.24 The following is a chronology of the events, literary works, and poets from my reconstruction of the Jinglong wenguan ji. Since most of the academicians’ recent-style poems comply perfectly with the rules of lüshi 律詩, or regulated verse, I will simply apply the terms of wulü 五律 (eight-line pentasyllabic regulated verse), qilü 七律 (eight-line heptasyllabic regulated verse), wupai 五排 (extended pentasyllabic regulated verse), wujue 五絕 (pentasyllabic quatrain), and qijue 七絕 (heptasyllabic quatrain). Some unimportant words in the titles, such as feng he 奉和 (respectfully write a poem in reply), ying zhi 應制 (by imperial command), and fen yun 分韻 (set the topic and allot different rhymes to different poets) will be omitted. 18   Taiping guangji (Beijing: Zhonghua shuju, 1961), 271.2132. 19   z ztj, 209.6629, 6638, 6641. 20   g zj, 7.17a–21b. 21   l s, 6.22b–25a. 22   s f, 77.12a. 23   c qsf, 46. 1a–3b. 24  The data are from wyyh, tsjs, and qts. See the following chronology.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1176

Jia

I. = Date and Year of Jinglong Reign-period25 II. = Events III. = Extant Works: Title / Genre / Number IV. = Authors V. = Sources I. 7/7, 2nd yr. (7/28/708) II. Banqueting in Liangyi Palace; Li Xingyan 李行言 sang a Taoist song, “Treading the Void.” III. Banqueting in Liangyi Palace on the Seventh Evening of the Seventh Month (七夕兩儀殿會宴) / wulü / 6. IV. Li Qiao, Zhao Yanzhao, Liu Xian, Li Yi, Su Ting, Du Shenyan. V. WYYH, 173.1–2a; TSJS, 9.114, 11.169–170; QTS, 58.692, 103.1087, 71.779, 92.994, 73.799, 62.732. I. 9/9, 2nd yr. (9/27/708) II. Visiting Ci’en Temple. Shangguan Wan’er presented a poem, and all courtiers followed. III. Ascending the Pagoda of Ci’en Temple on the Ninth Day of the Ninth Month (九月九日登慈恩寺浮圖) / wulü / 28. IV. Shangguan Wan’er,26 Li Qiao,27 Zhao Yanzhao,28 Liu Xian, Zheng Yin, Li Yi, Song Zhiwen, Xiao Zhizhong, Li Jiongxiu, Yang Lian, Xin Tifou, Wang Jing, Bi Qiantai, Qu Zhan, Fan Chen, Sun Quan, Li Congyuan, Zhou Liyong, Li Heng, Zhang Jingyuan, Zhang Xi, Xie Wan, Xue Ji, Ma Huaisu, Cui Riyong, Cen Xi, Lu Cangyong, Li Shi. V. WYYH, 178.1b–2; TSJS, 9.114; QTS, 5.60–61, 58.693, 103.1088, 71.780, 106.1106, 92.995, 52.631, 104.1091, 104.1093, 104.1094, 105.1099, 105.1099, 105. 1100, 105. 1100, 105.1100, 105.1101, 105.1101, 105.1102, 105.1102, 105.1102, 105.1102–1103, 105.1103, 93.1006, 93.1008, 93.558, 93.1004, 93.1002, 20.776.

25  Gregorian calendar in parentheses. 26  Shangguan Wan’er’s poem is also attributed to Cui Shi in qts (54.663), but according to wyyh (178.2), this poem should belong to Shangguan. 27  Li Qiao’s poem is also attributed to Song Zhiwen in qts (52.643), but according to wyyh (178.1b), this poem should belong to Li. 28  Zhao Yanzhao’s poem is also attributed to Zhao Yanbo in qts (882.9969), but Zhao Yanbo should be Zhao Yanzhao, see Zhou Zuzhuan 周祖譔 (ed.), Zhongguo wenxuejia da cidian: Tang Wudai juan 中國文學家大辭典 : 唐五代卷 (Beijing: Zhonghua shuju, 1992), 559.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1177

I. 9/9 (leap), 2nd yr. (10/27/708) II. Visiting Zongchi Temple. III. Your Majesty Visits Zongchi Temple and Ascends the Pagoda on the Ninth Day of the Leap Ninth Month (閏九月九日幸總持寺登浮圖) / wulü / 4. IV. Li Qiao, Song Zhiwen, Liu Xian, Li Yi. V. JLWGJ: GZJ, 7.18a; LS, 6.23a; WYYH, 178.2b–3a; TSJS, 9,114; QTS, 58.693, 52. 632, 71. 780, 92.995. I. 10/3, 2nd yr. (11/19/708) II. Visiting Sanhui Temple. III. Your Majesty Visits Sanhui Temple (幸三會寺) / wupai / 6. IV. Shangguan Wan’er, Li Qiao, Zheng Yin, Liu Xian,29 Li Yi, Song Zhiwen. V. WYYH, 178.3; TSJS, 9.114; QTS, 5.61, 61.724, 106.1107, 71.782, 92.999, 53.647. I. 11/15, 2nd yr. (12/31/708) II. Banqueting in the inner palace to celebrate Emperor Zhongzong’s birthday. III. Banqueting in the Inner Palace to Celebrate Your Majesty’s Birthday (帝誕辰內殿宴群臣) / heptasyllabic linked verse / 1. IV. Emperor Zhongzong, Li Qiao, Zong Chuke, Liu Xian, Cui Shi, Zheng Yin, Zhao Yanzhao, Li Shi, Su Ting, Lu Cangyong, Li Yi, Ma Huaisu, Xue Ji, Song Zhiwen, Lu Jingchu, Shangguan Wan’er. V. JLWGJ: CQSF, 46.2b; TSJS, 9.114; QTS, 2.24. I. II. III. IV. V.

11/21, 2nd yr. (1/16/709) Princess Anle married Wu Yanxiu. Banqueting in the Residence of Princess Anle (宴安樂公主宅) / wupai / 1. Song Zhiwen. JTS, 7.146; ZZTJ, 209.6630; WYYH, 176.10b–11a; TSJS, 9.114; QTS, 53.649.

I. 12/6, 2nd yr. (1/21/709)30 II. Visiting Jianfu Temple, formerly Emperor Zhongzong’s residence. Zheng Yin was the first to finish the poem, and Song Zhiwen the second. III. Your Majesty Visits Jianfu Temple (幸薦褔寺) / wupai / 7.

29  Liu Xian’s poem is also attributed to Xiao Zhizhong in qts (104.1092), but according to wyyh (178.3a), it should belong to Liu. 30   j ts (7.147) records this event on the fifteenth day of the first month in the third year of the Jinglong reign-period (3/1/709).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1178

Jia

IV. Li Qiao, Zhao Yanzhao, Song Zhiwen, Zheng Yin, Liu Xian,31 Li Yi. V. WYYH, 178.4–5; TSJS, 9.114; QTS, 61.724, 103.1089, 53.647, 53.648, 106.1107, 71.782, 92.1000. I. 12/19, 2nd yr. (2/3/709) II. Visiting the royal garden. III. Visiting the Royal Garden and Greeting the Spring on the Day of Beginning of Spring (立春日游苑迎春) / qilü / 8. IV. Emperor Zhongzong, Li Shi, Wei Yuandan, Yan Chaoyin, Shen Quanqi, Lu Cangyong, Ma Huaisu, Cui Riyong. V. TSJS, 9.114; QTS, 2.24, 70.777, 69.773, 69.771, 96.1041, 93.1003, 93.1009, 93.559. I. 12/21, 2nd yr. (2/5/709) II. Visiting Linwei Pavilion. III. Attending Your Majesty to Visit the Royal Garden and Linwei Pavilion and Catch a Snow (游禁苑陪幸臨渭亭遇雪) / wulü / 5. IV. Li Qiao, Li Shi, Li Yi, Xu Yanbo, Su Ting. V. WYYH, 173.9–10a; TSJS, 9.114; QTS, 58.693, 70.776, 92.994, 76.823, 73.799. i. 12/30, 2nd yr. (2/14/709) II. Visiting me ancient city of Chang’an. Staying all night on New Year’s Eve. Married the empress’ wet nurse to Dou Congyi. III. Your Majesty Visits Weiyang Palace in the Ancient City of Chang’an (幸長安故城未央宮) / wupai / 5. Staying All Night on New Year’s Eve ( 守歲) / qilü / 1. IV. Li Qiao, Zhao Yanzhao, Liu Xian, Song Zhiwen, Li Yi. Shen Quanqi. V. JLWGJ: GZJ, 7.18; LS, 6.23; WYYH, 174.3b–4, 169.11b–12a; TSJS, 9.114; QTS, 61.724, 103.1089, 71.782, 53.648, 92.999, 96.1043. I. 1/7, 3rd yr. (2/21/709) II. Banqueting in Qinghui Pavilion. This day everybody was very happy. The emperor asked the academicians to dance.

31  Liu Xian’s poem is also attributed to Xiao Zhizhong in qts (104.1092), but according to wyyh (178.4b), it should belong to Liu.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1179

III. Banqueting in Qinghui Pavilion and Catching a Snow on Human Day (人日清暉閣宴群臣遇雪) / wulü / 6. Enjoying the Snow on Human Day (人日玩雪)32 / qijue / 7. Song of Water Wave (回波詞)/ ci / 4. IV. Li Qiao, Zong Chuke, Liu Xian, Li Yi, Zhao Yanzhao, Su Ting. Liu Xian, Song Zhiwen, Shen Quanqi, Li Qiao, Xiao Zhizhong, Xu Yanbo, Zhao Yanzhao. Shen Quanqi, Li Jingbo, Cui Riyong, an anonymous court actor.33 V. JLWGJ: GZJ, 7.18b; SF, 46.1a; Sui Tang jiahua 階唐嘉話,34 Da Tang xinyu 大唐新語,35 Benshi shi, 24b–25; ZZTJ, 209.6633; WYYH, 173.8a–9b; TSJS, 9.114; QTS, 58.692, 46.560, 71.779, 92.994, 103.1087, 73.799, 71.783, 53.656, 97.1054, 61.729, 104.1092–1093, 76.826, 103.1090, 97.1054, 101.1078–1079, 869.9849, 890.10049.

32  This sequence has several titles. tsjs says, “On Human Day of the third year of the Jinglong reign-period,… Li Qiao and others wrote heptasyllabic poems (such as ‘One thousand goblets of royal wine present in the imperial banquet’ 千鍾聖酒御筵披)” (9.114). It does not indicate the title of the heptasyllabic poems and does not attribute the line mentioned to Li Qiao. qts attributes a poem which includes this line to Li Qiao under the title of “Ascending Qinghui Pavilion and Catching a Snow” 上清暉閣遇雪 (61.729–30) and also to Xu Yanbo and Zhao Yan bo under the title of “Catching a Snow in the Royal Garden: by Imperial Command” 苑中遇雪應制 (76.826, 104.1097). As I have mentioned, Zhao Yanbo should be Zhao Yanzhao. qts attributes another heptasyllabic quatrain to Zhao Yanzhao under the title of “Catching a Snow in the Royal Garden on Human Day: by Imperial Command” 苑中人日遇雪應制 (103.1090). It also attributes four pieces of heptasyllabic quatrains to Li Qiao, Song Zhiwen, Shen Quanqi, and Xiao Zhizhong under the title of “Catching a Snow in the Royal Garden: by Imperial Command” (61.729, 53.656, 97.1054, 104.1092–1093). The last one is also attributed to Liu Xian (71.783), but Liu has another quatrain entitled “Enjoying the Snow on the Human Day: by Imperial Command” 人日玩雪應制 (71.783). Since Emperor Zhongzong and his academicians only had two Human Days to enjoy themselves (the third year and fourth year of the Jinglong reignperiod), and the second Human Day was a sunny day (see below), I date these poems to this year. These are in the same sequence under different titles due to the long time circulation. I attribute them to the poets as follows: Liu Xian, “Enjoying the Snow on Human Day: by Imperial Command”; Zhao Yanzhao, “Catching a Snow in the Royal Garden on Hu man Day: by Imperial Command”; Li Qiao, Song Zhiwen, Shen Quanqi, Xiao Zhizhong, and Xu Yanbo, “Catching a Snow in the Royal Garden: by Imperial Command.” 33  This poem is also attributed to Pei Tan in qts (890.10049), but according to Meng Qi’s 孟槃 Benshi shi 本事詩 (skqs, 25a), it should belong to an anonymous court actor. 34  Liu Su 劉餗, Sui Tang jiahua (Shanghai: Gudian wenxue, 1956), 2.25. 35  Liu Su 劉肅, Da Tang xinyu (Shanghai: Gudian wenxue, 1956), 3.70.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1180

Jia

I. 1/17, 3rd yr. (3/2/709) II. Banqueting in the Pavilion of Pear Garden. III. IV. V. JTS, 7.147. I. 1/29, 3rd yr. (3/15/709) II. Visiting Kunming Pool. About one hundred courtiers wrote poems by imperial command. A silk building was made in front of the palace. The emperor asked Shangguan Wan’er to choose one poem as new palace song. Shangguan evaluated Song Zhiwen’s poem as the first. III. Your Majesty Visits Kunming Pool on the Last Day of the Month (晦日駕 幸昆明池) / wupai / 4. IV. Shen Quanqi, Song Zhiwen, Li Yi, Su Ting. V. WYYH, 176.9; TSJS, 9.114; QTS, 97.1045, 53.647, 92.999, 74.807. I. 2/2, 3rd yr. (3/17/709) II. Watching the palace maids playing tug-of-war in Xuanwu Gate, ordering them to make up a market of palace, and to trade and bargain with the courtiers and academicians. III. IV. V. JTS, 7.147; ZZTJ, 209.6631. I. 2/8, 3rd yr. (3/23/709) II. Seeing off Xuanzhuang and other monks to return to Jingzhou. III. Seeing off Monk Hongjing, Daojun, and Xuanzhuang to return to Jingzhou (送沙門弘景，道俊，玄奘還荊州) / wulü / 2. IV. Li Qiao,36 Li Yi. V. WYYH, 277.2b–3a; TSJS, 9.114; QTS, 58.694, 92.995 I. 2/11, 3rd yr. (3/26/709) II. Visiting Princess Taiping’s residence. III. Your Majesty Visits Princess Taiping’s South Residence in Early Spring (初春幸太平公主南莊) / qilü / 8.

36  Li Qiao’s poem is also attributed to Song Zhiwen in qts (52.643), but according to wyyh (177.2b), this poem should belong to Li.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1181

IV. Li Qiao, Shen Quanqi,37 Song Zhiwen, Su Ting,38 Li Yi, Wei Sili,39 Shao Sheng, Zhao Yanzhao.40 V. WYYH, 176.9–10; TSJS, 9.114; QTS, 61.723, 96.643, 52.645, 96.1041, 92.997, 91.987, 69.774, 115.1169. I. 2nd month, 3rd yr. (3/16–4/14/709) II. Banqueting in the palace, ordering the courtiers and academicians to perform their skills. Zhang Xi performed “Dance of Tan Rongniang”; Zong Jinqing performed “Dance of Huntuo”; Zhang Qia 張洽 performed “Dance of Huangzhang”; Du Yuantan 杜元談 chanted the Dhāraṇī of Brāhman (Boluomen zhou 波羅門咒); Lu Cangyong mimiced the presenting of Taoist prayer; Guo Shanyun 郭山惲 sang two songs of Shijing 詩經. III. IV. V. ZZTJ, 209.6631–6633. I. 3/3, 3rd yr. (4/15/709) II. Banqueting in Pear Garden. III. Attending a Banquet in Pear Garden on the Third Day (三日梨園侍 宴)41 / wulü / 1. IV. Shen Quanqi. V. QTS, 96.1029.

37  Shen Quanqi’s poem is also attributed to Su Ting in qts (73.804), but according to wyyh (176.9b), it should belong to Shen. 38   q ts (96.1041) ascribes this poem to Shen Quanqi, but according to wyyh (176.9b), it should belong to Su Ting. 39  Wei Sili’s poem is also attributed to Zhao Yanzhao in qts (103.1089), but according to wyyh (176.10a), it should belong to Wei. 40   w yyh (176.10b) ascribes this poem to Song Yong 宋雍, but Song lived during the reigns of Emperor Daizong 代宗 and Dezong 德宗. See Zhongguo wenxuejia da cidian: Tang Wudai juan, 401. qts (115.1169) ascribes this poem to Li Yong 李邕, but according to the biography of Li Yong in jts (190.5040–5041), Li was banished from the court during the Jinglong reign-period. Since Zhao Yanzhao was one of the academicians, and Wei Sili’s poem of the same title is also attributed to Zhao in qts (103.1089), this poem might belong to Zhao. 41  This poem has another title, “Attending a Banquet in the Pavilion of Pear Garden” 梨圈 亭 侍宴. The poem says, “The Lustral Hall Opens on the Third Day of the Third Month” 上巳禊堂開. Since the emperor and his academicians visited the Wei River on the Third Day of the Third Month in the following year, this poem must have been written that year.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1182

Jia

I. 7/7, 3rd yr. (8/16/709) II. Banqueting in the Pavilion of Pear Garden. III. IV. V. JTS, 7.147 I. 8/11, 3rd yr. (9/19/709)42 II. Seeing off Military Commissioner, Zhang Rendan, to Shuofang Defense Command. III. Your Majesty Sees off Military Commissioner, Zhang Rendan, to Shuofang Defense Command in Wangchun Palace (幸望春宮送朔方總管張 仁亶) / wupai / 4 // wulü / 2. IV. Li Qiao, Liu Xian,43 Li Yi, Su Ting, Zheng Yin, Li Shi. V. wyyh, 177.1b–2; tsjs, 9.114; qts, 61.724, 71.782, 92.999, 74.809, 106.1106, 70.777. I. 8/21, 3rd yr. (9/29/709)44 II. Visiting Princess Anle’s residence. III. Your Majesty Visits Princess Anle’s Villa (幸安樂公主山莊) / qilü / 15 and 2 unattached lines. IV. Li Qiao,45 Zhao Yanzhao, Zong Chuke, Lu Cangyong, Su Ting, Xiao Zhizhong, Cen Xi, Li Yi, Ma Huaisu, Wei Yuandan, Li Jiongxiu, Li Shi, Xue Ji, Liu Xian, Shen Quanqi, Li Rizhi. V. jlwgj: cqsf, 46.2b; wyyh, 176.4b–6a; tsjs, 9.114; qts, 61.723, 103.1089, 46.561, 93.1004, 73.804, 104.1091, 93.1005, 92/998, 93.1010, 69.773, 104.1093, 70.778, 93.1007, 71.781, 96.1041. I. 9/9, 3rd yr. (10/16/709) II. Visiting Linwei Pavilion. The emperor set the topic and gave different rhymes to different poets. III. Your Majesty Ascends Linwei Pavilion on the Ninth Day of the Ninth Month (九月九日幸臨渭亭登高) / wulü / 25 // prose / 1. 42   t sjs (9.114) records this event in the seventh month. Here I adopt the record of jts (7.148). 43  Liu Xian’s poem is also attributed to Xiao Zhizhong in qts (104.1092), but according to wyyh (177.1b), it should belong to Liu. 44   t sjs (9.114) records this event on the third day of the eighth month. Here I adopt the record of jts (7.148) and zztj (209.6637). 45  Li Qiao’s poem is entitled as “Attending a Banquet in Princess Taiping’s Mountain Pavilion: by Imperial Command” 太平公主山亭侍宴應制 in qts (61.723). I change the title according to wyyh (176.4b). Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1183

IV. Emperor Zhongzong (poem and preface), Wei Anshi, Su Gui, Li Qiao,46 Xiao Zhizhong, Dou Xijie, Wei Sili, Li Jiongxiu, Zhao Yanzhao, Yang Lian, Cen Xi, Lu Cangyong, Li Xian, Yan Chaoyin, Shen Quanqi, Xue Ji, Su Ting, Li Yi, Ma Huaisu, Lu Jingchu, Wei Yuandan, Li Shi, Zheng Nanjin, Yu Jingye, Lu Huaishen. V. JTS, 7.148; WYYH, 169.8b–9a; TSJS, 9.114; QTS, 2.23, 104.1094, 46.562, 58.696, 104.1091, 104.1095, 91.986, 104.1093, 103.1088, 104.1094, 93.1004, 93.1002, 104.1096, 69.770, 96.1030, 93.1006, 73.799, 92.994, 93.1008, 104.1095, 69.773, 70.776, 104.1096, 104.1097. I. 10/8, 3rd yr. (11/13/709)47 II. Banqueting on Princess Anle’s new residence. III. Attending a Banquet to Celebrate Princess Anle’s Moving into Her New Residence (安樂公主移入新宅侍宴)/ wulü / 4. Banqueting in Princess Anle’s New Residence at Night (夜宴安樂公主新 宅) / Qijue / 15 // prose / 1. IV. Zong Chuke,48 Zhao Yanzhao, Wu Pingyi, Shen Quanqi. Yan Chaoyin, Li Yi, Xu Yanbo (poem and preface), Su Ting, Liu Xian, Li Shi, Wei Yuandan, Wu Pingyi, Li Jiongxiu, Shen Quanqi, Xue Ji, Ma Huaisu, Cui Riyong, Cen Xi, Lu Cangyong. V. JTS, 7.148; WYYH, 176.4; TSJS, 9.114, 119; QTS, 46.561, 103.1088, 102.1084, 96.1030, 69.771, 92.1001, 76.826, 74.815, 71.783, 70.778, 69.774, 102.1085, 104.1093, 97.1054, 93.1008, 93.1010, 46.560, 93.1005, 93.1004. I. 11/13, 3rd yr. (12/18/709)49 II. Emperor Zhongzong performed the ritual of the South Suburbs; Xu Yanbo presented a fu. III. On the Ritual of the South Suburbs (南郊賦) / fu / 1. IV. Xu Yanbo. V. JTS, 7.148; ZZTJ, 209.6637; TSJS, 9.114; QTW, 267.17–20b.

46  Li Qiao’s poem is also attributed to Song Zhiwen in qts (52.632), but according to tsjs (1.8), this poem should belong to Li. 47   t sjs (9.114) records this event in the first day of the eleventh month. Here I adopt the record of jts (7.148). 48  Zong Chuke’s poem is also attributed to Li Shi in qts (70.777), but according to wyyh (176.4a), it should belong to Zong. 49   t sjs (9.114) records this event on the twenty-third day of the eleventh month. Here  I adopt the records of jts (7.148) and zztj (209.6637). Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1184

Jia

i. 11/15, 3rd yr. (12/20/709) II. Banqueting to celebrate Emperor Zhongzong’s birthday and the completion of the first month of Princess Changning’s daughter. III. Attending a Banquet to Celebrate Emperor Zhongzong’s Birthday and the Completion of the First Month of Princess Changning’s Daughter (中宗降誕日長寧公主滿月侍宴) / wulü / 2. IV. Li Qiao, Zheng Yin. V. WYYH, 169.9a; TSJS, 9.114, 1.9; QTS, 58.691, 106.1105. i. 12/12, 3rd yr. (1/16/710) II. Visiting Hot-Spring Palace in Xinfeng District. The emperor ordered Xu Yanbo, Prefect of Puzhou, to come as an attendant like other academicians. Xu, Wu Pingyi, and three others presented poems. III. Your Majesty Visits Hot-Spring Palace in Xinfeng District (幸新豐溫泉宮) / wupai / 2 // qijue / 3. IV. Xu Yanbo, Wu Pingyi, Shangguan Wan’er. V. ZZTJ, 209.6638; WYYH, 170.10; TSJS, 9.114–115; QTS, 76.825, 102.1084, 5.61. i. 12/14, 3rd yr. (1/18/710)50 II. Visiting Wei Sili’s villa. III. Attending a Banquet in Wei Sili’s Villa (幸韋嗣立山莊侍宴) / wupai / 10 // prose / 1. Your Majesty Visits Wei Sili’s Villa (幸韋嗣立山莊)/ qijue / 9. IV. Li Qiao, Li Yi,51 Shen Quanqi, Wu Pingyi, Zhao Yanzhao, Xu Yanbo, Liu Xian, Cui Shi, Zhang Yue (poem and preface), Su Ting. Li Qiao, Li Yi, Shen Quanqi, Wu Pingyi, Zhao Yanzhao, Liu Xian, Cui Shi, Zhang Yue, Su Ting. V. JLWGJ: GZJ, 7.19b, LS, 6.25a; WYYH, 175.8–11; TSJS, 11.154; QTS, 61.725, 92.999, 97.1044, 102.1085, 103.1090, 76.825, 71.783, 54.665, 88.963, 74.807, 61.729, 92.1000, 97.1054, 102.1086, 103.1090, 71.783, 54.667, 89.982, 74.815; QTW, 226.1–2a.

50   j ts (7.148) records this event on the eighteenth day of the twelfth month. Here I adopt the record of tsjs (9.115). 51  Li Yi’s poem is also attributed to Song Zhiwen in qts (53.648), but according to wyyh (175.8b), it should belong to Li.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1185

I. 12/15, 3rd yr. (1/19/710)52 II. Visiting Bailu Temple. III. Your Majesty Visits the Bailu Temple (幸白鹿觀) / wulü / 10 and one unattached line. IV. Li Qiao, Li Yi, Shen Quanqi, Wu Pingyi, Zhao Yanzhao, Liu Xian, Cui Shi, Zhang Yue, Su Ting, Xu Yanbo, Zong Chuke. v. JLWGJ: GZJ, 7.19b, LS, 6.24b–25a; TSJS; QTS, 58.693, 92.995, 96.1031, 102.1084, 103.1088, 71.780–781, 54.662–663, 87.942, 73.800, 76.823–824. I. 12/18, 3rd yr. (1/22/710) II. Visiting the tomb of the First Emperor of Qin. III. Your Majesty Visits the Tomb of the First Emperor of Qin (幸秦始皇陵) / wulü / 1. IV. Emperor Zhongzong. V. TSJS, 1.9; QTS, 2.24. I. 12/22, 3rd yr. (1/26/710) II. Ascending Mount Li. III. Ascending Mount Li and Looking into the Distance (登驪山高頂寓目) / wulü / 9 // wujue / 1. IV. Emperor Zhongzong, Li Qiao, Li Yi, Wu Pingyi, Zhao Yanzhao, Liu Xian, Cui Shi, Zhang Yue, Su Ting, Yan Chaoyin. V. WYYH, 170.9–10a; TSJS, 9.115, 1.10; QTS, 2.24, 58.693, 92.993, 102.1083, 103.1088, 71.780, 54.662, 87.942, 73.800, 69.772 I. 12/30, 3rd yr. (1/26/710) II. Banqueting on the New Year’s Eve and Celebrating the completion of the first month of Princess Anle’s daughter.53 III. Attending a Banquet on the New Year’s Eve and Celebrating the Completion of the First Month of Princess Anle’s Daughter (歲夜安樂公主滿月 侍宴) / wulü / 1. IV. Shen Quanqi.54 V. WYYH, 169.9b; QTS, 96.1030.

52   j ts (7.148) records this event on the eighteenth day of the twelfth month. Here I adopt the record of tsjs (9.115). 53  The following New Year’s Eve had other events, so I put this event in this year. 54   q ts (62.732) ascribes another poem of the same title to Du Shenyan. But Du died the preceding year, so the ascription in qts may be a mistake.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1186

Jia

I. 1/1, 4th yr. (2/4/710) II. Bestowing cypress leaves to court officials. III. Bestowing Cypress Leaves to Court Officials on New Year’s Day (元日賜群 臣柏葉) / wujue / 3. Iv. Zhao Yanzhao, Li Yi, Wu Pingyi. V. WYYH, 172.11; TSJS, 9.115; QTS, 103.1090, 92.1000, 102.1085. I. 1/5, 4th yr. (2/8/710) II. Meeting the envoy of Tibet and watching the show of horse riding at Daming Hall. III. Moving the Imperial Insignia to Penglai Palace and Meeting the Envoy of Tibet to Watch the Show of Horse Riding at Daming Hall: Again Writing in the Style of Bailiang (移仗蓬萊宮御大明殿會吐蕃騎馬之戲因重爲柏 梁體) / heptasyllabic linked verse / 1. IV. Emperor Zhongzong, Empress Wei, Princess Changning, Princess Anle, Princess Taiping, Li Chongmao, Shangguan Wan’er, Cui Shi, Zheng Yin, Wu Pingyi, Yan Chaoyin, Dou Congyi, Zong Jinqing, Ming Xilie. V. JLWGJ: GZJ, 18b–19a, LS, 6.23b, CQSF, 46.3b; TSJS, 1.10; QTS, 2.24–25. I. 1/7, 4th yr. (2/10/710) II. Banqueting in Daming palace and bestowing silk ornaments of human figures to the courtiers. Visiting Pear Garden and watching a polo game. III. Attending a Banquet in Daming Palace, Your Majesty Bestows Silk Ornaments of Human Figures on Human Day (人日侍宴大明宮恩賜采鏤人 勝) / qilü / 12. Your Majesty Visits Pear Garden and Watches a Polo Game (幸梨圍觀打 毬) / wulü / 3. IV. Li Qiao, Zhao Yanzhao, Cui Riyong, Wei Yuandan, Ma Huaisu, Su Ting, Li Yi, Zheng Yin, Li Shi, Shen Quanqi, Liu Xian, Yan Chaoyin.55 Wu Pingyi, Shen Quanqi, Cui Shi. V. WYYH, 172.2b–4, 175.5a; TSJS, 9.115; QTS, 61.723, 103.1089, 93.559, 69.773, 93.1009, 73.804, 92.998, 106.1107, 70.777, 96.1041, 71.781, 69.771, 102.1083, 96.1030, 54.663.

55  In qts (69.771) Yan Chaoyin’s poem is entitled as “Your Majesty Visits Wanchun Palace in a Spring Day: by Imperial Command” 奉和聖制春日幸望春宫應㓡. I change the title according to wyyh (172.4b).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1187

I. 1/8, 4th yr. (2/11/710) II. Banqueting and bestowing flowers of cut silk to the courtiers on the day of Beginning of Spring. III. Attending a Banquet on the Day of Beginning of Spring, Flowers of Cut Silk are Sent out from the Inner Palace (立春日侍宴內出剪綵豪) / wulü / 7 // qilü / 1. IV. Li Qiao, Zhao Yanzhao, Shen Quanqi, Song Zhiwen, Liu Xian, Shangguan Wan’er, Su Ting, Wu Pingyi. V. JLWGJ: GZJ, 7.18a, LS, 6.25a, CQSF, 46.1; WYYH, 169.7; TSJS, 9.115; QTS, 58.691–692, 103.1087, 96.1029, 52.631, 71.779, 5.60, 73.798–799, 102.1085. I. 1/29, 4th yr. (3/4/710) II. Visiting the Chan River. III. Attending a Banquet beside the Chan River on the Last Day of the First Month (正月晦日侍宴滻水) / wulü / 3. IV. Zong Chuke, Zhang Yue, Shen Quanqi. V. TSJS, 9.115; QTS, 46.561, 87.944, 96.1029. I. 2/1, 4th yr. (3/5/710)56 II. Seeing off Princess Jincheng to Tibet. III. Seeing off Princess Jincheng to Marry the Ruler of Tibet (送金城公主適 西蕃) / wulü / 17. IV. Li Qiao, Cui Shi, Liu Xian, Zhang Yue, Xue Ji, Yan Chaoyin, Su Ting, Wei Yuandan, Xu Jian, Cui Riyong,57 Zheng Yin, Li Shi, Ma Huaisu, Wu Pingyi, Xu Yanbo, Tang Yuanzhe, Shen Quanqi. V. ZZTJ, 209.6639; WYYH, 176.11–13; TSJS, 12.180; QTS, 58.691, 54.662, 71.780, 87.942, 93.1006–1007, 69.771, 73.800, 69.773, 107.1112, 46.560, 106.1105, 70.776, 93.1008–1009, 102.1084, 76.823, 69.774, 96.1030–1031. I. 2/3, 4th yr. (3/7/710)58 II. Visiting Wang Guangfu’s 王光輔 residence. That night, Cen Xi held a tea party and discussed classics with other academicians. Wu Pingyi lectured

56   j ts (7.149) and zztj (209.6639) record this event from the twentieth day of the first month to the second day of the second month. Here I adopt the record of tsjs (9.115). 57  Cui Riyong’s poem is also attributed to Zhao Yanzhao in qts (103.1088), but according to wyyh (176.12a), it should belong to Cui. 58   j ts (7.149) records this event on the first day of the second month. Here I adopt the record of tsjs (9.115).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1188

Jia

on the Chunqiu. Cui Riyong asked Wu to be his teacher and presented a poem to Wu. III. To Wu Pingyi (贈武平一) / Chu song / 2 unattached lines. IV. Cui Riyong. V. JLWGJ: CQSF, 46.3b; JTS, 7.149; TSJS, 9.115; QTS, 93.560. I. 2/21, 4th yr. (3/25/710) II. Zhang Rendan came from Shuofang Defense Command; banqueting in Peach Garden. III. Attending a Banquet in Peach Garden and Writing on the Peach Blossoms (侍宴桃花園詠桃花) / qijue / 6. Iv. Li Qiao, Su Ting, Zhang Yue, Li Yi, Zhao Yanzhao, Xu Yanbo. v. JLWGJ: CQSF, 46.2b–3a; TSJS, 9.115, 10.146; QTS, 61.729, 74.815, 89.981, 92.1000, 103.1090, 76.827. I. 2/22, 4th yr. (3/26/710) II. Banqueting in Chengqing Palace; palace singers sang poems on peach blossoms. The emperor ordered the Court of Imperial Sacrifices to choose twenty poems and named them “Songs of the Peach Blossoms.” III. iV. V. JLWGJ: GZJ, 7.20a, LS, 6.23a–34b; TSJS, 9.115 I. 2nd month, 4th yr. (3/5–4/3/710)59 II. Visiting the royal garden. III. Visiting the Royal Garden and Enjoying the Rain on a Spring Day (春日游 苑喜雨) / wulü / 2. IV. Li Qiao, Li Yi.60 V. WYYH, 173.6b; QTS, 58.696, 92.994. I. 2/30, 4th yr. (4/3/710) II. Bestowing sweet gruel, silk balls, and engraved eggs to courtiers. III. Cold Food Day (寒食) / wupai / 1.

59  Li Yi’s poem says, “The thunder is heard early in the second month” 二月早聞雷 (qts, 92.994). The poems must have been written in the second month of that year or the preceding year. 60  Li Yi’s poem is also attributed to Li Qiao in qts (58.692), but according to wyyh (173.6b), it should belong to Li.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1189

A Study of the Jinglong Wenguan Ji

IV. Zhang Yue. V. JLWGJ: GZJ, 7.20a, LS, 6.24b, CQSF, 46.2b; WYYH, 172.6; QTS, 88.963. I. 3/1, 4th yr. (4/4/710)61 II. Visiting Pear Garden and ordering high officials and academicians to play tug-of-war. III. IV. V. JLWGJ: GZJ, 7.20b, LS, 6.24b, CQSF, 46.2b; JTS, 7.149; ZZTJ, 209.6640. I. 3/2, 4th yr. (4/5/710)62 II. Visiting Wangchun Palace. III. Your Majesty Visits Wangchun Palace on a Spring Day (春日幸望春宮) / qilü / 13. Iv. Cen Xi, Cui Shi, Zhang Yue, Liu Xian, Su Ting, Zheng Yin, Xue Ji, Wei Yuandan, Cui Riyong, Ma Huaisu, Li Shi, Li Yi, Shen Quanqi. V. WYYH, 174.1–3; QTS, 93.1005, 54.665, 87.960, 71.781, 73.804, 106.1107, 93.1007, 69.773, 93.559, 93.1009, 70.777, 92.998, 96.1041. I. 3/3, 4th yr. (4/6/710) II. Lustrating beside the Wei River and bestowing the courtiers willow wreaths. III. Lustrating beside the Wei River on the Third Day of the Third Month (上巳日紱禊渭濱) / qijue / 6. Iv. Wei Sili, Xu Yanbo, Liu Xian, Shen Quanqi, Li Yi, Zhang Yue. V. JLWGJ: GZJ, 7.20b, LS, 6.24, SF, 77.12a, CQSF, 46.2; JTS, 7.149; WYYH, 172.8b–9; 2375, 9.115; QTS, 91.988, 76.826, 71.783, 97.1054, 92.1000, 89.981. I. 3/5, 4th yr. (4/8/710) II. Banqueting in Peach Garden. III. IV. V. JTS, 7.149.

61   j ts (7.149) and zztj (209–6640) record this event on the twenty-ninth day of the second month. Here I adopt the record of tlwgj. 62  Cui Riyong’s poem says: “Tomorrow morning there will be a lustral ritual beside the Wei River” 渭浦明晨修禊事 (qts, 93.559). That was what happened the following day, the third day of the third month.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1190

Jia

I. 3/8, 4th yr. (4/11/710) II. Ordering the academicians to look for fine scenes; banqueting in Dou Xijie’s pavilion. III. The Officials of the Southern Department Visit Minister Dou’s Mountain Pavilion to Look for Flowers and Willows and Have a Banquet: A Preface (南省就竇尚書山亭尋花柳宴序) / prose / 1. IV. Zhang Yue. V. TSJS, 9.115, 12.177; QTW, 225.12. I. 3/11, 4th yr. (4/14/710) II. Visiting Shangguan Wan’er’s residence. III. Your Majesty Visits Palace Lady Shangguan’s Residence (幸上官昭容院) / wulü / 4. IV. Zheng Yin. V. WYYH, 175.12b–13a; TSJS, 9.115; QTS, 106.1105. I. 3/27, 4th yr. (4/30/710) II. Li Qiao went to Luoyang to worship the imperial ancestors; other academicians saw him off. III. Seeing off Lord Specially Advanced Li Qiao to Go to Eastern Capital to Worship the Imperial Ancestors (送特進李嶠入都祔廟) / wupai / 1. IV. Xu Yanbo. V. 7575, 9.115; QTS, 76.825. I. 3rd month, 4th yr. (4/4–5/3/710) II. Banqueting in Lotus Garden. III. Attending a Banquet in Lotus Garden on a Spring Day (春日侍宴芙蓉園) / wulü / 4. Iv. Li Yi, Su Ting, Song Zhiwen, Li Qiao.63 V. WYYH, 169.8; QTS, 92.993, 73.799, 52.631, 58.692. I. 4/1, 4th yr. (5/4/710) II. Visiting Princess Changning’s residence. III. Attending a Banquet in Princess Changning’s Eastern Residence (侍宴長 寧公主東莊) / wulü / 6 and 4 unattached lines.

63  Li Qiao’s poem is also attributed to Song Zhiwen in qts (52.643), but according to wyyh (169.8b), it should belong to Li.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji



1191

Visiting Flowing-Cup Pool in Princess Changning’s Residence (游長寧公 主流杯池) / triasyllabic quatrain / 2 // tetrasyllabic quatrain / 5 // wujue /

// 9 // wulü / 6 // qijue / 3. IV. Li Qiao, Cui Shi, Li Shi, Liu Xian,64 Li Yi, Zheng Yin, Xu Yanbo. Shangguan Wan’er. V. WYYH, 176.7b–8a; TSJS, 9.115, 1.9; QTS, 58.691, 54.662, 70.776, 71.780, 104.1092, 92.995, 106.1104, 5.61–63. I. 4/5, 4th yr. (5/8/710)65 II. Visiting Fanglin Garden and letting courtiers ride and pick cherries; banqueting in Grape Garden and playing palace music; bestowing cherries to courtiers. III. IV. V. JLWGJ: CQSF, 46.3a; JTS, 7.149; ZZTJ, 209.6640. I. 4/14, 4th yr. (5/17/710)66 II. Visiting Longqing Pool, making a silk building, banqueting with the courtiers, playing boat race, and visiting Dou Xijie’s residence. III. Playing Boat Race in Longqing Pool (隆慶池戲競渡) / qilü / 11. Your Majesty Visits Minister of Rites Dou Xijie’s Residence (幸禮部尚書 竇希玢宅) / wupai / 4. Iv. Xu Yanbo, Li Shi, Wu Pingyi, Liu Xian, Su Ting, Shen Quanqi, Wei Yuandan, Zhang Yue, Su Gui, Li Yi, Ma Huaisu. Li Yi, Shen Quanqi, Su Ting, Liu Xian.67 V. JTS; ZZTJ; WYYH, 176.2b–3a, 175.12; TSJS, 12.177; QTS, 76.826, 70.777–778, 102.1085, 71.781, 73.805, 96.1042, 69.773, 87.960–961, 46.562, 92.998, 93.1010, 92.998–999, 97.1045, 74.807, 71.782.

64  Liu Xian’s poem is also attributed to Xiao Zhizhong in qts (104.1092), but according to wyyh (176.8a), it should belong to Liu. 65   j ts (7.149) records this event on the sixth day of the fourth month. Here I adopt the record of zztj (209.6640). 66   t sjs (9.115) records this event on the sixth day of the fourth month. Here I adopt the record of jts (7.149) and zztj (209.6640–41). 67  Liu Xian’s poem is also attributed to Xiao Zhizhong in qts (104.1092), but according to wyyh (175.12a), it should belong to Liu.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1192

Jia

I. 5/29, 4th yr. (6/30/710)68 II. Banqueting in the palace; Chancellor of the Directorate of Education Zhu Qingming performed “Dance of Eight Obscene Actions.” Lu Cangyong told other academicians that Zhu dragged the five classics in the dust. III. Iv. V. JLWGJ: GZJ, 7.21a; ZZTJ, 209.6641; TSJS, 9.115. I. 2nd–4th yrs (708–710) II. Banqueting in the palace; ordering the courtiers and academicians to compose ci. III. Title lost / ci / 1. Iv. Cui Riyong. V. Benshi shi, 25b–26a; TSJS, 10.132–133; QTS, 869.9849. I. 2nd–4th yrs (708–710) II. Xue Ji was in charge of the court library of belles lettres, Ma Huaisu the classics, Shen Quanqi the histories, and Wu Pingyi philosophy. III. Iv. V. JLWGJ: GZJ, 7.19a, LS, 6.23b. I. 2nd–4th yrs (708–710) II. Playing the popular song Hesheng 合笙 in the inner palace. Wu Pingyi admonished against it. III. Iv. v. JLWGJ: GZJ, 7.21a, LS, 6.24b, SF, 77.12a. 2 The imperial patronage of literature and the establishment and activities of the literary institute in the Jinglong period had an important impact on the development of Tang poetry. The national institutes of the Tang began with the founding of the Xiuwen guan, established by Emperor Gaozu 高祖 in the first month of the fourth year 68   t sjs (9.115) records this event on the twenty-ninth day of the fourth month. Here I adopt the record of zztj (209.6641).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1193

of the Wude 武德 reign-period (621), and the Wenxue guan 文學館, founded by Prince Qin 秦王, Li Shimin 李世民, in the tenth month of the same year. The latter was closed when Li Shimin ascended the throne in the ninth year of Wude (626); the former’s name was altered to Hongwen guan 宏文館 in the same year, and it was again altered to Zhaowen guan 昭文館 in the first year of the Shenlong reign-period (705) and resumed the name Xiuwen guan in 706. Before Emperor Zhongzong, the institute served as an academic resource for the emperors in various fields, such as Confucian classics, history, literature, and politics. It is only in the reign of Emperor Zhongzong that it was consigned to the status of merely a literary salon. Emperor Zhongzong’s transformation of the nature of the institute was significant in the evolution of Tang poetry. He recruited to it nearly all the famous poets of the time. Under the imperial auspices, the academicians obtained high political positions in the court and had very close relationships with the royal family. From the above chronology, we can seen that in a mere two years, the emperor held numerous banquets in the palaces and visited religious temples, royal families, and ancient relics in the vicinity of the capital. The academicians attended the emperor and composed poems for all of these occasions. Zhang Yue later says in his “Tang Zhaorong Shangguan shi wenji xu” 唐昭容 上官氏文集序 (Preface to the Collection of Shangguan, the Palace Lady of the Tang Dynasty): The high officials had profound knowledge and felt shame if they were not of literary talent. When visiting palaces and touring rivers and mountains, the emperor composed poems, and the courtiers followed. This was not only because of the emperor’s love of literature, but also due to the palace lady’s help.69 The biography of Li Shi in the Xin Tangshu says, Whenever the emperor held a banquet or went on an excursion, only the chief ministers and the academicians could attend…. When the emperor was inspired by something, he would compose a poem, and all academicians would write poems in reply. They were envied by other people at that time.70

69   q tw, 225.18. 70   x ts, 202.5748.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1194

Jia

The poets were promoted to high status, and literature became an independent and significant branch in the political system. It was even regarded as more important than the Confucian classics, as the Zizhi tongjian tells us, “Thereupon all people of the country strove to excel in literature, whereas the faithful Confucian scholars could not get promotion any more.”71 In addition, the popularization of poetry among educated people was further realized during this period. It had been developing gradually since the Six Dynasties. In the Jinglong reign-period, because of the imperial patronage and the rise of the status of poets, “all people of the country strove to excel in literature.” In an earlier article focusing on the date on which poetry became a test subject in the jinshi examination and on the establishment of regulated verse,72 I suggest that when Shen Quanqi and Song Zhiwen were in charge of the examination in the second year of the Chang’an 長安 reign-period (702) and the second year of the Jinglong reign-period (the same year as the establishment of the literary institute) respectively, they likely made poetry one of the test subjects, fixed the tonal and rhetorical rules and the number of words and lines for regulated verse, and named this new genre lüshi. This event marks a significant landmark in the development of Tang poetry. Thereafter, all educated people who wanted to pass the jinshi examination and serve in the government had to train themselves as poets. As a result, the popularization of poetry among educated people quickly reached its summit in the High Tang. The Jinglong emperor and academicians contributed greatly to this accomplishment. The poetic examinations must have been under Zhongzong’s auspices, and the two examiners were academicians themselves. The rise of literary status and the further popularization of poetry in Jinglong reign-period supplied a solid foundation for the examination. Without this foundation, it is impossible to require all educated people to take poetic examinations. Moreover, because the composition of court poetry was competitive, the winner got a reward and was honored, while the loser got forfeit and was shamed, the establishment of a regulated form and a set of composing techniques became exigent. This might to some degree have resulted in the final establishment of the regulated patterns. It is not unreasonable that, when discussing the causes of the flourishing of Tang poetry, Hu Zhenheng 胡震亨 emphasizes the patronage of emperors, and he especially appreciates that of Emperor Zhongzong: 71   z ztj, 209.6622. 72  Jia Jinhua, “Jinshi shishi yu lüshi dingxing” 進士試詩與律詩定型, Wenxue yanjiu 文學 研究 2 (1992): 114–120.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1195

A Study of the Jinglong Wenguan Ji

The emperors’ preference of poetry created fascination for it among the people, thus the custom was transformed. The most important period was Emperor Zhongzong’s reign. At that time, many talented poets gathered in the literary institute, and also there was a gifted lady in the inner palace. Banquets and outings were held for composing poems; poets were rewarded and promoted.73 3 Among the extant poems of the Jinglong wenguan ji, except for two heptasyllabic linked poems, two triasyllabic poems, five tetrasyllabic poems, and five ci poems, all are recent-style poems. This fact indicates that recent-style poems had become dominating genre at that time. Moreover, a statistical analysis of these recent-style poems shows that most of them perfectly follow the tonal rules of regulated verse: Genre

Eight-line pentasyl­ labic verse

Total Poems 158 114 Tonally Perfect Poems 72 % (of Matured Regulated Poems)

Extended pentasyl­ labic verse

Pentasyl­ labic quatrain

Eight-line Heptasyl­ heptasyl­ labic labic verse quatrain

All recentstyle poems

45 29

13 9

69 54

49 33

334 239

64

69

78

67

72

73  Hu Zhenheng, Tangyin guiqian 唐音癸簽 (Taibei: Muduo, 1982) 27.281.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1196

Jia

Several observations can be drawn from this table. First, seventy-two percent of the recent-style poems are perfect in terms of the tonal rules. In my article “The ‘Pearl Scholars’ and the Final Establishment of Regulated Verse,”74 I demonstrate that only twenty-eight percent of the recent-style poems of the “pearl scholars” during the reign of Empress Wu (r. 685–705) are matured regulated verse in the extant Zhuying xueshi ji 珠英學士集 (Anthology of the Pearl Scholars). Therefore, I concluded that the tonal pattern of regulated verse had not been generally accepted by the “pearl scholars” around the year 701, when the Zhuying xueshi ji was compiled. However, only a few years later, the matured regulated poems of the Jinglong poets surprisingly increased by a percentage of forty-four. Moreover, it is noticeable that most of the violations of the regulated rules are committed in the poems written by some high-ranking officials, who were not the academicians, or, in other words, who were not “poets.” This fact, accompanied with the fact that there are almost no ancient-style poems in the extant Jinglong wenguan ji, further supports my suggestion that the rules of regulated verse matured between 702 and 708. Second, compared with the immediately preceding period of Empress Wu, extended pentasyllabic verse also increased greatly in numbers. The most noticeable phenomenon is that among the forty-five extended pentasyllabic regulated poems, there are thirtytwo twelve-line poems. Since the twelve-line pentasyllabic regulated verse is exactly the genre used for the jinshi examination, the Jinglong poets’ interest in this genre supports my suggestion that the establishment of poetry as one of the test subjects took place at that time. Third, there is also a large number of eight-line heptasyllabic regulated verse, which comprises twenty-one percent of the total of three hundred forty-three extant poems, and the mature percentage of this genre is even higher than eight-line pentasyllabic regulated.75 The Jinglong poets’ interest in this genre and their contribution to its maturity are also remarkable. The Jinglong academicians have five pieces of ci poem extant. These are among the earliest ci we can see today. All the five ci poems were written by imperial command. The Zizhi tongjian records: “The emperor banqueted with the courtiers and asked them to compose ‘Song of Water Wave.’ ”76 The Benshi shi says, 74  Jia Jinhua, “The ‘Pearl Scholars’ and the Final Establishment of Regulated Verse,” Tang Studies 14 (1996), pp. 8–16. 75  In his article (p. 80; see n. 13), Takagi Masakazu makes a statistical study of the Jinglong poets’ seventy eight-line heptasyllabic poems and concludes that about seventy percent of those poems are perfect regulated verses. My statistical result of the percentage of matured poems is a little higher than his. 76   z ztj, 209.6633. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1197

At that time, if a courtier wanted to wear a fish figure (pei yu 佩魚), he had to get the emperor’s special bestowal. At a banquet, the emperor asked the courtiers to write ci. [Cui] Riyong wrote [a piece of ci]… Then, Emperor Zhongzong bestowed on him a golden fish figure.77 The members of royal house were fond of folk ci very much. The Jinglong wenguan ji records: “Song of Hesheng” was performed in the inner palace. Its words were very simple and obscene. Wu Pingyi admonished, “The barbarian shamaness, prostitutes, and street kids gossip about the emotion and appearance of imperial concubines and princesses as well as the names and events of the nobility. They also make songs and dances about those things and call them Hesheng. It should not be performed in the palace.”78 The imperial preference and the Jinglong court poets’ practice in ci were very important in the transplantation of this new genre from low folk song to high literature. 4 Since most works of the Jinglong court poets were written by imperial command, these are not of high quality and therefore have seldom been noticed. However, a careful observation of the extant three hundred fifty pieces of shi, ci, fu, and prose demonstrates some distinctive aspects of the Jinglong poets’ literary concept and interesting features of their works. Along with the rise of the status of both poetry and poets in this period, the Jinglong court poets’ concept of literature changed. They took poetry as a political tool whereby they obtained favor from the emperor, empress, and other members of the royal house in order to gain high political position in the court. The Da Tang xinyu records: “In the Jinglong reign-period, Emperor Zhongzong visited Xingqing Pool. The courtiers attended the banquet and danced one after another. They sang ‘Song of Water Wave’ to beg for official positions.”79 The biography of Cui Riyong in the Xin Tangshu says, “The emperor banqueted in the inner palace. While they were drinking to the full, Cui Riyong performed

77   Benshi shi, 25b–26a. 78   g zj, 7.21; ls, 6.24a; sf, 77.12a. 79   Da Tang xinyu, 3.70. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1198

Jia

‘Dance of Water Wave’ and begged to be an academician. The emperor appointed him immediately.”80 Cui Shi says: I, your subject [Dongfang] Shuo, am so lucky 臣朔真何幸 To attend Emperor of Han Wudi for outings frequently.81 常陪漢武游 Zhao Yanzhao says: I hope to attend Queen Mother of the West forever; As an attendant I feel ashamed without Dong[fang]  Shuo’s talent.82

常年願奉西王母 近侍慚無東朔才

They were voluntarily playing the court comedy role of Dongfang Shuo 東方朔 in the reign of Emperor Wudi of the Han dynasty (r. 141–87 bc). However, their role was a little different from Dongfang Shuo’s. Dongfang was only a low court comic subject, whereas the Jinglong court poets held good positions. As a result, many Jinglong poems are odes to the emperor and his family. Emperor Zhongzong was a famous fatuous and uxorious ruler. He was controlled by his wife, Empress Wei, and even by his sister, Princess Taiping, and his two daughters, Princess Changning and Anle. These four powerful women were lawless and stopped at no evil. However, in their poems, the court poets poured abundant praise on them. These poems, though elegant and smooth in style, are very superficial and sham. The Jinglong court poets also viewed poetry as a means of entertainment. Most of their poems were composed in feasts and holidays to entertain the royal family as well as the courtiers themselves. The Jinglong wenguan ji says: On the seventh day of the first month in the third year of the Jinglong reign-period, when the emperor ascended Qinghui Pavilion, it was snowing…. He asked the academicians to chant poems. That day everybody was very happy.83 In the spring of the fourth year of the Jinglong reignperiod, the emperor banqueted in Peach Garden. All courtiers attended. Academician Li Qiao and others presented poems on peach blossoms.

80   x ts, 121.4330. 81  Cui Shi, “Your Majesty Visits Bailu Temple: by Imperial Command.” qts, 54.662–63. 82  Zhao Yanzhao, “Attending a Banquet in Peach Garden: On Peach Blossoms.” qts, 103.1090. 83   c qsf, 46.1a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1199

A Study of the Jinglong Wenguan Ji

The emperor ordered the palace maids to sing these poems. The poems were lucid and graceful, and the songs were sweet and lovely.84 Shen Quanqi composed “Song of Water Wave.” … It made the emperor laugh heartily.85 Nearly all the ci poems of “Water Wave” are very humorous. The one written by an anonymous court actor reads: Wave of water, basket of bamboo, It is a good character to fear one’s wife. Outside there is only Pei Tan, Inside no one outdoes the respected Li.86

回波爾時栲栳 怕婦也是大好 外邊只有裴談 內里無過李老

This ci jokes about Emperor Zhongzong’s fear of his wife, Empress Wei. Its style is light and bantering, and vernacular is properly used. This style hardly appears in other kinds of poetry during the Early Tang, but in Emperor Zhongzong’s court it is not the only example. Another untitled ci by Cui Riyong goes: You should know the mouse in the Censorate, Which jumped up to the shrine on the wall. He knocked the lamp oil over and dirtied Zhang the Fifth; Again he gnawed Han the Third’s belt to get even with him. It’s not a boast that he is equal with the ministers. The emperor would be sure to bestow on him the golden turtle, And to sell the cat in order to reward him.87

台中鼠子直須諳 信足跳梁上壁龕 倚翻燈脂污張五 還來嚙帶報韓三 莫浪語，直王相 大家必若賜金龜 賣卻貓見相報賞

According to the Benshi shi, Cui Riyong was then vice censor-in-chief and wrote this ci to beg for a fish figure.88 Therefore, tai should refer to yushi tai 御史台, the censorate. This ci is notable for its use of metaphor, vernacular, and humorous style. All the ci poems written by the Jinglong poets carry this

84   c qsf, 46.2b–3a. 85   c qsf, 7.18b. 86   q ts, 890.10049. 87   q ts, 869.9849. 88   Benshi std, 25b–26a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1200

Jia

stylistic trace of folk song. The reason for this may be that these ci poems were among the earliest pieces of the transplantation of the popular ci form. Another stylistic feature of their poems is femininity. As we have seen, Emperor Zhongzong was under the control of his wife, sister, and daughters. Many banquets were held in the inner palace and the princesses’ residence. Those powerful women also attended most of the outings.89 Moreover, the court poets’ poems were commented on and evaluated by another powerful court woman, Shangguan Wan’er, Emperor Zhongzong’s concubine. The biography of Shangguan Wan’er in the Xin Tangshu says: Wan’er urged the emperor to expand the institute, add positions of academician, and appoint high officials and famous scholars to be academicians. The emperor often held feasts and composed poems, and the courtiers wrote in reply. Wan’er always wrote for the benefit of the emperor, the empress, and princesses Changning and Anle. She wrote several poems at the same time and her words were very florid and ingenious. She also evaluated the courtiers’ poems and granted them rewards. Therefore, all officials of the court turned their attention to poetry. Although most poets of the time presented a flashy style, their poems were still readable. This is only due to Wan’er’s influence.90 Thus, the establishment of the Jinglong literary salon came from Wan’er’s suggestion. Since she wrote for the benefit of the emperor and other members of the royal house, all the “by imperial command” poems of the courtiers were in fact written in reply to her poems. She also commented on all these poems, hence the court poets must have tried to follow her taste. On holidays, the academicians were often awarded some small gifts, such as cypress leaves on New Year’s day, silk ornaments of human figures on Human 89  For example, Shao Sheng’s “Your Majesty Visits Princess Taiping’s South Residence in Early Spring: by Imperial Command” (qts, 69.774) says: 二聖忽從鸞殿幸 雙仙正下鳳樓迎 The two sages come from the simurgh palace suddenly; The two immortals are going down to the phoenix residence to welcome them. “The two sages” refer to both the emperor and the empress. Zhang Yue’s “Dongshan Ji” 東山記 (Record of the East Mountain, a preface to the poems written during a visit to Wei Sili’s residence) says: “The emperor composed poems, and the empress, princesses, and palace lady followed” (qtw, 226.1a). 90   x ts, 76.3488.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1201

A Study of the Jinglong Wenguan Ji

Day, cut silk flowers on Beginning of Spring, sweet gruel, silk balls, engraved eggs on Cold Food Day, and so forth. These kinds of small gifts and feminine ornaments might have been bestowed by the female rulers. The court poets, therefore, had to describe those trivial and feminine objects to express their gratitude. The style of those poems is inevitably soft and gorgeous. Let us look at Song Zhiwen’s “Attending a Banquet on the day of Beginning of Spring, Flowers of Cut Silk are Sent out from the Inner Palace”: New apricot flowers decorate the golden pavilion; Filigree plum flowers are displayed in the jade banquet. These have not yet been seen in the human realm, But first blossom in Heaven above all of a sudden. Butterflies surround and stop on the fragrant threads; Bees love and return for the bright-colored pollen. Spring comes earlier this year, Only because the [palace maids’] scissors hasten it.91

金閣妝新杏 瓊筵弄綺梅 人間都未識 天上忽先開 蝶繞香絲住 蜂憐艷粉回 今年春色早 應爲剪刀催

In the opening couplet, “golden pavilion” and “jade banquet” refer to the theme “imperial banquet” in the title, and “apricot flower” and “plum flower” refer to the theme “cut silk flower.” The middle four lines further develop these two themes. The second couplet plays on the traditional analogue: the imperial palace is like Heaven. The third couplet, taking butterflies and bees as media, describes the charming odor, color, and shape of the cut silk flowers. Both couplets play between the truth and untruth of artwork. The last couplet uses an ingenious empathy to hint at the theme “Beginning of Spring”: spring does not come naturally but the palace maids’ scissors hasten or bring it out. The poet is skillful in using both conventional and original metaphors and in describing objects. His touch is minute and florid, and his style is of feminine delicacy. Since regulated verse had matured by that time, the poems of the Jinglong court poets are very harmonious and smooth in tone and syntax, especially those written by the academicians. The opening and ending couplets of their poems may sometimes be dull and hackneyed because these are often used to narrate the events and occasions of imperial banquets and outings, and to express the poets’ gratitude; but the middle couplets are often quite original in images, well matched in parallelism, and complex in syntax. We can pick up many graceful couplets from their poems:

91   q ts, 52.631.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1202

Jia

Flowers are blossoming with red charm, Leaves are fresh towards green sense.92

花隨紅意發

OO O X X 葉就綠情新

XXXOO

The tonal pattern and verbal antithesis are perfect. The simple meaning of red flower and green leaf is complicated by the ingenious and anthropomorphized words of “red charm” and “green sense” and by the unusual syntax of putting verbs in the second positions of the couplet. The neat antithesis of the two lines not only depicts a full picture of the colorful and dynamic spring world, but also conveys the joyful feeling of the poet. Along the iced Winding Pool the colored moss looks like liquid, Among the snow in the imperial garden the plum flowers are fragrant and charming.93

曲池苔色冰前液

XOOXOOX 上苑梅香雪里嬌

XXOOXXO

This couplet vividly describes the interwoven scene of spring and winter. The metaphor of the light-green moss in early spring like liquid is new and clever. The force of winter makes the fluid water solid, but the force of spring makes the solid moss fluid, hence it hints at the germination and energy of life at the early spring. 5 From this study of the Jinglong wenguan ji, we can see that the Jinglong reignperiod was a special transition-period in the development of Tang poetry. The imperial patronage raised the status of both poetry and poets. The popularization of poetry was further realized. The maturing of regulated verse and the establishment of poetry as a test subject in the jinshi examination may have happened in this period. In only about two years, many court literary activities were held, and hundreds of poems were composed. The Jinglong court poets

92  Zhao Yanzhao, “Attending a Banquet on the day of Beginning of Spring: Flowers of Cut Silk are Sent out from the Inner Palace,” qts, 103.1087. I use “O” for a level tone and “X” for an oblique tone in the transcriptions. 93  Cui Riyong, “Re-Opening a Banquet in Daming Palace on Human Day, Your Majesty Bestows Silk Ornaments of Human Figures: by Imperial Command,” qts, 93.559.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

A Study of the Jinglong Wenguan Ji

1203

displayed their own styles and features and made some peculiar contributions to Tang literature, even though their works are seldom great. Zhang Yue, a younger poet among the Jinglong academicians, later gave high praise to some of his fellow academicians: The literary works of Li Qiao, … Xue Ji, and Song Zhiwen are just like good gold and fine jade and can be employed everywhere…. Yan Chaoyin’s works are like well dressed beauties, embroidered silk clothes, and the singers of the Yan area and the dancers of the Zhao area. They make the audience forget their sorrow.94 The literary works of Brother Shen the Third (Quanqi) are really the foremost.95 Zhang and some other younger Jinglong academicians, such as Su Ting and Xu Jian, lived well into the early High Tang. Like their old colleagues in the Jinglong period, they occupied high political positions and continued to support literary activities and promote young poets. As some scholars have mentioned, many famous High Tang poets, such as Zhang Jiuling 張九齡 (678–740), He Zhizhang 賀知章 (659–744), Wang Han 王翰, and Wang Wan 王灣, were sponsored and promoted by Zhang Yue. It is not inappropriate to say that the Jinglong period made good preparations for and had an important impact on the advent of the High Tang, the zenith of Chinese poetry.

94   Da Tang xinyu, 8.146. 95   Sui Tang jiahua, 2.24.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old-Style fu of Han Yu David R. Knechtges It1 is a commonplace of scholarship on Chinese literature that each dynasty has its own literature. The Han was the golden age of the fu, the Tang saw the flowering of the shi, the Song brought the blossoming of the ci, and the Yuan produced the qu. Eager to study the prime blossoms of Chinese literature, scholars have concentrated their studies on the “golden ages.” However, with so much attention devoted to the great literary periods, histories of genres often stop with the golden ages and say virtually nothing about subsequent developments in the genre. A good example of a genre the later history of which scholars generally have neglected is the fu. The fu was the dominant literary form in the Han dynasty, and along with the shi was a common form of literary expression in the Six Dynasties and Tang periods. Modern scholarship has focused particularly on the fu of the Han period, and in the past few years a number of important books devoted to the study of the Han fu have been published.2 We also are beginning to see some good work being done on the fu of the Wei-Jin-Nanbeichao period.3 However, very little attention has been paid to the fu of the Tang. The prevailing view of the Tang fu is that most fu composition consisted of “regulated fu” (lü fu).4 Thus, the fu anthologist Zhu Yao 祝堯 (jinshi 1318), after examining thousands of fu in the collected works of Tang writers as well as Source: “The Old-Style fu of Han Yu,” T’ang Studies 13 (1995): 51–80. 1  An abridged Chinese version of this article has been published in Han Yu xueshu taolun hui zuzhi weiyuanhui 韓愈學術討論會組織委員會 (Guangzhou: Guangdong renmin chubanshe, 1988), 174–89. For other studies of Han Yu’s fu see Yamasaki Jun’ichi 山崎純一, “Kan Yu no kofu ni tsuite” 韓愈の古賦について Tōyō bungaku kenkyū 19 (1971): 22–34; Part ii, Tōyō bungaku kenkyū 20 (1972): 1–13; Huang Ting 黃挺, “Lun Han Yu fu” 論韓愈賦, Hanshan shizhuan xuebao 13.3 (1984): 44–48; Gong Kechang 龔克昌, “Lüe lun Han Yu cifu” 略論韓愈 辭賦, Wen shi zhe (1992: 3): 31–37. 2  For a survey of recent scholarship on the fu, see He Xinwen 何新文, Zhongguo fu lun shigao 中國賦諭史稿 (Beijing: Kaiming chubanshe, 1993). 3  See Cao Daoheng 曹道衡, Han Wei Liuchao cifu 漢魏六朝辭賦 (Shanghai: Shanghai guji chubanshe, 1989); Liao Guodong, Wei Jin yongwu fu yanjiu (Taipei: Wen shi zhe chubanshe, 1990). 4  For an account of the regulated fu, see Suzuki Torao 鈴木虎雄, Fushi taiyō 賦史大要 (Tokyo: Fusambō, 1936), 170–244.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_038

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old-Style Fu of Han Yu

1205

the Wenyuan yinghua 文苑英華, concluded that “regulated fu were many and ancient style fu were few” (lü duo er gu shao 律多而古少).5 The reason for the popularity of the regulated fu in the Tang is well known: it was required for the jinshi examination. The Qing dynasty fu authority Li Tiaoyuan 李調元 (1734–1803) aptly characterizes the extent to which the regulated fu became an important literary form by the middle of the Tang: At the beginning of the Tang, the jinshi were tested by the bureau for the examination of merit, and most important was the examination on excerpts from the classics. They also could substitute admonitions, disquisitions, memorials, and encomia, but during the period when men were not tested in poetry and rhyme-prose, those who specialized in composing regulated fu were still small in number. During the Dali (766–780) and Zhenyuan (785–805) periods, the writing of regulated fu gradually became the fashion. By the eighth year of Dahe (834), the composition exams entirely consisted of poetry and rhyme-prose, and famous specialists in the fu multifariously vied with one another. Li Cheng 李程 (ca. 765–ca. 841) and Wang Qi 王起 (760–847) lay claim to the greatest fame in their time. Jiang Fang 蔣防 (fl. 820–824) and Xie Guan 謝觀 (n.d.) were as flank horses following the lead. As a whole they adhered to the standards of lucid freshness and classical elegance. Of those that galloped on their flank and dashed in separate directions, Yuan Zhen 元稹 (779–831) and Bai Juyi 白居易 (772–846) were most venerable.6 Although the regulated fu was the preferred type of fu for most of the Tang, the gu fu 古賦, or “ancient style fu,” did not disappear. Such fu are in fact found in the collected works of some of the most distinguished Tang writers. Among the early Tang poets, Yang Jiong 楊炯 (650–ca. 695) has a long exposition of astral lore, the “Huntian fu” 渾天賦 (“Fu on the Enveloping Sky”),7 which is rivaled only by the “Tianwen daxiang fu” 天文大象賦 (“Fu on the Grand Simulacra of Astronomy”) by Li Bo 李播 of the Sui.8 Lu Zhaolin 盧照鄰 (ca. 634–ca. 684) is the author of a Taoist philosophical essay on a sick pear 5  See Gu fu bian ti 古賦辯體, in Siku quanshu zhenben (Taipei: Shangwu yinshuguan, 1975), 7.1a. 6  See Fu hua 賦話 (Taipei: Shijie shuju, 1961), 1.3. 7  See Xu Mingxia 徐明霞, ed., Yang Joing ji 楊炯集 (Beijing: Zhonghua shuju, 1980), 1.1–7. 8  This fu is sometimes wrongly credited to Zhang Heng 張衡 (ad 78–139) of the Later Han. For an edited text and commentary, see Sun Xingyan 孫星衍, ed., Xu Guwen yuan 續古文苑 (Taipei: Shijie shuju, 1973), 3.155–206.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1206

Knechtges

tree,9 as well as a highly affecting piece on his illness, written in the style of the Chu ci.10 The great shi poet Li Bai 李白 (701–762) was a highly skilled fu writer as attested by his powerful ode on that noble bird of Taoist myth, the peng 鵬,11 and his rousing fu on the Tang imperial hunt, which he wrote deliberately to outdo the hunting fu of Sima Xiangru 司馬相如 (179–117 bc) and Yang Xiong 揚 雄 (53 bc–ad 18).12 Even in the period when the regulated fu was at its zenith of popularity, some writers showed a clear preference for the gu fu. This is particularly true of the guwen advocates, notably Han Yu 韓愈 (768–824) and Liu Zongyuan 柳宗元 (773–819). Zhu Yao, who had a declared predilection for the ancient style fu, praised Han and Liu as the foremost ancient style fu writers of the Tang. In Zhu’s estimation, they even surpassed the great Li Bai: Li Taibo was a genius of outstanding talent, and the ancient style fu that he wrote are passably good, but even though the creeper vines of the parallel-style have been removed, the roots of the regulated fu still remain. Although there are brilliant sparks when he puts his brush to paper, and he occasionally writes unusual words, his works are merely Six Dynasties fu. Only the ancient style of Han and Liu, which take the Sao as their standard, soar far beyond the parallel-style and regulated fu…. Amongst the ancient-style pieces of the Tang fu, none is more faithful to the ancientstyle than their fu.13 Zhu Yao’s anthology of Tang dynasty gu fu is quite small. It includes one piece by Luo Binwang 路賓王 (ca. 640–684), seven by Li Bai, two each for Han Yu and Liu Zongyuan, and one fu by Du Mu 杜牧 (803–852).14 A much larger selection of Tang dynasty gu fu is contained in the Song anthology, the (Tang) wencui 9  See “Bing lishu fu” 病梨樹賦 in Xu Mingxia, ed., Lu Zhaolin ji 盧照鄰集 (Beijing: Zhonghua shuju, 1980), 1.6–7. For an English translation, see Stephen Owen, “The Barren Tree: From Yü Hsin to Han Yü,” Chinese Literature: Essays, Articles and Reviews 1.2 (1979): 16–62. 10  This is his “Shi ji wen” 釋疾文 (“Essay Dispelling Illness”). See Lu Zhaolin ji 5.59–68. 11  This is the “Da peng fu” 大鵬賦 (“Fu on the Great Peng-bird”); see Wang Qi 王琦 (1696– 1774), ed. and comm., Li Taibo quannji 李太白全集 (Hong Kong: Zhonghua shuju, 1972), 1.1a–6a. For an excellent translation, see Paul W. Kroll, “Li Po’s Rhapsody on the Great P’eng-bird,” Journal of Chinese Religions 12 (1984): 1–17. 12  See “Da lie fu” 大獵賦 (“Fu on the Great Hunt”) in Li Taibo quanji 1.28b–43b. This fu has been translated into German by Erwin von Zach; see “Lit’aipo’s poetische Werke: Buch I,” Asia Major 3 (1926): 452–66. 13  See Gu fu bian ti 7.2b. 14  This is Du Mu’s famous “Ebang gong fu” 阿房宮賦 (“Fu on the Ebang Palace”).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old-Style Fu of Han Yu

1207

唐文粹, compiled by Yao Xuan 姚鉉 (968–1020). Yao, who had a strong prefer-

ence for ancient-style prose and poetry, included fifty-five fu in the first nine juan of the anthology.15 Although Yao nowhere explains the term gu fu, it is clear from remarks in his preface that he intended his selection as an alternative to the collections of regulated fu that were popular in his time: For the fu, there are such collections as the Jia fu 甲賦, Fu xuan 賦選, and Gui xiang 桂香, which contain mostly regulated fu and rarely touch on the Way of antiquity. This probably is because they were intended to provide quick reference for the newly arrived younger scholars who avidly sought to make names for themselves in the examinations.16

The model for the gu fu was the fu of the Han dynasty, which actually encompassed a broad range of forms and subjects. There were first of all the grand, epideictic court pieces such as Sima Xiangru’s “Shanglin fu” 上林賦 (“Fu on the Imperial Park”) and Yang Xiong’s “Ganquan fu” 甘泉賦 (“Fu on the Sweet Springs Palace”), in which the authors display (puchen 鋪陳) their vast learning and rich vocabulary. There were also the so-called xiao fu 小賦, or “lesser fu,” which usually were short descriptive yongwu 詠物 compositions.17 Finally, there was the fu of “the scholar who has failed in his ambition” (xianren shi zhi zhi fu 賢人失志之賦), which in contrast to the court fu, usually are expressions of the poet’s sorrows and despairs. As an ardent devotee of ancient literature (guwen 古文), Han Yu certainly must have had a profound knowledge of the ancient fu. Scholars have in fact observed that some of his shi exhibit features peculiar to the fu.18 Among his extant writings, there are five pieces titled fu. One fu, the “Ming shui fu” 明水賦 (“Fu on Bright Water”), is a regulated fu Han wrote for the jinshi examination of Zhenyuan 8 (792).19 Han Yu has four ancient-style fu, all included in juan 1 15  The selection of fifty-five pieces probably was intended to emulate the Wen xuan, which also has fifty-five fu. 16  See Tang wencui, Sibu congkan, “Preface,” p. 3a. 17  See Wang Qisun 王芑孫 (1755–1818), Du fu zhiyan 讀賦卮言, in Ho P’ei-hsiung 何沛雄, ed., Fu hua liuzhong 賦話六種 (Hong Kong: Wanyou tushugongsi, 1975), 10–11. 18  See Gu Yisheng 顧易生, “Shi tan Han Yu de shang qi ji Han wen yu cifu pianwen de guanxi” 試談韓愈的尚奇及韓文與辭賦駢文的關係, Wenxue yichan zengkan 10 (1962): 66–72; Charles Hartman, Han Yü and the Tang Search for Unity (Princeton: Princeton Univ. Press, 1986), 271–74. 19  This fu was not part of Han Yu’s original collection. It now is the first piece of the wai ji 外集 (supplemental collection). See Ma Tongbo 馬通伯, ed. and comm., Han Changli wenji jiaozhu 韓昌黎文集校注 (Hong Kong: Zhonghua shuju, 1972), 381–82.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1208

Knechtges

of his collected works. Han wrote these pieces in the nine-year period between Zhenyuan 11 (795) and Zhenyuan 20 (804), when he experienced much frustration in his official career. They are all variations on the theme of the xianren shi zhi. This theme, which originated in the “Sao” poems of Qu Yuan and the “Gui shi” 佹詩 of Xunzi, by Han times developed into a special subgenre of the fu.20 In contrast to the epideictic court poems, the xianren shi zhi fu are more personal, and at least make a good pretense of expressing genuine feeling and emotion. The best known of the Han dynasty xianren shi zhi fu are the “Bei shi buyu fu” 悲士不遇賦 (“Fu Lamenting the Neglected Scholar”) and the “Shi buyu fu” 士不遇賦 (“Fu on the Neglected Scholar”) attributed to Sima Qian 司馬遷 (ca. 145–90 bc) and Dong Zhongshu 董仲舒 (179–104 bc) respectively.21 In both of these pieces, the authors complain about the unlucky fate of the man of talent who fails to obtain proper recognition in his time: How sad the scholar so untimely born! Filled with shame, he abides alone watching his shadow. . . . . . . . . . . . . . . . . . . . .  Though his deeds are grand, he is unrenowned; All he has is ability that he cannot display. Failure and success—how easily they are confused; Truly beauty and ugliness are hard to distinguish. As time goes on and on, swiftly passing, He is held in check, unable to stretch out. (“Bei shi buyu”) Alas! Alas! How far, how distant! The proper time, how slow it comes! How swiftly it departs! 20  The locus classicus for the term xianren shi zhi as applied to the fu is in the “Shi fu lue” 詩賦略 (“Summary of Songs and Rhapsodies”) of Liu Xin 劉歆 (ob. ad 23), an abridged form of which has been preserved in the “Yiwen zhi” 藝文志 of the Han shu. See Han shu 漢書 (Beijing: Zhonghua shuju, 1962), 30.1756. For a survey of the xianren shizhi fu in the Han period, see Nakajima, Fu no seiritsu, 418–526; and Cao Shujuan 曹淑娟, “Lun Handai fu zhi xie wu yan zhi chuantong” 論漢代賦之寫物言志傅統, Guoli Taiwan Shifan daxue guowen yanjiusuo jikan 27 (1983): 48–72. 21  The text of “Bei shi buyu fu” is found in Ouyang Xun 歐陽詢 (557–641), comp., Yiwen leiju 藝文類聚 (Beijing: Zhonghua shuju, 1965), 30.541. The “Shi buyu fu” is contained in Guwen yuan 古文苑, Dainan ge congshu, 1.13a–14a. The authenticity of either piece is not certain.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1209

The Old-Style Fu of Han Yu

Those who bend their will to follow others Are not my kind. Upright, awaiting the proper time, I shall approach the grave. If on and on, I accord with the times, How will they ever be awakened? . . . . . . . . . . . . . . . . . . . .  I was born not to meet the splendrous glory of the Three Dynasties, But meet instead the decadent customs of their final reigns. (“Shi buyu”) Besides lamenting time’s fate, some of the poets who write on the xianren shi zhi theme use the fu as a means of self-consolation and introspection. Beginning at the end of the Former Han, some of these fu are even autobiographical. A notable example is Liu Xin’s “Sui chu fu” 遂初賦 (“Fu on Fulfilling My Original Resolve”).22 Liu Xin wrote this fu around 6 bc after being forced to leave the court for submitting his “Letter Berating the Taichang Erudites” (“Yishu rang taichang boshi” 移書太常博士), in which he advocated the official recognition of the guwen texts, including the controversial Zuo zhuan. From the remote frontier post of Wuyuan 五原 (northwest of modern Baotou, Inner Mongolia), he wrote the “Sui chu fu” recounting his journey to Wuyuan, which took him through numerous historic sites of the ancient state of Jin. Rather than complain about his fate, Liu ends his fu on an optimistic note, and declares his intention to make the best of his situation and above all preserve his integrity. Han Yu must have been intimately familiar with the xianren shi zhi tradition. The Chuci of course was well known to him and obviously provided him much inspiration. He also must have known such deeply introspective pieces as Ban Gu’s “You tong fu” 幽通賦 (“Fu on Communicating with the Hidden”) and Zhang Heng’s “Si xuan fu” 思玄賦 (Fu on Contemplating the Mystery”), which were easily accessible to Han in the Wen xuan.23 Both fu begin with expressions of despair and frustration, but end with strong assertions of confidence and optimism. It is well known that Han Yu encountered much failure and frustration throughout much of his life. Both his mother and father died when he was an infant. About age eleven he saw his brother, Han Hui 韓會 (739–ca. 780), 22  For the text see Yiwen leiju 27.490. There is a longer text in the Guwen yuan 2.11a–4b. 23  For the “You tong fu” see Wen xuan (Taipei: Zhengzhong shuju, 1971), 14.11a–20a; for “Si xuan fu,” Wen xuan, 15.1a–19b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1210

Knechtges

who raised him, die as an exile in the remote area of Shaozhou 韶州 (modern Shaoguan 韶關, Guangdong). After burying Han Hui in the ancestral plot in Heyang 河陽 (south of modern Meng 孟 xian, Henan), he and his sisterin-law had to return south to Xuancheng 宣城 to escape the threat of insurrections and unrest that had spread across much of north China. In spite of his great learning and literary skill, Han Yu attempted the jinshi examinations three times before he finally passed in Zhenyuan 8 (792). His three attempts at the boxue hongci 博學宏辭 examination in Zhenyuan 9–11 (793–94) also were unsuccessful. Utterly frustrated in his attempt to obtain a position, in the early part of Zhenyuan 11 (795) he wrote three letters to the ministers requesting employment.24 When he obtained no response, Han left the capital in May 795 for his home in Heyang. It was at this time that Han Yu wrote his first fu, “Gan er niao fu” 感二鳥賦 (“Feelings Roused by Seeing Two Birds”).25 Han Yu’s prose preface explains the circumstances of its composition: In the eleventh year of Zhenyuan (795), on xuchen of the fifth month (May 24), I returned east. On guiyou (May 29), I departed from Tong Pass to rest on the southern bank of the Yellow River.26 When I first left the capital, I sighed the sorrow of not meeting a timely opportunity. On my journey I saw men heading west with an albino crow and albino mynah in a cage. They shouted on the road: “He who holds a certain office of a certain territory sends an emissary to make a presentation to the Son of Heaven.” Those traveling east and west all got out of the way, and no one dared to gaze directly upon them. I secretly grieved to myself: I am fortunate to have been born at a time when the empire is without conflict, and carrying on the work of my forebears I did not engage in the toil of shield and lance, plow and plowshare, attack and defense, plowing and reaping, but instead I have read and composed from the age of seven until now, twenty-two years in all. In my conduct I have never dared do anything that would disgrace the moral way, and while dwelling in leisure, I have pondered matters past and present, but I have barely remembered one or two of the most important of them. When I was 24  For texts of the letters, see Ma Tongbo, Han Changli wenji jiaozhu, 3.89–95. 25  For the text see Ma Tongbo, Han Changli wenji jiaozhu, 1.1–3; Wei Zhongju 魏仲舉 (ob. post-1200), Wubaijia zhu Changli wenji 五百家注昌黎文集, Siku quanshu zhenben (Taipei: Shangwu yinshuguan, 1973), 1.1a–4b. There are also notes in Tong Dide 童第德, Han Yu wen xuan 韓愈文選 (Beijing: Renmin wenxue chubanshe, 1980), 1–5. 26  Tong Pass was located east of modern Huayin, south of the Yellow River.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1211

The Old-Style Fu of Han Yu

selected for presentation to the authorities,27 to pass or fail along with a hundred-odd others, I never had my name mentioned in the recommendation letters, or even ranked as a petty official at court to gaze upon the Son of Heaven’s lustrous brilliance. Now these birds, merely by virtue of the uniqueness of their plumage, not because they have moral virtue and wise plans, or can offer advice and counsel, or assist in teaching and transformation, have been privileged to be selected and promoted, recommended and presented, and be gloriously honored to this extent. Thus, I have composed a fu to commiserate with myself, and also to make clear that a man who encounters a timely opportunity, though his goodness be small, shall achieve success; but for him who does not encounter a timely opportunity, his accumulated goodness will not obtain him acceptance anywhere. The fu proper is forty lines long and is written mostly in modified sao-style (without the xi) six-syllable lines. The fu begins with two rhymed couplets in which Han unambiguously states the desperate situation he finds himself in: he cannot make a living for himself in the capital, and he has no choice but to go home. The first line is a four-syllable question, which is an unusual opening for a fu. The second line is the antithesis of Lun yu 5/19, which says of Ji Wenzi 季文子that “he reflected three times before acting.” Oh, where shall I return? I shall act first and then reflect. Truly, I cannot live on; Wherever I find sustenance, I shall go.28 In the following six couplets, Han first briefly recounts his journey from Chang’an across Tong Pass, to the southern bank of the Yellow River, where he met the tribute procession carrying the albino birds to be presented to the emperor. He is indignant and resentful that such ignorant creatures, merely because of their “external adornment,” receive more favor and good fortune than he himself has obtained: Departing the capital gates, I gallop east, And encounter the blazing light of the bright sun.

27  Han here refers to the boxue hongci examinations he took in 793, 794, and 795. 28  By “sustenance,” Han Yu means employment.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1212

Knechtges

From time to time I look back and shed tears,29 As I ponder how truly long is the road west.30 Crossing Tong Pass, I sit down and rest, And gaze at the rushing fury of the yellow stream. I am moved by two ignorant birds, Who have just received imperial favor and entered unto good fortune. Thinking on the vast difference between advancement and withdrawal,31 Increases the vexation in my breast. What fine qualities lie in their hearts? It is merely external adornment that they display. In the next section, consisting of four couplets, Han Yu explains the cause of his failure: fate that has so far doomed him to a wandering existence, with limited material sustenance: My life has been one of obstacles and hardship, Utterly unlike that of these two birds. I have dashed east and west, south and north, For ten years with never a fixed abode. I am mortified that I have to eat a fixed amount; How much more so not having my name listed on the recommendation list?32 Whatever the world prefers is considered “wise”; Thus, how could anyone say I am not stupid? In the concluding and longest section, consisting of eight couplets, Han Yu cites the example of the Yin dynasty minister Fu Yue 傅説 to show that it is not political influence that enables the man of worth to obtain office, but fate and timely opportunity. Emperor Wuding 武丁 (Gaozong 高宗) of the Yin one night dreamed that the Lord of Heaven had given him a “good assistant.” When he awoke, he described the man’s appearance to his subordinates and ordered 29  Cf. “Li sao”: “Suddenly I look back and shed tears”; Chuci buzhu (Hong Kong: Zhonghua shuju, 1963), 1.23b. 30  He means of course the road back to the capital. 31  The birds “advance” while Han “withdraws.” 32  Tong Dide (Han Yu wen xuan, p. 4) argues for the variant rong 榮 (honor, honorable) instead of ce 策 (to list). Following his interpretation, the translation should read: “How much more so not attaining an honorable name on the recommendation list?”.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old-Style Fu of Han Yu

1213

them to search for him throughout the empire. Fu Yue, who was working as a builder at Fu Crags 傅巖, fit the description exactly, and Wuding appointed him his prime minister.33 From the example of Fu Yue, Han Yu concludes that unless “timely fortune has come,” there is no use continuing the quest for political patrons. Attempting to advertise oneself only “brings blame and hastens trouble.” Nevertheless, although he has been unsuccessful in his career up to this point, Han is optimistic, for he feels that Heaven “has great hopes” for him. He also seeks solace from such ancients as Fu Yue, who rose from obscurity to become prime minister. Thus, he will no longer bemoan his lack of position. In the penultimate couplet, Han in effect declares that he will no longer be resentful of those who obtain undeserved positions, “for they will become the playthings of the gods.” Confident that he has many years ahead in which to achieve recognition on the basis of merit, he no longer feels envious of “such creatures.” “Such creatures” of course has a double meaning: it refers both to the tribute birds and unworthy men who obtain undeserved positions. Of old, Gaozong of the Yin Obtained a good assistant while asleep at night. Which of the king’s attendants paved the way for him? Truly it was Heaven that united them and the gods that matched them. But if timely fortune had not come, Though two men sought each other out, nothing could have brought them together. Though one goes from house to house stating his case, He would only bring himself blame and hasten trouble. When Heaven above gave birth to me, It had great hopes for me on the earth below. Why not seek an example among the ancients? Why sadly complain about lack of position? Verily men who attain one without proper ability Will become playthings of ghosts and spirits. Fortunately, I have not come to the evening of my years; I hope not to be envious of such creatures. “Gan er niao fu” is a relatively straightforward piece. Like such Han dynasty works as “Sui chu fu,” “You tong fu,” and “Si xuan fu,” it begins as a sorrowful complaint, but ends on a note of confident optimism. Such a shift in effect 33  See Shang shu, “Yue ming” 説命.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1214

Knechtges

represents a retraction of the gloomy pessimism of the first part of the poem. It is similar to the palinode of medieval European poetry, which the poet wrote to retract something said in an earlier poem. Han Yu’s piece is written in a much simpler style than the Han dynasty fu. Even though it uses a sao-style prosodic pattern, it has few allusions to the Chuci. Han Yu also avoids the use of the rare words that one usually finds in the Han dynasty xianren shi zhi pieces. Han Yu did not remain in Heyang long. In Zhenyuan 12 (796), he received his first appointment: on the staff of the former prime minister, Dong Jin 董晉 (724–799). In August of 796 Dong was appointed cishi 刺史 (prefect) of Bianzhou 汴州 (modern Kaifeng), guancha shi 觀察使 (regional inspector) of Bian, Song 宋, Bo 亳 and Ying 穎, and jiedu shi 節度使 (military governor) of Xuanwu 宣武.34 Han Yu’s position was that of guancha tuiguan 觀察推官 (judge for the regional inspector), and he was assigned to Bianzhou. After about a year in Bianzhou, Han became ill and returned home to Heyang to recover. While at home, he wrote the second gu fu contained in his collection, the “Fu zhi fu” 復志賦 (“Fu on Restoring My Resolve”).35 The theme of this fu is similar to that of Liu Xin’s “Sui chu fu.” The terms fu zhi (restoring resolve) and sui chu (fulfilling original resolve), which are virtually synonymous, both express the idea of unwavering pursuit of long-held aspirations, which in the case of both Liu Xin and Han Yu meant the quest for fame and success as scholar-officials.36 The “Fu zhi fu,” which is ninety lines long, is Han Yu’s longest fu. The first part of it is autobiographical and supplies much information about Han’s early life. Han Yu tells of the hardships he faced as a boy, including the arduous journey south to Shaozhou, where his elder brother died in exile, and his subsequent return with his sister-in-law to Heyang in the north, where finding the “central plain in turmoil,” they had to return south (to Xuancheng):

34  Han Yu has a xingzhuang 行狀 (career description) for Dong Jin; see Ma Tongbo, Han Changli wenji jiaozhu 8.331–36. He also has biographies in Jiu Tang shu 舊唐書 (Beijing: Zhonghua shuju, 1975), 145.3934–37 and Xin Tang shu 新唐書 (Beijing: Zhonghua shuju, 1975), 214.6001–2. 35  For texts, see Ma Tongbo, Han Changli wenji jiaozhu 1.3–5; Wubai jia zhu Changli wenji 1.5a–9b. The fu also was included in Chao Buzhi’s 晁補之 (1053–1110) Bian Li sao 變離騷, which was incorporated by Zhu Xi 朱熹 (1130–1200) in his Chuci houyu 楚辭後語; see Chuci jizhu 楚辭集注 (Hong Kong: Zhonghua shuju, 1972), 4.8a–10a. 36  Note that Liu Xin’s use of the term sui chu was different from the sense it acquired in the Six Dynasties period, when it always implied “fulfilling the aspiration to retire.” See Nakajima Chiaki 中島千秋, Fu no seiritsu to tenkai 賦の成立と展開 (Matsuyama: Kankōsei, 1963), 438–39.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1215

The Old-Style Fu of Han Yu

Having dwelled in unrelieved sadness and sorrow, Alone I brood and heave long sighs. It is not that my morning repast is not filling, Nor that my winter furs are not whole. In days past, as I gained knowledge of the world, Truly the road was rough and the course difficult. Before the Jupiter cycle began to repeat, I accompanied my elder brother on his southern exile.37 We crossed the great Yangtze’s terrifying waves, And traversed Dongting’s vast expanse. Only at Qujiang did we rest,38 Beyond the mountain ranges of the southern quadrant. Alas! After only a few days and months, Hand in hand with the orphan and his widowed mother, I returned north.39 Encountering turmoil in the central plain,40 We then sought sustenance south of the Yangtze. The fu is remarkable for its statement of Han Yu’s strong devotion to ancient learning, which he claims to have had from the time he was a youngster living in Xuancheng: When I first began to concentrate on study, Only on the ancient teachings did I focus my mind. Scanning the traces left by ancient men of genius, Alone I soared, searching the darkness. Once I learned the road, I swiftly galloped; Who could have known my strength was insufficient?

37  Jupiter takes approximately twelve years to revolve around the sun. Thus, Han Yu was not quite twelve when his elder brother Han Hui was exiled to Shaozhou. 38  Qujiang 曲江 was a prefecture of Shaozhou. It was located south of modern Shaoguan City. 39  Han Hui was banished in the fifth month of Dali 12 (=June 10 to July 9, 777). After Han Hui’s death, Han Yu, his sister-in-law (née Zheng 鄭), and her son (Han Yu’s nephew) returned north to bury Hui at Heyang. 40  The turmoil (literally “incidents”) to which Han Yu refers is the insurrections that afflicted north China beginning in Jianzhong 2 (781).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1216

Knechtges

In the next section, Han tells how he naively believed he could easily obtain official position: I examined the garb worn by the ancients, And surveyed the fashions donned by the present age.41 Suddenly forgetting how unworthy I was, I thought “blue and vermilion” could be easily obtained.42 He who knows himself is discerning;43 This thus was the reason I was so confused.44 Han Yu then relates the disappointment he felt when he went to the capital to attempt the examinations. He found this “metropolis of fame and profit” full of opportunists and schemers, whose devious ways were hard to follow. Realizing that he could not compete with them without compromising his ideals, he resolved to dwell quietly and devote himself to scholarship and literature: Selecting an auspicious day, I traveled west. And I reached the capital city. The ruler’s gate could not be directly entered, And so I submitted to examination from the authorities. Verily this metropolis of fame and profit Is the place to which the crowd gallops. Vying to seize opportunity and cling to power, Their numerous transformations are hard to fathom. Preserving my pristine ignorance, I quietly dwelled, 41  Han Yu here borrows a sartorial metaphor from the “Li sao” (see Chuci buzhu 1.10b): “Oh, I model my garb from the moral men of the past; / It is not the fashion donned in the present age.” 42  Han Yu borrows both from the biography of Xiahou Sheng 夏侯勝 in the Han shu (75.3159) and Liu Jun’s 劉峻 “Bian ming lun” 辦命論 (“Disquisition on Fate”) in the Wen xuan (54.16b): “Whenever (Xiahou Sheng) lectured, he said to his students, ‘The scholar is most troubled by not understanding classical learning. If he can understand classical learning, he can obtain the blue and purple as easily as bending down to pick up a straw from the ground’;” “Looking at the sudden change in the status of Peng Yue and Han Xin, some would say their rapacious fierceness obtained them noble ranks, and looking at the vermilion sashes of Zhang Yu and Huan Rong, some would say it was because they understood the classics that they obtained the blue and purple.” 43  Cf. Laozi 33: “He who knows others is wise; he who knows himself is discerning.” 44  Han Yu seems to be saying that he did not know himself. Thus, he was confused. See Wubai jia Changli Wenji 1.7a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1217

The Old-Style Fu of Han Yu

To follow a different course from theirs. I wished to rush ahead in order to make up for lost time, But looking back at my original aims, I reproved myself. At dawn I galloped in the grove of letters; At dusk I soared in the garden of arts. Truly I walked backwards while trying to go forward;45 Rather than advancing nearer, I drew farther from my goal.46 In the next long section, Han Yu tells of the great despair he felt after his three failures of the boxue hongci examinations. Utterly discouraged, he left the capital for home. I sorrow that the bright sun would not consult with me; For ten years till now all is as it was in the beginning. It was not that I did not earn a degree in the examinations, But I never was able to repair my deficiencies. I advanced but did not fulfill my aspirations, And so I withdrew, to retire and dwell in poverty. Opening the gates of state, I departed for the east, Sighing at the slowness of my journey. At times resting on a height I looked back, And tears streamed down my face. On his way to Heyang, Han Yu stopped at Luoyang, where he performed divination with a tortoise shell: Arriving at Luo City, I sadly gazed; Here I briefly roamed and drifted, disconsolately pacing. I used a large tortoise to examine the oracle; And to seek the dwelling place of “the hidden man who remains firm.” The “hidden man who remains firm” is the hermit, and thus these lines seem to indicate that Han Yu was contemplating becoming a recluse. This part of the 45  Cf. Kongzijiayu 孔子家語, Sibu congkan, 1.15a: “When Confucius was in Wei, Ran Qiu said to Jisun, ‘When a state has a sage and is unable to use him, but still wishes to seek good government, this is like walking backwards while trying to catch up with the people ahead. One simply cannot succeed’.” 46  Cf. “Da siming” 大司命 (Chuci buzhu 2.13b): “Rather than advancing nearer, I draw farther away.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1218

Knechtges

fu is similar to Zhang Heng’s “Si xuan fu,” in which Zhang Heng likewise resorts to divination and receives an oracle that advises him to become a hermit.47 However, even though he was prepared to live in seclusion for the rest of his life, thanks to the beneficence of Dong Jin, he finally was able to embark on an official career: I was content to grow old and die in seclusion, And not obtain illustrious name and renown. Without the veritable goodness of my master,48 How could I have come to the city of Jun?49 In the next section, Han Yu expresses the great joy he felt at obtaining a position. He even claims to be so grateful to obtain employment, he would be happy in this lowly post for the rest of his life. To show his gratitude, he vows absolute devotion and gratitude to his patron. The petty man who yearns for favors,50 Still knows enough to offer his foolish devotion. Truly I am different from oxen and horses; How can I stop at drinking water and demanding hay? Bowing beneath his gate, silent and still, I shall end my years with contentment and joy. Now I shall avail myself of leisure to obtain advancement; My face beams with pleasure, and I am happy and glad. Admiring his splendid virtue, I shall be content with poverty; But how can my loyalty be fully expressed? Once I made a pact in my heart: Who without giving can obtain success?51 Detesting the foul filth of greedy artifice, I say, “First shall I work and then shall I eat.”52

47  See Wen xuan 15.4a. 48  “My master” is Dong Jin. 49  Jun refers to Junyi 浚儀, the administrative center of Bianzhou. 50  Cf. Lun yu 4/11. 51  Cf. “Chou si” 抽思 (Chuci buzhu 4.16b): “Who without giving can expect reward? / What without bearing fruit can be reaped?” 52  Cf. Li ji 禮記, “Ru xing” 儒行 (in Shisan jing zhushu 59.2b): “If one works first and then obtains salary, won’t the salary be easily earned?”.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1219

The Old-Style Fu of Han Yu

However, Han Yu’s recent experience reminds him that such resolve is hard to maintain. With this realization, his sadness returns, and his mind becomes confused. Since he cannot avoid such feelings, he concludes that it would be better to avoid food (=rich emolument and high position) and escape worldly involvement (“soar on high”). As a warning lest my resolve not be maintained, I shall love these words and not forget them. Sad and disappointed, I am lost in bewilderment; My heart is confused as if there is no place to turn. Since I cannot help feeling this way, Can anything be better than refusing to eat or soar on high? Han then declares that he can be happy and content, even in the humble position of gate-guard. This idea clearly is derived from Mengzi, who stipulated what kind of position would be appropriate for a man who had to take office on account of his own poverty: One does not take office because of poverty, yet there are times when one does take office because of poverty…. He who takes office because of poverty declines an honorable position and occupies a lowly one, and declines high salary and accepts a small one. For a man who declines an honorable position and occupies a lowly one, and who declines high salary and accepts a small one, what would be the most appropriate position? That of gate-guard or watchman.53 As an example of a man who was content in low position, Han Yu cites the ancient sage Yi Yin 伊尹, who according to Mengzi was happy with his life as a farmer: Yi Yin plowed in the lands of the Lord of Xin and delighted there in the way of Yao and Shun…. When Tang sent a man with presents to invite him to court, he calmly said, “What could I do with the gifts with which Tang invites me? Is it not better for me to dwell in the fields and from this delight in the way of Yao and Shun?”

53  See Mengzi 5B/5.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1220

Knechtges

Though my post of gate-guard is humble and low, I have the chance to release my feelings, content and happy.54 If Yi Yin could find delight in the fields, How do I merit wealth and honor? Han then vows that he will hold fast to his vow made earlier that he “will work before he eats.” He then ends the fu on an optimistic note, confident that the future will be brighter than the past: Fearing that I will not hold firm to my oath, I reproach myself with these words as I end my poem: “The past cannot be repeated, But I expect there is much hope for the future.” Like the “Gan er niao fu,” the “Fu zhi fu” is written in relatively simple language. Although the theme is similar to that of Liu Xin’s “Sui chu fu,” Han Yu’s piece avoids ornate language and extensive use of parallelism, both of which are typical of the Han fu. Although it uses more allusions than “Gan er niao fu,” references to earlier texts are much less than one finds in the Han fu. If it were not rhymed, one could easily mistake it for a guwen prose essay. What the “Fu zhi fu” lacks in verbal ornament it makes up in the directness of its expression. Ever the uncompromising moralist, Han Yu uses the fu not for artistic purposes, but to reaffirm his intention to pursue an official career. For the most part, the fu consists of an autobiographical narration interspersed with moral maxims. Although many fu have a sententious quality, Han Yu’s fu outdoes them all in sententiousness. Han Yu’s third fu, “Min ji fu” 閔己賦 (“Grieving for Myself”), is a short piece of thirty-four lines.55 Although it does not have a preface, the “supplementary commentary” (buzhu 補注) cited by Wei Zhongju gives the circumstances of composition as follows: Han Yu once assisted Dong Jin in Bian, but in a short time Jin died. He then assisted Zhang Jianfeng in Xu. When Jianfeng also died, he quit office and left to live in Luoyang. This is the reason he composed a fu lamenting himself. 54  Cf. Hou Han shu 後漢書 (Beijing: Zhonghua shuju, 1963), 81.2678: “The gatekeeper released his feelings guarding the gate.” 55  For texts, see Ma Tongbo, Han Changli wenji jiaozhu 1.5–6; Wubai jia zhu Changli wenji 1.10a–11b; Chuci houyu 4.10b–11b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1221

The Old-Style Fu of Han Yu

Dong Jin died in March 799 (second moon, Zhenyuan 15), and after accompanying Dong’s funeral cortege to Luoyang, Han took up residence in Xuzhou 許州 (modern Xuzhou) where in autumn of the same year the military governor Zhang Jianfeng 張建封 (735–800) appointed him to the same position he had held under Dong Jin, tuiguan. After Zhang’s sudden death in June 800, Han returned immediately to Luoyang. Thus, if the supplementary commentary is correct, Han must have written the “Min ji fu” sometime after his arrival in Luoyang.56 As Han Yu’s poetry written during the years 799–800 clearly shows, in spite of having attained his long-sought goal of official position, these were not happy times for him. For example, in “Cong shi” 從仕 (“Taking Office”), written just when he first arrived in Xuzhou, Han bitterly vents his frustration:57 When I dwelled in idleness, food was insufficient; But upon taking office, my strength was not up to the task. Both endeavors harm one’s nature, And all life long has my heart been tormented. At dusk I return to my own lodgings, And my sadness gives rise to sounds of sighing. To abandon the human world— From antiquity men have done so, not only now. The “Min ji fu” begins as a typical xianren shi zhi piece. Using almost the same language as the “Ai shi ming” 哀時命 of the early Han fu poet Zhuang Ji 莊忌, Han bemoans the fact that he was not born in the golden age of antiquity when worthy men were properly appreciated: I lament that I lived too late to meet the men of old;58 It is time and circumstance that have made this so! My lonely melancholy—when will it end? I shall rely on literature to express myself.

56  Chao Buzhi does not give a definite date for its composition, but implies that Han Yu wrote this fu in Yuanhe 7 (812) when he assumed the position of guozi boshi 國子博士 (professor at the College of the Sons of State). See Chuci houyu 4.10b. 57  See Qian Zhonglian 錢仲聯, ed. and comm., Han Changli shi xinian jishi 韓昌黎詩繫年 集釋 (1975; rpt. Shanghai: Shanghai guji chubanshe, 1984), 1/113. 58  Cf. “Ai shi ming” 哀時命 (Chuci buzhu 14.1b): “I lament that time fated me not to meet the men of old; / Why was I born not to meet the proper time?”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1222

Knechtges

In the following seven couplets, Han Yu contrasts himself with Yan Hui 顏回, whom Confucius praised for not allowing difficult living conditions to affect his happiness: Worthy indeed was Hui! A single bowl of rice and a single gourd of drink, he lived in a humble lane. Others could not have endured the misery, but Hui did not allow it to alter his joy. Worthy indeed is Hui!59 Han Yu dismisses Yan Hui’s feat as “a petty matter for a wise man.” He had enough food and drink to sustain himself, and he had the great sage Confucius who gave him moral support. Here Han Yu seems to repeat the same criticism he made of Yan Hui in his “Yu Li Ao shu” 與李翱書 (“Letter to Li Ao”): Confucius praised Yan Hui as follows: “A single bowl of rice and a single gourd of drink. Others could not endure the misery, but Hui did not allow it to alter his joy.” He had a sage who gave him support, and he had food and drink enough to keep him from dying. That he was not distressed, but joyful, was it not easy? A person such as I, who has no one to give him support, no bowl of rice, no gourd of drink, and no resources—it wouldn’t be hard for me to die of starvation, would it?60 Unlike Yan Hui, Han Yu cannot be content with life as an impoverished recluse, for he still clings tenaciously to his ideal of service to the state. Han further considers that he has suffered more adversity than Yan Hui, but has no one like Confucius to whom he can turn. Although he feels as if he has been sculling a boat across the water without any sense of direction, he resolves to emulate the ancient worthies and continue his pursuit of office. Of old, Master Yan almost achieved perfection;61 Though living in obscurity and straitened circumstances, he was calm and composed.62 59  See Lun yu 6/9. 60  Ma Tongbo, Han Changli wenji jiaozhu 3.105. 61  See Zhou yi 周易, “Xici xia” 繫辭下 “Master Yan Hui, isn’t he one who has nearly achieved perfection?” 62  Cf. Han Yu, “Sheng shi Yanzi bu erguo lun” 省試顔子不貳過論 (“Disquisition for the Departmental Examination on the Topic ‘Yanzi Did Not Err Twice’ ”): “Master Yan considered himself to be like this. Thus, he lived in a humble lane in order to perfect his sincerity, and he drank from a single gourd in order to seek his aims. He did not obstruct

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1223

The Old-Style Fu of Han Yu

This certainly was a petty matter for a wise man, But the Master praised him as a worthy. Though he ate and drank poorly in a humble lane, It was enough to nurture his spirit and preserve his years. He had the supreme sage to give him support; How could he not be content in the face of adversity? I say, “Muddled and confused, and without companions, I gaze at that man who is far removed from me.”63 Guiding my boat with no sense of direction, I have crossed the great water’s vast expanse. I shall devote myself to the legacy of my forebears, And strive to emulate the words of former worthies. In the following eight lines, Han Yu bemoans the failure of his contemporaries to join him in promoting the Way of the ancients. He also complains that in the vast empire there is no place to which he can go for support or safety: Although I strode forth and trod the Way, I sorrow that there was no one to join me. The crowd all rejected it, and I alone used it, And suddenly I myself was confused about right and wrong. The earth below stretched so vast and far, I utterly did not know to what I could cling. Finding water and grass, I would take a rest; But never was I safe, only in danger.64 In spite of his adversity, however, Han Yu declares that he has maintained his resolve. Why does he do this? Because he realizes that Heaven’s decree dictates his lot, and thus yielding to fate, he is content to bide his time and wait for the proper opportunity. As in the “Fu zhi fu,” Han Yu recants the pessimism expressed at the beginning of the fu, and he declares that it is futile to lament time’s fate:

his Way with wealth and honor, and he did not alter his resolve because of obscurity and straitened circumstances.” In Ma Tongbo, ed., Han Changli wenji jiaozhu 2.27. 63  “That man” is Yan Hui. 64  The commentator Sun 孫氏 (=Sun Ru? 孫汝) in Wubai jia zhu Changli venji (1.11b) thinks Han Yu is referring to the insurrections he encountered while serving in Bianzhou and Xuzhou.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1224

Knechtges

Why have I been resolute so long? It also is the basic principle of Heaven’s decree. Verily stagnation and prosperity are opposite poles; Always while one arrives the other departs. A gentleman sometimes loses his proper place, While the petty man sometimes obtains timely opportunity. For now I shall firmly watch and quietly wait; Truly, not meeting the ancients—wherein lies the sorrow? Like the “Fu zhi fu,” the “Min ji fu” is written throughout in sao style (including the xi). However, except for the opening line and such phrases as qian xiu 前脩 (former worthies), this fu derives little from the sao tradition. There are none of the sartorial metaphors or plant allegories of the Qu Yuan poems, and except for rhyme, the fu is more prosaic than poetic. The long disquisition on Yan Hui, which is more typical of a prose essay than a fu, again reveals Han Yu the moralist issuing his sententious pronouncements. The fourth gu fu in Han Yu’s collection is the “Bie zhi fu” 別知賦 (“Fu on Parting from a Friend”).65 All commentators agree that Han Yu wrote this fu while he was living in exile in Yangshan 陽山 (modern Yangshan, Guangdong). He probably wrote it to send off one Yang Yizhi 楊儀之, who visited Han Yu in Yangshan perhaps in Zhenyuan 20 (804). The exact identity of Yang Yizhi is not precisely known. Cai Yuanding 蔡元定 (1135–1198) claims he was the son of Yang Ning 楊凝 (ob. 803), whose elder brother Yang Ping 楊憑 (ca. 750–815) served as guancha shi (regional inspector) for the Hunan circuit.66 Although the genealogical table of the ministers’ clans that Cai cites does not actually contain Yang Yizhi’s name,67 it is quite possible that Yang Yizhi did belong to the family of Yang Ping. According to a Song dynasty commentator, Yang Ping appointed Yang Yizhi to his staff sometime after he took up his post as guancha shi of Hunan.68 Han Yu must have been on good terms with Yang Yizhi, for in a farewell preface that he must have written about the same time as the “Bie zhi

65  For texts, see Ma Tongbo, Han Changli wenji jiaozhu, 1.6–7; Wubai jia zhu Changli wenji 1.11b–13b; Chuci houyu 4.11b–12b; Tang wencui 9.4b–5a. 66  See Wubai jia zhu Changli wenji 1.12a. 67  See Xin Tang shu 71B.2381. This error has been pointed out by Yamasaki Jun’ichi, “Kan Yu no kofu ni tsuite,” Part ii, p. 6. 68  See the commentary of Sun cited in Wubai jia zhu Changli wenji 20.8b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1225

The Old-Style Fu of Han Yu

fu,” Han praises him as equal to Cui Qun 崔群 (772–832) and Li Bo 李博 (jinshi 792), the most worthy guests of Xuanzhou 宣州.69 Except for its sao-style prosody, the “Bie zhi fu” is almost indistinguishable from a song bie shi 送別詩 (parting poem). Han Yu begins the fu declaring how hard it was to find a truly understanding friend: Since I first began choosing friends in the world, It has been almost two revolutions of the Jupiter cycle.70 Below to what depth did I not go? Above what height did I not search? Diversely mixed, many they were; All appreciated talent and were fond of good character. How can one be content with illustrious position and enjoy prosperity alone? Seeing someone in straitened circumstances, one should share his sorrow. But a truly understanding friend is hard to find; Such a person is one in a hundred.71 In the following lines, Han Yu tells how he met such an understanding friend in Yang Yizhi. After he had been exiled to Yangshan in the year guiwei (803), Han Yu received a visit from Yang, who traveled to Yangshan on an official mission. In this land where Han had only “insects and snakes for companions,” he found the profound and sympathetic friend he had long sought. Han refers to Yang as a man who could draw out “profound words” from Han’s unsettled mind and relieve his “many sorrows.” In the guiwei year I was exiled, With insects and snakes for companions in a corner by the sea. Here I met a man coming on a mission, 69  See “Song Yang zhishi xu” 送楊支使序, in Ma Tongbo, ed., Han Changli wenji jiaozhu 4.148–49. 70  It is not clear when Han Yu “first began choosing friends.” Cai Yuanding (see Wubai jia zhu Changli wenji 1.12a) says the period to which Han Yu refers began in Xingyuan 1 (784), which was a jiazi year. At that time he fled south to Xuancheng. From that time to Zhenyuan 19 (803), when he was exiled to Yangshan, twenty years had passed. 71  Cf. Lu Ji 陸機 (261–303), “Tan shi fu” 歎逝賦 (“Fu Lamenting the Departed”), Wen xuan 16.16a: “I look for old acquaintances from those that survive, / And I find one tenth out of a thousand and a hundred.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1226

Knechtges

And I opened the official lodge and spread a banquet. He searched for profound words in my confused thoughts, And elicited a solitary smile from my numerous cares. What object is so deep it cannot be reflected in a mirror? What principle lies so hidden it cannot be drawn upon? At first at odds, our views conflicted, But in the end our differences dissolved and together we flowed. However, Yang could not stay in Yangshan forever, and as Han Yu thinks of the arduous journey his friend must undertake over steep mountains and through dense forests, he expresses sorrow not only that he will be separated from Yang, but that he himself has no way to leave this place. Han Yu then portrays himself standing at the outskirts of the town, where he parted from Yang, trying to hold back his tears. But how can this happiness be enjoyed for long? Then, he harnessed his horses and wheeled round his carriage. I know that in the future such comings will be infrequent, And I sorrow there is no way for me to leave this place. I lean on the outer wall and wipe my tears; All day long I slowly linger in vain. Unlike the three fu previously discussed, the “Bie zhi fu” ends in sorrowful weeping. The tearful conclusion of course is common in parting poems.72 However, this ending may be more than mere convention, for there is no doubt about the sincerity of Han Yu’s admiration for Yang Yizhi.73 Furthermore, as Yamasaki Jun’ichi points out, the sadness that Han Yu expresses in this fu may reflect the increased frustration that he felt after being exiled to Yangshan.74 He had held high positions in the capital, and now because of his forthright criticism of government policy,75 he suffered the ignominy of demotion to a 72  See Stephen Owen, The Poetry of the Early T’ang (New Haven: Yale Univ. Press, 1977), 296–97. 73  Cf. Han’s “Song Yang zhishi xu” (Ma Tongbo, ed., Han Changli wenji jiaozhu 4.149): “Those who say that because I am a town chief here I flatter this man (=Yang Yizhi) do not know the truth.” 74  See “Kan Yu no kofu ni tsuite,” Part ii, pp. 9–10. 75  In late 803, Han and two other members of the censorate submitted a memorial urging that the capital area, which was suffering from drought and famine, be excused from the winter taxes. See “Yushi tai shang lun tian han ren ji zhuang” 御史臺上論天旱人饑

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1227

The Old-Style Fu of Han Yu

prefectureship located in an area Han called “the most desolate place in the world.”76 He also was deeply moved by the loss of his nephew Han Laocheng 韓老成 (ob. 803) and several good friends, who had died not too long before he was sent into exile.77 Thus, Han Yu does not sorrow so much for Yang’s departure, but rather laments his own isolation and separation from friends, who have either died or live in remote places, far from Yangshan. The “Bie zhi fu” does resemble the other three fu in its simple language. The use of such phrases as “Below to what depth did I not go, / Above what height did I not search” 下何深之不即, 上何髙之不求 and “What object is so deep it cannot be reflected in a mirror, / What principle lies so hidden it cannot be drawn upon” 物何深而不鏡, 理何隱而不抽 again reveal Han Yu’s sententiousness and his fondness for prosaic lines. More typical of the fu style is the four-line landscape description depicting the harsh terrain and weather that not only make Yang’s journey difficult but also form barriers that separate Han from his friend. Mountains, rocky and steep, press one upon another; Trees, dense and thick, twine together. Rain, pouring heavily, does not stop; Clouds, swollen large, constantly drift. 山磝磝其相軋 樹蓊蓊其相摎 雨浪浪其不止 雲浩浩其常浮

The last two lines, at least syntactically, are modeled after lines in the “Bie fu” 別賦 (“Fu on Separation”) by Jiang Yan 江淹 (444–505); (“Presentation from the Censorate Submitting a Disquisition on Drought and Famine”), Ma Tongbo, ed., Han Changli wenji jiaozhu *. 338–39. 76  See “Song Ou Ce xu” 送區冊序 (“Preface Sending Off Ou Ce”), in Ma Tongbo, ed., Han Changli wenji jiaozhu 4.156. For a complete translation of this piece, see Hartman, Han Yü, 59. 77  See “Ji shier lang wen” 祭十二郎文 (“Offering for My Nephew”) in Ma Tongbo, ed., Han Changli wenji jiaozhu 5.195–98. In 803, Han Yu wrote “Ku Yang bingbu Ning Lu Xizhou Shen” 哭楊兵部凝陸歙州參 (“Weeping for Military Minister Yang Ning and Lu Shen of Xizhou”), a poem lamenting the death of Yang Ning, who died in the first month of Zhenyuan 19 (January–February 803), and Lu Shen, who died in the fourth month of Zhenyuan 18 (May–June 802). See Qian Zhonglian, ed., Han Changli shi xinian jishi 2.152–53.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1228

Knechtges

The wind, soughing and sighing, strangely sounds; The clouds, stretching on and on, take on unusual hues. 風蕭蕭而異響 雲漫漫而奇色

This four-line passage is about as close as Han Yu ever gets to writing in the epideictic Han fu style. It almost reads as an intrusion into the essentially prosaic style of the remainder of the piece. Qian Mu 錢穆 has in fact commented on the prosaic character of Han Yu’s fu style: I maintain that not only did Han Yu use prose to write verse, he actually used the spirit and manner of prose to write fu. Try to recite the fu in Han’s collection and his laments and offerings, even the inscriptions in his epitaphs as well as other genres such as eulogies, encomia, admonitions, and inscriptions, and all literary genres that properly should be classified as fu. As Han Yu writes them, whether they are rhymed or not, he actually uses a prose manner and style to create the piece.78 It is certainly true that Han Yu’s fu lack the verbal embellishments contained in the fu of a Sima Xiangru or Yang Xiong. Although he was capable of using ornate language in his shi and in some of his prose pieces, even when compared with the xianren shi zhi fu of the Han and Six Dynasties, Han Yu’s fu are much more direct and written in a simpler style. Georges Margouliès even goes so far to say that Han Yu’s fu are so prosaic, they more closely resemble his guwen prose writings than they do “true fu”: Les fou de Han Yu, nous l’avons dit, malgré les hi and les phrases à six caractères, malgré les citations historiques à la manière des T’ang, les mots redoublés à la façon des Tcheou et les synonymes accouplés à radicale semblable comme le faisaient les Han, malgré tous les artifices extérieurs, restent toujours prosaïques et ressemblent peu aux vrais fou, faisant corps avec les autres ouvrages de Han Yu qu’assemble le style uniforme.79

78  See “Za lun Tangdai guwen yundong” 雜論唐代古文運動, Xinya xuebao 3.1 (1975): 145–46. 79   Histoire de la littérature chinoise: Prose (Paris: Payot, 1949), 182.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old-Style Fu of Han Yu

1229

Regardless of one’s opinion of the worth of the fu, there is no question that traditionally a strong appeal of the fu lay in its flowery style and rich language. Han Yu’s style certainly is bland when compared with the pre-Tang fu, or even with many fu of the Tang period. I think it is fair to conclude that Han Yu did not place the highest value on the fu, and he seems to have used the form for only a brief period. When he did turn to the fu, he used it purely as a means of self-expression and moral statement. In later years, Han Yu did write other fu-like pieces. Madeline Spring has published a study of the “Jin xue jie” 進學 解 (“Justification for Advancing in Learning”), treating it as a fu.80 Other fulike works include “Song qiong wen” 送窮文 (“Farewell to Poverty”), which must have been inspired by Yang Xiong’s “Zhu pin fu” 逐貧賦 (“Fu on Expelling Poverty”), and possibly “Song feng bo” 訟風伯 (“Accusing the Lord of the Wind”). The literary quality of these later works is much higher than the four pieces he titled fu, and a careful study of them as fu would offer a better guide to Han Yu’s achievement as a fu writer.

80  Madeline K. Spring, “Han Yü’s Chin-hsüeh chieh: A Rhapsody on Higher Learning,” T’ang Studies 4 (1986): 11–27.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Another Go at the Mao Ying chuan Elling O. Eide I recently had occasion to fish out and freshen up a translation of Han Yü’s Mao Ying chuan 毛穎傳 that I had laid aside some twenty years ago. I offer it here in the thought that others might find it useful (or at least amusing) to see my attempt at making the effect of the puns and wordplay in that comic pseudo-biography accessible to the general reader. The piece is, of course, particularly awkward to translate because the wordplay involves personal and proper names. To make matters worse, even the classical Chinese only just barely lets Han Yü get away with treating rabbits and rabbit fur as if the two categories were identical. As will be obvious, I have cheerfully bent my rule that “a translator should not get in the way,” and have shamelessly intruded to create new names and ersatz literary allusions. I feel better about this than might otherwise be the case because there now exist at least two excellent, scholarly translations of the piece: one by James R. Hightower in “Han Yü as Humorist,” Harvard Journal of Asiatic Studies 44.1 (1984), 5–27, and the other by William H. Nienhauser, Jr., “An Allegorical Reading of Han Yü’s ‘Mao-Ying Chuan’ (Biography of Fur Point),” Oriens Extremus 23.2 (1976), 153– 174. My notes here below are only those that might be wanted for the general and rather casual reader. I have followed the text established by Kao Pu-ying 高歩瀛 in his T’ang-Sung wen chü-yao 唐宋文舉要, vol. 1 (Peking: Chung-hua, 1963), 260–269, preferring two or three minor variants from the Wen-yüan ying-hua, 793.9b–11a.

Source: “Another Go at the Mao Ying chuan,” T’ang Studies 8–9 (1990–91): 105–11.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_039

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Another Go at the Mao Ying chuan

1231

The Biography of Tipp O’Hair1 by Han Yü (768–824) Tipp O’Hair was a native of Chung-shan. His ancestor, Perspicacity D. Cottontail, assisted Emperor Yü in governing the East2 and was conspicuous for his role in providing sustenance to the ten thousand creatures. In recognition of this he was enfeoffed with the House of Coney, and, at his death, he became one of the Twelve Calendrical Divinities.3 Perspicacity once said, “My children and grandchildren, as the descendants of a divinity, should not be like common creatures; let them therefore be born from the mouth.”4 And it was so. Eight generations removed from Perspicacity was a descendant named Leveret. Tradition holds that he lived at Chung-shan during the Shang dynasty and acquired the arts of the spirits and immortals so that he could obscure the light and have a mysterious effect on things. To spy on the moon princess, Ever Beautiful,5 he rode away to the moon on the back of the Lunar Toad. Thereafter his descendants lived in retirement and did not serve. Some say, however, that there was a Wylie Hare living at Eastside, a clever fellow and an excellent runner, who once had a contest with the greyhound Lu from Han. When Lu lost, he was so angry that he plotted with Magpie, a dog from Sung, and together they killed Wylie, making mincemeat of his family.

1  This biography of the first writing brush is written in imitation of the biographies in the Shih chi of Ssu-ma Ch’ien (died about 85 bc), China’s first great historian, who bore the title of Lord High Astrologer. 2  In Chinese cosmology the East is governed by the cyclical sign mao 卯, the sign of the rabbit. The “House of Coney” in the next sentence translates what is literally “the land of mao.” 3  The Chinese calendrical system uses twelve cyclical signs (“terrestrial branches”) associated with animals, among which the rabbit is the fourth in the sequence. 4  There was a Chinese tradition that the rabbit conceived by sucking its fur and gave birth by spitting its young out of its mouth. One guesses that this belief had its origin in a combination of folk etymology and observation: the word for “rabbit” 兔 is homonymous with the verb “to spit” 吐, and a doe rabbit will pull out some of her belly fur to line her nest. The Chinese may also have misconstrued their observation of a frightened doe eating her newborn young. 5  In Chinese myth the moon is inhabited by Ever Beautiful (Heng O 恆娥, a.k.a. Ch’ang O 常娥), a rabbit, a toad, and a sweet-olive tree (Osmanthus species) which is usually misidentified as a cassia.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1232

Eide

At the time of the First Emperor of the Ch’in, General Meng T’ien6 stopped at Chung-shan on his way south to attack Ch’u. Preparing for a great hunt to intimidate the people of Ch’u, he summoned his adjutants and his commanders of the left and right to conduct a divination with milfoil stalks according to the Lien-shan Manual.7 They got a configuration relating to the designs of Heaven and the affairs of Man, and the diviner congratulated the general saying, “Today’s bag, neither tusked nor horned, shall be of those who are clothed in brown. With only eight of the body holes,8 with defective lip and whiskers long, it shall sit by squatting down. Take only the hair of highest grade and for documents it shall be an aid. When the world has unified written words, shall Ch’in not unite the feudal lords?” Subsequently, when they went hunting, they surrounded the O’Hair clan. Plucking out the best O’Hair,9 they brought Tipp back and presented their captive at the Chang-t’ai Palace. They also assembled his clansmen who were bound and tied. Then the Ch’in emperor had Meng T’ien give Tipp a ritual bath for his installation in Ferule City. Calling him the Viscount of Ferule City, the emperor daily favored him with responsibility. Tipp, as a man, was quick witted and had a good memory. From the age when knotted cords were used for keeping accounts down to the affairs of the Ch’in, there was nothing that he did not record: works on Yin and Yang, divination, and physiognomizing; medical prescriptions, genealogies, orographies, and geographies; dictionaries, maps, and pictures; the writings of the geniuses of the Nine Sects and the Hundred Schools; even the doctrines of the Buddha, Lao-tzu, and the foreign lands—all were set down in complete detail. He was also thoroughly versed in contemporary affairs. For keeping the ledgers on the government bureaus or recording the transactions of the market, he was the one the emperor employed. From the Ch’in emperor himself to the heirapparent Fu-su and the second son Hu-hai, from Prime Minister Li Ssu to livery officer Chao Kao, down even to the common people, there was none who did not love and respect him. He was, moreover, very good at carrying out people’s wishes. Upright or devious, clever or stupid, he followed his man; and although he might later be cast aside, he would remain silent to the end, letting nothing 6  General Meng T’ien is credited with invention of the writing brush. The best rabbit hair for brushes was said to come from the Chung-shan area. 7  The Lien-shan Manual, now lost, is thought to have been the Hsia dynasty divination manual. There were several such manuals in antiquity, one of which was the Chou dynasty manual known as the I ching or Book of Changes. 8  Only eight body orifices because the young are born from the mouth. 9  “Best O’Hair” translates the word hao 豪, which means both “fine hair” and “an outstanding person or leader.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Another Go at the Mao Ying chuan

1233

leak out. It was only the military men that he did not serve gladly, but even then, when called upon, he would sometimes go along. Through successive promotions, Tipp rose to the Directorship of the Secretariat and was on increasingly intimate terms with the emperor who addressed him as “Mr. Secretary.” When the emperor was personally deciding affairs of state and had weighed out his daily quota of documents with a weighing stone, even the palace ladies were not allowed to stand around, but Tipp and the candle holder were always in attendance, staying until the emperor retired. Tipp was on the friendliest of terms with Mark Black of Chiang, Potter E. Poole of Hung-nung, and Mr. Bond of Kuei-chi,10 and they supported one another, going out together wherever they went. Thus, when the emperor summoned Tipp, they would always respond as a group without awaiting further orders, and the emperor never thought it strange. In later years, on an occasion in the imperial presence, the emperor was having him groomed for an important assignment, but when Tipp took off his cap to express his appreciation, the emperor could see that he was losing his hair and that his copy work would not be up to the imperial expectation. The emperor laughed jovially and said, “Mr. Secretary is old and bald and can no longer afford me service. I once called you ‘Mr. Secretary,’ but now you are ‘Mr. Too-Sick-to-Tarry.” Tipp replied, “I have, as they say, given my heart.” Thereafter he was no longer summoned. He returned to his fiefdom and ended his days in Ferule City. Tipp’s descendants are very numerous and are scattered throughout the Middle Kingdom and the barbarian lands. They all trace their line back to Ferule City, but only those still living in Chung-shan are able to perpetuate the work of their ancestors. The Lord High Astrologer observes: The O’Hairs belong to two different clans. One group originally had the Chi surname of the Chou royal family. These are the descendants of King Wen of the Chou who were enfeoffed at Mao, as is mentioned in Tso’s Commentary where it speaks of the fiefs of Lu, Wei, Mao, and Tan. In the Warring States period there were O’Hairs of this clan known as Master Mao and Mao Sui.11 It is not known where the Chung-shan 10  That is, with Mr. Inkstick, Mr. Inkstone, and Mr. Paper. The place mentioned with each was famous for that product. 11  Although there were several notable men surnamed Mao who lived prior to Han Yü’s day, throughout the piece Han Yü confines himself to people, names, and anecdotes that were, or could have been, known to Ssu-ma Ch’ien, the Lord High Astrologer. Master Mao, a sage recluse (in chapter 77) and Mao Sui, a bold retainer (in chapter 76), are the only Maos who receive memorable attention in Ssu-ma Ch’ien’s Shih chi. It does no harm to Han Yü’s

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1234

Eide

clan originated, but its descendants are flourishing and abundant. They were pushed aside by Confucius when he laid down his brush upon completion of the Spring and Autumn Annals, but this was through no fault of theirs. After General Meng plucked out the best O’Hair at Chung-shan and the First Emperor enfeoffed him with Ferule City, they were famous in subsequent generations, but nothing more has been heard of the O’Hairs who had had the surname Chi. Although Tipp O’Hair first appeared before the emperor as a prisoner, in the end he was honorably employed. Tipp played a role in the Ch’in conquest of the feudal lords, but his reward was not commensurate with his service. How lacking was the Ch’in in gratitude!

humorous intent that both these men were unorthodox and slightly comical individuals. The surname Mao is also the Chinese word for “hair,” and except here where it was necessary to preserve the identity of a real fiefdom and two real people, it has been translated as “O’Hair” to show the wordplay that was involved. The surname Mao meaning “hair” is unrelated to the cyclical character mao associated with the rabbit or the hare.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old Men [of the Early Ninth Century] Stephen Owen Longevity is an inconvenience for literary history. Poets can live on past their most creative years characterized by poetic innovation and achievement. Secure in their fame, such senior poets often remain untouched by the new directions taken by younger poets. These old poets can join with other figures of cultural power to become a literary establishment, an enclosed world of often complacent celebrity. This more or less describes the circle of aging poets and political figures around Bai Juyi. In the case of Bai Juyi and his friends, it was a literary establishment situated in a retirement community, realized in sinecures in Luoyang, far from the hopes and perils of political life in Chang’an or the wanderings of younger poets looking for preferment. Bai Juyi enjoyed such semi-retirement, never forgetting that he was now out of public life. He had removed himself from more active public life by choice, but he celebrated his decision so frequently that it is hard not to feel that he “protested too much.” There is a genuine charm in many of the poems Bai Juyi wrote as an old man in Luoyang, but it is a charm that can be sustained only by limiting one’s reading to a few poems. In large doses these poems become repetitive, facile, and self-absorbed—and the aging Bai Juyi wrote very many poems indeed. Bai Juyi’s spontaneity and ease were a cultivated style, but in poetry these are precarious virtues. One can admire such poems, while at the same time understanding why younger poets seem to have reacted against the style. Poets were always coming and going in Chang’an, where in the 830s the dominant fashion seems to have been the finely crafted regulated verse, centered around figures like Yao He and Jia Dao. The other major center of poetic activity was Luoyang. The latter was the Eastern Capital, with dilapidated palaces and a substantial bureaucracy that had very little to do. No emperor had visited Luoyang for as long as anyone could remember; this fact became something of a theme in the city’s poetry. Since the succession was far from certain, positions in the crown prince’s establishment, the “Regency Office,” were already at some remove from the exercise of power, either in the present

Source: “The Old Men,” in Stephen Owen, The Late Tang: Chinese Poetry of the Mid-Ninth Century (827–860), Cambridge, Mass.: Harvard University Asia Center, 2006, 41–88. Reprinted by permission of the Harvard University Asia Center, © The President and Fellows of Harvard College, 2006.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_040

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1236

Owen

or in the anticipated future. However, positions in the “Luoyang assignment” of the crown prince’s establishment were sinecures in the fullest sense; their incumbents were secure in the thought that a real crown prince would never call upon their services. In the spring of 827, at the beginning of Wenzong’s reign, Bai Juyi completed his term as governor of Suzhou. The poet was called to the capital to serve as Director of the Imperial Library. In the spring of 828 he moved up a step to Vice Minister of Justice. Up to this point in his career, Bai Juyi had been very much a part of the political world. However, in the spring of the following year, at his own request, Bai was transferred to the post of Adviser to the Heir Apparent, “Luoyang assignment.” The older as well as some of the younger holdovers of the Yuanhe generation were passing away. Around 827 Li Yi, a grand old man of letters and already a well-known poet in the last quarter of the eighth century, passed away; and around 830 Wang Jian and Zhang Ji died. More unexpectedly, on September 2, 831, Bai Juyi’s dearest friend Yuan Zhen passed away at the relatively young age of fifty-three. Yuan Zhen had been considered, together with Bai, one of the leading poets of the Yuanhe generation, though in his later years he became increasingly preoccupied with political life and had become a powerful figure. By that point Bai had moved to the post of Metropolitan Governor (yin 尹) of Luoyang, a post that required some administrative involvement. Late in 831 or early 832 Liu Yuxi, another survivor of the generation and a fellow poet and close friend, stopped over in Luoyang on his way to assume his new post as governor of Suzhou, the post Bai himself had held just five years earlier. Liu Yuxi was to fill the void left by Yuan Zhen’s death, becoming Bai Juyi’s closest poetic correspondent. In the following year Bai, complaining of “illness,” resigned as Metropolitan Governor and resumed his old post as Adviser to the Heir Apparent. Liu Yuxi served out his term in Suzhou, and after passing through other short stints as a prefectural governor, he arrived in Luoyang to take up Bai Juyi’s post as Adviser to the Heir Apparent. Bai Juyi was elevated to the distinguished title of Junior Mentor (shaofu 少傅) to the Crown Prince, no less a sinecure. That same year Li Shen, another well-known poet and friend, came to Luoyang to take Bai’s earlier post as Metropolitan Governor. Li Shen was to move on to other distinguished political posts, but Bai Juyi and Liu Yuxi would remain in Luoyang until their deaths, Liu dying in 842 and Bai in 846. These were all men in their late fifties, sixties, and (for Bai Juyi) seventies. All were famous and highly connected. Luoyang was on the most popular travel route from Chang’an, and through it passed the eminent statesmen of the day, including the very senior Pei Du, Linghu Chu, and the dominant younger

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old Men [ of the Early Ninth Century ]

1237

figures of current political life, including Niu Sengru and Li Deyu (though Li Deyu was reputed to have disliked Bai Juyi and was much closer to Liu Yuxi). Such figures would stop by on their way to take up new posts, banqueting and exchanging poems with the Luoyang poets; sometimes they would occupy Luoyang posts, which were effectively vacations or rustications from their still active political careers. Bai’s and Liu’s companion pieces to poems by these eminent statesmen show that they were much more active poetically than their surviving poems often seem to indicate. Despite the fierce political feuds of the day, they represented a unified cultural community. Moreover, with the exception of Linghu Chu, who always had an eye for younger talent, they did not exchange many poems with the aspiring younger poets of the age. These younger poets sometimes came through Luoyang as well and would address poems to the senior statesmen or Bai Juyi (less often to Liu Yuxi); sometimes their poems were acknowledged, but more often they were ignored. Social and poetic relationships play a large role in Bai Juyi’s poetry dating from this period. This is true not only in poetry intended for social occasions or explicitly sent to friends; ostensibly solitary poetry, always concerned with the poet’s self-representation, was also widely distributed in his circle (and always carefully copied into the ever-growing manuscripts Bai was preparing). Liu Yuxi’s poetry composed during this period is overwhelmingly social, and his relationships with Bai and Linghu Chu stand at the center of his poetic network.1 Other poets of the day speak of culling their poems—often radically—for the poetry collections that would preserve their work for posterity. Particularly in his later years, Bai Juyi seems to have included in his collection most of what he wrote. We do find companion pieces (primarily by Liu Yuxi) to Bai Juyi’s poems that no longer survive; considering the quality of some of Bai Juyi’s pieces that do survive, it is hard to believe that the exclusions were based on aesthetic grounds. When we read Bai Juyi’s easygoing poems written in Luoyang in the early 830s—poems that not infrequently concerned concubines and singing girls— we should remember that in Luoyang during this very same period Du Mu wrote his famous ballad for Zhang Haohao and the young Li Shangyin wrote his densely difficult “Yan Terrace” poems. Although Du Mu’s ballad for Zhang Haohao derives in part from Bai Juyi’s younger work in the narrative ballad, Du Mu and Li Shangyin represented very different worlds of poetry from that of Bai Juyi. Moreover, these distinct poetic communities, located in the same 1  The case of Li Shen is more complex because we do not have the full range of his social poems. What we have is Recollecting Past Travels 追昔遊 in three juan with a preface dated 838; I discuss this later.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1238

Owen

city and active at the same time, never seem to have crossed paths (apart from a few courteous exchange poems). In other words, although the characteristic poetry of the Late Tang was taking shape nearby in the same city, Bai Juyi seems to have been completely unaware of it. Li Shangyin did meet Bai Juyi in Luoyang (probably in 829–30), but Bai does not mention it.2 It is difficult to imagine what Bai Juyi would have made of Li Shangyin’s “Yan Terrace” poems—though they seem to have lit a fire under the Luoyang demimondaine whom Li genetically called “Willow Branch.” It is not hard to imagine the existence of a community of young people in Luoyang who probably knew of Bai Juyi but had very different tastes in poetry. Du Mu seems to have shared something of his friend Li Kan’s deep dislike of Bai Juyi’s poetry. In his famous “Epistle on Presenting My Poems” he implicitly but clearly distinguished his poetic aims from the “familiar and common” style (xisu 習俗) that would have instantly been associated with Bai.3 Li Shangyin did not personally comment on Bai’s poetry. Li paid his tribute to Du Fu, Han Yu, and Li He, but he never tried his hand at the distinctive style of Bai Juyi. We may presume that it was for Li Shangyin’s mastery of parallel prose that Bai Juyi’s adopted heir called on him to write the tomb inscription, which mentions Bai’s literary fame in other countries and the size of his collection but provides no details about his achievements as a poet. Yao He was old enough, sufficiently politically elevated, and poetically bland enough to enter the orbit of the group surrounding Bai Juyi. But, apart from Yao He, it is striking how completely divorced the older Bai Juyi and his friends were from other poets working at the time. Although the poems of Liu Yuxi and Li Shen have enough in common with Bai Juyi to warrant their inclusion in the same chapter (commonalities that in no small measure reflect Bai Juyi’s influence), Bai Juyi was clearly the major poet. His poetry is unified by a set of recurrent concerns that in the aggregate lend a degree of depth to poems that can sometimes seem trivial on the surface.

2  On this point see Liu Xuekai 劉學鍇 and Yu Shucheng 余恕誠, Li Shangyin wen biannian jiaozbu 李商隱文編年校注 (Beijing: Zhonghua shuju, 2002), vol. 4, p. 1801; see also Xie Siwei 謝思煒, “Bai Juyi yu Li Shangyin” 白居易與李商隱, in Wang Meng 730–34. We must dismiss the later legend that Bai Juyi much admired Li Shangyin’s poetry and prose; see Liu Xuekai et al. (2001) 25. 3  Fancbuan wenji 242. That everyone could recognize Bai Juyi behind the “familiar and common” style was, of course, tacit recognition of Bai’s fame.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old Men [ of the Early Ninth Century ]



1239

Bai Juyi

Any critical survey of Bai Juyi’s poetry will disclose a sharp divide between those who considered him one of the greatest of all poets and those who were openly hostile to his poetry. Such a division was already apparent in the ninth century. We have already mentioned Li Kan’s criticism (and will return to it); in Du Mu’s account Li Kan wished for absolute political power simply to expunge Bai Juyi’s insidiously popular poetry from the empire. Later in the ninth century Sikong Tu 司空圖 (837–908) described Bai and Yuan Zhen as “overbearing in force, yet feeble in [natural] energy [qi], like domineering merchants in the marketplace.”4 Tang writers tended to be relatively restrained in their criticism of other poets, so these are strong words indeed. In later criticism Bai Juyi was often used as a minatory case of the style the careless poet can fall into. Pi Rixiu 皮曰休 (ca. 834–883), no less eminent than Sikong Tu, had only the highest praise for Bai. Zhang Wei’s 張爲 Schematic of Masters and Followers Among the Poets 詩人主 客圖 places Bai Juyi at the head (“master”) of his first category: “extensive and grand civilizing power” 廣大敎化. What we see here is no ordinary disagreement about the quality of a poet but rather extreme positions on either side. No other Tang poet divided critics so sharply. Such a split was a consequence of different understandings of what poetry was and should be. These general assessments of Bai Juyi’s work were directed primarily at that body of his poetry written in the Yuanhe reign and immediately thereafter, particularly the “New Yuefu” and the two famous ballads. However, we can see from passages cited by critics from the Song onward that Bai Juyi’s later poems were indeed read. The term that was applied to Bai Juyi’s poetry was su 俗, meaning “common,” “vulgar,” appealing to popular taste. In Bai Juyi’s case this meant many things, from the actual popularity of his famous narrative ballads (which were easy to understand but not particularly vernacular), to the rambling clarity of his longer personal poems in “old style” verse, to some of the poems of his old age that were aggressively su in many ways. Roughly half of Bai Juyi’s immense poetic oeuvre dates from the reigns of Wenzong and Wuzong; that is, from 827 onward.5 In 827, the first year of the Taihe reign, Bai Juyi, aged fifty-six, was already established as a poet. To put this in perspective, Han Yu died at fifty-six, while Liu Zongyuan, Yuan Zhen, Du Mu, and Li Shangyin (not to mention Li He) all died at a younger age. It 4  Chen Youqin 陳友琴, Bai Juyi shi pingsbu huibian 白居易詩評迷匯編 (Beijing: Kexue chubanshe, 1958), 10. Although Bai’s name is conventionally paired with that of Yuan Zhen, it was Bai Juyi who was primarily praised or condemned. 5  Only a small number of poems from the last years of his life have survived.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1240

Owen

is fair to say that had Bai Juyi died at fifty-six, not only would his literary collection have been far smaller, but his reputation and contribution to Chinese poetry would have remained unchanged. We speak here only in literary-historical terms; in his later work Bai Juyi effectively invented a new poetry of old age and wrote some wonderful poems. At the same time, he preserved too many poems, which collectively give the impression of carelessness and repetitiousness.6 In part his style reflects a studied casualness; when he writes lines that are pure vernacular, there can be little doubt that it is a conscious gesture.7 Rather, this “carelessness” was the result of his habit of writing so many poems over so many years that certain patterns of verse seem to have come to him automatically. Other poets wrote as succinctly as possible, whereas Bai Juyi did not hesitate to use unnecessary words and to add a second line when the point had already been made in the first line. Bai Juyi’s facile style was a topic in Song critical discourse, but it was understood as a conscious choice. This was implied by the famous (and certainly apocryphal) anecdote that whenever Bai composed a poem, he would recite it to an old woman and change whatever she did not understand. In Shiren yuxie this anecdote is coupled with a report by Zhang Lei 張耒 (1054–1114) that he had seen the drafts of several poems by Bai, and that these had been extensively revised.8 Facility was obviously a self-conscious value in much, though not all, of Bai Juyi’s poetry throughout his career, and it is easy to believe that many of his earlier poems were carefully revised to achieve the fluent transparency that has made them memorable. However, a large number of the later poems have a roughness and redundancy that suggest a first draft. These poems were rarely anthologized or noted in “remarks on poetry” shihua 詩話. Such apparent haste goes hand in hand with Bai’s growing interest in the quantity of his poetry. The poetry of the last seventeen years of Bai Juyi’s life, written during his residence in Luoyang, was the result of a decision that was, in its own way, every bit as radical as Tao Qian’s decision to withdraw from public life. As was the case with Tao Qian, justifying and celebrating that decision became a 6  Not only do we find repetition among the later poems, but Bai would also redo earlier poems. For example, compare “Too Lazy to Be Able To” 慵不能 (23237; Zhu 1505) of 830 with “On Laziness” 詠慵 (21999) of 814. 7  See, e.g., the last line of “A Question for the Young Man” 問少年 (24056; Zhu 2188), which is explicitly framed as speech: 作个狂夫得了無. 8  Wei Qingzhi 魏慶之, Shiren yuxie 詩人玉屑 (Shanghai: Gudian wenxue chubanshe, 1958), 175. The anecdote and Zhang Lei’s note are given with a comment by Hu Zi, which I have not been able to locate in Tiaoxi yuyin conghua.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1241

The Old Men [ of the Early Ninth Century ]

recurrent, almost obsessive topic in Bai Juyi’s subsequent poetry. Bai Juyi could have played the role of Tao Qian, whom he greatly admired, were it not for one striking difference of which Bai was intensely aware. When Tao Qian decided to withdraw, he had to farm for himself; he had to constantly worry about his survival, and he sometimes faced near starvation. Tao Qian’s decision was a radical one whose consequences were always a source of anxiety. Bai Juyi effectively sought retirement in the “Luoyang assignment” as a court official with a 3a rank, that is, he was a Regency official in the top echelons of the imperial bureaucracy, wore the “gold and purple,” and was on intimate terms with ministers and senior statesmen. He didn’t wind up on a ramshackle farm, worrying about how his bean crop was doing. He retired to a modest urban estate (with garden and pool) in the empire’s second city, with sufficient savings, a considerable salary, and a domestic establishment that included singing girls. He no longer rode the political roller-coaster of Chang’an, but he was not Tao Qian. He had to establish a new poetic position for himself and his situation. He did so by means of a humorous twist on the old distinction between “greater” and “lesser” hermits: he was the “hermit in between” 中隱. 白居易, 中隱

Bai Juyi, The Hermit in Between9 大隱住朝市 小隱入丘樊 丘樊太冷落 朝市太囂諠 不如作中隱 隱在留司官 似出復似處 非忙亦非閑 不勞心與力 又免飢與寒 終歲無公事 隨月有俸錢

The greater hermits stay in court and market, the lesser hermits enter the cage of the hills.10 The cage of the hills is too cold and dreary, court and market are too noisy.11 It’s better to be a hermit in between, hermit in an auxiliary post. It resembles service as well as retirement, not too busy, not idle either. Without taxing mind or energy, I can also avoid hunger and cold. No public duties all year long, but I get my salary every month.

9  23223; Zhu 1493. See the discussion of this poem in Xiaoshan Yang’s Metamorphosis of the Private Sphere: Gardens and Objects in Tang-Song Poetry (Cambridge, Mass.: Harvard University Asia Center, 2003), 38–39. 10  The “cage of the hills” or “fenced-in area in the hills” was a Southern Dynasties term for the place to which a recluse would withdraw. 11  As an example of repetition, see 24416; Zhu 2483.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1242

Owen 君若好登臨 城南有秋山 君若愛遊蕩 城東有春園 君若欲一醉 時出赴賓筵 洛中多君子 可以恣歡言 君若欲高臥 但自深掩關 亦無車馬客 造次到門前 人生處一世 其道難兩全 賤即苦凍餒 貴則多憂患 唯此中隱士 致身吉且安 窮通與豐約 正在四者間

If you like to go climbing for the view, there are autumn mountains south of the city. If you love wild excursions, there are spring gardens to the east. If you want to get drunk, you can go out often to parties. With the many good gendemen of Luoyang you can talk merrily, as much as you please. And if you want to rest in solitude, just hide deep behind barred gates. There are also no visitors in fine coaches unexpectedly showing up at your door. A person lives only one life, and you can’t have the best of both worlds. If poor, you suffer from cold and want, when rich, you have many troubles and cares. Only such an in-between hermit can bring himself peace and good luck. Failure and success, opulence and straits— he is right in between these four.

Although old-style poems such as this one generally have more logically lucid expositions than regulated poems, Bai Juyi can be particularly pellucid. He is trying to make a space for himself between equally uncomfortable alternatives—and the issue is indeed his personal comfort, put forward in explicit terms that have few precedents in the tradition. This was one of the problems of scholars from “poor families” entering the bureaucracy: they depended on their salaries for a comfortable life and could not, like officials from wealthier backgrounds, go into high-minded withdrawal from public life without jeopardizing not only their own livelihood but that of the often considerable domestic establishments they had acquired. No Tang poet talks about his salary and his domestic establishment as much as Bai Juyi. No doubt many others were equally preoccupied with this subject, but Bai Juyi immortalized it in his verse and prose. As Bai Juyi well knew, court service was unstable. Apart from the aftermath of the Sweet Dew Incident of 835, few lost their lives; but for a poet in his late fifties and sixties—one who liked his creature comforts—the prospect of being sent off to some remote prefecture in the far south because a friend or marriage relation fell from power was not an attractive one. The “Luoyang assignment” was thus ideal. It was an escape that Bai had himself chosen; and he

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old Men [ of the Early Ninth Century ]

1243

can be forgiven some gentle gloating when those who mocked his choice were later packed off to the pestilential south in advanced old age. The following couplet probably refers to Li Zongmin’s exile to Chaozhou in 835:12 今日憐君嶺南去

Today I pity you,  going off south of the Alps, 當時笑我洛中來 back then you laughed at me  for coming to Luoyang. To enjoy such unencumbered economic security and leisure was one thing; to write about it was another. The recurrent discourse of personal comforts and money was perhaps one aspect of his work that divided Bai Juyi’s readers. This was, according to one view, the very essence of vulgarity (recall Sikong Tu’s image of Bai as a “domineering merchant in the marketplace,” suggesting someone with power and money). Bai Juyi’s somewhat uncomfortable poetic genius derived from a certain social blindness: he stood at the center of his universe; poetically he often stepped back a few paces to contemplate and admire himself, to comment on his great good fortune. If, in old age as in youth, he occasionally experienced embarrassment when encountering the sufferings of people of lower social status, there was something peculiarly egotistical and self-congratulatory about it, not unlike the whiff of envy we catch when he writes of those who are wealthier.13 As the “hermit in between,” he is always measuring where he is and comparing himself to others, either real people or speculative models. As many critics have observed since the Song, Bai Juyi likes to count things: he counts his age; how many years are left to him; how many years between the present and some moment in the past; how many things he has; or, implicitly in the poem above, the advantages of his position.14 He compares himself to those who are better and worse off. The changing world is quantifiable, so that one knows one’s gains and losses. It is important always to take stock.

12  24042; Zhu 2176. 13  In “A Poem I Wrote on Being Stirred by a Newly Finished Fine Silk Padded Jacket” 新制綾 襖成感而有詠 (23826; Zhu 1986) Bai does express the wish, in very conscious imitation of Du Fu, to have a great cape to cover all Luoyang; but in Bai Juyi’s case, in striking contrast to Du Fu, it follows a vivid celebration of his personal comfort. See also “Year’s End” 歲暮 (23863; Zhu 2016). 14  See Hong Mai’s 洪邁 (1123–1202) Rongzhai suibi 容齋隨筆 for a dramatic list of lines on his age. See also Chen Youqin, Bai Juyi shi pingshu huibian, 118.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1244

Owen

The Tang imperial system of bureaucratic and social rank was intended to control the “place” of the elite by claiming a monopoly on social value. For such a system to function through the agency of the government—rather than social value inhering in a person through birth—the government had to continually exercise its power over individual members of the elite to add and subtract social value. The imperial government invested a great deal of wealth and energy in conferring rank and title and in constandy moving members of the bureaucracy, both between positions and geographically (far more frequently in this part of the ninth century than in earlier eras). Members of the bureaucracy, in turn, sought to move “up,” and when they were demoted “down,” they sought to move back up. It was a system predicated on mobility, realized through a currency of prestige that was a government monopoly. And since income was a function of rank, rank-prestige could be linked to financial security for those who had no independent source of family wealth. A system of social mobility through quantifiable value, even on a restricted social scale, clearly has formal counterparts in early modern Europe. In the Song—particularly in the Southern Song—a commercial culture would develop in competition with and woven into the rank-salary structure of value that remained a government monopoly. In the Tang, however, the restriction of merchants and the contempt in which they were held was perhaps a function of the recognition that this represented a structure of value that might compete with the very basis of the central government’s social power. Although we can see in prose, in anecdotes, and in history that relations of social “value” and mobility were central concerns of the Tang elite, this was generally hidden in poetry or expressed through sanctioned tropes, such as the desire for someone (of higher standing and power) to recognize one’s worth or the plea for preferment. The “poetic” world generally involved a circumscribed range of sentiments and a physical world made up of things possessing a poetic aura. Such a version of the “poetic” had its own social function as an alternative to the hierarchy of social value that was so much a visible part of the Tang world. Circles of poets often ignored differences in social hierarchy. Although he had withdrawn from the precarious mobility of high officialdom, of all Tang poets Bai Juyi most fully internalized that social structure of value and brought it to the surface—in the sense of always tallying up what he possessed and comparing that with what others had. Like Tao Qian before him, he claimed to be satisfied with what he had. Yet he “had” so much more of everything—rank, years, friends, poems, salary, and possessions—that his claims of satisfaction take on a very different tone. Bai Juyi often celebrates both utilitarian and ornamental domestic objects that were not part of the usual repertoire of “poetic” things treated by other

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1245

The Old Men [ of the Early Ninth Century ]

poets. The way in which he poetically figures his relation to these things is often of considerable interest. 白居易, 别氈帳火爐

Bai Juyi, Parting from My Felt Curtain and Brazier15 憶昨臘月天 北風三尺雪 年老不禁寒 夜長安可徹 賴有青氈帳 風前自張設 復此紅火爐 雪中相暖熱 如魚入淵水 似兔藏深穴 婉軟蟄鱗蘇 溫燉凍肌活 方安陰慘夕 遽變陽和節 無奈時候遷 豈是恩情絶 毳簾逐日卷 香爐隨灰滅 離恨屬三春 佳期在十月 但令此身健 不作多時别

I recall recently in late winter weather the north wind and three feet of snow. Getting old, I couldn’t stop feeling cold, how was I to get through the long nights? Luckily I had a green felt curtain, I hung it up against the wind. Also there was this red brazier that warmed me up in the snow. I was like a fish diving into deep water, like a rabbit hiding deep in his hole. Tender and gentle, the wintering scales revive, poached in warmth, frozen flesh revitalized. But then those dark and gloomy evenings changed instantly to a time of balmy light. It’s the seasons moving inevitably on— of course my affection has not ceased. The frizzy curtain is rolled up with the days, the ashes die in the fragrant brazier. Parting’s pain belongs to springtime, our tryst will be in the tenth month. If only this body stays healthy, we will not be parted for long.

Hidden behind this ninth-century poem is the famous fan poem attributed to Lady Ban, in which the fan—figured as or a figure for the harem favorite— worries about being “put away” when the cool autumn wind comes, when passion slackens. Bai Juyi’s domestic objects protect him from the cold but likewise suffer rejection, in this case when warm weather comes. In this now domestic drama between a person and his objects, Bai Juyi has placed himself at the center as the speaker and the person with the power to give or withhold favor, which is value. He humanizes his objects, giving them eyes to look at him with desire: they want to be used and valued. His drama of 15  23189; Zhu 1455.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1246

Owen

ownership is one in which his own pleasure in use and possession is displaced into inanimate things that look upon him with a desire to be possessed and cared for. He does, however, want to be benevolent. He begins in personal discomfort, and his comfort is provided for. (Recall the rhetoric of imperial “anxiety,” the sovereign’s discomfort that sends armies out to conquer places and restore the imperial sense of well-being.) However, in contrast to the fan poem, in which gratitude is desired but not expected by the fan, Bai Juyi is generously grateful. When the season comes to put away the instruments of his comfort, he assures them that they are not being rejected; they will only experience a brief period of separation, after which they will be received again with equal love. Bai Juyi’s enduring concern for his body and health returns as a final qualification of his reassurances: if the object-as-lover wants to be reunited with the beloved Bai, it must hope for Bai’s continuing good health. From shivering helplessness Bai has humorously empowered himself. The figure addressed in Lady Ban’s poem, the person who has the power to give or withhold favor, is the emperor. The poet has written himself into this imperial role among his household possessions. They are his harem or his officials, who gain or lose value through his favor and his need. We have here an economics of variable value granted by favor of a central authority in recognition of utility. To this a second system of value should be added, a system of capital that touches on the first system in various ways. Ye 業 is perhaps best translated as “capital” in a broad sense—whether financial, scholarly, cultural, or karmic. It is something one can inherit, accumulate, and pass on. Property, bureaucratic and religious merit, and learning can be understood this way. At the beginning of Wenzong’s reign Bai had been called to the capital to become Director of the Imperial Library (rank 3b). For the first time in his life he received the “gold [seal] and purple [sash],” the mark of a high court officer. This was, as the poet wrote, “glory in this age.” If he had had sons who had lived, they could have inherited office by yin privilege, a fact about which he remarked elsewhere. But the capital of “name” that might have gone to his descendants in the male line was wasted. Later, when a son died in infancy, here is how he ended a poem to Yuan Zhen and Cui Xuanliang:16 文章十帙官三品

Ten cases of writings,  an office of the third rank— 身後傳誰庇廕誰 to whom will I pass the one on after my death,  who will get the yin privilege? 16  23816; Zhu 1978.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1247

The Old Men [ of the Early Ninth Century ]

The yin privilege, his capital accumulated through long service in the bureaucracy, was wasted. His writings, as a quantifiable legacy, would go somewhere and bear his name, even if not to a descendant in the male line. If he wrote—or saved—too many poems it was his chanye 產業, the “capital produced,” the “inheritance” for the future. A more conventional chanye would be quantifiable; his poetic production became that. In his numerous prefaces and letters from this period he is always counting how many poems he has produced and how many have been added to the current store. It is as if instead of working at accumulating capital of goods and prestige for his descendants (who would bear his “name after his death”), he is accumulating poems for that “name.” 白居易,初授秘監幷賜金紫閒吟小酌偶寫所懷

Bai Juyi, On First Being Appointed Director of the Palace Library and on Being Granted the “Gold and Purple,” a Leisurely Verse on Drinking a Little, at Which I Chance to Describe What Is on My Mind17 紫袍新秘監 白首舊書生 鬢雪人間壽 腰金世上榮 子孫無可念 產業不能營 酒引眼前興 詩留身後名 閒傾三數酌 醉詠十餘聲 便是羲皇代 先從心太平

A purple gown, new Library Director, white-haired, a former student. Snow at my temples, old age among mortals, gold at my waist, glory in this age. Nothing to brood on about descendants, no busying myself with providing inheritance. Ale brings elation to my eyes, poems will preserve my name after my death. Idly I pour myself a few cups and drunk sing out a dozen lines. Even if it is the age of a sage-king, one should first set the heart at peace.

A few years later he “set his heart at peace” in the “Luoyang assignment,” but he retained a strong sense of his literary oeuvre as a physical and quantifiable capital that would be passed on. His bookcase resembled nothing so much as a miser’s chest.

17  23471; Zhu 1711.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1248

Owen 白居易, 題文隼櫃

Bai Juyi, On the Cabinet for My Literary Collection18 破柏作書櫃 櫃牢柏復堅 收貯誰家集 題云白樂天 我生業文字 自幼及老年 前後七十卷 小大三千篇 誠知終散失 未忍自棄捐 自開自鎖閉 置在書帷前 身是鄧伯道 世無王仲宣 只應分付女 留與外孫傳

I broke up cypress to make a book cabinet, the cabinet sturdy and the cypress strong. Whose collection is stored there?— the heading says “Bai Letian.” My lifetime’s capital is in writing from childhood on to old age. Seventy scrolls from beginning to end, in size, three thousand pieces. I know well that at last they will be scattered, but I cannot bear to rashly throw them away. I open it up, I lock it tight, placing it by my study curtain. I am childless Deng You, and there is no Wang Can in this age.19 I can only entrust it to my daughter to keep and pass on to my grandchild.

Though translated as a noun, the “capital” of the fifth line, ye, is used as a verb here, loosely translated as “to build capital” or “to build a legacy.” It is literally inscribed with his “name.” It is quantified and—most interesting—is locked up when the poet is not using it. The fatalistic recognition that it will be scattered someday could temporarily be countered by a son who would inherit both the accumulated cultural capital and the name. In Bai Juyi’s case this capital will have to go “out,” wai (the untranslatable qualifier of the grandchild in the final line, who is on the “distaff” side, “outside” the main line of the family as a result of having a different surname). Bai Juyi’s passion for counting and measuring overflows his spiritual economy of value: it becomes a form of representing the world and goes hand in hand with his poetic loquacity. The true Bai Juyi of the 830s and 840s does not appear fully in the handful of poems commonly anthologized or translated but rather in the sea of verse in which these better-known poems float.

18  23915; Zhu 2072. See the discussion in Christopher Nugent, “The Circulation of Poetry in Tang Dynasty China” (Ph. D. diss., Harvard University, 2004), 254–55. 19  This probably refers to the anecdote of the famous literary man Cai Yong 蔡邕 meeting the adolescent Wang Can; Cai was so impressed with Wang Can’s talent that he immediately thought he should give the young man all his books. Sanguo zbi 597. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1249

The Old Men [ of the Early Ninth Century ]

白居易, 答崔賓客晦叔十二月四曰見寄 (來篇云, 共相呼喚醉歸來) Bai Juyi, Answering the Poem Sent by Tutor Cui Huishu [Xuanliang] on the Fourth Day of the Twelfth Month (The poem he sent had the line: “Let us call to each other to go back drunk”)20 今歲日餘二十六



來歲年登六十二



尚不能憂眼下身



因何更算人間事



居士忘筌默默坐



先生枕麴昏昏睡



早晚相從歸醉鄉



醉鄉去此無多地



The days that remain in the present year  are twenty-six, next year’s rich harvest—  the age of sixty-two. Still I cannot worry about  this body that I see, why should I calculate any more  affairs in the mortal world? This recluse has forgotten the fish-trap  and sits silently,21 you, sir, pillow your head on mash  and sleep in a stupor. Sooner or later we will go together  back to the Land of Drunkenness— from here to the Land of Drunkenness  is not very far.

Some of the points mentioned earlier should be apparent here: an eight-line poem performs three calculations (ll. 1, 2, 8), and in the fourth line the poet invokes calculation by refusing to “calculate.” Bai Juyi is always reckoning his years and time; probably only he would have noticed the inverse symmetry of 26 and 62. The loquaciousness of the above poem—despite the poet’s claims to have “forgotten the fish-trap” of language—is of a very special sort. The apocryphal anecdote in which Bai revised his poetry to make it understood by an old woman undoubtedly grew out of the recognition that Bai Juyi’s poetry was widely comprehensible aurally. This comment requires some nuance. Middle Chinese of the Tang had many more phonemes than modern Mandarin; thus, there was less potential ambiguity in recognizing both individual words and compounds. If we recognize “poetic Chinese” as a special idiolect with its own habitual situations, grammatical patterns, and lexicon, then most Tang poetry was comprehensible aurally by those fully familiar with the idiolect. Although 20  23193; Zhu 1461. 21  Recall Zhuangzi’s parable of language: when you get the fish, you forget the fishtrap; when you get the meaning, you forget words. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1250

Owen

Bai Juyi kept the formal constraints of “poetic Chinese,” he preferred a lowregister lexicon that made it accessible to a wider audience. If standard poetic Chinese preferred hua hong 花紅, “the red of flowers,” Bai Juyi would write hong hua 紅花, “red flowers.” Apart from a few compounds that involve allusions (ironically including “forgetting the fish-trap”), Bai Juyi’s entire poem above is made up of very simple compounds and compound phrases. This worked best in the long line (which remained the dominant form of oral popular poetry from the Tang through the twentieth century). Bai Juyi was quite capable of writing complex, erudite verse—though he had difficulty sustaining the “high style” throughout a poem. He was predisposed to write the kind of poem cited above, of which he was fully conscious. The poem is careless, but it is self-consciously so; it is a style “marked” as su 俗 (the “common,” the “low,” the “popular,” the “vernacular”).22 In his old age he wrote far less about the art of poetry than he did in his earlier poetry (although he continually spoke of composing poems). We have some texts reflecting on the changhe 唱和 form, in which poets “matched” poems by friends. But perhaps the most significant piece for his later view on poetics can be found in an 828 poem and preface for the monk Daozong, whose Buddhist poetry is written for the salvation of souls rather than “for the sake of poetry” 不爲詩而作.23 He praises Daozong’s poetry for its formal perfection and clarity of exposition: “No error in a single tone, / four lines in organized sequence” 一音無差别, 四句 有詮次. He praises how Daozong “lets his words go freely in a relaxed manner, / drifting off and leaving [the art of] the written word” 從容恣語言, 縹緲離文字. The description perfectly describes Bai Juyi’s own poetry—though without Daozong’s religious purpose. Bai Juyi further favorably contrasts Daozong’s kind of Buddhist poetry with the works of the poet-monks of the late eighth century, masters of regulated verse in the short line, who did indeed write “for the sake of poetry.” Bai Juyi was in Chang’an when he composed this poem, and he can hardly have been unaware of the popularity of regulated-verse masters such as Jia Dao and Yao He, poets who clearly much admired those very poet-monks who wrote “for the sake of poetry.” What we see here in the second quarter of the ninth century is the first stirring of an opposition of high and low registers as a type of stylistic choice.24 22  Here one might contrast Du Fu, famous for his use of vernacular phrases in poetry. In Du Fu’s works the vernacular and high “poetic” are easily integrated rather than articulated against one another. 23  23180; Zhu 1445. 24  Earlier poets had sometimes engaged in individual play on register. Beginning in the late eighth century there was a publicly recognized opposition between registers associated Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old Men [ of the Early Ninth Century ]

1251

The poetics of the “low style,” su, had profound consequences in Chinese poetry. By foregrounding ease and immediacy in composition, the poetics of the “low style” contributed negatively to the “high style,” which increasingly celebrated the traces of effort and time spent in composition. As we shall see in the next chapter, this was the poetics of “bitter chanting” or “painstaking composition,” kuyin 苦吟. The dynamics of register set in motion went far beyond two groups of poets with different values. Kuyin in regulated verse was very much the art of the parallel couplet, and these poets would often frame their “high style” couplets with a closing couplet (and occasionally an opening couplet) in a “low” register, often very vernacular and reminiscent of Bai Juyi. Such interplay between high and low register passages became an essential part of the poetics of song lyric (ci 詞). In embryonic form the emergence of a “low” or vernacular aesthetic led to the opposition between it and the “classical” or “poetic” that was to play such an important role in later literary culture. Bai Juyi’s adoption and cultivation of a low poetic register was not a neutral act but rather a stylistic instantiation of a set of values that are thematically reiterated throughout his poetry, especially his later poetry. As the poet often says, he is doing what suits him and following his nature. He acts on whim and responds to situations spontaneously. In short, in Bai Juyi the low register is part of the claim of the “natural,” and it remained part of various versions of such a claim later in the tradition. A claim about behaving “naturally” or composing poetry “naturally” is very different from simply “being natural.” From a broad perspective it would be hard to argue that a human being is ever capable of not being “natural.” A claim about behaving naturally is based on an idea of what the “natural” is; for such a person the “natural” has certain attributes; it is a value and thus something to be desired. It is negatively defined against some state or behavior—usually social—conceived as somehow “unnatural,” constraining or violating the self. In its essential negativity the “natural” must break out of constraint; it must appear against something else. It is confirmed only in being seen from the outside. (After all, how can one know one is “behaving naturally” without recognizing it from the perspective of an imagined alternative?) Bai Juyi and many poets that followed in his lineage had a genius in picturing themselves. The poet sees himself as he must look. Indeed, from his earlier poetry Bai Juyi possessed a remarkable ability to picture and imagine himself as the object of the perception of others. The following poem, from with the “ancient,” gu 古, and the “modern.” The “ancient” style, however, claimed an aura of ethical value, lacking in Bai Juyi’s “low” style. The “poetic” or “high” style was very much in the lineage of the “modern” in the opposition between “ancient” and “modern.” Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1252

Owen

the Luoyang period, captures one of Bai Juyi’s characteristically humorous domestic moments, ending with the poet’s witty observation, which is also an observation of self. 白居易, 偶眠

Bai Juyi, Falling Asleep by Chance25 放盃書桉上 枕臂火爐前 老愛尋思事 慵多取次眠 妻教卸烏帽 婢與展青氇 便是屏風樣 何勞畫古賢

I set down my cup on the desk, rest my head on my arm before the burning stove. Old, I am fond of thinking back on things, lazy, I often fall asleep at random times. My wife makes me take off my black cap, the maid spreads a green blanket on me. Exactly the pose you see on a screen!— why bother to paint ancient worthies?

To write a poem about nodding off, you have to be awake. Apparently the poet’s wakefulness is the result of his loving but officious household—his wife making him remove his hat (or removing it herself) and the maid covering him with a blanket. To fall asleep this way is indeed “natural,” but clearly from another perspective, provided by the women of the household, it is all too natural and needs to be done the “right” way. What is, of course, striking about the poem is that once the person dozing is properly attired (or unattired), he is no longer the sleeper but the outer eyes seeing himself as a sleeper. In precise terms it is a yang 樣, translated as “pose,” but also a “fashion” or “manner.” He sees himself as a figure in a screen painting, and in this pose he obviates the need for actual paintings. The representation of being seen occurs often in Bai Juyi’s poems, particularly in the closing section. In some cases it is less seen than listened to. 白居易, 偶吟

Bai Juri, Chanting by Chance26 里巷多通水 林園盡不扃 松身為外户 池面是中庭

Much water is passing through the ward lanes, none of the groves and gardens are barred. The pine tree is my outer door, my courtyard is the surface of a pool.

25  23487; Zhu 1725. 26  23674; Zhu 1887.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1253

The Old Men [ of the Early Ninth Century ] 元氏詩三帙 陳家酒一瓶 醉來狂發詠 鄰女映籬聽

Three cases of Yuan Zhen’s poems, a bottle of Mr. Chen’s ale. When I get drunk, I sing out wildly, a neighbor girl listens, half hidden by the hedge.

Bai Juyi gives us a flooded city, in which everyone is apparently isolated and the poet is kept at home and barred from his usual excursions. What remain to him are the two things most commonly put in parallel positions (as here) in party poems: “poems” and “ale” (conveniently, a level tone balancing a deflected tone). Inaccessible friends in Luoyang are replaced by the poems of a more remote friend, Yuan Zhen. Left to his own devices, the poet recites the poems “wildly,” kuang 狂, heedless of normal restraint. His heedlessness is, however, heeded and he heeds the person heeding his heedlessness. There is a characteristic satisfaction in noting the neighbor girl “half hidden by the hedge,” listening to him and no doubt thinking: “what a crazy old man!” Kuang is used often in Bai Juyi’s poetry to describe the poet’s recitation or behavior. It is an interesting term in that it already incorporates an outside perspective. Perfectly “natural” and spontaneous behavior would have no outside from which to recognize itself: the poet who knows he is kuang is already picturing himself in relation to some other standard. To picture oneself is one thing; to be pictured by another is quite another. One of Bai Juyi’s finest poems, written in 810 when he was at the height of his poetic powers, consists of a playful contemplation of his own portrait. 白居易, 自題寫眞 (時爲翰林學士) Bai Juyi, On My Portrait (at the time I was a Hanlin Academician)27 我貌不自識 李放寫我眞 靜觀神與骨 合是山中人 蒲柳質易朽 麋鹿心難馴 何事赤墀上 五年爲侍臣 况多剛狷性 難與世同塵

I didn’t recognize my own face, Li Fang painted my portrait true. Calmly I observe the spirit and the bone structure: this must be some man of the mountains. Willow wood and rushes easily rot away, the heart of a deer is not to be tamed. Why then in the court’s red pavements have I served in attendance for five years? Worse still, this nature so hard and blunt cannot share the dirt of the world.

27  21968; Zhu 311.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1254

Owen 不惟非貴相 但恐生禍因 宜當早罷去 收取雲泉身

Not only not a face with signs of nobility, I fear only it will be the cause of misfortune. Best then to quit and go as soon as I can, to take away this body of clouds and streams.

Here Bai Juyi wittily plays the role of the physiognomist, reading his nature in his portrait. He pretends to discover his “real,” “natural” self in the process, a self that does not belong in the court role in which he is represented. We know that this is play, that his “discovery” is what he already knows about himself; and yet the poetic move he makes is a significant one, seeing himself from the outside in order to discover what is “true” or “genuine,” zhen 眞, the term for portraiture, xiezhen 寫眞, “depicting the true.” In 829, when still Vice Director of the Board of Punishments in Chang’an, Bai Juyi saw that same portrait again—and perhaps remembered his old poem. 白居易, 感舊寫眞

Bai Juyi, Moved by an Old Portrait28 李放寫我眞 寫來二十載 莫問眞何如 畫亦銷光彩 朱顏與玄鬢 日夜改復改 無嗟貌遽非 且喜身猶在

Li Fang painted my portrait true, since he painted it, twenty years have gone by, Don’t ask how the true original looks, the painting has also lost its gloss. Rosy complexion and dark locks of hair change and keep changing day and night. Sigh not that the face is suddenly not what it was— just be glad that the body is still here.

Bai Juyi begins with an exact quotation from his earlier poem, but the “true” quality of the portrait has become complicated. In place of physiognomy that reads unchanging nature, the poet now reads aging. Even though he tells us not to ask about the “true [original],” zhen, the use of “also” lets us know that the “true” has changed, like the portrait itself. Still, the painting and its original have parted company: the painting may have lost its gloss, but it still has traces of the “rosy complexion and dark locks of hair” that are gone from the original— an original that had not yet taken the playful advice of 810 to get out of office. Although here the once-legible body has simply become the surviving body, Bai Juyi’s fascination with stepping outside himself and taking a snapshot of 28  23219; Zhu 1491.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1255

The Old Men [ of the Early Ninth Century ]

himself exerted a profound influence on the tradition; from the Southern Song on, there is a familiar type of quatrain in which the poet pictures himself in the last line. Here is Bai Juyi in a quatrain from 838. 白居易, 東城晚歸

Bai Juyi, Returning Late in the Day in the East of the City29 一條邛杖懸龜榼

A single staff of Qiong bamboo,  a tortoiseshell mug suspended, 雙角吳童控馬銜 a Wu servant lad with a pair of tufts  leads the horse’s bit. 晚入東城誰識我 Late I enter the east of the city,  who recognizes me?— 短靴低帽白蕉衫 short boots, a low hat,  a tunic of white plantain fiber.

It is a gaffer’s fashion show, and we know this gaffer is a bibulous poet from his outfit. It is, in effect, iconography inviting painting, legible even when portraiture that captures the “true” appearance fails to account for a changing body. Such a vignette of self-portrayal, involving not the features of the face but visible markers of a “type,” would itself become a poetic “type” later in the tradition. 白居易, 自詠

Bai Juyi, On Myself30 鬚白面微紅 醺醺半醉中 百年隨手過 萬事轉頭空 臥疾瘦居士 行歌狂老翁 仍聞好事者 將我畫屏風

Whiskers white, the face faintly red, in a tipsy, half-drunken state. Life’s hundred years pass in a snap, ten thousand cares gone in the nod of a head. Lying sick, a gaunt man in retirement, going singing, a wild old man. And I heard that some who are interested have painted me on screens.

29  24255; Zhu 2359. 30  24259; Zhu 2362.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1256

Owen

Bai Juyi’s theatrical “naturalness” is, to his credit, only one aspect of his later poetry—though a very large one. He could also be unctuously polite, celebrate the austerities of Chan, and lament family and friends. In addition, he could also laugh at himself. Bai Juyi was immensely fond of his Luoyang home, his Taihu rock, his cranes, his garden, his comfortable clothes and blankets, his small pleasures. Sometimes he let things go, always celebrating his act of renunciation in a poem. One of his favorite things was the skiff he had transported to Luoyang from Suzhou, after he finished his term as governor there. It served him well on his little pond. Bai was usually cheerful—or tried to be—when facing old age, though the topic recurs with such frequency that we can see it was always on his mind. In 839 he had a stroke that left him partially paralyzed in one leg. As any householder knows, constant vigilance is necessary; otherwise things break down. After he became somewhat mobile again, he went out into his garden and surveyed the scene: 白居易, 感蘇州舊舫 Bai Juyi, Stirred by My Old Skiff from Suzhou31 畫梁朽折紅窗破

My painted beams have decayed and snapped,  the red window is broken, 獨立池邊盡日看 I stand alone beside my pool  looking all day long. 守得蘇州船舫爛 The skiff that I kept  from Suzhou is rotten, 此身爭合不衰殘 then how should not this body of mine  be wasting away?

Like his portrait and his body, his “things” also deteriorate. His attachment to Buddhism was sincere and seems to have deepened in his later years. We might leave Bai Juyi by recalling a poem of 845, the year before his death. The old sensualist clings to attachments and calls back pleasures in memory, even as he claims to let them go.

31  24310; Zhu 2399.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1257

The Old Men [ of the Early Ninth Century ] 白居易, 齋居春久感事遣懷 Bai Juyi, Living in Abstention Spring Lasts Long; Stirred, I Get Things off My Chest32 齋戒坐三旬 笙歌發四鄰 月明停酒夜 眼闇看花人 賴學空爲觀 深知念是麈 猶思閒語笑 未忘舊交親 久作龍門主 多爲兔苑賓 水嬉歌盡日 雪宴燭通晨 事事皆過分 時時自問身 風光抛得也 七十四年春

I have stayed in abstention for thirty days as song and piping emerge from all the neighborhood. Moon bright on nights I have quit drinking, the eyes dim of the one who looked at flowers. Luckily I study Void for my viewpoint, and well understand that Thought is dust. Still I think of idle chatter and laughter and have never forgotten my old friends. Long I was master at Dragongate,33 and often have been a guest in Rabbit Park.34 Sporting on the water, song lasted all day; feasting in snow, candles burned until dawn. In everything I exceeded the measure, again and again I ask myself why. All spring’s glory has been cast away in this spring of my seventy-fourth year.

Interlude To single out the distinctive concerns of a poet like Bai Juyi is to overlook a significant amount of purely social verse, which was rarely read even though it was the currency of the network in which someone with a reputation as a poet maintained his social connections. For example, any notable public event in the life of a highly placed friend or acquaintance called for a poem—just as we might write a note or send a card in modern times. Each common social situation had a set of topics that should or could be addressed with poetic grace, but an established poet would be expected to provide at least one fine couplet to display his skill. Without delving too deeply into this mode, we might consider just one poem. 32  24506; Zhu 2561. 33  Dragongate was the western side of Fragrant Mountain, Xiangshan, near whose temple Bai had a retreat. 34  This was the park of the Prince of Liang, in the Western Han, and stands for a patron of poets.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1258

Owen

In the spring of 832 Linghu Chu was transferred from a command in the east to become Regent, Metropolitan Governor, and military commissioner in Taiyuan, the Northern Capital. We might recall that it was this same spring that Li Ting, a general with a long history of distinguished service—and one humiliating defeat—was given command of the Wuning army, headquartered in Xuzhou, and declined the appointment on the grounds of ill health, after the Wuning army ate one of his subordinates.35 The army at Taiyuan was less troublesome, and Linghu Chu had a special affection for this city where his father had served as administrator. It is said that the locals had a reciprocal affection for Linghu Chu. When someone of Linghu Chu’s stature was given such a post, congratulations, usually in the form of poems, would come from his friends in the official world. From Suzhou Liu Yuxi sent “Lord Minister Linghu Transfers His Command from Tianping to Taiyuan: Extending My Congratulations in a Poem” 令狐相公自天平移鎮太原以詩申賀, concluding grandly with a comparison of Linghu Chu to Wang Shang of the Han:36 夷落遙知眞漢相

Far away the barbarian tribes  know a true Minister of the Han 爭來屈膝看儀形 and compete to come on bended knee  to behold his deportment.

Bai Juyi, Linghu Chu’s counterpart as Regent in the Eastern Capital, sent his own congratulations in the form of a parting poem. 白居易, 送令狐相公赴太原 Bai Juyi, Seeing Off Lord Minister Linghu, Who Is Setting Off for Taiyuan37 六纛雙旌萬鐵衣

Six pennants, paired banners  ten thousand suits of armor, 並汾舊路滿光輝 the former road of Bing and Fen  fills with radiance. 青衫書記何年去 A clerk in blue student gown,  what year did you leave? 35   Zizhi tongjian 7879. 36  19029. After Wang Shang’s bearing intimidated the Shanyu of the Xiongnu, the emperor exclaimed that he was a “true Minister of the Han.” 37  23627; Zhu 1864.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1259

The Old Men [ of the Early Ninth Century ] 紅旆將軍昨日歸



詩作馬蹄隨筆走



獵酣鷹翅伴觥飛



北都莫作多時計



再爲蒼生入紫微



now a general with red streamers  you recently returned. In poems you make the horse’s hooves  speed along with the brush; hunting, you intoxicate the hawk’s wings,  flying together with drinking horns. Do not make long-term plans  to stay in the Northern Capital— for the sake of the people once again  enter the Purple Precincts.

Such poems possess a distinct quality of social ritual: they contain a series of topics that must be mentioned in a roughly predetermined order. The particulars of the situation and Linghu Chu’s stature map most of Bai Juyi’s poem for him. He needs an opening couplet on the entourage to represent Linghu Chu’s prestige in transit. Voyaging to take up an important provincial post was political theater: the incumbent would travel with banners, insignia, an entourage, and troops commensurate with the importance and nature of the post. Since Taiyuan was where Linghu Chu started out as a much-admired young graduate, Bai could not resist the couplet contrasting the young student with the mature Regent. Finally, a parting poem begs for a conclusion expressing the hope that the person leaving will return soon. Bai Juyi nicely twists that convention into a gracious compliment, asking that, “for the sake of the people,” Linghu Chu return soon to Chang’an and the imperial palace.38 This leaves Bai Juyi only one couplet (the third) to demonstrate his prow‑ ess as one of the empire’s best-known senior poets. Generals entertain themselves by galloping in the hunt; literary men (and civil officials) entertain themselves by writing poetry and drinking. Linghu Chu combines the virtues of a military man (a military commissioner) and a literarily inclined civil servant. When someone with poetic talent writes, his brush “speeds,” as do his horse’s hooves. At a drinking party the goblets are said to “fly,” as does the hunting hawk, metaphorically intoxicated by the hunt. In its peculiar way the third couplet combines all these elements while maintaining perfect parallelism. The ingenuity, however, fills one of the requirements of such poems: it must

38  The choice of words often has a weight of association that borders on allusion but does not rise to explicit allusion. To choose the phrase cangsheng 蒼生 for “the people” recalls Shishuo xinyu xxv.26 where Gao Song hopes the great statesman Xie An will give up retirement and come to the capital for the sake of “the people” (cangsheng).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1260

Owen

somehow refer to the post (or, in this case, the multiple posts) that the incumbent is to assume. Linghu Chu would have written back to both Bai and Liu, either separately or jointly. Linghu Chu’s response does not survive, though we do have a poem from Linghu Chu to both Bai and Liu, responding to poems they wrote earlier when he assumed command of the Xuanwu army.39

Liu Yuxi 劉禹錫, 寓興 Liu Yuxi, Figurative (second of two)40 世途多禮數 鵬鷃各逍遙 何事陶彭澤 拋官爲折腰

The world’s ways have many different customs: peng bird and sparrow each roam free. Why, then, did Tao Qian of Pengze drop his post because he had to bow low?

Although Liu Yuxi has, in the past half century, become known and approved for his imitations of Southern folksongs (and for his youthful political affiliation with the Wang Shuwen government of 805), these hardly represent his poetic oeuvre as a whole and do not capture the poet in his later years. Like Bai Juyi, a chronological overview of his works reveals that about half his poems were composed beginning in Wenzong’s reign. Yet the numerous modern anthologies of Liu’s poetry tend to represent that prolific period of his career with only a few poems, often seemingly chosen out of a sense of embarrassment about leaving his last years in silence. Far more than for Bai Juyi, Liu Yuxi’s later poetry was produced in response to his extensive social network. More often than not, Liu is responding to the poems of others; when not responding, he is sending a poem hoping for a response. This is not necessarily a bad thing, but Liu was often corresponding with men who were both politically powerful and poetically mediocre. Many of his poems are responses to those of senior statesmen, such as Pei Du and Linghu Chu, and somewhat younger men at the height of their power, such as Li Deyu and Niu Sengru; the surviving poems of Pei Du and Linghu Chu set a level of banality that Liu Yuxi matches, with an added touch of servility. Liu was obviously prominent—even more so than Bai—as an exchange poet, but one can only sympathize with Wenzong’s 39  17712. 40  19176; Qu (1989) 560.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1261

The Old Men [ of the Early Ninth Century ]

silence at the prospect of Liu Yuxi as one of his speculative Academicians of Poetry. Liu Yuxi has a few justly famous poems, but the same is true of many other contemporary poets who were less fortunate in the eyes of literary history. He was more fortunate than he deserved, as Zhang Hu, the striking counterexample, was less fortunate. A significant proportion of Tang poetry was sent to friends or in answer to poems sent by friends. In this period we begin to find many collections of such poetry, called “Song and Response Collections” (changhe ji 唱和集), circulating independendy in the groups of senior poets around Bai Juyi. Liu Yuxi participated with three men in such collections: the Pengyang Song and Response Collection 彭陽唱和集 with Linghu Chu; a collection with Li Deyu; and collections with Bai Juyi. Only a modest number of Linghu Chu’s poems survive. Judging both from those extant poems and the poems Liu Yuxi wrote in response to Linghu Chu, this was no great loss to Tang poetry. In the poems exchanged between Bai Juyi and Liu Yuxi, however, most of the sets are still extant. Along with the poems exchanged between Bai Juyi and Yuan Zhen, these are our best examples of how exchange poetry worked. Bai Juyi’s greatest poetic gift was invention, and even if Bai’s inventive powers declined somewhat in his old age, his poems still provided interesting topics on which Liu could base his responses. Even when Liu writes first, Bai Juyi overwhelms him. Bai is always himself: he can be spiritually “vernacular” in strict form. 白居易, 南園試小樂 Bai Juyi, Putting on a Small Musical Performance in My South Garden41 小園班駮花初發

A riot of color in my small garden,  the flowers begin to bloom, 新樂錚鏦敎欲成 the clang and blare of new music  almost fully learned. 紅萼紫房皆手植 The red stamens and purple calyces  all planted by my own hand, 蒼頭碧玉盡家生 my servant “Sapphires,”42  all born in my own household.

41  23592; Zhu 1821. 42  Sapphire was a famous singer (plural in the text because Bai boasts a number in his service). Thus, Bai Juyi’s household musicians are the children of older bondservants.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1262

Owen 高調管色吹銀字

The colors of the pipes tuned high,  they blow the “silver word”;43 慢拽歌詞唱渭城 song lyrics slowly drawn out  as they sing “Wei City.” 不飲一杯聽一曲 If you don’t drink a cup with me  and listen to a tune, 將何安慰老心情 how will this old man’s feelings  ever be consoled?

Bai Juyi claims to control his domestic spectacle in a peculiar conjunction of disorderly display and ownership. It is the theater of domestically circumscribed excess. Everything that he wants to show is “his.” This is Tang social display in a new key, and it seeks someone to recognize it. Liu Yuxi gives Bai what he wants. 劉禹錫, 和樂天南園試小樂

Liu Yuxi, A Companion Piece for Letian’s “Putting on a Small Musical Performance in My South Garden”44 閒步南園煙雨晴



遙聞絲竹出牆聲



欲拋丹筆三川去



先敎清商一部成



花木手栽偏有興



歌詞自作别生情



多才遇景皆能詠



當日人傳滿鳳城



Walking calmly in the southern garden  the misty rain clears, from afar I hear the strings and piping  come out over the walls. About to give up the red brush  and go off to Three Rivers,45 he first has a performance learned  in the clear Shang mode. Flowering trees planted by his hand  are particularly inspiring, song lyrics he wrote himself  have a distinct mood. When one of great talent encounters a scene  he can always make a verse, and this very day people will pass it  all through Phoenix City.46

43  This is a pipe with an inscription in silver inlay. 44  18998; Qu (1989) 1086. 45  Bai Juyi is about to give up the “red brush” for drafting convictions in his post on the Board of Punishments and to take up a post as Adviser to the Heir Apparent in Luoyang. 46  Chang’an. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1263

The Old Men [ of the Early Ninth Century ]

The third line dates the poem to the spring of 829, just before Bai Juyi left Chang’an to take up his Luoyang post. The poem is, in effect, an extended compliment, praising various elements raised in Bai’s poem. The music passing over the walls recapitulates from an outside perspective Bai’s description of the musicians practicing in his garden. He mentions Bai Juyi’s new office, the “Luoyang assignment.” He praises the effects of Bai as gardener and lyricist. Finally, he praises Bai’s poem itself, whose quality is such that it will be repeated and passed throughout the capital in the course of the day. In his old age Bai Juyi was often a poet of the senses. In the preceding poem he could see the analogy between the riotous display of colors in the flowers and music being practiced for his little performance. No earlier poet wrote so affectionately of his comfortable, soft clothes and blankets. He appreciated the warmth where the sunlight touched his body or the developing bouquet of steeping tea. 白居易, 間臥寄劉同州

Bai Juyi, Resting at Leisure: to Liu of Tongzhou47 軟褥短屏風 昏昏醉臥翁 鼻香茶熟後 腰暖日陽中 伴老琴長在 迎春酒不空 可憐閒氣味 唯欠與君同

A soft quilt, a short folding screen, an old man lying in a drunken daze. After tea is steeped, sweet smell in the nose, in the sunlight my midsection warm. Old age’s companion, a zither ever here, to welcome spring, the ale is not drunk up. Yet too bad, for the flavor of leisure lacks only being together with you.

The poem essentially celebrates how good the poet feels. In order to fully enjoy his pleasures, Bai likes to display them to others. He therefore sends the poem to his friend, here as elsewhere tacking on a polite note in the last couplet, saying, in effect: “I just wish you were here.” In contrast to Bai Juyi’s celebration of what lies close at hand, in his answering poem Liu Yuxi thinks of effects: the tea serves a purpose; ale is not simply meant to “welcome spring” but to work together with herbal medicine. The third couplet is a “poetic” one, not unlike those of the contemporary regulatedverse masters; Liu Yuxi could easily have written it at some other time and dropped it in here to show off.

47  24121; Zhu 2242. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1264

Owen 劉禹錫, 酬樂天閒臥見寄 Liu Yuxi, Answering Letian’s “Resting At Leisure”48 散誕向陽眠 將閒敵地仙 詩情茶助爽 藥力酒能宣 風碎竹間日 露明池底天 同年未同隱 緣欠買山錢

Blissfully content, you lie facing the sun, in leisure rivaling an immortal on earth. Tea aids the briskness of poetic feelings, ale can bring out the effect of medicines. Wind shatters the sunlight amid bamboo, dew brightens the skies at the pool’s bottom. Alike in age, but not alike in retirement all because I lack money to “buy a mountain.”

Bai Juyi is the self-proclaimed “hermit in between” who still draws a comfortable salary. Liu has chosen to read Bai Juyi’s polite “I wish you were here” at least semiseriously. The desire to “buy a mountain,” satirizing a certain attitude toward retirement in its original Shishuo xinyu context, had become in the Tang simply a way to refer to the savings needed to retire in comfort. Liu, the prefectural governor with serious duties, seems to be reminding his friend that not everyone can enjoy such leisure. Liu Yuxi was indeed the same age as Bai Juyi, but he seemed always to be following the more famous poet: Bai had been governor of Suzhou, and Liu was given the same position; later Liu received Bai Juyi’s post as Adviser to the Heir Apparent, Luoyang assignment, as Bai was moved up to Junior Mentor. The coincidence in offices was somehow appropriate for Liu, who was “following” Bai Juyi’s poetic lead in other ways. The impulse to write a “companion piece,” in which a prior poem provides some direction, is nicely demonstrated in the following poem, sent from the capital to Bai Juyi in Luoyang and Yuan Zhen in Zhedong. 劉禹錫, 月夜憶樂天兼寄微之

Liu Yuxi, Moonlit Night, Recalling Letian and Sent to Weizhi As Well49 今宵帝城月 一望雪相似 遙想洛陽城 清光正如此

The moon tonight in the imperial city, in my gaze it is like the snow. I imagine far off there in Luoyang its clear rays are exactly like this.

48  18870; Qu (1989) 1215. 49  18648; Qu (1989) 111.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1265

The Old Men [ of the Early Ninth Century ] 知君當此夕 亦望鏡湖水 展轉相憶心 月明千萬里

And I know that you, Weizhi, this evening are also gazing into Mirror Lake’s waters. This heart, recalling you, tosses and turns, the moon is bright for thousands of leagues.

This is, of course, a hidden “companion piece” to another poem of the same title from a different era. 杜甫, 月夜 Du Fu, Moonlit Night50 今夜鄜州月 閨中只獨看 遙憐小兒女 未解憶長安 香霧雲鬟濕 清輝玉臂寒 何時倚虛幌 雙照淚痕乾

The moon tonight in Fuzhou she just watches alone in her chamber. I am moved by my children far off there who don’t yet know to recall Chang’an. Fragrant fog, her coils of hair damp, its clear glow, her jade-white arms are cold. When will we lean at the empty window, doubly lit, the tracks of our tears dry?

The epigone’s poem stands with some embarrassment beside the original. Du Fu’s domestic imagination is redeployed to Liu’s circle of friends, with Liu placing himself in Du Fu’s part in Chang’an. All three are at this moment gazing at the moonlight, which is either figured as snow or reflected in Mirror Lake’s waters. In place of Du Fu’s children, who are too young to be troubled by memories of Chang’an and their father, Liu Yuxi gives us Luoyang in moonlight in the second couplet. Significantly, he does not indicate that Bai Juyi there is gazing at the moon. In place of Du Fu’s gazing wife in the moonlight, in the third couplet Liu puts Yuan Zhen, who is gazing. Absent is Du Fu’s vision of a future reunion; there is only Liu’s longing and the moonlight that spans the distances of separation. Bai Juyi lived on after his greatest poetic achievements and their era had passed. Only rarely, however, do we catch him looking down upon the present generation and longing for the good old days. Liu Yuxi, by contrast, was often nostalgic. Nostalgia for the “old music” was something of a poetic commonplace, one that seems to have appealed to Liu. 50  10974; Qiu 309.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1266

Owen 劉禹錫, 與歌者米嘉榮

Liu Yuxi, For the Singer Mi Jiarong51 唱得涼州意外聲

For singing the unbelievable notes  of Liangzhou 舊人唯數米嘉榮 of former people only Mi Jiarong  is worth considering. 近來時世輕先輩 Recently our age belittles  the previous generation— 好染髭鬚事後生 best to dye your whiskers  to serve younger men.

This piece is probably from 828, which is somewhat earlier than the period in which we begin to find attacks on the Yuanhe style. Mi Jiarong’s skill in singing “Liangzhou” (one of the then-lost border prefectures, whose music would evoke lost Tang glory in 828) is certainly not the Yuanhe style, but Liu’s comment in the third line of the quatrain is a broader judgment than mere musical taste. The numerous variants of this poem suggest its popularity, and this is the kind of poem that might easily have made its way into the palace and perhaps reached the ears or eyes of the twenty-year-old, heavily bearded emperor. To search for the precise reasons for Wenzong’s apparent dislike of Liu Yuxi as a poet is futile, but Liu Yuxi generously supplies several possibilities. As a young man, Liu Yuxi had participated in the Wang Shuwen government during the brief reign of Shunzong in 805. Whether this was a reform movement, a virtual coup d’etat, or a little of both, members of that government were sent off into long-term exile when Xianzong assumed the throne at the end of 805. In 815 Liu was briefly back in the capital, only to be sent off to the distant provinces again. Only in the early years of Wenzong’s reign did he serve again in the capital. Apart from the short period in 815, Liu Yuxi had been away from Chang’an for well over two decades. In 805, aged thirty-four, he had been a bright young star on an upward career path; back in Chang’an in 828, aged fifty-seven, he had little hope of reaching the highest echelons of government. Some nostalgia was understandable, and hearing the old musicians obviously touched him deeply.

51  19270; Qu (1989) 783. This quatrain has many variants. I have followed the text given in Qu. I have treated Mi as if it were a surname; the name is clearly Central Asian and may be the aural representation of a single name.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1267

The Old Men [ of the Early Ninth Century ] 劉禹錫, 與歌者何戡 Liu Yuxi, For the Singer He Kan52 二十餘年别帝京

Parted from the emperor’s capital  for more than twenty years, 重聞天樂不勝情 hearing again Heaven’s music  is emotion I cannot bear. 舊人唯有何戡在 Of those I once knew only  He Kan remains 更與殷勤唱渭城 and once again he does his best to sing  “Wei City” for me.

We cannot date this poem with precision, but “Wei City” was a parting song, and it is tempting to place this late in 831, when Liu Yuxi was again sent out to be a prefectural governor—though this time of Suzhou, one of the most desirable prefectures. The pain at hearing the old music performed by one of the old singers also suggests that this might have occurred not long after his return to the capital. The old music conjured up the end of the long Zhenyuan reign, when Liu was a successful young man with a promising career. 劉禹錫, 聽舊宮中樂人穆氏唱歌 Liu Yuxi, Listening to the Former Palace Musician Mu Sing a Song53 曽隨織女渡天河



記得雲間第一歌



休唱貞元供奉曲



當時朝士已無多



You used to accompany the Weaver Woman  fording the River of Stars, and recall the finest song of all  from up there in the clouds. Sing no more those songs for the Emperor  of the Zhenyuan reign— the court officials of those times  are not many anymore.

52  19283; Qu (1989) 786. 53  19272; Qu (1989) 784.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1268

Owen

One of Liu’s most famous poems touches on such a sense of a life somehow missed, a life that could have been fully realized only in Chang’an. Liu Yuxi is only the recurrent visitor. 劉禹錫, 再遊玄都觀 (幷引)

Liu Yuxi, Again Visiting Xuandu Lodge (with preface)54 余貞元二十一年爲屯田員外郎, 時此觀未有花. 是歲出牧連州, 尋貶朗州司馬. 居十年, 召至京師, 人人皆言, 有道士手植仙桃, 滿觀如紅霞, 遂有前篇以志一時 之事. 旋又出牧, 今十有四年, 復爲主客郎中. 重遊玄都觀, 蕩然無復一樹, 唯兔葵 燕麥動搖於春風耳. 因再題二十八字, 以俟後遊, 時大和二年三月.

In the twenty-first year of the Zhenyuan reign (805), I was Vice Director of the State Farms Bureau. At that time there were as yet no flowers at the Lodge. That year I went out to govern Lianzhou and soon was banished to the post of Assistant in Langzhou. After staying there ten years, I was summoned to the capital. Everyone said that a Daoist master had planted immortal peaches that filled the lodge like red clouds. In consequence I wrote a piece to commemorate the moment. Soon I was once again sent out to govern a prefecture; and now, fourteen years later, I am back as Director of the Bureau of Receptions. Again I visited Xuandu Lodge, and it was swept bare, without a single tree left. There was nothing but rabbit mallow and wild wheat waving in the spring breeze. Thus I wrote another poem of twenty-eight syllables in anticipation of some later visit. At the present it is the third month of the second year of the Taihe reign. 百畝中庭半是笞

In the hundred-acre courtyard  half is moss, 桃花淨盡菜花開 peach blossoms have been cleansed away,  wild vegetables are blooming. 種桃道士歸何處 The Daoist master who planted the peaches,  where has he gone?— 前度劉郎今又來 young master Liu of the previous times  today comes once again.

54  19269; Qu (1989) 703.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old Men [ of the Early Ninth Century ]

1269

Lang 郎, “young master,” was not quite appropriate for the vice director of a bureau in his mid-thirties in 805; it has a comic dissonance for the aging poet in the Taihe reign. The ironic poetic game Liu Yuxi is playing is that he has not changed, but the “immortal” world of the magic peaches has come and gone.55 Not insignificantly, the moment of splendor that came and vanished during Liu Yuxi’s long absences from the capital was in the Yuanhe reign of Xianzong, witnessed by Liu only for a few months in 815. It was a glory seen only in passing; and behind the wit and pathos, there is nostalgia. One phrase, usually unnoticed, mitigates the nostalgia: the quatrain is written “in anticipation of some later visit.” The poet anticipates being sent out to the provinces again— as indeed he was—and returning. Young master Liu will check to see if the peach blossoms have returned. We have no record that they did return.

Li Shen and “Recollecting Past Travels”

Li Shen was an exact contemporary of Bai Juyi and Liu Yuxi, dying in 846, the same year as Bai Juyi. He passed the jinshi examination in 806, the first year of the Yuanhe reign, and belonged to that generation of poets. Little remains of his work from the Yuanhe era, though he does make an appearance at a very important moment. At the end of “Ying-ying’s ‘Story” 鶯鶯傳 by Yuan Zhen, the speaker “Zhang” approached Li Shen, who then wrote a companion piece entitled “Yingying’s Song” 鶯鶯歌 to go with the prose story. This was clearly a narrative-ballad version, like Bai Juyi’s “Song of Lasting Bitterness” 長恨歌. Li Shen’s ballad does not survive intact, but pieces were preserved, some in the Jin “Account of the Western Chamber in All Modes” 西廂記諸宮調, by Scholar Dong 董解元. A few of his other poems that have been preserved suggest that Li Shen was once very much a poet of the Yuanhe generation. Li Shen’s modest place in the history of Tang poetry is due to a peculiar and unprecedented collection of poems that did survive, with a preface dated to 838. This collection was entitled Recollecting Past Travels 追昔遊 in three juan. Some of the poems were composed on the occasions they describe, but most were written retrospectively in 838 or slightly earlier, covering his life between 820 and 836.56 In the preface Li Shen gives a succinct account of his life and 55  The poem also plays on the well-known story of Liu Chen and Ruan Zhao, who met two goddesses on Mount Tiantai among the peach blossoms, stayed with them awhile, and then left. When they returned, they could not find the grotto of peach blossoms again. 56  A few of the poems in this collection have prefaces explicitly stating that they were composed on the occasion they describe and that the author has included (“compiled”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1270

Owen

career. The poems, arranged chronologically, are supplied with copious internal notes explaining whatever circumstances might be obscure to a later reader. The preface ends: “There are things that concern me in the phrases, and the inspiration arose from resentment; thus, the words are not all of the same order, sometimes obscure and sometimes obvious; they are nothing more than hope for someone who truly understands (zhiyin) in some future time” 詞有所 懷, 興生於怨, 故或隱或顯, 不常其言, 冀知音於異時而已.57 Yuan Zhen seems to have been the first to actively compile his own collected works, with explanatory materials (in the form of prefaces or letters), at a relatively young age in the 810s.58 Bai Juyi (along with others in his circle) continued and extended the practice, writing his own prefaces for the ever expanding versions. As Bai Juyi’s collection grew in his later years, the poems added were arranged more or less chronologically, though they were sometimes divided between regulated and nonregulated groups.59 In short, Bai’s collection was the prototype of the poetic diary. Li Shen’s Recollecting Past Travels is, by comparison, the prototype of poetic autobiography. We do not know how much Li Shen wrote in the years covered by Recollecting Past Travels, but it seems likely that, in contrast to Bai Juyi, his efforts in poetic composition were sporadic. He could not simply arrange existing poems in chronological order; when he had a past poem representing some significant moment, he included it. Rather, Li Shen is supplementing an apparently sketchy poetic record of his life since 820 with the poems that he “should have” written. Indeed, Recollecting Past Travels bian 編) them in the collection. In the preface to the series entitled “On the Day I Finished My Term at Shouyang” 壽陽罷郡日, he justifies such inclusion by saying that the poems are “no different from recollection” 與追懷不殊 (Wang Xuanbo 37). Many poems are explicitly or implicitly marked as memory; the explicit notation that a poem was composed on the occasion represented suggests that the majority of the remaining poems are also retrospective, though not marked as such. The poems in the last part of the collection, closer in time to the compilation of the collection, were most likely compiled close to the occasions they describe. 57  Wang Xuanbo 157. 58  Since we know that various versions of certain authors’ works (as opposed to individual pieces or small “samplings”) circulated during their lifetime even in the eighth century, the practice was, on some level, older. What generally distinguishes the ninth-century practice is the care and publicity given to the act of self-editing. There are, of course, exceptions, such as Yuan Jie’s 元結 collection, with its 767 preface. 59  Most collections of Tang poetry have been lost, and those that have survived probably do not reflect their original form. Poems may have circulated in rough chronological sequence within generic divisions before Bai Juyi, but from what survives Bai represents the strongest model of chronology as a principle of organization.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Old Men [ of the Early Ninth Century ]

1271

looks very much like an attempt to retrospectively construct a “poetic life,” a chronological oeuvre that looked like Bai Juyi’s oeuvre. The use of poetry to chronologically “document” a life was an important new way of thinking about poetry. We see here the ancestor of a long tradition of scholarship that painstakingly tries to date every poem in a poet’s work and arrange each chronologically within a collection. In the Song dynasty Du Fu became the central figure whose poetic oeuvre was read this way, but that form of reading can be traced back to the group around Bai Juyi. Li Shen’s unique act is part of a larger interest in making poetry collections to publicize, preserve, and document. In many ways the spread of exchange collections marks a shift in attitude from considering such poems as ephemera that might be preserved in a poet’s literary remains to the publicized documentation of a long-term literary relationship. Exchange collections between Bai Juyi and Yuan Zhen and between Bai and Liu Yuxi were already in circulation in 833 when Liu Yuxi received the following letter from Linghu Chu: Mr. Bai, the governor of Three Rivers [Luoyang] compiled the poems sent and answers received between you and him. He wrapped this in a bag of pale blue silk and gave them to me. Mr. Bai wrote some prefatory words at the beginning and called it the Liu and Bai Collection. Long I pondered over how the pieces I have written with you also fill trunks and wrappers—why not put them in order to silence mockery from Three Rivers [i.e., Bai Juyi]?60 Linghu Chu was a very literary senior statesman who had first entered public service because of his writing. Although Linghu Chu is speaking playfully, there is something competitive here, the desire to publicize a poetic relationship which we assume had been carried on for many years for the sheer pleasure it provided. In response to Linghu Chu’s request, Liu Yuxi compiled the Song and Response Collection with Pengyang 彭陽唱和集.61 Later in 835 Linghu Chu would compile his exchange poems with Li Fengji 李逢吉, the canny and somewhat ruthless former minister who had died that year.62 We see the same desire to publicly document a relationship through poetry in the anecdote, quoted earlier, in which Wenzong had hoped to add a poem from the dying Pei Du to his own poetry collection. 60  Qu (1989) 1496. 61  Liu also compiled a later version of this collection, as well as a collection of poems exchanged with Li Deyu. 62  Fu (1998), vol. 3, 114.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1272

Owen

Documentation of a life or relationship in poetry might be undertaken with an eye to current prestige, but it was clearly directed outward, to readers who might not know the particulars that could be assumed between old friends exchanging poems for the pleasure of it. The author’s note explaining the circumstances behind a line is the clearest indication of such a general audience. Earlier we mentioned Linghu Chu’s poem answering Bai Juyi and Liu Yuxi after assuming command of the Xuanwu army in 828. The opening line reads: 蓬菜仙監樂天客曹郎劉爲主客

Penglai’s librarian immortal (Letian), the Reception’s Director [Liu was in the Bureau of Receptions] It seems likely that this was originally preserved through the Song and Response Collection with Pengyang, and although the identifying notes may have been added later, it seems to suggest that the collection circulated with explanatory notes. No group of poems at this time was provided with such rich explanatory notes as Recollecting Past Travels, and few were as explicit as Li Shen in stating that his collection was intended for future readers. The idea, however, was implicit in Bai Juyi’s editorial efforts. We have seen how Bai conceptualized his literary collection as “capital”; and the multiple copies of his collection that he had made and deposited in various temples (as Sima Qian planned to deposit the Shiji at some “famous mountain” to await future readers) suggests just such an eye to future readers. Indeed, Bai offers what was perhaps the most singular and interesting anticipation of future readership in the preface to his Luoyang Collection 洛中集, to be deposited at the Xiangshan Temple: “How do I know that I won’t revisit this temple in some future life and, looking at these writings again, recapture memory of this past life?”63 Li Shen was not a major poet, but he deserves more credit than he is generally given. One reason we might willingly believe that he compiled Recollecting Past Travels in the shadow of Bai Juyi’s model is that Li Shen so often wrote with prior poetry in mind. Older poets were in his head, and he sometimes cites poems and lines of past Tang poets associated with sites he visits. Old poems were the occasion for new poems. Transferred from exile in the far south to a better position as prefect of Chuzhou, Li Shen was visited by an unnamed pipa player; in the musician’s style Li Shen detected the legacy of the court musician Maestro Cao, who 63  Zhu 3806. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1273

The Old Men [ of the Early Ninth Century ]

together with Mu (mentioned above in Liu Yuxi’s poem) were the stars of the court musical establishment at the beginning of the ninth century. Any lover of Tang poetry knows what to expect from Li Shen on this occasion: a ballad in the long line in which Bai Juyi’s “Ballad of the Pipa” meets Du Fu’s “On Seeing a Student of Mistress Gongsun Dance the ‘Sword Dance.’ ”64 Like Du Fu, Li Shen recognizes the teacher in the disciple, provides a long section on hearing Maestro Cao earlier in his life, tells of his own exile, the death of Maestro Cao, and the present performance. The requisite description of pipa playing, however, derives from Bai Juyi. 李紳, 悲善才 Li Shen, Lament for the Maestro65 余守郡日, 有客遊者, 善彈琵琶. 問其所傳, 乃善才所授. 頃在内庭日, 別承恩顧, 賜 宴曲江, 敕善才等二十人備樂. 自余經播遷, 善才已歿. 因追感前事, 爲悲善才.

When I was governing the district [Chuzhou], a visitor came through who was good at the pipa. Asking where he learned the style in which he played, it turned out that he had learned it from the Maestro. During the time I was in the Inner Court, I was shown particular favor; an imperial banquet was given in Twisting River Park, in which the Maestro and some twenty other musicians were appointed to provide the music. During the period of my exile the Maestro passed away. Thereupon I recollected and was moved by those past experiences and wrote a lament for the Maestro. 穆王夜幸蓬池曲



金鑾殿開高秉燭



東頭弟子曹善才



琵琶請進新翻曲



翠蛾列坐層城女



By night King Mu paid a visit  to the bends of Penglai Pool,66 the Hall of Golden Bells was opened,  candles were held high. The eastern ensemble’s  Maestro Cao begged to present a newly composed  tune for the pipa. Lined in seats with kingfisher brows  girls from Tiered Walls,67

64  10818; Qiu 1815. 65  25586; Wang Xuanbo 30. 66  Muzong, figured as King Mu of Zhou. 67  Tiered Walls was the dwelling of the immortals in the Kunlun Mountains. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1274

Owen 笙笛參差齊笑語

pipes and flutes of different lengths,  equal to laughing chatter. 天顏靜聽朱絲彈 His Majesty’s visage calmly listened  to vermilion strings plucked, 眾竽寂然無敢舉 the other musicians were silent  and did not dare to play.

銜花金鳳當承撥



轉婉撒玄促揮抹



花翻鳳嘯天上來



徘徊滿殿飛春雪



抽弦度曲新聲發



金鈴玉佩相磋切



流鶯子母飛上林



仙鶴雌雄唳明月



A golden phoenix, flowers in beak,  was there where the plectrum was plied, with arching wrist he gathers the strings,  to speed the “sweep” and “rub.”68 Flowers floated, the phoenix shrilled  coming from Heaven, and lingering there, filling the hall,  spring snow flew. He pulled the strings composing a melody,  and the newest music emerged, golden bells and jade pendants  clinking against each other. Fluent orioles, mother and chicks,  flew through the Imperial Grove, immortal cranes, hen and mate,  shrill to the bright moon.

此時奉詔侍金鑾

At this time I received a summons  to wait on the golden-belled palanquin, 别殿承恩許召彈 in the detached palace I received grace  and was permitted to call him to play. 三月曲江春草綠 In the third month at the Twisting River  the springtime grasses were green, 九霄天樂下雲端 from nine-tiered wisps heaven’s music  came down from the edge of clouds.

紫髯供奉前屈膝

The purple-whiskered imperial servant  came forward and knelt, 盡彈妙曲當春日 he played all the most wondrous melodies  during the spring day.

68  “Gathering,” long (or rang) 揮, “sweeping,” hui 揮, and “rubbing,” mo 抹, were all technical terms for fingering techniques: long is a left-hand technique; hui and mo are right-hand techniques. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1275

The Old Men [ of the Early Ninth Century ] 寒泉注射隴水開



胡雁翻飛向天沒



日曛麈暗車馬散



爲惜新聲有餘歎



明年冠劍閉橋山



萬里孤臣投海畔



離禽鎩羽強迴飛



白首生從五嶺歸



聞道善才成朽骨



空餘弟子奉音徽



南譙寂寞三春晚



有客彈弦獨淒怨



靜聽深奏楚月光



憶昔初聞曲江宴



心悲不覺淚闌干



更爲調弦反覆彈



Cold fountains gushed out,  Longtou’s waters began,69 Turkish geese winged in flight  and sank away in the sky. The sunlight faded, dust muffled the light,  horses and carriages scattered, and because I cherished that new music,  I heaved continuing sighs. The following year his cap and sword  were enclosed in Bridge Mountain,70 ten thousand leagues away this lone official  lodged by the oceanside. The isolated bird’s ruined plumage  struggled to turn in flight, and white-haired I returned  from over the Five Alps. I heard then that the Maestro  had become rotting bones, and there only remained his disciple  to inherit those dulcet tones. Southern Jiao is still and lonely  late in the third month of spring, there was a visitor plucking the strings,  uniquely sharp and bitter. Calmly I listened as deeply he played  in the light of the moon in Chu, and recalled long ago when first I heard it  at a Twisting River banquet. The heart was touched and unawares  my tears were streaming, then he tuned the strings once more for me  and played it over again.

69  This line both describes the music and an old yuefu about Longtou’s waters. 70  That is, Muzong passed away. The reference is to Han Wudi’s question as to why, if the Yellow Emperor became an immortal and rose to Heaven, he had a tomb at Bridge Mountain. The answer was that only his cap and sword were buried in the tomb. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1276

Owen 秋吹動搖神女佩



月珠敲擊水晶盤



自憐淮海同泥滓



恨魄凝心未能死



惆悵追懷萬事空



雍門琴感徒爲爾



Autumn gusts sets swaying  the goddess’s pendants, bright moon pearls strike upon  a plate of crystal. I pity myself, this Huaihai region  sharing its muck and mud, my angry soul fixes the thought  that I have not been able to die. Depressed in these thoughts of the past,  everything is gone, being stirred by the zither of Yongmen  was all for nothing.71

One could point out the correspondences in detail: as Bai complained about the rusticity of Xunyang in “Ballad of the Pipa,” Li Shen complains about the Huaihai region. One small point of correspondence, however, deserves particular attention. Toward the end of Bai Juyi’s “Ballad of the Pipa,” he asks the woman to play one more time: 莫辭更坐彈一曲

Don’t refuse to sit once more  and play a melody, 爲君翻作琵琶行 and I will compose for you  a “Ballad of the Pipa.” Here, too, we have the motif of the encore: 心悲不覺淚闌干

The heart was touched and unawares  my tears were streaming, 更爲調弦反覆彈 then he tuned the strings once more for me  and played it over again.

This encore of the encore is interesting in that while Bai Juyi specifically asks for it, the musician here seems to offer the encore spontaneously. We do not want to give too much weight to the representational accuracy of the line, but it suggests a significant possibility. It may not be simply Li Shen who is 71  A reference to the Lord of Mengchang’s response to hearing Yongmen Zizhou play the zither; it seemed to the Lord of Mengchang as if Yongmen Zizhou were someone whose kingdom and home city had fallen.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1277

The Old Men [ of the Early Ninth Century ]

poetically reenacting Du Fu’s and Bai Juyi’s poem: the unnamed pipa player may also know the poems, and may be playing his role according to the old poetic “script.” Earlier poetry (and particularly “Ballad of the Pipa”) was becoming part of a general cultural legacy in which later poets participated along with others. Buddhism was clearly an important force in Li Shen’s life, and Recollecting Past Travels tells of vows made in youth and fulfilled. Bai Juyi characteristically enjoyed the intellectual consolation of Buddhism, the company of monks, and the perversely sensuous pleasures of austerities. Li Shen sometimes seems more truly religious. One of his finest poems is based on a famous Suzhou singer who, according to her abiding wish (zhi), was buried in the precincts of a Buddhist temple. It was a topic that invited the confrontation of values and worlds, with the Sanskrit chanting of sutras enduring as the singer’s voice faded into nothingness. 李紳, 眞娘墓 Li Shen, The Tomb of the Pure Miss72 吳之妓人, 歌舞有名者. 死葬於吳武丘寺前, 吳中少年從其志也. 墓多花草, 以滿 其上, 嘉興縣前亦有吳妓人蘇小小墓. 風雨之夕, 或聞其上有歌吹之音.

She was a singing girl of Wu and famous for her singing and dancing. When she died, she was buried in front of the Wuqiu Temple in Wu. The young men of Wu honored her abiding wish to be buried there. Many flowers grow on her tomb, covering the whole top. In Jiaxing county there is also the tomb of the Wu singing girl Su Xiaoxiao. On evenings when there are rainstorms, the notes on song and piping are sometimes heard from upon it. 一株繁艷春城盡



雙樹慈門忍草生



愁態自隨風燭滅



A trunk of thick and voluptuous color  the springtime city is gone, the “paired trees” Gate of Compassion73  suffers the flowers to grow. Her melancholy charm was extinguished  with a candle in the wind,

72  25649; Wang Xuanbo 100. For another poem on this theme, see Zhang Hu, 27282; Yan Shoucheng 16. 73  It was under a pair of śāla trees that the Buddha entered nirvana.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1278

Owen 愛心難逐雨花輕



黛消波月空蟾影



歌息梁塵有梵聲



還似錢塘蘇小小



祇應回首是卿卿



a passionate heart can never be so light  as Heaven’s rain of flowers.74 Moonbeam glances have melted from brows,  there is only the light of the moon, her songs let the dust on the beams rest,  but there are sounds of Sanskrit chanting. She is like Su Xiaoxiao  of Qiantang— but one should only look back  on this little darling.

The poem is built around a series of contrasts: the “Pure Miss” (her virginal name clearly at odds with her profession, but perhaps related to her wish to be buried in the temple grounds) is a flowering tree set against the śāla trees under which Buddha achieved enlightenment, a compassion that allows flowers to keep growing on her tomb. The weight of sensuality of a “passionate heart” melts away in each particular, her flowers replaced by the rain of flowers from Heaven, her “wavelike” glances replaced by the moonbeams from a brow-like crescent moon, her songs fading into the chanting of sutras. The last couplet compares her to another courtesan with a famous tomb, but the latter lacks the enduring commitment to the truth of Buddhism that distinguished the “Pure Miss.”75 The retrospective stance of Li Shen’s collection as a whole invited not only the “meditations on the past,” huaigu, so popular in the ninth century, and reminiscences of his youth; it also disposed him to see himself from the outside, as Bai Juyi so often did, only in a different key. We do not know if the following was one of the poems actually composed on the occasion described, but the title suggests that this is the Li Shen of 838, recalling passing his native region (modern Wuxi) in 833 on his way to his post as surveillance commissioner of Zhedong, at that time recalling his youth spent studying at Hui Mountain Temple.

74  The flowers that fell from Heaven when the Buddha preached. 75  The sense of the last line is very uncertain.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1279

The Old Men [ of the Early Ninth Century ] 李紳, 憶題惠山寺書堂 Li Shen, Recalling My Study Hall in the Hui Mountain Temple76 故山一别光陰改



秋露清風歲月多



松下壯心年少去



池邊衰影老人過



白雲生滅依巖岫



青桂榮枯託薜蘿



惟有此身長是客



又驅旌旆寄煙波



Once I left the mountains of home,  time passed, autumn dew and clear breeze,  the years and months grew many. A bold heart under the pines,  the young man went away, a wasted reflection in the pool,  an old man passing by. White clouds appear and vanish  beside the cliffs and crags, green cassia flourish and wither,  the hanging moss still on them. There is only this body of mine,  ever a traveler, hurrying on with banners raised  into the misty waves.

Lines like the opening couplet are so dreadful that it is hard to read on. Yet when we do, the poem redeems itself in its multiple retrospective visions from both 833 and 838. Zhuangxin 壯心, “bold heart” or “heart of someone in his prime,” is disembodied in memory: it belonged to the young man who did not look at himself but only thought to go off to Chang’an and enter public life long ago—indeed, to become the very high official who traveled in 833 “with banners raised.” The young man went away; an older man returned, remembering the young man and seeing his own aging reflection in the pool. He is, however, only “passing by,” bound to the constant changes in post to which the young man, who stayed here long in study, aspired. We can view the white clouds as constant change, the ultimate emptiness of this life, or perhaps even as the figure of riches and honor, which were to Confucius “as drifting clouds.” The cassia tree, which is usually flourishing in poetry and whose boughs are figuratively plucked by the successful examination candidate, is here clearly withered. Nevertheless, it must remain standing, providing support for the dependent 76  25609; Wang Xuanbo 57.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1280

Owen

hanging moss that still lives by clinging to it. The successful official acquires his “family,” going far beyond the “nuclear family” and even the genetic family proper, and their dependency in turn holds him in place, whether “flourishing” or “withered.” To provide the necessary income, he is always a traveler, in the constantly shifting posts of Tang bureaucratic life. In the end he (whether the “he” of 833 or 838) sees himself leaving the site of youth, the new surveillance commissioner with banners flying, fading off into the river mist. This was his home; but once left, he became “always a traveler”—though the eyes that view the scene at the end remain at home to see himself fade away. None of the younger generation—the poets we think of as belonging to the “Late Tang”—achieved anything like the public distinction of these aging poets of the Yuanhe generation. Their relative success and continuous engagement in the mid-to-upper echelons of political life made them unique among Tang poets. In many ways this was the true end of an era in which poetry could still have a place in public life. The man who in his intemperate youth wrote “Yingying’s Story” and was likely its protagonist, Yuan Zhen, became a minister. The man who wrote the ballad for Yingying sailed off with banners flying as a surveillance commissioner. The man who satirized social abuses in the Yuanhe, Bai Juyi, ended up in a well-paid sinecure, enjoying a salary that ultimately derived from Southland peasants—though he sometimes still felt guilty over the fact that his quilts provided warmth while the common folk were freezing.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription of Emotion in Mid-Tang Collegial Letters Anna M. Shields Ah, Ziqing! In knowing others, we most value knowing one another’s hearts and minds. 嗟乎子卿! 人之相知, 貴相知心。1

In his “Letter in Reply to Director Xue of Daozhou Discussing the Etiquette of Letters,” the mid-Tang writer Liu Yuxi 劉禹錫 (772–842) left us a tantalizing fragment in which he worries about the altered styles of letter-writing in recent decades. Although he appears most interested in the correct terms of address for men of different ranks, he notes that, “after I had been in exile for a decade, I dwelled in obscurity and meanness, hearing nothing of the world, and the only people to whom I wrote letters of inquiry were intimate relations and close friends, those who would not change [their styles]; but contemporary styles [for letters] were changing, and I had no way to learn them.”2 Liu’s concern about the changes in letter conventions expresses not merely the distress of an exile long out of the capital hoping not to embarrass himself in correspondence; it also points to a broader interest in the early ninth century for socially and culturally “correct” writing, an interest indicated also by the proliferation of etiquette manuals (shuyi 書儀) in the second half of the Tang dynasty.3 Liu’s offhand comment that the styles of letters that he exchanged with his “intimate relations and close friends” did not change is equally illuminating, since it suggests quite reasonably that the social interactions of letters Source: “The Inscription of Emotion in Mid-Tang Collegial Letters,” in Antje Richter (ed.), A History of Chinese Letters and Epistolary Culture, Leiden: Brill, 2015: 675–720. 1  From “Letter [att. to Li Ling] in Reply to Su Wu [Ziqing]” (“Da Su Wu shu” 答蘇武書), Wen xuan 41.1848. 2  Liu Yuxi ji jianzheng, 1:267. 3  As Zhao Heping demonstrates, Liu was right to worry: in his analysis of terms of address for officials in three Tang shuyi, Zhao shows that within a century (the span of time covered by three Tang shuyi he examines), those conventions shifted significantly. See Zhao, Dunhuang shuyi yanjiu, 190–91. This 2011 volume is Zhao’s expansion on some of the same research topics found in Zhou and Zhao, Tang Wudai shuyi yanjiu.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_041

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1282

Shields

to those recipients were more important than attention to changes to formal epistolary etiquette. Since his emotional ties with those people did not change, he implies, he had no need to alter his customary style in writing them. Liu Yuxi’s anxiety was focused on the proprieties for addressing those outside his close circle, and therefore, his brief remarks tell us little about the stylistic or rhetorical expectations of letters sent to friends and colleagues in the mid-Tang period, or the influence of the style of those more familiar letters on epistolary practices more broadly considered. The collections of mid-Tang writers contain many more letters, and many more “collegial” letters (which I define below), than the collections of individual writers either before the An Lushan Rebellion (755–63) or from the last few decades of the dynasty, a period for which letters in general are relatively scarce. Some of the largest individual collections from the entire dynasty date from the mid-Tang, including those of Bai Juyi 白居易 (772–846), Han Yu 韓愈 (768–824), Liu Zongyuan 柳宗元 (773–819), Yuan Zhen 元稹 (779–831), and Liu Yuxi, and letters are well-represented in those collections, particularly those of Han Yu and Liu Zongyuan. With respect to their role in culture, letters became an important venue for mid-Tang intellectual and literary debate, and for many ninth-century political exiles (which includes all of the major mid-Tang writers at one moment or another), letters were sometimes the only means by which they could conduct those debates. Moreover, evidence found within mid-Tang letters suggest that shu 書 (letters) of various sorts, along with a greater variety of prose genres, were becoming more prominently featured in the literary portfolios men used to seek patrons and forge reputations.4 Most extant mid-Tang letters are, like most extant letters from before the era of printing, both formal and political, concerned with an individual’s career and official duties. But we also find in mid-Tang corpora more letters sent to colleagues and friends alongside the many letters sent to patrons and political superiors. The prominent presence of letters in certain mid-Tang collections and the discussions of letter exchanges found within the texts themselves reveal that mid-Tang literati circulated their letters, even ones addressed to individual colleagues or friends, within their circles and beyond them, deploying them as proof of their literary talents and intellectual interests. Letters to colleagues and friends were also critical social tools for negotiating delicate

4  See, e.g., the letter by Li Guan 李觀 sent to then-Chief Examiner Lu Zhi 陸贄, in which he lists ten pieces he is submitting for Lu’s perusal; of the ten, three are letters on different topics. Quan Tang wen 533.5415; see also Fu Xuancong, Tang dal keju, 264–65; Moore, Rituals of Recruitment, 142–43; Ditter, “Genre and the Transformation of Writing,” 138–39.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1283

relationships over distance and time.5 The presence in mid-Tang corpora of more familiar letters—such as those written to colleagues and friends— suggests that mid-Tang literati had a greater interest in preserving texts that documented private life and sentiments, an impression confirmed by the greater presence of episodes from private life in other mid-Tang literary genres, such as poetry, funerary texts, and parting prefaces.6 At the same time, the etiquette manuals that included models for letters to friends (pengyou shuyi 朋友書儀) or even the letter models in more comprehensive manuals for “auspicious and inauspicious [occasions]” (jixiong shuyi 吉凶書儀), though rich in both literary and social detail, bear almost no resemblance to the extant letters to friends and colleagues in mid-Tang corpora.7 This study examines the linguistic and rhetorical features used to inscribe emotion in mid-Tang “collegial” letters, focusing on anger and affection in particular. I define “collegial letters” as letters sent between men of similar rank and status who were already known to one another. The goal is not only to learn more about the function of emotion in the genre of letters specifically, but also to contribute to our understanding of emotional expression in literary writing in the mid-Tang period more broadly. Given the turn toward private life that we see in poetry and other texts from the era, what degree of personal detail, including personal feeling, was common to collegial letters? How might we reconstruct the conventions for emotional expression in such letters? What larger objectives of collegial letters might have influenced the expression of 5  For a discussion of the conventions and circulastion of Tang letters and the subgenre of “cover letters” in particular, see Ditter, “Genre and the Transformation of Writing,” 101–7 and Ditter’s essay in this volume. 6  On the greater presence of private life in mid-Tang writing, see Owen, The End of the Chinese ‘Middle Ages’ and McMullen, State and Scholars in T’ang China. 7  Patricia Ebrey notes this in her important early article, “T’ang Guides,” 604, 608–9. Wu Liyu in Tang li zhiyi has examined a number of pengyou shuyi from Dunhuang, dating to the early medieval period and to the Tang, and transcribes and analyzes a number of the monthly correspondence models, which rely heavily on parallel prose and formulae for expressing feelings toward the friend, as I will discuss below. She also notes, judging both from extant examples from Dunhuang as well as the bibliographical evidence for Tang shuyi, that the earlier forms of pengyou shuyi do not seem to have been especially popular in the 9th c., despite ninth-century authors’ interest in other issues of social ritual and etiquette, as evidenced by the many titles for shuyi texts dating to the first half of the 9th c. (pp. 4–29). Wu also points out that letters by ninth-century writers such as Li Shangyin and Linghu Chu, advocates of parallel prose style in other forms as well, tend to show greater resemblance to shuyi models, and that the so-called guwen writers of the mid-Tang seem to have had little influence on letter styles in the succeeding generations. The question of the longterm influence of mid-Tang collegial letters (e.g., on letter-writers in the Northern Song) remains unexplored.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1284

Shields

emotion? I propose that the pervasive concerns expressed by mid-Tang literati about sincerity, cheng 誠, authenticity, zhen 真, and trust, xin 信, and their concomitant suspicion of hypocrisy and falseness, led many writers to strive for an affective style in collegial letters that avoided the emotional excess and parallel prose form of conventional social letters. The extant letters also reveal that mid-Tang writers did not avoid affective display altogether, as they might in letters sent to superiors, but instead experimented with a variety of syntactic and rhetorical techniques, using emotional expression to support intellectual content, and weaving their emotional dispositions into the fabric of the text as a whole. The styles we find in these letters can be read collectively as a powerful counter-aesthetic that set itself not only against the parallelistic style that was expected in many mid-Tang prose forms, including letters, but also against the relative lack of affect in the more formal, political letters of Tang public epistolary culture. These mid-Tang collegial letters draw upon emotional responses as a way to frame their self-representations and intellectual debates, and in so doing, they intensify the urgency of the concerns that writers hoped to convey. Perhaps more than any other genre, medieval letters functioned as sociotexts, used both to communicate information and to negotiate and sustain social relations. In fulfilling those roles, letters required an appropriate expression of feeling from the writer toward the recipient, whether of humility, concern, affection, shame, or any other emotion, and the feelings expressed had to match the needs of specific social situations.8 My focus on emotions in letters draws on recent research in the humanities on the critical relationship between emotion and intellect,9 as well as the ancient Chinese link between qing 情, the emotions, and literary creativity, whether defined as poetry, shi 詩, or more broadly as wen 文.10 In the case of emotions in Tang letters, I focus less 8  For a useful discussion of early modern letters’ function as sociotexts, see Schneider, The Culture of Epistolarity, 15–18. 9  There are many different versions in the humanities and social sciences of “the affective turn.” Contemporary approaches to affect and the emotions range from the social constructionist (inspired by Deleuze and Guattari, e.g.) to those that draw on research in cognitive psychology and neuroscience (such as Reddy, The Navigation of Feeling or Thrailkill, Affecting Fictions), with many other positions in between. Scholars of Chinese literature and culture have in recent decades begun to historicize the study of qing for different eras and texts; see, e.g., the essays collected in Eifring, Love and Emotions and Santangelo, Love, Hatred and Other Passions. 10  The traditional theoretical link between qing and wen (that literature was born out of one’s responses to circumstances) appears very early in the tradition, perhaps most famously codified in the “Great Preface” to the Classic of Poetry, and was a commonplace of medieval literary culture; see, e.g., mid-Tang writer Liu Mian’s 柳冕 comment in his

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1285

on the theory of emotional states—what were called either the “Six Emotions” or “Seven Emotions” in medieval texts11—and instead consider the controlled display of affect in crafted literary utterances. The degree to which literati believed that literary writing should express feeling—especially personal feeling—varied widely in Chinese history not just from period to period but also according to social occasion, as well as according to the gender and class of writers and recipients.12 As I discuss below, medieval etiquette guides for letters were filled with forms for both positive and negative emotional expression (affection and grief, for example). Their models suggest that statements of feeling were expected in letters written for occasions such as betrothal, mourning, and separation, and furthermore, that the specific terms for degrees of feeling had to be chosen carefully for one’s audience and one’s relationship (superior/ inferior, familiar/distant) to the addressee. However, in contrast to the need for affect in letters for social occasions, the great body of letters concerning official matters extant from the Tang and earlier periods suggests that emotional display—like personal information more generally—was either absent or greatly attenuated in letters. Letters composed for formal political contexts emphasized persuasion and evidence as well as appropriate politeness and deference; they contain few references to personal feeling unless they were immediately relevant to the topic under “Letter to Lu of Huazhou Discussing wen” (Yu Huzhou Lu dafu lun wen shu 與滑州盧 大夫論文書): “Now wen is born from qing, and [the various forms of] qing are born from grief and joy. Grief and joy are born out of order and disorder. Therefore the junzi is moved by grief and joy and thereby composes writing in order to understand the root of order and disorder” 夫文生於情, 情生於哀樂, 哀樂生於治亂。 故君子感哀樂而為 文章以知治亂之本. Quan Tang wen 527.12a. 11  A search of the Quan Tang wen yields almost even numbers of uses of 六情 and 七情, although the “seven emotions” term appears more frequently in texts from the midand late Tang. The “Liyun” chapter of the Li ji is often noted as the source of the “seven” list; it includes happiness, anger, grief, fear, affection, hate, and desire (喜, 怒, 哀, 懼, 愛, 惡, 欲); Xunzi gives both a list of six (without affection, ai 愛) and a list of seven (with le 樂 in place of the Li ji’s “fear”). For a good introduction to the early sequences and expansion of the basic emotions, see Middendorf, “Basic Emotion Terms.” Many modern scholars of emotions consider love or affection to be a secondary or social emotion rather than a primary or basic emotion; however, Middendorf’s research suggests that the emotion sequences and lists in early China tended to develop conceptually as positive/negative binaries, and ai appears paired with wu early in Warring States texts (yu, “desire,” was latest and without a pair; p. 138). For a study of the broader range of the term qing in early China, see Harbsmeier, “The Semantics of Qíng.” 12  David R. Knechtges discusses this in his study of anthology compilation in the early medieval period, “Culling the Weeds,” 207–11.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1286

Shields

discussion. Mid-Tang collegial letters, however, fall between these two major types of letters both socially and linguistically. They were of non-specific occasion (could be written for any purpose) and therefore opened up a greater range of style registers and affects; they were composed within a relationship where excessive attention to politeness and deference was unnecessary, even unwanted, and direct speaking more possible; and they often concerned issues whose meaning was under discussion, allowing writers to offer more questions, doubt, and reflection in lieu of reporting, flattery, or persuasion. The more open social space of collegial letters also gave writers greater room for literary experiment since, from the perspective of medieval genre conventions, letters were among the least regulated of any prose form widely practiced in the Tang. The inscription of emotion in mid-Tang collegial letters also raises the question of representing the public and private spheres in Tang literary writing, and where letters were conceptualized with respect to those realms. David Pattinson has argued that one reason we have few “private letters” extant from the early medieval period was that the wide circulation of letters discouraged writers from entrusting truly private matters to such texts; furthermore, if trivial or personal matters were entrusted to letters, that might also have made them less likely to be preserved in collections. Regarding letters to friends, he notes, “well-written letters expressing friendship were preserved, but the tone was often morally uplifting rather than intimate, and they were anyway not numerous.”13 The question of the ways that emotional expression intersected with the issue of public and private spheres in medieval China is important but as yet little-studied. It is certainly the case that most Tang letters on political and official matters tend not to include expressions of sentiment other than the merely conventional, usually at the openings and closings of letters. However, the many extant medieval letters expressing sorrow, longing, and grief—which would have been shared with contemporaries even before being placed into collected works—demonstrate clearly that emotional expression was not limited to private or intimate exchanges, nor was it a marker of informality. We may simply not possess enough evidence of familiar letters from medieval corpora to conclude more about the question. But this returns us to the significance of mid-Tang letters in the history of epistolary practices in China. If, as we know, a bibliographical distinction between formal and informal letters does not emerge until the Northern Song, with the preservation

13  Pattinson, “Privacy and Letter-Writing,” 97–98.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1287

of brief letters known as chidu 尺牘 or shujian 書簡,14 the range of emotional expression found in mid-Tang letters strongly suggests that the mid-Tang was a transitional moment in the history not merely for the forms and expectations of letters but also for their preservation. With respect to the collections of Bai Juyi and Yuan Zhen, we know that both authors were meticulous about the selection (and deletion) of items in their corpus, and therefore that the letters that have survived represent deliberate choices.15 In the cases of Liu Zongyuan and Han Yu, on the other hand, we know less about the authors’ own preservation of materials but can assume that in the process of collecting their works undertaken by Liu Yuxi for Liu Zongyuan and by Li Han 李漢, Han’s son-in-law, and others for Han Yu,16 the intellectual content and literary merit of certain letters addressed to friends and colleagues were seen as worthy of preservation and transmission despite—perhaps even because of—the glimpses of the author’s personal feeling they revealed.17 At the very least, this suggests a changing ninth-century view of the boundaries of an author’s identity that a literary collection was intended to transmit. Beyond that, the extraordinary diversity of letters we find within the corpora of a few authors such as Han Yu and Liu Zongyuan— letters to patrons, monks, higher-ranking officials, as well as to relatives, friends, and students—demonstrates the power of letters to capture the complexity of unique personalities and talents. 1

Emotion in Medieval Models from the Wen xuan and Tang shuyi 書儀

When considering the style and moods of mid-Tang letters, it is important to consider the possible influence of medieval models for letter-writing on 14  For this shift, see Zhao Shugong, Zhongguo chidu wenxue shi, 236–85; Egan, “Su Shih’s ‘Notes’,” 561–88; and Egan’s article in this volume. 15  For the work of Bai and Yuan on their own corpora and the practices of compiling individual collections more generally in the Tang, see Nugent, Manifest in Words, 236–84. 16  See Wan Man, Tang ji xu lu, 167–83 and 188–200, for brief discussions of the compilation, transmission, and editions of their collections. 17  The exclusion of some letters that were later collected in Han’s works that survived outside the main collection (waiji 外集) sheds light on this issue as well—e.g., the letter that prompted Liu Zongyuan’s outrage (see below) against Han’s positions, “Letter to Floriate Talent Liu [Ke],” has been much criticized by later readers for its weak argument and position on historiography. See the comments collected by Luo Liantian in Han Yu guwen jiaozhu, 4:3438–40.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1288

Shields

mid-Tang writers. Although it was but one anthology among many, the highly influential model of the Wen xuan gave mid-Tang writers examples of literary letters exchanged among medieval elites.18 But the letters collected in shu category in the Wen xuan are largely formal and political in nature; they provide only a few examples of emotion directed at the recipients of letters, and they are generally quite constrained, consistent with the “moderate” vision of belles lettres that this anthology transmitted.19 In strong contrast, models of letters for social occasions found in jixiong shuyi, etiquette manuals for letters for specific “auspicious and inauspicious” occasions, and pengyou shuyi, manuals for composing letters during the year and at specific seasons, are replete with emotion, and in fact seem characterized by an excess of sentiment. The Wen xuan models may have been less influential to the routine exchange of collegial letters in Tang culture, but they certainly were widely known among Tang literati due to their standing in the medieval curriculum. Their model of literary style in letters, including their preference for a formal and reserved register, surely shaped Tang writers’ understanding of the degree of emotion appropriate to letters they considered to be “literary writing,” wenzhang 文章. Conversely, the high degree of sentiment in models for letters exchanged between friends, as captured both in seasonal pengyou shuyi and the social occasions on which friends would exchange letters, might have acted as a negative influence on emotional expression in collegial letters. Many examples of medieval letters will need to be examined in order to understand the variables affecting emotional expression in letters; my discussion should therefore be seen as a preliminary step in this research. The twenty-four Wen xuan letters preserved as shu (in contrast to other letter forms such as qi 啓) include many that can be considered “collegial” or exchanged among colleagues and friends, but we find few examples of emotional expression directed at the recipients of the letters. Two important exceptions are found in the Han general Li Ling’s 李陵 letter to Su Wu 蘇武, cited in the epigram, and a Jin-era letter from the official Zhao Zhi 趙至 to his friend and colleague Xi Fan 嵇蕃,20 both of which use parallel prose, numerous allusions, 18  See also David Knechtges’s article on the Wen xuan letters in this volume. 19  See Knechtges, “Culling the Weeds,” 207–12. In a future expansion of this research, I plan to consider the influence on Tang letter-writing of medieval letters outside the Wen xuan, particularly the influence of letters transmitted in widely-read texts such as the standard histories of the pre-Tang dynasties. 20  Zhao Zhi’s letter is also cited in full in his Jin shu biography (92). He is said to have died from frustration and sorrow at age 37. Both letters are introduced in David Knechtges’s article in this volume.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1289

and phrases for feeling in a manner similar to conventions found in medieval etiquette manuals. The Wen xuan letter from Li Ling was suspected by Tang and later readers, including the Tang historian Liu Zhiji 劉知幾 (621–771), to be a later forgery, an opinion debated by many scholars since that time; some have suggested that its use of parallel prose and its high degree of emotional expression filiate it to a later period style.21 In both of these letters, the writers explain their particular extreme situations (for Li Ling, a life exiled among the Xiongnu; for Zhao Zhi, service in Liaodong), and both writers lament the separation from their friends in highly emotional and dramatic terms. For example, in the Li Ling letter, the writer exclaims in his closing: Ah, Ziqing! What else is there left to say? We are ten thousand li apart, cut off from each other, on different roads. In this life, I have become someone parted from our generation; in death, I will become a ghost in foreign lands. I bid farewell to you in life and death forever. 嗟乎子卿, 夫復何言? 相去萬里, 人絕路殊。生為別世之人, 死為異域之鬼, 長與 足下生死辭矣。22

Zhao Zhi closes his letter in a similar fashion: I am gone from you, Master Xi, parted far, and forever cut off! Alone and solitary, I am whirled away, to the edge of the desert sands! Stretching into the distance three thousand li, the road is hard to cross! As for a 21  An early discussion of this in English was that of Whitaker, in “Some Notes on the Authorship.” Chinese scholars continue to debate the question; a recent article rehashes some of the most prominent positions (the majority of which tend to argue for a post-Han date of composition), but goes on to argue that it is a Western Han piece. Ding Hongwu, “Li Ling ‘Da Su Wu shu’.” Stephen Owen notes that the Li Ling-Su Wu story, as one of parting and separation of friends due to political conduct, “was a topic that invited persona composition, as can be seen in the song in the [Han shu] biography and in the famous letters, of uncertain date, supposedly exchanged between the two men.” The Making of Early Chinese Classical Poetry, 244. 22  “Da Su Wu shu” 答蘇武書, Wen xuan 42.1853. Throughout this essay, I cite the Chinese text with the punctuation given in modern editions, since that punctuation reflects the opinions of modern Chinese commentators on the vehemence of an utterance, in the case of exclamation points, or the interrogative intent, as in the case of question marks. For authors without modern punctuated editions, I use only periods and commas to mark sentence structure in the Chinese, but I introduce exclamation marks and question marks in the English where they seem appropriate.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1290

Shields

meeting when we could clasp hands again, it is remote, no day that I can foresee! My feelings of longing tie me ever closer to you—who could say they would ever dissolve? 去矣嵇生，遠離隔矣。煢煢飄寄，臨沙漠矣。悠悠三千，路難涉矣。攜手之 期, 邈無日矣。思心彌結，誰云釋矣!23

In these two letters, the intensity of the writers’ emotion seems to come from the finality and peril of the separation between two friends, a situation in which elaborate displays of sorrow and distress would be expected. The letters thereby fulfill the social function of performing grief at parting, rhetorically identical to the function of parting poems. More commonly, however, expressions of emotion in Wen xuan letters come from the writer in regard to a situation that does not directly affect the recipient of the letter—as in Sima Qian’s 司馬遷 letter to Ren An 任安, which is of course filled with anguish, regret, and bitterness. Cao Pi’s 曹丕 and Cao Zhi’s 曹植 letters to Wu Zhi 吳質, on the other hand, articulate a milder feeling of melancholic nostalgia that is largely directed outward at a group of men and a moment in time, rather than being a personal response directed toward Wu Zhi himself. As an example of a letter exchanged within a dyadic relationship, the letter from Xi Kang 嵇康 to Shan Tao 山濤 severing their friendship displays little anger or emotion of any sort.24 Although Xi Kang does reveal indignation and resentment of the Sima regime, as Thomas Jansen has shown in his research on medieval “severing friendship” letters, by cutting off his relationship with Shan Tao, Xi Kang presents a highly political argument about service to a corrupt court and the need to choose his own path of personal integrity. Compared to the letters of parting, the letter severing friendship is remarkable for its reserve about the personal dimension of the mens’ relationship; Jansen suggests that this stems directly from the public nature of the social

23  “Yu Xi Maoqi shu” 與嵇茂齊書, Wen xuan 43.1942. 24  The twenty-four shu collected in the Wen xuan are found in j. 41–43; see Knechtges’s essay on these letters in this volume. The letter from Li Ling to Su Wu leads off the collection (Wen xuan 41.1847–53); Sima Qian’s very long letter follows (“Bao Ren Shaoqing shu” 報任 少卿書, 41.1854–66); Cao Pi’s two letters to Wu Zhi are very brief (“Yu Zhaoge ling Wu Zhi shu” 與朝歌令吳質書, “Yu Wu Zhi shu” 與吳質書, 42.1894–98); Cao Zhi’s letters to Wu Zhi is also short (“Yu Wu Jizhong shu” 與吳季重書, 42.1905–7). Xi Kang’s letter to Shan Tao is among the longer letters in the shu collection (“Yu Shan Juyuan juejiao shu” 與山 巨源絕交書, 43.1923–30).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1291

interaction.25 Furthermore, one could argue that writers might be averse to showing excessive sentiment in such letters. In letters composed in such serious political and social contexts, we would expect literati to defend their choices with arguments more substantial than mere personal feeling, explaining their reasoning gravely and with sufficient evidence. In contrast to the relatively unemotional quality of anthologized literary letters, medieval etiquette models found in the Dunhuang manuscripts, particularly the forms preserved in pengyou shuyi, overflow with repeated expressions of affection, longing, and wishes to be reunited. (There are no examples for the expression of anger, however, in the seasonal friendship letters.) Although the Dunhuang fragments of etiquette texts date from different eras—the earliest is dated to the Western Jin—scholars have pointed out that the mid-Tang appears to have been a significant transitional moment in the interest in and production of such texts.26 The Dunhuang copy of a shuyi text composed by the mid-Tang literatus Zheng Yuqing 鄭餘慶, The Newly Established Auspicious and Inauspicious Letters and Etiquette for the Great Tang, Da Tang xinding jixiong shuyi 大唐新定吉凶書儀 (S 6537), shows the wide range of social occasions and interactions that Zheng’s manual was meant to address, with six broad categories of events for which the work provided letter models. Notably, Han Yu, one of the most innovative mid-Tang authors of letters, is listed as one of the compilers of this text.27 Zhao Heping argues in fact that the flowery expression of sentiment through specific formulae is a constitutive element of the models.28 Parallel prose in the four-six form is a consistent formal feature of all of the models. Examples from these texts are filled with repeated parallel phrases about separation and longing, such as the following statements from different letter models, taken from the Texts Exchanged through the Twelve Months (Shi’er yue xiangbian wen 十二月相辯文):

25  Jansen, “The Art of Severing Relationships,” 356–61. Of the letter by Han dynasty literatus Zhu Mu on severing friendship, Jansen also argues, “in the Han discourse on human relations the termination of a friendship is not a matter of private concern only, but a political statement that resonates with the overall political situation by which it is influenced and which it itself tries to affect” (p. 355). 26  Wu Liyu, Tang li zhiyi, 16–17; Jin Chuandao, “ ‘Shuyi’ neirong bianzheng,” 127–30. See also Huang Zhengjian, Zhong Wan Tang shehui, 209–43. 27  Jin, “ ‘Shuyi’ neirong bianzheng,” 126. The compilers are also listed in Zheng’s Jiu Tang shu biography, 158.4165. The work was completed in 811 or 812, but Zhao notes that the Dunhuang copy was made, according to the text’s internal date, in 827. 28  Zhao, “Dunhuang xieben ‘pengyou shuyi’,” in Zhao and Zhou, Tang Wudai shuyi yanjiu, 110–12.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1292

Shields

From afar I think of my friend—when could I ever forget? 遠念朋友, 何時可忘?

My sorrow flies up as qi, increasing the clouds; my tears fall as pearls, adding to the dew. 愁飛氣而添雲, 淚垂珠而益露。

I think of you across a thousand li, despising that letters from friends are cut off. 千里相思, 恨朋書之隔絕。

In the desolate court, I sigh alone, holding back tears as I think on my companion; in the thatched hut, I lament in solitude, tears running down as I remember my friend. 荒庭獨嘆, 收淚思朋, 草室孤嗟, 行啼憶友。29

In these models, emotional expression is foregrounded through direct statements of the sentiments “I am thinking of you” and “I am sad without you,” and with images of separation sorrow being realized materially or physically through sighing, tears, and the “sending” of thoughts across long distances. Although here I have only selected phrases from the texts, the models consistently employ strings of such phrases, a pattern that only heightens the emotional tenor of the letters. The question remains, however, as to the influence of such models on elite letter-writing in the early medieval period and the Tang. Were these in fact generally observed conventions for the expression of sentiments toward friends, or do they represent prescriptive texts that, while significant as models for conducting friendship in texts, had a limited impact on epistolary practice among elites? The question may in fact be one of intended audience and the writer’s ambitions for the text: would such conventions have been adapted by elite men and women for their own social purposes, but largely avoided by literati attempting to compose letters with greater literary, intellectual, or philosophical content? The Wen xuan texts suggest that emotional statements in letters should be restricted to close relationships and extreme situations; but in those contexts, feeling should then be expressed clearly and in multiple ways. The 29  Ibid., 118–29. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1293

shuyi models concur with this position in some respects, in that they suggest that statements of affection and longing had to appear in letters between friends composed at any time. In mid-Tang collegial letters, in contrast, we find writers being both more creative and indirect with their expressions of affection and displeasure. They also avoid parallel prose almost entirely, composing in a looser, sometimes more narrative style; and yet they very clearly convey the overall mood of the author and his disposition toward the recipient. The question then becomes, what epistolary techniques did mid-Tang writers extend or create to do so, and how did those feelings support the other objectives of the letters? 2

Methodology and Questions of Evidence for Mid-Tang Letters

My analysis is based on three stages of reviewing mid-Tang texts: compiling a corpus of collegial letters, identifying within that corpus letters that displayed anger and affection toward the recipients, and then analyzing the two sets of texts for lexical, syntactic, and rhetorical patterns for the respective emotional expressions.30 Out of the texts titled shu 書 in 270 juan of Quan Tang wen (j. 440–710, or in terms of approximate dates, roughly the 780s to the 830s), I identified forty-four texts by nine writers as “collegial letters” according to the following criteria: the letter’s mode of salutation and closing indicated that it was addressed to a peer (someone of similar official rank or age, or perhaps some combination);31 and the addressees were already known to the writer, a fact either revealed in the letter (in statements such as “when I saw you last a few years ago…” or “we’ve been separated for a long time…”) or established from external evidence. These criteria therefore eliminated letters to patrons or to men seeking to meet and study with the writer, as well as formal letters reporting on political affairs or making official requests. Only two letters that I include deviate from these criteria slightly. One was Han Yu’s letter to a young man he had not yet met, but because he had been “introduced” to the young 30  For the importance of focusing on the lexicon for emotions in this kind of cross-cultural analysis, see the discussions in linguist Anna Wierzbicka’s Emotions Across Languages and Cultures. 31  Ebrey, “T’ang Guides,” notes the plainness of salutations and closings for peer-to-peer letters in one extant Tang shuyi (604); the models given in Zhao Heping are equally simple and brief (such as “surname, personal name reports” 姓名白), in comparison to the longer and more honorific or humilific conventions for superiors. I also eliminated letters without either a salutation or a closing from the analysis, even in the case (as in Huangfu Shi’s letters to Li Ao) where we know from external evidence that the writer was a peer of the recipient. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1294

Shields

man and his literary work via a close friend, Li Guan (who was deceased at the time of the letter), Han takes pains to assure the man that he feels as if he already knows him, and addresses him in the same manner as he addresses colleagues and friends already known to him. The second was Li Ao’s letter to his younger male cousin, which, though in the tradition of “family instruction” letters, seems to violate the expectations of that subgenre by speaking to the young man as if he were a member of Li Ao’s literary circle.32 Within this corpus, I identified five letters that expressed some degree of anger and fourteen expressing affection toward the recipient of the letter, and examined these two sets of “anger” and “affection” letters for consistent lexical, syntactic, and rhetorical patterns. As I will show, using indirectness to convey emotion was common to both sets; however, the rhetorical, syntactical, and lexical techniques used to suggest anger or affection rather than state it explicitly were surprisingly distinct. In working with texts from before the era of printing, we must keep in mind potential problems of textual integrity, both in the collections and in individual texts. As noted above, the largest mid-Tang corpora tend to preserve more letters, and thus the probability of finding collegial letters in those corpora increases. As a consequence of the relative size of collections and other biographical and textual transmission issues, certain writers are better represented than others in the group of forty-four collegial letters. Han Yu’s corpus, with fifty-four texts titled shu, yielded nine collegial letters; Liu Zongyuan’s, with thirty-five shu, yielded eleven; Bai Juyi’s corpus, with over 3400 poetic and prose texts, contains only nine shu, and yielded three collegial letters; Yuan Zhen’s corpus, which suffered significant losses after the Song, has only four letters, and yielded none.33 The question of the textual integrity of mid-Tang letters is especially troublesome in part due to the diverse forms for letters in the medieval period. To what degree were letters—perhaps letters containing personal information in particular—truncated, edited, or otherwise altered before being placed in literary collections? Unlike the case of poetry, we have little evidence from the Dunhuang material (or elsewhere) that would help us understand the transmission practices for letters; however, some internal features of these letters at least allow us to speculate. 32  See the essay by Antje Richter in this volume. 33  There are some obvious reasons for these different ratios: Liu Zongyuan’s long exile, e.g., meant that his letter-writing was largely addressed to people he already knew or in response to younger literati seeking to study with him. Han Yu’s body of letters includes several letters sent to members of his close circle, making them easily recoverable for his collected works, if Han himself did not have a copy. For losses to Yuan’s corpus, see my article “Defining Experience,” 62–66. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1295

Two features of this group of collegial letters would suggest that they are largely intact internally, that is, not abbreviated or excerpted. First, of the fortyfour I examined, all but five contain salutations and closings appropriate for letters between peers, according to models found in etiquette manuals and used consistently in the letters themselves. Second, the letters’ overall tendency toward hypotaxis and tightly structured arguments would be likely to reveal a “missing” component, unless that were a brief paragraph of incidental personal detail that might be placed at the end of a letter and excised without disrupting the flow or coherence of the text. The collegial letters are among the most personal of extant mid-Tang letters; however, as a rule, they contain very little information about daily life of the kind found in Song and later informal letters (chidu).34 It seems at least possible that brief sentences on mundane affairs might have been omitted in the final transcription of letters, but it seems unlikely that editorial excisions would have included the writers’ expressions of feeling. More critical is the fact that the corpus of forty-four yielded texts by such a small number of writers, many of whom were known to one another and shared intellectual positions, and that these texts fell within a time frame of roughly twenty years, despite the larger initial scope of the search. In other words, this corpus is coherent in ways that may shed light only on the epistolary practice of several writers rather than a sense of period style or preferences. 3

“In my heart, I am deeply displeased” 私心甚不喜: Expressing Anger and Disagreement

The social norms for expressing anger or other negative emotions in Tang culture are difficult to reconstruct, given the nature of the sources we have. Although third-person historical accounts and anecdotes record many incidents of rage and violence, particularly in martial settings, we rarely find first-person utterances of anger or even irritation in literati writing, and fewer instances still of such feelings communicated directly toward the reader or intended recipient. Xunzi’s admonition to keep anger out of noble conduct— the junzi 君子 “does not become angered yet maintains his authority” 不怒而威35—was entirely consonant with other social conventions of medieval culture that were shaped by the Confucian discourse of moderation,

34  On the prominence of mundane detail in the chidu of Su Shi, see Egan, “Su Shih’s ‘Notes’ ” and his essay in this volume. 35   Xunzi, “Wang xiao,” 8.11. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1296

Shields

self-control, humility, and deference, as well as by Daoist and Buddhist teachings that advocated quietude, stillness, and imperturbability. Anger in social intercourse signaled a loss of control and face on the part of the one angered. Although righteous anger, or moral outrage, had considerable rhetorical power and could be a sharp tool in political discourse, that degree of emotion was unlikely to appear in the kinds of texts most common to Tang social interaction—exchange poetry, letters, parting prefaces, and funerary compositions, among others. When we see traces of anger in such texts, as in these five mid-Tang letters, the source is frequently the writer’s disagreement with the recipient on a matter of consequence. In these letters, the terms, syntax, and rhetorical devices serve to express anger along a spectrum of mere vexation to moral outrage, but they never spill into violence or threat. However, given that each writer felt justified in revealing more emotion in each of these contexts, we need to consider the risks and rewards they weighed in their letters, and what they might have seen as the ultimate goal of the epistolary exchange. In contrast to the performance of separation sorrow or mourning, two emotional displays of great social importance for which we have many ritual and literary texts, we have no surviving cultural script for the expression of anger in medieval China. As one historian of early modern England has written of English letters, “anger was a forceful invitation to renegotiate unsatisfactory aspects of relationships. It spotlighted deficiencies in duty, unacceptable conduct, disrespect, broken promises, and frustrated expectations.”36 To express anger or annoyance in a text was also to seize the right to criticize and rebuke others. Expressing anger, therefore, could be a claim of moral superiority, one supported by indignation. But in the context of the collegial relations of these letters, expressing anger may also have been a sign of familiarity and intimacy; the risk of losing face by giving way to anger could be mitigated in a relationship in which there was some trust or at least an expectation that whatever provoked the anger could be resolved. In these letters, the driving issue is conduct: the letter-writer criticizes the conduct of the recipient, supporting his arguments with reason, historical precedent, appeals to common sense, and personal experience. At the same time, each writer uses every technique possible to avoid directly stating his anger toward the recipient. The five letters give us a truly narrow slice of mid-Tang literati culture—they were exchanged within a very small group of people, and Han Yu is a party to four of the five exchanges—and yet we should keep in mind that these were likely read well beyond the circle of close acquaintance, and thus were 36  Pollock, “Anger and the Negotiation of Relationships,” 568.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1297

providing models for other writers to consider. There is also a clear pattern here of what we might call “annoyance at having to repeat oneself,” in that three of the letters (by Li and Han) are apparently the second in a series of rebuttals or responses to the recipient. Such letters must have been quite common; but these are among the few extant examples of “second replies,” di er shu 第二書 or chong da 重答, from ninth-century corpora.37 Author

Recipient

Date (est.)

Title

Content summary

Liu Zongyuan

Han Yu

815

Liu utterly rejects Han’s position on the role of the historian, mocks his statements and position, criticizes his conduct

Li Ao

Hou Gao

Bef. 805

李翱

侯高

“Yu Han Yu lun shi guan shu” 與韓愈論史官書a (Letter to Han Yu discussing the [duties of] the Historian) “Da Hou Gao di er shu” 答侯高第二書b (Second Letter in Reply to Hou Gao)

Second letter to Hou (first is not extant) rejecting his advice to “go along with the world”; Li defends “my Way” (吾道), gives historical, philosophical, and personal reasons why Hou Gao is wrong

37  One excellent example of a letter that was clearly one in a series of exchanges can be found in Liu Zongyuan’s corpus, in his “Letter [in Reply] to Du Wenfu’s Multiple Letters” (Fu Du Wenfu shu 復杜溫夫書, which also contains strong expressions of annoyance and anger at Du. I excluded this piece from my analysis since the relationship between Du and Liu appears to be that between a prospective student (Du) and master (Liu), rather than a collegial relationship, along the lines of many relationships documented in Han Yu’s corpus; nor can the relationship between the two men be more clearly elucidated, since there is neither internal nor external evidence for it beyond the single letter. Liu calls Du “student,” 生, throughout, and castigates him severely for seeking him out (and Liu Yuxi and Han Yu) to preach of “the Zhou” and “Confucius.” The letter also contains a useful point of medieval grammar, when Liu takes Du to task for not distinguishing clearly between particles expressing doubt or question (疑辭: 乎, 歟, 耶, 哉) and particles expressing definition or completion (決辭: 矣, 耳, 焉, 也). Liu Zongyuan ji, 3:889–891.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1298

Shields

(cont.) Author

Recipient

Date (est.)

Title

Content summary

Han Yu

Zhang Ji 張籍(ca. 766–ca. 830)

798

“Chong da Zhang Ji snu” 重答張籍書c (Again Replying to a letter from Zhang Ji)

Han Yu

Li Yi 李翊 801 or 802

Second letter to Zhang in response to Zhang’s criticism of Han’s idle, unserious behavior (gambling, making jokes, telling stories); much shorter than first letter, a more pointed rebuke of Zhang Second letter to Li (following the first letter on 文); Han criticizes Li for his attitude and conduct, presumption of special treatment Response to Li Ao’s letter in which Li urged Han to return to the capital; outrage and affection mixed as Han tries to convince Li of the correctness of his position (this letter also contains expressions of affection, and is examined in both sets)

Han Yu

Li Ao

800

“Chong da Yi shu”

重答翊書d

(Second Reply to a Letter from Yi)

“Yu Li Ao shu”

與李翱書e

(Letter to Li Ao)

a Ibid., 3:807–11; Quan Tang wen 574.2b–4.b. b Quan Tang wen 635.15a–17b. c Han Yu quanji jiaozhu, 3:1333–41; Han Yu guwen jiaozhu, 3:564–73; Quan Tang wen 551.3a–4b. d Han Yu quanji jiaozhu, 3:1460–61; Han Yu guwen jiaozhu, 3:738–41; Quan Tang wen 552.6a–6b. e  Han Yu quanji jiaozhu, 3:1386–91; Han Yu guwen jiaozhu, 3:773–86; Quan Tang wen 552.10a–11b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1299

Three considerations seem to shape the relative force of anger or irritation expressed in a given letter: the nature of the disagreement or criticism and whether it implicated both parties’ personal conduct; the degree of intimacy between the two writers; and the ultimate object of the letter in the mind of the letter writer—that is, did the writer wish to convince the recipient of his faults or bring the relationship to an end? In all five cases, the issue of personal conduct comes into play at some point—either the recipient’s conduct toward the letter writer or, in two cases, some actions of the letter writer that had previously been criticized by the addressee (Han Yu’s responses to Zhang’s and Li’s criticism and advice). In other words, it appears there needed to be some interpersonal conflict, not merely intellectual debate, for disagreement to require emotionally charged language. Not surprisingly, greater intimacy in the social relationship seems to have allowed for more vehemence in terms of exclamation, as we see most vividly in Han Yu’s letter to Li Ao. But formal techniques in a less familiar letter, such as the repetition of fixed phrases and the recipient’s name in Liu Zongyuan’s letter to Han Yu, could create a sarcastic, caustic tone that both mocked the recipient and closed the distance between the two men. Finally, a writer’s ultimate objective in composing the letter, insofar as we can determine it, seemed to play a critical role in the modulation and display of anger, particularly in the opening rhetorical moves and in the closing of a letter where the optative mood often appeared, and where a writer could amend or recuperate the tone.38 The most negative closing of the five appears in Li Ao’s letter to Hou Gao, which ends abruptly (and is perhaps truncated) with, “As for your Way, you should follow it yourself—but do not lecture me with it” 子之 道, 子宜行之者也, 勿以誨我.39 This closing and Li’s defensive statements leading up to it imply that there was little left to lose in his relationship with Hou Gao. We find examples of both positive and negative goals in Han Yu’s second letters to Zhang Ji and Li Yi, both of whom were students of his (Li Yi may have been merely a prospective student) when the letters were written. In Han’s second letter to Zhang, his irritation at Zhang’s persistent criticism is offset by the reasoned manner in which he offers historical evidence to defend his 38  In a study of displays of anger among European medieval aristocrats, Richard E. Barton has argued that “the performance of lordly anger was frequently linked to compromise and negotiation…[and that] anger signaled the initiation of a process of restructuring social relationships.” Barton, “ ‘Zealous Anger’,” 162. Although here the anger in question is aimed at men of roughly equal status, the desire to renegotiate or shift the relationship between writer and recipient emerges both implicitly and explicitly in the letters. 39   Quan Tang wen 635.17b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1300

Shields

actions, and by his calling Zhang “my fellow/my follower/student” (wu zi 吾子), though he does take one parting shot at Zhang’s lack of reflection on the issue: “How could [laughter and teasing] harm the Dao? How could you, my fellow, not have considered this?” 惡害於道哉? 吾子其未之思乎.40 But in his second letter to his student Li Yi, Han Yu’s obvious annoyance at what seems to have been Li’s presumption or demands for attention seems in no way tempered by concern at losing a student—the criticism of the closing, in fact, seems aimed at driving Li Yi away. There is truly a distinction between worthiness and meanness; how can you be so anxious to get something by which you can be established, and not disregard it if people do not know you? I have never heard of [someone] making a great echo with a tiny sound, let alone with [the help of] my being so earnest on your behalf? I have had a stomach illness of late, and I am in poor spirits, therefore I do not write this myself. Yu reports.41 賢不肖固有分矣，生其急乎其所自立，而無患乎人不知己，未嘗聞響大而聲 微者也。況愈之於生懇懇耶? 屬有腹疾，無聊。不果自書。愈白。42

In all five letters, the writers’ display of emotion blends highly literate strategies, such as hypotaxis, rhetorical questions, logical reasoning and evidence, with features that imitate oral discourse, such as exclamations, calling the recipient’s name, and posing a rapid series of questions that mimics the behavior of “talking over” someone in argument. However, the choice to color a disagreement with some kind of anger, beyond mere differences of intellectual or philosophical opinion, had to be made carefully. If the emotional expression were too excessive, a letter could devolve into mere feeling and fail to be persuasive. For this reason, the prominent features of these letters are those that exercise tight control over the argument in order to place the fault of poor conduct or misguided criticism firmly onto the recipient. Only rarely do we see what we might think of as a lexicon for anger in the letters, terms for annoyance, displeasure, or similar negative emotions used by the writers to describe their own states. As the historian of early modern English letters cited above notes, “sentiments could be conveyed without being 40   Han Yu quanji jiaozhu, 3:1341; Han Yu guwen jiaozhu, 3:571. 41  Han Yu uses the term sheng, “student,” to address Li in both letters, rather than zuxia 足下, wuzi 吾子, or Li’s personal name. 42   Han Yu quanji jiaozhu, 3:1461; Han Yu guwen jiaozhu, 3:741. Early editions of the text do not have the word guo in the final line, and commentators suggest it is an inserted error.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1301

named. Writers of letters rarely stated ‘I am angry’ in the letter; rather, they vented their complaints and accusations with sufficient vehemence so that the recipient would be fully cognizant of their state of mind.”43 In Liu Zongyuan’s letter to Han Yu about the duties of a historian, Liu’s unusually flat statement of displeasure, shen buxi 甚不喜, sets the tone for his attack on Han Yu’s words, conduct, and potential influence: In the letter I obtained earlier that spoke of matters concerning history [historiography and the duties of the historian], you said that you had detailed all of it in your “Letter to Floriate Talent Liu”; now that I have seen the original draft, in my heart I am seriously displeased, and I think that it is even more erroneous than the previous year’s letter on the duties of the historian.44 前獲書言史事，云具與柳秀才書，及今乃見書稿，私心甚不喜，與退之往年 言史事甚大謬。45

This is the only declarative statement of personal displeasure in the five letters; however, words for negative emotions are used often, usually in response to the action or words of the recipient, and positive emotions are scarce (except in Han’s letter to Li Ao, which is why it is also considered in the affection letters). For example, in this letter Liu Zongyuan goes on to caricature Han’s worry about the burden of “praise and blame” (baobian 褒貶) that was the historian’s duty, accusing Han five times of “being afraid” (kong 恐, ju 懼, and kongju 恐懼) to speak. In effect, Liu charges Han with moral and personal cowardice: “[You] Tuizhi should reflect on this again, and what you can do, you should do swiftly; 43  Pollock, “Anger and the Negotiation of Relationships,” 523. 44  Liu’s reference to seeing an original draft (presumably in Han’s hand) of Han’s letter after seeing what was presumably a copy is a tantalizing and rare clue about the ways in which letters circulated among circles of friends and acquaintances during the mid-Tang. In his letter to Li Jian of 809, Liu mentions receiving one letter from Li directly from a messenger and two others, presumably copies, via Liu Yuxi, their mutual friend; in the same letter, Liu also protests that he cannot write directly to someone higher in rank (Li Jian’s brother) from whom he has not first received a direct letter; and finally he bids Li Jian to copy out his letter and show it to another mutual friend, Cui Qun, who cannot receive letters at his post. Elsewhere, he tells a recipient not to circulate his letters for fear of attracting ridicule. Liu Zongyuan ji, 3:801–2. Brief statements such as these indicate that mid-Tang literati expected their letters to be regularly shared within a circle of friends and would often make copies for others to do so. 45   Liu Zongyuan ji, 3:810–11.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1302

Shields

but if in the end you do not dare because of your fear, then within a day you can resign—so then why do you speak of ‘going along but planning [to resign]’ [citing a phrase from Han Yu’s earlier letter]?” 退之宜更思，可為速 為，果卒 以為恐懼不敢，則一日可引去，又何以云行且謀也.46 Other negative feelings of plaint and grief appear as responses to actions or words of both parties, as when Han Yu acknowledges four times Li Ao’s “anguish,” bei 悲, for him; or when, in his letter to Hou Gao, Li Ao admits that Hou “will surely resent me and repeat your words [of criticism]” 必將憤予而復其辭 when he reads Li’s letter. Liu Zongyuan accuses Han four times of being “deluded,” huo 惑, and Han Yu suggests that others will accuse him of being both “foolish” and “wild,” kuang 狂, in his letter to Zhang Ji. The few appearances of “joy,” le 樂, in these letters come in negative rhetorical questions, such as Han Yu’s repeated question to Li Ao, “How could I be happy [doing the thing you recommend]?” Beyond these content words for negative emotion, however, the most prominent and consistent lexical feature of the five letters is their use of emphatic particles, both exclamation particles and question final particles used repeatedly and in conjunction with exclamations. This feature of the text is not only a consequence of the writers’ constant use of rhetorical questions, but is also a reminder of the letters’ latent orality, the fact that they would often be read aloud. Whether or not the writers were aware of it, when read aloud, exclamations and question particles effectively forced the reader to perform the anger inscribed in the text through multiple exhalations of qi 氣, the embodiment of fa fen 發憤. The frequency of these specific particles in only five letters is astonishing, particularly when we compare the use of the same particles in the letters of affection. There are eighteen uses of the exclamation particle zai 哉 (compared to only three uses in the set of fourteen affectionate letters); thirtytwo uses of hu 乎 as a final particle in both questions and exclamations (compared to forty-one in fourteen affectionate letters, but fourteen of those appear in Han Yu’s letter to Li Ao, included in both sets); and fourteen uses of ye 耶 as 46  Han’s “Letter in Reply to Floriate Talent Liu” was only transmitted in the waiji of Han’s collection; later commentators largely agree with Liu Zongyuan’s criticism of Han’s statements in that letter. Liu Zongyuan’s repeated taunts about the “fear” Han has seem somewhat unfair, given that Han Yu uses the phrase weiju 畏懼, “to be awed and fearful,” only once, and with respect to the weighty responsibility of writing Tang history. Overall, Han’s tone in the letter to “Floriate Talent” Liu is more self-deprecating—appropriately for a response to a young man who had clearly asked him for advice and guidance about history—and less intellectually engaged with the topic of history than Liu’s attack suggests. Han Yu quanji jiaozhu, 4:1920; see especially the comments collected in Han Yu guwen jiaozhu, 5:3426–39.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1303

a question final particle (compared to three).47 (The exclamation “Ah” or “Alas,” jiehu 嗟乎, appears four times also in Han’s letter to Li Ao.) At the very least, the frequency of these particles made for rather noisy readings; more speculatively, they served to articulate the vexation of the writer, fulfilling the promise of letters to convey the intentions and the presence of the writer in the text. But expostulation alone did not suffice in these letters. In order to demonstrate the moral integrity of one’s anger, the relationship between the addressee’s conduct and the writer’s response had to be extremely clear and grounded not only in personal but also contextual and even historical evidence. We find this concern for evidence and clarity reflected consistently in the syntactic patterns of the letters: first, the writers show an overall preference for hypotaxis, including carefully constructed, often long complex sentences. Hypotaxis in these letters is indicated by the frequency of subordination, most often in conditionals (if-statements with 若 / 如 / 苟; “then” clauses with 則 / 乃), concessives (use of 雖 and 非 in “even though X [is or is not the case]…”), and counterfactuals (if-then statements with the premise in the subjunctive and the consequence in the condition, used to refute unacceptable or unrealistic possibilities, as in “if it were the case that…”; often introduced with 若使 / 使 / 設使).48 The writers also use syntactical variation as a way to bolster the strength of their argument: throughout the letters, we find writers varying strong declaratives, especially true/false statements using fei X ye 非 X 也 and A, B 也, with interrogatives (both real and rhetorical questions) and exclamations in rapid series, a technique that affirms the writer’s position while simultaneously calling into doubt the recipient’s views.49 Finally, the intensity and frequency of questions grows in each of the five letters as the writer approaches the close, which adds to the urgency of the writer’s feeling as well as rhetorically “silencing” the recipient. With respect to the features of hypotaxis and syntactical variation, and even to shifting the argument (“X is wrong, [my position] Y is 47  Moreover, all of these particles have mid- or open vowel finals—using David Branner’s reconstruction, we have tsei 哉, ghuo 乎, and ya 耶—adding to the sense of exhalation that would appear at the end of multiple sentences. (Yintong database accessed November 2014: http://yintong.americanorientalsociety.org/public/index.php). 48  On the debate over counterfactuals and their use in classical Chinese, see Harbsmeier, Science and Civilisation in China, 116–18. As Harbsmeier notes, jiashi 假使 and ruoshi 若使 always mark counterfactual clauses; in Liu Zongyuan’s letter to Han Yu, we see sheshi 設使, and in Li Ao’s letter to Hou Gao, we find shi 使 and rushi 如使 used in sequence. 49  From the perspective of speech act theory, rhetorical questions can also function as constatives; the variation in syntax that we see when the writer alternates assertions of fact (constatives) with rhetorical questions does not necessarily alter the illocutionary act.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1304

Shields

correct”), these are also common to political discourse, critiques of policies or others’ views; the notable difference in these letters is the way the friend or colleague is being personally attacked, doubted, or refuted by the writer. Li Ao’s opening discussion of his “Way” demonstrates the syntactic patterns of complex, hypotactic sentence structure and the alternation of truth claims with questions aimed at the recipient: My Way is not the Way of a single school; it is the Way that derives from the ancient sages. If my Way is blocked, then the Way of the superior person will disappear; if my Way is made manifest, then the Way of Yao, Shun, King Wen and King Wu, and Confucius has not yet been cut off from the world. If, blurring matters, I had agreed with the words of your last letter, that would have been like calling gong and shang the same note, and by what means would the Way become clear? Therefore I refuted your words, even knowing that you would surely resent me and repeat your words [of criticism]. As for your repeatedly instructing me to go along with the times to follow the Way—your so-called “times,” are they the times of humaneness and rightness? Or are they the crass, frivolous “times”? If they are humane and right, then how could my Way be obscured in them? If I were to go along with crass, frivolous times, then I would surely be riding their waves to follow the flow, watching how the wind blows while rising and receding. If this is [what you mean], then although you see me, you indeed do not recognize [understand] me. How much more so for other people in the world? 吾之道非一家之道，是古聖人所由之道也。吾之道塞，則君子之道消矣。吾 之道明，則堯舜文武孔子之道未絕於地矣。前書若與足下混然同辭，是宮商 之一其聲音也，道何由而明哉? 吾故拒足下之辭，知足下必將憤予而復其辭 也 。足下再三教我適時以行道，所謂時也者，乃仁義之時乎? 將浮沈之時乎? 苟仁且義，則吾之道何所屈焉爾。如順浮沈之時，則必乘波隨流望風而高下 焉。 若如此，雖足下之見我，且不識矣。況天下之人乎?50

By alternating his strong claims about his Way with rhetorical questions aimed at exposing Hou Gao’s ignorance, Li Ao’s tone turns sarcastic, a quality heightened by his quoting of Hou’s words. The strategy of repeating the other’s words in order to attack him is found in four of the five letters, and is especially frequent in Liu Zongyuan’s attack on Han Yu, as in this passage (Liu cites words 50   Quan Tang wen 635.15a–15b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1305

and phrases from Han’s “Letter to Floriate Talent Liu,” which is extant in Han’s collection, throughout his letter). Liu Zongyuan at this time had been in exile for a decade and was writing to Han Yu in Chang’an. Now you have said, “I’m but one person—how can I make it clear [praise and blame of human conduct, as a historian]?” If all your fellow [historians] also were to say the same thing, and those to come after them also were to say the same thing, and everyone all said, “I’m just one person,” then in the end who would be able to record and transmit [history]? 今退之曰，我一人也何能明?則同職者又所云若是，後來繼今者又所云若 是，人人皆曰，我一人，則卒誰能紀傳之耶?

In throwing Han’s words back at him, Liu Zongyuan here and through his closing arguments constructs a reductio ad absurdum, suggesting the collapse of Tang historiography under Han’s influence: Now if [everyone] were to be as learned as [Han] Tuizhi, speak and write like Tuizhi, enjoy debating like Tuizhi, valiantly calling themselves correct and upright, “bold and soldierly,” like Tuizhi,51 and yet still spoke in this manner [“fearing” to assign praise and blame], then will there not ultimately be no one to whom the writing of Tang history can be entrusted? 今學如退之，辭如退之，好議論如退之，慷慨自謂正直行行焉如退之，猶所 云若是, 則唐之史述，其卒無可託乎?

The personal nature of the charge and the fact that Liu Zongyuan sees Han Yu as the emblem of a larger cultural failing are made even more clear by Liu Zongyuan’s use of Han’s style name, Tuizhi (rather than zuxia, which Liu Zongyuan uses for other peers and colleagues), which he repeats no fewer than sixteen times in the letter. The importance of indirectness in expressing anger is best demonstrated by the frequent use of rhetorical questions throughout the letters, often in rapid series with question particles that expect a negative response, such as 豈… 乎 / 耶, or 豈… 乎哉, and 安… 耶 / 乎? This technique communicates negative 51  Liu Zongyuan alludes here to a description of Confucius’s disciple Zilu (from Analects 11.13), who stood in a “bold and soldierly” fashion (hang hang 行行), when the Master commented, “Someone like Zilu will not get to live out his years.” Liu appears to be mocking Han Yu’s self-representation (zi wei…) as well as commenting on his impetuosity.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1306

Shields

emotion indirectly in two ways. First, the repetition of questions suggests that the recipient is foolish enough to hold the positions the writer is questioning and needs to be instructed multiple times (thereby implicitly insulting him); second, the piling up of a string of questions in the text mimics the effect of a rising voice in a debate, intensifying the speaker’s demands on the listener without allowing for response or a break in the attack. Han Yu’s questioning of Li Yi’s motives conveys his annoyance and his suspicion in this manner. In the preceding sentence, Han Yu suggested that he has welcomed Li Yi like Mencius did his students, “not refusing any who come” 來者不距.52 Nevertheless, is it your intention to seek knowledge of me? Or to seek profit from me? Is it that you think to spread the Way of the Sages? Or is it that you desire to improve yourself so that other people cannot match up to you? 雖然，生之志求知於我耶? 求益於我耶? 其思廣聖人之道耶? 其欲善其身而使 人不可及耶?53

In the center section of a different letter, that to Li Ao, Han Yu’s repeated questions capture not only his frustration with Li Ao’s lack of understanding but also his distress about his lack of choices at the time: But this is only one issue—what benefit would there be in my going to the capital as you said? If in my relationship with you there are still things you do not understand, then how can other people know me? To gather up all I have and rush off, fleeing to mingle with officials and ministers, opening my mouth to debate and discuss—how could I possibly fit in?… Those who know me are certainly few, and those who know and love me and yet are not spiteful toward me are even fewer. So if I had nothing with which to support my family, and I had no one to follow [in some employ], how in the end would I manage?

52   Mencius 7B30. One of Mencius’s followers said of him, “When you give lessons, Master, you do not chase after those who leave, nor refuse any who come. You simply accept anyone who comes with the right heart.” Han Yu refers to Li as “one who comes,” laizhe, twice in the letter, implying that Li does not in fact have the “right heart.” Translation from Van Norden, Mengzi, 191. 53   Han Yu quanji jiaozhu, 3:1461; Han Yu guwen jiaozhu, 3:740–41.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1307

此一事耳。足下謂我入京城，有所益乎? 僕之有子猶有不知者，時人能知我 哉? 持僕所守，驅而使奔走伺候公卿間，開口論議, 其安能有以合乎?・・・其 知我者固少，知而相愛不相忌者又加少，內無所資，外無所從，終安所為乎?

This letter in particular shows us how a disagreement that may have originally been a matter of philosophical or practical concerns could escape the boundaries of the intellectual and spill into the realm of the two men’s personal relationship in the context of a collegial letter. When charges and rhetorical questions are aimed at the recipient’s own words, views, or former actions, as they commonly are in these letters, the force of the writer’s anger or irritation tightens the textual focus on the dyadic relationship of writer and addressee. The immediate anger of the text could also obscure other resentments and jealousies latent in the collegial relationship that the writers do not acknowledge. We do not have the subsequent responses from the recipients for any of the five letters, but we do have Liu Zongyuan’s subsequent letter to Han Yu, in which he implicitly apologizes to Han by moderating his tone, acknowledging Han’s own trials, keeping to a discussion of facts, and even including comments on his illness and disgrace (he had at that point been in exile for a decade) and inability to serve as a historian ever again.54 Certainly Han Yu and Li Ao continued to be close companions in subsequent years, despite this emotional dispute. This ability to compromise, or to reexamine the source of conflict, may indeed have been prompted by the strength of the feelings provoked by conflict.55 In these mid-Tang letters, it appears that expressing anger in disagreement was allowable within limits; the posture of moral superiority and indignation, if grounded in evidence drawn both from personal experience and from history or social context, could justify emotional outbursts and make them more persuasive—if not to the recipient, then to a broader group of readers the writers hoped to convince. This points to what was surely a positive function of medieval letters as sociotexts: letters could serve as “social buffer,” a means of negotiating conflict that did not require face-to-face shouting, in the same way that shame or disgrace might also be more easily acknowledged in texts 54   Liu Zongyuan ji, 2:811–12. Jo-shui Chen argues that this later period of the two men’s relationship, during Liu Zongyuan’s time in Yongzhou, was a period in which Han came to admire Liu’s writing, despite his opposition to Liu’s activity in the Shunzong regime. Han Yu certainly defended Liu’s talent and character vigorously in his 819 epitaph for Liu. Chen, Liu Tsung-yuan, 74 n. 30. 55  One scholar of anger in medieval European texts and society makes this link as well, see Barton, “ ‘Zealous Anger’,” 162–63.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1308

Shields

than by bowing and weeping in person.56 But these letters also show that allowing anger to creep into disagreement, even in its milder versions, required a highly self-conscious and crafted rhetoric—from careful lexical choices, to varied syntax, to finely tuned strategies of repetition and recursion—to be acceptable among close colleagues in mid-Tang literati society. In this way, the writers were able to display precise control of their language while they were venting disruptive emotion. 4

“All I had stored up in my heart I then wished to quickly tell you” 心所蓄者便欲快言: Affection and Self-Representation

Disagreements over conduct or ideas could provoke caustic language, even when writing to correspondents far away, but the work of sustaining social relations, one of the key functions of letters, required a softer tone, one that could express affection, commitment, and concern for friends and colleagues. Rather than being prompted by a particular season or extreme event, as we saw in the etiquette manuals and in the Wen xuan letters, or aroused by a personal conflict, mid-Tang collegial letters in which we find expressions of affection were largely composed for non-specific occasions. The openings to these letters indicate that they were part of an ongoing if sometimes sporadic correspondence between men known to one another for some time who were (in all but a few cases) separated by great distances. Liu Zongyuan’s opening to his letter to Li Jian 李建 captures the pleasure and the value of this kind of exchange and reminds us of the ways letters could be exchanged within a circle of acquaintance: To Biaozhi [style name of Li Jian]: the provincial courier arrived unexpectedly, and I received your letter, and I also received through Liu Yuxi your letter from before that; your thoughts in both of them were earnestly generous. Zhuang Zhou said, ‘Those who withdraw to wild, brambled places, hearing the tramping sound of human footsteps, are indeed happy.’ Since I have been among the southern barbarians, obtaining your two letters was better than the best medicine, making me more happy than I can say.

56  Schneider, The Culture of Epistolarity, 132.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1309

杓直足下。州傳遞至，得足下書，又於夢得處得足下前次一書，意皆勤厚。 莊周言，逃蓬藋者，聞人足音則跫然喜。僕在蠻夷中，比得足下二書，及致 藥餌，喜復何言。57

As Liu’s Zhuangzi reference confirms, letters embodied the person of the writer, and this particular epistolary trope emerges prominently in letters expressing affection, which are filled with rhetorical techniques that invoke or replicate the presence of the friend. The acknowledgement of receipt of letters was not merely practical information—it also affirmed connection and continuity, and expressed gratitude implicitly without having to resort to more formal thanks.58 Nine of the fourteen letters here refer to receiving a letter (or not receiving one, and regretting that fact) from the addressee, and seven of those go on to cite directly from the letter or to discuss its contents. Letters in these collegial relationships were of course gifts, a fundamental currency of friendship, and responding with a letter both repaid the gift and continued the exchange. The underlying dynamic of gift exchange could color the letters in unexpected ways.59 Although the letters do on occasion contain requests or responses to requests (for opinions on literary work and one request from Liu Zongyuan for books), the letter-writers generally portray themselves as making no demands of the reader, depicting the recipients as the perfect audience, the ones who will listen sympathetically and understand completely. Within the texts, expressions of affection and trust helped to create a rhetorical space in which writers could naturally unburden themselves. However, this unburdening was frequently intended as an implicit request for aid, particularly for those in political exile. Through the content and the physical text of the letter, the writers’ self-revelation became a burden of knowledge that the recipient had to manage carefully and circulate appropriately. As we saw in the shuyi models, constructing the distant friend as the zhi yin 知音 was a standard convention of friendship letters, as was the practice of assuring the recipient that he was remembered and missed. The existence and the likely pervasiveness of these conventions for expressing affection between 57   Liu Zongyuan ji, 2:801. The passage Liu paraphrases from Zhuangzi comes from chapter 24, “Xu Wugui” 徐無鬼, Zhuangzi jiaoquan, 3:921. 58  On the importance of conventions for acknowledging letters within a social community, see Schneider, Culture of Epistolarity, 15–16. The acknowledgement often continues within the body of the letter, as the writer cites or refers to parts of a previous letter in discussion. 59  On this issue, see Xiaofei Tian’s article in this volume.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1310

Shields

friends presented mid-Tang literati with competing social and literary imperatives: although they needed to sustain emotional ties with their colleagues in friendly letters, they had to do so in a way that seemed authentic, original, and in no way perfunctory. As Liu Zongyuan’s description of receiving letters from Li Jian tells us, this was emotional expression intended to be seen by others; Liu Zongyuan’s happiness at receiving Li Jian’s letters as well as the sense of humiliation he describes later in the text would also have been conveyed to Liu Yuxi and others in their circle. What we find in these letters, therefore, is a common set of gestures and tropes for expressing affection among friends and colleagues, a set also found in abbreviated forms in medieval etiquette manuals, that mid-Tang writers adapted and transformed. Some consistent gestures beyond portraying the recipient as the perfect listener included: referring to previous meetings or memories of the recipient; expressing a wish to be reunited in the future; and suggesting unsettled feelings at being separated from the recipient. However, the two conventions we saw in the pengyou shuyi as well as the Wen xuan examples—direct statements of sorrow and a physicalization of feeling through sighing and weeping—are notably absent in these letters. These rhetorical patterns of the collegial letters should make us even more aware of their literariness, their representation of the writer’s craft. Furthermore, their avoidance of the many mundane details that must have been negotiated by letters with friends shows the likely bias for selecting and transmitting only letters that captured ideal versions of the author’s literary talent, not simply his character—or perhaps a later process of editing letters that were to be included in collected works. The letters are also coherent with respect to their overall avoidance of parallel prose, again with the exception of certain passages in Bai Juyi’s letter. Until we can analyze a broader selection of Tang collegial letters, I do not think we can conclude that this is necessarily a guwen 古文, “ancient-style prose,” choice, particularly since the set of fourteen letters includes texts by Bai Juyi and Pei Du 裴度 (765–839), both of whom knew Han Yu but were not associated with his literary style; but it is clearly a conscious decision. This style choice alone may have seemed sufficient to convey authenticity, since avoiding the metric regularity and ornament of parallel prose required writers instead to reference real-life events, people, and places in their communication, though allusions are still very common. In addition to style choices, the writers’ context also shaped the content of the letters: twelve were written to friends or colleagues who were at a great distance from the writer, a fact that surely heightened the communicative urgency and possibly the emotional burden of the texts.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

Author

Recipient

Liu Zongyuan

Xiao Mian 811

Liu Zongyuan

Li Jian

Liu Zongyuan

Lü Wen

Liu Zongyuan

Yang Huizhi

蕭俛

呂溫

楊誨之

Liu Yuxi

Liu Zongyuan

Date (est.)

Title

“Yu Xiao Hanlin Mian shu” 與蕭翰林俛書a (Letter to Xiao Mian of the Hanlin Academy) 809 “Yu Li Hanlin Jian shu” 與李翰林建書b (Letter to Li Jian of the Hanlin Academy) 808– “Yu Lü Daozhou lun 11 Fei Guoyu shu” 與呂 道州溫論非國語書c (Letter to Lü Wen of Daozhou on [my] Against the Discussions of the States) 811 “Yu Yang Huizhi shu” or 與楊誨之書d 812 (Letter to Yang Huizhi)

808? “Da Liu Zihou shu” 答柳子厚書e (Letter in Reply to Liu Zihou)

1311

Content summary

(in exile) Received letter with good news of Xiao’s promotion, writing to congratulate; sharing his state of mind and current unhappiness with Xiao (in exile) Acknowledging exchange of letters, happiness of hearing from Li; sharing his current state (in exile) Brief letter ostensibly about his reasons for writing the Fei Guoyu, but also a discussion of his disgrace, expresses need to have Lü Wen to listen to him (in exile) Brief personal letter accompanying longer letter containing the parable of the cart, an allegory of how to get along in contemporary society, and explaining why Liu wrote this story for Yang (who was known to Liu since childhood) (in exile) Brief response to Liu Zongyuan about his recent work, which lzy asked him to evaluate

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1312

Shields

(cont.) Author

Recipient

Date (est.)

Pei Du

Li Ao

801

Title

Content summary

“Ji Li Ao shu”

Praise of Li Ao’s recent work and modest praise of Li as a person (they are cousins by marriage); discussion of good writing (wen) and its qualities; criticism of Han Yu’s frivolity and fondness for unusual language (Pei accuses Han of “making a game out of literary writing” 以文為戲) (in exile) Long discussion of literary writing, prefaced and concluded by personal exchange (in exile) Long selfexplanation and defense of actions that had him demoted in 815; discussion of Yang’s worthiness, character; Bai describes his resignation to current fate (in exile) Emotional letter sent to Yuan after 3 years apart, two years without contact Written to peer after multiple failures to pass the selection exam (xuan 選) or gain a post; need to explain himself, express his despair and loss of hope for future, despite Cui’s praise of him as an “uncarved jade”

寄李翱書f

(Letter Sent to Li Ao)

Bai Juyi

Yuan Zhen 815

Bai Juyi

Yang Yuqing 楊

816

虞卿

Bai Juyi

Yuan Zhen 818

Han Yu

Cui Lizhi 崔立之

795

“Yu Yuan Jiu shu” 與元九書g (Letter to Yuan the Ninth) “Yu Yang Yuqing shu” 與楊虞卿書h (Letter to Yang Yuqing)

“Yu Yuan Weizhi shu” 與元微之書i “Letter to Yuan Weizhi” “Da Cui Lizhi shu” 答 崔立之書j “Letter in Reply to Cui Lizhi”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1313

Author

Recipient

Date (est.)

Title

Content summary

Han Yu

Meng Jiao

800

“Yu Meng Dongye shu” 與孟東野書k “Letter to Meng Dongye”

800

“Yu Li Ao shu”

(in separation) Brief note praising Meng, expressing sadness at separation, description of current life, plans for future (in separation) Response to Li Ao’s letter in which Li urged Han to return to the capital; outrage and affection mixed as Han tries to convince Li of his position Long, crafted text sent to a “same-[examination pass] year” (tongnian 同 年) friend, senior official, discussion of friendship, social knowledge, praise of Cui; Han expresses concern about his own future, current difficult circumstances Brief note sent to a tongnian, pained at hearing criticism and unpleasant gossip from Feng, but thanking him for trying to help and requesting that he listen and pass on anything else

孟郊

Han Yu

Li Ao

與李翱書

(Letter to Li Ao)

Han Yu

Cui Qun

802

崔群

Han Yu

Feng Su

807

馮宿

“Yu Cui Qun shu” 與 崔群書l (Letter to Cui Qun)

“Da Feng Su shu”

答馮宿書m

(Letter in Reply to Feng Su)

a b c d e f g

Liu Zongyuan ji, 2:797–800. Ibid., 2:800–803. Ibid., 2:822–24. Ibid., 3:847–49. Liu Yuxi ji jianzheng, 1:265–66. Quan Tang wen 538.1b–4a. Bai Juyi ji jianjiao, 4:2789–2905.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1314

Shields

h Ibid., 4:2769–75. i Ibid., 4:2814–17. j Han Yu quanji jiaozhu, 3:1261–68; Han Yu guwen jiaozhu, 1:691–711. k Han Yu quanji jiaozhu, 3:1425–30; Han Yu guwen jiaozhu, 1:574–78. l Han Yu quanji jiaozhu, 3:1532–39; Han Yu guwen jiaozhu, 1:803–21. m Han Yu quanji jiaozhu, 3:1711–15; Han Yu guwen jiaozhu, 1:853–61.

As we saw in the letters expressing anger, the expressions of affection in these letters tend to be indirect, even on a lexical level. Terms for happiness—both “pleasure,” xi 喜, and “joy,” le 樂—are common in the letters; there are eighteen uses of xi and twenty-four of le, including the negative version bu le, compared with three and four appearances, respectively, in the letters with anger. Stating pleasure or other positive emotions inspired by the recipient’s communication or memory is one common convention, as we saw in Liu Zongyuan’s letter. Positive words of praise for the recipient are also frequent, but neither affection nor praise is given gratuitously in these letters; on the contrary, the letter writers often take pains to describe the specific causes of pleasure and praise. In Liu’s letter to Yang Huizhi, for example, he recalls seeing Yang mature physically and in his literary talent: “At first I did not know you, and then I arrived in Tanzhou, where I saw that your spirits had become more calm, your work more focused; your demeanor was more weighty, and you spoke less. And in my heart I was happy.” 僕未始知足下, 及至潭州, 乃見足下氣益和, 業益專, 端重而少言, 私心乃喜. In Pei Du’s letter to Li Ao, Pei praises Li’s work in conventional terms as “truly excellent, truly excellent” (shen shan shen shan 甚 善甚善), then goes on to substantiate his assessment of Li’s work with a cautious evaluation of his character: “However, in knowing you, I have not yet come to know more than [what you disclose in your literary work], that you are diligent in your learning, delight in writing, and rectify yourself through the Six Classics… I presume to think that you are to be treated with directness and honesty and not to be handled with pleasing flattery” 然僕之知弟 也, 未知其他, 直以弟敏於學而好於文也, 就六經而正焉… 竊料弟亦以直諒見 待, 不以悅媚相容. By explaining his personal understanding of Li Ao, Pei Du

forestalls suspicions that his words might be taken as mere “pleasing flattery,” which would give his assessment even more weight as testimony to others, a fact that Pei Du surely realized. In his lengthy letter of 816 discussing literature, Bai Juyi’s statement to Yuan Zhen, “since I love your poetry…” 僕既愛足下詩, becomes the pretext for his encyclopedic discussion of literary history, poetic values, and their shared poetic tastes. Expressing praise and pleasure toward

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1315

the recipient, in other words, was often the starting-point of these letters, not the final goal. Recalling the emotional charge of a shared event at the opening or close of a letter was another indirect technique to convey affection, establish a mood of intimacy, and also underscore social bonds between writer and recipient, as in Bai Juyi’s letter to Yang Yuqing: When I returned to the capital again [in 814, after mourning for his mother], you had been posted in Hu district and were fettered with official work, so we rarely saw one another; within the space of a half a year or so, we only met to talk and laugh three or four times. But when the order for my demotion came down, I had to head east the very next day; you came from the western part of the city to see me, all the way to Zhaoguo ward, but you did not make it in time. Riding your horse all the way to the Chan river, we were finally able to clasp hands; with pity you bade your farewell, and we spoke of nothing but that. 自僕再來京師，足下守官鄠縣，吏職拘絆，相見甚稀。凡半年餘，與足下開 口而笑者，不過三四。及僕左降詔下，明日而東，足下從城西來，抵昭國 坊，已不及矣。走馬至滻水，才及一執手，憫然而訣，言不及他。

In the next section of the letter, Bai states that because their parting had been so urgent, he is now forced to write in order to explain his actions and the consequences more fully. The memory of their parting therefore stages Bai’s subsequent self-revelation as yet another intimate encounter between close friends. In an even more indirect fashion, Han Yu opens his letter to Meng Jiao by suggesting that the bond between himself and Meng, who Han claims knows him so well, is so deep that they mirror one another and can understand each other’s emotions without stating them directly: I’ve been parted from you a long time now. From my own longing for you, I know that you must be anxiously worried about me. We are both bound by work, and can’t get together. When I am around others, since I do not get to see you and spend my days with you, you know if my heart is happy or not! 與足下別久矣。以吾心之思足下，知足下懸懸於吾也。各以事牽，不可合 并，其於人人，非足下之為見而日與之處，足下知吾心樂否也。

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1316

Shields

In both his letters to Yuan Zhen, Bai Juyi makes a similar claim about their intimate understanding of one another’s emotional states, though without expressing affection directly. His 816 letter discussing literature includes the repetition of the exclamation (read either as a question or an imperative), “Weizhi, Weizhi, do you know my heart?” 微之微之, 知我心哉. Although they avoid direct utterances of affection toward the recipients, in these letters, the writers use key terms for feeling and mutual understanding so as to establish the emotional conditions—not simply the practical reasons—for the selfdisclosure that follows. Whereas in the letters expressing anger we found a consistent preference for hypotaxis, repetition, and recursion, in these letters we see a greater variation of parataxis and hypotaxis from letter to letter, which seems largely to be a consequence of the writer’s choice of style register and degree of intimacy with the addressee. Affection, in other words, could appear in many forms and in letters of many different styles. The style register of the letters is communicated in many linguistic features, but modes of first- and second-person address are critical to establishing the relative distance or closeness of a relationship. One feature that indicated intimacy was shifting in and out of more informal or intimate terms for “you,” as Han Yu does in his letter to Li Ao, as he moves back and forth between the default term for “you” in letters to peers, zuxia, and zi 子, thereby shifting his tone toward Li from that of teacher to student to that of friend to friend.60 Letters that aim for a more formal style register overall, such as Han Yu’s letter to Cui Qun (which addresses the topic of friendship and knowledge in both a personal and theoretical manner) or Liu Zongyuan’s letter to Xiao Mian (in which Liu congratulates Xiao on being appointed to the Hanlin Academy and laments his own fate), tend toward hypotaxis, longer sentences, and more complex chains of argument. Conversely, letters to friends with a lower, more informal style register (such as Han Yu’s letter to Meng Jiao or Liu’s letter to Li Jian) tend to have shorter, less clearly connected sentences. 60  Three letters use only a single mode of address for the recipient and for themselves throughout the letter (zuxia 足下 and pu 僕), which is also the most common choice in the larger body of collegial letters. However, in the letters with affection, shifting down in register with pronouns (addressing the recipient as zi 子 and referring to oneself as wu 吾, wo 我, or yu 余), which occurs in nine letters, seems to have been a simple way to vary the register and also the intimacy of address. The repetition of the recipient’s style name has the same effect on register. Perhaps the most telling is the use of multiple terms of address suggesting familiarity or intimacy, as we see in Liu Zongyuan’s letter to Lü Wen, where he calls Lü zuxia and by his style name, Huaguang 化光, and also calls him junzi and zhi wo zhe 知我者, “one who knows me.” Liu Zongyuan ji, 2:822–23.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1317

Thus in Han’s letter to Cui Qun, we see Han’s argumentation laid out elaborately (using constructions that highlight Han’s reasoning, such as 雖… 抑…; 以此而退之…; 所以言者…), to the point where Han seems somewhat defensive in explaining his praise of Cui: In my humble foolishness, there is nothing I truly understand, and yet there are no works of the sages that I have not read. Although I am not exhaustively familiar with their precision and breadth, their scope and details, you also could not say that I haven’t entered deeply into their flows. Thus, if we use [my learning] to examine this issue [of valuing you], use it to assess this question, then I truly know that your greatness stands above the crowd, but do not ask where I attained this knowledge. Given the true feeling between us, is it even necessary for me to say this and only then have you understand it? The only reason I say all this is that I fear you’ll think I have too many deep friendships, and that I do not take to heart the difference between white and black. 僕愚陋無所知曉, 然聖人之書, 無所不讀。 其精麤巨細, 出入明晦, 雖不盡識, 抑 不可謂不涉其流者也。 以此而推之, 以此而度之, 誠知足下出群拔萃, 無謂僕何 從而得之也。 與足下情義, 寧須言而後自明耶? 所以言者? 懼足下以為吾所與 深者? 多不置白黑於胸中耳。

In Liu Zongyuan’s letter to Li Jian, in contrast, the simple paratactic style manages to make Liu’s state of mind equally clear: Even if I could get rid of my illness, and my body were strong again, I am far away from this age, and I could not overcome the fact of being a forty-year-old traveler [an exile]. I just passed my thirty-seventh year, and it was no different from the blink of an eye. As for the worthlessness of what I have gotten thus far [in life], I have indeed already laid this out. Biaozhi, do you not think this is true? 假令病盡已，身復壯，悠悠人世，越不過為四十年客耳.61 前過三十七年，與 瞬息無異. 復所得者，其不足把翫，亦已審矣。杓直以為誠然乎?

The conclusion of Han Yu’s letter to Meng Jiao proceeds in a similarly plain manner, providing important information in a way that resembles the more 61  The original text reads “thirty” 三十, but the editors of Liu Zongyuan ji note that some editions have the variant “forty” 四十, which I adopt here.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1318

Shields

quotidian style of later chidu, including the conventional comments on the weather, a wish for Meng, and Han’s health: Li Xizhi [Ao] is to marry my deceased brother’s older daughter a month from now—we are waiting for him to arrive soon. Zhang Ji is in Hezhou, in mourning, and his family is very poor. I feared that you would not know, and so I tell you all this. He hopes that you will come so you can see one another. From there to here is a long way, but if you travel by boat all the way, you can get here. My hope is that you’ll plan this quickly. Spring is almost over, and the seasonal weather is turning hot; I wish you good fortune. My eye problem is worse, which is really a bother; but I will not detail everything here. Yu bows twice. 李習之娶吾亡兄之女，期在後月，朝夕當來此。張籍在和州居喪，家甚貧，恐 足下不知，故具此白。冀足下一來相視也。自彼至此雖遠，要皆舟行可至。速 圖之，吾之望也。春且盡，時氣向熱。惟侍奉吉慶。愈眼疾比劇，甚無聊。不 復一一。愈再拜。

In contrast to these uses of parataxis in letters to familiar friends, which more resemble oral discourse, Bai Juyi includes sections of paratactic parallel prose into his 817 letter to Yuan Zhen, opening his letter with lines of affection and longing that seem close to shuyi models in some ways: Weizhi, Weizhi! I have not seen your face for three years now; I have not received your letters for almost two years now. How long is human life? That the distance between us be so broad! It’s even worse for us, whose hearts are bonded like lacquer, bodies consigned to the far realms of Hu and Yue; when we were near, we could not get together; now secluded, we cannot forget each other. Bound and tied together yet blocked and divided, we each are growing white-headed. Weizhi, Weizhi—how can it be like this! How can it be like this! 微之微之! 不見足下面已三年矣。不得足下書欲二年矣。人生幾何? 離闊如 此。況以膠漆之心。置於胡越之身，進不得相合, 退不得相忘。牽攣乖隔，各 欲白首。微之微之，如何如何!62

Interestingly, though Bai Juyi uses this parallel style to lament the distance between them and evokes the intimacy of the friendship with the image of “hearts 62   Bai Juyi ji jianjiao, 4:2814.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1319

bonded like lacquer,” he utters no direct statements of affection for Yuan here or elsewhere in the letter. He says that he feels “anguished,” cece 惻惻, when he reads Yuan’s descriptions of his illness and poor conditions, and yet Bai avoids words for longing or sadness, such as si 思, huai 懷, or chou 愁. Even in this dramatic piece, one that we recognize to be replete with feeling, Bai Juyi closes with the indirect utterance that is as much question as wish: “Weizhi Weizhi, do you understand my heart on this night?” 微之微之，此夕此心，君知之乎. Throughout the letters, emotional indirectness appears to be a constant, no matter whether a writer adopts a hypotactic or a paratactic style. The ultimate goal of many of these letters that express affection, which the writers make explicit at different points in the texts, is for the writer to express his feelings, his recent history, and his current condition to the addressee and, presumably, to the larger circle of people who might read the letter. Eight of the fourteen contain a direct statement of that goal, as when Bai writes to Yuan upon his receipt of a bundle of Yuan’s writings: … upon opening the scrolls and understanding their meaning, suddenly it was as if I saw your face; and all that I had stored in my heart, I wished to quickly tell you. Though I often doubted myself, I did not feel that we were separated by ten thousand li. But soon I felt a spirit of sad vexation, and I wished to have some way to release it, so I went back to reflect on my former ambitions, and endeavored to write this letter. … 開卷得意，忽如會面。心所蓄者，便欲快言，往往自疑，不知相去萬里 也。 既而憤悱之氣思有所洩，遂追就前志，勉為此書。63

In his letter to Li Ao, even though his wish to defend himself (as well as express his anger at Li Ao’s proposals) is perfectly clear, Han Yu closes his letter with a similar gesture: “I have therefore tasked a courier to deliver this, awaiting your thoughts [in reply], and also explaining myself” 故專使馳此候足下意，  并以自解.64 In his letter to Cui Lizhi, Han Yu even implies that a fuller discussion of their personal experience (and not merely their intellectual positions) is needed for the two men to understand one another: “However, there still seems to be areas where we don’t understand one another; is this not why you intentionally sought to provoke me [to reveal my thoughts]? If not, then why do you not treat me as a man of substance? I cannot remain silent and would like to clarify myself.” 然尚有似不相曉者，非故欲發余乎? 不然，何子之不以 63  Ibid., 4:2790. 64   Han Yu quanji jiaozhu, 3:1388; Han Yu guwen jiaozhu, 1:784.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1320

Shields

丈夫期我也? 不能默默，聊復自明.65 Han Yu then proceeds to give Cui a his-

tory of his trials by fire, as his youthful learning and his beliefs about success and failure were tested in what he sees as the biased, ludicrous process of the placement examination, which he had failed repeatedly at the time of writing the letter in 795. Where the writers had goals in addition to recounting their personal feelings and circumstances, such as giving their addressees instruction or assessment, they explain those carefully as well, framing their advice in ways meant to lessen any sting of rebuke. When writing to Yang Huizhi, Liu Zongyuan first notes his concern for Yang’s impetuous character, fearing that Yang would suffer a fate similar to Liu’s own, and he prefaces the advice with implied affection: Good people are few, and bad people many, and therefore those who care for you are few, whereas those who would harm you are many. I truly wish you to make yourself square in the middle but round outside [moral but flexible with others, in order to get along], so I have written you this “Discussion of the Cart” for you to read closely. This discussion of the cart will benefit you in your dealings with the world. 善人少，不善人多，故愛足下者少，而害足下者多。吾固欲其方其中，圓其 外，今為足下作說車，可詳觀之。車之說，其有益乎行於世也。66

These gestures also underscore the reciprocity expected in the back-and-forth of letters, and the assumption that the exchange of knowledge about one another would continue, even if that exchange required sometimes painful self-examination. As the examples demonstrate, affection expressed in letters was ultimately only one component of the larger structure of the social relationships, but it was essential to creating the climate of trust and security that writers invoked so often. References to trust also reveal the potential burden of self-disclosure on the recipient, in that he then had to decide how best to handle the knowledge or implied request contained in the letter, and whether or not to share the letter with those who might be of help. This issue particularly affected men writing from political disgrace or poor circumstances, as is the case for twelve of the letters. The rhetorically bounded space of the collegial letter allowed for direct speech, and the exchange of letters provided both the pretext and the stage on which a writer could unburden himself. As Han Yu concludes in his 65   Han Yu quanji jiaozhu, 3:1261; Han Yu guwen jiaozhu, 1:691. 66   Liu Zongyuan ji, 2:848.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1321

letter to Cui Lizhi, “The scholar indeed trusts in those who know him; if it were not for you, I would not have had reason to utter my wild [intemperate] words.” 士固信於知己，微足下無以發吾之狂言.67 In Bai Juyi’s long defense of his actions to Yang Yuqing, Bai proclaims his trust and intimacy with Yang: “Shigao! As for these words of mine, I do not utter them to other people—I only utter them to you, Shigao. You, Shigao, are one who knows me; how can there be shame between us? If I were to feel shame before you, then I certainly would not utter these words.” 師臯! 僕之是言，不發於他人，獨發於師臯。師臯知 我者，豈有愧於其間哉? 苟有愧於師臯，固是言不發矣. Since just prior to this statement Bai Juyi had regaled Yang Yuqing with a list of high-ranking colleagues to whom he had turned for help, we are not meant to take this as a serious request for secrecy. Rather, Bai Juyi is using the rhetorical framework of speaking “privately” to the trusted friend—one before whom he need not feel shame, despite his public humiliation—as the condition for a full disclosure of the events he had witnessed and the reasons for his actions, which he hopes will then be more widely circulated in his defense. In the expressions of affection to their friends and colleagues, these midTang letters appear to work against medieval conventions that exaggerated feelings of longing and affection and conveyed them in fixed phrases and terms. However, the situations in which the letters were composed made expressing affection or any positive feeling toward the recipient a delicate matter: in the case of Liu Zongyuan’s letters, for example, all four were written during Liu’s years of demotion to Yongzhou. Although only Li Jian and Xiao Mian would have been in a position to aid Liu politically, Liu Zongyuan had no way to ensure a “safe” circulation of his writing. The same is true of Bai Juyi’s letters, which were all composed during a two-year period at the start of his five-year demotion. Han Yu, in his letters to Cui Qun, Li Ao, and Meng Jiao, was still struggling to find supporters and an official post in the ranks. The challenge for these men when writing to friends and former colleagues, therefore, was to avoid the appearance of flattery, hypocrisy, or begging while still defending themselves and reassuring the recipients of a continued bond. Being reserved in showing affection was thus critical; but at the same time, since they were writing from a place of political, social, and even economic vulnerability (none except Liu Zongyuan had a substantial income), these writers had few resources other than personal feeling and literary talent to draw on. Ensuring that their letters of self-expression also reinforced their bonds with friends and colleagues or displayed their sincere feeling to others, at a time in Tang 67   Han Yu quanji jiaozhu, 3:1263; Han Yu guwen jiaozhu, 1:704.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1322

Shields

political history when those assets were essential for survival, was at the very least a wise strategy. At the core of this research lies the very question Bai Juyi posed of Yuan Zhen in his letters: “do you know my heart?” 知我心哉. How did mid-Tang writers convey their “hearts”—and how can we be sure we understand them? In these mid-Tang collegial letters, we find ambitious and diverse goals, including self-revelation, intellectual debate, political positioning, and social role negotiation. The writers’ awareness of the larger audience for their letters shapes their literary craft at every level, and certainly led to the preservation of the texts in their collected works. If, as I have suggested, medieval models for letters between friends and colleagues were focused on ritual occasions and constrained by formulaic expressions of sentiment, we can better see how difficult mid-Tang writers might have found the task of expressing emotions with sincerity and clarity in these epistolary contexts, and how concerned they might have been about being misread. Their goal was to convince their readers—both the intended recipient and the wider group of readers beyond him—of their authenticity, their originality, and the necessity of their communication. By inscribing feeling indirectly at many levels, from rhetorical techniques down to a careful choice of emotion words and question particles, these midTang writers attempted to make their inner states clear without risking either hypocrisy or coldness. They did so by referring to specific people and events from their lives, including memories of past encounters or references to certain actions and words they found objectionable, to reinvigorate the sense of immediate understanding that letters sent over long distances had to convey. They used a flexible, vigorous prose style that varied significantly within the letters and from letter to letter in order to signal originality and to surprise the reader, who could have no way of predicting from the form or style of the text what information or revelation would appear at any given moment. And they heightened the communicative urgency of the letters by engaging the recipient constantly in the exchange with questions, reassurances, critiques, and exclamations. As the many letters here that are titled “reply” and “second reply” remind us, these mid-Tang letters were part of exchanges; perhaps even more than the letters expressing anger, the letters that suggest affection give us the sense of hearing just one side of an ongoing conversation, one that continued in the many texts now lost to us. The narrowness of this set of letters may, in the end, only tell us about these mid-Tang letters and writers. It is to be hoped that future research on emotions in Song and later texts will allow us to compare mid-Tang letters to a broader set of exchanges, or even to emotions expressed

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1323

in other literary forms, and show us the ways that anger and affection could be negotiated over time and within a broader range of social relationships. Bibliography Bai Juyi ji jianjiao 白居易集箋校. Edited by Zhu Jincheng 朱金城. Shanghai: Guji chubanshe, 2003. Barton, Richard E. “ ‘Zealous Anger’ and the Renegotiation of Aristocratic Relationships in Eleventh- and Twelfth-Century France.” In Anger’s Past: The Social Uses of an Emotion in the Middle Ages, edited by Barbara H. Rosenwein, 153–70. Ithaca: Cornell University Press, 1998. Chen, Jo-shui. Liu Tsung-yuan and Intellectual Change in T’ang China, 773–819. Cambridge: Cambridge University Press, 1992. Ding Hongwu 丁宏武. “Li Ling ‘Da Su Wu shu’ zhenwei zai tantao” 李陵 “答蘇武書” 真 偽再探討. Ningxia daxue xuebao 34.2 (2012): 47–53. Ditter, Alexei. “Genre and the Transformation of Writing in Tang Dynasty China.” PhD diss., Princeton University, 2009. Ebrey, Patricia. “T’ang Guides to Verbal Etiquette.” HJAS 45 (1985): 581–613. Egan, Ronald C. “Su Shih’s ‘Notes’ as a Historical and Literary Source.” HJAS 50 (1990): 561–88. Eifring, Halvor, ed. Love and Emotions in Traditional Chinese Literature. Leiden: Brill, 2004. Fu Xuancong 傅璇璁. Tang dai keju yu wenxue 唐代科舉與文學. Taipei: Wenshizhe, 1994. Han Yu guwen jiaozhu huiji 韓愈古文校注彙輯. Edited by Luo Liantian 羅聯添. Taipei: Guoli bianyi guan, 2003. Han Yu quanji jiaozhu 韓愈全集校注. Edited by Qu Shouyuan 屈守元 and Chang Sichun 常思春. Chengdu: Sichuan daxue chubanshe, 1996. Huang Zhengjian 黃正建. Zhong Wan Tang shehui yu zhengzhi yanjiu 中晚唐社會與政 治研究. Beijing: Zhongguo shehui kexue chubanshe, 2006. Jansen, Thomas. “The Art of Severing Relationships (juejiao) in Early Medieval Culture.” JAOS 126 (2006): 347–65. Jin Chuandao 金傳道. “ ‘Shuyi’ neirong bianzheng” 書儀內容辨正. Nei Menggu daxue xuebao 42:5 (2010): 127–30. Jiu Tang shu 舊唐書. Beijing: Zhonghua shuju, 1975. Knechtges, David R. “Culling the Weeds and Selecting Fine Blossoms: The Anthology in Early Medieval China.” In Culture and Power in the Reconstitution of the Chinese Realm, 200–600, edited by Scott Pearce, Audrey Spiro, and Patricia Buckley Ebrey, 200–241, 322–34. Cambridge, Mass.: Harvard University Asia Center, 2001.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1324

Shields

Liu Yuxi ji jianzheng 劉禹錫集箋證. Edited by Qu Tuiyuan 瞿蛻園. Shanghai: Shanghai guji chubanshe, 1989. Liu Zongyuan ji 柳宗元集. Beijing: Zhonghua shuju, 2006. McMullen, David. State and Scholars in T’ang China. Cambridge: Cambridge University Press, 1988. Moore, Oliver. Rituals of Recruitment in Tang China: Reading an Annual Programme in the Collected Statements by Wang Dingbao (870–940). Leiden: Brill, 2004. Harbsmeier, Christoph. Science and Civilisation in China, vol. 7: The Social Background, part 1: Language and Logic. Cambridge: Cambridge University Press, 1998. Nugent, Christopher M. B. Manifest in Words, Written on Paper: Producing and Circulating Poetry in Tang Dynasty China. Cambridge, Mass.: Harvard University Asia Center, 2010. Owen, Stephen. The Making of Early Chinese Classical Poetry. Cambridge, Mass.: Harvard University Asia Center, 2006. Owen, Stephen. The End of the Chinese ‘Middle Ages’: Essays in Mid-Tang Literary Culture. Stanford: Stanford University Press, 1996. Pattinson, David. “Privacy and Letter Writing in Han and Six Dynasties China.” In Chinese Concepts of Privacy, edited by Bonnie McDougall and Anders Hansson, 97–118. Leiden: Brill, 2002. Pollock, Linda A. “Anger and the Negotiation of Relationships in Early Modern England.” The Historical Journal 47 (2004): 567–90. Quan Tang wen 全唐文. Compiled by Dong Gao 董誥 (Qing). Rpt. Beijing: Zhonghua shuju, 1983. Reddy, William. The Navigation of Feeling: A Framework for the History of Emotions. Cambridge: Cambridge University Press, 2001. Santangelo, Paolo, ed. Love, Hatred and Other Passions: Questions and Themes on Emotions in Chinese Civilization. Leiden: Brill, 2006. Schneider, Gary. The Culture of Epistolarity: Vernacular Letters and Letter-Writing in Early Modern England, 1500–1700. Newark, n. j.: University of Delaware Press, 2005. Shields, Anna M. “Defining Experience: The ‘Poems of Seductive Allure’ (yanshi) of the Mid-Tang Poet Yuan Zhen (779–831).” JAOS 122 (2002): 61–78. Thrailkill, Jane. Affecting Fictions: Mind, Body, and Emotion in American Literary Realism. Cambridge, Mass.: Harvard University Press, 2007. Van Norden, Bryan, trans. Mengzi. Indianapolis, Ind.: Hackett Publishing, 2008. Wan Man 萬曼. Tang ji xu lu 唐集敘錄. Beijing: Zhonghua shuju, 1980. Whitaker, K. P. K. “Some Notes on the Authorship of the Li Ling/Su Wu Letters.” BSOAS 15 (1953): 113–37, 566–87. Wierzbicka, Anna. Emotions Across Languages and Cultures: Diversity and Universals. Cambridge: Cambridge University Press, 1999. Wu Liyu 吳麗娛, Tang li zhiyi 唐禮摭遺. Beijing: Shangwu yinshu, 2002.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Inscription Of Emotion In Mid-tang Collegial Letters

1325

Wen xuan 文選. Shanghai: Shanghai guji chubanshe, 1986. Zhao Heping 趙和平. Dunhuang shuyi yanjiu 敦煌書儀研究. Shanghai: Shanghai guji chubanshe, 2011. Zhao Heping and Zhou Yiliang 周一良. Tang Wudai shuyi yanjiu 唐五代書儀研究. Beijing: Zhongguo shehui kexue chubanshe, 1995. Zhao Shugong 趙樹功. Zhongguo chidu wenxue shi 中國尺牘文學史. Shijiazhuang: Hebei renmin chubanshe, 1999. Zhuangzi jiaoquan 莊子校詮. Edited by Wang Shumin 王叔岷. Taipei: Academia Sinica, 1988.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘The Story of Ying-ying’ James R. Hightower During the T’ang dynasty the Tale of the Marvelous (ch’uan ch’i) became an identifiable literary form that attracted the attention of the best writers of the period. The tales were collected in anthologies with titles like Record of the Occult and Bizarre 玄怪錄, Collection of Strange Accounts 集異記, and Transmitted Marvels 傳奇,1 this last having become the accepted generic label for all these stories. All these titles emphasize the unusual nature of the occurrence which provides the plot; yet while few of the stories are free of supernatural or incredible episodes, they are not for the most part presented as tall tales or ghost stories that deliberately challenge belief. The author appears anxious rather to persuade us that we are reading about an actual occurrence, and he often makes an effort to establish the credentials of his tale by anchoring it in the experience of a narrator. It is rarely told in the first person throughout,2 but at the end of the story the narrator frequently puts in an appearance to make a generalization or to moralize for the author or in the author’s own name: “Alas, that a singing girl could be capable of virtuous conduct not even surpassed by the exemplary women of antiquity! Who can fail to sigh in admiration?” He may conclude by giving the circumstances under which he heard the story and how he came to write it down: My great-uncle was at one time Governor of Chin-chou, then transferred to the Ministry of Finance, before becoming Transport Officer. In all three posts he succeeded this young man, and so came to have some knowledge of his affairs. During the Chen-yüan period (785–805), I was discussing with [Li] Kung-tso of Lung-hsi the quality of loyalty in women, and so Source: “Yüan Chen and ‘The Story of Ying-ying’,” Harvard Journal of Asiatic Studies 33 (1973): 90–123. 1  Collections of stories under these titles are attributed respectively to Niu Seng-ju 牛僧孺, Hsüeh Yung-jo 薛用弱, and P’ei Hsing 裴鉶 in the Essay on Bibliography in the T’ang shu and Sung shih. Over a hundred stories in the T’ai-p’ing kuang-chi were taken from these collections. (Abbreviated tpkc; references are to the Peiping Wen-yu-t’ang shu-fang ed. of 1934.). 2  Notable exceptions are “Dalliance in the Abode of Immortals” 遊仙窟 by Chang Cho 張 鷟, Shen Ya-chih’s 沈亞之 “Dream of Ch’in” 秦夢記, and “The Record of a Trip to Chou and Ch’in” 周秦行記. Problems of authorship are connected with the first and last of these; see below p. 1357.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_042

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1327

came to recount the story of [the Lady of] Chien-kuo. Kung-tso listened appreciatively and attentively and asked me to write it down. So I have taken my brush and wetted it in ink to preserve the story in its rough outline. The time is autumn, the eighth month of the year i-hai (795). Po Hsing-chien, of T’ai-yüan.3 Here, as the author lends the narrator his name, his voice ceases to be impersonal and anonymous. This formal commitment to the veracity of the story sometimes involves calling others to witness, listing the names (and ranks) of those who were present at the telling, or even recording an actual meeting with the protagonist of the story, as in “The Governor of South-branch.”4 “[I, Li] Kung-tso, in the eighth month, autumn, the eighteenth year of Chen-yüan (802), on the way from Wu to Lo-yang, stopped over at Huai-p’u, where I chanced to meet Ch’un-yü Fen. I questioned him closely about the evidence for what had occurred, and on going into everything carefully, found that it was all confirmed. Then I wrote it up into a story for all who might be interested.” This last step involves identifying a fictional character with someone encountered in real life who has an existence independent of the story. The reverse process, using the name of a known contemporary for a character in a story, is another way of providing a link between the world of the story and the real world, a link that can be multiplied, as when Li I is made the protagonist of “The Story of Huo Hsiao-yü”5 and his friend Wei Hsia-ch’ing is introduced as a minor character; one is less ready to dismiss as a fabrication a story which includes two people one knows to have lived. For all that it includes the heroine’s prophetic dream and the appearance of her ghost to the contrite Li I, “The Story of Huo Hsiao-yü” is a credible effort at realistic writing. Li I’s reputation for jealousy is recorded in his biography: “As a young man I was insanely jealous. He exercised extremely strict surveillance over his wife and concubines. The gossip about his scattering ashes and bolting doors was widespread, and contemporaries referred to jealousy as ‘Li I’s failing.’ ”6 Chiang Fang may have invented his story, or he could have been giving literary form to a current piece of gossip about a well-known figure. It is

3  “The Story of Li Wa,” tpkc 484.7b; translated by C. C. Wang, Traditional Chinese Tales, pp. 60–74. 4  t pkc 475.7a, where the story is entitled “Ch’un-yü Fen” 淳于棼; translated by Gladys Wang and Wang Hsien-yi, The Dragon King’s Daughter, pp. 44–56. 5  t pkc 487; translated by C. C. Wang, op. cit., pp. 48–59. 6  T’ang shu 137.13a (T’ung-wen ed.).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1328

Hightower

a carefully constructed fiction, presented as a straightforward job of reporting, with emphasis on character and psychological motivation. Two other T’ang love stories dispense altogether with supernatural apparatus, and both make a claim that is lacking in Huo Hsiao-yü, that the author/ narrator had access to special knowledge of the circumstances and so can vouch for their veracity. “The Story of Li Wa” is supposed to have happened some fifty years before the author was writing, and the detailed knowledge of the events is said to come from the narrator’s great-uncle. The family name of the hero is ostentatiously withheld, a discretion which invites us to believe it all must really have happened. The events are unusual enough, some of them, but nothing outside the range of normal human experience. “The Story of Ying-ying” puts even less strain on credulity, as far as the events it recounts are concerned. Even in societies where the sexes are more strictly segregated than in T’ang China, it must have happened many times, if not commonly, that young people meet, fall in love, and after a brief affair go their separate ways, whether parted by circumstance or deliberate choice of one of the partners. That is not to say that the liaison between Chang and a girl of good family was not unconventional, though this aspect receives surprisingly little emphasis in the narrative. It is a Tale of the Marvelous only by convention or by courtesy; of all the T’ang stories it seems to justify its claim to be a story of real life, and we are supplied a surfeit of proofs in the form of references to persons, places, and events known to history as well as poems and testimony by friends of the protagonist. The introduction of the author’s name into the story, as the narrator Yüan Chen, an especially close friend of the hero, insists on the fact, and for a historically minded reader in search of the identity of the characters, it was not unreasonable to read the story as a piece of autobiography, simply identifying Yüan Chen with the hero, Chang, and treating the narrator as a transparent device designed to protect the author from his own indiscretion. The arguments for the identification of Chang with Yüan Chen are keyed to details of the story, of which I should like to present a translation before discussing their validity.7

7  t pkc 488; T’ang-tai ts’ung-shu 120. I have taken a few variant readings from the latter as noted. There are several translations of Ying-ying chuan; that by C. C. Wang (op. cit., pp. 75–86) in particular is accurate and readable. However, like all the others it omits the long poem by Yüan Chen, an omission amply justified on aesthetic grounds, but its place in the story is of some importance to my discussion. There are also some details of phrasing which are significant and which call for comment.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’



1329

The Story of Ying-ying

During the Chen-yüan period (785–805) there lived a young man named Chang. He was agreeable and refined, and good-looking, but firm and selfcontained, and capable of no improper act. When his companions included him in one of their parties, the others could all be brawling as though they would never get enough, but Chang would just watch tolerantly without ever taking part. In this way he had got to be twenty-two years old without ever having had relations with a woman. When asked by his friends, he explained, “Teng T’u-tzu8 was no lover, but a lecher. I am the true lover—I just never happened to meet the right girl. How do I know that? It’s because all things of outstanding beauty never fail to make a permanent impression on me. That shows I am not without feelings.” His friends took note of9 what he said. Not long afterward Chang was traveling in P’u,10 where he lodged a few miles east of the city in a monastery called the Temple of Universal Salvation.11 It happened that a widowed Mrs. Ts’ui had also stopped there on her way back to Ch’ang-an. She had been born a Cheng; Chang’s mother had been a Cheng, and when they worked out their common ancestry, this Mrs. Ts’ui turned out to be a rather distant aunt on his mother’s side. This year Hun Chen12 died in P’u, and the eunuch Ting Wen-ya proved unpopular with the troops, who took advantage of the mourning period to mutiny. They plundered the citizens of P’u, and Mrs. Ts’ui, in a strange place with all her slaves and chattels, was terrified, having no one to turn to. Before the mutiny Chang had made friends with some of the officers in P’u, and now he requested a detachment of soldiers to protect the Ts’ui family. As a result all escaped harm. In about ten days the Imperial Commissioner of Enquiry

8  The character ridiculed in the Sung Yü fu, “The Lechery of Master Teng-t’u,” Wen hsüan 19.7a–8a (sppy ed.), partial translation by Arthur Waley in One Hundred and Seventy Chinese Poems, pp. 43–44. 9  For 識 there is a variant 哂 “scoffed at him.” 10  P’u-chou is Yung-chi hsien 永濟, Shansi, about 120 km. east-northeast of Ch’ang-an. 11   The Lives of Eminent Monks, Second Collection 高僧傳二集, 39.1112 (Taiwan: Yin-chingch’u, 1960) mentions a P’u-chiu ssu 普救寺 in P’u-chou. Ch’en Yin-k’o, “Tu Ying-ying chuan,” reprinted in 元白詩箋證稿, 1955, p. 109. 12  Hun Chen 渾⊙ (for this graph, see Morohashi No. 21105) (Governor-General of Chiangchou) died in Ho-chung in the twelfth month, 799 (Tzu-chih t’ung-chien 7585.18). He has a biography in T’ang shu 134.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1330

Hightower

Tu Ch’üeh13 came with full power from the throne and restored order among the troops. Out of gratitude to Chang for the favor he had done them, Mrs. Ts’ui invited him to a banquet in the central hall. She addressed him, “Your widowed aunt with her helpless children would never have been able to escape alive from these rioting soldiers. It is no ordinary favor you have done us; it is rather as though you had given my son and daughter their lives, and I want to introduce them to you as their elder brother so that they can express their thanks.” She summoned her son Huan-lang, a very attractive child of ten or so. Then she called her daughter, “Come out and pay your respects to your brother, who saved your life.” There was a delay; then word was brought that she was indisposed and asked to be excused. Her mother exclaimed in anger, “Your brother Chang saved your life. You would be a slave if it were not for him—how can you give yourself airs?” After a while she appeared, wearing an everyday dress and no makeup on her smooth face, except for a remaining spot of rouge. Her hair coils straggled down to touch her eyebrows. Her beauty was extraordinary, so radiant it took the breath away. Startled, Chang made her a deep bow as she sat down beside her mother. Because she had been forced to come out against her will, she looked angrily straight ahead, as though unable to endure the company. Chang asked her age. Mrs. Ts’ui said, “From the seventh month of the fifth year of the reigning emperor to the present twenty-first year,14 it is just sixteen years.” Chang tried to make conversation with her, but she would not respond, and he had to leave after the meal was over. From this time on Chang was infatuated but had no way to make his feeling known to her. She had a maid named Hung-niang with whom Chang had managed to exchange greetings several times, and finally he took the occasion to tell her how he felt. Not surprisingly, the maid was alarmed and fled in confusion. Chang was sorry he had said anything, and when she returned the next day he made shamefaced apologies without repeating his request. The maid said, “Sir, what you said is something I would not dare repeat to my mistress or let anyone else know about it. But you know very well who Miss Ts’ui’s relatives are; why don’t you ask for her hand in marriage, as you are entitled to do because of the favor you did them?” Chang said, “From my earliest years I have never been one to make any improper connections. Whenever I have found myself in the company of young 13  Tu Ch’üeh 杜確, appointed Prefect of T’ung-chou and Imperial Commissioner of Chiang-chou. 14   貞元庚辰 is the year 800.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1331

Yüan Chen and ‘ The Story of Ying-ying ’

women, I would not even look at them, and it never occurred to me that I would be trapped in any such way. But the other day at the dinner I was hardly able to control myself, and in the days since, I walk without knowing where I am going and eat without hunger—I am afraid I cannot last out another day. If I were to go through a regular matchmaker, taking three months and more for the exchange of betrothal presents and names and birthdates15—you might just as well look for me among the dried fish in the shop.16 Can’t you tell me what to do?” The maid replied, “Miss Ts’ui is so very strict that not even her elders could suggest anything improper to her. It would be hard for someone in my position to say such a thing. But I have noticed she writes a lot. She is always reciting poetry to herself and is moved by it for a long time after. You might see if you can seduce her with a love poem. That is the only way I can think of.” Chang was delighted and on the spot composed two stanzas of Spring Verses which he handed over to her. That evening Hung-niang came back with a note on colored paper for him, saying, “By Miss Ts’ui’s instructions.” The title of her poem was “Bright Moon on the Night of the Fifteenth”: I await the moon in the western chamber Where the breeze comes through the half-opened door. Sweeping the wall the flower shadows move: I imagine it is my lover who comes. Chang understood the message: that day was the fourteenth of the second month, and an apricot tree was in bloom next to the wall east of the Ts’ui’s courtyard. It would be possible to climb it. On the night of the fifteenth Chang used the tree as a ladder to get over the wall. When he came to the western chamber, the door was ajar. Inside, Hungniang was asleep on a bed. He awakened her, and she asked, frightened, “How did you get here?” “Miss Ts’ui’s letter told me to come,” he said, not quite accurately. “You go tell her I am here.” In a minute Hung-niang was back, “She’s coming! She’s coming!” Chang was both happy and nervous, convinced that success was his. Then Miss Ts’ui appeared in formal dress, with a serious face, and began to upbraid him, “You did us a great kindness when you saved our lives, and that is why my mother entrusted my young brother and myself to you. Why then did 15  To determine an astrologically suitable date for a wedding. 16  An allusion to the parable of help that comes too late in Chuang-tzu 9.2b–3a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1332

Hightower

you get my silly maid to bring me that filthy poem? You began by doing a good deed in preserving me from the hands of ravishers, and you end by seeking to ravish me. You substitute seduction for rape—is there any great difference? My first impulse was to keep quiet about it, but that would have been to condone your wrongdoing, and not right. If I told my mother, it would amount to ingratitude, and the consequences would be unfortunate. I thought of having a servant convey my disapproval, but feared she would not get it right. Then I thought of writing a short message to state my case, but was afraid it would only put you on your guard. So finally I composed those vulgar lines to make sure you would come here. It was an improper thing to do, and of course I feel ashamed. But I hope that you will keep within the bounds of decency and commit no outrage.” As she finished speaking, she turned on her heel and left him. For some time Chang stood, dumbfounded. Then he went back over the wall to his quarters, all hope gone. A few nights later Chang was sleeping alone by the veranda when someone shook him awake. Startled, he rose up, to see Hung-niang standing there, a coverlet and pillow in her arms. She patted him and said, “She is coming! She is coming! Why are you sleeping?” And she spread the quilt and put the pillow beside his. As she left, Chang sat up straight and rubbed his eyes. For some time it seemed as though he were still dreaming, but nonetheless he waited dutifully. Then there was Hung-niang again, with Miss Ts’ui leaning on her arm. She was shy and yielding, and appeared almost not to have the strength to move her limbs. The contrast with her stiff formality at their last encounter was complete. This evening was the night of the eighteenth, and the slanting rays of the moon cast a soft light over half the bed. Chang felt a kind of floating lightness and wondered whether this was an immortal who visited him, not someone from the world of men. After a while the temple bell sounded. Daybreak was near. As Hung-niang urged her to leave, she wept softly and clung to him. Hung-niang helped her up, and they left. The whole time she had not spoken a single word. With the first light of dawn Chang got up, wondering was it a dream? But the perfume still lingered, and as it got lighter he could see on his arm traces of her makeup and the teardrops sparkling still on the mat. For some ten days afterwards there was no word from her. Chang composed a poem of sixty lines on “An Encounter with an Immortal” which he had not yet completed when Hung-niang happened by, and he gave it to her for her mistress. After that she let him see her again, and for nearly a month he would join her in what her poem had called the “western chamber,” slipping out at dawn and returning stealthily at night. Chang once asked what her mother thought

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1333

about the situation. She said, “She knows there is nothing she can do about it, and so she hopes you will regularize things.”17 Before long Chang was about to go to Ch’ang-an, and he let her know his intentions in a poem.18 Miss Ts’ui made no objections at all, but the look of pain on her face was very touching. On the eve of his departure he was unable to see her again. Then Chang went off to the west. A few months later he again made a trip to P’u and stayed several months with Miss Ts’ui. She was a very good calligrapher and wrote poetry, but for all that he kept begging to see her work, she would never show it. Chang wrote poems for her, challenging her to match them, but she paid them little attention. The thing that made her unusual was that, while she excelled in the arts, she always acted as though she were ignorant, and although she was quick and clever in speaking, she would seldom indulge in repartee. She loved Chang very much, but would never say so in words. At the time she was subject to moods of profound melancholy, but she never let on. She seldom showed on her face the emotions she felt. On one occasion she was playing her cither alone at night. She did not know Chang was listening, and the music was full of sadness. As soon as he spoke, she stopped and would play no more. This made him all the more infatuated with her. Some time later Chang had to go west again for the scheduled examinations. It was the eve of his departure, and though he had said nothing about what it involved, he sat sighing unhappily at her side. Miss Ts’ui had guessed that he was going to leave for good. Her manner was respectful, but she spoke deliberately and in a low voice, “To seduce someone and then abandon her is perfectly natural, and it would be presumptuous of me to resent it. It would be an act of charity on your part if, having first seduced me, you were to go through with it and fulfill your oath of lifelong devotion. But in either case, what is there to be so upset about in this trip? However, I see you are not happy and I have no way to cheer you up. You have praised my cither playing, and in the past I have 17  I have taken the T’ang-tai ts’ung-shu variant 知 for 我 (according to a note in the Peking edition of tpkc, p. 4014, the variant is also attested in a Ming ms. of tpkc). With wo, the natural way of reading the speech would be, “I (Ying-ying) am unable to do anything (to make her resigned to the situation).” Then the rest would have to mean “and so I hope you will marry me.” If the speech ends with the first phrase, as it is punctuated by Wang P’i-chiang 王辟疆 in T’ang-jen hsiao-shuo, p. 137, the conclusion would almost have to be a statement about Chang’s intentions: “Therefore he wanted to do the right thing.” This is a non sequitur and does not fit into the story. For chiu ch’eng “bring to fulfillment” in the sense of “fulfill the rites of marriage,” see the earlier T’ang story Yu hsien k’u, p. 7: 就成大 禮 (Shanghai: Ku-tien wen-hsüeh, 1955). 18  For 情 read 詩 with T’ang-tai ts’ung-shu 120.4b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1334

Hightower

been embarrassed to play for you. Now that you are going away, I shall do what you so often requested.” She had them prepare her cither and started to play the prelude to the “Rainbow Robe and Feather Skirt.” After a few notes, her playing grew wild with grief until the piece was no longer recognizable. Everyone was reduced to tears, and Miss Ts’ui abruptly stopped playing, put down the cither, and ran back to her mother’s room with the tears streaming down her face. She did not come back. Next morning Chang went away. The following year he stayed on in the capital, having failed the examinations. He wrote a letter to Miss Ts’ui to reassure her, and her reply read roughly as follows: I have read your letter with its message of consolation, and it filled my childish heart with mingled grief and joy. In addition you sent me a box of ornaments to adorn my hair and a stick of pomade to make my lips smooth. It was most kind of you; but for whom am I to make myself attractive? As I look at these presents my breast is filled with sorrow. Your letter said that you will stay on in the capital to pursue your studies, and of course you need quiet and the facilities there to make progress. Still it is hard on the person left alone in this far-off place. But such is my fate, and I should not complain. Since last fall I have been listless and without hope. In company I can force myself to talk and smile, but come evening I always shed tears in the solitude of my own room. Even in my sleep I often sob, yearning for the absent one. Or I am in your arms for a moment as it used to be, but before the secret meeting is done I am awake and heartbroken. The bed seems still warm beside me, but the one I love is far away. Since you said goodbye the new year has come. In the spring Ch’ang-an is a city of pleasure with chances for love everywhere. I am truly fortunate that you have not forgotten me and that your affection is not worn out. Loving you as I do, I have no way of repaying you, except to be true to our vow of lifelong fidelity. Our first meeting was at the banquet, as cousins. Then you persuaded my maid to inform me of your love; and I was unable to keep my childish heart firm. You made advances, like that other poet, Ssu-ma Hsiang-ju;19

19  An allusion to the story of Ssu-ma Hsiang-ju, who enticed the young widow Cho Wenchün to elope by his cither playing (Shih chi 117.5, Takigawa ed.).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1335

I failed to repulse them as the girl did who threw her shuttle.20 When I offered myself in your bed, you treated me with the greatest kindness, and I supposed, in my innocence, that I could always depend on you. How could I have foreseen that our encounter21 could not possibly lead to something definite,22 that having disgraced myself by coming to you, there was no further chance of serving you openly as a wife? To the end of my days this will be a lasting regret—I must hide my sighs and be silent. If you, out of kindness, would condescend to fulfill my selfish wish, though it came on my dying day it would seem a new lease on life. But if, as a man of the world, you curtail your feelings, sacrificing the lesser to the more important, and look on this connection as shameful, so that your solemn vow becomes dispensable, still my true love will not vanish though my bones decay and my frame dissolve: in wind and dew it will seek out the ground you walk on. My love in life and death is told in this. I weep as I write, for feelings I cannot express. Take care of yourself, a thousand times over, take care of your dear self. This bracelet of jade is something I wore as a child: I send it to serve as a gentleman’s belt pendant. Like jade may you be invariably firm and tender; like a bracelet may there be no break between what came before and what is to follow. Here are also a skein23 of tangled thread and a tea roller of mottled bamboo. These things have no intrinsic value, but they are to signify that I want you to be true as jade, and your love to endure unbroken as a bracelet. The spots on the bamboo are like the marks of my tears,24 and my unhappy thoughts are as tangled as the thread: these objects are symbols of my feelings and tokens for all time of my love.25 Our hearts are close, though our bodies are far apart and there is no time I can expect to see you. But where the hidden desires are strong enough, there will be a meeting of spirits. Take care of yourself, a thousand times over. The springtime wind is often chill; eat well for your health’s sake. Be circumspect and careful, and do not think too often of my unworthy person.

20  It was Miss Kao who repulsed Hsieh K’un’s advances by throwing her shuttle in his face. He lost two teeth (Chin shu 69.31b, Chin shu kou-chu, I-wen ed.). 21  The phrase 旣見君子 is common in the Shih ching, e. g., No. 10/2. 22   定情 is used to mean a betrothal present; here it is a euphemism for marriage. 23  Emend 絢 “thread” to 絇. 24  Alluding to the legend of the two wives of Shun, who stained the bamboo with their tears. 25   永以爲好也 is another Shih ching line, No. 64/1.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1336

Hightower

Chang showed her letter to his friends, and in this way word of the affair got around. One of them, Yang Chü-yuan, a skillful poet, wrote a quatrain on “Young Miss Ts’ui”:26 For clear purity jade cannot equal his complexion, On the iris in the inner court snow begins to melt. A romantic young man filled with thoughts of love. A letter from the Hsiao girl, brokenhearted. Yüan Chen of Ho-nan wrote a continuation of Chang’s poem, “Encounter with an Immortal,” also in thirty couplets:27

10

Faint moonbeams pierce the curtained window Fireflies glimmer across the blue sky. The far horizon begins now to pale Dwarf trees gradually turn darker green. A dragon song28 crosses the court bamboo A phoenix air brushes the well-side tree.29 The silken robe trails through the thin mist The pendant circles tinkle in the light breeze. The accredited envoy accompanies Hsi-wang-mu30 From the cloud’s center comes Jade Boy.31 Late at night everyone is quiet

26  The poem is included in Ch’uan T’ang shih 333.3737 (Peking: Chung-hua shu-chü, 1960 ed., pagination consecutive; abbreviated cts), with Yang Chü-yüan’s other poems. 27  The text of this poem is not in Yüan Chen’s Ch’ang-ch’ing chi, but it is one included in the selection by Su Chung-hsiang 蘇仲翔, 兀白詩選 (Shanghai: Ku-tien wen-hsüeh, 1957), with a few notes. 28   龍吹: someone playing a flute with a dragon design? Su Chung-hsiang says “dragon” means old bamboo 指竹之古者. chapter 26 of Hung lou meng the words 龍吟 are used for the rustling of bamboo; see Jen-min wen-hsüeh ed. (1957), p. 269, note 1. 29   鸞歌: the luan bird is singing? Su says, “The leaves of the t’ung tree are like the wings of the luan.” The well-side tree is 井桐, the wu-t’ung by the well. 30   金母 is the Western Mother, metal representing that direction, hence a euphemism for 西王母 (Su). Does she stand for Ying-ying’s mother? 31   玉童 can be 仙童; it could represent Ying-ying’s brother. I do not know how to take 雲心; from the parallelism it should be a name: “Cloud-heart supports Jade Boy”—Ying ying leaning on Hung-niang’s arm?

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1337

Yüan Chen and ‘ The Story of Ying-ying ’

20

30

At daybreak the rain drizzles.32 Pearl radiance shines on her decorated sandals Flower glow shows off the embroidered skirt.33 Jasper hairpin: a walking colored phoenix Gauze shawl: embracing vermilion rainbow. She says she comes from Jasper Flower Bank And is going to pay court at Green Jade Palace. On an outing north of Lo-yang’s34 wall By chance he came to the house east of Sung Yü’s.35 His dalliance she rejects a bit at first, But her yielding love already is disclosed. Lowered locks put in motion cicada shadows36 Returning steps raise jade dust.37 Her face turns to let flow flower snow As she climbs into bed, silk covers38 in her arms. Love birds in a neck-entwining dance Kingfishers in a conjugal cage. Eyebrows, out of shyness, contracted Lip rouge, from the warmth, melted. Her breath is pure: fragrance of orchid buds Her skin is smooth: richness of jade flesh. No strength, too limp to lift a wrist Many charms, she likes to draw herself together. Sweat runs: pearls drop by drop, Hair in disorder: black luxuriance. Just as they rejoice in the meeting of a lifetime They suddenly hear the night is over. There is no time for lingering

32  The couplet appears to be descriptive, but there is no literal counterpart to this line in the story. I prefer not to read any other meaning into it. 33  Emend 龍 to 襲. 34  A reference to the Goddess of the Lo River? 35  In “The Lechery of Master Teng-t’u,” Sung Yü tells about the beautiful girl next door to the east who climbed up on the wall to flirt with him (Wen hsüan 19.7a–b). 36  Referring to her hairdo in the cicada style 蟬鬢. 37  It is her steps that confer value (hence the epithet 玉) to the dust. 38  I have translated 綺叢 from the context of the story (though it was Hung-niang who carried the bedding) and the occurrence of the phrase in a Fan Ch’eng-ta poem (from pwyf) where it parallels 瑤席.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1338 40

50

60

Hightower

It is hard to give up the wish to embrace. Her comely face shows the sorrow she feels With fragrant words they swear eternal love. She gives him a bracelet to plight their troth39 He ties a lovers’ knot as sign their hearts are one. Tear-borne powder runs before the clear40 mirror Around the flickering lamp are nighttime insects.41 Moonlight is still softly shining. As the rising sun gradually dawns. Riding on a wild goose she returns to the Lo River42 Blowing a flute he ascends Mt. Sung.43 His clothes are fragrant still with the musk perfume The pillow is slippery yet with the red traces. Thick thick, the grass grows on the dyke Floating floating, the tumbleweed yearns for the isle. Her plain cither plays the Resentful Crane Song44 In the clear Milky Way she looks for the returning wild goose.45 The sea is broad and truly hard to cross The sky is high and not easy to traverse. The moving cloud is nowhere to be found—46 Hsiao Shih stays in his chamber.47

All of Chang’s friends who heard of the affair marveled at it, but Chang had determined on his own course of action. Yüan Chen was especially close to him and so was in a position to ask him for an explanation. Chang said, “It is a general rule that those women endowed by Heaven with great beauty48 invariably

39  The word “bracelet” 環 is a homonym for 還 “to return” (Su). 40  For 宵 “evening” read 淸, with T’ang-tai ts’ung-shu 120.7b. 41   暗蟲, e. g., moths. For 遠 “keep away” read 遶 “circle,” with ibid. 42  Again the Goddess of the Lo River theme. 43  There are no pronouns, but the allusion to Wang-tzu Chin, who ascended Mt. Sung playing the flute, suggests that this refers to Chang. 44  Su Chung-hsiang identifies 怨鶴 with 離鶴操, a cither piece. 45  Which might be bringing a message. 46  A reference to the Goddess of Mt. Kao-t’ang? 47  Hsiao Shih was the flute-player immortal who taught Nung-yü. 48  The term 尤物 was used in Chang’s first speech in the story, where it seemed to have a wider connotation than “beautiful woman.” Here it is associated with its Tso chuan occurrence (Chao 28): “Great beauty entails great evil…. Where there are beautiful creatures,

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1339

either destroy themselves or destroy someone else. If this Ts’ui woman were to meet someone with wealth and position, she would use the favor her charms gain her to be cloud and rain or dragon or monster49—I can’t imagine what she might turn into. Of old, Emperor Hsin of the Shang and King Yu of the Chou50 were brought low by women, in spite of the size of their kingdoms and the extent of their power; their armies were scattered, their persons butchered, and down to the present day their names are the objects of ridicule. I have no inner strength to withstand this evil influence. That is why I have resolutely suppressed my love.” At this statement everyone present sighed deeply. Over a year later Ts’ui was married, and Chang for his part had taken a wife. Happening to pass through the town where she was living, he asked permission of her husband to see her, as a cousin. The husband spoke to her, but Ts’ui refused to appear. Chang’s feelings of hurt showed on his face, and she was told about it. She secretly sent him a poem: Emaciated, I have lost my looks, Tossing and turning, too weary to leave my bed. It’s not because of others I am ashamed to rise, For you I am haggard and before you ashamed. She never did appear. Some days later when Chang was about to leave, she sent another poem of farewell:

such as can move men, if they are not moral and virtuous, there will inevitably be disaster.” 甚美必有甚惡・・・夫有尤物，足以移人，苟非德義，則必有禍. 49   蛟 is defined as a rain dragon, also as a female dragon. It occurs in combination with 螭 (itself variously defined) as a water spirit. The saying which provides the point of departure for this speech is probably Chou Yü’s remark about Liu Pei (San-kuo-chih chi-chieh 54.10a), “I fear if the dragon gets clouds and rain, it will be no creature of the pond,” that is, given scope, Liu Pei will be dangerous. Shu yen ku-shih 6.47b書言故事 (quoted by Morohashi 33009.28) has an entry “To exploit a situation to undergo a complete metamorphosis is expressed by ‘The water dragon gets clouds and rain’ ” 乘勢變化曰蛟璃得 雲雨. In applying the expression to Ying-ying, Chang underlines the sexual connotations of “cloud and rain” by using the phraseology of the “Kao-t’ang fu”: 旦爲朝雲，暮爲行雨. 50  Hsin is the familiar last ruler of the Shang dynasty, whose misrule and fall are attributed to the influence of his favorite Ta-chi. King Yu, last ruler of the Western Chou, was misled by Pao-ssu. The irrational behavior of both rulers was the result of their own infatuation and the wicked frivolousness of the women they loved.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1340

Hightower

Cast off and abandoned, what can I say now, Whom you loved so briefly long ago?51 Any love you had then for me Will do for the one you have now. After this he never heard any more about her. His contemporaries for the most part conceded that Chang had done well to rectify his mistake. I have often mentioned it among friends so that, forewarned, they might avoid doing such a thing, or if they did, that they might not be led astray by it. In the ninth month of a year in the Chen-yüan period (785–805) when an official, Li Shen,52 was passing the night in my house in Chingan Street, the conversation touched on the subject. He found it most extraordinary and composed a “Song of Ying-ying” to commemorate the affair. Ts’ui’s child-name was Ying-ying, and Li Shen used it for his poem.53 The story begins as a third-person narrative, with the action viewed through Chang’s eyes. Until we read Ying-ying’s letter, we depend on his insights to know her state of mind, though we have already heard her reproachful speech when he is about to abandon her. It is then she first mentions—and without contradiction—Chang’s oath of lifelong devotion. By the end of the story the narrative voice has become identified with Yüan Chen, Chang’s friend. The transition from impersonal third-person narrative to first-person is made almost imperceptibly. The first poem by a friend (Yang Chü-yüan) leads to the disproportionately long one by Yüan Chen, with its flowery erotic passage and excessively allusive conclusion. It serves at least to convince us that “Yüan Chen was especially close to Chang” and prepares us to accept him in the role, first as interlocutor and then as moralizing commentator. Except as compositions by witnesses out of real life, these two feeble poems have scarcely any function, the long one in particular impeding the flow of narrative without contributing either illumination or emotional depth. Translators understandably omit 51  This line is subject to several interpretations. Either “you loved me then,” or “I loved you then”; 且 can mean “temporarily” or “basically” (this is Chang Hsiang’s reading; see 詩詞 曲語匯釋, p. 88). 52  Li Shen 李紳 (tzu Kung-ch’ui 公垂) died 846. He appears in Po Chü-i’s poem-titles as 小李. 53  Four stanzas of the poem (a total of 42 lines) are quoted in Tung Chieh-yüan’s Medley, Hsi-hsiang chu-kung-tiao (Peking: Jen-min wen-hsüeh, 1962), pp. 10, 36, 69, 91. The first stanza was collected in cts 5493; attention was called to the others by Tai Wangshu 戴望舒，小說戲曲論集 (Peking: Tso-chia, 1958), pp. 2–3.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1341

it, but it is of interest when we consider the role in the story of the narrator it serves to introduce. In the conclusion the narrator not only endorses Chang’s course of action, but even claims to cite it as an example to be emulated. It is here, in what appears to be an unnecessary epilogue, that a basic contradiction in attitudes is introduced. For there is nothing within the frame of the narrative to persuade the reader to accept Chang’s rejection of Ying-ying as justified, morally, emotionally, or even as a matter of practical convenience. That is not to say that he did not have his reasons, and they may well be conveyed in the remarkable farrago he delivers in response to the narrator’s query; but it is hard to accept the narrator’s assertion that Chang “did well to rectify his mistake.” It is tempting to try to extract a grain of psychological sense from Chang’s speech, and to guess that he has at last found himself sexually inadequate to the girl he began by seducing: “This Ts’ui woman… would… be cloud and rain or dragon or monster….” The sexual overtones are clearly there, though the expression primarily suggests transformation on accession to power. Chang turns it into an attack on Ying-ying’s moral character, accusing her of being a yao nieh, “an evil influence,” a witch. Nothing we have been told about her in the story remotely prepares us for such a charge; it seems intelligible only as a product of the speaker’s state of mind. Consider Chang’s situation: he has “determined on a course of action” which will surprise and perhaps disappoint his friends, who, thanks to his indiscretion, are only too well informed about his love affair and the extent of his commitment to Ying-ying. He has this chance to explain himself and to justify an act which may seem heartless and disloyal. What can he say? That he is tired of her? It is perhaps more respectable, if less convincing, to claim that he is afraid of her. However, if he were free to invent, a powerful and unanswerable excuse was ready to hand: he could claim to be betrothed; a stern parent has committed him irrevocably to someone else. This was the predicament of the unfortunate Li I in the “Story of Huo Hsiao-yü,” and though Li I is blamed by his friends for the way in which he abandoned the courtesan Hsiao-yü, there is no suggestion that he was free to flout his mother’s wishes. And in fact, in our story, Chang did marry within the year. Though we are not told so, it may well have been a more advantageous match; and in marrying to further his career he might have found an acceptable excuse—Ying-ying’s letter admits that to “a man of the world” (ta jen, and the term is not necessarily pejorative) this would be an understandable course of action. Since there were available to Chang other forms of self-justification, his claim of one that goes against the evidence of the story itself invites the reader to be skeptical: perhaps Chang is not telling the truth. We could look on him

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1342

Hightower

either as self-deluded and unaware of his real reasons, or hypocritical and unwilling to reveal them. In the first case, we are not tempted to commend him for moral rectitude, but we can understand his action as necessary to his selfpreservation. It would even be possible to feel pity for his weakness, which his subsequent attempt to see Ying-ying after her (and his) marriage further confirms. Such a reading is not impossible to reconcile with the narrator’s moral, which becomes a cynical comment on Chang’s character. Against it is the assumption that a T’ang story could be written on such a high level of sophistication and deviousness. If Chang is to be taken as a mealymouthed hypocrite, a Tartuffe trying to sanctify his own lechery, we must assume an attitude of strong disapproval on the part of the author of the story, at variance with the expressed attitude of the narrator. This superimposed irony is really more weight than the structure of the story can sustain; if it was intended, the result must be judged a failure. The difficulty is not in finding reasons for Chang’s behavior, but in reconciling it with the narrator’s unqualified approval. To claim, in the face of the evidence, that Chang’s action was the morally correct one, that he should be emulated by all young men who find themselves in his position, makes one suspect that the narrator was not speaking objectively, and one is tempted to look for reasons that would make Yüan Chen (or his spokesman) biased in Chang’s favor. It may have been this impression that the author was trying too hard to rehabilitate the hero of his story that originally led to the identification of Yüan Chen with Chang, and when we examine it with this in mind, he seems almost to invite the identification: “Yüan Chen was especially close to him and so was in a position to ask him for an explanation.” “Chang composed a poem of some sixty lines on ‘An Encounter with an Immortal,’ ” but the text of the poem of that title which we are given is Yüan Chen’s “continuation,” also of sixty lines and hardly a sequel to anything, except that it continues the narrative to the lovers’ final separation. (It is interesting, considering the otherwise generous documentation of the story, that we are presented with no single specimen of Chang’s poetic talent.) As early as the twelfth century, Wang Chih,54 apparently reading the story as history, felt that Chang could only be Yüan Chen, and sought confirmation in biographical details. Yüan Chen was born in 779; he would have been

54   王銍 (tzu Hsing-chih 性之). His Ch’uan-ch’i Set to Rights is quoted by Chao Ling-chih 趙令畤 in Hou ching lu 侯鯖錄 5.1a (Chih-pu-tsu chai ts’ung-shu, vol. 172).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1343

twenty-one in the year 800, when Chang was also twenty-one.55 Family names correspond: the hero’s mother was a Cheng, as was the mother of Ying-ying; so was Yüan Chen’s mother, who was the daughter of Cheng Chi, and there was another daughter of Cheng Chi who married Ts’ui P’eng; she would qualify as the mother of Ts’ui Ying-ying, cousin of Chang/Yüan. The most persuasive congruence was a hearsay report of a “Grave Inscription for Aunt Cheng” by Yüan Chen, where the author wrote that he had preserved her family from a troop mutiny. Unfortunately no one, including Wang Chih, has been able to find the text of such a composition, and it must be dismissed as a misguided attempt to be helpful by Wang Chih’s friend Yang Fu-kung. None of these verifiable correspondences is as impressive or as interesting as the one Ch’en Yin-k’o pointed out,56 between the names of the protagonists of Chang Cho’s “Dalliance in the Abode of Immortals” and those of the chief characters of Yüan Chen’s story. This serves as a useful reminder that “The Story of Ying-ying” is after all a story, that it has literary dimensions as well as (possibly) personal ones. But it is for a literary purpose that we are pursuing the question of the relation between author and his characters, to see whether it provides a better understanding of a puzzling story. Wang Chih may have had no better motive than simple curiosity;57 in any case he found evidence of another kind in a collection of over a hundred “Antique Love Poems” 古豔詩 by Yüan Chen,58 of which he possessed a copy. He quotes the text of nine poems written by Yüan Chen to a girl whom he identifies with Ying-ying, or about a love affair which could have provided the material for “The Story of Ying-ying.” Several are vague enough to apply to any love affair. Two depend on the occurrence of the word ying “oriole,” a not uncommon witness of secret meetings as well as a component of Ying-ying’s name; but since they are given the tide “Spring Song” 春詞, which was the generic term used in the story for the erotic verses Chang wrote to seduce Ying-ying, Wang Chih thinks these may be the ones:59 55  He refers to the line that says Chang had reached the age of twenty-two (twenty-three sui) without ever having any experience of women, but makes the age twenty-one (twentytwo sui). 56  Op. cit., p. 102. 57  “Whenever I hear of something cryptic and unclear, or clear but contradictory, it is like a stone in my breast: I cannot rest until I have investigated and reduced it to a consistent explanation” (Hou ching lu 5.1a). 58  No such title exists or is listed in the T’ang shu “Essay on Bibliography,” nor are the poems he quotes to be found in Yüan-shih Ch’ang-ch’ing chi. But they all occur in chapter 5 of the tenth-century anthology Ts’ai tiao chi (abbreviated ttc; sptk ed.). 59   t tc 5.11a, with the alternate title 古豔詩 in cts 4644–4645.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1344

Hightower

I When spring came I often went to the house east of Sung Yü’s. With loose sleeves and opened robe I await the good wind. Oriole hidden in willow shade, all silent, no one speaks; Only the blossoms over the wall envelop the tree in pink. II In the secluded courtyard, no one; grass and trees shimmer. Pretty oriole does not speak, but hides in the foliage. Idly she dabbles in the water, setting petals adrift That float out through the gate to entice Master Juan. The first line of the first stanza is reminiscent of lines 19–20 of the “Encounter with an Immortal”: “From an outing north of Loyang’s wall / By chance he came to the house east of Sung Yü’s,” and the tree in bloom over the wall might just be that convenient apricot. The second stanza is less relevant. Another, entitled “Spring Morning” 春曉, is easily read as a recollection of the events of the story:60 The sky halfway bright, halfway not yet bright, Drunk I smell the flowers’ emanation, asleep I hear the oriole. The dog is roused as the bell sound stirs— Twenty years ago it happened in the early morning temple. The temple setting gains in significance because of the peculiar word wa-erh

娃兒61 for “dog” in line 3, a word which turns up in the “Dream of an Outing in

Spring” discussed below. The poem most obviously connected with the story, as quoted by Wang Chih, has the title “Ying-ying”; it appears in Ts’ai tiao chi as the first of six “Absent Love” 離思 poems. There are several variants, and I shall give first the Ts’ai tiao chi version:62 Deep red, pale green, her old clothes; She has neglected to comb her hair, her makeup is faded. Magnolia girdled with mist, catching the early morning moon Peony after rain weeps for the setting sun.

60   t tc 5.10b; cts 4642. 61   c ts writes the variant with the dog radical instead of the woman radical. 62   t tc 5.11a; cts 4642 includes the variants.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1345

Yüan Chen and ‘ The Story of Ying-ying ’

The fleeting hidden smile is no smile at all The fragrance you seem to smell is not really perfume. She keeps turning her sparkling eyes to chide her mother: So lax as to introduce her to a young man. It is hard to imagine that this could have been composed independently of the scene in the story which describes Ying-ying’s appearance and behavior when summoned by her mother to meet Chang. Even as quoted by Wang Chih, without the decisive mention of the mother, it fits very well: Her lively eyes are charming without speech: How careless to introduce her to a young man! This is the only poem with the name “Ying-ying”—which occurs in the story only as an afterthought, to explain another poem title—and even here the title is suspect, or anyhow not confirmed. But there is a poem “For Shuang-wen” 贈雙文63 and Shuang-wen could of course be a euphemism for Ying-ying (or for any other reduplicative name, for that matter). Shuang-wen is mentioned again in each of the five stanzas of “Recollections” 雜億,64 and certainly stands for a partner in a love affair, real or imagined, but the relevance of these poems to the situation of the story is tenuous at best. Of all these poems quoted by Wang Chih, the most interesting is a series of three with the evasive title, “Three Songs of Separation in the Ancient Style” 古決絕詞三首.65 The word ku in the title may be taken to mean “of antiquity,” hence putting the situation commemorated in the poem far back in ancient times. And chüeh-chüeh can mean “rejection” as well as “a definitive separation.” Although these poems do not fit any part of “The Story of Ying-ying,” they express the contrasting points of view of two estranged lovers, each ready to blame the other for what has happened. The first and third are in the person of the girl who feels herself neglected and abandoned; the second gives the young man’s side of the story. 63   t tc 5.6b. 64   t tc 5.12a; cts 4643. 65   t tc 5.4a; cts 4637. These songs are interesting metrically. The first is basically in a 5-word meter; of the six 7-word lines, four begin with 2–word phrases that invite a strong caesura and could be treated as belonging outside the meter. The first long 10-word line reduces easily to 3+7, and the second, of 8 words, divides 3:5. The second song is sao-type throughout, a striking and effective contrast to the first. The third reverts to 5-word, with four 7–word lines, two of which are 2:5, in the same metrical idiom as the first.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1346

Hightower

I I would rather we were Spinning Girl and Cowherd, stars in the sky above; I do not want to be The pink flower that fades in a single day. They meet once every seventh day of the seventh month66 And between meetings their hearts never change. Not like the flower, blooming this morning and gone by night, At the mercy of the wind, east, west, north, south. Fated not to stay together I regret we will love no more. So it is when we are close— Can it be different when we are apart? The spring wind stirs feelings, the shrike chatters— And this is the season when you are going to leave. Clasping hands I implore you But you never mention a future meeting. Since you are resolved to break it off My hopes were vain67 from the start. Well, if it were death that separated us, I would not mourn for you forever. One can imagine Ying-ying expressing her resentment in such terms, and ending with the defiant resolution of the last couplet. The next poem takes the other side: II Though the ice of spring is about to melt, My feelings still are tied in knots. There is a lovely person Far away, alas. One whole day I do not see her— That one day is like three years,68 And three years this separation has endured. The water touched by wind—even a little and already there are waves;

66  The traditional day (or night) of reunion for the Spinning Girl and the Cowherd. 67   參差 “out of phase”—with reality. 68  The preceding line and the comparison are from Shih ching No. 72 / 3.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1347

Yüan Chen and ‘ The Story of Ying-ying ’

The bamboo in leaf grows tall but shows no measured nodes: What can one expect of peach and plum in springtime69 Striving for attention from everyone?70 I see myself as a wandering cloud, How can I keep you pure white as the snow? I am touched that a broken mirror71 shows a clear reflection; I see the remaining red traces of shed tears. Fortunate that no one preceded me, How can I keep others ultimately from taking my place? It is all over: The Spinning Girl has left the Cowherd.72 Once a year they crossed for a brief meeting— Living on each side of a river, anything can happen. The charge of inconstancy is unmistakable: who knows what she has been up to while he has been away? She is as responsive as water to wind, her desires are as unmeasured as the bamboo shoot that springs up with no nodes, she is as beautiful as a flowering tree in spring, noticed and courted by everyone. He may have been her first lover, but during his absence there must have been others. He does not appear to be asking for reassurance, and the next poem, if it is intended as an answer to his, does not provide it. III When every night we slept in one another’s arms We still were bound by ties of deepest love. How can the events of a year Be talked out in a single evening? All I know is constant yearning for you; When is there time to be happy together? The rainbow bridge was built73 in early evening, The dragon chariot is waiting at dawn. How they hate the wild magpies’ vacillation 69  From the proverb (quoted in Shih chi 109.21): “Peach and plum do not speak, but paths grow up under them”—because of the visitors attracted by their beauty. 70  “Competing for people to break [their branches].” 71  A metal mirror broken in half, like a tally, to serve as a pledge of love. 72   黃姑 for 牛星, for the rhyme. 73  The bridge for the lovers to pass over the Heavenly River is usually improvised by magpies, referred to in the next line but one.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1348

Hightower

Resent the careful time kept by the Heavenly Cock. The light of dawn grows gradually clear, The shining stars fade one by one. Once separated, another whole year A whole year, how can one get through? Better than this far-off rendezvous Would be the separation death brings. The Lord of Heaven must be jealous of lovers Or he would let them break it off for good. She concludes that meetings as infrequent as theirs are not worth the pain it costs in hopeless longing, and she cannot accept the lot of the Spinning Girl and the Herd Boy, though in the first poem that is what she said she would settle for. In this sequence, Yüan Chen must be given credit for a high degree of imaginative understanding of the states of mind of mutually reproachful lovers. The evidence of these lyric poems is hardly conclusive, nor, in the nature of things, is it likely to be. They suggest that Yüan Chen was preoccupied with secret meetings and unfaithful lovers, but his knowledge of such things could have been limited to an acquaintance with tales like “Huo Hsiao-yü” or the gossip of his friends. The last poem quoted by Wang Chih, however, is of a different sort. It is not a love poem, but a sort of autobiography, though hardly straightforward narrative. A Springtime Outing: A Dream74 Years ago I dreamed of an outing in spring An outing in spring, and what did I find? I dreamed I entered a deep, deep cavern And there achieved my lifelong desire. 5

10

Clear and cool the shallow waters flow, A painted boat with magnolia oars drifts by. I pass a myriad peach trees in flower And stroll along the bamboo grove path. A long covered corridor surrounds a little house Where doors and windows show at every turn.75

74   夢遊春 ttc 5.1a–3a, cts 4635–4636. Su Chung-hsiang, op. cit., p. 201, provides a few notes. 75  I am not sure I understand this banal line: 門牖相囘互.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1349

Yüan Chen and ‘ The Story of Ying-ying ’

Below the house all sorts of flower bushes, Around the bushes circling egrets. The pool shines with reflected sunrise And then the morning sun is bright76 and warm. 15

20

25

30

35

40

I dare not climb the steps directly But pace back and forth beside the pool. The black dog makes no sound; Green Jade is the one I always yearned for. I approach at length the curtained door And hesitate, feeling fearful still. I take time to peek in the rooms west and east— Everywhere rare objects set out in view: The partition painted bright green, The hump-hooks gilt and purple.77 Presently the sun rises gradually higher; From the light and sounds, people are about to wake. Hungry, the parrot cries aloud, Sleeping still, the little dog is angry. A curtain opens and the serving girl gets up, Sees me, and understands without a word. She spreads a red embroidered floor covering And sets out inlaid and decorated vessels. She quietly raises the kingfisher bed-curtain And discloses a coral tree to view. One cannot make out the flower-like girl And is only surprised at the fragrance like mist. Her body turns: a magnolia bloom aslant Her form contracts: sunrise clouds cluster. Her sleepy face: peach blossom bursting in the breeze The beads of sweat: dewdrops on lotus. Her hair she combs in a hundred folds,78

76  Read 明 with cts for ttc 鳴. 77  This mysterious couplet, 隔子碧油糊, 駝鉤紫金鍍, may conceal an elaborate euphemism, for at this point he seems to have summoned the courage to get into her bed. 78   t tc records what may be the author’s own note, “the modern style hairdo.” It is literally “hundred-leaf hair-bun” 百葉髻.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1350

45

50

55

60

65

Hightower

Her double-lotus shoes are covered with gold foil,79 Soft woven blouse of filigree80 Bright trousers of a shared-joy print.81 But now her once fresh makeup is worn thin Her faded clothes are grown old,82 Just like the red peony flower When the rains come at the end of spring. The dreaming soul is easily awakened In the spirit world it is hard to stay long.83 Every night I gaze at the Milky Way— No way again to follow that stream.84 After the mind has been fixed on what the heart desires To return is like the sudden enlightenment of Zen. During the eight or nine years after awakening I did not turn again to look at flowers. In springtime encounters in the capitals Chattering, the birds all congregate. But when it came my time for flower-viewing I could only write lines in memory of the Fairy. The more frivolous this floating life became, The more firm my moral nature grew. Recently I still wrote poems on Dreaming of the Fairy But I know now it is a wearying of the flesh. When was a dream ever worth talking about? When the right time came, I got married.

79  Original note: “Walking-in-the-palace style.” Ch’en Yin-k’o op. cit., p. 88, explains the phrase “double tower shoes” 重臺履 as deriving from the expression “double tower lotus,” a sort of double-tiered lotus flower, appropriate both as an epithet for high-heeled shoes and as a reference to the Consort P’an 潘妃 of the last ruler of the Ch’i dynasty, one of whose extravagances was to plant lotus flowers of gold so that as she walked over them he could say “A lotus springs up at her every step” (Nan shih 5.20a). 80  The note reads 琴瑟色 (a material? a color?), perhaps to explain the incongruous 鈿頭 “inlaid” (of a garment). 81  Here the gloss is 夾纈 “name of a print-dyed (pattern),” a technique invented in the first half of the eighth century, according to T’ang yü-lin 4, quoted by Ch’en Yin-k’o, op. cit., p. 88. 82  This couplet seems to summarize the passage of time. 83  The dream narrative is over; what follows is reflection on the affair, ending with determined renunciation. 84  Cowherd and Spinning Girl theme?

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1351

Yüan Chen and ‘ The Story of Ying-ying ’

70

75

80

85

I was just twenty-three at the time,85 On the auspicious occasion of our wedding night.86 Hibiscus’87 jade pendants came to greet me On the tall pine the vine found support. The Wei Clan was just at its most prosperous Having entrée gave much pleasure. By the mansion the clear pond swelled, Neighing colts drew the red carriage; Many were dancing beside spacious pavilions, At the long table guests seated together. Green spring—how many days does it last? Secret worms hide in the flowers and fruit. The autumn moon shines on Master P’an88 The empty hills enfold Tutor Hsieh.89 The red house, alas, its walls are fallen, Gold Valley Garden90 is lost in rank growth. Stones press against the broken railing The gate leans against the steps and hitching post. Dream and reality—you may say they differ, Both are equally hard to preserve forever. What are my feelings like, after all?

85  Converting 二紀 “two cycles,” twenty-four years, to the Western way of reckoning age. Arthur Waley used this date (802) as the basis for his statement that Yüan Chen was only fifteen at the time of the love affair recounted earlier in the poem, because of the lines “During the eight or nine years since I awoke, / I have not turned to look at flowers,” which he interprets to mean that Yüan Chen for this period “led a life of complete celibacy” (Life and Times of Po Chü-i, p. 70). This is assuming too much (though Su Chung-hsiang makes the same assumption, op. cit., p. 100); Yüan Chen would hardly have included his wife, the well-connected Wei Ts’ung, among the “flowers.” There is no need to exaggerate either his fidelity or his precociousness by taking the known date of his marriage as coming eight or nine years after the end of his love affair. 86   三星 gets its meaning from Mao shih No. 118. 87   朝蕣: the hibiscus is an ephemeral blossom; it symbolizes his bride, Wei Ts’ung. 玉佩 refers to Mao shih No. 83, another marriage song. 88  P’an Yüeh, who lost his wife. 89  Tutor Hsieh 謝傅 stands for Yüan Chen’s father-in-law, whose last post was Tutor to the Heir Apparent (Su Chung-hsiang). 90   金谷: the site, near Lo-yang, where Shih Ch’ung had his garden, scene of luxurious festivals.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1352

90

95

Hightower

Tangled thread that never makes a skein. Wen-chün’s “White Hair Song,”91 A-chiao’s “Golden Chamber Rhapsody.”92 Sheng-chi in her Jade Tower93 Ming-fei in her green tomb.94 Consigned for ever, their bones to the eternal dust All follow the flowing waves to pour into the sea. Past and present are too much alike, How contrast white and undyed silk?95

Especially since I, in my prime From early times did my duty. Imperial notice placed me first among the outstanding 100 I offered council in the palace, setting out right and wrong. At thirty I again came up to court, No sooner came up than down I fell again.96 Fame and favor came to me early Reversals and retreat were also frequent. 105 The instinct for directness is in my vitals This humor grows daily more chronic. If I fail to speak, my feelings are not satisfied, When I satisfy my feelings, my words are often obstinate. Obstinacy is truly what people dislike 91  Cho Wen-chun is supposed to have composed the “White Hair Song” 白頭吟 to complain of Ssu-ma Hsiang-ju’s infidelity. 92  The “Ch’ang-men Palace fu” attributed to Ssu-ma Hsiang-ju (Wen hsüan 16.6b–8a) is written in the person of the Empress Ch’en, after she had lost the Emperor’s favor. She was the former Princess A-chiao, for whom the child Liu Ch’e, the future Wu Ti, offered to build a golden chamber (Han Wu ku-shih 1.1b). 93   盛姬 was a favorite of the Emperor Mu, for whom he built the Tower of the Double Jade Disk 重璧之臺 She died young (Mu T’ien-tzu chuan 6.1b). 94  Ming-fei was Wang Chao-chün, the Palace Lady of the Han Emperor Yüan, married off to the Hsiung-nu ruler, who died of grief and whose grave in the northland is ever green. 95  I. e., there is no difference between old love and new wife, insofar as both are gone. The line alludes to the anoymous yüeh-fu poem about a divorced wife: “The new wife is skilled at weaving chien 嫌, / The old is skilled at weaving su 素; Of chien she weaves a bolt a day, / Of su she weaves fifty feet and more. If you compare the chien with the su, / The new wife is not the equal of the old” (Ch’üan Han shih 3.8b). 96  In 809 Yüan Chen was recalled to Ch’ang-an from his post in Lo-yang, was reprimanded, and sent off to Chiang-ling in disgrace. See Arthur Waley, op. cit., pp. 65, 70.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1353

Yüan Chen and ‘ The Story of Ying-ying ’

But my disposition was Heaven’s bequest. Better be submerged and fragrant97 Than floating, an empty gourd. In truth I was bent on keeping my standards, I was not wise about what I should have done. 115 I have learned from much experience How wrong it is to exert oneself in plans. Fine jade, polished, makes a courtier’s insignia, Gold metal goes into the arsenal for war. I only thought to be firm and pure 120 Not knowing I would have to be cast and polished. A long rope restrains the wild horse A fine net traps the rabbit in hiding, But as long as they stay away from the world Who can fetter them from afar? 125 Time passes as though it were flying Trouble is quick as if racing. 110

Since my former understanding was not refined To whom can I complain of this trip? In all haste I leave for Chiang-ling 130 With whom can I talk there or laugh? The river flowers may be lovable But they are not what my heart desires. Carnations show themselves sly and crafty Turnips spread over acres of fields. 135 They are alike in that they grow in poor soil Shallow or deep, they are not worth envying. The leaves of the lotus grow above the water Round and firm they dwell in the water. Water you may pour upon the leaves, 140 But look! You cannot sully me. It is only the first part of this long poem, the part reported as a dream, that tells of a love affair and the poet’s attitude toward it; but the rest deserves attention for what Yüan Chen volunteers about himself. There is more justification than usual in taking the poem as autobiographical and deliberately selfrevelatory, for it prompted a response and a comment by Yüan Chen’s friend 97  Like 沈水香, aloes, a heavy wood used for incense that sinks in water.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1354

Hightower

Po Chü-i. His “One Hundred Couplets in Response to the Poem ‘Dream of a Spring Outing’ ” 和夢遊春詩—百韻 has the following introduction:98 After Wei-chih got to Chiang-ling, he sent me a poem of seventy couplets, “A Dream of a Spring Outing,”99 and introduced it with these words: “What I have written should not be disclosed to any but my friends, but it is essential that my friends know it. Lo-t’ien is my friend, and I dare not keep it from him.” I read these words and keep repeating his message, which is, essentially, to regret what is done, and to be wiser in the future. In my opinion, if one does not regret, one learns nothing, and there’s an end to it. But if one regrets this, then one must understand that. If one turns away from that, understanding where one went wrong, then one ought to revert to the truth—you especially, who long have externally worn the robes of a Confucian while inside honoring the conduct of the Buddhists. From this time on, whither shall you go if not on the Road of Awakening, to what shall you revert, if not to the Gate of Emptiness? The poem which I have written in response is ultimately to this point. For if the feeling is not deep, the repentence will not be ripe. And if the feeling is not intense, the enlightenment will not be profound. So I have expanded the original seventy couplets to a hundred, repeating what in the dream excursion made the deepest impression and recounting what was most moving in marriage and public service, wishing to follow in all its ramifications the irrationality, to make wholly clear the error, after which comes the return to the truth. It is the same idea as rushing from the burning house to dwell in the City of Change stated in the preface to the Lotus Sutra,100 and the entry into the place of lust and the visit to the wine shop in the Vimalakirti Sutra.101 This poem of mine is even more not to be shown to any but friends; please keep it secure. The most definite information conveyed in this preface is the fact that Yüan Chen wrote his poem as a private communication and that he regarded one 98   Works, 14.20ab. 99  This would seem to date Yüan Chen’s poem in late 809 or in 810. 100  An unreal city devised by the Buddha, to give rest and comfort to weary seekers after salvation (Miao-fa lien-hua ching 妙法蓮花經 3.25b–26a). The “burning house” 火宅 as a symbol for the human condition appears ibid. 2.16a–b. 101   維摩詰所說經, Taisho No. 475; vol. 14, p. 539.1: “He enters into the houses of lust to show the error of desire. He enters into the wineshops and there can make his determination prevail.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1355

episode as not altogether to his credit, though the suggestion that he derived a moral lesson from it is more reminiscent of the concluding paragraph of “The Story of Ying-ying” than anything in the poem itself, which sounds rather selfcongratulatory. It is significant that Yüan Chen’s poem is not included in his collected works, but like the love poems quoted above occurs only in the T’ang anthology Ts’ai tiao chi; Po Chü-i’s preface guarantees the authenticity of this one and makes the attribution of the others credible.102 The “Springtime Outing” falls into three parts, the title being an appropriate label for the first part only. The clandestine love meeting is followed by a brief account of the poet’s marriage, with much emphasis on the position, influence, and wealth of his father-in-law. His wife’s death is deplored—the reality of marriage turns out to be as illusory as the dream of romance—that is to say, both women are lost to him. Meanwhile duty calls. He treats the ups and downs of official career in summary fashion, finding the explanation for his ultimate failure in his own stubborn integrity, unappreciated by a corrupt world. This has a familiar sound, and the remaining thirty-two lines are the stock lament of the courtier out of favor, in the tradition of Ch’ü Yüan. It remains for Po Chü-i to supply the consolations of religion to one disappointed in love, marriage, and political ambitions. I find it hard to admire either Yüan Chen’s poem, or Po Chü-i’s elaboration of it; the picture they offer of complacency and self-congratulation is not particularly attractive. Yüan Chen’s poem establishes one fact for which the poems quoted earlier were inadequate evidence, that he did participate in a love affair before his marriage, an affair that made a profound impression on him. It does not, of course, prove—even though the juxtaposition of the other poems suggests it—that “The Story of Ying-ying” was directly based on that experience, and it is here that we must part company with Wang Chih, who set out to prove just that. For him the story was a record of something that happened, and he is concerned to show that real people are hidden behind the names of the characters of the story. If Yüan Chen were writing about someone else’s experience, not his own, how could he have written so circumstantially? This is surely to oversimplify the relation between fiction and the experience which the writer draws upon in creating it. To say that Yüan Chen is Chang is as irrelevant to

102  Yüan Chen himself mentioned “over a hundred love poems” which he had written that “offend against moral teachings”; this in a letter to Po Chü-i which describes the 800-odd poems he had written from the age of fifteen to the year 812, when he was thirty-three years old (Works 30.2b). They presumably circulated separately to survive into the twelfth century, when Wang Chih still possessed a copy.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1356

Hightower

the story as to suggest, as Ch’en Yin-k’o103 did, that Ying-ying is a prostitute, for within the framework of the story itself both propositions are obviously false. There are inconsistencies within the story which may conceivably reflect something of the actual situation which Yüan Chen used as the basis for his story, as for example the ambiguous role played by Ying-ying’s mother, but they are there by inadvertence, not as hints to the reader. The problem which we are concerned with is one of understanding; to know how to take the story, we must make an assumption about the author’s attitude toward his material, and to do that intelligently we need to go outside the story itself. Yüan Chen’s “Springtime Outing” does throw light on this essentially psychological problem of how did he expect his readers to react to “The Story of Ying-ying”? That there is a problem is clear: a story is recounted with considerable objectivity: a young man seduces a girl of good family; they are in love; nevertheless he abandons her ostensibly because of some unspecified flaw in her character which he fears will make him behave badly if he ever achieves a position of power. The readers’ sympathies are all with the girl. She has been given the chance to present her side, and she does so with dignity and feeling. Then the narrator interposes to assure us that his hero did the right thing after all and should be the model for all young men in his situation. The discrepancy between expectations and denouement is so striking that we are taken aback: did the author really expect us to be persuaded by that hysterical speech of Chang’s, in spite of all he has told us about Ying-ying, and the letter we were invited to read (along with all the young gallants of Ch’ang-an)? Surely this must be a piece of heavy-handed irony, and we are being invited to despise Chang’s friend as well as Chang himself? It is at this point that the identity of the narrator becomes crucial. The story gives him the name Yüan Chen, and, so far as I know, only Arthur Waley has refused to make the obvious assumption that this is also the name of the author, and that the author and narrator are the same person, the poet-friend of Po Chü-i. Waley gave no reasons for his skepticism,104 but it is possible to pursue the implications of taking the narrator as Yüan Chen, while making the author someone else. In the first place, it opens up the possibility of the ironic reading which makes us disapprove of the narrator; the author, who would have been no friend of Yüan Chen, wrote the story to cast doubt on his moral judgment. There is reason to believe that ch’uan ch’i stories were used as the vehicle for 103  Op. cit., p. 104. 104  “I do not believe that ‘The Story of Ying-ying’ is really by Yüan Chen; but the question is too complicated to discuss here” (op. cit., p. 222, Additional Notes 70/1).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1357

personal or political attacks on an individual, and those in which the name of a well-known person is introduced either as main character or as author are often open to the suspicion. The most notorious example is the Chou Ch’in hsing chi,105 attributed to Niu Seng-ju, a first-person narrative in which the main character identifies himself as Niu. The documented attempt by Li Te-yü to use the story to accuse Niu of disrespect toward the ruling dynasty106—not a light charge—makes it likely that the story was written in the first place to discredit a political rival. In “The White Ape”107 the use of the name Ou-yang Ho for the husband of the woman who bears the ape’s child sounds like a deliberate insult aimed at his son Ou-yang Hsün, though here the motive is not obviously political. Likewise the attribution of the unedifying tale Yu hsien k’u to Chang Cho (the name of the first-person narrator of the story), has been suspected of being similarly motivated.108 The possibility that “The Story of Ying-ying” was another such indirect attack on a prominent statesman is not impossible, but the degree of indirection does seem excessive. The same end could have been achieved with greater economy and much more certainty by giving the protagonist of the story the name Yüan Chen. Irony is a dangerous weapon and would seem to have misfired completely in this case, if the genesis of the story were to be explained in these terms. In fact, a story written for the purpose of discrediting Yüan Chen could have dispensed with the moral, or it could have been couched in terms more unambiguously damning. The hypothesis should be rejected as raising more problems than it solves. What about the identification of the narrator Yüan Chen with the poet of the same name, and making him the author? This would eliminate one possibility of irony, for the narrator then speaks with the author’s voice. In favor of the identification is the poem written by the narrator (the “Encounter with an Immortal”) which has stylistic and verbal similarities with poems generally accepted as Yüan Chen’s. Yang Chü-yüan as well as the official Li Shen (mentioned in the last line of the story as a friend of the narrator’s) were in fact Yüan Chen’s friends,109 and part of Li Shen’s “Song of Ying-ying” survives.110 None of 105   t pkc 489. 106  Quoted by Wang P’i-chiang, T’ang-jen hsiao-shuo (Hongkong: Chung-hua shu-chü, 1958), p. 154. 107   t pkc 444, translated by C. C. Wang, Traditional Chinese Tales, p. 12. 108  Wang Chung-han, “The Authorship of the Yu-hsien k’u,” hjas 11 (1948). 153–62. 109  Yüan Chen, Works 21.5a has a poem entitled “A Farewell to the yüan-wai Yang Chü-yuan”; ibid. 19.6b: “Dreaming of Li Shen at Ch’ang-t’an.” 110  See note 53.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1358

Hightower

this rules out a well-informed and skillful forger, but this is surely the easier and more economical assumption. (That the author Yüan Chen would lend a spokesman his name and then have him speak with a voice quite at variance with his own intentions is not absolutely excluded, but is unlikely.) If the narrator is in fact the author, Yüan Chen, irony would seem to be ruled out: the author really did mean us to accept his statement that Chang did the right thing. There is another identification already argued for, the triple identification of the protagonist Chang, the narrator Yüan Chen, and the poet Yüan Chen, who is the author. If we accept it as the basis for a reading of the story, we are left with the same disturbing contradiction: the story does not lend itself to the moral that the narrator-author draws from it, though the situation presented in the story derives to some extent from his own experience. The fact that Yüan Chen gives his hero a different name and relates the events in the third person suggests that, however closely he identified himself with Chang, and however true to fact the events narrated, Yüan Chen has made an effort to view himself from the outside. What he tells us about Chang we can believe, in the sense that this is the way he saw himself at some remove in time. The young man he presents is rather straitlaced, self-consciously and self-righteously moral, not easily influenced by his friends. He claims to have under all this a passionate nature, a claim not wholly borne out by his subsequent behavior. A chance meeting with Ying-ying leaves him infatuated, and in a speech to Ying-ying’s maid he claims passion so far overrides morality that he cannot even consider a regular betrothal and marriage. But he retires after his first rebuff by Yingying, ready to give up—apparently his situation was not so desperate as he had made out. He accepts Ying-ying’s visit happily enough and is depressed at having to leave her some time later to pursue his career. In spite of her hints, he makes no gestures in the direction of marriage. His final decision to repudiate Ying-ying is put on grounds of expedient-morality: he thinks she might be dangerous. Having renounced her, after his own marriage (and hers), he attempts to see her again, and behaves like a disappointed child when she refuses. Not a flattering portrait. Did the narrator find it so? In the first place, he wrote a long erotic poem about the “Encounter with an Immortal” in questionable taste, but with no suggestion of disapproval. He greeted Chang’s statement of repudiation with a sigh. He “often mentioned it among friends so that, forewarned, they might avoid doing such a thing, or if they did, that they might not be led astray by it.” So he really disapproved of the affair, in spite of the vicarious enjoyment reflected in the poem, and thought Chang did right to terminate it as he did. Clearly he was on Chang’s side all the time. What about the author? Yüan Chen is quite as naively outspoken and selfrevealing as his narrator. The poems he wrote about love affairs, one of which

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Yüan Chen and ‘ The Story of Ying-ying ’

1359

may or may not have been the model for “The Story of Ying-ying,” are yen shih, erotic poems, a genre about which he later expressed strong disapproval and which he excluded from his own collection of his poems. In his three “Songs of Separation” he demonstrated the ability to write understandingly in the voice of a girl neglected by her lover, even while presenting the suspicious lover’s case persuasively. The certainly autobiographical “Springtime Outing,” written after the death of his wife and at the time of his demotion to Ching-chou, after an elaborate description of a rendezvous with his mistress, dismisses her brusquely: “When was a dream ever worth talking about? / When the right time came, I got married.” His wife’s death was a blow, chiefly because it entailed the loss of the support and patronage he had enjoyed from her father. The smug tone of self-congratulation on his making a good marriage, combined with outrage at the blows dealt him by unkind fate, would have been perfectly in character for both Chang and the narrator of his story. One is reluctantly led to the conclusion: the author of the story was Yüan Chen, who wrote it out of his own experience and intended it both as self-justification and warning. While there is no room for irony in a reading of the story, there is remarkable objectivity in presenting both sides of the affair, and perhaps it also served the emotional needs of its author to present in a sympathetic light the girl he had such difficulty forgetting.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia and History in Mid-Ninth-Century Verse: Cheng Yü’s Poem on “The Chin-yang Gate” Paul W. Kroll Introduction Within the three centuries of the T’ang, the forty-four-year reign of Li Lung-chi 李隆基 (685–762, reg. 712–56), the monarch posthumously known as Hsüantsung 玄宗, represents the cultural high noon of the dynasty, when China was at its acme of political power, economic prosperity, and cultural influence visà-vis the surrounding states of East, South, and Central Asia. This is the socalled “High T’ang” (or, more accurately, “fullness of T’ang” 盛唐), comprising the K’ai-yüan 開元 (713–42) and T’ien-pao 天寶 (742–56) reign-eras, the age of the great poets Li Po 李白 (701–762?), Wang Wei 王維 (701–761), Tu Fu 杜甫 (712–770), and a host of others, when the capital city of Ch’ang-an 長安, boasting a population of over a million souls within the thirty square miles enclosed by its outer walls, was an international metropolis and indeed the grandest city of the whole world, when the emperor and the scholar-officials of his court enjoyed unparalleled luxury and license—until the rebellion of An Lu-shan 安祿山, beginning in December 755, brought all the splendid scene to a crashing end, including the capture in July 756 of Ch’ang-an by rebel forces and the ignominious flight of Hsüan-tsung himself and consequent death of his “Precious Consort,” Lady Yang 楊貴妃. Although the dynasty would eventually recover, with loyalist forces retaking the capital for a new emperor a year and a half later and control being reasserted over the whole nation again by 763, the T’ang world from then on was a changed and diminished realm. For those who lived in this later world there was little doubt, when they looked back, that a watershed had been crossed and that there were significant differences between their age and the grander one that had preceded it. This can be seen perhaps most noticeably in the poems and the prose accounts meant to evoke that former time which was chronologically still near but psychologically already distant. The drift of history into legend was already underway in the generation immediately following the An Lu-shan Source: “Nostalgia and History in Mid-Ninth-Century Verse: Cheng Yü’s Poem on ‘The Chin-yang Gate’,” T’oung Pao 89 (2003): 286–366.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_043

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1361

rebellion. One discovers it, for instance, in poems on the vanished splendors of Hsüan-tsung’s years by the likes of Wei Ying-wu 韋應物 (737–792?)1 and others. Whether this was done with regret, nostalgia, fascination, rebuke, or varying mixtures of these feelings and views, scholars made of Hsüan-tsung’s reign a conservatory for memories and images both radiant and monitory. Of that recently departed age one might say, as Talleyrand was to remark of prerevolutionary France, that those who had not lived then could not fully know life’s pleasure (“Qui n’a pas vécu dans les années voisines de 1787 ne sait pas ce que c’est le plaisir de vivre”). Indeed, the remembrance and re-creation of the High T’ang’s aura of sun-splashed joys and extravagance was to become, by the ninth century, a topical subgenre of T’ang poetry. As those who had lived through the events in their youth were one and all taken from life, and as the central government continued to remake itself, it became possible to invest the subject with even more consequence. One could choose to focus on the theme of dynastic fortunes sliding from the peak of pomp and prosperity to the pit of dethronement and civil war, or on the overweening pleasures of the court as opposed to the crushing miseries of the plebs, or on Hsüan-tsung’s besotted infatuation with his voluptuous Lady Yang. Political or personal, this was a tale impossible to ignore. How it happened (and, to a lesser degree, why) was a subject that attracted scholars habitually till the final fall of the T’ang in 907. There are over a hundred extant poems, both shih 詩 and fu 賦, on different aspects of this topic, dating from the late eighth century to the end of the dynasty. Of course Po Chü-i’s 白居易 (772–846) “Song of Lasting Regret” (Ch’anghen ko 長恨歌), written in early 807,2 and Yüan Chen’s 元稹 (779–831) “Lyric 1  Wei’s dozen or so poems on this topic set the tone for later treatments. Especially interesting are the following works, cited to Ch’üan T’ang shih 全唐詩 (Peking: Chung-hua shu-chü, 1960; hereafter CTS): “Wen-ch’üan hsing” 溫泉行 (194.2001), “Pai-sha t’ing feng Wu sou ko” 白沙亭逢吳叟歌 (195.2004), “Li Shan hsing” 驪山行 (195.2005). Echoes of certain of his phrasing and imagery can be discovered in Po Chü-i’s “Song of Lasting Regret,” revealing Po’s indebtedness to him. Even the phrase ch’ang-hen 長恨, “lasting regret,” in its emotional coloring, seems to have been borrowed from Wei Ying-wu by Po; see the final line of the former’s “Hsing-lu nan” 行路難 (194.1998), a poem in which a pair of linked bracelets symbolize a past love. Although one may think of Tu Fu as the first poet to look back longingly at Hsüantsung’s reign, this is neither accurate (some other contemporaries of his did so also) nor did his words strongly influence those of later T’ang poets. 2  Among the many English versions of the poem, I cannot but recommend my own rendering, with commentary, in “Po Chü-i’s ‘Song of Lasting Regret’: A New Translation,” T’ang Studies 8–9 (1990–91): 97–105; slightly revised version (with some notes abridged) reprinted in The

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1362

Kroll

on the Lien-ch’ang Palace” (Lien-ch’ang-kung tz’u 連昌宮辭), written in 816, are the most famous, popular—and melodramatic—of such poems. But they are not necessarily the most interesting, either as verse or as historical reflection. The hundred other poems reveal more, in various ways, of how ninthcentury writers recalled Hsüan-tsung’s era. The majority of them focus on one of two sites—either the hot-springs resort of the Palace of Floriate Clarity (Hua-ch’ing kung 華清宮) on Blackhorse Mountain (Li shan 驪山), twenty-five miles east of Ch’ang-an, where the emperor usually removed with the court for a portion of the winter months and which was celebrated as the place of his most notorious idylls with Lady Yang, or the post-station at Ma-wei 馬嵬, thirty miles west of the capital, where on 15 July 756 Hsüan-tsung was forced to have his paramour strangled in order to secure the continuing loyalty of the troops accompanying him on his inglorious flight from An Lu-shan’s rebel forces. I have discussed many of these works elsewhere, and presently I will turn our attention closely to the single most remarkable composition among them.3 It is worth noting here, incidentally, that twice during the ninth century the topic of “Rainbow Skirts and Feather Vestments” (Ni-shang yü-i 霓裳羽衣), the musical suite to which Yang Kuei-fei particularly loved to dance, was set as a composition subject in the prestigious chin-shih 進士 examination, once (in 837) as a shih-poem, once (in 838) as a fu. The subject was expected to elicit images of Hsüan-tsung’s magnificent court as well as measured deprecation of the monarch’s neglect of proper rule during the T’ien-pao years. This attests to an official assimilation of the topic into bureaucratic morality in the late T’ang. It also highlights the personal interest of the reigning emperor, Wen-tsung 文宗 (reg. 827–40), who wished to revive at his court the music of Hsüan-tsung’s era. Poems explicitly dealing with historical themes cannot help but involve us in issues beyond those of verbal artistry and literary appreciation. While it is rarely safe to cling too tightly to so-called disciplinary divisions— derived from the questionable subject-matter designations found in modern Columbia Anthology of Traditional Chinese Literature, ed. V. Mair (New York: Columbia Univ. Press, 1994), 478–85. 3  A convenient selection of such poems from the T’ang, though not a comprehensive collection, is Chin Chi-ts’ang 靳極蒼, “Ch’ang-hen ko” chi t’ung t’i-ts’ai shih hsiang-chieh 長恨歌及 同題材詩詳解 (Cheng-chou: Chung-chou ku-chi ch’u-pan-she, 1989). An earlier selection of texts including prose works as well as poetry from the T’ang to the Ch’ing is Hu Feng-tan’s 胡鳳丹 1877 Ma-wei chih 馬嵬志, recently republished as Ma-wei chih: T’ang Ming-huang Yang Kuei-fei shih-chi 唐明皇楊貴妃事跡, ed. Yen Chung-i 嚴仲義 (Nanking: Chiang-su ku-chi ch’u-pan-she, 1990). Hu simply gathers quotations, providing no commentary or annotation, but he brings together many items that one might otherwise overlook.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1363

universities—when studying any aspect of premodern Chinese (or Western) culture, the boundary between “literature” and “history” is doubtless among the most permeable. Of course, we know that from early times the Chinese have displayed an ardor and an anxiety with regard both to the production and preservation of historical records that is unmatched in continuity by any other civilization. The stately progression of twenty-five “official” histories, which even in modern printings occupies row upon row in our libraries, is but the most visible (because imperially validated) expression of this multi-millennial concern. And the impressive testimony of these standard or “orthodox” histories (cheng-shih 正史), compiled for the most part by succeeding generations of scholars employed in the imperial Office of Historiography, floats upon a broad sea of other texts swept by additional or competing currents that usually remain beneath the surface of the sanctioned narratives. It is to some of these unofficial or submerged histories of the High T’ang, written during the ninth century, that we may turn for more evidence of how Hsüan-tsung’s era was represented in retrospect. The compiling of such works, largely anecdotal in nature, was a characteristic of ninth-century scholarship that is particularly relevant for our purposes here.4 These texts enhance and complicate our look backward. To a great extent, the attitudes suggested in them came to dominate Chinese thought about Hsüan-tsung’s reign in later centuries. Ssu-ma Kuang 司馬光 (1019–1086), for instance, was to draw on them heavily in the T’ang chapters (originally drafted by Fan Tsu-yü 范祖禹) of his comprehensive chronicle, Tzu-chih t’ung-chien 資治通鑑. These unofficial collections, often claiming to record memories passed down orally since Hsüan-tsung’s time or to transcribe items noted in obscure sources, contain accounts in varying length and detail about private pastimes, public celebrations, personal foibles and desires, objects of art and of addiction, fashions in food, drink, dress, and recreation, uncensored conversations and critiques from high and low—in short, everything one wanted to know about the real life of the period but which was or would be left out of official records. The fact that much of it was based on rumor and oral tradition only increases its interest, human nature being what it is. The great T’ang historiographer Liu Chih-chi 劉知幾 (661–721) classified texts of this sort among what he calls “disparate accounts” (tsa-shu 雜述) in 4  I do not, of course, mean to suggest that unofficial histories and anecdotal collections were in any way a new scholarly development in the ninth century—far from it. However, the textual remains of such works from the ninth century are relatively more numerous and generally less fragmentary than those of earlier centuries.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1364

Kroll

his Shih-t’ung 史通. The third of ten kinds of tsa-shu that Liu comments on are the “[records of] unattached, or misplaced, events” (i-shih 逸事). He says, “The responsibility of state historians is to record both events and speech, but since they cannot see and hear everything, there must be collateral items that have escaped their attention. In this regard, scholars interested in such things supply the deficiencies [in the historical record]… these are what we call i-shih.”5 He also tells us that the authors of such works, who are usually from a generation or more after the events they narrate, “search through odd tales to augment and make addition to fact, making use of whatever doubtful items they come upon, haphazardly copying down rumor and hearsay without making a critical assessment or selection. Because of this, reality and fiction 真偽 may not be distinguished, and the true and the false are jumbled together.”6 Nevertheless, Liu concludes, the scholar who wishes the broadest view possible of the past will necessarily consult such works (bringing with him a critical eye), for the standard histories alone do not show the whole picture.7 There are nearly two dozen works of this nature, pertaining specifically to the great days of the K’ai-yüan and T’ien-pao eras, still extant today. In what follows we shall make abundant use of them. A suggestive point of intersection of the two tendencies referred to above— the interest of ninth-century scholars in versifying about Hsüan-tsung’s golden era and in collecting forgotten anecdotes about the period—is represented by the poem called “The Chin-yang Gate” (Chin-yang men 津陽門), written by Cheng Yü 鄭嵎 in late 851 or early 852. This heptametric poem comprises two hundred lines. It is thus nearly as long as Po Chü-i’s “Lasting Regret” and Yüan Chen’s “Lien-ch’ang Palace” combined (the former being 120 lines, the latter 90). And whereas those two poems are composed in stanzas reflecting frequent rhyme-changes (the chief stanzaic blocks being AABA quatrains and independent rhyming couplets), Cheng Yü adheres to one rhyme-group throughout his poem, a real tour de force.8 “The Chin-yang Gate” is one of the 5  Shih-t’ung t’ung-shih 史通通釋, comm. P’u Ch’i-lung 蒲起龍 (1679–1762) (Shanghai: Shanghai ku-chi ch’u-pan-she, 1978), 10.274. 6  Ibid., 10.275. 7  Ibid., 10.276. 8  Cheng’s rhymes are across the 支, 脂, 之, and 微 categories of the Kuang-yün 廣韻, belonging to the 止 she (攝). This is normal t’ung-yung 同用 practice in “old-style” (ku-t’i 古體) verse, although rhyming practice in “recent-style” (chin-t’i 近體) poetry separates out the 微 category. Ninety-seven different rhyme-words are used by Cheng in the poem’s one hundred rhyming positions, there being only three repeats: 旗 in lines 2 and 60, 輝 in lines 54 and 94, and 帷 in lines 76 and 160. For a study of the poem’s rhymes, see Hsü Shih-ying 許世瑛, “Lun

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1365

longest of all T’ang shih and by far the longest of all retrospective poems on Hsüan-tsung’s reign. Besides, Cheng Yü has also provided the poem with his own interlinear prose commentary. This auto-commentary, running to more than 2,200 characters (half again as long as the poem’s 1,400 characters), explains and expands upon the images presented in the poem and is at pains to offer the information that Cheng Yü considers necessary to put his verses in their historically accurate context. The furnishing of a self-commentary to a long poem was a practice that some writers were beginning to exploit occasionally in the first half of the ninth century. Examples one may point to include Po Chü-i’s eighty-eight-line “Song of the Rainbow Skirts and Feathered Vestments” (for more on which, see appendix 1),9 or Yüan Chen’s hundred-line poem to his maternal cousin Hu Ling-chih 胡靈之,10 or, perhaps most interestingly, Li Te-yü’s 李德裕 (787–850) eighty-line poem called “Retelling a Dream,” in which the author reconstructs from remembered fragments a lengthy poem he says he had composed while dozing (shades of Coleridge);11 but in none of these does the self-commentary overmatch the poem itself. Thus we see that just as Cheng Yü’s poem outdoes in length every similar work previously done on the general topic it treats, the extent of its inserted commentary also outdoes all its forerunners of any sort. In several respects it may be considered a culminating composition, containing much more historical detail than earlier ones and surpassing them in the sustained quality of its writing. It is unfortunate, therefore, that it has been largely neglected by scholars for centuries.

Poem and Commentary

We shall have something to say later about the rather shadowy figure of Cheng Yü himself, but let us first introduce the particulars of the poem. An annotated translation of this most substantial of all the poems on the glory and fall of Hsüan-tsung’s reign, and of its interlinear commentary, is offered below, for it has not previously been translated into any language. Before we immerse Cheng Yü ‘Chin-yang men shih’ yung-yün” 論鄭嵎【津陽門詩】用韻, Yu-shih hsüehchih 幼獅學志 7.4 (1968). 9  “Ni-shang yü-i ko,” CTS, 444.4970–71. Yüan’s “Lien-ch’ang kung tz’u” also includes some interlinear commentary by the poet. 10  “Ta i-hsiung Hu Ling-chih chien-chi, wu-shih yün, ping hsü,” CTS 406.4523–24. 11  “Shu-meng shih, ssu-shih yün, ping hsü,” CTS 475.5390–91.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1366

Kroll

ourselves in it, it will be useful to survey what the text includes. Cheng Yü begins with a prose preface of 122 words in which he explains the title (identifying the Chin-yang Gate as the main portal on the north side of the Hua-ch’ing palace complex at Mount Li, looking down on the east-west road that ran to the capital) and tells that in the years 836–41 he frequently took various selections of historical documents and ensconced himself to study them in the quiet of the Shih-weng 石甕 (Stone Urn) monastery on the northeast side of Mount Li (or Blackhorse Mountain). This alerts us that Cheng has a personal attachment to this area as well as specialized knowledge of the past. Most importantly, the preface sets out a frame-story for the poem. This frame-story says that “during the winter of this present year” (which was 851; see below) he was on his way from Kuo 虢 to Ch’ang-an and stopped for the night at an inn located at the foot of Mount Li. The innkeeper was a man of many years who, as it turned out, had been a palace guard when young at the court of the great emperor Hsüan-tsung, and he regaled our poet deep into the night with reminiscences of that bygone time. The next day, as Cheng Yü, resumed his journey, he “swiftly shaped and sculpted the talk of this plain old man,” making of it a heptametric poem in a hundred rhymes (i.e., two hundred lines; as mentioned, the poem uses only one rhyme-group throughout). Note the open recognition in these words of the poet’s own, necessary role in formally fashioning the old man’s recollections. Now, of course, the story is unlikely if not impossible on its face: the old man would need to have been at least 110 years old by this time, to have served at Hsüan-tsung’s court. But this is a simple staging ploy, which had been used previously—most famously by Yüan Chen in his “Lyric on the Lien-ch’ang Palace,” which is told largely in the words of an old man who speaks to that poet (a generation before Cheng Yü’s day) of his experiences as a servant more than half a century earlier in Hsüan-tsung’s temporary palace of Liench’ang.12 Thus, the poet’s lines are cast as the faithful memories of an eyewitness. In Cheng Yü’s case, the position of his interlocutor as a former palace 12  Other impossibly old interlocutors are not unknown in T’ang narrative verse. Cf. Ts’ui Hao’s 崔顥 (chin-shih 723, ob. 754) 38–line “Chiang-p’an lao-jen ch’ou,” CTS, 130.1324–25, in which we find an old man who claims to have seen when young the change of dynasties from Liang to Ch’en in 557, i.e., some 170 or more years previously (though he states, in line 32, that he is now “one hundred five,” which we should probably understand as 150—still not old enough to have witnessed the fall of Liang). As to prose works, we might cite the purportedly ninety-eight-year-old survivor of Hsüan-tsung’s reign who is the narrative center of the ninth-century ch’uan-ch’i 傳奇 story “Tung-ch’eng lao-fu chuan” 東城老父 傳; this tale concerns itself with events of Hsüan-tsung’s time (and the years immediately following) other than those described in Cheng Yü’s poem. For a study and translation of

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1367

guard allows the poet to bring into our view a very wide range of persons and activities—much of which information must have been gleaned, one assumes, from Cheng’s diligent reading when shut away at the Shih-weng monastery. When we turn to the poem itself, we find that seven-eighths of the lines—from verse 15 through 188—are spoken by the ancient innkeeper, only twenty-six lines being spoken by the poet. The narrative outline of the work is as follows. Lines 1 through 24 set the scene of the wet, snowy night that brings Cheng to the “old puffer’s” tavernhouse. The old graybeard (he is not named) begins speaking in line 15. He states first how in his lusty youth, an orphan at fifteen, he became a member of the imperial Yü-lin 羽林 guard-troop. With line 25 he begins his specific recollections of Hsüan-tsung’s court. Lines 25 through 108, nearly half the poem, center on the Hua-ch’ing compound and those who were privileged to enjoy themselves there in the monarch’s entourage. We meet the emperor’s princely brothers13 on their royal hunts (ll. 35–44), then visit the luxuriously fashioned thermal pools that were the centerpiece of the palace grounds (ll. 45–52). From lines 54 through 76 the focus is on the extravagant doings and gorgeous trappings of Lady Yang’s indulged relatives, specifically her cousin Yang Kuo-chung 楊國忠 who held sway in the government from 753 to 756 and her three older sisters who were advantaged notoriously by her preferment. The Precious Consort herself does not make an appearance till as late as lines 83–84, where she is seen playing her rosewood lute in accompaniment to the emperor’s purple cross-flute. This single couplet is, surprisingly, the only place in the poem where Yang Kuei-fei is seen alive. Later, she will reappear only as a blood-clotted corpse (l. 131), and, a bit further on (ll. 143–44), as moldering flesh to be reburied, ornamented by a still-fragrant perfume sachet. In marked contrast with Po Chü-i’s “Ch’anghen ko,” in which she is the dominant presence, Lady Yang hardly makes it on stage in Cheng Yü’s poem, for his interest is not in the great romance: he wishes to encompass a larger scene than that. (Indeed, Cheng Yü’s relative lack of attention to the emperor’s romance, along with his seeming disdain of the historical innacuracies in Po Chü-i’s poem [see below], which had by the mid-ninth century become widely known and very popular and which had focused so centrally on the love affair, may have been a factor in the neglect of the tale, see Robert Joe Cutter, “History and ‘The Old Man of the Eastern Wall,’ ” Journal of the American Oriental Society 106 (1986): 503–28. 13  Since two of these who are mentioned died in 726 and 729, respectively, and our witness was a teenager when he joined the guard, he would—if reality be required—actually have to be approximately 140 years old when Cheng Yü encountered him!

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1368

Kroll

Cheng Yü’s poem shown by later critics: it was not as dramatically appealing as “The Song of Lasting Regret.”) Next (ll. 85–92) we are presented with various sites of the Hua-ch’ing Palace and Mount Li, and with the seemingly magical feats of the Taoist master Lo Kung-yüan 羅公遠 and Buddhist monk Vajratripiṭaka 金剛三藏 as they tried to discomfit each other before the emperor. Following on rather logically from this, we have eight lines (93–100) about Hsüan-tsung’s fabled trip to the moon and how he learned there the music that would come to be used for the “Rainbow Skirts” suite. Then there are eight lines (101–8) describing the magnificent Festival of a Thousand Autumns (Ch’ien-ch’iu chieh 千秋節), the birthday celebration for Hsüan-tsung, held annually at the Hsing-ch’ing 興慶 palace in Ch’ang-an. An Lu-shan makes his appearance, “attend[ing] at the autocrat’s side,” in line 109 and he occupies our attention for fourteen lines, culminating in the emperor’s flight from the capital and the sad events at the Ma-wei poststation (ll. 123–36). Cheng’s account of the latter episode is more circumstantial than Po Chü-i’s quick snapshot. Lines 137 through 160 tell of Hsüan-tsung’s return to Ch’ang-an as emeritus emperor a year later, the reburial of Yang Kueifei, the now poignant ruin of many beautiful places, animals, and people that had thrived during the marvelous years of K’ai-yüan and T’ien-pao, and finally the death of the great monarch and the eventual dispersal of items plundered from the imperial household. The last forty lines of the poem bring us back to the present time. First there is a relatively long section (through line 184) in which we hear of the current, wasted condition of the Shih-weng monastery, owing to the Buddhist persecution carried out under Wu-tsung during the recent Hui-ch’ang 會昌 era (841–47). Although it is still ostensibly the old man speaking, we know from the preface the personal connection that Cheng Yü himself had with this retreat during the K’ai-ch’eng era immediately preceding the Hui-ch’ang period. Its decline seems to be another indicator of the unkind drift of the times. The graybeard then ends: The route that lies before me now is still of an unknown length; I can merely live out whatever is left in failing vigor and health. To encounter you, sir, and speak of all this is spilling tears in vain, Yet I remember the joys and delight of a time that will not be met with again. Choosing not to close on this rueful note, the poet tries to hearten the old fellow by assuring him that the present ruler is returning us to the great days of yore, and that he shall therefore have the unwonted good fortune of living twice in

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1369

one lifetime under a true sage-king. Clearly, given this statement, Cheng Yü’s angle of admiration (or ironic criticism?) for Hsüan-tsung is equal to that from which he sees his own monarch. This politically safe, or personally anxious, finale reminds one of the comparably low bow to the reigning monarch that concludes Yüan Chen’s Lien-ch’ang poem.14 This may strike us as somewhat unsatisfying, but it is a legitimate denouement. It may even put us in mind of the similarly didactic endings to some Han fu. Among his words of intended comfort to the aged innkeeper, the poet refers (ll. 193–96) to a specific event that allows us to date the poem quite precisely. This is the recent (“yesternight”) conquest of the Kansu corridor, retaken from the Tibetans—who had held it since 76315—by Chang I-ch’ao 張義朝 in the tenth month (28 October–26 November) of 851. The triumph was reported to the throne sometime in November and was widely regarded as a sign of T’ang military resurgence (though it was not to be a lasting victory). Accordingly, the “winter of the present year”—the same year that had just witnessed Chang I-ch’ao’s success—must point to December 851 or January 852, the final months of that year.16 As we shall see, this was also the year in which Cheng Yü took his chin-shih degree.17 The above summary of the poem gives only half the picture—that is, only the two hundred lines of verse. We could pause here to examine the stylistic and imagistic niceties of this composition, for it is a virtuoso accomplishment and fully deserves scrutiny for its merits as a poem. But we must not lose sight of the other half of the picture, namely the extended prose commentary with 14  An earlier T’ang use of such a conclusion is in Wang Ch’ang-ling’s 王昌齡 (ca. 690– ca. 756, c.s. 727) poem “Tai Fu-feng chu-jen ta,” CTS, 140.1425–26, in which a military veteran of the frontier campaigns, who has given his whole life to the army and finds no family or friends alive when he finally returns home, nevertheless lauds the reigning Son of Heaven (Hsüan-tsung) for whose glory his years have been spent. 15  It had been part of their booty from helping the T’ang defeat the rebel forces of An Lu-shan and Shih Ssu-ming, hence there is a significant historical tie here to Hsüantsung’s day. 16  For more details and necessary documentation on this, see nn. 230–32 to the translation below. 17  This latter fact seems to weigh most heavily for some scholars who insist on having Cheng Yü write the poem in late 850, while on his way to Ch’ang-an to sit for the examination of spring 851. See, e.g., Wu Tsai-ch’ing 吳在慶 and Fu Hsüan-ts’ung 傅璇琮, T’ang Wu-tai wen-hsüeh pien-nien shih: Wan-T-ang chüan 唐五代文學編年史: 晚唐卷 (Shen-yang: Liao-hai ch’u-pan-she, 1998), 327–28. However, there is no getting around the more important fact that Chang I-ch’ao’s great victory in the northwest, clearly referred to in the poem, occurred late in 851.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1370

Kroll

which Cheng Yü glosses the verse at regular intervals. This commentary is not meant to be read after the poem but along with it. Once Cheng’s self-annotation starts (after line 24), the remarks come after almost every quatrain, sometimes even after successive couplets: commentary follows lines 24, 28, 32, 34, 44, 48, 60, 68, 72, 76, 80, 84, 88, 92, 100, 108, 112, 116, 118, 120, 124, 128, 136, 140, 144, 148, 152, 156, 172, 176, 180, and 184. As mentioned above, the interlinear annotation runs to more than 2,200 words, compared with the 1,400 words of the poem. Cheng Yü’s passion for history is at least as keen as his taste for poetry. The commentary is well integrated into the fabric of the work and does not seem an intrusion. Indeed, the work as a whole should be seen most properly as a hybrid historical account. More of this anon. For the moment, let us note what sort of information Cheng Yü places in his commentary. Very few of the comments are simple scholia merely meant to identify potentially unfamiliar terms or names. Instead, Cheng Yü seems to relish opportunities to offer additional facts or causeries about people and places, almost as though he is compiling his own anecdotal collection. These remarks usually have a vividness and a personal quality to them that bring to life the Realien of the past. To adduce a small selection of some of the more curious items, there is information on the hunting hawks of Hsüan-tsung’s brothers Li Ch’eng-i 李成義 and Li Fan 李範 (after l. 44); on the master builder of the Lady of Kuo’s country estate (after l. 72); on Taoist and Buddhist thaumaturgy (after 1. 92); on Hsüan-tsung’s moon-journey and music-making, being a version of this incident different from others (after l. 100); on the specific favors and indulgences granted to An Lu-shan (after ll. 112, 116); on the reunion of Hsüan-tsung and the new emperor, his son Su-tsung, following the T’ang recapture of Ch’ang-an, and the nice matter of precedence touching on who should enter the capital gates first (after l. 140);18 on the exhumation and reburial of Yang Kuei-fei by the emperor’s eunuch confidant Kao Li-shih 高力士, who had earlier carried out the order to strangle her (after l. 144); on a pet parrot of Lady Yang, a pet deer of Hsüan-tsung, and the otherwise forgotten Princess Yen-sa 罨颯公主 who was fond of sleeping late (after l. 152); plus a harvest of obscure but fascinating material on the history and practices of the Stone Urn Monastery (after ll. 172, 176, 184). Reference is made in several places to the famous tune called “Melody of the Waters” (Shui-tiao 水調) that had special significance for Hsüan-tsung. One one occasion (after l. 148) Cheng Yü sternly takes to task Po Chü-i for the latter’s “egregious mistake” in portraying the Hall of Protracted Life (Ch’ang-sheng tien 18  See Appendix 2 below for a translation of the Tzu-chih t’ung-chien 資治通鑑 account of this unusual event.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1371

長生殿) on Mount Li as the emperor’s sleeping chamber when it was actually

a hall for ritual purifications; obviously it would have been easy for Cheng to point out more of Po’s errors in the fanciful “Ch’ang-hen ko.” Cheng Yü does not cite the sources of his information, but some of his comments seem to be culled and reworded from texts we can identify. Chief among these is the Ming-huang tsa-lu 明皇雜錄 (Disparate Records of the Illuminate Monarch) of Cheng Ch’u-hui 鄭處誨, a relative of Cheng Yü. This text, which has not been preserved intact, is one of the more engaging and reliable unofficial histories pertaining to Hsüan-tsung’s time. The author was the son of Cheng Huan 鄭澣 (c.s. 794) and grandson of Cheng Yü-ch’ing 鄭餘慶 (746–820), both of whom were well-known officials and scholars.19 Cheng Ch’u-hui passed the chin-shih exam in 834 and went on to a successful bureaucratic career himself.20 Like his father, who had compiled the twenty-chüan Ching-shih yao-lu 經史要錄, he had an interest in history. He completed the Ming-huang tsa-lu in 845.21 One of his motives for writing it was discontent with the veracity of Li Te-yü’s Tz’u Liu-shih chiu-wen 次柳氏舊聞, a collection of anecdotes about Hsüan-tsung’s court that was presented to the throne in 834 and purported to be based on several layers of oral transmission stretching back to Kao Li-shih.22 Cheng Ch’u-hui’s work seems to have been well received. It stimulated the compiling of even more such collections recalling the dynasty’s halcyon days: most of the extant collections focusing on Hsüan-tsung’s reign postdate the Ming-huang tsa-lu. While retrospective poems on this topic, as we have seen, began to be written within a generation after the T’ien-pao years, a discernible 19  For their biographies, see Chiu T’ang shu 舊唐書 (Peking: Chung-hua shu-chü, 1974; hereafter cited as ChTS), 158.4163–68, and Hsin T’ang shu 新唐書 (Peking: Chung-hua shuchü, 1974; hereafter cited as HTS), 165.5059–62. 20  Official biographies in ChTS 158.4168–69 and HTS 165.5062. 21   Ho-yin Ssu-k’u ch’üan-shu tsung-mu t’i-yao chi Ssu-k’u wei-shou shu-mu chin-hui shu-mu 合印四庫全書總目提要及四庫未收書目禁燬書目 (rpt. Taipei: Shang-wu yin-shukuan, 1978), 27.2888. 22  Li Te-yü’s prefatory remarks say that the incidents he recorded came to him from his father Li Chi-fu 李吉甫, who had them from an official colleague Liu Mian 柳冕, who was the son of the mid-eighth-century court historian Liu Fang 柳芳, who was told the tales by Kao Li-shih sometime during the Shang-yüan 上元 period (760–62). See Tz’u Liushih chiu-wen, 1, in K’ai-yüan Tien-pao i-shih shih-chung 開元天寶遺事十種 (Shanghai: Shang-hai ku-chi ch’u-pan-she, 1985); also ibid., 27.2886. Political conflict between Li Te-yü’s party and that of the Chengs may also have played a role in Ch’u-hui’s discontent with Li’s work.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1372

Kroll

urge to assemble historical or pseudo-historical remembrances of the period begins only in the second quarter of the ninth century. It then continues, and indeed escalates, for the next hundred years. This corresponds with an understandable human impulse and need. As the last long-lived survivors of a momentous age go to their graves, perhaps sixty or seventy years following the events, there is often anxiety and a consequent desire to “catch” or “fix” the memories of that time before only imagination remains. We might think, for instance, of the increased interest in the history of the Civil War shown in the United States from about 1930 onward, or our recent reattachment to the history of World War II. In ninth-century China, Cheng Yü’s poem and commentary, both of which are at pains to suggest a fidelity to history, ride an early wave of the topical tide that began to swell in about the 830s. Perhaps Cheng Yü’s fondness for historical records, which, he says in his preface, brought him often in 836–41 to periods of concentrated study at the Stone Urn Monastery, was excited at least partly by discussions with his kinsman Cheng Ch’u-hui who would then have been gathering material for what would be his Ming-huang tsa-lu. This is speculation; but it is quite clear that more than a few portions of Cheng Yü’s commentary are beholden to Cheng Ch’u-hui’s work (though not to his work only). The most notable and extended examples of this are the descriptions of the Hua-ch’ing bathing pools (after l. 48) and the emperor’s birthday entertainments, including especially the splendid vignette of the dancing horses and their sad end (after l. 108). Other instances will be apparent in the notes to the translation below.23 An ancillary factor during the generation of Cheng Ch’uhui and Cheng Yü that may have encouraged compilations of the kind we have been considering is the mid-830s effort by the reigning sovereign, Wen-tsung,

23  Given the somewhat checkered textual history of the Ming-huang tsa-lu, which has resulted in the existence of several different editions as well as a pieh-lu 別錄 and groups of “pu-i” 補遺 and “i-wen” 逸文, is it possible that segments of Cheng Yü’s commentary might have been borrowed to fill gaps in later editions of Cheng Ch’u-hui’s work and that therefore when we now look at Ming-huang tsa-lu we are looking in a few places at “The Chin-yang Gate”? This is a difficult question that probably requires a separate study. To adduce just one point: since the text of Cheng Yü’s composition seems to be relatively stable through time, one would expect that if portions of it had been imported into Minghuang tsa-lu, there would be a rather closer verbal correspondence in the relevant passages than is now displayed between extant editions of Cheng Ch’u-hui’s work and that of Cheng Yü. On the other hand, if the influence ran the other way—Cheng Yü reworking and abridging passages from Cheng Ch’u-hui—we would perhaps see something quite like the lexical and syntactic variations that the texts now do present.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1373

to restock and increase the holdings of the imperial library.24 The possible effects of that endeavor on contemporary scholarship reach into areas that must wait on a different opportunity to study. Author Like Cheng Ch’u-hui, Cheng Yü was a member of the Cheng clan of Hsing-yang 滎陽. The Hsing-yang Chengs, an extended and much-ramified kin-group, was one of the handful of super-elite clans whose status was traditionally considered higher than that of the T’ang imperial family. The influence of the great clans had waned considerably by the mid-ninth century. But female Chengs had long been regarded as valuable marriage partners because of the clan’s prestige, and male Chengs of Hsing-yang figure prominently and often in lists of T’ang chin-shih graduates and among the bureaucratic rosters at court and in the provinces. Indeed, it is of some interest to note that a generation after Cheng Ch’u-hui and Cheng Yü, another chin-shih graduate of the Hsing-yang Chengs compiled a collection of historical anecdotes on Hsüan-tsung’s reign. This was Cheng Ch’i 鄭綮 (ob. 899), who was the last representative of the clan to serve at the T’ang court, being an active player in the political world under Hsi-tsung 僖宗 and, especially, Chao-tsung 昭宗.25 His book, completed around 880, is the K’ai-T’ien ch’uan-hsin chi 開天傳信記 (Accounts of Veracities Passed On from the K’ai[-yüan] and T’ien[-pao] Eras), and it is as useful a work as Ming-huang tsa-lu and “The Chin-yang Gate.” It is likely that, owing to their regular involvement in government matters, the Cheng clan had been able to assemble over the decades their own archive of unofficial historical material on the T’ang court. This would help explain the unusual incidence of works on Hsüan-tsung’s period that all read rather like nei-chuan 內傳, written by three Chengs of Hsing-yang within half a century. In any event, the subject was practically becoming a specialty of the house. Until recently, though, we were not certain of Cheng Yü’s affiliation with a specific choronym (unless one knew of the rather rare compendium of Hsingyang texts to be mentioned in the succeeding subsection). All we seemed to know was that Cheng Yü’s byname (tzu 字) was Pin-kuang 賓光 and that he took the chin-shih degree in the fifth year of the Ta-chung 大中 era, 851. And 24  David L. McMullen, State and Scholars in T’ang China (Cambridge: Cambridge Univ. Press, 1988), 237; Jean-Pierre Drège, Les bibliothèques en Chine au temps des manuscrits ( jusqu’au au Xe siècle) (Paris: École française d’Extrême-Orient, 1991), 68. 25  See ChTS 179.4662–63 and HTS 183.5384–85 for his official biographies.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1374

Kroll

the earliest surviving source of that information was the mid-twelfth century Chün-chai tu-shu chih 郡齋讀書志 of Ch’ao Kung-wu 晁公武.26 The little else suggested about him by those few scholars who bothered to take notice of him and his one surviving composition was conjecture. However, examination of the recently published rubbings of T’ang tomb inscriptions from the Ch’ien T’ang chih-chai 千唐誌齋 collection27 has revealed an epitaph for a woman who was Cheng Yü’s elder sister, and this text furnishes some tantalizing particulars about his ancestry and his career. The epitaph is no. 1130 in the two-volume edition of the Ch’ien T’ang chih-chai texts;28 it has also been transcribed, punctuated, and printed in the T’ang-tai mu-chih huipien 唐代墓誌彙編.29 The lady in question died on 15 December 854, and the epitaph was composed by her eldest son, Li Shu 李述. Her name was Cheng Kuan 鄭琯, of the Hsing-yang Chengs, and she was the widow of Li Kung-tu 李公度, sometime magistrate (ling 令) of Ying-shang district 潁上縣, Yingchou, who had predeceased her by two years.30 She was in her sixty-fourth year at the time of her death; hence she was born in approximately 791. Most interesting to us is the passage in the inscription that says: There is a younger brother of hers, named Yü 嵎. When very young he became enthralled by the classics and histories, and when older was proficient in literature. He passed the chin-shih examination with a high ranking. He was subsequently commissioned as coadjutor militant (ts’anchün 參軍) on the staff of the governor-general of Yang-chou 楊州. This gives us just enough information to recognize our man and to wish for more. That, however, is all that is said about Cheng Yü. At least it is plain from this that his early passion for history was notable (despite the somewhat formulaic quality of such a statement), that he was still alive in 855, and that he was then serving in Yang-chou. Although we are not told his age or date of birth, we may surmise that he came into the world sometime after 791, when 26   B SS edn., 4B.398. [See list of abbreviations at the end of the article]. 27  For an exemplary study of how these valuable materials can be used as complements or correctives to the official histories, and what their limitations are, see David L. McMullen, “The Death of Chou Li-chen: Imperially Ordered Suicide or Natural Causes?” Asia Major 3rd ser., 2.2 (1989): 23–82. 28   Ch’ien T’ang chih-chai ts’ang-chih 千唐誌齋藏誌 (rpt. Peking: Wen-wu ch’u-pan-she, 1989). 29  Ed. Chou Shao-liang 周紹良 (Shanghai: Shang-hai ku-chi ch’u-pan-she, 1992), 2320. 30  Li Kung-tu’s epitaph is no. 1120 in Ch’ien T’ang chih-chai ts’ang-chih; also transcribed in T’ang-tai mu-chih hui-pien, 2305.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1375

his elder sister was born, and before roughly 816, for he should have been not much younger than his mid- or late teens (probably a bit older) during his study retreats in the years 836–41 at the Shih-weng monastery. Elsewhere in the inscription we learn that the lady’s—and Cheng Yü’s—father was one Cheng Hung-min 鄭弘敏. He is stated to have been exceptionally well-versed in scholarly (or classicist, ju 儒) concerns as a youth, later to have been a mingching 明經 graduate, and to have served successively as sheriff (wei 尉) in Huat’ing 華亭, Soochow, as sheriff in Hsüan-ch’eng 宣城, Hsüan-chou, and then as district magistrate in T’ang-shan 唐山, Hangchow. The personal name of the paternal grandfather is also given—Huan 寰, but there is no further information about him or any earlier ancestors.31 Apparently this was not one of the more eminent branches of the clan. It would seem that Cheng Yü was a second-generation scholar, who may not have had his father’s bureaucratic abilities but surpassed him in literary gifts. The dating of the poem to the winter following Cheng Yü’s successful examination sitting may prompt another supposition. Might the poem and commentary have been written as a self-conscious display of erudition, meant to be presented to those responsible for making the initial assignments of bureaucratic positions for recent chin-shih graduates? We are well aware of the T’ang practice of “circulating scrolls” (hsing-chüan 行卷) by chin-shih candidates hoping to gain the backing of influential patrons.32 The compositions presented in such fashion were usually poems.33 But passing the examination was simply a first step, merely a foot in the door; it did not guarantee one an immediate appointment to government service. Before being eligible for appointment one had next to pass the hsüan 選 (“selection”) exam, which took account of personal traits—appearance, speech, bearing, calligraphy—as well as scholarly ability.34 And after clearing this hurdle, one might still wait for months before being offered a post. It is not unreasonable to surmise that 31  Some of this material is also summarized by Ch’en Shang-chün 陳尚君and T’ao Min 陶敏 in their respective comments on Cheng Yü, in T’ang ts’ai-tzu chuan chiao-chien 唐才子傳校箋, vol. 5 (Peking: Chung-hua shu-chü, 1995), 386–87. 32  See esp. Victor H. Mair, “Scroll Presentation in the T’ang Dynasty,” Harvard Journal of Asiatic Studies 38 (1978): 35–60. 33  Mair, ibid., 39–40, noting that almost all references to presentation scrolls speak only of poems, discounts the theory advanced by some scholars that mixed portions of verse and prose in presentation pieces played a role in the development of T’ang ch’uan-ch’i. 34  This was normally an individual exam, conducted under the aegis of the Department of State Affairs (shang-shu sheng 尚書省) between the fifth and tenth month of the year. See Robert des Rotours, Le Traité des Examens, traduit de la Nouvelle histoire des T’ang (chap, xliv, xlv) (Paris: Librairie Ernest Leroux, 1932), 42–44.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1376

Kroll

Cheng Yü’s large-scale and virtuoso composition was used by him as a selfadvertisement to spur a quicker and better appointment than he might otherwise receive. The poem’s concluding praise of the current emperor’s excellent administration may thus have also some very personal and hopeful relevance to the author’s own fortunes. Provenance Unlike virtually all other T’ang poems, Cheng Yü’s poem and commentary circulated for centuries as a separate publication in its own right, not as part of a larger collection. The earliest bibliographic mention of it (“ ‘Poem on the Chin-yang Gate,’ by Cheng Yü, in one chüan”) is in the Ch’ung-wen tsung-mu 崇文總目, cataloguing the Sung imperial library of the mid-eleventh century.35 Hsin T’ang shu’s bibliographic monograph, probably borrowing from the former work, records it also,36 as do the two most important catalogues of Southern Sung private libraries, the mid-twelfth-century Chün-chai tu-shu chih37 and the mid-thirteenth-century Chih-chai shu-lu chieh-t’i 直齋書錄解題;38 likewise the bibliographic monograph of the mid-fourteenth-century Sung shih 宋史.39 The latter also records separately a three-chüan work called Piao-chuang lüeh 表狀略 by Cheng Yü,40 which sounds like an epitome of official, memorial writings. But neither this work nor any other by Cheng Yü, apart from “The Chin-yang Gate,” is now extant. At some time before the fall of the T’ang in 907,41 the text of the poem and commentary was carved on a stele erected by the Hung-wen-kuan 弘文館, just outside (i.e., north) of the Chin-yang Gate.42 Sung Min-ch’iu 宋敏求 (1019– 1076), writing in the mid-eleventh century, said that “traces of it exist even at the present day.”43 Lo T’ien-hsiang 駱天驤 refers to it also in the fourteenth 35  Compiled by Wang Yao-ch’en 王堯臣 (1001–1056) et al., between 1034 and 1038. TSCC edn., 5.363. 36   H TS, 60.1613. 37  Compiled by Ch’ao Kung-wu 晁公武, preface dated 1151. See BSS edn., 4B.398. 38  Compiled by Ch’en Chen-sun 陳振孫 (fl. 1235). BSS edn., 19.542–43. 39   Sung shih (Peking: Chung-hua shu-chü, 1985), 208.5342. 40  Ibid., 208.5352. 41  See P. W. Kroll, “The Last Year of T’ang, 907—Not 906,” T’ang Studies 18–19 (2000–01): 107–10. 42  The terminus ante quem is provided by the fact that the Hung-wen-kuan ceased to exist with the end of T’ang. 43   Ch’ang-an chih 長安志 (SKCS), 10.8a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1377

century.44 But no excavations have yet been undertaken around the location of the old Chin-yang Gate, and the stele has not been seen for at least half a millennium. The carving of this stele, placed at a government site, would have been an occasion of some moment and would certainly have enhanced Cheng Yü’s reputation and that of his family. If this event occurred within the three years between the poem’s composition and the death of the poet’s elder sister, it might help explain why he is mentioned in her epitaph as being specially obsessed with history and proficient in literature. Then again, if a composition by a chin-shih graduate of only one or two years previously had been honored by having the work inscribed on such a monument, one might expect Cheng Yü’s name to be mentioned (if even as an object of envy) in some document relating to those years. More likely, the stele was erected posthumously. The earliest extant exemplar to which present-day versions of the poem and commentary may trace their filiation is that included in the twelfth-century T’ang-shih chi-shih 唐詩紀事, compiled by Chi Yu-kung 計有功 (c.s. 1121). The text printed in the 1972 edition of this anthology45 is preferable to and more reliable than the one offered in the more recent T’ang-shih chi-shih chiaochien 唐詩紀事校箋.46 The latter includes useful collation notes but regularly adopts readings from the version of the poem found (sans commentary) in Ho Hsi-wen’s 何谿汶 twelfth-century Chu-chuang shih-hua 竹莊詩話,47 an inferior text that has emended nearly every lectio difficilior. The 1972 printing of T’ang-shih chi-shih was primarily based on Hung P’ien’s 洪楩 1545 edition (later reprinted in Ssu-pu ts’ung-k’an), which was in turn based on Wang Hsi’s 王禧 1234 edition. Its version of Cheng Yü’s poem and commentary is also eclectic, but the variants accepted are relatively few and defensible. In the translation below, I have used this as my base-text, explaining those few occasions where I have chosen to deviate from it. Of course the version of the poem and commentary most easily accessible today is that in the Ch’üan T’ang shih 全唐詩. Although the frequent problems with this collection are well known, in this particular instance the text it reproduces is a trustworthy one and its readings can sometimes correct those of Hung P’ien. One may consult the early Ch’ing collection, T’ang-shih pai-ming-chia ch’üan-chi 唐詩百名家全集, compiled by Hsi Ch’i-yü 席啟寓 (1650–1702), for still another version.

44   Lei-pien Ch’ang-an chih 類編長安志, ed. Huang Yung-nien 黃永年 (Peking: Chung-hua shu-chü, 1990), 10.316. 45   T’ang-shih chi-shih (Hong Kong: Chung-hua shu-chü, 1972; hereafter TSCS), 62.932–37. 46  Ed. Wang Chung-yung 王仲鏞 (Chengtu: Pa-Shu shu-she, 1989), 62.1671–86. 47   S KCS edn., 11.16a–20a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1378

Kroll

Certainly the rarest version of the poem and commentary is that contained in the Hsing-yang tsa-tsu 滎陽雜俎. This compendium, compiled by Ch’eng Ting-yüan 程定遠 during the K’ang-hsi reign-period (1662–1722), is devoted to the works of eight Chengs from Hsing-yang—three dating from the T’ang, one from the Sung, two from the Ming, and two from the early Ch’ing. One of these works is “The Chin-yang Gate” of Cheng Yü (another is Cheng Ch’i’s K’ai-T’ien ch’uan-hsin chi), thus establishing, for any who had the good fortune to encounter this collection prior to publication of the Ch’ien T’ang chih-chai inscriptions, the clan relationship of our Cheng Yü.48

Looking Backward

Works like “The Chin-yang Gate” and the anecdotal histories it resembles force us to examine not only the particular past being evoked, but also our approach to the past itself. In his famous essay, “The Task of Cultural History,” the great Dutch historian Johan Huizinga (1872–1945) averred that the unity which we construct to allow ourselves to examine a determined historical topic “can never lie in an arbitrary slice of past reality itself. The mind selects from tradition certain elements it synthesizes into a historically coherent image, which was not realized in the past as it was lived.”49 From this perspective, I find the increasing ninth-century interest in re-creating images of Hsüan-tsung’s era, in the different gravities of poetry and prose, useful as commentaries on the perpetual human pursuit of memory and invention—the two defining qualities of our being. Coming at the mid-point of a century of retrospective regard for the dynasty’s greatest days, Cheng Yü fashioned a tour de force combining, in a unique way, narrative verse and anecdotal history. But the mixed form of his masterpiece has probably worked against its popularity in later centuries. Neither as melodramatically compelling as Po Chü-i’s “Ch’ang-hen ko” and Yüan Chen’s “Lien-ch’ang-kung tz’u,” nor as discretely instanced as prose collections of unofficial vignettes, it occupies a middle border that does not fold neatly into standard genre categories. Still, love of scholarship tied to love of poetry is not 48  There now seems to be but a sole copy of this collection extant. Chung-kuo ts’ung-shu tsung-lu 中國叢書總錄 (Shanghai: Shang-hai ku-chi ch’u-pan-she, 1959–62), 1: 994, located it forty years ago in the Peking Library. See Addendum. 49  In Huizinga’s Men and Ideas: History, the Middle Ages, the Renaissance, tr. James S. Holmes and Hans van Marle (Princeton: Princeton Univ. Press, 1984), 26.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1379

as irregular a combination in premodern China as in the premodern (or, for that matter, modern and even postmodern) West.50 Cheng Yü’s amalgam in “The Chin-yang Gate” smacks both of the scholar’s desire to tell all he knows and the poet’s urge to charm. As we said, memory and invention. However, it may have been precisely this hybrid, seemingly aberrant form that disinclined scholars of later dynasties to give the work its due. As purely affective verse it fell short of the emotion expressed in “Ch’ang-hen ko,” and as history the commentary was less satisfying than a clearly identified collection of anecdotes. Yet, the carving of the work on the Hung-wen-kuan stele indicates the value that at least some scholars of the latter half of the ninth century attached to it. Although Cheng Yü’s self-commentary first of all glosses his verse, it is not a lexical or grammatical but rather a reflective commentary, shaping a deeper or, often, parallel account of events. In our view of his work, we must stay balanced between two stresses, just as Cheng Yü himself tries to balance his composition, and just as anyone must who would recall the past to devise a present. In this regard I cite the words of Arnaldo Momigliano who, in speaking of the antiquarian scholarship of the fifth-century bc Sophists, reminds us that “An element of play and pastime was inherent in erudition from its inception,” that “erudite pleasure is always ambiguous.”51 This quality applies to Cheng Yü as well. Impulses severally hortative, aesthetic, learned, and ludic are present in his work, and we should be alert to them all. Content to be neither historian merely nor merely a poet, Cheng Yü must sing in verse because of “the universal belief that words arranged in a certain order have a potency denied to words arranged in another order,”52 and must also say in prose because of a need to amplify and inform more freely. The ninth-century fascination with Hsüan-tsung’s era, picturing and reframing it as a time of stories to be savored, of characters vividly defined, betokens a sense of difference between the age described and the current one.53 50  In the West, this unusual alliance is most notable in Alexandrian poetry of the third century bc and French poetry of the sixteenth century. Arnaldo Momigliano, The Classical Foundations of Modern Historiography (Berkeley and Los Angeles: Univ. of California Press, 1990), 64. 51  Ibid., 63. 52  George Sampson, “Truth and Beauty,” in his Seven Essays (Cambridge: Cambridge Univ. Press, 1947), 102–3. 53  For a survey of the changing views of ninth-century poets regarding Hsüan-tsung and Yang Kuei-fei, see Wu Ho-ch’ing 吳河清, “T’ang-jen Ma-wei shih chih wo-chien” 唐人馬 嵬詩之我見, in T’ang-tai wen-hsüeh yen-chiu 唐代文學研究 8 (Kweilin: Kuang-hsi shihfan ta-hsueh ch’u-pan-she, 2000), 179–85.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1380

Kroll

Perhaps this is an additional argument for a periodization that sees the An Lu-shan rebellion as the effective end of medieval China. Ninth-century writers, though only a few generations removed from the K’ai-yüan and T’ien-pao periods, seem to regard that age across a divide rather similar to that which subsequently separated even later centuries from it. This is another way of saying that ninth-century writers fixed the angles from which most scholars for the succeeding millennium would look upon the reign of Hsüan-tsung and characterize it in their works. That reign seemed to include so much, promise so much, achieve so much, even though it ultimately plunged into a swirl of disorder. In the reminiscences of its highest virtues and most sweeping refinements, it may recall to us Walter Pater’s paean to quattrocentesco Italy, in the preface to his celebrated study of The Renaissance: There come, however, from time to time, eras of more favorable conditions, in which the thoughts of men draw nearer together than is their wont, and the many interests of the intellectual world combine in one complete type of general culture. The fifteenth century in Italy (read: the time from 712 to 756 in China) is one of these happier eras; and what is sometimes said of the Age of Pericles is true of that of Lorenzo (read: of Hsüan-tsung)—it is an age productive in personalities, many-sided, centralized, complete. Here, artists and philosophers (read: poets) and those whom the action of the world has elevated and made keen do not live in isolation, but breathe a common air, and catch light and heat from each other’s thoughts. There is a spirit of general elevation and enlightenment in which all alike communicate.54 Of course, our current view of the Renaissance has acquired more shading, become more complicated since Pater’s or Burckhardt’s days; in the same way we continually refine and alter our understanding of eighth-century China.55 As Jacques Le Goff has said, “History is a matter of transformation and memory, memory of a past that continues to live and to change as one society succeeds

54  Walter Horatio Pater, The Renaissance: Studies in Art and Poetry (1873; rpt. New York: Mentor Books, 1963), xv–xvi. 55  An example of this is the way that the discovery and publication of new inscriptional material dating to the T’ang is gradually but surely changing scholarship in both history and literature.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1381

another.”56 When that history or detached pieces of it are represented in verse, additional formal tensions come into play. Ninth-century poets, like ninthcentury collectors of anecdotes, worked out an increasingly detailed and nuanced portrait of Hsüan-tsung’s reign. Cheng Yü’s contribution to this is among the more exacting of such efforts, even as it acknowledges in design and diction its own highly constructed nature. Its long neglect by scholars allows us now to approach it seemingly afresh, almost as though it were a new document. And as we see upon reading it, it yields more than a few original insights. Before turning to a full presentation of the text, I should like to suggest a possible analogy for our—and Cheng Yü’s—recreation of the past. In the Kuoshih pu 國史補, compiled by Li Chao 李肇 (fl. 813) in the early ninth century, we find the following account of how a citizen of Ma-wei 馬嵬 was able to enrich herself through a saved fragment of Precious Consort Yang’s clothing, in the years following the great lady’s tragic death at this place: When he had reached the post-station at Ma-wei on his progress to Shu 蜀, Hsüan-tsung ordered Kao Li-shih57 to strangle the Precious Consort, which he did beneath a pear tree that stood in front of the Buddha hall.58 An elderly shop-mistress from Ma-wei [later] retrieved the upper portion of one of [Lady Yang’s] stockings, of parti-colored silk tabby. Tradition 56  Jacques Le Goff, The Medieval Imagination, tr. Goldhammer (Chicago: Univ. of Chicago Press, 1988), 11. Perhaps I may juxtapose here a suggestion by Nabokov, spoken by Van Veen in the latter’s meditation on time, in Part Four of Ada: “The Past, then, is a constant accumulation of images. It can be easily contemplated and listened to, tested and tasted at random, so that it ceases to mean the orderly alternation of linked events that it does in the large theoretical sense.” This is a nice, if unwitting, description of Chinese anecdotal histories. Ada or Ardor: A Family Chronicle (New York: McGraw-Hill, 1969), 580. 57  Hsüan-tsung’s eunuch confidant; see n. 85 below. 58  The date of this sad event—forced upon Hsüan-tsung by his mutinous troops—was 15 July 756, during the emperor’s flight from Ch’ang-an to Szechwan, after he abandoned the capital to An Lu-shan’s rebel forces. See lines 123–32 of the poem below. The most accurate, as well as quite moving, narrative of these highly charged days, is that in Ssuma Kuang’s 司馬光 (1019–1086) Tzu-chih t’ung-chien 資治通鑑 (Hong Kong: Chung-hua shu-chü, 1976; hereafter TCTC), 218.6970–74, translated, with Ssu-ma Kuang’s “examination of differences” (k’ao-i 考異) and Hu San-hsing’s 胡三省 (1230–1302) commentary, in P. W. Kroll, “The Flight from the Capital and the Death of Precious Consort Yang,” T’ang Studies 3 (1985): 25–53. To depend for one’s knowledge of these events mainly or solely on the fancies of Po Chü-i’s “Ch’ang-hen ko” (as many do even now) is like believing one has the facts of Mozart’s and Salieri’s careers after watching the film Amadeus.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1382

Kroll

has it that, for a fee of a hundred cash per sitting, she would allow visitors to fondle it. The profit she took in from this was, all told, so much that the old woman made a fortune.59 Perhaps that shred of a stocking that once covered Lady Yang’s leg may serve us as an apt symbol for the remains of the past. And the fondling of it by those who desired a momentary connection with departed realities may correspond suggestively to all attempts, scholarly or emotional, at recapturing lost time. (One can hardly avoid thinking here of Proust’s petite madeleine dipped in lemon tea.) More than seven centuries later, this anecdote of Li Chao’s would inspire a scene in Hung Sheng’s 洪昇 (1605–1704) famous dramatic romance, “The Hall of Protracted Life” (Ch’ang-sheng tien 長生殿), the most popular theatrical presentation of Hsüan-tsung’s story in late imperial times.60 In searching out ways to bring the past to life, we deal with intangible words powerful through the ages, fading colors brilliant in imagination, palpable artifacts redolent of sensuality. We grasp at all resources, be they text or textile. In doing so, while we may not become wealthy—as the Ma-wei shopkeeper did—we may still fulfill in ourselves a more deeply human need. For such a purpose, let us now listen to Cheng Yü and the garrulous old gaffer he claims to have encountered one snowy evening in 851. In what follows, I have where necessary added a subcommentary in explanatory notes, to elaborate on many of the interesting matters evoked in Cheng Yü’s wording or references.

The Chin-yang Gate: An Annotated Translation61

Preface. The Chin-yang Gate 津陽門 was the outermost gatetower of the Palace of Floriate Clarity (Hua-ch’ing kung 華清宮). South of it lay the forbidden portal [to the palace compound]; north from it ran the road to the capital. During the K’ai-ch’eng

59   T’ang kuo-shih pu; Yin-hua lu 唐國史補因話錄 (Shanghai: Shang-hai ku-chi ch’upan-she, 1979), 1.19. 60  Hung Sheng took Li Chao’s brief notice and imaginatively expanded it into one of the more provocative scenes in his play. For a translation, see Stephen Owen, An Anthology of Chinese Literature, Beginnings to 1911 (New York: W.W. Norton and Co., 1996), 1063–67. 61   Ch’üan T’ang shih, 567.6561–66; T’ang-shih chi-shih, 62.932–37. On these different editions see discussion above.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1383

開成 era (836–41), it was often my practice to take documents of sundry sorts and shut myself up62 in the Monastic Close of the Stone Urn (Shih-weng seng-yüan 石甕僧院);63

at that place I learned a great deal about the traces of timeworn events. During the winter of this present year, I was returning from Kuo 虢.64 At nightfall, when I had reached the foot of the mountain,65 I loosened my saddle and took thought for a meal. I sought lodging at an inn for travelers, the elderly proprietor of which was far advanced in years.66 He said that he had himself been in service to Ming-huang 明皇.67 As the night drew on and the wine flowed plentifully, he spoke to me at length of the verities of that olden time when the world had enjoyed tranquility. The following day, while on horseback, I swiftly shaped and sculpted the talk of this plain old man, making a poem in long lines of seven syllables. It has 1,400 words in all, since I ended after completing a hundred rhymes.68 I have entitled it with the name of the gate; it runs as follows:

62  “Shut myself up”: the idiom is literally “lower the curtain” (下帷), which carries the implication of “closing oneself off from all outside distractions, for the purpose of concentrated study.” The locus classicus is a passage in the biography of Tung Chung-shu 董仲舒 (176–104 bc) in the Han shu 漢書 (Peking: Chung-hua shu-chü, 1962), 56.2495. 63  The Stone Urn Monastery was located in the northeast area of Li Shan 驪山. The name derived from the Vale of the Stone Urn (Shih-weng ku 谷), so called because of the narrow rock-lined bottleneck through which spurts a mountain stream running down as the waterfall of Stone Fish Cliff (Shih -yü yen 石魚巖), below the Embroidered Ridge of the East (Tung hsiu-ling 東繡嶺). See Lin-t’ung hsien-chih 臨潼縣志 (1776 ed.; rpt. Chungkuo fang-chih ts’ung-shu; hereafter LTHC) 2.36a/b; also Cheng Yü’s commentary following line 176 below, and Sung Min-ch’iu 宋敏求 (1019–79), Ch’ang-an chih 長安志 (SKCS edn.; hereafter CAC), 15.12b–13a. 64  The time is the winter of 851–52; see note 232 below. Kuo-chou 虢州 was in the extreme western part of the present-day province of Honan, somewhat south of modern Ling-pao 靈寶 district, on the route between Lo-yang and Ch’ang-an. 65  Sc. Li Shan 驪山, “Blackhorse Mountain,” on the north side of which the Hua-ch’ing Palace 華清宮 stood. 66  Given the memories the old man claims to have of events nearly a century before, 艾 (ai, “moxa-gray”) cannot here mean “fifty years old” or even “seventy years old” as the word is often defined, but must refer generally to advanced age. 67  “The Illuminate Resplendent One,” or “Augustus Illuminate”—the popular abbreviation of Li Lung-chi’s 李隆基 posthumous title, “Great Paragon of the Ultimate Tao, Greatly Illuminate and Filial August Thearch” (Chih-tao ta-sheng ta-ming hsiao huang-ti 至道 大聖大明孝皇帝). Of course he is more often known by his posthumous temple-name, Hsüan-tsung 玄宗, so designated because of his passion in late years for Taoist teachings. 68  See note 8 above on the poem’s rhymes.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1384

4

8

12

16

20

Kroll

The Chin-yang Gate overlooks to its north a connecting crossway, Where snow and wind whipped and gusted, flurrying the wine-shop flags.69 Cold and muddied, my slow-gaited mount stumbled, advancing no more; My tired servant pulled back to ask, “How are we to go on?” A steward from a tavernhouse hurried out to loosen my packs; Then, before me, on a table, he ranged goblet and flagon in order. The bits and leavings of my copper coins were hardly enough to be counted, But the soothing taste of plain unstrained wine was sweet as sugary syrup. From out the brazier we drew full cups, pledging and toasting each other; Famished stomach, dried lungs lost now the hunger felt since morning. Worries and cares seemed to have slipped out the door and departed, As I felt a sense of springtime gradually spread through my four limbs. Then the graybeard host came to his guest, raised the wick of the ornate lamp; His hunched-over shoulders kept his knees from sight, his crowblack cap was askew. He chortled and said, “This ancient puffer no longer comports with the rites;”70 His snowy sidelocks, in bedraggled tangle, hung down both sides of his  jowl. He asked me, “Where is it you go, braving so the cold and wintry sun? You’d never guess, from my feeble condition, how I was when I was young. For this graybeard once was hale and lusty, though my guest cannot perceive it; Just let me describe it for you, sir, that time of long ago.

69  Cf. K’ai-yüan T’ien-pao i-shih 開元天寶遺事 (hereafter KYTPIS), compiled by Wang Jenyu 王仁裕 (880–956), in KYTPIS shih-chung 十種 (Shanghai: Shang-hai ku-chi ch’u-panshe, 1985), 2.93, on the wine-shops along the road from Li Shan to Ch’ang-an. 70  “Ancient puffer” is t’ai–lao 鮐老, t’ai being the classical term for the globe-fish, swellfish, or puffer (Tetraodontidae Spheroides), now commonly known as the “river pigling” (ho-t’un-yü 河豚魚), all the names suggesting its bloated, unattractive appearance. When he says he “no longer comports with the rites,” he means he shall not stand on ceremony in his conversation with the poet.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

24

1385

The times were tranquil, and the personal guard was called ‘The Forest of Plumes,’71 As for me, I was but ten and five when I became a fatherless lad.72 But in shooting the brownbear or grappling the tiger, none among them was my match; In drawing the bow, passing into court or out, I was with the ‘Apt and  Volant.’73

During the K’ai-yüan 開元 era (713–42), there were no Armies of Divine Tactics (Shents’e chün 神策軍) of the East and of the West. Only men from the “Six Armies” were employed as personal guards [of the emperor].74 71  The “personal guard” (ch’in-wei 親衛) was the emperor’s personal honor-guard and bodyguard. See Robert des Rotours, Traité des fonctionnaires et traité de l’armée, 2nd, rev. ed. (San Francisco: Chinese Materials Center, 1974), 503, and P. W. Kroll, “The Flight from the Capital and the Death of Precious Consort Yang,” 36–37, where I rendered the term more formally as “Proximate Paladins.” “Plume Forest” (yü-lin 羽林) does not here refer to the army group of that old title but to a special company of fifty imperial guards drawn from local militia in service at the capital; see des Rotours, 526. 72  Compare the phrasing in this line and the next with a passage in Han shu 8.260, where Hsüan-ti 宣帝 (reg. 73–48 bc) levies “the bowmen of the Apt and Volant, the fatherless lads of the Plume Forest.” A note by Ju Shun 如淳 (fl. 189–265) comments that the sons of those soldiers who died in service were raised up by the state to a guaranteed post in the Plume Forest army. Cheng Yü has nominally merged the two different Yü-lin groups (see preceding note) of Hsüan-tsung’s time. 73  “Apt and Volant” (tz’u-fei 佽飛) was the name given to a select group of imperial archers, beginning in Han Wu-ti’s 漢武帝 (reg. 140–86 bc) time. The name (variously written as 茲非) is claimed to have come from that of a courageous Yüeh 越 warrior in Chou times. Han shu 8.260. The semanticized phrase “Apt and Volant” or perhaps “Let Fly Aptly” is certainly fitting for a troop of archers; I also intend “volant” here in its old military sense of “so constituted as to be capable of rapid movement” (o.e.d.). Cf. des Rotours, 531. 74  The Shen-ts’e army was formed only in 753, near the end of the T’ien-pao 天寶 (742–56) period. In 765 it was divided into Left and Right wings and quartered in the imperial parks of the capital. From that time on it played an important role in T’ang military history, being under eunuch control. See des Rotours, 565–69, 844–81, esp. 565–66, 845–46. The phrase “Six Armies” (liu-chün 六軍) refers to the capital-based soldiers of the Armies of the Dragon’s Might on the Left and on the Right (tso, yu Lung-wu chün 左右龍武軍) and the Armies of Divine Might on the Left and on the Right (tso, yu Shen-wu chün 左右 神武軍), in addition to the Left and Right Armies of the Plume Forest; des Rotours, 567 n. 3, shows that there is confusion regarding the HTS naming of the Shen-ts’e army as part of the liu-chün, which in any case was not a group of six until at least 757. Ch’en Yin-k’o pointed out in 1950 that references in ninth-century poetry and historical texts to the “Six Armies” as existing in Hsüan-tsung’s reign are anachronistic, numerous though they be.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1386

Kroll

This was the time when first was raised the Tower for Observing the Winds, Its eaves being a hundred feet high, undergrouped by ornate rafters. South of that building next was reared the Great Hall for Cock-fights; 28 The morning light and mountain shadows there were mixed and  intermingled. The Tower for Observing the Winds (Kuan-feng lou 觀風樓) was situated at the northeast point outside the palace proper, connected to the Supreme Interior by means of a subjoined pincer-wall.75 It overlooked, in front, the Fleet-way76 and took in on all sides a view of mountains and streams. During the Pao-ying 寶應 (762–63) era, Yü Ch’ao-en 魚朝恩 had it razed in order to improve his [Monastery of] Emblazoned Reverence.77 Today the ruins of it are yet in evidence. But only the Great Hall for Cock-fights (Chitou tien 雞鬥殿) and the polo fields, one next to the other, remain as they were.78 See his Yüan Po shih chien-cheng kao 元白詩箋證稿, rpt. in Ch’en Yin-k’o hsien-sheng lunwen chi 陳寅恪先生論文集 (Taipei: Chiu-ssu ch’u-pan-she, 1977), 718–20. 75  On the Kuan-feng lou, see CAC 15.9a, LTHC 2.35b. The “Supreme Interior” (shang-nei 上內) refers to the inner grounds of the imperial compound. 76  “Fleet-way” is ch’ih-tao 馳道, the road (or part of the road) reserved for the emperor’s use. “Speedway” is unfortunately not a tenable translation, owing to present-day commercial and automotive connotations. 77  The eunuch Yü Ch’ao-en 魚朝恩 (d. 770) was the most influential and feared figure at court during the first half of Tai-tsung’s 代宗 reign (763–80). He was in command of the Shen-ts’e armies (see n. 74 above). On land that had been granted to him outside the T’ung-hua Gate 通化門 (the northernmost of the gates in the east wall of Ch’ang-an) he ordered built a monastery dedicated to the merit of Tai-tsung’s deceased mother, naming it the Chang-ching (Emblazoned Reverence) Monastery after her posthumous title (Chang-ching huang-hou 章敬皇后). Since there was not in the city sufficient timber of the high quality needed to fulfil Yü’s lavish design, certain of the buildings at the Huach’ing complex, plus others by the famous Serpentine Lake (Ch’ü-chiang ch’ih 曲江池) in the southeast corner of the city, and various other official edifices, were demolished, so that their beams and rafters could be used in the creation of the Chang-ching Monastery. Cheng Yü dates this event to the Pao-ying period (May 762–August 763), but other sources place it in the second year of Ta-li 大曆 (i.e., 767). See ChTS 184.4764, HTS 207.5865, TCTC, 224.7195. 78  See CAC 15.9a/b, LTHC 2.35b, 36a; these sources indicate there were two polo fields, a larger one and a smaller. On polo as an imperial amusement in the T’ang, see the classic studies of Hsiang Ta 向達, T’ang-tai Ch’ang-an yü hsi-yü wen-ming 唐代長安與西域文明 (Peiping: Harvard-Yenching Institute, 1933), 74–81, and Lo Hsiang-lin 羅香林, T’ang-tai wen-hua-shih yen-chiu 唐代文化史研究 (Shanghai: Shang-wu yin-shu-kuan, 1946),

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

32

1387

In the tenth month of that year the adyta’s honor-guard was shifted;79 At the mountain’s foot, set close as comb’s-teeth, diverse administrators were ranged.80 When the Gallery for Homage to the Prime was completed, Lord Lao was seen; As the ‘District of Assembled Glories’ was ’New Prosperity’ recast.

At the time there was an imperial order to change Hsin-feng 新豐 (New Prosperity)81 to Hui-ch’ang 會昌 (Assembled Glories) District, shifting the county seat from the former city of Yin-pieh 陰鱉 to a site below the mountain.82 In the tenth month of the next year Lord Lao appeared, to the south of the Ch’ao-yüan ko 朝元閣 (Gallery for Homage to the Prime, i.e., to Lao-tzu); at that spot was dedicated the Belvedere of the Descended Saint (Chiang-sheng kuan 降聖觀).83 The name of the old Hsin-feng 136–66. On cockfighting in the T’ang, see Lo, 127–35, and Robert Joe Cutter, The Brush and the Spur: Chinese Culture and the Cockfight (Hong Kong: Chinese Univ. Press, 1989), 57–107. 79  The year was T’ien-pao 6, the tenth month corresponding to 7 November–6 December 747. “Adyta” translates chin 禁, the palace precincts and living quarters forbidden to all but the imperial family and their personal retinue, protected by the “honor-guard” (chang 杖). 80  It was in the last month of T’ien-pao 6 (that is, 5 January–3 February 748) that lodges for the “diverse administrators” (pai-ssu 百司) were erected outside the imperial compound, signifying that the monarch’s stays at Mount Li were thereafter intended to be longer than the two- or three-week excursions they had previously been and that the official business of the court would need to be carried out en bloc in the vicinity. See T’ang hui-yao 唐會要 (hereafter THY) (BSS), 30.599; HTS 37.962. 81  Hsin-feng had been the district north of Mount Li. The town of “New” Feng had been created at command of Han Kao-tsu 漢高祖 (reg. 206–194 bc) in the seventh year of his reign as a replica for his homesick father of their original hometown of Feng in the prefecture of P’ei 沛. Old Feng was rebuilt here in exact detail and its inhabitants transported to populate the new town. See Han shu 28A. 1543, esp. the note by Ying Shao 應紹 at 1544; also Yüan-ho chün-hsien t’u-chih 元和郡縣圖志 (Peking: Chung-hua shu-chü, 1983), 1.7. The redrawing of the administrative geography of the area, which produced the new district of Hui-ch’ang, is recorded in ChTS 28.1396 as occurring in 743, and in HTS 37.962 in 744. According to the chronology implied in the next sentence of Cheng Yü’s commentary, however, this change would have been adopted in 747. 82  Yin-pieh refers to the county seat more commonly known as Yin-p’an 陰槃, in existence during the Han and Western Chin periods, very near the Hsin-feng of T’ang times. 83  Lao-tzu’s epiphany is dated equivalent to 26 December 748, the second day of the twelfth (not the tenth) month of T’ien-pao 7, in THY 30.599. See also ChTS 9.222, TCTC 216.6892. However, the renowned Taoist priest and ecclesiastical historian Tu Kuang-t’ing 杜光庭 (850–933) dated the manifestation to 19 January 747 (T’ien-pao 5, twelfth month, fourth

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1388

Kroll

district was changed again, to Chao-ying 昭應 (Splendid Response) District.84 Upon the completion of the edifice, the Great General Kao Li-shih 高力士85 was ordered to lead the musicians of the adyta during the inaugural ceremony.86

At daybreak horses from Yu-chou were presented for deference and devoir;87 day), describing it as follows: “Before the sun had come forth, there suddenly appeared cloud-mottlings at the summit of Mount Li, portentous in their accumulation. In a moment the clouds dispersed and the Sainted Forefather of Integral Prime (Hun-yüan sheng-tsu 混元聖祖) appeared above the Gallery for Homage to the Prime. The thearch, along with the women of the inner precincts, beheld him with respect and awe for a good while.” Li-tai ch’ung-tao chi 歷代崇道記 (hy 593), 10a. Tu’s dating is likely mistaken, for Hsüan-tsung honored the mountain deity with a new title commemorating the event in early 749 (see the following note); it is unlikely that he would have waited two years to do so. The Ch’ao-yüan ko was located on the Embroidered Ridge of the West (Hsi hsiu-ling 西繡嶺); LTHC 2.39b. 84  It retained this name throughout the rest of the T’ang period. At this time Hsüan-tsung enfeoffed the god of Li Shan as “Duke of Mystic Virtue” (Hsüan-te kung 玄德公), in recognition of the epiphany granted on his territory. THY 47.834 gives the date of this enfeoffment as T’ien-pao 7.xii.9, or 2 January 749, a week following Lao-tzu’s appearance. 85  The eunuch Kao Li-shih (d. 762) was Hsüan-tsung’s most faithful servant, having been in his employ since before the latter assumed the throne. He would endure with him to the end, through the debacle of July 756 (when it was he who had to carry out the order to execute Lady Yang), the exile in Shu and renunciation of the throne, to the emeritus years following the return to Ch’ang-an. See also the sad task entrusted to him in lines 140–44 below. He had received the title ta chiang-chün 大將軍 early on; ChTS 184.4758, HTS 207.5858. A connoisseur of music, Kao was often involved in the planning and presentation of musical matters at court. On his ancestry, see Tu Wen-yü 杜文玉, “Kao Lishih chia-tsu chi ch’i yuan-liu” 高力士家族及其源流, T’ang yen-chiu 唐研究 4 (1998): 175–97. 86  For the phrase lo chih 落之 meaning “perform inaugural ceremonies in celebration of a newly completed official building,” see Tso chuan, Chao kung 昭公 7. Ch’un-ch’iu Tsochuan chu 春秋左傳注, ed. Yang Po-chün 楊伯峻 (Peking: Chung-hua shu-chü, 1981), 4: 1285; James Legge, The Ch’un Ts’ew with the Tso Chuan (The Chinese Classics, vol. 5; 1872, rpt. Taipei: Wen-shih-che ch’u-pan-she, 1971), 612 (tr. 616). 87  Yu-chou 幽州 (also called Fan-yang 范陽) was the administrative region centered on present-day Peking, where An Lu-shan enjoyed sole control from 744, building his power steadily. An ever-increasing number of war-horses were reared and trained under his command. “For deference and devoir” (kung-feng 供奉), an official descriptive used for those persons specially allowed to be in close attendance upon the emperor, is here attached to the horses contributed by An Lu-shan for the palace stables.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse 34

1389

Jade rang out on their jewelled bridles, the halters of yellow gold.

Each time An Lu-shan presented horses [to the court], they were sure to be decidedly outstanding and their snaffles and bridles ornamented to the maximum.88

Where the cortèges of the Five Princes lined the road that skirted the wall,89 36 The sounds of the barricade hunt were rumored by the verge of Wei  Water.90 The Plume Forest and the rest of the Six Armies each went out for the shooting; Boxing the mountain, enclosing the countryside, spreading out cords for the snares. The carved bows, embroidered bow-cases—in numbers beyond knowing; 40 Bending their backs, racing in a blur—all in gracefulest lines.91

88  Among numerous studies on T’ang horses, see Edward H. Schafer, The Golden Peaches of Samarkand: A Study of T’ang Exotics (Berkeley and Los Angeles: Univ. of California Press, 1963), 58–70, and Li Shu-t’ung 李樹桐, “T’ang-tai chih chün-shih yü ma” 唐代之軍事與 馬, rpt. in Chung-kuo shih-hsüeh lun-wen hsüan-chi 中國史學論文選集 2 (Taipei, 1977), 353–406. 89  The “Five Princes” refers to Li Lung-chi and his four royal brothers, whose fraternal closeness was as well known as it was unusual. (A sixth brother had died during the rule of Wu Tse-t’ien.) See ChTS 95.3009–19, HTS 81.3596–3603. However, if the reference here follows chronologically after the events of the preceding lines, it reveals either Cheng Yü’s scholarship or the old codger’s memory to be frayed, for all of the emperor’s brothers were dead by the year 741. 90  The Wei 渭 river ran north of the mountain. A “barricade hunt” (chiao-lieh 校獵) is one in which the area of the hunt is fenced in so as to keep the prey accessible. 91  The diction in this line calls for some comment. The soldiers bend their backs (fanshen 翻身) to shoot their arrows upward; cf. the similar use of this phrase in line 11 of Tu Fu’s 杜甫 poem “Ai chiang-t’ou,” Tu shih hsiang-chu 杜詩詳註, ed. Ch’iu Chao-ao 仇兆鼇 (Peking: Chung-hua shu-chü, 1979), 4.329. The phrase mieh-mo 滅沒 connotes the speeding blur of a “heavenly horse”; here the referent is the fine steeds of the imperial huntsmen. See Lieh-tzu chi-shih 列子集釋, ed. Yang Po-chün (Peking: Chung-hua shuchü, 1979), 8.255, for the locus classicus; cf. the usage in Li Po’s 李白 poem “T’ien-ma ko,” Li Po chi chiao-chu 李白集校注, ed. Ch’ü T’ui-yüan 瞿蛻園 and Chu Chin-ch’eng 朱金城 (Shanghai: Shang-hai ku-chi ch’u-pan-she, 1980), 3.234. The common phrase o-mei 蛾眉, “moth eyebrows,” describing the elongated eyebrows of beautiful ladies, seems to be used here to image the handsomely arrayed columns of the mounted hunters. It is also possible

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1390

44

Kroll

When the red goshawk and yellow falcon came down from amidst the clouds, The uncanny fox and wily hare had no way to protect themselves. As the men grew weary and the horses were spent, beasts and fowl were obliterated— A noisome reek for a hundred li, where the millet and grain was scarce.

The Prince of Shen 申王 had owned a red hawk from Koguryǒ, the Prince of Ch’i 岐王 a yellow falcon from the Northern Mountains.92 With pinions unrestricted, in bearing uncommon, they were markedly different from others of their kind. His Highness doted on them and whenever he engaged in the hunt or the chase placed them before him on his carriage. He called them the “Sure Victors” (chüeh-sheng-erh 絕勝兒).93

48

In the warming mountains, Hallows Month was measured with a faint breeze from the east;94 The palace beauties were granted leave to bathe in the long thermal pools. Lotus flowers incised from jade spouted forth scented liquids; Circling, gurgling, misted ripples swirled and wavered over the deeps.

that the phrase should be read as metonymy for the ladies of the court, participating in the hunt themselves as archers. 92  The Prince of Shen was Li Ch’eng-i 李成義, second son of Jui-tsung 睿宗 (reg. 710–12) and elder brother of Hsüan-tsung. The Prince of Ch’i was Li Fan 李範 fourth son of Jui-tsung and younger than the emperor. As the former died in 729 and the latter in 726, we must suppose their hunting-birds to have outlived them a good twenty years. See E. H. Schafer, “Falconry in T’ang Times,” T’oung Pao 46 (1958): 293–338, for a study of the sport. 93  The commentary following line 44, minus only the second sentence, appears (unattributed) in KYTPIS, 2.101. In that better known work the nickname given the two raptors by the sovereign is “Cloud Bursters” (chüeh-yün-erh 決雲兒). Both names are credible designations. 94  “Hallows Month” is la [yüeh] 臘[月], the last month of the year, time of the la-sacrifice to the “hundred spirits,” which serves to send away the yin-principle that has been growing in power from the summer to the winter solstice. See Derk Bodde, Festivals in Classical China: New Year and Other Annual Observances during the Han Dynasty, 206 BC–AD 220 (Princeton: Princeton Univ. Press, 1975), 49–74. The “faint breeze from the east” is, of course, a hint of spring’s arrival with the new year.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1391

Within the palace, besides the two thermal pools provided in deference [for the emperor], there were in addition sixteen “long thermae.” These “long thermae” were each granted in use to various concubines of the autocrat. In length and breadth they were dissimilar to the various other thermae. They were tiled with patterned gems and precious stones. In the very center there were jade lotuses holding up thermal fountains that spouted forth to form the pools. What is more, in the waters there were ducks and wildgeese that had been stitched and sewn together in filigreed embroidery. His Highness occasionally floated in their midst in a small boat chased with gold and silver, drifting to delight among them.95 95  I emend the first sentence of Cheng Yü’s comments, based on the relevant passage in KYTPIS 2.101. The Ming-huang tsa-lu 明皇雜錄 (hereafter MHTL) of Cheng Ch’u-hui 鄭 處誨, completed in 845 (in KYTPIS shih-chung), 2.25, gives the following, more detailed description of the bathing pools. The translation is Edward H. Schafer’s, from his “The Development of Bathing Customs in Ancient and Medieval China and the History of the Floriate Clear Palace,” Journal of the American Oriental Society 76 (1956): 76, with some necessary corrections in the final sentence.  “The Mystic Ancestor graced the Floriate Clear Palace and renewed and broadened the thermal pools with fabrication and construction magnificent and lovely. In Fan-yang, An Lu-shan made fish, dragons, ducks, and geese of white jade-stone, and even built stone bridges and stone lotus flowers, and made offerings of them. The chiseling and carving were so cunning and miraculous as scarcely to be human craft. The Highest was greatly pleased, and commanded that they be laid out in the midst of the thermae. Moreover he set the stone bridges transversely over the thermae, while the lotus flowers emerged slightly from the water’s edge. Then the Highest graced the Floriate Clear Palace, and coming to this place, undid his garments, and made to enter. But fishes, dragons, ducks and geese, all seemed to be agitating their scales or raising their pinions, being imaged on the point of flight and motion, and the Highest was very frightened, and forthwith commanded that they be eliminated and done away with. But those lotus flowers have still survived down to the present. Moreover he had ‘Chambers of the Long Baths,’ several tens of them, in the midst of the palace, and ringed them about with slabs of patterned stone. And he made boats of lacquer chased with silver, and also boats of white aromatic wood, and placed these in their midst. As to the paddles and sculls, all were adorned with pearl and jade. Further, he piled lapis lazuli and ‘sinking aromatic’ (i.e., aloeswood) in the midst of the thermae to make hills, in the image of [the paradise isles] Ying-chou and Fang-chang.”  This entire passage is not original with Cheng Ch’u-hui but is rather a quotation from Ch’en Hung’s 陳鴻 “Hua-ch’ing t’ang-ch’ih chi” 華清湯池記 (Record of the Thermal Pools at the [Palace] of Floriate Clarity), Ch’üan T’ang wen 全唐文 (Kuang-ya shu-chü ed., 1901; rpt. Taipei: Ta-t’ung shu-chü, 1979; hereafter CTW), 612.5a/b. It begins Ch’en’s record and is followed immediately by the quotation of Cheng Yü’s commentary to this section of his poem, attributed to “Chin-yang men shih-chu” 津陽門詩注 and reproduced (with

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1392

Kroll

Rhinoceros-horn screens and ivory mats were ranged in intermixed order; Brocaded ducks and embroidered wildgeese followed after one another. Broken hairpins and splintered adornments were worthlessly discarded; a few variants) in full. This would seem to be the earliest surviving reference to Cheng Yü’s poem. Ch’en’s text then concludes with these lines: “The next [pool] to the west was called the ‘Bath of the Grand Heir’ (t’ai-tzu t’ang 太子湯), the next beyond that the ‘Bath of Lesser Yang’ (shao-yang t’ang 少陽湯), and next beyond that were the sixteen sites of the Long Thermae. Today only two of the thermae—those of the Grand Heir and ‘Lesser Yang’—are preserved there. In reaching the extreme of luxuriousness and fully satisfying cupidity there has indeed rarely been a match for this, whether in ancient or modern times.” Cf. two entries on the hot springs in KYTPIS 2.101. If Ch’en Hung was in fact the author of the “Hua-ch’ing t’ang-ch’ih chi,” he could not have written it prior to 852 at the earliest, at which time he himself must have been nearly seventy years old. Ch’en Hung was famous as the author of the “Ch’ang-hen ko chuan” 長恨歌傳, composed in 807 as a prose companion-piece to Po Chü-i’s 白居易 “Song of Lasting Regret,” and of the tale called “Tung-ch’eng lao-fu chuan” 東城老父傳, also dealing with Hsüan-tsung’s era. Given his known interest in the period, his name may have been attached later to the account of the Li Shan bathing pools, after that of the actual author was lost. On Ch’en Hung, see Cutter, The Brush and the Spur, 91–94.  Recent excavations at the Hua-ch’ing Palace have brought to light a dozen or more of the T’ien-pao bathing pools, including that associated with Lady Yang (rather keyholeshaped, it is described as having the outline of a “fully opened crabapple-flower,” although its obvious similarity in shape to the vagina should not go unremarked), that of the Grand Heir (approximately sixteen and a half feet long by nine feet wide), and the “Thermae of the Lotus Flowers” (lien-hua t’ang 蓮花湯). This latter, measuring about thirty-four and three-quarter feet by nineteen and two-third feet, or over 370 square feet, is the largest yet excavated. Its conduits for guiding the inflow and outflow of the bubbling waters have been discovered in the process, as have those of Lady Yang’s pool and several others. It is these that were topped at the water’s surface with carved lotuses out of whose center the water spurted. All the pools seem to have had beneath the surface a low, outjutting ledge running round the circumference, in the manner of the narrow sitting-ledge in modern hot-tubs. For more detailed information on the excavations, see the report with photographs in the 18 March 1988 issue of Chung-kuo wen-wu pao 中國文物報, entitled “ShengT’ang huang-chia yüan-lin Hua-ch’ing kung i-chih tsai Lin-t’ung fa-hsien” 盛唐皇家園林 華清宮遺址在臨潼發現. The fullest collection of site reports, including hundreds of illustrations and drawings, is T’ang Hua-ch’ing kung 唐華清宮, ed. Lo Hsi-che 駱希哲 and the Shensi Provincial Administration of Affairs Pertaining to Cultural Relics 陝西省 文物事業管理局 (Peking: Wen-wu ch’u-pan-she, 1998). For a very interesting and thorough discussion of the origin of the name “Hua-ch’ing,” with special attention to its Taoist connotations, see pp. 525–28 of this book.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse 52

1393

In the sullied outflow from the metal drains tassels and cap-bands were  mingled. His August Highness was generous and tolerant, easily accepting events, As the Ten Houses and Three Principalities vied for a gleam of his light.96 Around the game-table, calling for the ‘black,’ they indulged in chaupar;97

96  “His August Highness” is shang-huang 上皇, the epithet applied to Hsüan-tsung as retired emperor after the throne was claimed by Su-tsung in August 756. Here it is used anachronistically, as it is in a few other places in the poem. The “Ten Houses” (shih-chia 十家) is metonomy for members of the various households related by ancestry and marriage to the royal family. The “Three Principalities” (san-kuo 三國) refers to the elder sisters of the Precious Consort, who had been ennobled as “Ladies” (fu-jen 夫人) of the principalities of Han 韓, Kuo 虢, and Ch’in 秦. They enjoyed free access to the palace and to the sovereign, and were continual recipients and conspicuous spenders of imperial treasure. Each one was granted, for instance, a thousand strings of cash yearly just to defray the cost of cosmetics (ChTS 51.2179). The younger two sisters will be mentioned specifically a few verses farther on and the eldest in the commentary following line 180. 97   Chaupar (shu-p’u 樗蒲 [M.C. t’yo bu] or, as here, shu-po 樗博 [t’yo pak]) is an Indian board-game (also called chausar or chaupad) played by two teams of two participants each over a cross-shaped design. It is similar to, but rather older and more complex than, pachisi. Medieval legend had it that Lao-tzu invented the game during his sojourn in South Asia, after leaving China; see Anna K. Seidel, La divinisation de Lao tseu dans le taoïsme des Han (Paris: École française d’Extrême-Orient, 1969), 49 n.2. Ma Jung 馬 融 (79–166) wrote a “Rhapsody on Chaupar” in the second century; it is quoted in the early T’ang commonplace book, I-wen lei-chü 藝文類聚 (Taipei: Wen-kuang ch’u-panshe, 1974), 74.1278. There Lao-tzu is said to have eased his melancholy when among the “barbarians” by playing the game. However, the game being referred to by Cheng Yü is not the classic Indian game of chaupar in which each team has sixteen counters and three dice are used. The early ninth-century Kuo-shih pu 國史補, compiled by Li Chao 李肇 (fl. 813), has a description of shu-p’u as played in the T’ang; this has been translated by E. D. Edwards, in her Chinese Prose Literature of the T’ang Period, A.D. 618–906 (sic), vol. 1: Miscellaneous Literature (London: Arthur Probsthain, 1937), 198–99. We can see from this passage that the players use five wooden dice (as opposed to three dice in chaupar), which are painted black on one side and white on the other, and that each player controls only six tokens (Edwards’s translation is in error on this point, saying there are 360 “men”; but this actually refers to the number of spaces on the board). A throw of all blacks (lu 盧) is the best possible in shu-p’u, allowing the greatest movement of one’s draughtsmen, and this is what Cheng Yü is referring to in line 55. In the following line of the poem, the “tricking and bluffing” I take to be part of the game. See T’ang kuo-shih pu, Yinhua lu 因話錄 (Shanghai: Shang-hai ku-chi ch’u-pan-she, 1979), 3.61–62. The dicing game of shup’u crops up often in early medieval texts. A particularly relevant occurrence, which may

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1394 56

60

Kroll

Setting out lamps, they outlasted the day, tricking and bluffing each other. The Minister to the Sovereign, shameless in intent, gave rein to pride and perversity; Daily accompanied by Ladies Ch’in and Kuo, he was wont to jaunt out for amusement. Vermilion tunics before his horses were not enough to bring satisfaction—98 He even forced martial footmen ahead to display his flags and banners.

Yang Kuo-chung 楊國忠 served as a minister of state (tsai-hsiang 宰相), concurrently invested as Order-and-Rule Commissioner for Chien-nan (Chien-nan chieh-tu-shih 劍南節度使).99 He regularly went on outings in company with the Ladies of Ch’in and

have been in the back of Cheng Yü’s mind, is that in Ko Hung’s 葛洪 (283–343) Pao-p’u tzu wai-p’ien 抱朴子外篇, “Pai-li” (SPPY), 28.1b, where mention is made of “those who neglect their governmental responsibilities through playing wei-ch’i 圍棋 and shu-p’u.” For another ninth-century description of shu-p’u, see Li Ao’s 李翱 (ca. 772–836) Wu-mu ching 五木經 (in T’ang-tai ts’ung-shu), which confirms much of the information in the Kuo-shih pu regarding the game, but also includes additional facts—e.g., that the game can be played by up to five individuals (again note the contrast with chaupar). Clearly, “chaupar” had by this time been sinicized and the term had acquired the general sense of “dicing game” (cf. the modern usage of the word, now usually written 摴蒱), no longer denoting specifically an Indian import. This topic awaits a proper study. One of Lien-sheng Yang’s classic articles, “An Additional Note on the Ancient Game liu-po,” Harvard Journal of Asiatic Studies 15 (1952): 124–39, glances briefly at T’ang shu-p’u (pp. 132–34). 98  Vermilion tunics were worn by the emperor’s servants. 99  Yang Kuo-chung, second cousin to Yang Kuei-fei, controlled the court and government after the death of Li Lin-fu 李林甫 in early 753. Yang Kuo-chung had spent his early years in Szechwan (i.e., Chien-nan 劍南); late in 751 he was made chieh-tu shih of the area and proceeded to build up there a private base of military power. This was the main reason that the emperor chose to take refuge there, when forced to flee the capital in the summer of 756. It was popular knowledge that Yang Kuo-chung was carrying on an incestuous relationship with the Lady of Kuo. His extravagance and that of the Precious Consort’s three sisters was extreme, provoking much resentment during the years before the great rebellion. See, e.g., MHTL, 2.25. Probably the most famous verse comment on these gorgeous persons is that of Tu Fu, in his poem “Li-jen hsing” 麗人行, Tu shih hsiang-chu, 2.156–60, written in spring 753 to describe just such an outing as is referred to by Cheng Yü here; see also his “Kuo-kuo fu-jen,” 2.162.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1395

Kuo, all riding side-by-side.100 He also used the banners and verge of the Two Rivers (liang-ch’uan 兩川101) to show the way before his horse.

The painted wheels, jewelled axletrees, must have come here by way of heaven; The laughter and talk, from out of the clouds, gave voice to perfused bliss. Cracking their whips, the riders behind—how daintily they paced!102 64 Palace adornments on lappets and sleeves were in the style of the  transcendents. By the azure gates and purple lanes the springtime breezes were full;103 Amid the breezes, for a number of days, fading springtime would linger. Where the ebony colts, in frothy sweat, had galloped all hot in haste, 68 Along the road, wielding brooms, folk vied to sweep up gemstones and  beads.104

100  Literally, “with all their reins joined”—i.e., riding side-by-side in equality, not single-file according to precedence. Cf. ChTS 106.3245: “There were times when Kuo-chung entered the court, reins matched with those of the Lady of Kuo Principality, the two of them brandishing whips and racing their horses for sport and lark.” 101  Tung-ch’uan 東川 and Hsi-ch’uan 西川, both Pa 巴 and Shu 蜀, that is, the entire Szechwan area. 102  “The riders behind” refers to Yang Kuo-chung and his female companions, coming after the standard-bearers who clear the way for them. The line is reminiscent of Tu Fu’s verse portraying Yang Kuo-chung’s mounted arrival at the feast, in his “Li-jen hsing”: “The saddled horse of he who comes last—how languidly it parades!” ChTS 106.3245 gives the following picture of the high minister’s outings at Li Shan: “Kuo-chung had the banners and verge for [his position as military governor of] Chien-nan lead forth in the front. When he went out there was a feast for the road; when he returned there was another to ‘soothe the viator.’ ” The Chu-chuang shih-hua reading, adopted in TSCS chiao-chien, says “The frontcriers (lit. “crying-hurriers,” referring to those who call out to clear the road in advance) and the riders behind—how daintily they paced.” 103  The “purple lanes” (tzu-mo 紫陌) are the streets of the imperial city, purple being the symbolic color for all associated with the monarch. 104  Cf. ChTS 51.2179: “In the tenth month of each year, when Hsüan-tsung graced with his presence the Palace of Floriate Clarity, the five households of [Yang] Kuo-chung and the sisters [of Yang Kuei-fei] followed in procession. Each household formed a single host, wearing garb of an identical color. When the five houses combined their hosts, they shone radiantly like the iridescent display of the hundred flowers; and the gauds that had fallen off, the slippers that had dropped, bluestones, pearls, halcyon feathers, all glittered and gleamed, fragrantly redolent on the road.” Cf. the similar description at ChTS 106.3245.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1396

Kroll

These matters are thoroughly indited in the state history. Here below are recounted yet more of such.105

72

Eighth Sister raised up anew a hall of Concordant Joy,106 Which the hovering stork and congratulant swallow could barely descry.107 Ten thousand in gold to pay the builder was rejected by him out of hand; Proud of his competence, sure of his craft, he just sniffed self-pleased.

The Lady of Kuo Principality had a hall designed, the expense and outlay for which was a myriad gold-pieces. With the completion of the hall she sought to reward the expertise of the builder’s skill beyond the recompense promised: he received as a bonus five

105  The references cited in some of the preceding notes suffice to indicate Cheng Yü’s familiarity with documents that were ultimately used in composing the official history of the dynasty. 106  The “Eighth Sister” (pa-i 八姨) is what Hsüan-tsung called the Lady of Ch’in, the youngest of Yang Kuei-fei’s elder sisters (see n. 96 above). The oldest, the Lady of Han, was called “Ta-i” 大姨 (Elder Sister) by him and the middle of the three, the Lady of Kuo, “San-i” 三姨 (Third Sister). See ChTS 51.5178. Cheng Yü seems to have confused the Lady of Kuo in his commentary with the Lady of Ch’in in line 69. As we have noted, it was the Lady of Kuo who was said to have been intimate with Yang Kuo-chung. “Concordant Joy” (ho-huan 合歡) was the name of one of the women’s palace residences of Han Ch’angan, mentioned in Pan Ku’s 班固 (32–92) “Hsi-tu fu” 西都賦, Wen hsüan 文選 (Shanghai: Shanghai ku-chi ch’u-pan-she, 1.12; hereafter WH); tr. David R. Knechtges, Wen xuan, or Selections of Refined Literature, vol. 1: Rhapsodies on Metropolises and Capitals (Princeton: Princeton Univ. Press, 1982), 123. The name evokes the sumptuousness of the dwellings commissioned by the profligate sisters. All three sisters had country estates at Mount Li, as did Yang Kuo-chung; the four mansions were very near one another, to the south of Hua-ch’ing’s east gate. See ChTS 106.3245; also LTHC 2.38a. 107  That is, the buildings were so grandly imposing that they exceeded the range of the birds’ vision. The “hovering stork” is mentioned in a similar context in Chang Heng’s 張衡 (78–139) “Hsi-ching fu” 西京賦, WH 2.56–57, where, in a description of Han Wu-ti’s “Sky-piercing Tower” (通天臺), we read: “The hovering great fowl [or stork], neck craned upward, was unable to reach the spire”; tr. Knechtges, Wen xuan, 1: 195. As to the “congratulant swallow,” it is mentioned in Huai-nan tzu that, upon the completion of a large building the swallow comes to offer its felicitations. See Huai-nan tzu chu 淮南子注 (Taipei: Shih-chieh shu-chü, 1962), 17.295.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1397

thousand lengths of scarlet silk-gauze. Contemptuous, he disdained even to look at it. The Lady was astonished by this and asked for an explanation. The builder said, “I have here expended the competence gained in a whole lifetime. Should you not perchance believe me, I pray you take such things as mole-cricket, ant, skink, wasp, or scorpion and, having put out their eyes, cast them into the hall. If there be any crack or fissure into which a single creature can disappear,108 then speak no more of my expertise as builder.” At this she showered him in addition with silk-stuffs of variegated hues and costly treasures. The people who live by the mountain tell of these old incidents even to the present day, and they still refer to the several sisters simply by their clan rank.109

76

Veritable moguls from throughout the empire bent to fawn on and truckle to her; Utter extravagance, consummate squandering shopped favor and personal interest. Specially placed inside the hall was a pillow that shone in the night; The silver tapers were no longer set out, for it lit up mirror and curtains.

The Lady of Kuo’s pillow that shone in the night was placed in this hall, where it lit up and illumined the whole chamber. It had been presented to her by the Order-and-Rule Commissioner for Hsi-ch’uan 西川.110 The matter is indited in the state history, there being briefly reported.111

108  Here I read 使有閑隙得亡一物 with TSCS, instead of 使有隙失一物 with CTS. 109  On the building projects of Yang Kuei-fei’s powerful relations, ChTS 51.2179 has this to say: “For the framing of any one hall the expense surpassed a thousand myriads in the reckoning. But if they saw a layout and measurements more spacious and grand than their own, they had theirs demolished and built anew. The landscapers and carpenters never rested, day or night.” See particularly MHTL, 2.25–26, regarding Lady Kuo’s appropriation and dismantling for her own purposes of Wei Ssu-li’s 韋嗣立 former residence. In referring to the sisters by their clan rank (ti-hang 第行)—that is, their order of seniority in their generation, e.g., “Miss Three”—instead of by their court titles, the locals speak of them with casual familiarity. 110  That is, Yang Kuo-chung. See n. 99 above. 111  Cf. KYTPIS, 2.101: “The Lady of Kuo Principality had a pillow that shone in the night. It was placed in the middle of the hall, where it lit up and illuminated the whole chamber, unaided by lamp or taper.” Pillows typically being fashioned of carved stone, this one must have been made of a block with fluorescent properties, such as possessed, for instance, by some varieties of fluorite known in T’ang times.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1398

80

Kroll

South of the Tower of Gemmy Light spread the purple adyta, Where sylphine revelers from the Pear Garden looked out on flowering boughs. Cozy Damsel’s caroling throat was a supple suavity of jade, And Odalisque’s dancing sash a tasseled tendril of gold.

The Tower of Gemmy Light (Yao-kuang lou 瑤光樓) was the north portal of the Great Hall of Drifting Frost (Fei-shuang tien 飛霜殿).112 Cozy Damsel (Ying-niang 迎娘) and Odalisque (蠻兒) were famously notorious members of the Apprentices of the Pear Garden.113

84

The purple cross-flute of the Third Young Lord played with the hazy moonlight, Plaintive as a crane set apart, calling out to its lone-roaming hen. The round lute of Jade Slave, with plectrum of dragon’s aromatic, Falling in with the song, as wine was urged, gave voice to a delicate grief.

His August Highness was expert at performing upon the flute; he regularly favored a unique pipe of purple jade. The Precious Consort played wondrously upon the round lute; her instrument, one renowned in the human realm, had a sound-box made of Lhasa rosewood, with a plectrum made of dragon’s-aromatic cypress.114 When His 112  This structure was east of the Chin-yang Gate; CAC 15.8a. It was the emperor’s sleeping hall; see the commentary following line 148. 113  On the renowned “Pear Garden” (Li yüan 梨園), with its group of “apprentices” (ti-tzu 弟子) in song and dance, established by the ruler in 714 and sometimes engaging his personal instruction, always his appreciation, the best summary and discussion of sources is Kishibe Shigeo 岸辺成雄, Tōdai ongaku no rekishiteki kenkyū: gakuseihen 唐代音楽 の歷史的硏究: 楽制篇, 2 vols. (Tokyo: Tôhô daigaku shuppankai, 1960), 1: 449–90. The dancer referred to familiarly here as “Odalisque” is Hsieh A-man 謝阿蠻, a Serindian performer who particularly excelled in dancing to the tune “Skimming the Waves” (“Ling-po ch’ü” 凌波曲). See MHTL, “pu-i” 補遺, 35; also the remarks of Wang K’o-fen 王克芬 in Ou-yang Yü-ch’ien 歐陽予倩, T’ang-tai wu-tao 唐代舞蹈 (Shanghai: Shang-hai wen-i ch’u-pan-she, 1980), 145–46, on “Ling-po ch’ü” and Odalisque’s participation in Hsüantsung’s musical fêtes. 114  Sanderswood, or as it was termed in the T’ang, “purple rosewood (or purple sandalwood)” (tzu-t’an 紫檀), was “the preferred substance for making stringed musical instruments, above all the lute.” Schafer, Golden Peaches, 135. The tree is native not to China but to Central Asia, hence the reference here to “Lhasa (i.e., Tibetan) rosewood” (邏逤檀). The fourteenth-century Wen-hsien t’ung-k’ao 文獻通考 (BSS), 137.1218b, records the following: “During the T’ien-pao era of the T’ang, the eunuch-official Po Hsiu-cheng 白秀正

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1399

Highness would take up a flagon of wine he would command Cozy Damsel to sing a turn of the “Melody of the Waters,”115 while T’ai-chen 太真116 plucked the lute-strings, falling in with the song—thus they “delivered up” the wine to His Highness. Within the inner palace all referred to His Highness as the “Third Young Lord” (San-lang 三郎).117 “Jade Slave” (Yü-nu 玉奴) was the child-name of T’ai-chen.

By the side of Deer-drink Fountain, springtime dew dried in the sun, As powdery prunus and sandal-hued apricot wafted through the vermilion parvis.118 was commissioned to Shu 蜀. Upon his return he proffered a paired-phoenix lute the sound-box of which was made of Lhasa rosewood. Warmly lustrous, with a glossy sheen, its surface shadings resembled those of sceptre (圭) or circlet (璧); there was a crimson patterning, threaded in gold, forming in passement a pair of phoenixes. When the Precious Consort played it herself in the Pear Garden, its sound and resonance were clear and piercing; wafted as though beyond the clouds, it was scarcely of a kind with anything in the human realm. The several princes and honored princesses vied to be the Precious Consort’s apprentices on the lute.” Shu, of course, would be the most logical Chinese locale in which to acquire an item crafted from Lhasa rosewood. This entry is copied over, with minor changes, from a passage attributed to MHTL in T’ai-p’ing yü-lan 太平御覽 (Taipei: Shang-wu yin-shu-kuan, 1968), 583.5a; MHTL “i-wen” 逸文, 37. 115  The tune “Melody of the Waters’ (“shui-tiao” 水調) is said to have been created by Emperor Yang of the Sui 隋煬帝 (reg. 605–17) after his retreat to Yangchow—the Sui Chiang-tu 江都 or “Metropolis on the Kiang”)—in the darkening last year of his reign. The tune is characterized as “plaintive and sharp,” and when Yang-ti heard the lyrics that were added to it he is said to have remarked that they made him feel he would never be returning north, as indeed he never did. The T’ang tune of this name is reported to have had eleven “repetitions” (tieh 疊), rhythmically framed so that the first five constituted a ko 歌 and the final six a ju-p’o 入破 that is, in the form of a proto-ta-ch’ü 大曲 (see n. 132 below). The lyrics of these sections, presumably those that would have been sung by Cozy Damsel herself, are preserved. We do not know, however, whether the tune was that of Yang-ti’s original, which was still known in T’ang times, or a new variant of the tune which is also reported. Both versions were classed in the shang 商 mode, seasonally identified with autumn (hence, plaintive). See Sung-pen Yüeh-fu shih-chi 宋本樂府詩集 (Taipei: Shih-chieh shu-chü, 1961), 79.6b. For more on the “Melody of the Waters,” see n. 167 below. 116  T’ai-chen (“Greatest Perfection”) was Yang Kuei-fei’s religious name, taken during her brief tenure as a Taoist priestess, after she was removed from the side of Hsüan-tsung’s eighteenth son, Li Mao 李瑁, Prince of Shou 壽王, and purified—as it were—to share the emperor’s bed. 117  Because he was the third son of his father, Jui-tsung. 118  “Powdery prunus” (fen-mei 粉梅) refers to the flowers of Prunus mume, the so-called “Japanese apricot,” not a “plum” as usually translated. See esp. E. H. Schafer, “Mildewed

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1400

Kroll

At the mouth of the Grotto of Gold Granules was the Hall of Protracted Life; 88 At the top of the Peak of Jade Stamens, the Bethel of the Spirit  Mother.119 Within the mountain walls there were numerous tame deer and a flowing gill that was styled Deer-drink (Yin-lu 飲鹿). There was the Hall of Protracted Life (Ch’ang-sheng tien 長生殿), it being a basilica for ritual purification. When there were events to be held in the Gallery for Homage to the Prime, it was to the Hall of Protracted Life that the autocrat repaired for ablution and lustration.120

92

In the forbidden Court adepts of artifice multiplied unreal transmutations; Matching their prowess before His Highness, all to gain the upper hand. The ju-i of Sire Lo was stripped of its guise and appearance, And the kaṣāya of Tripiṭaka was turned into raveled threads.

Apricots,” Schafer Sinological Papers 25 (10 May 1985). However, it will not do to translate mei as “apricot,” unless we do otherwise for hsing 杏, the Prunus armeniaca, which already enjoys that designation rightfully. The problem becomes acute when, as in this verse, the two trees appear in immediate proximity: to call them both “apricot” will clearly not work. I reserve “apricot” for 杏 and apply the synonymous “prunus” to 梅. “Sandalhued apricot” (t’an-hsing 檀杏) calls attention to the light pinkish color of the hsing’s flowers. The mei and the hsing were, according to tradition, the first plants to bloom in the spring. “Vermilion parvis” (chu ch’ih 朱墀) denotes the red-lacquered entrance porch of a great hall. 119  The “Spirit Mother” in whose honor the bethel (tz’u 祠) is named is, of course, the Taoist goddess Hsi Wang Mu 西王母, the Spirit Mother of the West. In rendering her title this way, rather than the more normal but mistaken “Royal (or Queen) Mother of the West,” I am following Paul R. Goldin, “On the Meaning of the Name Xi wangmu,” Journal of the American Oriental Society 122 (2002): 83–85. For the site of the bethel, see LTHC 2.36a. 120  The Ch’ang-sheng tien, built in 742, was behind (i.e., southeast of) the main enclosed compound, farther up the mountain. CAC 15.9a, LTHC 2.38b–39a. For a fine study of the practice of ritual purification or purgation (chai 齋) in T’ang Taoism, see Roman Malek, Das Chai-chieh lu: Materialien zur Liturgie im Taoismus (Frankfurt am Main: Peter Lang, 1985); a good historical and inclusive survey of the concept is presented on pp. 1–34. The complementary pair of grotto and peak nearby were named by the emperor appropriately in suggestion of elixir items conducing to long life—gold granules (chin-sha 金沙) and jade stamens (yü-jui 玉橤). These were in the vicinity of the Embroidered Ridge of the East. On the Gallery for Homage to the Prime, see n. 83 above.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1401

His Highness quite esteemed Lo Kung-yüan 羅公遠, while Consort Yang put her trust specially in Vajratripiṭaka 金剛三藏.121 Once when His Highness graced with his presence the Close of Pious Works (Kung-te yüan 功德院) and was about to pay a visit to the Temple of the Seven Paragons (Ch’i-sheng tien 七聖殿),122 his back suddenly became itchy. Kung-yüan broke off a branch of bamboo and transmuted it into a seven-treasure ju-i,123 to be presented [for the sovereign’s use]. His Highness was greatly delighted. Turning to Tripiṭaka, he said, “Can Your Eminence top this?” Tripiṭaka said, “This is merely an unreal artifice. Let a monk bring the actual item to Your Majesty’s hand.” He then produced from his sleeve a ju-i whose seven jewels shone glittering and sparkling. Right away the one presented by Yüan became again merely a bamboo branch.124 On a certain day afterward Consort Yang proposed to determine which of the two men excelled the other. At the time a lesser hall was being built within the adyta. Tripiṭaka proceeded to levitate an enormous beam into the air, ready to drop on Kung-yüan’s head. Kung-yüan’s countenance remained unchanged, and His Highness had to order Tripiṭaka to stop. Kung-yüan then sent a talisman flying to a spot elsewhere, to snatch from its coffer Tripiṭaka’s gold-stitched kaṣāya.125 The guardian there noticed nothing.

121  Numerous stories circulated in T’ang times about Lo Kung-yüan, a thaumaturge who alternately charmed and frustrated the emperor. He was later to become the center of a local rain-cult in Ch’eng-tu. See the admirable study of Franciscus Verellen, “Luo Gongyuan: Légende et culte d’un saint taoïste,” Journal asiatique 275 (1987): 283–332, for translation of and critical commentary on the relevant texts. On p. 299 Verellen identifies Vajratripiṭaka plausibly as the Tantric master Vajrabodhi 金剛智 (671–741), perhaps best known as the teacher of Amoghavajra (705–74). The phrase 奪顏色 in line 91 of the poem refers to Lo Kung-yüan’s conjured ju-i (see n. 123 below) losing its appearance and resuming its original shape as a bamboo branch, not, as Verellen translates, to Lo’s “losing face.” 122  The Kung-te yüan was just south of the Ch’i-sheng tien, within the Hua-ch’ing compound. For more on them, see Cheng Yü’s commentary following line 156. 123  As pointed out by E. Zürcher, The Buddhist Conquest of China: The Spread and Adaptation of Buddhism in Early Medieval China, 2 vols. (Leiden: Brill, 1972), 2: 407 n. 59, the ju-i 如意 sceptre, “the most venerable attribute of the Buddhist priest,” had its origin as a simple back-scratcher, for which purpose Lo Kung-yüan produces one here. For the Temple of the Seven Paragons, see the commentary following line 156 below. 124  Following for this and the preceding sentence the TSCS text. 125  The kaṣāya 袈裟 (M.C. kǎsrǎ) or cassock embroidered in gold recalls that created for Śākyamuni by Mahāprajāpati. According to the usual version of the story, Śākyamuni refused to accept it, preferring it to be offered to the monks assembled with him, all of whom declined the honor except Maitreya who was thereby revealed as the “future Buddha.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1402

Kroll

Infuriated, Tripiṭaka made an incantation to reclaim it and it shortly came to him. Kung-yüan next spit out on the kaṣāya a water-dragon talisman that decomposed it so completely that it became just a frazzle of threads.

By the P’eng-lai Pool, He gazed at the autumn moon afar,126 Unclouded for a myriad li, shedding its clear brilliance. In mid-night the High Radiant One set off in the moonlight; 96 The six and thirty palaces feared he might never return home.127 As orphic music from in the moon was diffused in mid-sky, A tinkling of jade stones harmonized with vessel-flute and flageolet.128 The Blessed Percipience harked and heard, but the tune was not finished;129 100 Arrived back among humankind, He could not make out fact from  fiction. Yeh Fa-shan 葉法善 conducted His Highness into the Moon Palace, when autumn was deep in season. As His Highness found the intense cold there to be painful, he was not able to linger long and had to return homeward. Transcendent music still sounded

126  The P’eng-lai Pool 蓬萊池, deriving its name from that of one of the three paradise isles in the Eastern Sea, was located in the northern sector of the Ta-ming Palace 大明宮 compound extruding from the northeast side of the city-wall of Ch’ang-an. It seems to have been an alternate name for the T’ai-yeh 太液 or “Grand Fluid” Pool, which had an artificial “P’eng-lai Mountain” with a “P’eng-lai Pavilion” in its midst. 127  The san-shih-liu hung 三十六宮 refers to the various halls and dwellings of the royal compounds and, by synecdoche, their residents. The locus classicus is Pan Ku’s “Hsi-tu fu,” WH 1.10; tr. Knechtges, Wen xuan, 1: 113: “The detached palaces and separate lodges/ Are thirty-six in number.” 128  These are instruments of the celestial orchestra heard by the emperor. “Vessel-flute” is hsün 塤, sometimes rendered “ocarina,” but see L. E. R. Picken, “T’ang Music and Musical Instruments,” T’oung Pao 55 (1969): 118. “Flageolet” is ch’ih 箎. Schafer’s description of this couplet, in his Pacing the Void: T’ang Approaches to the Stars (Berkeley and Los Angeles: Univ. of California Press, 1977), 196, as showing “the deposed monarch in his declining years” hearing the lunar music in distant memory is in error, as is his dating of Cheng Yü and the poem to the late ninth century. 129  Blessed Percipience” is ch’en ts’ung 宸聰, denoting specifically the aural faculties of the exalted ruler.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1403

in his ears when he was in mid-sky. Upon his return he could recall from memory only the half of it, which he consequently reproduced upon the cross-flute.130 As 130  The story of Hsüan-tsung’s trip to the moon exists in numerous versions, involving various guides. The one referred to here by Cheng Yü, with the wonder-worker Yeh Fa-shan (631–720; for biographical accounts, see Huan-hsi chih 幻戲志 [T’ang-tai ts’ung-shu 唐代 叢書, 1806 ed.]; Tu Kuang-t’ing’s Tao-chiao ling-yen chi 道教靈驗記 [HY 590], 14.8b–9a; ChTS 191.5107–8; HTS 204.5805) as outworldly director, has its closest cousin in Chiang Fang’s 蔣防 (fl. 806–21) Huan-hsi chih, 8a. That version is repeated in T’ai-p’ing kuang-chi 太平廣記 (Taipei: Ku-hsin shu-chü, 1976), 26.75d, embedded in a clutch of anecdotes regarding Master Yeh’s wondrous deeds. The relevant portion of the tale reads: “Once, it being the full-moon night of the eighth month (i.e., the mid-autumn festival, 中秋節), the Master journeyed with Hsüan-tsung to the Moon Palace. Giving ear to the celestial music to be heard in the moon, [the monarch] inquired about the name of the tune being played. He was told, ‘Purple Clouds.’ Hsüan-tsung had long been an aficionado of ‘notes and pitches’ and he silently memorized the sound of the piece. Upon his return [to earth], he carried over its tonalities. The resulting piece was called ‘Rainbow Skirts and Feather Vestments.’ ” (This version continues with a report of Hsüan-tsung’s incognito playing of the tune on his flute above Lu-chou 潞州 [present-day Ch’ang-chih 長治, Shansi] during a mid-air pause on the trip home, followed by his sprinkling of gold coins over the town walls; ten days later both the heavenly melody and the miraculous coins were presented to the court in awe-struck tribute.).  A postscript appears after this story in Huan-hsi chih, recording a somewhat more mundane narrative in which the emperor is said to have had the song, here called “Purple Clouds of the Divine Transcendents,” played for him by a troupe of a dozen transcendent musicians who appeared to him in dream, proclaiming it as the “sound of the True Inception of the incomparable T’ang” (sheng-T’ang cheng-shih yin 聖唐正始音). The next day he was so distracted at an audience with his chief ministers Yao Ch’ung 姚崇 and Sung Ching 宋璟 as to leave them in quivering trepidation over his dismissive attitude toward them. When queried later by Kao Li-shih, the emperor confessed that his inattention was not because of any displeasure with his ministers but was rather owing to his desire to concentrate on recapturing from memory the wonderful dream-tune he had heard the night before, which he then performed for Kao on a jade flute. This rendition is similar in outline (though quite different in linguistic particulars) to that contained in the Li Mo ch’ui-ti chi 李謨吹笛記 (T’ang-tai ts’ung-shu), 4b–5a, attributed to Yang Chü-yüan 楊巨源 (c.s. 789). A nearly identical version appears in K’ai-T’ien ch’uan-hsin chi 開天傳 信記, compiled by Cheng Ch’i 鄭棨 (d. 899), in KYTPIS shih-chung, 54; it has been translated by Schafer, Pacing the Void, 199–200. For more on Yeh Fa-shan, see Russell Kirkland, “Tales of Thaumaturgy: T’ang Accounts of the Wonderworker Yeh Fa-shan,” Monumenta Serica 40 (1992): 47–86. The Tun-huang narrative text known (improperly) as Yeh Chingneng shih 葉凈能詩 (S. 6836) includes a version of the lunar jaunt that adds several items of Buddhist significance. See Tun-huang pien-wen chi 敦煌變文集, ed. Wang Chungmin 王重民 et al. (Peking: Jen-min wen-hsüeh ch’u-pan-she, 1957), 225; on the error in

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1404

Kroll

it happened, Yang Ching-shu 楊敬述, Intendant-General for Western Liang (西涼 都督),131 had presented to court the musical piece “Brahman” 婆羅門 the tone and mode of which rather tallied with that [of the celestial music]. Consequently, what had been heard in the moon was made into a free prelude (散序) and the tune presented by Ching-shu employed as a main theme (腔), and this tune was named “Rainbow Skirts and Feather Vestments” (Ni-shang yü-i 霓裳羽衣).132 the ascribed title, see Victor H. Mair, T’ang Transformation Tales: A Study of the Buddhist Contribution to the Rise of Vernacular Fiction and Drama in China (Cambridge, Mass.: Council on East Asian Studies, Harvard Univ., 1989), 12.  The Taoist master Lo Kung-yüan, mentioned in line 91 of our poem and in the commentary following line 92, is Hsüan-tsung’s guide and companion in another account of the moon-trip, found in the I shih 逸史 (preface dated 847) of Lu Chao 盧肇 as preserved in citations in late encyclopedias. See translations by Michel Soymié, “La Lune dans les religions chinoises,” in La Lune: Mythes et rites (Paris: Seuil, 1962), 310, and especially F. Verellen, “Luo Gongyuan,” 327–28, with notes. Here the journey is still dated to the fullmoon night of the eighth month, though the year is now given as being in the beginning of the T’ien-pao period. There are other unique and interesting details in this version— e.g., at the outset Lo tosses into the air a branch of osmanthus which turns into a silver bridge providing him and the emperor with a causeway to the moon, the bridge later vanishing behind them step-by-step on the return trek to earth; and here the celestial tune is named by the transcendents themselves as “Rainbow Skirts and Feather Vestments.”  Still another, and longer, depiction of the moon journey is recorded in the late Sung Lung-ch’eng lu 龍城錄 (T’ang-tai ts’ung-shu), 6b–7a, with a Heavenly Master Shen 申天師 functioning as trip-leader and another adept added to the party. Translation in Soymié, “La Lune,” 308–9; summarized and discussed in Schafer, Pacing the Void, 200 (where one must correct “Heavenly Master T’ien” to … “Heavenly Master Shen”). 131  The border protectorate of Western Liang comprised the northwest sector of present-day Kansu, roughly from Chiu-ch’üan 酒泉 to Tun-huang 燉煌. The music of Liang-chou, part exotically Kuchean, part traditionally Chinese, made up one of the ten main divisions of popular (as opposed to “formal”) music and was particularly favored in Ch’ang-an during the first half of T’ang rule. Among other discussions, see esp. Kishibe, Tōdai ongaku, 2: 210–16; also Ou-yang Yü-ch’ien, T’ang-tai wu-tao, 67–69. According to ChTS 29.1073, the orchestration for “Brahman” added to the standard ensemble two lacquered pi-li 篳篥 or double-reed pipes (about which, see Picken, “T’ang Music and Musical Instruments,” 118–20), and one ch’i-ku 齊鼓, a drum resembling a lacquered churn (see description at ChTS 29.1079). The earliest extant reference to Yang Ching-shu’s role in presenting the tune “Brahman” seems to be one of Po Chü-i’s interlinear explanations to his poem “Ni-shang yü-i ko,” CTS 444.4971 (see appendix 1 below): “During the K’ai-yüan era, it was composed by Yang Ching-shu, Order-and-Rule Commissioner for the Archivate of Western Liang,” this commenting on the verse “Mr. Yang created the music, the sovereign devised the setting.” HTS 22.476 miswrites Yang’s personal name as Ching-chung 敬忠. 132  The dance-suite “Ni-shang yü-i,” most famous of all T’ang musical pieces and reputedly the chief vehicle for the Precious Consort’s gestic talents, has been much written about,

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1405

The autocrat’s festival of a Thousand Autumns took place in the eighth month, A gathering that joined a myriad states, both Chinese and foreigners at court. South of the Tower of Blossom and Calyx was a great concert of music;133 104 To the complete performance of all instrumentation, simurghs in ceremony  came.134 although not always accurately. From the description here, and also that in Po Chü-i’s long poem on the dance, it is clear that “Ni-shang yü-i” was an extended suite or ta-ch’ü 大曲. T’ang ta-ch’ü typically consisted of three major sections—the “free prelude,” the “central prelude” (chung-hsü 中序), and the “broaching” (p’o 破, sometimes called “entering broaching” [ju-p’o 入破]), with each of these three sections containing a number of subsidiary divisions. According to Cheng Yü, the moon music recalled by the emperor was played as the “free prelude,” an introductory sequence of solo parts, and the Serindian melody “Brahman” provided the musical subject (literally “thorax” or “main bodycavity”) that was embroidered (or, following the somatic metaphor, filled out) during the chung-hsü and p’o divisions. Cf. Martin Gimm, Das Yüeh-fu tsa-lu des Tuan An-chieh: Studien zur Geschichte von Musik, Schauspiel und Tanz in der T’ang-Dynastie (Wiesbaden: Harrassowitz, 1966), 224–30; also Picken, “T’ang Music and Musical Instruments,” 83–87. For more details on “Ni-shang yü-i,” see Appendix 1 below. 133  Hsüan-tsung held official celebrations deriving from personal events within the grounds of the Palace of Ascendant Felicity (Hsing-ch’ing kung 興慶宮), a compound of halls and pavilions directly northeast of Ch’ang-an’s “eastern market.” This neighborhood, where Hsüan-tsung had resided during the years 701–12, was designated a “palace” proper after his assumption of the throne and continued to be visited by him on special occasions. Renovations and expansions of the several structures of the Hsing-ch’ing kung were carried out at various times. The Tower of Blossom and Calyx in Mutual Radiance (Hua-o hsiang-hui lou 花萼相輝樓), was a multi-storeyed building located in the southwest corner of the compound. It took its name from the fact that from it the emperor could see the abodes of his princely brothers, who still resided in this quarter, strung in sequence close by, like the leaves of the calyx round a flower. Immediately next to the Hua-o lou was the Tower of Zealous Administration, mentioned in Cheng Yü’s commentary to this section of the poem. As stated there, the celebration described in these lines was that for the emperor’s birthday. On the Hsing-ch’ing palace, see esp. CAC 9.3a–4b; Hsü Sung 徐松 (1781–1848), T’ang liang-ching ch’eng-fang k’ao 唐兩京城坊考 (Peking: Chung-hua shuchü, 1985), 1.25–27; and, for detailed drawings of the areal layout, Hiraoka Takeo 平岡武 夫, Chōan to Rakuyō: chizuhen 長安と各陽: 地図篇 (Kyoto: Jimbun Kagaku Kenkyūjo, Kyoto Univ., 1956), maps 23 and 24. 134  “All instrumentation” paraphrases pa-yin ハ音, “eight timbres,” referring to the characteristic sounds of the eight families of Chinese musical instruments: metal (represented by bells), stone (lithophone), silk (strings), bamboo (pipes), gourd (syrinx), earth

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1406

Kroll

The Tu-lu pole-climbers were, in truth, most nimbly adroit,135 And Madame Kung-sun’s skill with the sword just uncannily wondrous. The horses knew when their dance was done, descending from upheld platforms; 108 And everyone grudged the melody’s ending, when the feathered vestments were changed. His Highness had begun to use the day of his birth as the “Festival of a Thousand Autumns,” each time holding a large celebratory gathering below the Tower of Zealous Administration (Ch’in-cheng lou 勤政樓) which he made sure was open to view by both Chinese and foreigners.136 There was a Madame Kung-sun 公孫大娘 who danced (vessel-flute), skin (drums), and wood (churn-box). The whole line recalls the description in Shang shu 尙書, “I Chi” 益稷, of how phoenixes (feng-huang 鳳凰, instead of Cheng’s “simurghs” [luan 鸞]) “in ceremony came” (lai i 來儀) in response to the “ninefold perfection [of the sage-king Shun’s] hallowed hymnody” (hsiao-shao chiu-ch’eng 蕭紹九成, chiu-ch’eng implying “complete performance”). See also Shih chi 史記 (Peking: Chung-hua shu-chü, 1962), 2.81, with accompanying commentary (2.82) of P’ei Yin 裴駰 (372–451) using the synonymous term chiu-tsou 九奏. 135  Tu-lu 都盧 as the home of Burmese pole-climbers is first mentioned in Chang Heng’s “Hsi-ching fu,” WH 2.58; tr. Knechtges, Wen xuan, 1: 197. Knechtges’ commentary notes that, while the term tu-lu 都盧 may transcribe the name of the locality whence the acrobats originally hailed, it is possible that it may itself mean “pole-climber” in a purely Chinese context. Indeed, there are T’ang examples of this; see, e.g., the following passage in Liang She’s 梁涉 (fl. 730–50) “Rhapsody on the Long Pole” (“Ch’ang-kan fu” 長干賦, CTW 407.3a–4a), celebrating such an act by a female acrobat in Hsüan-tsung’s time: “There is a lovely person—/ Come from the [palace ladies’] Purple Portals; /To perform as a pole-climber (為都盧)—/ Clothed in brocade, with a simple cloak. / She has applied her painted make-up to resemble jade, / Boosts up her lithe body, oh, as though taking flight! / Swiftly, a dragon coiling, she twines and twists; / Then is a blossom, falling amid fine flurries.” For another contemporary literary depiction, sited at the Hsing-ch’ing Palace, see Wang Yung’s 王邕 (c.s. 751) “Ch’in-cheng lou hua-kan fu” 勤政樓花干賦, CTW 356.16b–17a; and, from the seventh century, Chang Ch’u-chin’s 張楚金 (fl. 630–80) “T’ou-ch’uang t’ung-erh fu” 透橦童兒賦, CTW 234.2a–3a. See also the account in MHTL, 17, again remarking a Hsing-ch’ing entertainment for Hsüan-tsung; and Chiao-fang chi chien-ting 教坊記箋訂, compiled by Ts’ui Ling-ch’in 崔令欽 (fl. 710–30?), ed. Jen Pant’ang 任半塘 (Shanghai: Chung-hua shu-chü, 1962), 46–48. In our poem, tu-lu is evidently a proper name, modifying hsün ch’uang 尋橦 (“pole-climber”), in parallel with the next line’s proper name Kung-sun. 136  The emperor’s birthday was the fifth day of the eighth month. It was proclaimed from 729 a three-day national holiday, pursuant to a memorial submitted by Chang Yüeh 張 說 (667–731) and Sung Ching 宋璟 (663–737), prime ministers (ch’eng-hsiang 丞相) of

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1407

with the sword and was proclaimed at the time the “intrepid wonder.”137 Also, joined platforms were set out, upon which horses were made to dance. The horses were caparisoned in taffeta silks, decked out with grelots large and small. They tossed their heads and shook their forelocks, lifting their hooves and flaring their tails, changing their

the “left” and “right,” respectively; see CTW 223.11a–12a. Schafer termed such p’u-huei 酺會 holidays “bacchanals”; see his “Notes on T’ang Culture, II,” Monumenta Serica 24 (1965): 130–34. For an overview of the birthday entertainments presented below the Ch’in-cheng lou, see MHTL, 2.23, and esp. Harada Yoshito 原田淑人, “Senshūsetsu enraku kō” 千秋節 宴樂考, in Shiratori hakushi kanreki kinen: Tōyōshi ronsō 白鳥博士還曆記念: 東洋史論 叢 (Tokyo: Iwanami, 1925), 833–43, with four plates. 137  On Madame Kung-sun’s renowned sword-dance, see T’ai-p’ing yü-lan, 569.6a and 575.8b, citing a passage from MHTL that is not in the current text; the former quotation, however, is nearly identical with a note found in the “dance” (wu-kung 舞工) section of Tuan Anchieh’s 段安節 (fl. 894–98) Yüeh-fu tsa-lu 樂府雜錄, which is in turn given as source of the lost passage in MHTL, “i wen,” 37; see Hung Wei-chu 洪惟助, ed., Tuan An-chieh Yüeh-fu tsa-lu chien-ting 箋訂 (Taipei: Chung-hua hsüeh-yüan, 1972), 125. The most widely known reference to Kung-sun’s dance is Tu Fu’s famous poem, “Kuan Kung-sun ta-niang ti-tzu wu chien-ch’i hsing” 觀公孫大娘弟子舞劍器行), Tu shih hsiang-chu, 20.1815–18, in which the poet, seeing in 767 the performance of an already middle-aged former student of the great dancer, recalls with emotion his viewing fifty years earlier Madame Kung-sun herself performing her celebrated dance. In his lengthy preface to the poem, Tu Fu remarks that Kung-sun was the only one of all the many artistes to perform at court who knew how to do the sword-dance, and also reports how the work of the great calligrapher Chang Hsü 張旭 (ca. 700–750) was said to have been influenced by watching Kung-sun’s vigorous and unreserved dancing. Tu Fu’s poem begins: Long ago there was a lovely lady, née Kung-sun, Who quickened the four quarters whenever she danced with the sword. Observers, many as the hills, turned pale of face, aghast, 4 As heaven and earth seemed for long to drop and lift with her movements. Streaks of light—like Yi as he shot the nine suns down; Poised high—like a host of gods hovering with teams of dragons. Advancing, like rolling thunder gathering in a crashing rage; 8 Halting, like river and sea freezing into a clear glow. The rest of the poem turns to a meditation on how quickly the years have passed. For full translations, see Erwin von Zach, Tu Fu’s Gedichte, ed. J. R. Hightower, 2 vols. (Cambridge, Mass.: Harvard Univ. Press, 1952), 2: 670–72; and William Hung, Tu Fu: China’s Greatest Poet (Cambridge, Mass.: Harvard Univ. Press, 1952), 251–52. There is some dispute about whether the dance was a pantomime without real swords or did in fact involve one sword, or even two. See Ou-yang Yu-ch’ien’s comments in T’ang-tai wu-tao, 106–9, for a summary of opinions. It seems most likely that a real sword was indeed used. There would be little reason otherwise to praise her intrepidity.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1408

Kroll

gait and shifting their aspect, completely in keeping with the concord of the notes.138 Also, palace artistes with their chignons combed as nine-fold cavalier-transcendents,139 clothed in vestments of peacock and halcyon feathers, with pendent necklaces of seventreasure gemstones,140 performed in the manner of “Rainbow Skirts and Feather Vestments.” When the melody ended, jewels and halcyon feathers could be swept up from the ground. Of the dancing horses, a number were taken in [after the capture of Ch’ang-an by his forces] by An Lu-shan who exercised them for his personal amusement. Afterward, when T’ien Ch’eng-ssu 田承嗣 had supplanted An, one still survived. One day, upon hearing the sound of drums while in the stable, it stamped and nodded to the music. The stable-boy was horrified at this and lifted a broom to strike it, at which the horse, thinking the anger was owing to its not doing beautifully enough, redoubled its effort with dip and swerve, until the music brought its posturings to a close. Frightened, the stabler reported this to Ch’eng-ssu, who considered the animal bewitched and accordingly had it flayed. From which time the dancing horses were no more.141

Lu-shan, during this time, attended at the autocrat’s side; A painted screen of a golden cockerel fronted the ‘covert for second thoughts.’142 In embroidered drapes and swaddling clothes he was each day more lusty and strapping; 138  On these wonderful animals, originally a troop numbering one hundred, see P. W. Kroll, “The Dancing Horses of T’ang,” T’oung Pao 67 (1981): 240–68. The “upheld platforms” mentioned in line 107, on which the horses sometimes did their posturings, were hefted by strong men. The tune, played by drums and flutes, to which they danced was called “The Upturned Cup” (ch’ing-pei 傾盃), at the close of which the horses lifted in their mouths a wine-cup and tossed back their heads—“bottoms up!”. 139  That is, resembling all manner of transcendent persons. A standard listing of the nine ranks of hsien 仙 includes “highest transcendents” (shang-hsien 上仙), the “eminent” (kao-hsien 高仙), the “great” (ta-hsien 大仙), the “mystical” (hsüan-hsien 玄仙), the “celestial” (t’ien-hsien 天仙), the “realized” (chen-hsien 真仙), the “divine” (shen-hsien 神仙), the “numinous” (ling-hsien 靈仙), and the “accomplished” (chih-hsien 至仙); see Yün-chi ch’i-ch’ien 雲芨七籤 (HY 1026), 3.6b. 140  Identification of the “seven treasures” (Skt. saptaratna) varies in different Buddhist texts; a typical list, from the Lotus Sutra, includes gold, silver, vaidūrya, giant clam-shell, coral, pearl, and carnelian. 141  This story is essentially as told in MHTL, “pu-i,” 34–35; see Kroll, “The Dancing Horses of T’ang,” 244–46, for translation. 142  The “covert for second thoughts” ( fu-ssu 罘罳) was a screened area outside the doorway to court where, according to tradition, a subject might have a last, private moment of introspection ( fu-ssu 復思) before entering to lay his business before the ruler. See

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse 112

1409

Through sweetened words and sly devices he was ever more petted and  indulged.

Whenever His Highness sat down to a banquet he was sure to have Lu-shan seated to the side of His own place, set off by a golden-cockerel screen,143 and it was permitted him to sit with his legs spread apart.144 T’ai-chen, moreover, treated him as her child, sometimes adding to the sport by bundling him in swaddling clothes.145 His Highness, for his part, called him “Lu-boy” (Lu-erh 祿兒). Whenever Lu-shan entered the palace, he would make obeisance first to the Precious Consort and only afterward to His Highness. Upon being asked by His Highness the reason for this, he quickly replied, “Your subject is by birth a man from foreign realms, where the ritual is to make obeisance first to one’s mother and after to one’s father; this is why I do so.”146 Shih-ming shu cheng-pu 釋名疏證補, ed. Wang Hsien-ch’ien 王先謙 (1896 ed., rpt. Shanghai: Shang-hai ku-chi ch’u-pan-she, 1984), sect. 17, 5.17b–18a. The term is also applied to the small tower sometimes placed above the corner of a city gate to screen the wall; see H. H. Dubs, The History of the Former Han Dynasty, vol. 1 (Baltimore: Waverly Press, 1938), 250–51, n.3. On the significance of the golden cockerel, see the following note. 143  A gilt cockerel was set atop a long pole by the palace gates during the formal announcement of imperial amnesties or “acts of grace” (she 赦). See des Rotours, Fonctionnaires, 364–65. The golden cockerel painted on the screen by which An Lu-shan sat signifies that his conduct enjoyed prior sufferance from the monarch and that he could therefore act however he wished, with full impunity. On this mark of Hsüan-tsung’s special favor, cf. KYTPIS, 2.102; ChTS 200A.5368; HTS 225A.6413; and Yüeh Shih’s 樂史 (930–1007) Yang T’ai-chen wai-chuan 楊太真外傳, in KYTPIS shih-chung, 2.141. 144  Sitting with legs spread apart (chi-chü 箕距) was traditionally regarded as crude and disrespectful. Even this liberty was granted An Lu-shan in the emperor’s presence. The most notorious incident of someone flaunting such an inappropriate posture is of course when Chuang-tzu was found by Hui-tzu sitting with legs spread apart, drumming on a plate and singing, after the death of his wife; see Chuang-tzu chi-shih 莊子集釋, ed. Kuo Ch’ing-fan 郭慶藩 (1844–96) (Peking: Chung-hua shu-chü, 1989), 18.614. 145  Lady Yang’s treatment of An Lu-shan as an “adopted” son—including the infamous incidents of her wrapping him in baby clothes and, three days after the proclamation of adoption, engaging playfully in the ceremonial “washing of the new-born” (TCTC 216.6903)—was seen by contemporaries and by later historians as nothing short of scandalous. Some have suggested this was a cynical means of licensing untoward intimacy, possibly extending even to sexual relations. E. G. Pulleyblank, The Background of the Rebellion of An Lu-shan (Cambridge: Cambridge Univ. Press, 1955), 97, regards the idea of a love affair as “too grotesque,” an opinion I choose to share. 146  Cf. An Lu-shan shih-chi 安祿山事跡 (hereafter ALSSC), by Yao Ju-neng 姚汝能 (fl. 840–60?) (Hsüeh-hai lei-pien edn.), 1.6a; tr. Robert des Rotours, Histoire de Ngan Louchan (Ngan Lou-chan che tsi) (Paris: Institut des Hautes Études Chinoises, 1962), 45; see also K’ai-T’ien ch’uan-hsin chi, 55.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1410

116

Kroll

Fiat ordered that by the high road a fine mansion for him be erected, Its buildings to be reached by steeds as fleet as flashing pheasants. The Inner Storehouse was laid wide open to supply and equip him at whim: Vessels of jade, golden canisters, and winnowing-baskets of silver.

At that time a fine mansion was erected for Lu-shan’s sake, by the south cross-street of the Ch’in-jen 親仁 ward.147 Trusted persons of the palace administration were commanded to oversee the task and even told [by the emperor], “My stewards, do your very best in arranging and deploying things. Lu-shan’s eyes are huge, and I would not have him mock me.”148 Even such items as containers, winnowing-baskets, and pots were all made of gold and silver. At present the Belvedere of Turning to the Prime (Hui-yüan kuan 回元觀) is on the former site of the mansion.149

His contrary schemes smoldered unseen, as he hastened back to his post;150 118 The belt bestowed him when he attended court stuffed in ten full  girths. Lu-shan’s grossness and obesity were unexampled. His paunch hung down flabbily, and his belt measured fifteen girths before it could enclose his bulk.151 147  The Ch’in-jen ward was in the east-central sector of Ch’ang-an. On An Lu-shan’s residence there, see T’ang liang-ching ch’eng-fang k’ao, 3.60–61; CAC 8.5b–6a. 148  “His eyes are huge” means Lu-shan sees and wants everything. The emperor does not wish to be mocked for being stingy. Cf. HTS 225A.6413, and T’ang liang-ching ch’eng-fang k’ao, 3.60. 149  Accepting the TSCS reading, instead of the CTS 四元觀. Sung Min-ch’iu, writing in the eleventh century, had the name of this Taoist abbey as Hui-yüan kuan and cited Cheng Yü’s poem as his source (CAC 8.6a), thus indicating that the error must have entered the text sometime later. However, the ninth-century ALSSC 1.7b, gives the name as Hsüanyüan 玄元, a very common Taoist term, which may be the true title of the establishment. 150  As the supreme military governor of the northeast frontier, with his headquarters at Fan-yang 範陽 (in the vicinity of present-day Peking), An Lu-shan had roughly 150,000 troops at his command, most of whom were permanent, professional soldiers—many of non-Chinese origin—who had served under him for a decade or more. His visits to court ended in 754 when, fearing the increasing enmity of the prime minister, Yang Kuo-chung, he left the capital in alarm and returned to his base post-haste. See ChTS 200A.5369–70; HTS 225A.6416; TCTC 217.6924–25. 151  Other sources claim that in the later years of T’ien-pao his stomach sagged down to his knees and that he required servants to support him under each shoulder in order to walk. See ChTS 200A.5368; HTS 225A.6413. His weight is estimated in ChTS, loc. cit., as 330

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

120

1411

Lord Chang, a loyal magnate, had discerned the look of rebellion; Urgently he protested, day after day, but His Highness brooked no  suspicion.

Chang Ch’ü-chiang 張曲江 had first discerned the look about Lu-shan of one certain to turn in rebellion. On numerous occasions he spoke of it to His Highness, but His Highness would say, “My good man, you should not wrong Lu-shan by seeing in him what Wang I-fu 王夷甫 discerned in Shih Le 石勒.”152

A pool was prepared, An was summoned to soak, but now he did not come;153 And T’ung Pass finally spilled over with legions from out of Yü-yang.154 catties (chin 斤) or approximately 500 pounds. ALSSC 1.7a, and Yang T’ai-chen wai-chuan, 2.141, up it to 350 catties. An anecdote often joined to the report of An Lu-shan’s sagging belly has Hsüan-tsung asking, “What is there in your barbarian stomach that makes it so large?” To which Lu-shan winningly replied, “Only a true heart and nothing else!” See K’aiT’ien ch’uan-hsin chi, 55; HTS 225A.6413. 152  Chang Ch’ü-chiang is Chang Chiu-ling 張九齡 (678–740). On his recognizing at first meeting that Lu-shan was destined to bring about calamity, see An Lu-shan shih-chi, 1.2a/b (tr. des Rotours, Ngan Lou-chan, 18); and TCTC 214.6814–16. The former text gives the date as 733 (emending 開元十一年 at 1.2a6 to 開元二十一年), the latter as 736. Both include the emperor’s misguided analogy involving Shih Le and Wang I-fu. Shih Le (274–333) was founder of the Later Chao 後趙dynasty (329–52), the second of the non-Chinese “Sixteen States” (shih-liu kuo 十六國) that ruled in North China after the Chin 晉 lost Lo-yang and fled south in 311. Shih Le was a Hsiung-nu who originally served under the Liu 劉 rulers of the Former Chao (first of the Sixteen States), before supplanting them and declaring his own dynasty. Chin shu 晉書 (Peking: Chung-hua shu-chü, 1974), 104.2707, records that when Shih Le was fourteen, he was one day in Lo-yang, trading. Wang Yen 王琰 (tzu I-fu; 256–311), a great ch’ing-t’an 清談 champion of the time and perceptive judge of character, caught sight of the boy and, detecting an unusual aura about him, told his companions that this boy would in future bring disaster to the realm. Wang’s prediction was only too true, and he himself—as a Chin official—was later captured and executed by Shih Le when the barbarian forces overran Lo-yang. The analogy with An Lu-shan is neat and, of course, Hsüan-tsung does not realize how apt it is as he rebukes Chang Chiu-ling with it. Cf. ALSSC 1.2b; TCTC 214.6814. KYTPIS 2.89 has an anecdote telling of Chang Chiu-ling’s prescience regarding the eventual fall of Yang Kuo-chung and the expected downturn of favor for those who seek to “warm themselves by his fire”; but this is anachronistic, since Kuo-chung had by the time of Chiu-ling’s death in 740 gained no significant power. 153  The bathing pool mentioned here refers to a newly fashioned pool at the Hua-ch’ing compound. 154  Yü-yang 漁陽, about seventy miles east of present-day Peking, here designates An Lushan’s military base of operations. The T’ung Pass 潼關, about seventy-five miles east of

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1412

Kroll

On the imperial avenue, throughout the night, was no drumbeat of interdiction; 124 The jade conveyance stirred smartly away, speeding westward and  south.155 That year an envoy, sent to bestow on Lu-shan a gift of sourpeel tangerines, returned to report tremulously that Lu-shan had the look of revolt, confessing, “I barely was able to return here alive.” His Highness still doubted such words and dispatched another envoy to notify Lu-shan that, “I have provided a bathing pool for you and am awaiting your arrival.”156 When the envoy returned, he acknowledged Lu-shan’s look

Ch’ang-an, protected access to the capital from the central plains. When it fell to the rebel forces—after a several-months-long stalemate—on 10 July 756, the capital’s last line of defense was gone and An Lu-shan’s revolt, begun in Yü-yang on 16 December of the preceding year, was ready to reach its crest. Note Cheng Yü’s word-play with i 溢, “spill over; flood,” continuing in a sinister way the bathing imagery of the foregoing lines. 155  Hsüan-tsung, along with those family members, high officials, and courtiers closest to him, and escorted by about a thousand cavalrymen, abandoned Ch’ang-an in the predawn hours of 14 July 756. This he did after making a public declaration from the Ch’incheng lou on the preceding day that he would personally lead out the capital armies to fight the rebels now advancing on the eastern approach to Ch’ang-an. The hoped-for destination of the fleeing imperial retinue was Ch’eng-tu, which, as we have seen (n. 99 above), Yang Kuo-chung had for several years been provisioning and fortifying as a potential refuge if court politics turned against him. The fullest account of the events immediately preceding Hsüan-tsung’s departure from the capital and those of the memorable first days of the flight is to be found in TCTC 218.6970–74; for translation of the text and commentary, see Kroll, “The Flight from the Capital and the Death of Precious Consort Yang” (n. 58 above). 156  In ALSSC 2.4a (des Rotours, JVgan Lou-chan, 149), the first envoy sent to Fan-yang, identified as Fu Ch’iu-lin 輔璆琳, is said to have received a bribe from An Lu-shan and to have brought back, consequently, a favorable report, whereas it was the second envoy (2.6b–7a; des Rotours, 166), identified as Feng Ch’eng-wei 馮承威, carrying the invitation for An Lu-shan to come and sport in the new pool, who reported that he had barely returned with his life. Cf. TCTC 217.6930–32, 6933–34, including in the k’ao-i versions of these events from other sources available to Ssu-ma Kuang but now lost, such as the Ming-huang shihlu 明皇實錄 and Hsüan-tsung hsing Shu chi 玄宗幸蜀記, which mainly agree with the extant ALSSC accounts. Ssu-ma Kuang dates the visits of both imperial envoys to 755. Note that in his main text Ssu-ma Kuang names the second envoy as Feng Shen(神)-wei. ChTS 200A.5369–70, places the mission of Fu Ch’iu-lin in 753, and gives no details of a second mission to An Lu-shan, other than to say that in 755 “Hsüan-tsung again summoned him, but he pleaded illness and did not come.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1413

of revolt. Only after this did His Highness grow distressed with doubt; but the marauding army had already reached T’ung Pass.157

On gazing back at the Nine Gates—dust and soot was deepening;158 The Six Dragons were driven till dark—troops and paladins were spent.159 Of the district officers there were none that had provided for His army’s halts; 128 So the rooms of the traveling palace must be stripped, His wyvern of the  clouds slaughtered. At that time the territory of the surrounding countryside was in terrific turmoil, and there were no proper dispositions for the autocrat’s halts. His Highness ordered the wood stripped from the traveling palace [for use as firewood] and his own horse butchered in order to supply refection for the soldiers.160

Before the Ma-wei post-station, the royal rig could not set off; The minister of state was shot at and slain by those who sought vengeance.161 157  The timing is exaggerated for effect. The second envoy had to have completed his mission before An Lu-shan’s army made its first strike in late December 755. The taking of T’ung Pass and fall of Ch’ang-an did not occur till mid-July 756. 158  “Nine Gates” is synecdoche for Ch’ang-an, there being three main gates each in the city’s eastern, southern, and western walls. There is also classical precedent for the term’s referring to the emperor’s palace. 159  The “Six Dragons” is traditional metonymy for the horses that draw the imperial carriage. 160  Cf. TCTC “k’ao-i,” 218.6972, quoting from T’ang li 唐曆 by the court historiographer Liu Fang 柳芳 (fl. 735–70); tr. Kroll, “The Flight from the Capital,” 42. The incident took place at mid-day on 14 July, when the progress reached the Wang-hsien kung 望賢宮 at Hsienyang 咸陽, about thirteen miles west of the capital. Although an envoy had been sent ahead to inform the district magistrate of the emperor’s coming, by the time of arrival all local officials had fled—a pattern that was to continue. “Traveling palace” (hsing-kung 行宮) denotes any temporary dwelling place or rest-house for the emperor. “Wyvern of the clouds” (yün-ch’i 雲螭), i.e., a sky-soaring dragon, refers to the emperor’s horse. 161  For the fateful happenings of 15 July at the Ma-wei 馬嵬 post-station, some thirty miles west of Ch’ang-an, see TCTC 218.6973–74; tr. Kroll, “The Flight from the Capital,” 48–52. When Yang Kuo-chung was seen conversing on horseback with a group of Tibetan guards, an angry member of the imperial cavalry let fly an arrow that struck his saddle. Yang tried to flee but was pursued and killed by soldiers under the overall command of Ch’en Hsüanli 陳玄禮. General Ch’en then demanded the death of Yang Kuei-fei, as the root of all

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1414

Kroll

Then elongate eyebrows and deep-black hair turned to clotted blood;162 132 Of no avail was the sovereign king, who could only weep to himself. Later, from over Blue Mud Slope, he reached the land of the Three Shu;163 And beside the walls of the Metal Embankment he rested the Nine Banners.164 In the text of a despatch was a tearful oblation for the tomb of a former vassal; 136 To a measured tune sadly was sung the lyric on the wildgeese of autumn.

troubles; otherwise, the troops would go no farther with Hsüan-tsung. The emperor could do naught but consign his love to death, ordering (with her consent, all sources say) that she be strangled with a silk cord wielded by Kao Li-shih, his chief eunuch and long-time confidant. Only when her corpse was shown to Ch’en Hsüan-li and his lieutenants would the troops agree to continue the journey. However, by this time the heir-designate, Li Heng 李亨 (pht. Su-tsung 肅宗), was persuaded to break off with a contingent of soldiers and head northwest, in order to organize loyalist resistance to the rebels. Hsüan-tsung and a diminished retinue trekked on to refuge in Ch’eng-tu as planned, arriving there on 28 August. Messengers from Li Heng reached him there on 10 September, informing the old monarch that his son had declared himself emperor on 12 August, three days after making Ling-wu 靈武 (present-day Yin-ch’uan 銀川, in Ninghsia) his base of operations. Hsüan-tsung, now a broken man, bowed to the run of events and issued a rescript on 14 September, formally abdicating the throne. 162  Cf. lines 37–38 of Po Chü-i’s “Ch’ang-hen ko”: “And then the Six Armies would go no farther—there was no other recourse / But the fluently curved moth-eyebrows must die before the horses.” P. W. Kroll, “Po Chü-i’s ‘Song of Lasting Regret’: A New Translation,” 98 (see n. 2 above). 163  Blue Mud Slope (Ch’ing-ni pan 青泥阪), northeast of Lüeh-yang 略陽 in western Shensi, leading to a pass that brought one to the northeastern entrance to Shu, was the site of a post-station. Li Po exclaimed, in his “Shu tao nan” 蜀道難, Li Po chi chiao-chu, 3.199, “So twisted and tortuous is Blue Mud Pass—/ Nine turnings every hundred paces wind round its rugged crest.” The “Three Shu” designates Shu commandery (Shu chün 蜀郎), along with the regions of Kuang-han 廣漢 and Ch’ien-wei 犍爲. For a detailed study of the emperor’s route to Ch’eng-tu, see Okano Makoto 岡野誠, “Tō Gensō no Shoku mōjinji ni tsuite” 唐玄宗の蜀蒙塵路について, Meiji Daigaku Shakaikagaku Kenkyūjo kiyō 明治 大学社会科学研究所紀要 32 (1993): 55–67. 164  Hsüan-tsung has now reached his stopping-point. The Metal Embankment (Chin-t’i 金堤) was constructed by the third-century bc governor of Shu, Li Ping 李冰, who channeled the Min 岷 river into two branches around the city of Ch’eng-tu. “Metal” simply suggests the strength of the stone walls. “Nine Banners” (chiu-ch’i 九旂) connotes all the various flags and pennons of the imperial retinue.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1415

When the royal rig arrived in Shu, an edict was given to trusted persons of the palace to speed away and make oblation at Chang Ch’ü-chiang’s tomb, out of remorse that his protests had not been heeded.165 Again, when passing below Saber Gallery,166 the emperor, gazing out upon the hills and streams, suddenly recalled aloud the lyrics to the “Melody of the Waters” that went: “Hills and streams fill up one’s eyes, and tears wet one’s cloak; / Wealth and esteem, honor and glory, last for how long a time? / Don’t you see, just at present, above the Fen River, / There are autumn’s wildgeese flying as they do year upon year.” His Highness wept, his face glistening with moisture, and turning round to his attendants asked if anyone knew who the author of this poem was. “It is a poem by Li Chiao,” replied a follower. His Highness tried to hide his tears, and he said, “Li Chiao may truly be called a genius.”167

165  Cf. lines 119–20 and n. 152 above. Chang Chiu-ling’s tomb was located at the foot of Mount Lo-yüan 羅源山, outside the present-day city of Shao-kuan 韶關 in northern Kwangtung. This tomb was excavated in 1960. For reports, see Yang Hao 楊豪, “T’ang-tai Chang Chiuling mu fa-chueh chien-pao” 唐代張九齡墓發掘簡報, Wen-wu 文物 1961.6: 45–51; and Okazaki Takeshi 岡崎敬, “Tō Chō Kyūrei no funbo to sono boshimei” 唐張九齡の墳墓 とその墓誌銘, Shien 史淵 89 (1962): 45–83. 166  Saber Gallery (Chien-ko 劍閣) is the name of a lofty pass in northeastern Shu, which must be traversed by a narrow cliffway, in order to come into the central region of the province. It was traditionally regarded as the natural boundary between the old state of Ch’in (Shensi) and that of Shu (Szechwan). 167  These lines are the concluding quatrain of Li Chiao’s 李嶠 (644–713) “Fen-yin Ballad” (汾陰行), a poem of forty-four lines describing Han Wu-ti’s visit to Fen-yin long ago, an event that briefly—all too briefly—bestowed glory and renown on the place, which was only to be forgotten again as the centuries advanced. The perpetual flux of earthly splendors, against the recurring cycle of nature itself, is the poem’s ground-note. The final four lines, summing this directly, were apparently set as lyrics to the plaintive “Melody of the Waters” (see n. 115 above). Some other sources record a variant version of this anecdote, in which the emperor, just prior to abandoning Ch’ang-an, enjoyed a poignant last party at the Hsing-ch’ing Palace. As one of the performers sang the “Melody of the Waters,” Hsüan-tsung wept feelingly over the four lines in question and asked who had composed them. Upon being told they were Li Chiao’s creation, he praised the poet as a genius and departed, overcome with emotion, before the tune came to its end. Tz’u Liu-shih chiu-wen 次柳氏舊聞, in KYTPIS shih-chung, 7; MHTL, 41; also CTS 57.690. Cf. TSCS 10.145, which combines both versions into a longer narrative. Whichever version one prefers, the point hits home: the emperor suddenly realizes the truth in Li Chiao’s lines—that all glory, even his own, must necessarily slip away (is slipping away, has already).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1416

Kroll

A year later the Magistral Father forwarded a writ of triumph,168 Having repurified the outlooks and gatetowers, regathered the royal domain. Then the two sovereigns met each other at the halt of Viewing-theWorthies; 140 Sovereigns and subjects, amid drumming and dancing, all gave way to sobs. The Palace for Viewing the Worthies (Wang-hsien kung 望賢宮) was located several li to the east of Hsien-yang.169 At this time, when Ming-huang returned from Shu, Su-tsung 肅宗 welcomed his carriage there. His August Highness delivered over to His Highness the state-transmitting seal,170 which His Highness, giving way to sobs, accepted in obeisance. All the attendants and acolytes wept, and said, “We could not have hoped for this day—to see again the two sovereigns, meeting each other in ceremony.” As his rig was about to enter the Gate of Opening-to-the-Distance (K’ai-yüan men 開遠門),171 His August Highness was uncertain lest the precedence regarding who should enter the gate first not be correct. Looking round, he inquired of the vassals in his entourage, but they could make no reply. Kao Li-shih came before him and said, “Although Your August Highness is revered, you are now a subject; though the August Thearch is your son, he is now the ruler. Your August Highness should advance first, but should enter through the side gateway; the August Thearch should enter through the main gate, but proceed afterward.” With elders and seniors all shouting “A myriad years!” everyone at the time thought this was correct.172 168  The honorific title “Magistral Father” (shang-fu 尚父) was classically associated with Lü Wang 呂望 (alias Lü Shang 呂尚, T’ai-kung Wang 太公望), counselor to King Wen and, later, King Wu during the founding of the Chou dynasty. The phrase is here applied to the great general Kuo Tzu-i 郭子儀 (697–781), who was responsible for retaking Ch’ang-an from An Lu-shan’s rebels in November 757, as well as Lo-yang a month later. Kuo’s distinguished military career spanned a quarter of a century and concluded with supreme honors bestowed by Te-tsung 德宗 (reg. 779–805), including conferral of the title shang-fu in July of 779, a month after that sovereign took the throne. See ChTS 120.3465, HTS 137.4608, TCTC 225.7259. Cheng Yü’s use of the title here is technically anachronistic. 169  The Wang-hsien kung lay between Ch’ang-an and the old site of Hsien-yang, a few miles northwest of the T’ang capital. It may have been meant originally as a stopping-place for imperial journeys to the dynastic tombs beyond Hsien-yang and across the Wei River. 170  The “state-transmitting seal” (ch’uan-kuo hsi 傳國璽) refers to the seal of imperial authority, presumably modeled on the famous ch’uan-kuo hsi of Ch’in Shih Huang-ti. See T’ai-p’ing yü-lan, 682.1b–7b. 171  This was the northernmost of the three gates in Ch’ang-an’s west wall. 172  Ch’ang-an was recaptured by imperial troops, significantly aided by Uighur contingents, on 13 November 757. Lo-yang was recaptured on 3 December. On 4 December, Su-tsung set out from his headquarters in Feng-hsiang 鳳翔, about seventy miles west of Ch’ang-an,

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1417

From the palace the monarch called in person for Kao of the Alert Cavalry,173 Ordered in stealth the burial to be changed of Realized Consort Yang. Her flower-fresh skin and snow-white allure could no longer be seen; 144 In vain was there a scented sachet that comported a dampness of  tears. At this time Su-tsung gave a fiat, ordering the burial-site of T’ai-chen to be changed. Kao Li-shih knew the place of her inhumation, a dozen or so paces northwest of the Ma-wei post-station. At that earlier time, in the haste in which [Hsüan-tsung] sped off in his carriage, no appropriate disposition was made for her corpse, which was simply wrapped up in a purple bed-roll. When Kao Li-shih came to move her body for reburial, he found it moldering and decayed. There was, however, a scented sachet, of purple embroidery, upon her bosom; it still gave off the aroma of curdled musk. This he took and presented to His August Highness, who wept on seeing it and wore it thereafter as a pendant.174 to return to the capital. Four days later he re-entered Ch’ang-an amid general rejoicing. Hsüan-tsung, returning from Shu, reached Feng-hsiang on 6 January 758. The TCTC account of his return is worth quoting in extenso. For this, see Appendix 2 below. 173  This is Kao Li-shih (see nn. 85 and 161 above). He received the title P’iao-ch’i ta chiangchün 驃騎大將軍, Great General of the Cavalry on the Alert, in 748. See A. F. P. Hulsewé, China in Central Asia, The Early Stage: 125 B.C.–A.D. 23; An Annotated Translation of Chapters 61 and 96 of the History of the Former Han Dynasty (Leiden: Brill, 1979), 74 n. 35, for an explanation of this title, first bestowed by Han Wu-ti in 121 bc on Huo Ch’ü-ping 霍去病 (ca. 146–117 bc). Both Dubs’ rendering of p’iao-ch’i as “Agile Cavalry” in his volumes of translations from the Han shu (as noted by Hulsewé) and Schafer’s Tolkienesque rendering as “Roan Rider” (in a passage referring to Kao Li-shih) in his “Wu Yün’s ‘Cantos on Pacing the Void’ ”, HJAS 41 (1981): 380, are in error. 174  Although Cheng Yü is definite in stating the order for Yang Kuei-fei’s reburial was given by Su-tsung, other sources, most notably ChTS 51.2181 and Yang T’ai-chen wai-chuan, 143, say it was (as one would expect) Hsüan-tsung who sought to have her corpse moved from its makeshift grave. Su-tsung’s advisors, especially his eunuch confidant Li Fu-kuo 李輔國 and Li K’uei 李揆, vice-president of the Board of Rites, spoke against the idea, with the result that Su-tsung withheld his approval. Hsüan-tsung, however, is reported to have “commanded secretly a court commissioner” to rebury Lady Yang. Both these accounts include the details of the purple bed-roll and the perfume sachet. The descriptions of her decayed corpse are more graphic: “Her skin and flesh by now were moldering” (ChTS), and “Her skin and flesh by now were decomposing and falling away” (wai-chuan). The latter text has Hsüan-tsung preserving the sachet in his sleeve instead of hanging it from his waistband as a pendant. Cf. a quatrain by Chang Hu 張祜 (fl. 820–45) on “T’ai-chen’s Scented Sachet” (“T’ai-chen hsiang nang-tzu,” CTS 511.5844):

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1418

Kroll

When the simurgh-belled carriage again made its way to the Palace of Floriate Clarity, Red-pink drupes in the mountains around hung down in longing for her. Before the Hall of Drifting Frost, the moonlight was sinkingly silent; 148 Beneath the Pavilion to Welcome Spring, breezes whispered coolness.175 The Hall of Drifting Frost was the hall for repose, but in Po’s telling of “The Song of Lasting Regret” he takes the Hall of Protracted Life as the hall for repose, which is an egregious mistake.176 When [winter of] the next year [after the return to Ch’ang-an] came, His August Highness again graced with his presence the Hua-ch’ing Palace,

Of gold-passement, the little flowered sachet of the Consort; Upon her bosom, wasted now to ruin, it tied up an aromatic. Who, for the sake of the sovereign king, will loosen it again? Regrets for all the rest of one’s life will bind a feeling heart. 175  The Ying-ch’un t’ing 迎春亭 (Pavilion to Welcome Spring) is presumably the I-ch’un t’ing 宜春亭 (Pavilion Befitting Spring), located outside the K’ai-yang 開陽 (Opened to Yang) gate, the main gate on the eastern wall of the Hua-ch’ing compound. See CAC 15.8a, LTHC 2.35a. 176  Cheng Yü’s criticism is of lines 115–18 of Po Chü-i’s poem, in which the etherealized Lady Yang recounts to the Taoist thaumaturge-envoy of Hsüan-tsung the midnight vow that only she and the emperor could know: On the seventh day of the seventh month, in the Hall of Protracted Life, At the night’s mid-point, when we spoke alone, with no one else around— “In heaven, would that we might become birds of coupled wings! On earth, would that we might be trees of intertwined limbs!” (Kroll, “Po Chü-i’s ’Song of Lasting Regret’,” 100.) The Ch’ang-sheng tien (see n. 120 above) on Mount Li was a “purification hall” for quiet study, contemplation, and preparation for Taoist devotions; see CAC 15.9a, LTHC 2.36b–37a. The Fei-shuang tien (see n. 112 above) contained the imperial bedchamber; Sung Min-ch’iu paraphrases Cheng Yü’s original objection to Po Chü-i’s mistake; see CAC 15.8a. This matter is explored farther by Ch’en Yin-k’o, Yüan Po shih chien-cheng kao, 727–28. Ch’en notes that, although the retirement chambers in the Ch’ang-an and Lo-yang palace complexes were called Ch’ang-sheng tien, the Ch’ang-sheng tien on Mount Li was, exceptionally, a hall for purification rites and would not therefore have been the site for a nighttime rendezvous between the emperor and his consort. Since the lovers’ pledge was purportedly made in the seventh month, it would not have been made at Li Shan which was not usually visited in the summer or early fall. Thus, the Ch’ang-sheng tien referred to should be the emperor’s bedchamber in Ch’ang-an.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1419

where he sojourned with satisfaction and then went back [to the capital].177 Hence­ forth his residence was shifted to within the Western Interior.178

The snow-caped maiden was gone, but her cage of jade remained; The long-life deer was now gaunt, its bronze medallion dangling. On the ivory couch the dust was thick on Yen-sa’s coverlet; 152 Under painted eaves insects had webbed the plaque of crystal-glass. T’ai-chen had kept for pet a white parrot sent as tribute from the western lands, which was quickly clever and loquacious. His Highness was particularly fond of it, and dubbed it “Snow-caped Maiden” (Hsüeh-i nü 雪衣女).179 His Highness once captured a white deer in the Garden of Lotuses (Fu-jung yüan 芙蓉園).180 Only a hermit from 177  Hsüan-tsung’s visits with Lady Yang to the hot springs had normally taken place in winter. See T’ang Hua-ch’ing kung, 604–5, for a complete list of his removals there. In mid-winter of the first year following the court’s return to Ch’ang-an, he revisited the site of those idylls, staying nearly a month, from 20 November to 13 December 758. TCTC 220.7063. Cheng Yü’s comment makes it appear that Hsüan-tsung’s removal from the Hsing-ch’ing complex occurred immediately following his visit to the hot springs; but this did not happen till some twenty months later. 178  The “Western Interior” (Hsi-nei 西內) refers to the Palace of Grand Culmination (T’ai-chi kung 太極宮) located in the north-central section of Ch’ang-an. Before construction of the Ta-ming Palace (in relation, the Eastern Interior [Tung-nei 東內]), it had been the residential complex for the entire imperial family. Hsüan-tsung’s transferral from his beloved Hsing-ch’ing palace (the Southern Interior [Nan-nei 南內]) was forced on 22 August 760 by Li Fu-kuo, who had previously and unsuccessfully advocated it to Su-tsung on the pretext that Hsüan-tsung was having too much unsupervised contact with the populace. Having feigned an invitation from Su-tsung requesting Hsüan-tsung’s presence at an outing at the T’ai-chi palace, Li Fu-kuo confronted Hsüan-tsung’s party outside the gate of the Hsing-ch’ing complex with five hundred horsemen showing unsheathed swords. He advised the emeritus emperor that Su-tsung wished his permanent removal to the T’aichi palace. Kao Li-shih spoke up for his frightened master, temporarily shaming Li Fu-kuo and his troops, and then proceeded to lead Hsüan-tsung’s horse, holding its bridle protectively, to the Western Palace. Hsüan-tsung there took up residence at the Sweet Dew Hall (Kan-lu tien 甘露殿), with a much reduced suite of attendants and guards, soon minus even Kao Li-shih who was forbidden to stay with his long-time master. Hsüan-tsung obediently professed that the move had been his wish all along. The deed being done, Sutsung pardoned and commended Li Fu-kuo’s actions. See TCTC 221.7093–95. 179  Yang Kuei-fei’s pet parrot (actually a cockatoo) was painted by the artist Chou Fang 周昉 and was said to have once been released by her to upset purposely a game-board when the emperor was about to lose to an opponent. See Schafer, Golden Peaches, 101. 180  See LTHC 2.34b–35a; also T’ang Hua-ch’ing kung, 533.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1420

Kroll

the mountains, Wang Min 王旻, could identify it; he said, “This is a deer from Han times.”181 His Highness thought it extraordinary and ordered his attendants to examine it thoroughly. In the snowy fur at the base of one of its antlers was discovered a small bronze medallion, carved with the words “White deer from the Befitting Spring Park” (I-ch’un yüan chung pai-lu 宜春園中白鹿).182 Because of this, His Highness became even fonder of it. He had it moved to the northward mountain and referred to it as “Transcendent Visitor” (Hsien-k’o 仙客).183 When His Highness stopped over at Huach’ing, the Princess Yen-sa 罨颯184 once was summoned by him at daybreak to hear him try out a new version of “The Melody of the Waters.”185 The Princess, who was inordinately fond of sleeping late, came out flustered with a pearl-bejewelled coverlet around her. [Some years afterward, ] upon the arrival of [An Lu-shan’s] marauders, she left the palace in fear and frenzy, trailing after the royal rig; later she became unbalanced.186 Upon His Highness’s homecoming to the Southern Interior, one morning he went again into her rooms, and there was Yen-sa’s coverlet from that time past, quite the same but with dust piled on it. His Highness was affected exceedingly by it. The [inscription-] plaque at the Hall of the Hot Springs (Wen-ch’uan t’ang 溫泉堂) was of a mineral of hyaline transparency, through which one could see a man’s shape and shadow.187 Within the palace it was referred to as the “crystal-glass plaque.” 181  Wang Min was a Taoist adept and master of occult practices, reputedly several centuries old when he was summoned to court by Hsüan-tsung. It was thus no wonder that he recognized this immortal deer. For Taoist biographies of him, see Li-shih chen-hsien t’itao t’ung-chien 歷世真仙體道通鑑 (HY 296), 32.3b–4b, and Mao-shan chih 茅山志 (HY 304), 15.10a/b; other incidents are recorded in T’ai-p’ing kuang-chi, 72.144c/d, quoting the T’ien chi wen 田紀聞. The identification of the long-lived deer is elsewhere attributed to Chang Kuo 張果, for whom see n. 196 below). 182  The I-ch’un Park was established during the Ch’in dynasty and was continued under the Han. On its old site were laid out the T’ang Serpentine (Ch’ü-chiang 曲江) and the Garden of Lotuses (Fu-jung yüan 芙蓉園). T’ang liang-ching ch’eng-fang k’ao, 3.92. 183  The “northward mountain” refers to the area of Li Shan north of the Hua-ch’ing complex. “Transcendent visitor” (hsien-k’o 仙客) is a stock term for a superior Taoist master, regarded as visiting this world from far off in the heavens and existent from long in the past. 184  I have been unable as yet to locate any other information about this unfortunate and tardy princess. If she caught up with Hsüan-tsung’s train, she was probably not one of the mere twenty-four palace women who made it all the way to Shu with the fleeing emperor; ChTS 9.234. 185  See nn. 115 and 167 above, for this famous piece of music and poetry. 186  “Unbalanced” is pu-chih-hsing 不知省, literally “unreflective,” “incapable of examining oneself,” i.e., non compos mentis. 187  The plaque was reputedly made for the nobleman Yüan Ch’ang 元萇 of the Northern Wei, when prefect of Yung-chou 雍州 around 512–14; for Yüan’s official biographies, see Wei

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1421

Caltrop-blossoms of cyan-blue cover Cloud-Mother Tumulus;188 Storm-torn bamboo and chrysanthemum droop, staggered and snapped off. The valance of the Realized One’s icon is grown about with weeds; 156 To the medicine hall of Kuo the Venerable it is now futile to close the  door. The Realized One, Li Shun-hsing 李順興, had cultivated the Tao on the north mountain in the time of the Latter Chou 後周.189 When the August Thearch of [the Clan of] the Divine Yao accepted the abdication [of the last Sui ruler],190 the Realized One gave

shu 魏書 (Peking: Chung-hua shu-chü, 1974), 14.351–52, and Pet shih 北史 (Peking: Chunghua shu-chü, 1974), 15.548. The plaque was preserved at least into the Sung dynasty, at the nearby Abbey of the Numinous Fountain (Ling-ch’üan kuan 靈泉觀); LTHC 2.33b–34a. See Schafer, Golden Peaches, 235–37, for T’ang “glass” and “paste.” Transparent glass, of the type out of which this inscribed plaque was made, was considered an exotic item in T’ang times. 188  Although the syntax and general sense of this line are clear enough, I cannot identify the specific referent of the “Mica Tumulus” (Yün-mu ling 雲母陵), if it is indeed a proper name. Perhaps it is an elegant epithet (mica signifying the richness of the mountain) for the slope on which the Hua-ch’ing palace was located. The caltrops are puzzling too: are they, as water plants, growing wild now in the bathing pools or are they proliferating in the streams outside the palace compound? 189  On Li Shun-hsing, a Taoist adept who could foretell the future, see Pei shih, 89.2929, and, more circumstantially, Li-shih chen-hsien t’i-tao t’ung-chien, 30.9a–12a. The latter source places the date of his transcendence from this world in 540, in his thirty-eighth year; the Pei shih has it sometime after 547. Both agree that Mount Li was his last earthly residence, Pei shih averring that he had been granted two mou (about a third of an acre) of land near the hot springs by Wen-ti 文帝 (reg. 535–51) of the Western Wei 西魏 dynasty. Western Wei became the Northern (also Latter) Chou in 557. 190  The full title of the T’ang founding emperor, Li Yüan 李淵 (566–626), posthumously Kaotsu 高祖 (High Progenitor), was Shen-Yao ta-sheng ta kuang-hsiao huang-ti 神堯大聖大 光孝皇帝 or “Greatly Glorious and Filial August Thearch and Great Sage of [the Clan of] the Divine Yao.” This is the title that was conferred by order of Hsüan-tsung in early 754 (see ChTS 1.18). The legendary sage-king Yao was said to have been enfeoffed in the state of T’ang, which the Li clan took as their ancestral seat (hence the designation of Li Yüan’s dynasty). Kao-tsu formally accepted the abdication of Yang Yu 楊侑 (pht. Kung-ti 恭帝, the thirteen-year-old grandson of Sui Yang-ti and puppet monarch for six months (he had been placed on the throne by Li Yüan when the latter captured Ch’ang-an or Ta-hsingch’eng 大興城, as it was then called), on 18 June 618, becoming the first emperor of the T’ang dynasty.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1422

Kroll

notice in secret of talismanic concurrence.191 There is to this day a fane to him below the mountain.192 Within the palace was the Temple of the Seven Paragons193—from the Divine Yao to Jui-tsung 睿宗 and his Empress Tou 竇后,194 each was represented, wearing imperatorial robes. The pomegranate trees around the temple, every one of them planted by T’ai-chen herself, have all grown rank and rife. To the south was the Close of Pious Works,195 in whose midst were to be found both an altar-platform of

191  This signifies spiritual support of and assent to Kao-tsu’s assumption of the throne. Apart from this comment of Cheng Yü, there is no other extant reference to the apotheosized Li Shun-hsing validating Kao-tsu’s imperial accession. On Taoist prophecies and confirmation of T’ang rule, see Stephen R. Bokenkamp, “Time After Time: Taoist Apocalyptic History and the Founding of the T’ang Dynasty,” Asia Major 3rd ser. 7 (1994): 59–88. 192  It was still there in 1776. See LTHC 3.19b. 193  It stood in the north part of the compound, north of the sixteen “Long Thermae.” LTHC 3.34a/b. 194  The “Seven Paragons” traditionally refers to one or another grouping of ancient sage-rulers (the most common list includes Yao 堯, Shun 舜, Yü 禹, T’ang 湯 [founder of the Shang dynasty], and the Chou kings Wen 文 and Wu 武, plus the Duke of Chou 周公). There were also Buddhist and Taoist groupings. That is why Cheng Yü specifies that the reference here is to none of these but rather to seven paragons of the T’ang—viz., Kao-tsu (“the Divine Yao”), T’ai-tsung 太宗 (reg. 626–49), Kao-tsung 高宗 (reg. 649–83), Chung-tsung 中宗 (reg. 684, 705–10), Jui-tsung 睿宗 (reg. 710–12), and, unusually, the latter’s consort née Tou 竇. Madame Tou was the mother of Hsüan-tsung and, along with another of Juitsung’s consorts, née Liu 劉, was put to death by Wu Tse-t’ien in 693 for supposed blackmagic directed against that reigning monarch. Empress Wu did not allow the corpse of either lady to be viewed and had them interred without ceremony. When Jui-tsung himself regained the throne in 710 (see also n. 226 below) he raised Lady Tou’s posthumous rank to Chao-ch’eng huang-hou 昭成皇后 (August Empress Perfected in Splendor) and Lady Liu’s to Su-ming huang-hou 肅明皇后 (August Empress Enlightened in Dignity). CAC 15.8b quotes Cheng Yü’s note as saying “… from the Divine Yao to Jui-tsung and his August Empresses Chao-ch’eng and Su-ming.” Including Empress Su-ming would make up the complement of seven “paragons,” and indeed Hsüan-tsung had decreed in 732 that her spirit be sacrificed to in the T’ang ancestral shrine, an honor he had already granted his mother’s spirit in 716. On empresses Chao-ch’eng and Su-ming, see ChTS 51.2176, HTS 76.3489–90, and TCTC 205.6488, 210.6661–62, 211.6719–20, 213.6799. If Lady Liu should not be among the exalted seven, the distinction must fall upon either the redoubtable Wu Tse-t’ien herself (unlikely, given that Lady Tou was put to death by her) or on Shao-ti 少帝 (Li Ch’ung-mao 李重茂) who reigned for a mere twenty days in July 710. 195  For this cloister, see n. 122 above.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1423

chalcedony and a plumed valance. Shun-hsing’s icon hall and Kuo the Venerable’s medicine room were likewise within the adyta.196

By Cauldron Lake one day the king’s bow and sword were left behind, And grasses in Bridge Mountain’s mists were erelong a rampant tangle.197 196  Apparently a statue or icon (ying 影) of Li Shun-hsing was placed in the Kung-te yüan near or on an altar of semi-precious stone that was surrounded by feathered drapes (the plumes of which recall Shun-hsing’s ethereal status). “Kuo the Venerable” (Kuolao 果老) is Chang Kuo 張果, a renowned occultist and physician of prodigious abilities who claimed to be several hundred years old. He had refused an imperial summons from Empress Wu by feigning death, but finally came to court in 733 (or 735) after two invitations from Hsüan-tsung. He captivated the monarch, who tried without success to have Kuo take in marriage the princess Yü-chen 玉真, a sister of his who had devoted herself to Taoist practices for more than two decades. (On her, see E. H. Schafer, “The Princess Realized in Jade,” T’ang Studies 3 [1985]: 1–23; the monograph of Charles D. Benn, The Cavern-Mystery Transmission: A Taoist Ordination Rite of A.D. 711 [Honolulu: Univ. of Hawaii Press, 1991], regarding the ceremonies surrounding her initiation and that of her sister—the Princess Transcendent in Gold 金仙—into the ranks of the Taoist religious, is not always reliable.) Chang Kuo was ultimately given permission to repair to Heng Shan 恆山, bearing with him gifts and titles, including the exalted designation of “Prior-born with Access to the Mysterium” (T’ung-hsüan hsien-sheng 通玄先生). For Chang Kuo’s official biographies, see ChTS 191.5106–7 and HTS 204.5810–11; but he appears more fully and colorfully elsewhere, in T’ang anecdotal literature. For which, see Jean-Pierre Diény, “La légende, le conte et l’histoire: Le cas du vénérable Zhang Guo (VIIIe siècle),” in En suivant la Voie Royale: Mélanges en hommage à Léon Vandermeersch, ed. Jacques Gernet and Marc Kalinowski (Paris: École française d’Extrême-Orient, 1997), 315–28. Diény’s illuminating article traces the various tales about Chang Kuo back to early ninth-century sources. He also notes that the long account of Chang Kuo in the present MHTL 2.26–28 is probably a later addition to the text, compounded of other sources. For more on Chang Kuo, see Meng Nai-ch’ang 孟乃昌, “Chang Kuo k’ao” 張果考, Tsung-chiao-hsüeh yen-chiu 宗教學 研究 11 (1985): 15–27. The “medicine hall” (yao-t’ang 藥堂) or “medicine room” (yao-shih 藥室) denotes a chamber where Chang Kuo would have carried out his work in mundane and celestial physick. In his Li-tai ming-hua chi 歷代名畫記 (TSCC, 9.295, Chang Yen-yüan 張彥遠 (fl. 847) notes a portrait of Chang Kuo done in 734 by the court painter Chu Pao-i 朱抱—, copies of which were circulating a century later; see also William R. B. Acker, Some T’ang and Pre-T’ang Texts on Chinese Painting, vol. 2, part 1 (Leiden: Brill, 1974), 251. 197  The famous story of the Yellow Emperor’s ascension to heaven is used as metaphor for Hsüan-tsung’s demise. After the Yellow Emperor cast a massive cauldron at the foot of Mount Ching 荊山, out of copper taken from Mount Shou 首山, a dragon appeared

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1424 160

Kroll

Helpless I hear of jade bowls brought to the gold-piece market, Merely look on as a ewer of bronze drifts out the halcyon curtains.198 From Opened Prime to the present day ten cycles have now run through;199 The traces of events from that former time are all withered and fading away. Only on flowers of bamboo feeds the phoenix that roosts in the wu-t’ung,

which carried him and seventy-odd vassals and palace ladies aloft to the empyrean. Some subjects tried to clamber on the dragon by grabbing hold of its whiskers but they, the whiskers, and Huang Ti’s bow (dislodged in the bumpy take-off) fell to earth. The place where this occurred came to be called Cauldron Lake (Ting hu 鼎湖), in presentday Shou-hsiang district, Honan. See Shih chi, 28.1394. “Lake” (hu 湖) would seem to be a mistake for “whiskers” (hu 胡, as in 胡髯); see Shui-ching chu 水經注 (Taipei: Shih-chieh shu-chü, 1962), 4.47, and Yüan K’o 袁珂, Chung-kuo ku-tai shen-hua 中國古代神化(1951; rpt. Shanghai: Shang-wu yin-shu-kuan, 1957), 138, 140 n. 11. Huang Ti’s bow was named Wu-hao 烏號 (“Crow-caw”); for discussion of different interpretations of the name, see Edouard Chavannes, Les Mémoires historiques de Se-ma Ts’ien, vol. 3 (rpt. Paris: AdrienMaisonneuve, 1967), 489 n. 2. Bridge Mountain (Ch’iao Shan 橋山), in present-day Huangling 黃陵 district, Shensi, is said to be the site of the Yellow Emperor’s grave—or, if he was indeed translated on high, of his robe and cap. Shih chi, 28.1396. Here it stands for Hsüantsung’s final resting place, begloomed in fog and weeds. Cf. a similar allusion to Huang-ti’s ascension in analogy with Hsüan-tsung’s departure, in lines 17–18 of Wei Ying-wu’s 韋應 物 (737–92?) poem “Wen-ch’uan hsing,” CTS 194.2001. 198  The precious utensils of Hsüan-tsung’s household are now to be found in the shops of merchants and hands of curio-seekers. Some of the images and the intent behind this couplet and the preceding one seem indebted to (inspired by?) a piece of parallel prose composed by Shen Chiung 沈炯 (502–60) upon passing by the ruins, more than half a millennium old, of Han Wu-ti’s “Sky-piercing Tower” (on which see esp. San-fu huang-t’u chiao-cheng 三輔黃圖校正 [Taipei: Shih-chieh shu-chü, 1975], 5.38; also n. 107 above). The piece, which is included in Shen’s biography in Nan shih 南史 (Peking: Chung-hua shu-chü, 1974), 69.1678, begins with references to Bridge Mountain and Cauldron Lake; midway through, musing on Wu-ti’s death, Shen writes “The beaded curtains of his principal enclosure [for the spirits] one day were fallen to shreds; bowls of jade from the Luxuriant Tumulus 茂陵 [of his grave] were subsequently brought out to the world of men.” 199  That is, ten Jupiter-cycles 十紀 or approximately one-hundred-twenty years. The poet is reckoning from the K’ai-yüan period rather than T’ien-pao because he sees the former as the era of Hsüan-tsung’s greatness, the latter of his decadence.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse 164

1425

All through water-milfoil swims the tortoise that nests in the leaves.200 In the reign of Assembled Glories the throne decried the ‘Inner Canons,’ Purged one teaching, preferred the other, distinguishing yellow-capped from ebon-robed.201 Mount Felicitation’s pool was defiled, the Stone Urn was dismantled,202

200  The phoenix that will roost only on the wu-t’ung 梧桐 tree and only feed off the “flowers” 花 (elsewhere the “fruit” 實) of the bamboo is traditionally regarded (as in Mao Shih 252) as metaphorical of the worthy minister who can only find recognition and support from a wise ruler. The tortoise that nests in the leaves of water-plants and roams through the pondweed is the holy tortoise whose shell is used in divination (see Shih chi, 128.3227 for the locus classicus, which would suggest that “leaves” 葉 might be a mistake for “lotus” 蓮). Here it symbolizes a far-seeing minister who requires the proper environment to thrive. 201  The Hui-ch’ang 會昌 or “Assembled Glories” period (841–47) under Wu-tsung 武宗 witnessed a severe government repression of the Buddhist clergy and large-scale confiscation of its accumulated wealth, owing variously to political maneuverings at court, economic concerns, and the emperor’s personal fascination with and partiality for Taoist teachings. For a good overview of the related issues and events, see Stanley Weinstein, Buddhism under the T’ang (Cambridge: Cambridge Univ. Press, 1987), 114–36. The “two teachings” (erh-chiao 二教) are represented, as often, by the distinctive colors of their clothing: the “yellow[-capped]” are the Taoists, the “ebon[-robed]” the Buddhists. The “throne” in line 165 is literally “the autocrat’s precincts; the Imperial House” (yü-yü 御宇). The “Inner Canons” (nei-tien 內典) is a term used by Buddhists to refer to their scriptures, non-Buddhist writings being identified as “outer canons” (wai-tien 外典). 202   H TS 30.910 reports that in the ninth month of Ch’ui-kung 垂拱 2 (23 September– 22 October 686) a new mountain pushed upward, in an earthquake accompanied by a storm of thunder and lightning, amidst the hills near Mount Li. It attained a height of some two hundred feet above the surrounding hills and featured a pool about fifty acres large near which shapes of dragon and phoenix were seen, along with unusual sproutings of grain. Regarding it as an auspicious response of the natural order to her supervision of the government, Empress Wu had it named “Mount Felicitation.” One courtier, Yü Wen-chün 俞文俊, who was bold and foolish enough to state that such an anomalous event was rather a sign of disapprobation of a woman exercising the rightfully male prerogatives of imperial authority, was exiled to the far south for his temerity. TCTC 203.6442, which adds that Yü was murdered en route to Kwangtung by a hireling of the empress, includes among its “varia” (k’ao-i) the disparaging remarks of the Liang-ching tao-li chih 兩京道里志 that Ch’ing Shan only protruded at first six or seven feet upward and was gradually mounded up by forced labor to a height of three hundred feet. Both HTS and TCTC say the mountain lifted itself on the chi-ssu 己巳 day of the ninth month; but there was no chi-ssu day in Ch’ui-kung 2.ix. An error for i-ssu 乙巳 (30 Sept.) is possible. The

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1426

Kroll

168

The crimson tower and the green gallery both lapsed into ruin. Exceptional pine-trees, impressive cypress were used for firewood and fuel; The mountain bare-headed, the vale gouged blind, lost their awesome aspect.203 Within the haze the walls have crumbled, with paintings by Mo-chieh on them; 172 Amid the clouds, the calligraphy of verses by Hsüan-tsung is lost. The Monastery Supporting the State (Ch’ih-kuo ssu 持國寺) was originally named the Monastery of Mount Felicitation (Ch’ing-shan ssu 慶山寺). It was Te-tsung 德宗 who first changed the name-placard.204 The placard was green and hung above the covered way.205 The court of the Heavenly Empress206 employed minions consigned to the adyta to frame the monastery according to measurements on a palatial scale. During the K’ai-yüan period, the Monastery of the Stone Urn was renovated and refitted using spare materials from the construction of the Hua-ch’ing Palace.207 All of the remarkable emergence of the mountain and its divine pool was later cited several times in the important commentary to the Great Cloud Sutra 大雲經 presented to the throne in 690, which confirmed Wu Tse-t’ien as a Cakravartin king and Bodhisattva. See Antonino Forte, Political Propaganda and Ideology in China at the End of the Seventh Century: Inquiry into the Nature, Authors and Function of the Tun-huang Document S. 6502 Followed by an Annotated Translation (Naples: Istituto Universitario Orientale, 1976), 194 n. 66, 236, 238. The Monastery of the Stone Urn, where, as Cheng Yü notes in his prefatory remarks, he had spent time in the late 830s studying the records of the past, was founded on Mount Felicitation. Cheng’s comments following the present section of the poem provide more details about it. Lines 167–72 picture the disfigurement of the place subsequent to the Hui-ch’ang acts of persecution. This is itself now another example of past pleasure gone forever. 203  With trees lopped, the mountain now looks like a young, uncapped boy (t’ung-shan 童山); with its pool choked nearly dry, the gorge now looks eyeless (yüan-ku 眢谷). 204  Thus, sometime between 779 and 805. 205  The “covered way” ( fu-tao 複道) refers to the elevated road built by the First Emperor of Ch’in between his capital and Mount Li; see LTHC 2.29b. The monastery on Ch’ing Shan overlooked the route. 206  That is, Empress Wu. 207  A lü-shih by one Fan Chao 范朝 describes the monastery during its heyday under Hsüan-tsung. Entitled “Inscribed at the Stone Urn Monastery,” it reads as follows (CTS 145.1469–70): A precinct of surpassing beauty—fit for lengthy gazing; In slow-paced springtime—good for casting out sadness. From the links of the mountain-barriers four passes rise; 4 From the river’s belt the eight streams flow. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1427

images for its Buddha-hall that were made of precious stone were brought in from Yuchou, arriving on the same day as the Taoist images for the Gallery for Homage to the Prime.208 They were absolutely wondrous and incomparable. When one rapped them with one’s knuckles they resounded as though they were chime-stones. The remaining [stone] images were alike sculpted by the hand of Yang Hui-chih 楊惠之.209 All of the hollow-limbed [bronze] images were cast by Yüan Ch’ieh-erh 元伽兒,210 wondrous of manner and beautiful in their delicacy, forever unprecedented and without peer. The crimson tower was located on the cliff west of the Buddha-hall, overlooking a sheer bluff. Within the tower were verses in the calligraphy of Hsüan-tsung, each poem in its own style, either cursive (ts’ao 草) or outspreading (pa-fen 八分). There were landscape paintings of Wang Yu-ch’eng 王右丞211 on two walls. After the dismantling of the monastery, all of these things have been lost. Mo-chieh 摩詰 was Wang Wei’s byname.212

At the base of Stone Fish Cliff was a well a hundred armspans deep,213 And a silver frame uncoiled downward a length of crimson cord.214 Doubled ledges support the incense-filled galleries, And layered cliffs reflect the high multicolored buildings. Comported to look upon the vicinity of the warming pools, 8 Inclined to the pleasures of the sagely sovereign’s rambles. (The “four passes” 四塞 of the Kuan-chung 關中 area are Han-ku 函谷 to the east, Wukuan 武關 to the south, San-kuan 散關 to the west, and Hsiao-kuan 蕭關 to the north. The “eight streams” 八川 of the region are, in traditional listings, the Ching 涇, Wei 渭, Pa 灞, Ch’an 滻, Feng 酆, Hao 鎬, Liao 潦, and Chüeh 潏.). 208  On Yu-chou, see n. 87 above. On the Gallery for Homage to the Prime, see n. 83; this may be the “green gallery” (lü ko 綠閣) mentioned in line 168. 209  On Yang Hui-chih as a famous sculptor, see Li-tai ming-hua chi, 9.286; tr. Acker, T’ang Texts on Painting, vol. 2, part 1, 233. Images crafted by him of Taoist divinities were also installed during the K’ai-yüan period in the Hsüan-yüan 玄元 abbey north of Lo-yang; see T’ang yü-lin chiao-cheng 唐語林校證 (Peking: Chung-hua shu-chü, 1987), 5.489; also T’ai-p’ing kuang-chi, 212.439c. 210  I have as yet no further information on Yüan Ch’ieh-erh. 211  The famous poet and painter Wang Wei 王維 (701–61), here called Yu-ch’eng because of his sometime holding of the position of Right Deputy in the Bureau of State Affairs (shang-shu yu-ch’eng 尚書右丞). 212  Cheng Yü notes Wang Wei’s byname here because of its Buddhist significance: Mochieh preceded by the personal name Wei yields the Chinese transcription for the Sanskrit Vimalakīrti, the name of the ideal Buddhist layman celebrated in the popular Vimalakīrti-nirdeša-sūtra. 213  Stone Fish Cliff (see n. 63 above) is the bluff over which stood the crimson tower of Stone Urn Monastery. On the cliff-face hung “a waterfall running straight down several dozen rods, resembling a sheet of dressed silk” (LTHC 2.36a). 214  The “silver frame” (yin ch’uang 銀床) is that housing the axle and pulley mechanism of the well-windlass described in Cheng Yü’s commentary. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1428 176

Kroll

In those times limpid shadows lay on the leaves of crimson, Till one morn blowing grit buried the pure-white disks.215

Below Stone Fish Cliff there were naturally striated stones, the formation of which resembled an urn storing up the waters of the soaring wellspring.216 Hence His Highness gave the name “Stone Urn” to the monastery.217 The monks of the monastery would carefully unwind, by means of a windlass poised in the uppermost storey of the lofty tower, a lengthy braid of bamboo [with attached bucket] stretching more than two hundred feet to draw water. The wellspring of the urn emerged from the nibs of upturning trees near the crimson tower.218 Now that the monastery has been laid to waste, the urn of stones has come to be buried in obliteration.

The candle spire of the household of Han was nigh the nibs of the grove, And its thousand branches glistened like the display of a mountain daybreak. In years long ago the brilliance of its light eclipsed the moon in the sky; 180 By yesterday all had melted away, at the forking of the paths.219 [The Lady of] Han Principality220 had a thousand-branch lantern spire made. It was eighty feet tall and was placed on the mountain. When the night of the First Prime arrived,221 she would have it lit up. Its thousand lights eclipsed the moon. Anyone who was within a hundred li could catch sight of it.222 215  The “crimson leaves” (hung yeh 紅葉) belong to the waterplants in the pool (or “well”) beneath the cascade, here seen shaded by the soft light of the moon. But with the destruction and pollution brought by the Hui-ch’ang oppression those once unsullied circlets (su kuei 素規) of moonlight reflections have been completely effaced. 216  Cf. n. 63 above. 217  As usual, here “His Highness” is Hsüan-tsung. 218  That is, the waterfall began above the height of the trees surrounding the tower. 219  The “forking of the paths” is the change of eras, the dissolution of the glorious past. 220  On the Lady of Han Principality, see line 69 and nn. 106 and 111 above. 221  First Prime (shang-yüan 上元) is the full-moon night of the first month of the year, the so-called Lantern Festival when the capital’s regular curfew was suspended, the gates of all wards remained open throughout the night, and households everywhere kept displays of lanterns burning to rival and celebrate the light of the moon. On the T’ang practices in the capital, see Ishida Mikinosuke 石田幹之助, Chōan no haru 長安の春 (Tokyo: Heibonsha, 1967), 65–76, 146–48. 222  Cf. the nearly identical account in KYTPIS 2.102.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1429

The autocrat’s nameplate over the dragon demesne was astonishingly high,223 But fire scorched it, then oxen pulled it down from above the crested crags. The peacock-pines are now no more, nor the red amber;224 184 Mandarin-duck tiles are shattered, and the blue lapis-glaze.225 The monastery’s placard had been handwritten by Jui-tsung at his feudatory residence.226 It showed forth conspicuously from atop the steep tower. Tradition tells that below the peacock-pines there was the red pachyma fungus which, if it remains underground for a thousand years, will turn to amber;227 and on the peak in front of the monastery were ancient pine-trees, old cypresses, and even mioga ginger.228 Now everything has been cleaned out and uprooted by fuel-gatherers and plant-foragers.

The route that lies before me now is still of an unknown length; I can merely live out whatever is left in failing vigor and health. To encounter you, sir, and speak of all this is spilling tears in vain, 223  The “dragon demesne” (lung-kung 龍宮) is the Shih-weng Monastery. 224  Peacock-pines (k’ung-ch’üeh sung 孔雀松) are pine-trees of peacock-green. For the red amber, see Cheng Yü’s commentary. 225  “Mandarin-duck tiles” are roof-tiles overlapped closely, like nestling duck and drake. “Lapis-glass” (liu-li 琉璃) can refer to authentic lapis lazuli, beryl (Skt. vaidūrya, one of the Buddhist “seven treasures”), or to colored glass or paste in the fashion of authentic gemstones. Schafer, Golden Peaches, 235–37. The blue lapis-glass referred to here ornamented the monastery’s fixtures and statues. 226  That is, during the years 698–705 when Jui-tsung, after serving for six years (684–90) as nominal emperor under Wu Tse-t’ien’s control and eight years (690–98) as figurehead— and effectively incarcerated—heir-apparent of the Empress’ new Chou dynasty, was replaced as successor by Chung-tsung, demoted to Prince of Hsiang 相王, and removed to a residence outside the imperial palace. 227   Fu-ling 茯苓 (Pachyma Cocos or tuckahoe), a fungus much used in medicinal and macrobiotic prescriptions, which grows on the roots of fir trees. See G. A. Stuart, Chinese Materia Medica: Vegetable Kingdom (1911; rpt. Taipei: Southern Materials Center, 1987), 298. Chang Hua 張華 (232–300), among others, records that the resin of pine and cypress, having seeped underground for a thousand years, will turn to pachyma fungus and then amber. See Po wu chih chiao-cheng 博物志校證, ed. Fan Ning 范寧 (1981; rpt. Taipei: Ming-wen shu-chü, 1984), 4.48. On the color of the millennial amber, signifying the glow of cinnabar and hence the elixir of immortality, see T’ai-shang ling-pao wu-fu hsü 太上靈寶五符序 (HY 388), 2.12a. 228  Mioga ginger is literally chia-ts’ao 嘉草, “finest of herbs.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1430

Kroll

188

Yet I remember the joys and delight of a time that will not be met with  again.” “Weep not, my graybearded host, but harken to my words: Lay down the burden of feelings for the past, and cease your somber sighs. The Ho running clear, the sea at rest, is not an unthinkable sight,229 192 For our August Lord has raised the foundations already of prosperous  peace.230 The land around Huang-chung was engulfed and lost long ago,231 But yesternight it was taken back, no longer a source of discomfort.232 229  The Ho (i.e., the Yellow River) is said to run clear and the sea to be at rest only in times of great peace, when a sage rules the empire. 230  The reference is to the reign of Hsiuan-tsung 宣宗 (transcribed this way to prevent confusion with our protagonist, Hsüan-tsung 玄宗), begun 25 April 846. It followed that of Wu-tsung which was an era of much social and political unrest, including not only the Hui-ch’ang oppression of Buddhism but also a serious external threat from the Uighurs in the early 840s (on which, see esp. Michael R. Drompp, “The Writings of Li Te-yü as Sources for the History of T’ang-Inner Asia Relations” (Ph.D. diss., Indiana Univ., 1986), and the same author’s “A T’ang Adventurer in Inner Asia,” T’ang Studies 6 [1988]: 1–23) and a turbulent provincial rebellion in Chao-i 昭義 in 843–44. In the succeeding lines of the poem Cheng Yü makes special note of the conflict with the Uighurs. 231  Huang-chung 湟中 was about forty-five miles due east of Kokonor (approximately eight miles south of present-day Lin-fan ch’eng 臨蕃城, Kansu), in the T’ang “circuit” of Lungyu East 隴右東道. The Kansu corridor had fallen out of T’ang control and under Tibetan domination in 763. hts 216A.6087, tctc 223.7146; also Christopher I. Beckwith, The Tibetan Empire in Central Asia: A History of the Struggle for Great Power among Tibetans, Turks, Arabs, and Chinese during the Early Middle Ages (Princeton: Princeton Univ. Press, 1987): 146–47. 232  This refers to the great success of Chang I-ch’ao 張義朝, a native of Tun-huang, in reclaiming for China suzerainty over the Kansu corridor in late 851. See The Cambridge History of China, vol. 3: Sui and T’ang China, 589–906 (sic), part 1, ed. D. Twitchett (Cambridge: Cambridge Univ. Press, 1979), 678–79, and Beckwith, The Tibetan Empire, 170–71. Although ChTS 18B.629 says Chang’s campaign was completed in the eighth month of the year, HTS 8.249 and TCTC 249.8048–49 say the tenth month (i.e., 28 October to 26 November), the k’ao-i of the latter source proving the correctness of the later date. The date of composition of Cheng Yü’s poem is established by this event, which was reported to the throne sometime in November 851. Thus, the “winter of the present year” referred to by Cheng in his prefatory remarks corresponds to December 851 or possibly January 852. “Yesternight” is not of course to be taken strictly. “Source of discomfort” is literally “ulcerous sore” (ch’uang-i 瘡痍); the loss to Tibet ninety years previously of the T’ang avenue to Central Asia had been a sore point for dynastic self-esteem.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1431

The king of the savages has fled northward, abandoning the Green Mound, 196 And the caitiffs’ horses have run off westward, deserting the Yüeh-chih!233 Twice to encounter the ‘years of Yao’ is surely no easy chance;234 So take good care, graybeard, and may you stay in the best of health.” At the first light of dawn, our heads cleared of wine, we parted company; 200 But tonight I’ll have a goblet in pledge that the old man not be  forgotten.

Appendix 1



On “Rainbow Skirts and Feather Vestments”

While it is now generally accepted that the suite “Ni-shang yü-i” was, as described in n. 132 above, the product of promiscuous conjunction, most of the accounts of Hsüantsung’s lunar journey cited in n. 130 above aver that the emperor transmitted the piece whole and intact just as he heard it in the moon. The beauty of the music was clearly thought by many to be deserving only of an unprofaned pedigree. A third, less detailed version of the music’s genesis is offered in a poem by Liu Yü-hsi 劉禹錫 (772–842) on the subject of looking out from a tall building of the San-hsiang post-station 三鄉驛 (near present-day I-yang 宜陽, Honan) westward toward Mount Nü-chi 女几山, just as Hsüan-tsung was said to have done more than half a century earlier: The Son of Heaven, during “Opened Prime,” was content in all affairs, Except he grudged the swift passing of the time’s brilliant scenes. 233  The “Green Mound” (Ch’ing-chung 青塚) is the reputed burial place of Wang Chao-chün 王昭君 (d. 31 bc), the Han-dynasty princess given in political marriage to the leader of the Hsiung-nu. The barrow heaped over her bones was said always to be green with vegetation, despite the aridity of the surrounding area. Present-day Huhehot in Inner Mongolia claims the site, south of the city. See Kwong Hing Foon, Wang Zhaojun: Une heroine chinoise de l’histoire à la légende (Paris: Institut des Hautes Études Chinoises, 1986). The Yüeh-chih 月支 were a nomadic people living in the Kansu region during the Former Han, the greater part of whom later migrated westward. See, among others, Hulsewé, China in Central Asia, 119–23. Cheng Yü uses the name anachronistically, to suggest that the war-steeds of the Tibetans bolt in fear of Chang I-ch’ao’s forces, leaving their masters stranded behind. 234  The “years of Yao” (堯年) indicates the consummate reign of a sage-king, like Yao himself. The old man has had the unique good fortune to experience two such reigns—that of the great Hsüan-tsung a century ago and now that of the present emperor.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1432 4

8

Kroll From cross-paths of San-hsiang he looked off to a mountain of transcendents And came back to compose the tune “Rainbow Skirts and Feather Vestments.” From then on his transcendent heart rested by the Chalcedony Pool; The Three Clarities and Eight Phosphors chased round about before him. In Heaven, up above, he suddenly mounted a white cloud and went off; While in the mortal realm remain vainly the lyrics of the autumn wind.

(cts 356.3999. The Chalcedony Pool 瑤池 is a famous haunt of Hsi Wang Mu on her cosmic mountain K’un-lun. The Three Clarities 三清 represent the lords of the Taoist heavens, Greatest Clarity 太清, Highest Clarity 上清, and Jade Clarity 玉清 [in ascending order]; the Eight Phosphors 八景 are the bright spirits of the eight directions of space as well as the luminous protectors of key junctures within the human body. Line 6 of the poem suggests Hsüan-tsung’s obsession with such matters. The “lyrics of the autumn wind” 秋風詞 in the concluding line may refer both to the earthly atmosphere now left behind by the etherealized monarch and perhaps also to the wellknown song of “The Autumn Wind” composed by Hsüan-tsung’s great counterpart from the Han dynasty, Wu-ti, a song that laments the unavoidable loss of youth and passage of time [Hsien-Ch’in Han Wei Chin Nan-pei-ch’ao shih 先秦漢魏晉南北朝詩, ed. Lu Ch’in-li 逯欽立 (Peking: Chung-hua shu-chü, 1983), 94]—predicaments apparently transcended by the T’ang ruler.) Attention is called to this poem and its unique explanation of the creation of “Rainbow Skirts” in Yüeh Shih’s self-commentary to his Yang T’ai-chen wai-chuan, 131. A horribly botched translation of the poem has been published in Howard S. Levy, Harem Favorites of an Illustrious Celestial (Taipei: Chung-t’ai, 1958), 144. Yüeh Shih also notes that the dance was specially performed for the emperor by Lady Yang on the day in the seventh month of 745 when she was formally installed as “Precious Consort” (kuei-fei). By the early ninth century it was already something of a rarity for the suite to be presented in its original form. Po Chü-i 白居易 (772–846) was particularly fond of it and attempted to assemble a suitable company for performance. His long, eighty-eight line poem, “The Song of the Rainbow Skirts and Feather Vestments” (“Ni-shang yü-i ko,” cts 444.4970–71), gives an account of his efforts and of his correspondence about it with his friend Yüan Chen 元稹 (779–831). Waley made a summary of this, though his words mask the fact that he is actually abstracting Po’s poem, in The Life and Times of Po Chü-i (London: Allen and Unwin, 1949), 154–55. The poem also contains the best surviving description of the suite, enhanced by Po’s own scholia. Here are the most relevant verses in this regard (lines 13–30) with the attached self-commentary: Lithophones, syrinx, koto, and cross-flute support each other in series; Struck, finger-stopped, plucked, and blown, they are voiced one after another.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1433

« At the beginning of all fa-ch’ü 法曲 [pseudo-classical music from the Sui era, liked by Hsüan-tsung and taught by him to selected musicians among the Pear Garden performers; the instrumentation consisted of handbells, cymbalets, large bells, lithophones, streamered syringes, and lutes; see hts 22.476] the various instruments do not play in consort. Rather, those made of metal, stone, strings, or bamboo produce their music serially in order. The prelude (hsü 序) at the beginning of “Rainbow Skirts” also was done this way. » Throughout the free prelude’s sextuple execution, the dancers betray no movement: The overnight clouds of the Sunlit Terrace indolent before taking flight. « The six segments of the free prelude (san-hsü 散序) are unmeasured [i.e., a time-marking clapper is not used], hence there is no dancing during it. [Note the importance, for dance movements, of the measured beat. The comparison of the motionless dancers with the “overnight clouds of Yang-t’ai” 陽臺宿雲 makes them one with the rainbow goddess of Wu Shan 巫山 who dallied with a king of Ch’u, revealing herself to him as the morning’s “dawn clouds” and the evening’s “moving rain”; see Sung Yü 宋玉, “Kao-t’ang fu” 高唐賦, wh 19.393.] » Now, as though ripped by a thumbnail the middle prelude brings the beat into play, Like an autumn bamboo’s stem torn apart or the cracking open of ice in spring. « Not until the middle prelude (chung-hsü 中序) is there a measured beat. It is also called the “measured prelude” (p’o-hsü 拍序). » Swirling awhirl so driftily, the circling snow comes lightly; Unheld, unhindered so beguilingly, the roaming dragons are skittish. Behind tiny hands dropping down—a willow with no strength at all; As lappets are dragged all aslant, the clouds seem about to come forth! « These four verses pertain to the initial posturings of the dance. [The “clouds” almost emerging here are the breasts of the dancers, a metaphor encountered not infrequently in T’ang erotic verse; cf. the lines from Li Tung’s 李洞 (fl. 893) poem “To Present to Refined Mistress P’ang” (“Tseng P’ang lien-shih”): “Both sides of your face suffused with wine—pink almond-blossoms are jealous; / Half-shown bosom creamy soft—the white clouds are luxuriant.” cts 723.8296.] » Hazy moth-eyebrows pulled together, in bearing unsustainable; Windblown sleeves fall and rise, as with a feeling of their own. The Shang-yüan Lady nods her hair-coil, summoning O-lü, The Royal Mother shakes out her sleeve-cuffs, parting from Fei-ch’iung. « Hsü Fei-ch’iung 許飛瓊 and O Lü-hua 萼綠華 are both female transcendents. [The former is best known from her appearance in the Han Wu-ti nei-chuan 漢武帝內傳 where she plays the “numen-quaking reed-flute” 震靈簧 in Hsi Wang

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1434

Kroll Mu’s orchestra (hy 292, 3a). The latter is a Shang-ch’ing goddess recorded in Chen kao 真誥 (hy 1010, l.la–2a) as attempting in December 359 to acquaint Yang Huan 羊懽, who later proved unworthy, with celestial lore. Shang-yüan 上元 is the “Lady of the Highest Prime,” one of the most exalted Shang-ch’ing divinities (for more on her, see Kroll, “Three Taoist Figures of the T’ang Dynasty,” Society for the Study of Chinese Religions Bulletin 9 (1981): 30–34, and “Li Po’s Transcendent Diction,” Journal of the American Oriental Society 106 (1986): 100–102.] » Intricate tones and surging cadences—for all of a dozen turns; Jounced beads and joggled jades—how they tinkle and clink! « The “broaching” (p’o 破) of “Rainbow Skirts” normally concludes after twelve turns (pien 遍). [Cf. the following description in another poem of Po’s (“Wot’ing fa-ch’ü ni-shang,” cts 449.5069) depicting the “Rainbow Skirts” music: “Fluently flexuous, the tender notes during the ‘entering broaching.’ ” 宛轉柔聲 入破時] » When the soaring simurgh’s dance is over, it retreats, gathering in its wings; As the calling crane’s music ends, it prolongs a last drawn-out note. « Whenever a piece is nearing its close, the beat of the music usually quickens to allegro. Only in the finale of “Rainbow Skirts” is there a single note prolonged and drawn out. [Cf. the similar statement in hts 22.476.] »

This may be the closest we are likely to get to this celebrated suite. The version known to Li Yü 李煜, last emperor of the Southern T’ang dynasty, in the tenth century, was recognized even at the time as containing interpolations. Chiang K’uei’s 姜夔 (ca. 1155– ca. 1235) reconstruction of the “middle prelude,” known from Japanese tōgaku repertoire and recently recreated and recorded by Picken, is but guesswork—more than less accurate, one hopes, but guesswork nonetheless. For a good, concise summary of “Rainbow Skirts” scholarship, see the essay by Tung Hsi-chiu 董錫玖 in T’ang-tai wutao, 141–43.



Appendix 2



Hsüan-tsung’s Return to Ch’ang-an

Cheng Yü’s commentary following line 140 describes some of the events and protocol surrounding the return to Ch’ang-an of the retired emperor Hsüan-tsung. The situation was rather delicate, since Su-tsung had proclaimed himself emperor on 12 August 756, only obtaining Hsüan-tsung’s agreement a month afterward (see n. 161 above). The capital was retaken from the rebels on 13 November 757, and Su-tsung, the reigning monarch, returned first, on 8 December. Hsüan-tsung seems to have waited for the approval or invitation of his son before setting out from Shu. He reached Feng-hsiang Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1435

鳳翔, seventy miles west of Ch’ang-an, on 6 January 758 (see n. 172 above). The meeting

of father and son and the former’s reentry into the capital he had abandoned a year and a half previously are moments of some drama and interest. tctc 220.7044–45 gives the following account. (“His August Highness” refers to Hsüan-tsung, “His Highness” to Su-tsung.)

On the ping-shen day (equiv. 6 January) His August Highness reached Fenghsiang, with a military entourage of over six hundred men. He ordered them to lay up all their armor and weapons in the commandery’s depot. His Highness set out with three thousand of the finest cavalrymen to offer respectful welcome. On the ping-wu day of the twelfth month (16 January) His August Highness reached Hsien-yang. His Highness made ready a formal carriage to welcome him at the Wang-hsien palace (see line 139 of the poem). His August Highness was at a pavilion on the south side of the palace. His Highness had removed his [imperial] yellow robe and put on a purple robe [suitable merely for a prince]; when he came in sight of the pavilion, he dismounted and hurried forward on foot [as a sign of respect], to perform an obeisance below the pavilion. His August Highness descended from the upper storey and embraced his son, weeping. His Highness clasped his father’s feet, unable to keep himself from sobbing loudly. His August Highness then unfastened his own yellow robe, to place it himself upon His Highness. But the latter louted low, with his head down, refusing obstinately. His August Highness then said, “The appetence of Heaven and the hearts of men have all turned allegiance to you. Should We gain protection and sustenance in the years remaining, it is owing to your filial devotion.” His Highness could do nothing but accept. The fatherly elders who stood outside the line of the guards cried out in joy and offered their salutations. His Highness commanded the guards to stand aside, thus allowing the thousand and more people to pay their respects to His August Highness. They said, “We, your subjects, have today beheld our two sages meeting each other again, and we can now die without regrets!” His August Highness was unwilling to abide in the principal hall [of the Wanghsien kung], saying, “This is the seat of the Son of Heaven only.” But His Highness insisted and took him by the arm up into the hall. When the magister of repasts presented their food, His Highness acted as taster for each dish and only then put it before Hsüan-tsung. On the ting-wei day (17 January), before they set out from the temporary palace, His Highness personally worked a horse for His August Highness before presenting it to him to ride. When his father mounted, His Highness himself held the bridle and advanced with him for several paces, at which time His August Highness made him give over [this excessive kindness]. His Highness then set himself on his own horse and led on ahead, but was careful not to encroach upon Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1436

Kroll the [imperial] causeway. His August Highness addressed the attendants, saying, “For fifty years I was the Son of Heaven but did not feel the [dignity of the] honor. Today, being father to the Son of Heaven, I feel honored at last.” The attendants all cried out, “A myriad years!” From the Gate of Opening-to-the-Distance (see n. 171 above), His August Highness proceded on to the Palace of Great Luminosity (Ta-ming kung, the grounds of the imperial residence), where he held sway at the Throne-Hall for Enclosing the Prime (Han-yüan tien 含元殿, the main basilica of the Ta-ming complex, located in the south-central part of the compound; see Li Hua’s 李華 [c.s. 735, d. 766] “Rhapsody on the Han-yüan Throne-Hall,” ctw 314.1a–9a, for a full contemporary evocation, and Fu Hsi-nien 傅熹年, “T’ang Ch’ang-an Ta-ming kung Han-yüan tien yüan-chuang te t’an-t’ao” 唐長安大明宮含元殿原狀的探討, Wen-wu 1973.7: 30–48, for a report on the excavation of the hall), where he gave consolation and solace to the hundred-fold officials. He then visited the ThroneHall of Lasting Jubilation (Ch’ang-le tien 長樂殿, in the central-west section of the Ta-ming complex, where the ancestral tablets were then temporarily housed, owing to the destruction of the Grand Temple for the Ancestors 太廟 during the rebel occupation) to beg indulgence before the spirit-tablets of the Nine Enshrinements [to the imperial ancestors], weeping heart-stricken there for a long time. Later the same day, he graced with his presence the Hsing-ch’ing palace (see nn. 133, 177 and 178 above), where he resided thereafter. Although His Highness repeatedly expressed the suggestion that he should ignore [questions of] status and come back to the Eastern Palace (i.e., the Ta-ming kung, where the reigning monarch resided), His August Highness would not agree.”

A portion of the opening section of this passage is translated by Robert des Rotours in his rendering of Lin Lu-tche (Lin Lü-chih 林旅之), Le règne de l’empereur Hiuantsung (713–756) (Paris: Institut des Hautes Études Chinoises, 1981), 478. Cf. the briefer accounts of these events in ChTS 9.235, 10.249.

Addendum Just as the proofs of this article reached me, I received from Ding Xiang Warner of Cornell University a copy made by her of relevant portions of the Hsing-yang tsa-tsu. As mentioned in note 48 above, the Peking National Library holds what seems to be the only edition of this collectanea. During a research visit to the library in October 2003, Professor Warner was given permission by the Rare Book Section to consult the work. Though not allowed to make a machine copy of it, she painstakingly transcribed in her own hand the text of Cheng Yü’s poem and interlinear commentary, as well as

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1437

the prefatory colophon by Ch’eng Ting-yüan, and has kindly placed this material at my disposal. For this I am much beholden to her. Given the constraints of time, I can make but a few brief remarks about this rare version of the text. Ch’eng Ting-yüan’s colophon is dated to early 1699 and includes the following statements of interest to us: The “Poem on the Chin-yang Gate” in one hundred rhymes was composed by [Cheng] Pin-hsien 賓先, personal name Yü 嵎, deputy director of the Bureau of Rites 禮部行郎 during the T’ang. He excelled in [the writing of] tz’u-fu 詞賦. For some time he enjoyed a full degree of fame together with Lord Li Tu 李君都, Lord Ts’ui Yung 崔君雍, and Lord Sun Huang 孫君璜, who were [collectively] called the “Four Lordlings” 四君子. Receiving fond glances, they all were able to be promoted and advanced. Because of this, contemporaries said of them, “If you would have success in life, inquire of Yü, Huang 瑝, Tu, and Yung.” In his time this is how greatly he was esteemed. The first thing we notice here is that Cheng Yü’s byname is given as Pin-hsien 賓先, not Pin-kuang 賓光 as in other sources. Pin-kuang, however, is the better semantic complement to the personal name Yü 嵎. We should regard this as a simple graphic error. A bit farther on we see two different graphs used in the space of two sentences for Sun Huang’s personal name; the correct graph is the second, 瑝. This looseness appears in the Hsing-yang tsa-tsu text of the poem and commentary also, where there are many occurrences of variant characters containing different phonetic or signific components than found in other editions. The reference to Cheng Yü as having held the position of deputy director in the Board of Rites is new. This was a relatively high office—rank 4a2, at the eighth rung of the T’ang thirty-rung bureaucratic ladder. Unfortunately, there is no indication of when Cheng Yü was in this position. If the information is reliable, he must have reached this position some time after 855 when, as we know, he was only a junior member of the staff of the Yang-chou governor-general. In light of the usual career path and timetable of successful bureaucrats, it is reasonable to assume that Cheng Yü is not likely to have become the number-two man in the Board of Rites before the early 860s. This allows us to extend his floruit dates to at least that time. The three men with whom Cheng Yü is associated here are an interesting and somewhat troubling group. While not major figures, all of them had notable early careers and seemed to be on their way to higher positions, perhaps thanks to patronage and family ties, as Ch’eng Ting-yüan’s wording suggests. However, each of the trio later incurred disgrace or misfortune. Sun Huang, after rising to second in command at the Censorate, was demoted to the far south because of his association with the high minister and Han-lin academician Liu Chan 劉瞻 who criticized the emperor too

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1438

Kroll

personally in 870.235 Ts’ui Yung, while serving as prefect of Ho-chou 和州 in the 860s, failed to resist P’ang Hsün 廳勛 and his forces during the latter’s rebellion, and was therefore “given the gift of death” upon order of the court.236 And Li Tu, while military commander, surrendered to the rebel Huang Ch’ao 黃巢 during the latter’s advance on Ch’ang-an in late 880.237 One wonders what sort of family tradition would have grouped Cheng Yü with these men. We should remember, as the anecdote encourages us to do, that it was their early successes—not their unmentioned, ignominious ends—that brought them to people’s attention. Are we meant to infer that Cheng Yü suffered similar dishonor late in life, or is it simply that a catch-phrase about the “four lordlings” was recorded and handed down? There are hints that family tradition, or perhaps Ch’eng Ting-yüan’s own faulty grasp of history, may have garbled the facts. Curiously, Ts’ui Yung is closely linked with Cheng Yü’s more famous kinsman Cheng Hao 鄭顥238 as youthful friends and prodigies, in an anecdote recorded in T’ang yü-lin.239 Elsewhere we find a popular saying about “attaining success in life,” nearly identical in its wording, applied to a different midninth-century foursome that includes a different member of the Cheng clan, namely Cheng Lu 鄭魯.240 Even more curiously, T’ang chih-yen contains a story also linking this Cheng Lu with Ts’ui Yung.241 During the last half of the T’ang, there were dozens of Hsing-yang Chengs who held various official positions. It is understandable that not all of them would be distinctly remembered eight centuries later. Confusion possibly continues when we read Ch’eng Ting-yüan’s next sentences, discussing Cheng Yü’s writings: His writings were plentiful, including a Collected Works 文集 in ten chüan, the Sheng cheng chi 聖政記 in eight chüan, and Piao-chuang lüeh 表狀略 in three chüan, as recorded in the [Hsin] T’ang shu bibliographic monograph, the Wenhsien t’ung-k’ao, and other sources. But today what we are able to see is nothing more than this poem alone. The second compilation mentioned here, Sheng cheng chi or “Records of Sagely Government,” nowhere else attributed to Cheng Yü, sounds suspiciously like the Sheng 235   C hTS 177.4606; HTS 181.5353; T’ang yü-lin chiao-cheng 4.374–75. 236   H TS 159.4963. 237   C hTS 182.4695, 4699; HTS 182.5364, 187.5435–36. 238  See ChTS 159.4181–82 and hts 165.5076 for his biographies. 239   T’ang yü-lin chiao-cheng 4.374. 240   H TS 160.4974. This group includes Ts’ui Hsüan 崔鉉, Yang Shao 楊紹, and Hsüeh Meng 薛蒙, in addition to Cheng Lu. 241   T’ang chih-yen 8.82.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Nostalgia And History In Mid-ninth-century Verse

1439

cheng lu 聖政錄, noted in the Sung shih bibliographic monograph as the work of Cheng Chü-chung 鄭居中 an important early twelfth-century official. Has he been carelessly confused with another, identically named Cheng Chü-chung who was active at court ca. 825 and a client of the Niu Seng-ju 牛僧儒 faction? However, the date of 825, established by both T’ang histories for this Chü-chung,242 is too early to force an identification with Cheng Yü. We might have here a notice of an authentic work by Cheng Yü; or it is possible that the obscure Cheng Yü has again been confounded with another member of the clan. The rest of the colophon paraphrases Cheng Yü’s own preface to the poem, closing with the comment by Ch’eng Ting-yüan that although the “Chin-yang Gate” is a work of poetry, “it is veritable history” 實史也 and can supply some of the lacunae in Cheng Ch’i’s K’ai-T’ien ch’uan-hsin chi, hence it is placed after it in this collectanea. As to the text of the poem and authorial commentary presented here, it contains approximately seventy variant readings compared with the standard, received version. The great majority of these variants are of a graphic nature involving the use of different significs in writing one and the same word, or involve the substitution of synonyms (e.g., 室 for 堂 in line 156, 中 for 間 in line 172). Here and there a word or two is dropped in the commentary, but no consequential changes in meaning occur. This again suggests that the frequent emendations evident in Ho Hsi-wen’s Chu-chuang shih-hua version, making any line that requires the least application of thought easier to understand, should be rejected; they likely represent Ho’s own “improvement” of the text.243 There are some obvious errors in the Hsing-yang tsa-tsu version, such as 梁楣 (“beams and lintels,” which makes no sense in context but is a common compound) for the rarer but clearly correct 罘罳 (“covert for second thoughts”) in line 110, or 神農 (“the Divine Husbandman,” again nonsense in this context but more familiar) for 神堯 (“the divine Yao,” referring here to the T’ang founder, Kao-tsu) in the commentary following line 156. On a few occasions the final two words of an oddnumbered (i.e., unrhyming) line are transposed. The one major anomaly in the text is a large-scale transposition of forty lines. The commentary following line 28 is cut short after the phrase “During the Pao-ying era, Yü Ch’ao-en had it razed,” and the text then jumps immediately to line 69. It continues on through line 92, after which the missing lines 29–68 are inserted. This arrangement destroys the logical development of the narrative and can be presumed to be an editorial mistake.

242   C hTS 168.4382, HTS 162.4993. 243  Only in lines 37–38 does the Hsing-yang tsa-tsu version reflect that of Ho Hsi-wen: “The Six Armies each went out to shoot rhinoceros and gaur;/ The bowmen enclosed the countryside, spread out cords for the snares.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1440

Kroll

Abbreviations Used in Notes

alssc An Lu-shan shih-chi 安祿山事跡 bss Basic Sinological Series (Kuo-hsüeh chi-pen ts’ung-shu 國學基本叢書) cac Ch’ang-an chih 長安志 cts Ch’üan T’ang shih 全唐詩 ChTS Chiu T’ang shu 舊唐書 ctw Ch’üan T’ang wen 全唐文 hts Hsin T’ang shu 新唐書 hy Consecutive numbering of texts in the Taoist Canon, according to the Harvard-Yenching Institute Combined Indices to the Authors and Titles of Books in Two Collections of Taoist Literature kytpis K’ai-yüan T’ien-pao i-shih 開元天寶遺事 lthc Lin-t’ung hsien-chih 臨潼縣志 mhtl Ming-huang tsa-lu 明皇雜錄 skcs Ssu-k’u ch’üan-shu 四庫全書 sppy Ssu-pu pei-yao 四部備要 sptk Ssu-pu ts’ung-k’an 四部叢刊 tctc Tzu-chih t’ung-chien 資治通鑑 thy T’ang hui-yao 唐會要 tscc Ts’ung-shu chi-ch’eng 叢書集成 tscs T’ang-shih chi-shih 唐詩紀事 wh Wen hsüan 文選

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao: The Construction of Mosaic Memory in Medieval Historical Miscellanies Manling Luo Medieval* Chinese men of letters developed a tradition of gathering and writing down anecdotes as fragments of information on the past. Often referred to as zashi 雜史 (miscellaneous histories),1 these miscellanies have traditionally been regarded as containing historical information that is valuable in supplementing the official histories. But what kind of active function may the historical miscellany have served as a genre in its social context, other than a passive repository of diverse information? Here we shall examine the historical miscellany from the perspective of cultural memory construction, or more specifically, “mosaic” memory construction, in terms of creating composite, multifaceted pictures of the past with manifold informational fragments. Source: Luo, Manling, “Remembering Kaiyuan and Tianbao: The Construction of Mosaic Memory in Medieval Historical Miscellanies,” T’oung Pao 97 (2011): 263–300. *  This essay has benefited from the insightful comments and suggestions of many people. I would like to thank Robert E. Hegel, Paul W. Kroll, James Hargett, Lynn Struve, Heather Blair, Kevin Tsai, Robert Eno, Aaron Stalnaker, Kevin Martin as well as the reviewers. I presented a preliminary version at the conference “T’ang Studies: The Next Twenty-five Years,” held in Albany in 2009, and received useful feedback from conference participants, especially Ding Xiang Warner, Anna Shields, and Yang Lu. I am also grateful for Julia Whyde’s patient, meticulous editorial assistance. 1  The Sui shu 隋書 established categories such as zashi, zazhuan 雜傳, jiushi 舊事 in its bibliographic monograph; Sui shu (Beijing: Zhonghua shuju, 1973), 33.959–67. Qing editors made zashi a broad rubric for marginal writings of historical value; see Ji Yun 紀昀 (1724–1805) et al., “Zashi lei” 雜史類, Siku quanshu zongmu 四庫全書總目 (rpt. Beijing: Zhonghua shuju, 1965), 51.406. Modern scholars have adopted varying labels. Hou Zhongyi 侯忠義, for instance, uses yishi xiaoshuo 軼事小說; Hou, Sui Tang Wudai xiaoshuo shi 隋唐 五代小說史 (Hangzhou: Zhejiang guji chubanshe, 1997). For discussions on terminology, see Du Huimin 杜慧敏 and Wang Qinghua 王慶華, “ ‘Xiaoshuo’ yu zashi,’ ‘zhuanji’: yi Siku quanshu zongmu wei li” “小說” 與 “雜史”丶“傳記”: 以《四庫全書總目》為例, Wenxue yanjiu 2008.4: 73–79; also Zhang Zikai 張子開, “Yeshi, zashi he bieshi de jieding jiqi jiazhi” 野史丶雜史和別史的界定及其價值, Mianyang shifan xueyuan xuebao 28.3 (2009): 1–9. I have used “historical miscellany” as a general reference throughout this essay, leaving aside issues of categorization and terminology, which are not the focus of the present study.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_044

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1442

Luo

Mosaic memory opens up diverse avenues of intimate encounters with the past and constitutes a communal discourse that intersects with official history but develops largely as a tradition of its own. In the following discussion, four late medieval collections commemorating the reign of Emperor Xuanzong 唐玄宗 (685–762; r. 712–756), otherwise known as the Kaiyuan 開元 (713–742) and Tianbao 天寶 (742–756) era, will serve as primary examples. We shall see how writers’ personal motives and cultural positions influenced their idiosyncratic constructions of mosaic memory, which overlapped and enriched the texture of a cumulative representation, a layered formation with both individual and communal imprints. Such an historicized understanding not only helps to shed light on how medieval historical miscellanies shaped the memory of the Kaiyuan-Tianbao era, but also demonstrates the discursive power of the genre as an important site of cultural production and transformation.

Anecdotes, Historical Miscellany, and Mosaic Memory

Historical miscellanies were part of medieval miscellaneous writings that are now most often referred to as biji 筆記 (notebook writings) or biji xiaoshuo 筆記小說 (notebooks of minor discourses), among other terms.2 Dubbed “the most formless of genres” by Rania Huntington, biji is notoriously hard to generalize or classify.3 Not surprisingly, the boundaries distinguishing Tang historical miscellanies from other categories of writing are far from clear. For instance, they can cover supernatural or anomalous occurrences, the signature of the zhiguai 志怪.4 They often include entries on the words and actions of historical personages, which are the focus of what has been termed the zhiren 2  Liu Yeqiu 劉葉秋, Lidai biji gaishu 歷代筆記概述 (Beijing: Zhonghua shuju, 1980). Ning Jiayu 寧稼雨 includes more than two thousand titles under the rubric wenyan xiaoshuo 文言小說; see Ning, Zhongguo wenyan xiaoshuo shumu 中國文言小說書目 (Ji’nan: Qi Lu shushe, 1996). 3  Huntington, “Chaos, Memory, and Genre: Anecdotal Recollections of the Taiping Rebellion,” Chinese Literature: Essays, Articles, Reviews 27 (2005): 61. 4  For study of the early zhiguai, see Robert Ford Campany, Strange Writing: Anomaly Accounts in Early Medieval China (Albany: State Univ. of New York Press, 1996). For studies of later materials, see Sing-Chen Lydia Chiang, Collecting the Self: Body and Identity in Strange Tale Collections of Late Imperial China (Leiden: Brill, 2005); Rania Huntington, Alien Kind: Foxes and Late imperial Chinese Narrative (Cambridge, Mass.: Harvard Univ. Asia Center, 2003); Judith T. Zeitlin, Historian of the Strange: Pu Songling and the Chinese Classical Tale (Stanford: Stanford Univ. Press, 1993); idem, The Phantom Heroine: Ghosts and Gender in SeventeenthCentury Chinese Literature (Honolulu: Univ. of Hawai’i Press, 2007).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1443

志人 tradition.5 Nevertheless, the historical miscellany is usually marked by

the author’s gathering of anecdotes on the past from predominantly oral sources, in the style of “what one has heard and seen” (wenjian 聞見 or jianwen 見聞) and with an explicit or implicit agenda to supplement the more formal histories. Wu Jing 吳兢 (670–749), a court historian, presented his Zhenguan zhengyao 貞觀政要 to the throne, explaining that he had “patched together what [he] had heard and thoroughly consulted existing histories [for verification]” 缀集所聞, 參詳舊史.6 In the Sui Tang jiahua 隋唐嘉話 (Fine anecdotes from the Sui and Tang), a statement by the compiler Liu Su 劉餗 (fl. 742–756) is also illustrative: “Ever since I was a little child, I have often heard stories of the past. They are not worthy to be grand records, so I have appended them at the end of the minor discourses” 余自髫丱之年, 便多聞往說, 不足備之大典, 故繫之小說之末.7 Whether or to what extent the information included can be said to be historical lies at the heart of the traditional reception of the historical miscellany. As Jack W. Chen has pointed out, anecdote is “the literary vehicle of gossip,” originating in “the horizontal axes of societal relationships and networks” and representing “a very different order of knowledge” from that in the state’s sphere of control.8 From Confucius to Ban Gu 班固 (32–92) and to medieval writers, historians’ ambivalence toward the tradition of xiaoshuo, constituted prominently of anecdotes, is well known to modern readers.9 Liu Zhiji 劉知幾 (661–721) exemplifies a pragmatic attitude toward these materials, what he calls pianji xiaoshuo 偏記小說 (partial records and minor discourses) or zashu 雜述 (miscellaneous narratives). He divides them into ten types and considers them indispensible supplements to more formal sources. Meanwhile, he criticizes irresponsible, inferior recorders for failing to discern the true from the false.10 Later critics often consider compilers of historical miscellanies to

5  For study of the zhiren genre, see Nanxiu Qian, Spirit and Self in Medieval China: The Shihshuo hsin-yü and Its Legacy (Honolulu: Univ. of Hawai’i Press, 2001). 6  Zhenguan zhengyao jijiao 貞觀政要集校 (Beijing: Zhonghua shuju, 2003), 7–8. 7  Sui Tang jiahua (rpt. Beijing: Zhonghua shuju, 1979), 1.1. 8  Chen, “Blank Spaces and Secret Histories: Questions of Historiographic Epistemology in Medieval China,” Journal of Asian Studies 69 (2010): 1071–73. 9  For some examples, see ibid, 1078–82; and Ming Dong Gu, Chinese Theories of Fiction (Albany: State Univ. of New York Press, 2006), 17–42. 10   Shitong tongshi 史通通釋, ed. Pu Qilong 浦起龍 (1679–ca. 1762) (rpt. Shanghai: Shanghai guji chubanshe, 1978), 10.273–77.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1444

Luo

be good or flawed historians, or even fiction writers, depending on the veracity of their anecdotes.11 Trustworthiness, however, is a problematic criterion. The anecdotes are stories after all, and they are often found to contradict other sources.12 Oliver Moore approaches the issue from a different perspective by arguing that Tang writers of historical miscellanies were indeed writing minor histories, the content of which may not always seem believable to later readers because the yardstick of trustworthiness has shifted over time. In the second half of the dynasty, when the system of compiling state history (leading to the shilu 實錄 or “veritable records”) broke down, the court admitted private works into the fold of official historical sources, such as the Dongguan zouji 東觀奏記 (Record presented to the throne from the Dongguan) by Pei Tingyu 裴庭裕 (fl. 890–898).13 The use of anecdotal sources, as Chen has demonstrated, greatly complicates the historiographic ideology and epistemological standards of official histories in medieval China.14 These complexities lead Stephen H. West to argue that “we should consider the boundary between shi and xiaoshuo more from the point of view of the site of production and rhetoric than from the nature of sources.”15 What kind of site of production did the historical miscellany constitute? What was its active cultural function as a genre beyond its supplementary role to official history, a conventional designation that tends to treat the genre as mere reservoir of miscellaneous information? The questions are important 11  For evaluations of writers as good or flawed historians, see Liu Zhiji, Shitong tongshi, 10.273–77; Siku quanshu zongmu, 51.463; Li Limin 李利民, “Lun ‘Kai Tian yishi’ de zhuanlu xintai” 論 “開天遺事” 的撰錄心態, Xueshu luntan 8 (2006): 41–43. For assessment of compilers as fiction writers, see Yu Minghua 禹明華 and Feng Xiaoping 馮小萍, “Tang Wudai Kaiyuan Tianbao ticai biji xiaoshuo shulun” 唐五代開元天寶題材筆記 小 說述論, Hunan chengshi xueyuan xuebao 26 (2005): 64–68; Shi Jiajia 史佳佳, “Tang Xuanzong leixing xiaoshuo de sanzhong moshi jiqi yanbian tedian” 唐玄宗類型小說的 三種模式及其演變特點, Xichang xueyuan xuebao (shehui kexue ban) 20 (2008): 39–42. 12  Yue Hong, “A Structural Study of Ninth-Century Anecdotes on ‘Original Events,’ ” T’ang Studies 26 (2008): 65–83. 13  Moore, Rituals of Recruitment in Tang China: Reading an Annual Programme in the Collected Statements by Wang Dingbao (870–940) (Leiden: Brill, 2004), 57–62. For official historiography in the Tang, see Denis Twitchett, The Writing of Official History under the T’ang (Cambridge: Cambridge Univ. Press, 1992), and David McMullen, State and Scholars in T’ang China (Cambridge: Cambridge Univ. Press, 1988), 159–205. 14  Chen, “Blank Spaces and Secret Histories,” 1071–91. 15  West, “Crossing Over: Huizong in the Afterglow, or the Deaths of a Troubling Emperor,” in Emperor Huizong and Late Northern Song China, ed. Patricia B. Ebrey and Maggie Bickford (Cambridge, Mass.: Harvard Univ. Asia Center, 2006), 606.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1445

because compilations of historical miscellanies constituted a continuous tradition throughout the Tang. In his Que shi 闕史 (The missing history), the compiler, who calls himself Shenliaozi 參寥子, and is thought to be Gao Yanxiu 高彥休 (b. 854), notes the popularity of such practices: “Since the Wude (618– 627) and Zhenguan (627–650) reigns, people who have taken up the writing brush to write ‘minor discourses,’ ‘minor records,’ ‘petty histories,’ ‘unofficial histories,’ ‘miscellaneous records,’ or ‘miscellaneous accounts’ indeed have been many” 故自武德丶貞觀而後, 吮筆爲小說丶小錄丶稗史丶野史丶雜 錄丶雜紀者, 多矣.16 More than fifty titles are known today, and they seem to have covered consecutive historical periods. Parallel to Liu Su’s focus on the period starting with the Sui (581–619) and continuing through to Xuanzong’s reign, the Chaoye qianzai 朝野僉載 (Records of the court and the commonalty) by Zhang Zhuo 張鷟 (660–732) contains anecdotes ranging from the reign of Emperor Gaozong 唐高宗 (r. 649–684) to that of Xuanzong.17 As a matter of fact, when Li Zhao 李肇 (fl. 806–820) put together his Guoshi bu 國史補 (Supplement to the state history), he wrote that he intended to continue the tradition by picking up where Liu Su left off: “In former times Liu Su collected minor discourses from the Northern and Southern Dynasties to the Kaiyuan reign and wrote the Zhuanji; I have compiled the Guoshi bu to cover the period from the Kaiyuan to the Changqing reign (821–825)” 昔劉餗集小說, 涉南北朝 至開元, 著爲傳記, 予自開元至長慶撰國史補.18 To Shenliaozi, there had been so many collections written that it became a challenge for latecomers to find anything new to record, or in his words, “there is not any anecdote left unrecorded before the Zhenyuan (785–805) and Dali (766–780) periods” 貞元丶大 歷已前, 捃拾無遺事. He pointed out that his Que shi had to focus on the time from the Dazhong 大中 (847–860) and Xiantong 咸通 (860–874) eras onward.19 These statements suggest that collections enjoyed relatively broad circulation among educated men, to the extent that later compilers were often aware of their textual predecessors. In what follows we shall examine how the popular, social writing practices in the form of the historical miscellany enabled writers to undertake a peculiar sort of cultural memory construction within the medieval context.20 Expanding 16   Que shi (rpt. Beijing: Zhonghua shuju, 1985), 1.1. 17   Chaoye qianzai (rpt. Beijing: Zhonghua shuju, 1979). 18   Tang guoshi bu (rpt. Beijing: Gudian wenxue chubanshe, 1958), 1.3. The Zhuanji is usually taken to be an alternative title to the now commonly known Sui Tang jiahua. 19   Que shi, 1.1. 20  This is an important aspect of historical miscellanies but by no means the only one. Carrie E. Reed’s analysis of Duan Chengshi’s 段成式 (?–863) Youyang zazu 酉陽雜组, for

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1446

Luo

Maurice Halbwachs’s famous concept of “collective memory,” Jan Assman has pointed out that when personal, everyday memory is crystallized into “objectivized culture”—texts, rites, and monuments—the structure of memory remains constant in that the social group bases its consciousness of identity upon this shared knowledge, or “cultural memory.”21 The ample investigations of memory in the Chinese field have largely concentrated on the late imperial period, with a focus on “the interplay between the remembering mind and the sociocultural factors” in autobiographical writings.22 Huntington’s study of memories of the Taiping Rebellion (1851–1864) demonstrates that biji is more accommodating than other literary genres of remembrance for allowing diverse and even contradictory impulses to work together in an attempt to make sense of chaos.23 Besides the power of accommodation, historical miscellanies make possible a special kind of cultural memory construction in that writers piece together anecdotes to create a composite, multifaceted picture of the past. I call this multifaceted picture “mosaic memory” to underline its distinct mode of representation as a pastiche of fragments and its function in shaping how the past is remembered.24 Not intended as a theoretical essay, this paper is instance, shows the complexities of a hodge-podge that encourages diverse approaches and readings. See Reed, “Motivation and Meaning of a “Hodge-podge”: Duan Chengshi’s ‘Youyang zazu,’ ” Journal of the American Oriental Society 123 (2003): 121–45. 21  Assmann, “Collective Memory and Cultural Identity,” New German Critique 65 (1995): 125–33. 22  Lynn A. Struve, “Introduction to the Symposium: Memory and Chinese Texts,” Chinese Literature: Essays, Articles, Reviews 27 (2005): 3. The symposium presented in this journal explored the rich dynamic of memory construction in late imperial and modern periods. For good examples in that issue, see Grace S. Fong, “Inscribing a Sense of Self in Mother’s Family: Hong Liangji’s (1746–1809) Memoir and Poetry of Remembrance,” pp. 33–58, and Philip A. Kafalas, “Mnemonic Locations: The Housing of Personal Memory in Prose from the Ming and Qing,” pp. 93–116. Elsewhere, see Struve, “Chimerical Early Modernity: The Case of ‘Conquest Generation’ Memoirs,” in idem, ed., The Qing Formation in WorldHistorical Time (Cambridge, Mass.: Harvard Univ. Asia Center, 2004), 335–80, and “SelfStruggles of a Martyr: Memories, Dreams, and Obsessions in the Extant Diary of Huang Chunyao,” Harvard Journal of Asiatic Studies 69 (2009): 343–94. For more examples, see Trauma and Transcendence in Early Qing Literature, ed. Wilt L. Idema et al. (Cambridge, Mass.: Harvard Univ. Asia Center, 2006). 23  Huntington, “Chaos, Memory, and Genre,” 59–91. For a discussion of self-conscious remembrances in poetry and prose essays, see Stephen Owen, Remembrances: The Experience of the Past in Classical Chinese Literature (Cambridge, Mass.: Harvard Univ. Press, 1986). 24  In Tang historical miscellanies, the fragments are assumed to be synecdochic of the bygone reality. For other discussions of the fragment-whole dynamic, see Owen, Remem

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1447

a historical study that uses “mosaic memory” as an interpretive lens for understanding the particular cultural function of historical miscellanies. Imbued with personal motives, the construction of mosaic memory is also a communal discourse formation that defines a shared memory of the past through a distinct mechanism, the pastiche and overlay of stories. Mosaic memory thus represents a different kind of historical engagement and imagination than those of official history. Chen has observed that the perspective of dynastic histories is “that of the state, which claims to represent the whole, if not in the individual biographical accounts, then in the aggregate of all the chapters.” By contrast, anecdotes embody the private knowledge born of the social network of gossip and they “[put] flesh on the bones of annalistic factuality”25 These comparisons are useful in showing the contrast between the more distant kind of overview at the macro-level of history and a local, close-up, and more intimate kind of historical imagination on the micro-level. As a compendium of anecdotes, mosaic memory represents the past in a discursive, kaleidoscopic mode, with snapshots portraying vivid details of a bygone social and political life, revealing diverse and even contradictory points of view. The tone is that of intimate one-way gossip as the writer passes on chit-chat that he assumes will be of interest to the reader, jumping more or less randomly from topic to topic. Even in a serious political collection such as Wu Jing’s Zhenguan zhengyao, the anecdotes offer a more intimate understanding of Emperor Taizong 唐太宗 (r. 626–649) and his officials by describing their close conversations and interactions, with a wealth of vivid details that the official annals cannot accommodate.26 Moreover, by offering samples of contemporary public discourse, mosaic memory also facilitates intimate encounters with the social community from which compilers draw their stories and for which they serve as spokesmen. These authors may be better characterized as story-tellers rather than fiction writers or historians.27 As they drew from a fund of anecdotal narratives that can be alternatively termed gossip and rumors to create composite pictures brances, 66–79; Campany Strange Writing, 1–17; Qian, Spirit and Self in Medieval China, 43–190. 25  Chen, “Blank Spaces and Secret Histories,” 1086–88. 26  Qian notes a similar sense of intimacy and vividness in the Shishuo xinyu; see Spirit and Self in Medieval China, 207–10. 27  Since “storytelling” usually refers to oral performances and “storyteller” to the performer, I use “story-telling” to encompass written modes of recounting stories and “story-teller” to designate the author-compiler. These adaptations are not to be confused with Mieke Bal’s narratological terms devoted to the study of Western narrative literature, especially fiction; see Bal, On Story-Telling: Essays in Narratology (Sonoma: Polebridge Press, 1991).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1448

Luo

of the past, their story-telling was not creative fiction writing in the modern sense.28 Although they often positioned themselves as historians, sometimes even evoking such high models as Confucius and Sima Qian, their works, as we shall see in more detail, were distinct from conventional historiography. Because it is difficult to draw clear lines between creativity and fidelity, or the individual and the collective, it is useful to examine the function of such storytelling as a distinct form of authorship, or how writers helped to sustain the cultural memory of the past by serving as story-tellers for their community. For the present analysis, I have chosen to focus on post-An Lushan Rebellion collections devoted to the memory of the Kaiyuan-Tianbao period. Just as the Yuanhe era (806–821) was seen as an important moment of political renewal and literary flowering in mid-ninth century and later, Kaiyuan-Tianbao was considered a historical watershed in the post-rebellion period.29 Noting the parallel to efforts to “fix” memories of a momentous age in the West, Paul W. Kroll sees the ninth-century literati interest in Emperor Xuanzong’s reign in general as embodying the universal human pursuit of “memory and invention.”30 Even though mid- and late-Tang writings on Kaiyuan-Tianbao as found in other genres have been well studied, the important role of historical miscellanies in shaping the cultural memory of this era has yet to be recognized.31 Although 28  Scholars have challenged the conventional model of reading Tang stories as fiction, arguing instead that they were narrative public property, products of a contemporary culture of oral storytelling and the written circulation of stories. See Sarah M. Allen, “Tang Stories: Tales and Texts” (Ph.D. diss., Harvard Univ., 2003); idem, “Tales Retold: Narrative Variation in a Tang Story,” Harvard Journal of Asiatic Studies 66 (2006): 105–43; also Glen Dudbridge, Religious Experience and Lay Society in T’ang China (Cambridge: Cambridge Univ. Press, 1995). For discussions beyond the Tang, see Campany, Making Transcendents: Ascetics and Social Memory in Early Medieval China (Honolulu: Univ. of Hawai’i Press, 2009); Alister D. Inglis, Hong Mai’s Record of the Listener and Its Song Dynasty Context (Albany: State Univ. of New York Press, 2006); Leo Tak-hung Chan, The Discourse on Foxes and Ghosts: Ji Yun and Eighteenth-Century Literati Storytelling (Honolulu: Univ. of Hawai’i Press, 1998). 29  Anna M. Shields, “Alternate Views of Literary History: The Yuanhe Reign Period in Tang and Five Dynasties Anecdotal Texts,” Chūgoku shigaku 20 (forthcoming). 30  Kroll, “Nostalgia and History in Mid-Ninth-Century Verse: Cheng Yü’s Poem on ‘The Chinyang Gate,’ ” T’oung Pao 89 (2003): 306–10. 31  For discussions of huaigu 懷古 poems on Kaiyuan-Tianbao and the famous twin compositions, “Changhen ge” 長恨歌 by Bai Juyi 白居易 (772–846) and “Changhen ge zhuan” 長恨歌傳 by Chen Hong 陳鴻 (fl.785–830), see Jin Jicang 靳极苍, Changhen ge jiqi tong ticai shi xiangjie 長恨及其同題材詩詳解 (Taiyuan: Shanxi guji chubanshe, 2002); Zhang Zhongyu 張中宇, Bai Juyi Changhen ge yanjiu 白居易《長恨歌》研究 (Beijing: Zhonghua shuju, 2005). For analysis of Zheng Yu’s 鄭嵎 (jinshi 851) poem “Jinyang men

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1449

anecdotes on Xuanzong and his reign appear in earlier collections by Zhang Zhuo and Liu Su and in many later ones, including Li Zhao’s Guoshi bu and Li Jun’s 李濬 (fl. 877–879) Songchuang zalu 松窗雜錄 (Miscellaneous records by the pine-shaded window), four monothematic collections became canonical representations of Kaiyuan-Tianbao because of their exclusive treatment of the subject.32 They are the Ci Liu shi jiuwen 次柳氏舊聞 (Jottings of tales heard from the Lius) by Li Deyu 李德裕 (787–849), Minghuang zalu 明皇雜錄 (Miscellaneous record of the Illustrious Emperor) by Zheng Chuhui 鄭處誨 ( jinshi 834), Kai Tian chuanxin ji 開天傳信記 (Record of transmitted facts from Kaiyuan and Tianbao) by Zheng Qi 鄭棨 (fl. late ninth century), and Kaiyuan Tianbao yishi 開元天寳遺事 (Leftover anecdotes from Kaiyuan and Tianbao) by Wang Renyu 王仁裕 (880–956).33 These collections span roughly a century from the first compilation in 834 to the last, completed sometime after 925. They offer valuable examples for looking into the layered formation of mosaic memory and its image of the past produced through a cumulative process of pastiche and overlay. I will discuss the idiosyncratic constructions of mosaic memory in the four collections and conclude with a reflection on its communal aspects and broader implications.

shi” 津陽門詩, see Kroll, “Nostalgia and History in Mid-Ninth-Century Verse,” 286–366. For the “Dongcheng laofu zhuan” 東城老父傳 and the culture of cockfighting, see Robert Joe Cutter, The Brush and the Spur: Chinese Culture and the Cockfight (Hong Kong: Chinese Univ. Press, 1989). 32  The emperor appears in more than three hundred anecdotes, making him the most talked about monarch of the Tang during the post-rebellion period. In addition to the collections mentioned here, there are more than twenty others, such as the Feng shi wenjian ji 封氏聞見記 by Feng Yan 封演 (fl. 756–800) and Da Tang xinyu 大唐新語 by Liu Su 劉肅 (fl. 806–820). 33  There was allegedly a fifth collection, Tianbao gushi 天寶故事 attributed to Zheng Shen 鄭審 (fl. 737–768), which is not extant but is cited in a few post-Tang sources. See Sima Guang 司馬光, Zizhi tongjian kaoyi 資治通鑑考異 (Sibu congkan ed.), 14.1a. In this essay, references to the four collections are marked by an abbreviated title with page number in parenthesis. The editions used are as follows: Ci Liu shi jiuwen, Congshu jicheng edition (Changsha: Shangwu yinshu guan, 1930 [hereafter Jiuwen]); Minghuang zalu, ed. Tian Tingzhu 田廷柱 (Beijing: Zhonghua shuju, 1994 [hereafter Zalu]); Kai Tian chuan­ xin ji (Congshu jicheng edition [hereafter Chuanxin ji]); Kaiyuan Tianbao yishi (Beijing: Zhonghua shuju, 2006 [hereafter Yishi]). Because my goal is to be illustrative rather than comprehensive, I do not discuss the textual diversity of extant editions but focus on the stable core of anecdotes in each collection to show how the writers’ idiosyncratic tendencies inform their constructions of mosaic memory.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1450

Luo

Li Deyu’s Ci Liu shi jiuwen: The Politics of Intimacy

Li Deyu’s collection exemplifies the fundamental dynamics of an individual construction of mosaic memory. By turning an oral history into a written record for his emperor, Li presents a composite picture of Xuanzong’s reign that caters to his intended audience and serves his personal agenda. Li Deyu was an important political figure of his time and his collection was politically motivated.34 According to his preface, Li first heard the seventeen anecdotes in his collection from his father, Li Jifu 李吉甫 (758–814), who in turn had heard them from his colleague Liu Mian 柳冕 (ob. 805). Liu Mian was the son of Liu Fang 柳芳 (fl. 741–760), who in exile had become acquainted with Gao Lishi 高力士 (684–762), the trusted eunuch of Xuanzong and the presumed original narrator of these stories. Though Gao Lishi’s reminiscences belonged to personal memory, what Li Deyu wrote down was already generations removed from the source. Meanwhile, this transmission of information did not follow official avenues but rather ran through a private lineage of family and friends, although Liu Fang had been an official historian. Putting together these anecdotes to present to Emperor Wenzong 文宗 (r. 827–840), who during a meeting with his ministers in the eighth month of 834 asked about the life of Gao Lishi, Li Deyu claimed that he intended “to supplement what might have been missed by the official historians” 以備史官之闕云 ( Jiuwen, 1). In other words, what Li recorded had become a form of historical knowledge, which he tried to inscribe into shared, cultural memory. Li Deyu, however, carefully tailored his version of mosaic memory for the political context. Along with this collection of anecdotes, Li presented another book to the throne entitled Yuchen yaolüe 御臣要略 (Outline of strategies for reining in court officials). The presentation occurred just one year before the tragic “Sweet Dew Incident” 甘露之變, in which officials who were secretly allied with Wenzong failed in an attempt to oust a coterie of powerful eunuchs and were met with bloody retaliation.35 This background may help explain the subtle discrepancy between what Wenzong had asked for and what Li Deyu 34  For biographies of Li Deyu, see Jiu Tang shu 舊唐書 (Beijing: Zhonghua shuju, 1975), 174.4509–30, and Xin Tang shu 新唐書 (Beijing: Zhonghua shuju, 1975), 180.5327–44. Sixteen editions of the collection are extant. See Yuan Xingpei 袁行霈 and Hou Zhongyi, Zhongguo wenyan xiaoshuo shumu 中國文言小說書目 (Beijing: Beijing daxue chubanshe, 1981), 56. 35   Jiu Tang shu, 17.562–65. For a discussion of the incident and its influence on literature, see Hu Kexian 胡可先, Tangdai zhongda lishi shijian yu wenxue yanjiu 唐代重大歷史事件 與文學研究 (Hangzhou: Zhejiang daxue chubanshe, 2007), 495–565.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1451

presented. The emperor had inquired about Gao Lishi, the most powerful eunuch throughout Xuanzong’s reign, who appears in eight of the seventeen entries as a trustworthy messenger and an intimate companion for Xuanzong. Judging from its title, Outline of Strategies for Reining in Court Officials, the additional presentation must have been a treatise on how to manage and control subordinates, although the book is not extant. Li Deyu’s combination of a book of disquisition with that of historical examples suggests that he anticipated Wenzong’s desire to reassert his royal power as the head of state. Because Li Deyu’s first intended reading audience was Wenzong, he was more interested in presenting Xuanzong as a role model for the emperor, giving Gao Lishi only secondary status in his collection. Xuanzong is cast as a loving brother and caring father and grandfather; respectful of his ministers, he also attracts the service of extraordinary Daoist and Buddhist figures and takes the interests of his people to heart. In his effort to portray the emperor in such a positive light, Li Deyu tries to minimize inconvenient elements. Take, for instance, the story in which the emperor, fleeing to Sichuan as a result of the An Lushan rebellion, rejects an offering of wine on the grounds that he had not enjoyed the taste of alcohol for more than forty years, ever since, in a drunken state, he killed a man soon after ascending the throne. Li Deyu comments at the end of the anecdote: “The emperor assiduously admonished himself like this. He presided over the kingdom for nearly fifty years. Did this not result from this very Way [of self-discipline]?” 上孜孜儆戒也如是 。 富有天 下, 僅五十載, 豈不由斯道乎? ( Jiuwen, 7). Li imposes commentaries to connect Xuanzong’s alleged virtue of self-discipline with his long stay in power, trying to submerge the irony that he had been forced to flee the capital for failing to defeat rebel forces and losing control of his country.36 In posing his interpretation as a rhetorical question, Li suggests that it is self-evident and invites the reader’s sympathy. The careful maintenance of Xuanzong’s image thus flatters Wenzong for his personal connection with this exemplary monarch and encourages him to identify with his illustrious ancestor. Among the monarchic ways of conduct that Li wishes Wenzong to emulate, the intimate relationship between Xuanzong and his early chief ministers is given prominence: six entries (one-third of the whole collection) focus on the emperor’s interactions with or appointments of chief ministers. Take the following entry, for instance:

36  For more details, see Kroll, “The Flight from the Capital and the Death of Precious Consort Yang,” T’ang Studies 3 (1985): 25–53; also Edwin G. Pulleyblank, The Background of the Rebellion of An Lu-shan (London: Oxford Univ. Press, 1955).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1452

Luo

When [the future] Emperor Xuanzong was Crown Prince, [his aunt] Princess Taiping felt threatened by him. She spied on him from morning to night and reported his slightest mistakes to [his father] Emperor Ruizong. The attendants of the palace secretly straddled both sides in deference to her power. At that time, [the future] Empress Yuanxian was favored by Xuanzong and had just become pregnant. Afraid of Princess Taiping, Xuanzong wanted Yuanxian to take medicine to abort the fetus, but there was no one for him to speak with [about the matter].37 Zhang Yue was able to enter the palace of the Crown Prince in his role as a Grand Tutor. When Xuanzong casually brought up the issue with him, Zhang secretly offered support. On another day, when Zhang came in again, he hid three doses of abortifacient within his robe and presented them to Xuanzong. Xuanzong was glad to get the medicine and, dismissing all of his attendants, lit a brazier in the hall in order to prepare it by himself. Before the medicine was ready, he was tired and while he started to doze off, he caught a glimpse of a god who was more than ten feet tall and wearing golden armor. With a dagger-axe in hand, the god circled the medicine three times and flipped the pot over, so that nothing was left. Xuanzong rose to check it and was amazed. After he stoked the fire and put in another dose, he returned to his couch and waited, pretending to close his eyes. He saw the god turn over the pot again. All three doses were [eventually] overturned. The next day when Zhang arrived, Xuanzong told him everything. Zhang descended the stairs [to the courtyard] to do obeisance and offer congratulations. He said, “This fetus is ordained by Heaven and cannot be eliminated.” Later on, when the pregnant mother had a craving for something sour, Xuanzong also told Zhang. Zhang then hid quinces in his sleeves [for her] whenever he came to deliver lectures on the classics [to the crown prince]. As a result, no one could rival the favor that Zhang enjoyed during the Kaiyuan reign. It is said that [the future] Emperor Suzong [who was the child born from this pregnancy] treated Zhang’s sons Jun and Ji like his own relatives and brothers. Liu Fang was recommended to office by Zhang and personally heard him relate this story, which was consistent with the words of Gao Lishi. 37  Normally a male heir would strengthen the status of a crown prince. Because Princess Taiping sought to undermine the future Xuanzong, the baby would have made her more desperate and hence more aggressive in doing harm. Princess Taiping was very influential during the reign of Emperor Ruizong 唐睿宗 (r. 710–712) as his favorite sister. She gradually developed an antagonistic relationship with her nephew, the crown prince and future Emperor Xuanzong, who eventually took actions against her and her allies.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1453

玄宗之在東宫 。 爲太平公主所忌 。 朝夕伺察 。 纖微聞於上 。 而宫闈 左右 。 亦潛持兩端 。 以附太平之勢 。 時元獻皇后得幸方娠 。 玄宗懼 太平 。 欲令服藥除之 。 而無可與語者 。 張說以侍讀得進太子宫中 。 玄 宗從容謀及說 。 說亦密贊其事 。 他日 。 說又入侍 。 因懷去胎藥三煑劑 以獻 。 玄宗得其藥 。 喜 。 盡去左右 。 獨搆火殿中 。 煑未及熟 。 怠而假 寐 。 盻蠻之際 。 有神人長丈餘 。 身披金甲 。 操戈繞藥三匝 。 煑盡覆而無 遺焉 。 玄宗起視 。 異之 。 復 增火 。 又投一劑 。 煑於鼎中 。 因就榻瞬目 以候之 。 而見神覆煑如初 。 凡三煑皆覆 。 乃止 。 明日 。 說又至 。 告其 詳 。 說降階拜賀曰 。 天所命也 。 不可去 。 厥後 。 元獻皇后思食酸 。 玄 宗亦以告說 。 說每因進經 。 輒袖木瓜以獻 。 故開元中 。 說恩澤莫之與 比 。 肅宗之於說子均 、 洎, 若親戚昆弟云 。 芳本張說所引 。 說嘗自陳 述 。 與力士詞協也 。 ( Jiuwen, 1–2)

Although in this anecdote Zhang Yue (667–731) is just a tutor, Emperor Xuanzong is still a crown prince, and Emperor Suzong 唐肅宗 (r. 756–762) an unborn child, the dynamics of the story are built upon their future identities as chief minister, emperor, and next emperor in line. The account projects an idealized vision of the intimacy between ruler and chief minister. Confinement of the crown prince in the palace and the key role of Zhang as his only contact with the outside world illustrate their symbiotic relationship in that the lord depends on the minister to carry out his intentions and rule the empire. He also counts on Zhang for authoritative interpretation and articulation: Zhang’s proclamation that the guardian god’s actions represent the mandate of Heaven asserts a confident voice of wisdom and intellectual power. Furthermore, his performance of obeisance also ritualistically fulfills and validates the prediction by formally treating the crown prince as an emperor. His contribution of the abortifacient first and quince later is highly symbolic in that the medicine to kill and the food to nurture represent the power to eliminate and to preserve the heir to the throne. In other words, he wields a power that can shape the political future of the dynasty. Zhang Yue’s critical role in the pregnancy of the crown prince’s consort thus exemplifies an idealistic fusion of the personal and the political by exhibiting the extent to which intimacy could develop between a minister and his monarch, a personal bond which, as the story suggests, deepens and undergirds the political success of both Zhang and Xuanzong. This thematic emphasis on personal and political intimacy runs through other anecdotes concerning Xuanzong and his chief ministers. We are told that when he first ascends the throne, the emperor pays special respect to Yao Chong 姚崇 (650–721) and Song Jing 宋璟 (663–737) by rising up to meet them when they arrive and seeing them off in person at the veranda, even entrusting Yao with the power to monopolize the promotion and demotion of

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1454

Luo

lower-ranking officials ( Jiuwen, 2). In another story, when an official informs against Yao Chong for not disciplining his own sons, the emperor allows Yao to justify himself and comes to trust Yao’s story rather than that of the informer ( Jiuwen, 2–3). When the emperor promotes Yuan Qianyao 源乾曜 (ob. 731), he does so because of his nostalgia for the deceased Xiao Zhizhong 蕭至忠 (ob. 713), whom Yuan happened to resemble physically ( Jiuwen, 3). When Xiao Song 蕭嵩 (ca. 668–749) requests retirement, the emperor instead gives him the post of the Right Minister, wrapping with white silk two tribute oranges newly arrived from the south to bestow on him ( Jiuwen, 3). In a story in which the emperor is said to always handwrite the names of chief ministers he is about to appoint, the intimacy is not physical, but the emphasis on his per­ sonal care for the appointees remains constant ( Jiuwen, 4–5). By foregrounding the intimate terms between the emperor and his chief ministers, these anecdotes constitute a systematic reiteration and idealization of the special bond between the two parties. The merging of personal and political intimacy in the anecdotes also informs the relationships between the internal characters, readers, and the author. Behind Zhang Yue’s story of smuggling abortifacient and quinces, for instance, are historical incidents in the turbulent power struggle between Princess Taiping and the future Emperor Xuanzong, in which Zhang as his major advisor played an important role in securing for him the final victory. Yet such bloody political battles are well-known facts related in the grand narrative of official discourse; the anecdote offers instead the vivid details of everyday life, pregnancy, food, and drugs associated with court figures, in other words, the delicious stuff of gossip. The retrospective projection of the minimelodrama also justifies the political status quo while endearing the characters to readers. The future emperor is unjustly driven to such a dejected state that he even attempts to abort his own child; small wonder that he struck back and succeeded with the blessing of Heaven. Meanwhile, Zhang Yue’s rise to prominence is a route with which everyone can identify: given the right place and time, anyone can deliver needed assistance and support. This anecdote, like many others, thus brings key historical moments and figures of the state to a personal level for readers; it is this kind of intimate encounter with the past that distinguishes the appeal of mosaic memory from that of official history. Likewise, the issue of intimacy also defines Li Deyu’s efforts to bond with his imagined readers. By explaining his two sources of the Zhang Yue story, one from Xuanzong to Gao Lishi and the other from Zhang Yue to Liu Fang, Li Deyu provides an epilogue that showcases his knowledge of the origins of the story proper as an insider. This new story again conflates personal and political intimacy: Liu Fang was recommended to office by Zhang and heard him relate the

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1455

story, while as the preface states, he also became a personal friend of Gao Lishi in exile and heard from him the emperor’s side of the story. The information not only serves to bolster the veracity of the entry and the narrative authority of Li Deyu, but also to highlight a continuous, authentic lineage of transmission from the protagonists of the story all the way down to Li himself, which he now extends to include his readers. Given that Li Deyu served as one of Wenzong’s chief ministers, the personal, political, and narrative intimacy that he promotes in his collection demonstrates that his book presentation was an unmistakable gesture to cultivate a special bond with the emperor.38 In this relationship, he strategically positions himself as a devoted advisor who offers a model of imperial conduct, an intimate insider who shares a privileged knowledge that is missing from official records, and a key transmitter who passes on information valuable to the emperor’s family and to state history. Li Deyu’s case thus illustrates how intimate fragments of casual conversations between colleagues, friends, and family members can be repackaged and turned into political capital, personal motives shaping the individual construction of mosaic memory. Although we do not know the effect of the book’s submission on the relationship between Wenzong and Li Deyu, it is interesting to note that after Wenzong’s brother, Emperor Wuzong 武宗 (r. 840–846), ascended the throne, Li became the ruler’s most trusted chief minister.39 Perhaps in this new relationship, Li fulfilled the idealized vision that he propagates in his collection.

Zheng Chuhui’s Minghuang zalu: Literary Talent and Officialdom

Compared to Li Deyu’s personalized version of mosaic memory, Zheng Chuhui’s evinces less the imprint of personal motives than the inscription of communal values. The prominent themes of literary talent and officialdom in the collection indicate how his background as a successful literary man and an active office-holder informs the outlook of his version, promoting the shared values of the literati community at large.

38  Li’s effort was probably encouraged by his previous success. In 825, he submitted a group of six poems entitled “Danyi zhen” 丹扆箴 to Emperor Jingzong 唐敬宗 (r. 824–827), the elder brother of Wenzong, to encourage him to devote himself to monarchic duties. Jingzong is said to have responded warmly to Li’s presentation. Jiu Tang shu, 124.4514–15. 39  For a study of Li’s role as Wuzong’s chief minister, see Michael R. Drompp, Tang China and the Collapse of the Uighur Empire: A Documentary History (Leiden: Brill, 2005).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1456

Luo

Zheng Chuhui was a grandson of the famous chief minister Zheng Yuqing

鄭餘慶 (748–820). He held several official posts during his lifetime after pass-

ing the highly competitive jinshi examinations in 834, the year in which Li Deyu presented his collection to the throne.40 Although a preface attributed to Zheng Chuhui has not survived, Song historians believed that his collection was prompted by his dissatisfaction with the limited scope of Li Deyu’s version.41 However, Zheng’s motivations may have been expressly political, for he seems to have belonged to the faction of Li Deyu’s enemies.42 Kroll points out that Zheng might have accessed archival sources when he worked as a collator in the imperial library and that the compilation could have been politically motivated.43 Zheng Chuhui modifies Li Deyu’s thematic orientation to underscore the role of literary writing. Anecdotes about Xuanzong’s appointments of chief ministers shift their focus to the performance of edict writers, in contrast to Li Deyu’s emphasis on the emperor’s intimate words and gestures toward the appointees. For instance, in the story of the appointment of Zhang Jiazhen 張嘉貞 (665–729), the instant composition of an elegant edict by Wei Kang 韋抗 (667–726) wins the emperor’s praise; Wei is also able to swiftly compose another after the emperor catches a mistake (Zalu, 12). By contrast, the story of the appointment of Su Ting 蘇頲 (670–727) serves as a negative example, in which Xiao Song makes a serious blunder by using a character appearing in the name of Su’s father and is too embarrassed by the emperor’s perceptiveness to make any revisions to his composition (Zalu, 34). Although Zheng Chuhui still reiterates the theme of personal and political intimacy between Xuanzong and his officials, he has substituted a gifted literary writer for Li Deyu’s ideal chief minister. The following entry on Su Ting is a case in point: Su Ting far surpassed others in intelligence and could recite [from memory] several thousand words in a day. Although his memory was superb, 40  For biographies of Zheng Chuhui, see Jiu Tang shu, 158.4168–69, and Xin Tang shu, 165.5062; also Kroll, “Nostalgia and History in Mid-Ninth-Century Verse,” 298. Five versions of the collection are extant. See Tian Tingzhu, ed., Minghuang zalu, 3–6; Yuan Xingpei and Hou Zhongyi, Zhongguo wenyan xiaoshuo shumu, 62. 41   Xin Tang shu, 165.5062. 42  See Kroll, “Nostalgia and History in Mid-Ninth-Century Verse,” 298, n. 22. Supporting evidence can be seen in the account of Li Deyu demoting Wei Wen 韋溫 to a provincial post for his association with Li’s enemies. Wei is said to have enlisted Zheng Chuhui on his staff, a move that further displeased Li. Jiu Tang shu, 168.4380. 43  Kroll, “The Dancing Horses of T’ang,” T’oung Pao 67 (1981): 244.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1457

his father Su Gui still disciplined him strictly; he was often ordered to prostrate himself in front of his bed wearing a black cotton robe and leaving his legs uncovered to be beaten. After Ting grew up, his erudition in literature and learning ranked at the top among his contemporaries, but he was unrestrained in his behavior and addicted to drinking. After Emperor Xuanzong had overcome the enemies of the state, he was troubled by the difficult choice of the right candidate to draft his edicts. He turned to ask Gui: “Who can write edicts well? Please think about it for me.” Gui answered, “I, your servant, don’t know of others, but my son Ting is rather quick-minded and could wait on Your Majesty’s orders. He is prone to drinking, but if fortunately he can avoid getting drunk, he can get the work done.” The emperor immediately summoned Ting. When he arrived, he had a hangover and could only make rough obeisance. [Earlier] he had vomited outside the hall, [so now] the emperor ordered the attendants to support him and lay him down in the imperial presence, and then personally covered him with a comforter. After Ting woke up, he picked up the paper and brush that had been provided and instantly finished the composition; its brilliant embellishments were unhindered and its diction was elegant. Greatly pleased, the emperor patted his back and said, “No one knows the son better than the father. Is there any case like this?” From this point on he thought highly of Ting and intended to put him in important positions. When Wei Sili was appointed Director of the Secretariat, it was Gui’s signature on the official announcement, Ting’s composition, and Xue Ji’s calligraphy, which the contemporaries called Three Masters par excellence. When Ting was just old enough to speak, a governor [yin 尹] of the capital visited Gui and asked Ting to compose a poem on the character yin 尹. In immediate response, Ting said, “Although the clown (丑) has a foot—his armor (甲) doesn’t cover his whole body./ When he sees you (君), he becomes speechless— Realizing that he (伊) has lost a companion.”44 Gui was in close contact with Zhou Yanyun, a Daoist priest at the Dongming Temple. When Zhou wanted to establish a memorial stele for his master, he said to Gui, “To fulfill my wish, I need only to trouble your sons: the composition of your 44  When “the clown (丑) has one foot,” the character is given an additional stroke at the bottom and becomes 尹. When “his armor (甲) doesn’t cover his whole body,” the character loses the left vertical stroke and also becomes 尹. The line “When he sees you (君), he becomes speechless” means that the character loses the bottom element 口 (mouth) and turns into 尹. The last line “realizing that he (伊) has lost a companion” means that the left element 人 (people) is removed and the character also turns into 尹.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1458

Luo

fifth son, the calligraphy of your sixth, and the carving of your seventh.” Gui laughed and although he didn’t say anything, he admired the fairness of Zhou’s evaluation. Ting was Gui’s fifth son, Xian the sixth, and Bing the seventh; Xian excelled at bafen style calligraphy.45 蘇頲聰悟過人, 日誦數千言, 雖記覽如神, 而父瓌訓厲至嚴, 常令衣青布襦伏於牀 下, 出其脛受榎楚 。 及壯, 而文學該博, 冠於一時, 性疎俊嗜酒 。 及玄宗既平内 難, 將欲草制書, 甚難其人, 顧謂瓌曰: 「誰可爲詔? 試為思之 。」瓌曰:「臣不知   其他, 臣男頲甚敏捷, 可備指使 。 然嗜酒, 幸免沾醉, 足以了其事 。」玄宗遽命 召來 。 至時宿酲未解, 粗備拜舞 。 嘗醉嘔殿下, 命中人扶臥於御前, 玄宗親爲 舉衾以覆之 。 既醒, 受簡筆立成, 才藻縱橫, 詞理典贍 。 玄宗大喜, 撫其背曰: 「知子莫若父, 有如此邪? 」由是器重, 已注意於大用矣 。 韋嗣立拜中書令, 瓌 署官告, 頲爲之辭, 薛稷書, 時人謂之三絕 。 頲纔能言, 有京兆尹過瓌, 命頲詠「 尹」字, 乃曰: 「丑雖有足, 甲不全身, 見君無口, 知伊少人 。」瓌與東明觀道士 周彥雲素相往來, 周時欲爲師建立碑碣, 謂瓌曰: 「成某志不過煩相君諸子: 五 郎文, 六郎書, 七郎致石 。」瓌大笑, 口不言而心服其公 。 瓌子頲第五, 詵第 六, 冰第七, 詵善八分書 。 (Zalu, 12–13)

The physical intimacy between the emperor and Su Ting not only represents an idealized vision of literary talent hewing a shortcut through the social thicket to the power center, but also sets the ideal bond between emperor and literary genius in parallel with the parent-child relationship. The account emphasizes the resemblance of Su Ting in his drunken stupor to a child fast asleep and the emperor’s paternal, loving gesture of covering him with a comforter. Su Ting’s drunkenness is symbolic in that by showing him in a moment of selfindulgence and, implicitly, of self-confidence and pride, the image highlights the right of literary talent to claim a temporary independence from political hierarchy.46 Yet this momentary freedom is meant to be an expression of subjectivity and individuality, not a serious challenge to the political structure, just as the impressive performance of the sober Su Ting proves his effective and indispensible role within the political system. This theme of affirming 45  The bafen style calligraphy evolved from the xiaozhuan 小篆 and lishu 隸書 and was the precursor of the later, regular script known as kaiti 楷體. See Zheng Xiaohua 鄭曉華, Hanyi shenfei: Zhongguo shufa yishu de lishi yu shenmei 翰逸神飛: 中國書法藝術的歷 史 與審美 (Beijing: Zhongguo renmin daxue chubanshe, 2003), 90–112. 46  Another drunken genius associated with Xuanzong’s court who appears in Tang historical miscellanies is Li Bai 李白 (701–762?), who was favored over Su Ting in post-Tang rewritings, such as the anonymous Jinghong ji 驚鴻記 and Tu Long’s 屠隆 (1541–1605) Caihao ji 綵毫記.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1459

an independent literary identity and reconciling it with the political system seems to parallel the trends of constructing a private space, which is both part of officialdom and above it, in mid- and late-Tang poetry.47 If the poetic counterparts choose to idealize the pleasant, liberating moments of the poetic self in the private space, the narrative explores the realistic condition of this literary freedom in political terms: it is contingent upon the emperor’s elegant accommodation in the mode of a doting father. The entry on Su Ting also illustrates Zheng Chuhui’s shift of focus from Xuanzong alone to the broader circles of the elite. The formal perfections of the official announcement of Wei Sili’s 韋嗣立 (654–719) appointment with Su Gui’s 蘇瓌 (639–710) signature, Su Ting’s composition, and Xue Ji’s 薛稷 (649– 713) calligraphy reflect the taste of a community of office holders and aspirants who are keenly interested in the political and cultural markers of prestige in officialdom. The Daoist priest’s request flatters Su Gui because it is a due recognition of the concentration of cultural capital within his family. The beginning anecdote on Su Ting’s strict upbringing offers an explanation for this family achievement, an idealized vision of family life in which the father exerts his authority to mold his precocious son into a talented writer and future pillar of the state. Su Ting’s precocity is proved by his ingenious poem on the character yin, which turns the subtle formal differences between characters into comic caricatures that also give a clever spin to their literal meanings. If the child prodigies in Liu Yiqing’s Shishuo xinyu show their talent through spontaneous and clever replies, capacities highly valued in the milieu of qingtan 清談 (pure conversation) culture, Su Ting’s extraordinary performance in poetic composition marks a shift in what was considered precocious talent. Another child prodigy included in Zheng Chuhui’s collection, the nine-year old Liu Yan 劉晏 (715–780), also proves his genius in a similar way: at the behest of Xuanzong’s favorite consort Yang Guifei 楊貴妃 (719–756), he instantly composes a quatrain on acrobatic performances (Zalu, 13).48

47  Owen, “Wit and the Private Life,” in his The End of the Chinese ‘Middle Ages’ (Stanford: Stanford Univ. Press, 1996), 83–106. Xiaoshan Yang, Metamorphosis of the Private Sphere: Gardens and Objects in Tang-Song Poetry (Cambridge, Mass.: Harvard Univ. Asia Center, 2003). 48  This story is evidently anachronistic in that Lady Yang would have been only five years old at the time. Although modern scholars often attribute such “mistakes” to textual corruptions or sometimes to the compiler’s “fictionalization,” the discrepancies also point to the nature of anecdotes as apocryphal stories with broad social origins rather than historical “facts.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1460

Luo

The entry on Su Ting’s success from child to adult also resonates with other anecdotes that reveal Zheng Chuhui’s persistent concerns about the efficacy of literary talent in officialdom. The presentation of a poem entitled “The Returning Swallow” by Zhang Jiuling 張九齡 (678–740) is said to have appeased the anger of Li Linfu 李林甫 (683–752), the most powerful chief minister during the second half of Xuanzong’s reign, by conveying Zhang’s unwillingness to compete for power with Li (Zalu, 25). In another case, Zhang Yue’s carefully timed delivery of a poem to Su Ting praising Su’s deceased father successfully reminds Su of his father’s friendship with Zhang, prompting him to help Zhang return from exile (Zalu, 28).49 The story of Li Linfu’s slander and murder of Li Shizhi 李適之 (ob. 746), however, shows Zheng Chuhui’s awareness of the limits of poetic power in the face of vehement political attacks as well as his efforts to achieve a sense of poetic justice. At the end of the account, Zheng Chuhui appends Li Shizhi’s two poems on the contentment found in drinking, one composed before and one after he was removed from the position of chief minister (Zalu, 16). Although these two poems do not seem to have any direct bearing on the political event recounted, their inclusion underscores Li Shizhi’s admirable spirit, seemingly untouched by political ups and downs.50 Showing sympathy for Li’s victimization, Zheng also offers a kind of poetic vindication by honoring Li’s poetic voice as a triumphant last note to his tragic life story. Consistent with his elevation of literary men, Zheng Chuhui is also highly critical of the unseemly behavior of powerful yet vulgar people at court. For example, Yang Guozhong 楊國忠 (ob. 756), the cousin of Yang Guifei, forces the chief examiner to pass his ill-educated son (Zalu, 13–14). Wang Zhun 王準, the son of one of Emperor Xuanzong’s favorite officials, is said to have bullied the emperor’s daughter and son-in-law (Zalu, 18). These stories show Zheng Chuhui’s clear disapproval of misconduct that disrupts the proper order of the official world. Moreover, Zheng tries to position officialdom in the greater scheme of the cosmos. He includes anecdotes in which the Daoist Xing Hepu 刑和璞 predicts the death of Fang Guan 房琯 (697–763), a key political figure 49  For discussion on how Tang stories represent poetic skills, see Graham M. Sanders, Words Well Put: Visions of Poetic Competence in the Chinese Tradition (Cambridge, Mass.: Harvard Univ. Asia Center, 2006), 203–78. 50  In a parallel version by Meng Qi 孟棨 (fl. 841–886), the poem composed after Li Shizhi’s demotion is given a more active role in the story: described as Li’s defiant gesture, it angers his enemy and leads to his death. Meng Qi, Benshi shi 本事詩 (Shanghai: Gudian wenxue chubanshe, 1957), 4.18. For translation and discussion, see Marc Nürnberger, Das “Ben shi shi” des Meng Qi (Wiesbaden: Harrossowitz Verlag, 2010), 211–12.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1461

during the transition from Emperor Xuanzong’s reign to that of his son (Zalu, 11), and another in which a certain Scholar Sun 孫生 foretells the highest official posts attained by Fang Guan and another man (Zalu, 33). These retrospective accounts of foresight reveal a characteristic fascination with the mysterious power of fate in determining one’s allotted life-span and official career, an interest that Zheng Chuhui shared with many other compilers of historical miscellanies. In contrast to Li Deyu’s collection, which is based on a privileged lineage of story transmission and focuses on Xuanzong’s life, Zheng Chuhui’s Minghuang zalu seems to have drawn from more extensive sources to present a broader picture of Xuanzong’s era. Preoccupied with literary figures in particular and the social and political life of the elite class in general, his version of mosaic memory reveals the perspectives and tastes of men of letters who remembered Kaiyuan-Tianbao in ways that affirmed their communal identity and interest.

Zheng Qi’s Kai Tian chuanxin ji: Amplification, Contradiction, and Humor

Zheng Qi’s collection not only continues and amplifies the thematic repertoire of his predecessors, but also includes contradictory and humorous variations. His construction of mosaic memory illustrates the mechanism of expansion and modification in the cumulative process of pastiche and overlay. Although the identity of Zheng Qi is disputed,51 his collection is characterized by a striking overlap with those of his predecessors in terms of thematic emphases. For example, in line with Li Deyu’s stress on the intimate relationship between Xuanzong and his chief ministers, Zheng Qi includes a new example in which the emperor adopts Yao Chong’s pragmatic recommendation to fight a plague of locusts. The intimacy between the two is embodied in the emperor’s unwavering trust in Yao and their alliance against the opposition expressed by the rest of the court (Chuanxin ji, 2). Meanwhile, Zheng Qi also inherits Zheng Chuhui’s stress on the special bond between the emperor and 51  Some believe that the compiler Zheng Qi 鄭棨 was the chief minister Zheng Qi 鄭綮 (ob. 899) because the two qi characters were interchangeable. Others disagree because the latter figure was not known to have held the post of Vice-Director in the Ministry of Personnel, the official title attached to the name of the compiler. For biographies of the chief minister, see Jiu Tang shu, 179.4662–63; Xin Tang shu, 183.5384–85. Eleven versions of the collection are extant. See Yuan Xingpei and Hou Zhongyi, Zhongguo wenyan xiaoshuo shumu, 76–77.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1462

Luo

gifted literary writers. In the anecdote in which a man submits a rhapsody ( fu) celebrating the royal trip to the hot-spring resort Huaqing Palace, the emperor immediately rewards the man with an official position (Chuanxin ji, 9–10). Zheng Qi’s reiteration of existing themes with new examples indicates the continuity in the public discourse on the Kaiyuan-Tianbao era. In addition to some new contents, such as anecdotes on Emperor Xuanzong’s trip to perform the fengshan 封禪 rituals on Mount Tai (Chuanxin ji, 2–3), Zheng Qi’s featuring of characters already seen in earlier collections creates thematic variations that add diversity and richness to the overlays of mosaic memory. Take his humorous entry on Su Ting, for instance: At first Su Gui didn’t appreciate Su Ting’s talent. He usually made Ting live in the stable and work with servants. One day, a guest visited Gui and waited for him in the hall. Going to the courtyard with a broom, Ting accidentally dropped a written document. The guest picked it up and read what turned out to be a poem on a Kunlun slave.52 Two of the lines were as follows: “His fingers are like ten ink-sticks—While his ears are like two spoons.” The guest was amazed. When Gui came out to entertain him, the guest recited the poem after cheerful conversation and, describing Ting’s appearance, asked, “Who is this? Could he be a descendant of your clan born of a concubine? If you adopt and raise him properly, he will surely become an outstanding son of the Su family.” From then on Gui became a little closer to Ting. It so happened that someone offered a rabbit to Gui and it was hung in the porch. Gui summoned Ting to compose a poem on it. Ting presented one in no time. It read as follows, “A rabbit, limp in death—was brought here and hung on a bamboo pole./ If we look at it against a bright mirror—how different is it from looking at [the rabbit in] the full moon?” Gui was so amazed that he immediately showed him respect and courtesy. As a result, Ting’s learning improved with each passing day and his literary compositions surpassed others of his generation. After Emperor Xuanzong had overcome the enemies of the state, streams of edicts came out overnight, all produced by Ting. He was called “the Little Duke of Xu” by later generations.53

52  “Kunlun” here refers to negrito peoples of the Malay peninsula. See Edward H. Schafer, The Golden Peaches of Samarkand: A Study of T’ang Exotics (Berkeley: Univ. of Calinfornia Press, 1963), 45–46. 53  Su Ting inherited his father’s ducal title and he was often referred to as the Lesser Duke of Xu, to be distinguished from his father.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1463

蘇瓌初未知頲, 常處頲於馬厩中, 與傭僕雜作 。 一日, 有客詣瓌, 候於廳所 。 頲 擁篲趍庭, 遺墜文書 。 客取視之, 乃詠崑崙奴詩也 。 其詞曰: “指頭十挺墨, 耳朶兩張匙 。” 客心異之, 而瓌出與客淹留 。 客笑語之餘, 因詠其詩, 并言形 貌, 問: “何人? 非足下宗族庶孽耶? 若加禮收舉, 必蘇氏之令子也 。” 瓌自是 稍稍親之 。 適有人獻瓌兔, 懸於廊廡間 。 瓌乃召頲詠之, 立呈詩曰: “兔子死 闌殫, 持來掛竹竿 。 試將明鏡照, 何異月中看 。” 瓌大驚奇, 驟加禮敬 。 頲 由是學問日新, 文章蓋代 。 及上平内難, 一夕間制詔絡繹, 無非頲出, 代稱小許 公也 。 (Chuanxin ji, 4)

This entry has discernible connections to the one in Zheng Chuhui’s collection cited above. On the one hand, the physical abuse that Su Ting endures from his father in the earlier entry is recast in social terms, with the father relegating the son to the status of a servant; on the other, the emperor’s recognition of Su Ting’s literary talent and subsequent promotion of him are replaced by similar actions by the father. Although both renditions demonstrate Su Ting’s literary talent by describing his poetic performance, they give the conventional trope of poetic ingenuity different twists. In Zheng Chuhui’s version, the poem that Su Ting allegedly composes after he just learns to talk is a sophisticated play with the formal and semantic aspects of written characters. His cleverness is underlined by his extemporaneous oral delivery that proves the spontaneity of his talent. By contrast, Zheng Qi’s entry is humorous with its foregrounding of the disparities between the poetic medium, forms, and content. Su Ting’s compositions, the first a dropped document and the second delivered on the spot upon his father’s request, are expressed formally in writing, yet his poetic lines are highly colloquial. While the lines fulfill the poetic rules of parallelism and tonal patterns, the images articulated, such as the analogies between fingers and ink-sticks and between ears and spoons, are so quotidian that they can be said to be a parody of high poetic elegance. This sense of parody is made even more apparent in the second poem: the dead rabbit hung on a bamboo pole in front of a mirror is staged to imitate the mythic white rabbit in the moon pounding the elixir of immortality, a hilarious contrast to its traditional poetic images of vitality, purity, and mysteriousness.54 The hilarity is pushed even further with Su Ting’s dramatic rise from the bottom of the social hierarchy to the top of it. Yet the anecdote is probably not meant to subvert the theme of literary talent or to mock Su Ting. His upward trajectory still reiterates the efficacy of poetic composition as a means of personal advancement. The ending of the 54  Allusions to the white rabbit (also as metaphor for the moon) are abundant in Tang poetry. For a few examples by Bai Juyi, see Quan Tang shi 全唐詩 (Beijing: Zhonghua shuju, 1979), 439.4893, 462.5254.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1464

Luo

entry with a brief summary of Su Ting’s important roles at Emperor Xuanzong’s court emphasizes Su’s positive image, which is in line with the representations of him as an outstanding writer and statesman in other historical miscellanies.55 In light of Su Ting’s celebrity status, Zheng Qi’s rendition in fact endears him to readers by offering a novel, humorous explanation of his early life. In addition to thematic variations, Zheng Qi also incorporates narrative variants that contradict earlier renditions. For instance, Zheng Chuhui uses the story of the child prodigy Liu Yan as a parallel example to that of the drunken Su Ting to corroborate the theme of the intimacy between the emperor and a gifted literary man: the emperor is said to have brought Liu Yan close to him and his favorite consort Yang Guifei pampers the child by setting him on her lap, making him up, and doing his hair (Zalu, 13). Zheng Qi retells the same storyline with a small but significant twist: though Liu Yan is said to be ugly in Zheng Chuhui’s version, he becomes a handsome boy in Zheng Qi’s rendition (Chuanxin ji, 2). Both versions focus on the intriguing combination and contrast of Liu Yan’s intellectual precocity, confirmed and rewarded by the emperor, and his physical childishness, highlighted by Yang Guifei’s treatment of him. Nevertheless, Zheng Chuhui’s version tries to reinforce the sense of contrast between the outer and the inner by adding another layer, Liu Yan’s ugly appearance and his talent. Zheng Qi’s version, by contrast, plays with subtle sexual innuendo by adding that numerous palace women gave him flowers and fruits as gifts. Though the gifts may be read as an innocent gesture to please a child, it unmistakably evokes the famous precedent of Pan Yue 潘岳 (247–300), who was said to have been so handsome that every time he went out, women fell in love with him and filled his cart with gifts of flowers and fruits.56 The contradictory accounts of an ugly and a handsome Liu Yan thus demonstrate how Zheng Qi’s “handsome” version plays with the sexual implications of the intimacy theme. Zheng Qi’s ready acceptance of inconsistency, contradiction, and humor shows his distinct personal touch in constructing his version of mosaic memory. His collection illustrates an important aspect of mosaic memory in terms of its power of accommodation. Mosaic memory allows for alternative and even contradictory accounts of the past without necessarily reconciling them, 55  For other examples, see Zhang Zhuo, Chaoye qianzai, 4.96–97; Liu Su, Da Tang xinyu, 1.9, 6.98; Li Jun, Songchuang zalu (Beijing: Zhonghua shuju, 1958), 1.6–7; Hu Qu 胡璩, Tanbin lu 譚賓錄 (Shanghai: Shanghai guji chubanshe, 1995), 6.2a/b. 56  The story is cited from the Yu lin 語林 and included in Liu Xiaobiao’s 劉孝標 annotations to the Shishuo xinyu. See Shishuo xinyu jiaojian 世說新語校箋, ed. Xu Zhen’e 徐震 堮 (Hong Kong: Zhonghua shuju, 1987), 335 (14/7).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1465

Remembering Kaiyuan and Tianbao

which constitutes a great fund of narrative resources for later re-tellers. The presence of humor shows another appeal of mosaic memory that distinguishes it from serious official history.57 Graham Sanders points out that poetic jokes are often the tool that people in lesser positions use to turn those in power into “fair game” for mockery through poetry.58 Apart from potential subversions, humorous anecdotes make possible light-hearted encounters with the past, embodying another form of intimacy. Humor not only humanizes the historical personages featured, but also reveals the most palpable presence of the bygone social community: although their laughter has long disappeared, their communal bond of humor still extends to latecomers, inviting future generations to join in the fun.

Wang Renyu’s Kaiyuan Tianbao yishi: Sensuality and Fantasy

Compiled after the end of the Tang, Wang Renyu’s collection indicates how a shift in historical horizon can affect an individual construction of mosaic memory. In addition to his continuation of the existing discourse and projection of personal interests, the prominent theme of sensuality in his collection shows a tendency to indulge in fantasizing about the past as the political relevance of the Kaiyuan-Tianbao era became more remote. Living through the demise of the Tang, Wang Renyu became a rather successful literary man under the subsequent regimes.59 According to his preface, traveling for an audience with Emperor Zhuangzong of the Latter Tang 後唐莊宗 (r. 923–926) allowed him to pass through the old Tang capital area in 925, inspiring his collection. The audience occurred shortly after the imperial army destroyed the Shu regime that Wang had served. Apparently aware of previous collections, he states that all his entries are “what the previous writings do not have” 皆前書所未有也 (Yishi, 9). He also points out that his collection “may not help to improve social customs and education but can still contribute to topics of conversation” 雖不助於風教, 亦可資於談柄 (Yishi, 9). Consistent with his self-positioning as a provider of conversational material, his entries in 57  As a matter of fact, Qing editors expunged Liu Su’s Da Tang xinyu from the division of the shi 史 because it includes humorous entries. Ji Yun et al., Siku quanshu zongmu, 140.1183. 58  Sanders, Words Well Put, 205–15. 59  For biographies of Wang Renyu, see Jiu Wudai shi 舊五代史 (Beijing: Zhonghua shuju, 1976), 128.1689–90; Xin Wudai shi 新五代史 (Beijing: Zhonghua shuju, 1974), 57.661–62. Ten versions of the collection are extant. See Yuan Xingpei and Hou Zhongyi, Zhongguo wenyan xiaoshuo shumu, 90.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1466

Luo

general are shorter and more fragmentary than those in previous collections. They often consist of a single anecdote rather than a cluster, bringing one’s reading experience even closer to that of overhearing social gossip. Wang Renyu’s continuation of his predecessors’ discourse can be seen in his inclusion of anecdotes that respond to earlier stories. For instance, in a story about Su Ting, Wang Renyu describes how he blows on the stove-fire in the stable to study at night while he works as a servant (Yishi, 46). Wang’s anecdote apparently tries to fill in a gap in Zheng Qi’s earlier story, that is, how Su Ting, miscast as a servant, could have managed to become educated on his own. Wang’s preference for this image of a self-made Su Ting may have had personal resonance. As someone who lost his father at a tender age, Wang is said to have revamped his life by beginning his learning in his late twenties; despite a late start, he embarked on a successful official career during the turbulent age of the Five Dynasties. The story of Su Ting’s rise from obscurity to eminence would have been a comforting affirmation of the promises held by self-determination and hard work. Carrying on his predecessors’ emphasis on the intimate relationship between Xuanzong and his officials, Wang Renyu also gives the theme a personal spin. Take the following anecdote as an example: Emperor Xuanzong was in a side-hall of the palace and wanted very much to discuss state affairs with Yao Chong. It was the fifteenth of the seventh month. It was raining continuously and the mud on the road was over a foot deep. The emperor ordered his attendants to carry his royal sedan-chair to summon Yao, who was a Hanlin scholar at that time. People inside and outside the palace saw it as a great honor. Among the emperors who were eager for and respectful of worthy gentlemen, there was none like this since ancient times. 明皇在便殿, 甚思姚元之論時務 。 七月十五日, 苦雨不止, 泥濘盈尺, 上令侍御 者擡步輦召學士來 。 時元之爲翰林學士, 中外榮之 。 自古急賢待士帝王, 如 此者未之有也 。 (Yishi, 11)

The emperor’s gesture of having Yao Chong carried into the palace in his own sedan chair reiterates the conventional theme of personal and political intimacy. The emphasis on Yao Chong’s title as a Hanlin scholar is not accidental, because Wang himself had held such a position. The Hanlin Academy was established in 738; scholars helped to draft imperial edicts and were selected for their literary talent, but did not seem to play a major role in politics during

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1467

Xuanzong’s reign. We should note that in Zheng Chuhui’s collection, the two anecdotes that feature edict writers Wei Kang and Xiao Song identify them as officials on night duty from the Secretariat rather than Hanlin scholars, while the one about Su Ting does not mention his title. During the post-rebellion era, however, because of their privileged proximity to the emperor, on call in an office right next to the inner palace, some scholars were able to develop a close relationship with the monarch and influence major political decisions, so much so that they acquired the nickname “inner chief minister” 内相.60 Given that Tang anecdotes on scholars almost invariably link them to emperors of the post-rebellion period and that Yao Chong’s biographies do not indicate that he held such a post, the twists in Wang’s version show a new shift in the intimacy theme, an emphasis that reveals his self-projection.61 Wang Renyu’s preference for stories about Hanlin scholars is consistent throughout his collection. He underlines the identity of Zhang Jiuling and Su Ting as scholars, both of whom were known as chief ministers in previous collections. Similarly, Wang also refers to Li Bai 李白 as a scholar, a reference which Fu Xuancong believes to be an error due to Li Bai’s service as a Hanlin gongfeng 翰林供奉, Academician-in-Attendance, which did not involve edict composition.62 Whether or not Wang Renyu was personally responsible for the discrepancies, the concentration of these anecdotes demonstrates Wang’s persistent promotion of the Hanlin scholar as the most visible post of Emperor Xuanzong’s court and the quintessential “worthy gentlemen.” Wang thus confutes Li Deyu’s exaltation of the chief minister and Zheng Chuhui’s stress on

60  The nickname was noted by Li Zhao in his “Hanlin zhi” 翰林志. See Hanxue sanshu 翰學 三書 (Shenyang: Liaoning jiaoyu chubanshe, 2003), 1–7. Also, F. A. Bischoff, La Forêt des Pinceaux: Étude sur l’Académie du Han-lin sous la dynastie des T’ang et traduction du Han lin tche (Paris: Presses Universitaires de France, 1963), and Mao Lei 毛蕾, Tangdai hanlin xueshi 唐代翰林學士 (Beijing: Shehui kexue wenxian chubanshe, 2000). 61  For example, Emperor Xuanzong 唐宣宗 (r. 846–859) is said to have ordered the use of his own golden lotus candles to accompany the scholar Linghu Tao 令狐綯 (fl. 830–859) back to his office after a lengthy conversation with him in the middle of the night. Pei Tingyu, Dongguan zouji (Beijing: Zhonghua shuju, 1994), 1.92. 62  Fu Xuancong 傅璇琮, “Tang Xuan Su liangchao hanlin xueshi kaolun” 唐玄肅兩朝 翰林 學士考論, Wenxue yichan 2000.4: 55–64. However, a seemingly anachronistic reference to Li Bai as Hanlin scholar may be an honorific tribute, as we can see in the tomb inscription composed by Fan Chuanzheng 范傳正 around 817. Fan Chuanzheng, “Tang zuo shiyi hanlin xueshi Li gong xin mubei bing xu” 唐左拾遺翰林學士李公新墓碑并序, in Li Taibai quanji 李太白全集 (rpt. Beijing: Zhonghua shuju, 1977), 31.1461–68.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1468

Luo

literary talent by turning the scholar into a new elite icon and the epitome of both political and literary genius. In addition, Wang Renyu’s version of mosaic memory is also preoccupied with a different kind of intimacy, one that seeks to relive the sensual details of the past. Although earlier references to the extravagant lives of Xuanzong and other members of the court, especially Yang Guifei’s family, are occasionally seen (Zalu, 29–30), Wang Renyu’s collection envisions sensual enjoyments as the defining feature of the entire bygone era. For example, we are told that the emperor enjoys sex-enhancing pills (Yishi, 21), sees himself and his consort as mandarin ducks in bed (Yishi, 40), and breaks flower branches to decorate her hair (Yishi, 21). This kind of strong voyeuristic attention to suggestive, erotic details concerning the Xuanzong-Yang Guifei affair is not seen in previous collections. Moreover, in addition to anecdotes on gestures of extravagance among aristocrats, such as the emperor’s brother Prince Shen 申王 having beautiful women press round him to keep him warm in winter (Yishi, 25) and the family of Yang Guozhong burning double-phoenix shaped charcoals (Yishi, 50), Wang also incorporates stories that portray the sumptuous life of women in the palace and people in the capital. The entertainment activities of palace women, such as listening to the sounds of crickets in golden cages in the fall (Yishi, 22) and playing on swings at the Cold Food Festival (Yishi, 41), paint a picture of peace, luxury, and idle enjoyment. Such a life of pleasure is not limited to the small court circle, as we can see in the following entry: During spring outings, men and women of Chang’an took walks in the outskirts of the city. When they ran into rare flowers, they would set up banquets on the grass and hang up red skirts as tents. Their extravagance and leisure were like this. 長安士女遊春野步, 遇名花則設席藉草, 以紅裙遞相插掛以為宴幄, 其奢逸如此 也 (Yishi, 49)

Consistent with such self-indulgent practices, Wang also includes a number of anecdotes on extraordinary objects. There are only a few such stories in previous collections: Zheng Chuhui mentions a jade dragon that Xuanzong receives from his grandmother Empress Wu 武后 (684–705) as a token of her belief in his later greatness (Zalu, 17), while Zheng Qi includes an entry on the unearthing of a precious white stone charm that predicts the length of the emperor’s reign (Chuanxin ji, 5–6). By contrast, Wang has more than ten such entries, such as a shining pearl that enables one to recall any memory (Yishi, 14), a wine cup that keeps the drink warm (Yishi, 15), and a pillow that is illuminated at

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1469

night (Yishi, 54).63 All these miraculous objects are ordinary articles imbued with wondrous powers.64 These anecdotes not only indicate Wang’s projection of Kaiyuan-Tianbao as an age of peace and prosperity but also his efforts to connect readers with the past through sensual imagining of its mundane details. Although stories of extravagance easily invite moralistic readings, Wang Renyu refrains from any explicit condemnation. Even when he occasionally comments “their extravagance and leisure were like this,” it is hard to tell exactly whether the tone is one of amazement, condemnation, or matter-offactness. The ambiguity is consistent with his vision of himself and his readers as “gossipers,” to whom such anecdotes provide occasions for stern upbraiding as much envy, admiration, disbelief, and vicarious enjoyment. Wang Renyu’s collection thus underscores a mode of intimate encounter with the past through a pastiche of sensual details. Reading the anecdote of an old female innkeeper at Mawei who makes a fortune by renting out a shred of Yang Guifei’s silk stocking for passers-by to fondle, Kroll points out that the story aptly captures the nature of sensuality in human efforts to connect with the past through textual or artefactual media.65 The violence and chaos of the end of the Tang and subsequent political instabilities might have intensified a desire to imagine what it was like to live in a world of peace and sensual pleasures. Whether or not these stories had already existed in the times of Wang’s predecessors, their appearance in his work suggests that he felt few constraints in incorporating suggestive and even erotic elements into accounts of Xuanzong’s court and his private life. Fantasizing about the Kaiyuan-Tianbao era in Wang’s compilation seems to coincide with the growing centrality of the Xuanzong-Yang Guifei affair in post-Tang narrative and dramatic literature. These works spanned the next millennium, from the Yang Guifei waizhuan 楊貴妃外傳 (The unofficial biography of Yang Guifei) by Yue Shi 樂史 (930– 1007) to the drama Changsheng dian 長生殿 (Palace of eternal life) by Hong 63  Pillows in the Tang were often made of porcelain, rare woods, jade, marble, or other precious stones, and even phosphorescent minerals. 64  Anecdotes of extraordinary objects can be seen in collections before Wang Renyu, often not associated with Xuanzong’s reign. Su E’s 蘇鶚 (jinshi 886) Duyang zabian 杜陽雜編, for example, lists extraordinary objects, animals, girls that each reigning emperor of midand late-Tang allegedly received as foreign tribute items; see Duyang zabian (rpt. Beijing: Zhonghua shuju, 1958), 15–60. Cf. also Duan Chengshi, Youyang zazu (Shanghai: Shangwu yinshuguan, 1937), 6.50–51, 10.73–81. For recent study and translation of Duan’s collection, see Carrie E. Reed, A Tang Miscellany: An Introduction to Youyang zazu (New York: Peter Lang, 2003); idem, Chinese Chronicles of the Strange: the “Nuogao ji” (New York: Peter Lang, 2001). 65  Kroll, “Nostalgia and History in Mid-Ninth-Century Verse”, 309–10.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1470

Luo

Sheng 洪昇 (1645–1704). In such later rewritings, the advance of historical horizons helps to explain the thinning of political relevance after the Tang, for Kaiyuan-Tianbao was no longer “ours” but “theirs,” a distant memory that could come out from under the burden of serious reflection to allow more melodramatic enjoyment.

Mosaic Memory as a Communal Discourse

The common interest of these writers in the Kaiyuan-Tianbao era demonstrates that they were part of a community set on inscribing certain values in cultural memory through the medium of narrative. How do we account for such continuous constructions of the cultural memory of Xuanzong’s reign? What were the implications of mosaic memory formation in the broader postrebellion context? The post-rebellion interest in Xuanzong had to do with the fascinating paradox he represented. His long reign produced great prosperity, but he also destroyed it, plunging the state into a turmoil from which the dynasty never fully recovered. This combination of brilliant success and catastrophic failure distinguishes him from Sui Yangdi 隋煬帝 (r. 604–618) who, as demonstrated by Arthur F. Wright, was stereotyped as the “bad last” ruler of a dynasty and later as a stock villain, or from Emperor Taizong, the great-grandfather of Xuanzong, who had been canonized in elite writing as a model of excellence.66 Mosaic memory substantiates a fundamental sense of historical judgment and fleshes out the details of the historical outline by offering diverse explanations of success and failure. For instance, Xuanzong’s tragic fall has been attributed to the would-be rebel general An Lushan’s ingenious deceptions of him, his refusal to heed Zhang Jiuling’s warning of An’s potential for betrayal, his misplaced trust in Li Linfu, his squandering of power, wealth, and attention on Yang Guifei and 66  Wright, “Sui Yang-Ti: Personality and Stereotype,” in idem, ed., The Confucian Persuasion (Stanford: Stanford Univ. Press, 1960), 47–76; also idem, “T’ang T’ai-tsung: The Man and the Persona,” in Essays on T’ang Society, ed. John Curtis Perry and Bardwell L. Smith (Leiden: Brill, 1976), 17–32. We should note that outside of the elite discourse, Taizong’s image is less flattering. In the “Tang Taizong ruming ji” 唐太宗人冥記 preserved among the Dunhuang manuscripts, for example, he is summoned to hell for killing his brothers and seizing the throne for himself; there he is manipulated by a clerk and has to bribe his way out. Wang Zhongmin 王重民 ed., Dunhuang bianwen ji 敦煌變文集 (Beijing: Renmin wenxue chubanshe, 1984), 209–15.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1471

her family, and so on.67 Sometimes, the compilers’ comments make the message explicit, such as Zheng Chuhui’s remark in the story of Xuanzong’s distancing himself from Zhang Jiuling: “Since Emperor Xuanzong had long been on the throne, he slackened a bit in his administrative duties” 玄宗既在位年 深, 稍怠庶政 (Zalu, 25). These anecdotes indicate the public’s understanding of the major power relations surrounding the monarch, revealing not historical truths but particular images, moods, points of view, and interpretations. They show that mosaic memory—like most collective memory—is less a representation of what the past was actually like than a wish for or imagination of what it might have been. In this sense, mosaic memory can be said to represent a history of the communal discourse. It is essentially a “kaleidoscopic history,” a peculiar kind of historical representation which, with its amalgamated fragments synedochic of presumed bygone historical totalities, captures snapshots of contemporary desires and anxieties as they are projected onto the past. Because these desires and anxieties reflect strong literati interest, mosaic memory reveals its nature as a discourse shaped largely by and for literati. By accounting for Xuanzong’s success and failure, the stories explore solutions to the post-rebellion crisis of sovereignty beset by threats from powerful military commissioners, eunuchs, and professional clerks. In other words, the stories have promoted figures of wen 文 (literature, culture, the civil) as the secret to a strong, prosperous sovereign rule. Attributing Xuanzong’s legendary success to a personal and political intimacy between the emperor and his civil officials— the chief minister in Li Deyu’s case, the literary genius in Zheng Chuhui’s, and the Hanlin scholar in Wang Renyu’s, these writers project themselves and their group onto the myth. It is highly symbolic that Xuanzong’s later failure has been tied to his rejection of Zhang Jiuling, a prominent man of letters who allegedly had foreseen the threat of An Lushan even when he was still a minor general. Meanwhile, none of the major power players at Xuanzong’s later court are presented as figures of wen: An Lushan is a barbarian general, Yang Guifei a beautiful, superficial woman, her cousin Yang Guozhong a vulgar upstart with little education, and Li Linfu a power-hungry man known for persecuting righteous scholar-officials such as the poet Li Shizhi. Taken together, the mosaic memory of Kaiyuan-Tianbao conveys a strong belief in the central role that wen and its representatives should play on the political stage.

67  For some examples, see Kai Tian chuanxin ji, 6; Da Tang xinyu, 1.12, 11.173; Minghuang zalu, 2.25–26, 2.28–30.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1472

Luo

Such a belief should be seen in the larger context of the post-rebellion efforts to revitalize wen as the key to solving sociopolitical and cultural problems of the time. Peter Bol has charted intellectual transformations from the early Tang to the Song, in which mid- and late-Tang was a critical transitional era with the emergence of guwen 古文 as an intellectual and literary trend that envisioned a unity between one’s literary writing, moral cultivation, and the power to transform society.68 Anthony DeBlasi expands our understanding of the picture by showing that the guwen values were marginal at the time and that mainstream scholars had promoted an alternative model of completeness that sought to accommodate diverse styles and opposing values.69 In parallel with these divergent theoretical discussions on the efficacy of wen, the high prestige of the jinshi (becoming essentially a literary examination in the postrebellion period) also indicates contemporary enthusiasms about wen as a practical life career.70 The mosaic memory of Kaiyuan-Tianbao reveals how the literati community envisioned the unfolding of such a career on the political stage and the ideal relationship with the monarch that came with it. From Li Deyu’s chief ministers to Zheng Chuhui’s literary talents and Wang Renyu’s Hanlin scholars, we can also glimpse a growing emphasis on the seamless fusion between the literary and political potency of wen as the ultimate goal of personal achievement for literati in the political realm. Bol has explained why early Song emperors decided to adopt civil-service examinations as the predominant mode of official recruitment, choosing the shi 士 as the representative of civil values to institutionalize imperial rule.71 The post-rebellion constructions of mosaic memory shed light on the perspectives of the shi themselves, who had optimistically envisioned their own empowerment before their political ascendency in the Song. The optimism is nowhere clearer than Wang Renyu’s story of Xuanzong bringing the Hanlin scholar Yao Chong to the palace in the royal sedan-chair: as an agent of intellectual authority, the scholar becomes a co-occupant of the seat of sovereignty with the monarch. 68  Peter K.  Bol, “This Culture of Ours”: Intellectual Transitions in T’ang and Sung China (Stanford: Stanford Univ. Press, 1992). 69  Anthony DeBlasi, Reform in the Balance: The Defense of Literary Culture in Mid-Tang China (Albany: State Univ. of New York Press, 2002). 70  Moore, Rituals of Recruitment in Tang China; Fu Xuancong, Tangdai keju yu wenxue 唐代科舉與文學 (Xi’an: Shanxi renmin chubanshe, 1986); Wu Zongguo 吳宗國, Tangdai keju zhidu yanjiu 唐代科舉制度研究 (Beijing: Beijing daxue chubanshe, 2010); Yang Bo 楊波, Chang’an de chuntian: Tangdai keju yu jinshi shenghuo 長安的春天: 唐代科舉 與進 士生活 (Beijing: Zhonghua shuju, 2007). 71  Bol, “This Culture of Ours,” 48–53.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1473

Post-rebellion literati did not foresee the future critique of the Song as a watershed in the rise of Chinese autocracy:72 they were confident that they could extrapolate what was valuable from the brilliance and imperfections of Emperor Xuanzong’s reign and apply those lessons to their own contexts. Mosaic memory came to shape in fundamental ways how the KaiyuanTianbao era was remembered. Such anecdotal imaginings were already influential in the Tang. The story of Yang Guifei’s costly taste for lichees, for example, appears in Li Zhao’s Guoshi bu.73 A number of poets, including Du Fu 杜甫 (712–770), Zhang Hu 張祜 (ca. 785–ca. 849), Du Mu 杜牧 (803–852), Zhao Gu 趙嘏 (806–853), Xue Neng 薛能 (jinshi 846), Zheng Gu 鄭谷 (ca. 849–911), and Qian Xu 錢翊 (fl. 895–900), evoke this story in their poems to convey their sentiments about the extravagance of Xuanzong’s court and its subsequent tragedy, though they did not necessarily draw from Li Zhao’s collection.74 For Song and later readers, late medieval versions of mosaic memory became important textual links to the bygone era. For instance, compilers of the Xin Tang shu (comp. 1060) drew from these sources extensively to correct what they saw as the deficiencies of the Jiu Tang shu (comp. 945). Zhang Yue’s new biography incorporates an anecdote from Zheng Chuhui’s collection about Zhang delivering a poem to Su Ting that sang the praises of Su’s father, a move that is said to have influenced Su to help Zhang out of his exile.75 The story is included probably because it fits well with the major narrative of Zhang Yue’s life by offering a plausible explanation for his political comeback. Likewise, the new biography of Yao Chong, in Xin Tang shu, adds stories from Li Deyu’s collection about Emperor Xuanzong personally welcoming Yao and seeing him off, entrusting him with what came to be his monopoly over the promotion and demotion of low-ranking officials, and allowing him to excuse his sons’ misconduct.76 These additions make the legendary intimacy between the emperor and Yao part of the official story. In his Zizhi tongjian 資治通鑑 (comp. 1084), Sima Guang 司馬光 (1019–1086) also included from Zheng Chuhui’s collection a story of Wang Zhun bullying Emperor Xuanzong’s son-in-law in order to demonstrate 72  In contrast to early medieval aristocracy, the civil administrative leadership chosen through the examination system, the full institutionalization of which started with the Song, is believed to have lacked the power and status to curb autocratic rulers. Miyazaki Ichisada 宮崎市定, Tōyōteki no kinsei 東洋的の近世 (Tokyo: Kyuiku taimususha, 1950). 73  Li Zhao, Tang guoshi bu, 1.19. 74   Quan Tang shi, 230.2518, 511.5843, 521.5954, 550.6365, 559.6488, 674.7722, 712.8197. 75   Xin Tang shu, 125.4407. 76   Xin Tang shu, 124.4384, 124.4387–88.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1474

Luo

the outrageous arrogance of the family members of Wang Hong 王鉷 (ob. 752), one of the emperor’s trusted officials during the Tianbao era. These examples illustrate the extent to which mosaic memory as an unofficial public discourse had influenced later official historians’ perception of historical figures and the era. Moreover, the medley of pieces that constitute mosaic memory are also recycled and adapted in other related post-Tang writings. The Tang Yulin 唐語林 (Forest of tales from the Tang) by Wang Dang 王讜 (fl. 1086–1110), for example, was based on fifty earlier historical miscellanies, including collections devoted to the memory of Kaiyuan-Tianbao. Wang’s collection presents a distinct case of mosaic memory construction in that it consists of a pastiche of fragments gleaned exclusively from existing writings. Another conspicuous example is the vernacular novel Sui Tang yanyi 隋唐演義 by Chu Renhuo 褚人獲 (fl. 1675– 1695), which incorporates more than forty anecdotes from the four collections on the Kaiyuan-Tianbao era.77 Such uses also extend to prose narratives, drama, encyclopedias, gazetteers, poetic commentaries, and even rhyme dictionaries, among others.78 These variations demonstrate how mosaic memory as objectivized culture is subjected to re-fragmentation and other historical transformations, providing for new needs and forms of cultural memory construction.79 The mechanism of recycling in this second stage of mosaic memory deserves a separate study. The multifaceted dynamic of mosaic memory formation as seen in the late medieval collections on the Kaiyuan-Tianbao era shows the working of both individual and communal forces and the role of historical miscellanies as a distinct medium of cultural production and historical engagement. Although

77  For a discussion of Chu’s novel, see Robert E. Hegel, “Sui T’ang yen-i and the Aesthetics of the Seventeenth-Century Suchou Elite,” in Chinese Narrative: Critical and Theoretical Essays, ed. Andrew H. Plaks (Princeton: Princeton Univ. Press, 1977), 124–59. 78  Examples include the Yang Guifei waizhuan by Yue Shi, the Changsheng dian by Hong Sheng, the monumental encyclopedia Taiping yulan 太平御覽 (comp. 983), the gazetteer Chang’an zhi 長安志 by Song Minqiu 宋敏求 (1019–1079), commentaries on Du Fu’s poetry by Qiu Zhao’ao 仇兆鼇 (1638–1717), and the rhyme dictionary Yunfu qunyu 韻府 群玉 (preface dated 1307) by Yin Jingxian 陰勁絃 and Yin Fuchun 陰復春. 79  Hegel argues that early Qing obsessions with the High Tang are “displacements of memory.” See Hegel, “Dreaming the Past: Memory and Continuity Beyond the Ming Fall,” in Trauma and Transcendence in Early Qing Literature, 345–71. In the same volume Zeitlin also discusses the adaptations of Tang images in Hong Sheng’s work; see her “Music and Performance in Hong Sheng’s Palace of Lasting Life,” 454–87.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Remembering Kaiyuan and Tianbao

1475

Tang writers did not invent the historical miscellany, the trend of continuous compilations during the Tang made this kaleidoscopic mode of representing the past an established cultural convention. Later writers continued to compile historical miscellanies up to the twentieth century, contributing to a rich tradition and a vital area in traditional Chinese culture that has yet to be fully examined.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T’ang Paul W. Kroll Sometime during the late summer or early autumn of ad 729, Chang Yüeh

張說 (667–731) submitted to the throne, under his name and that of his fellow statesman Sung Ching 宋璟 (663–737), a memorial requesting that the

imperial birthday—the fifth day of the eighth lunar month—be declared a national holiday.1 The memorial goes into some detail regarding various suggested activities thought suitable to the occasion, and one has the unmistakable Source: “The Dancing Horses of T’ang,” T’oung Pao 67 (1981): 240–68. Abbreviations Used in Notes: bss Basic Sinological Series (Kuo-hsüeh chi-pen ts’ung-shu 國學基本叢書). ChTS Chiu T’ang shu 舊唐書 (Peking, 1975). cts Ch’üan T’ang shih 全唐詩 (Peking, 1960). ctw Ch’üan T’ang wen 全唐文 (Taipei, 1979). hts Hsin T’ang shu 新唐書 (Peking, 1975). ssccs Shih-san-ching chu-shu 十三經注疏 (1871 ed.). tctc Tzu-chih t’ung-chien 資治通鑑 (Hong Kong, 1976). thy T’ang hui-yao 唐會要 (Ts’ung-shu chi-ch’eng ed.). ttts T’ang-tai ts’ung-shu 唐代叢書 (1806 ed.). wh Wen hsüan 文選 (Hong Kong, 1936; bss ed.).

1  For the text of this document (“Ch’ing pa-yüeh wu-jih wei ch’ien-ch’iu chieh piao”), see ctw, 223.11a–12a. tctc, 213.6786, is mistaken—as the text of the memorial itself clearly shows—in saying that the document was presented by Chang and Yüan Ch’ien-yao 源乾曜 (?–731). It is necessary to sort out here some bureaucratic shifts of office that took place in 729. Early in this year Chang Yüeh was restored to the post of Right Adjuvant Minister of the Magistery of Writing (shang-shu yu ch’eng-hsiang 尙書右丞相, in functional terms Under-Secretary of State), the very office that had been his when he was forced in 727 to resign from government service. At the time of Chang’s return to office in 729, Yüan Ch’ien-yao held the post of Left Adjuvant Minister (shang-shu tso ch’eng-hsiang 尙書左丞相, functionally Secretary of State) and was thus nominally Chang’s superior. But a few months later Yüan was “kicked upstairs” to the higher-ranking but less influential post of Junior Preceptor to the Grand Heir (t’ai-tzu shao-fu 太子少傅, functionally Under-Tutor to the Heir-Apparent); Chang then replaced him as tso ch’eng-hsiang, with Sung Ching being appointed yu ch’eng-hsiang. The three men received their promotions on the same day, and we have a royal poem composed in their honor by the emperor himself at a banquet laid to celebrate the occasion—“Tso ch’eng-hsiang Yüeh, yu ch’eng-hsiang Ching, t’ai-tzu shao-fu Ch’ien-yao t’ung-jih shang kuan, ming yen Tung-t’ang, hsi shih,” cts, 3.38.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_045

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1477

impression that Chang Yüeh’s proposal is founded firmly on what he knew to be the monarch’s own desires in this matter. To be designated the “ThousandAutumn Holiday” (ch’ien-ch’iu chieh 千秋節, thus proclaiming in the very name a wish for the sovereign’s longevity), the festival was to provide throughout the T’ang empire three days of leisure and rest from normal routines of work, during which time appropriate ceremonies honoring the emperor should be carried out at every village shrine and congratulatory gifts should be offered him by his contented subjects.2 Li Lung-chi 李隆基, posthumously known as Hsüan Tsung 玄宗 (reg. 712– 56), the sublime ruler to whom this homage was to be paid, approved the request of his ministers and ordered that the inaugural Thousand-Autumn celebration take place that fall.3 In so doing he established a practice that would be continued for nearly two centuries by his successors to the throne of T’ang, It should be noted that Chang’s biography in ChTS, 97.3056, along with the entry in Hsüan Tsung’s annals at ChTS, 8.193, is in error in stating that Chang’s post upon being recalled to office in 729 was that of tso ch’eng-hsiang (the biography in fact follows this statement immediately with the information that Chang was then soon promoted to tso ch’eng-hsiang, a redundancy that points up the obvious mistake). The correct sequence of events is given in Chang’s biography in hts, 125.4409. See also the pertinent information in Sung Ching’s biographies, ChTS, 96.3035 and hts, 124.4393, and Yüan Ch’ien-yao’s biography in ChTS, 98.3072, as well as the emperor’s poem mentioned earlier. Finally, for ad 731—instead of 730, given in most reference works—as the correct date of Chang Yüeh’s death, see Paul W. Kroll, “On the Date of Chang Yüeh’s Death,” Chinese Literature: Essays, Articles, Reviews 2.2 (July 1980), 264–65. 2  Chang’s memorial even specifies the kinds of offerings proper to different classes of society: from the ranks of courtiers, highest-quality wines; from the imperial relations, woven strapsashes to be attached to the backs of metallic mirrors. The lesser gentry and commoners were to exchange among themselves dew-catching pouches fashioned of silk. ctw, 223.12a. The emperor in turn bestowed on his courtiers metal-backed mirrors. Two poems by His Highness, one a lü-shih, one a quatrain, on the subject of the mirrors distributed on one occasion to his vassals are to be found in cts, 3.32 and 3.41; both poems are titled “Ch’ien-ch’iu chieh, hsi ch’ün-ch’en ching.” 3  In his proclamation mandating the new holiday (“Ta pai-liao ch’ing i pa-yüeh wu-jih wei ch’ien-ch’iu chieh shou-chao,” ctw, 30.5b–6a), the emperor notes that such a festival would also be appropriate as a seasonal thanksgiving, for it would take place at the time when “clear autumn is high and uplifted, and the hundred grains are just maturing.” Indeed, the Department of Rites (Li pu 禮部) petitioned the throne successfully the following year to incorporate with the birthday celebration the traditional agrarian festival known as ch’iu she 秋社, in which thank-offerings were made to the local tellurian deities for their aid in bringing to fullness the year’s crops. See the note appended to Hsüan Tsung’s poem “Ch’ien-ch’iu chieh yen,” cts, 3.38, and, especially, Ts’e-fu yüan-kuei 册府元龜 (Taipei, 1972), 2.7b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1478

Kroll

all but one of whom likewise declared their respective birthdays to be occasions of national jubilation.4 When Hsüan Tsung was resident in the western capital at Ch’ang-an for this annual event (as he was—for all but four years—from 729 to the end of his reign,5 the official festivities at court took place within the environs of the Palace of Ascendant Felicity (Hsing-ch’ing kung 興慶宮). This “palace” was not part of the great “palace city” at the extreme north of Ch’ang-an but a separate complex of halls and habitations making up the walled ward or “quarter” ( fang 坊) directly northeast of the metropolis’s Eastern Market (tung shih 東市). Hsüan Tsung had lived here from 701 till 712,6 when he assumed the throne and having formally designated the place one of his palaces continued thereafter to hold certain imperial ceremonies—usually those of a relatively personal nature7—in this precinct that was for him so full of fond memories of earlier years. In the southwest corner of the compound were two multi-storeyed structures, called the Loft of Zealous Administration (Chin-cheng lou 勤政樓)8 and the Loft of Blossom and Calyx (Hua-o lou 花萼樓).9 and it was on their grounds that the spectacular entertainment pageants, which were the most

4  The lone exception was Li Kua 李适, posthumously Te Tsung 德宗 (reg. 779–805), who was satisfied merely with a toast of long life offered by his assembled courtiers. hts, 22.477. 5  In the eighth month of 732, 734, 735, and 736 the emperor was holding court in the eastern capital at Lo-yang. 6  ChTS, 8.165; see also thy, 30.558. 7  Like, for instance, the rites dedicated to the marvelous Five Dragons, which had a particular fascination for Hsüan Tsung and which he instituted in 714. See Robert des Rotours, “Le culte des cinq dragons sous la dynastie des T’ang (618–907),” in Mélanges de Sinologie offerts à Monsieur Paul Demiéville (Paris, 1966), i, 264–65. 8  The formal—but rarely used—name of this building was Loft of Zealous Administration in Attention to Fundamentals (Chin-cheng wu-pen lou 勤政務本樓). It had been built upon imperial command in 720. See Hsü Sung 徐松 (1781–1848), T’ang liang-ching ch’eng-fang k’ao 唐兩京城坊考, 1.25 (Ts’ung-shu chi-ch’eng ed.). 9  Its full name was Loft of Blossom and Calyx in Mutual Resplendence (Hua-o hsiang-hui lou 花萼相輝樓), so called because from it the emperor could see the abodes of his princely brothers (who still resided in this quarter) strung in sequence close by, like the leaves of the calyx round a flower. Ibid. In 736 Hsüan Tsung had this building and its grounds expanded into the northeast section of the Eastern Market and the northwest section of the Tao-cheng 道政 quarter (directly south of the Hsing-ch’ing kung), necessitating a rather large-scale demolition and restructuring of the walls of the three wards. thy, 30.559; see also ibid. For detailed drawings of the layout of the entire Hsing-ch’ing kung, see Hiraoka Takeo 平岡武夫, Chōan to Rakuyō 長安と洛陽 (Kyoto, 1956), maps 23 and 24.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1479

notable events of the capital celebration of the Thousand-Autumn festival— and which were open to public view10—were annually held. These extravaganzas lasted nearly all night and included presentations of both courtly and secular music by various orchestras, processions of military orders and regalia, dances and songs from court-sponsored ensembles. But the most remarkable performances were the animal diversions—dancing (or at least shuffling) elephants and rhinoceroses11 and a troupe of one hundred dancing horses.12 The synchronized caperings of this last group, the dancing horses, were undeniably the most enchanting and memorable event of all. These marvelous steeds excited the imaginations and the pens of many contemporary writers, who saw in them something more than the corporal substance of mere trained beasts. But to begin to appreciate properly what these horses actually represented, romantically and symbolically, to the men of T’ang, we must examine the literary works, particularly those of poets, written in praise of them, and inquire also into their mythical and historical precursors. It is to this task that the present study, a modest exercise in cultural reanimation, is devoted. Certainly the best-known, and the fullest, notice of the dancing horses is that contained in the mid-ninth century Ming Huang tsa-lu 明皇雜錄, an interesting miscellany of tidbits relating to Hsüan Tsung’s glorious era, compiled by Cheng Ch’u-hui 鄭處誨 (chin-shih 834). Cheng’s Chiu T’ang shu biography remarks that this work was compiled by him during a turn as collator (chiaoshu lang 校書郎) in the imperial library, presumably from official documents 10   K’ai-T’ien ch’uan-hsin chi 開天傳信記, comp. Cheng Ch’i 鄭棨 (?–899), 2b–3a (ttts); also T’ang yü-lin 唐語林, comp. Wang Tang 王讜 (fl. 1110), (Taipei, 1979), 1.16. 11  For more on these talented behemoths, whose story is beyond the scope of this paper, see Edward H. Schafer, The Golden Peaches of Samarkand: A Study of T’ang Exotics (Berkeley, 1963), 82–83. 12   h ts, 22.477. The description of the holiday entertainments given in the official history derives ultimately from that in the mid-ninth century Ming Huang tsa-lu, 13a/b (ttts) (see below for comments on this work and its author). The Ming Huang tsa-lu account purports to describe the general program of amusements for all imperial fêtes, not only that of the Thousand-Autumn holiday. Although the text is slightly garbled, it mentions in addition to the sports noted above several other amusements, such as a parade of floats and displays of acrobatics and swordplay. The description of Hsüan Tsung’s fetes in tctc, 218.6993–94, is also based on this text. An especially astounding balancing act performed at the Thousand-Autumn Festival is described in Ming Huang tsa-lu, 14a/b. A good survey of all the entertainments is to be found in Harada Yoshito 原田淑人, “Senshūsetsu enrakukō” 千秋節宴樂考, in Shiratori hakushi kanreki kinen: Tōyōshi ronsō 白鳥博士還 曆記念, 東洋史論叢 (Tokyo, 1925), 833–43, with four plates.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1480

Kroll

that were at that time open to his scrutiny.13 The entry on the dancing horses is familiar to English readers, owing to Arthur Waley’s translation published three decades ago.14 But, for our present purposes, it deserves a more precise rendering. The following version, though rather less lyrical than Waley’s, is perhaps truer to the diction of the original text. Hsüan Tsung once decreed that four hundred hooves be trained to dance. [These hundred horses] were divided into companies of the Left and of the Right, and styled “So-and-so Favorite” or “Such-and-such Pride of the Household.” Occasionally there were also included excellent steeds that had been sent as tribute from beyond the border. His Highness had them taught and trained, and there was none but did not devote himself utterly to this wonderwork. Thence, it was decreed that the horses be caparisoned with patterned embroidery, haltered with gold and silver, and their manes and forelocks dressed out with assorted pearls and jades. Their tune, which was called “Music for the Upturned Cup” (傾盃樂) had several tens of choruses, to which they shook their heads and drummed their tails, moving this way and that in response to the rhythm. Then wood-plank platforms of three tiers were displayed. The horses were driven to the top of these, where they turned and twirled round as if in flight. Sometimes it was ordered that a doughty fellow lift one of the scaffolds, and the horse would [continue to] dance atop it.15 There were a number of musicians who stood to the left and right, before and behind; all were clothed in tunics of pale yellow, with patterned-jade belts, and all must be youths chosen for their handsome appearance and refined bearing. At every Thousand-Autumn Festival, beneath the Loft of Zealous Administration [the horses] danced by decree. 13   ChTS, 158.4169. While the Ssu-k’u editors point out one or two minor discrepancies in the Ming Huang tsa-lu, they are emphatic in approving its general reliability. Ho-yin ssu-k’u ch’üan-shu tsung-mu t’i-yao chi ssu-k’u wei-shou shu-mu chin-hui shu-mu 合印四庫全書 總目提要及四庫未收書目禁徵書目 (Taipei, 1971), iii, 2888–89. In any event, for the entry on the dancing horses quoted below there is ample corroboration, as we shall see, from contemporary (i.e., early to mid-eighth century) sources for all items. 14  Waley, The Real Tripitaka and Other Pieces (London, 1952), 181–83. 15  This conflicts with the statement in the hts, 22.477 description (which otherwise tallies with that in the Ming Huang tsa-lu and indeed seems to be based on it) that the horses “did not move” when their platforms were heaved up. The two accounts may be reconciled, though, if we imagine that the horses stood still when the scaffolds were lifted and then began their dance again.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1481

Subsequently, when His Highness graced Shu with his presence,16 the dancing horses were for their part dispersed to the human world. [An] Lu-shan, having often witnessed their dancing, coveted them at heart; because of this he had several sold [to him]17 in Fan-yang.18 Subsequently, they were in turn acquired by T’ien Ch’eng-ssu 田承嗣.19 He was ignorant of them (i.e., of their special talent). Confusing them with steeds of battle, he installed them in the outer stables. Unexpectedly one day, when the soldiers of his army were enjoying a sacrificial feast and music was struck up, the horses, unable to stop themselves, began to dance. The servants and lackeys considered them bewitched and took brooms in hand to strike them. The horses thought that their dancing was out of step with the rhythm and, stooping and rearing, nodding and straining, they yet [tried to] realize their former choreography. The stablemaster hurried to report this grotesquerie, and Ch’eng-ssu ordered that the horses be flayed. The more fiercely this was done, the more precise became the horses’ dancing. But the whipping and flogging ever increased, till finally they fell dead in their stalls. On this occasion there were in fact some persons who knew these were the [emperor’s] dancing horses but, fearful of [Ch’engszu’s] wrath, they never ventured to speak.20 Besides providing us with a vivid description of the horses’ apparel and performance, and a perhaps too vivid account of their pathetic end, this report also indicates the degree of the emperor’s own interest in the animals. It was he who ordered the troupe to be formed, he who determined their livery, and he who commanded their appearance at his birthday gala each year. They were indeed His Highness’s hoofers and no one else’s. In a rhapsody by Ch’ien Ch’i 錢起 (chin-shih 751) eulogizing the dancing horses, the emperor is portrayed at the height of the birthday celebrations, bidding their performance: 16  I.e., when he fled the capital to escape the advancing rebel forces of An Lu-shan. 17  The text as quoted in T’ai-p’ing kuang-chi 太平廣記 (Taipei, 1976), 435.916c–d, instead reads that An Lu-shan had several of the horses “[re]settled” 置 in Fan-yang. This is the only variant in the T’ai-p’ing kuang-chi version. 18  Fan-yang 范陽 was part of Yu-chou 幽州, the northeastern province (corresponding roughly to present-day Hopei) that was An’s original base of military operations. 19  T’ien Ch’eng-szu (704–78) was one of An Lu-shan’s most powerful lieutenants and continued to cause problems for the central government even after the great rebellion had been quelled. For his biographies, see ChTS, 141.3837–40 and hts, 210.5923–26. 20   Ming Huang tsa-lu, 8a–9a (ttts).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1482

Kroll

Spake the Thearch: Let the Gerent Equerries set dancing my mottled-grays and whitefoots,21 That they may display and spread wide august jubilation, That they may disclose and bring forth glad exultation. In the space of an instant Golden drums make the proposal, jade pipes carry the message, And suddenly, oh, Startled from dragon’s crouch into swan’s flutter— As dipping tassels make lightning fall on their vermilion manes, Prancing polls make stars flow on their white foreheads!22 The final two lines of this excerpt provide us with images of exceptional sparkle—the shaken tassels strung on the horses’ halters flashing like lightning against their manes and the tossing of their heads causing their forehead blazes to remind one, in their movement, of the streaking of meteors. The tune to which these talented steeds performed—“The Upturned Cup”— was an old air of foreign origin. It had been composed near the end of the Chen-kuan 貞觀 (627–50) period by one P’ei Shen-fu 裴神符, a p’i-p’a virtuoso from Kashgar 疎勒 then in the service of the T’ang court.23 Although originally written for lute, the tune was readily transposed to other instruments: it is, for instance, listed among the airs for the popular “wether drum” (chieh-ku 羯鼓) so favored by Hsüan Tsung24 and was also converted later for the reed-pipe 21  “My mottled-grays and whitefoots” 我騏馵 calls up the line from Shih ching #128: “Yoked be my mottled-grays and whitefoots” 駕我騏馵. On the word ch’i 騏, see Bernhard Karlgren, “Glosses on the Kuo Feng Odes,” bmfea 14 (1942), p. 230, gl. 364. Chu 馵 specifically refers to a horse whose left rear foot only is white. Shuo-wen chieh-tzu chu 說文解字 注 (Taipei, 1970), 10A.5b. 22  “Ch’ien-ch’iu chieh Chin-cheng lou hsia kuan wu-ma fu,” ctw. 23   t hy, 33.610. See also Hsiang Ta 向達, T’ang-tai Ch’ang-an yü hsi-yü wen-ming 唐代長安與 西域文明 (Peking, 1933; Yenching hsüeh-pao chuan-hao chih erh 燕京學報專號之二). 57–58. hts, 21.471, records that Chang-sun Wu-chi 長孫无忌 (?–659), one of the high ministers of Li Shih-min 李世民 (pht. T’ai Tsung 太宗, reg. 626–49), was commanded by that emperor to compose the ch’ing-pei ch’ü 傾盃曲. This statement can be reconciled with the emphatic attribution to P’ei Shen-fu in thy (which is the earlier of the two texts) if we understand that Chang-sun Wu-chi was composing lyrics to the tune. However, it would not be surprising to find that, while the tune was actually composed by the foreign musician, it was credited by a xenophobic scribe to the influential Chang-sun Wu-chi who was known to have a personal interest in music, having directed the compilation of the “Monograph on Music” 音樂志 of the Sui shu 隋書). 24   Chieh-ku lu 羯鼓錄, comp. Nan Cho 南卓 (mid–9th c.), 10a (ttts). In addition to providing a catalogue, arranged by mode, of the 128 tunes in the repertoire of this little Kuchean

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1483

(lu-kuan 蘆管) by Li Ch’en 李忱 (pht. Hsüan Tsung 宣宗, reg. 846–59).25 “The Upturned Cup” was in the shang 商 mode and hence well suited to performance at an autumn-time festival, since the note shang of the pentatonic scale corresponded, in the traditional system of wu-hsing 五行 correspondences, with the season of autumn.26 From the texts at our disposal, it is clear that the ensemble that played this tune while the horses danced emphasized drums and, to a lesser extent, flutes.27 But the tune was not part of the great suites of music—the shih-pu chi 十部伎 and erh-pu chi 二部伎28—that contained the standard pieces for “official” court concerts; it was rather a separate divertissement. The dancing horses were quite clearly an act in themselves. Chang Yüeh, whose memorial instigated the Thousand-Autumn Festival, has left us a set of six “Lyrics for Dancing Horses” (舞馬詞) which seem to be the words for as many choruses of the horses’ tune, “The Upturned Cup.” Chang was an old hand at writing lyrics for court music: in 725 he had been ordered by drum, this text also provides much useful information on the instrument’s history at the T’ang court. Both Hsüan Tsung and his musically-oriented brother Li Hsien 李憲 (Prince Ning 寧王) were especially proficient performers on it. 25   Yüeh-fu tsa-lu 樂府雜錄, comp. Tuan An-chieh 段安節 (late 9th c.), 18a (ttts). E. D. Edwards, Chinese Prose Literature of the T’ang Period, ad 618–906 (London, 1937), I, 40 mistakenly gives the credit for this to Li Ch’en’s more famous ancestor Li Lung-chi, apparently being confused by the identical Mandarin pronunciations of their posthumous titles. 26  The tune survives, though probably not in its pristine form, in the present gagaku 雅樂 repertoire in Japan. There it is still predominantly in the shang mode but modulates at two points into the chüeh 角 mode. L. E. R. Picken, “T’ang Music and Musical Instruments,” tp 55(1969), 81. 27  See #2 of Chang Yüeh’s series of three poems in praise of the horses, #1 of his series of six lyrics to their tune, and the excerpt from Ch’ien Ch’i’s fu, all quoted herein. Drums and flutes were the dominant instruments in the Central Asian music that enjoyed such popularity at Hsüan Tsung’s court. It appears that the lithophone common to classical Chinese music also had a part in the horses’ accompaniment. See the fifth of Chang Yüeh’s six “Lyrics for Dancing Horses,” in which mention is made of “the striking of the stones” and “the percussing of metal.” “Wu-ma tz’u,” #5, cts, 890.10050. Three of these lyrics are quoted below. 28  As Kishibe Shigeo 岸邊成雄 has convincingly demonstrated, the significance of these two terms has been badly misunderstood by most historians. They do not refer to distinct divisions or radically different kinds of court music, but rather to multi-part suites of music, performed in fixed sequence by groups of varying size and instrumentation. For detailed explication, see Kishibe, Tōdai ongaku no rekishiteki kenkyū 唐代音樂の歷 史的研究 (Tokyo, 1961), ii, 184–437. Kishibe’s magnificent study is crucial to a proper understanding of T’ang music, music schools and establishments, performers, and court entertainments (virtually all of which required music) in general.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1484

Kroll

Hsüan Tsung to amend the lyrics of the stately ya-yüeh 雅樂 while the sovereign himself revised the music,29 and in the same year was responsible for arranging the sacrificial music and dances (and, indeed, nearly everything else) for the grand feng-shan 封禪 sacrifice carried out with high pomp by the emperor at holy Mount T’ai.30 Since Chang died early in February of 731, we can date his lyrics for the dancing horses to either their first or second ThousandAutumn performance, that is, either 729 or 730. I shall quote the first, third, and fourth of the set, as it is these three verses that speak of the steeds in most detail. All the lyrics are quatrains, in lines of six words. #1. The “myriad jewels” pay court at the levee of the phoenix screen; “Thousand-guilders” guide and lead the Dragon’s Mediators. In promenade prancing, vigor concentrated, [the steeds] glance aside at the drums, Pace to and fro in sportive shadows, paying heed to the song.31 In line one, the “myriad jewels” (wan yü 萬玉) are the multitude of brilliant courtiers, the “phoenix screen” (feng i 鳳扆) the embroidered screen that stood near the imperial throne. The “thousand-guilders” (ch’ien chin 千金) of the second line are the young scions of the well-to-do elite; the “Dragon’s Mediators” (lung mei 龍媒) they lead into the arena are the dancing horses themselves, here called by a name that was attached long ago to the Heavenly Horses of Ferghana sent to the Han court in 101 b.c. I shall have more to say presently about the symbolic identification of Hsüan Tsung’s steeds with these earlier equines. The closing couplet of the poem above describes the beginning of the horses’ dance, which is seen more closely in the two quatrains to follow. #3. With colorful tail-pennants the eight dancing rows form up into  columns; The five hues of the temporal dragons are adapted to the directions. 29   t hy, 32.595. The term used to refer to the lyrics of this dignified music is yüeh-chang 樂章, in contrast to tz’u 詞 which denotes the lyrics of secular tunes. The songs Chang Yüeh composed pursuant to this order may be found in cts, 85.920–23. 30   t hy, 32.595. For Chang’s songs on this occasion, see cts, 85.918–20. Chang’s rather overbearing attitude during the imperial progress to Mount T’ai and encampment there created much resentment among lesser officials, particularly of the military, and helped to hasten his fall from power over the following year and a half. ChTS, 97.3054; hts, 125.4408; also tctc, 212.6766–67. 31  “Wu-ma tz’u,” #1, cts, 890.10050.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1485

Bending their knees, they clench winecups in mouth, attending to the rhythm; Inclining their hearts, they offer up longevity never ending.32 The eight rows (pa i 八佾) in which the gaudily arrayed horses are here drawn up are conciously reminiscent of the eight rows of ceremonial dancers traditionally regarded as the express prerogative of the Chou kings:33 these testify to the ordained virtue of the T’ang ruler just as those did for the sovereigns of old. The “temporal dragons” (shih lung 時龍) appear in Pan Ku’s 班固 (32–92) “Rhapsody on the Eastern Metropolis” as dragon-horses whose coloring is matched with the symbolic hues of the five directions;34 we understand thus that the dancing steeds are caparisoned in complementary colors. As to the winecups clenched in the horses’ mouths, this is no poetic fancy of Chang Yüeh but a well-verified part of the performance. It was in fact the most notable aspect of the dance, mentioned often in other poems celebrating these steeds. A fine pictorial representation of this very scene has recently been recovered: a prancing horse, with winecup in mouth, is figured in gold relief on the side of a silver wine-flask dating to Hsüan Tsung’s reign, unearthed during excavations in 1970 at the site of the Prince of Pin’s 邠王 (?–741) dwelling in Ch’ang-an.35 (A photograph of it is reproduced as Plate I.) The horses are thus quite literally acting out “The Upturned Cup”! #4. The dragon colts of the imperial fold are well-grouped,  well-composed; The thoroughbred foals from the astral corral are unwonted and uncommon. In easy expanse of gambade and prance, they answer to the rhythm, Full of high spirits, treading each other’s prints, and never wavering.36 These are no trotting hacks, no ordinary caballine canterers, they are “dragon colts” (龍駒) that abide in the “astral corral” (星闌)—both metaphors emphasize the transcendent nature of the animals. And, if we are reading with our 32  “Wu-ma tz’u,” #3, cts, 890.10050. 33  See Lun-yü, 3.1. 34  “Tung tu fu” 東都賦, wh, 1.18. 35  “Hsi-an nan-chiao Ho-chia-ts’un fa-hsien T’ang-tai chiao-ts’ang wen-wu” 西安南郊何家 村發現唐代窖藏文物, Wen-wu 文物, 1972.1, p. 31 and color plate 3. 36  “Wu-ma tz’u,” #4, cts, 890.10050.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1486

Kroll

ears as well as our eyes, we shall note that the dynamic, visual harmony of the dancing horses is neatly echoed in both the sense and the sound of the rhyming and alliterative binoms, respectively, with which the poet predicates his first two lines—沛艾, m.c. p’aai-aai, “well-grouped, well-composed,” and 權奇, m.c. gywen-giĕ, “unwonted and uncommon.” Both of these phrases had earlier lives in classical literature: the first was used in fu by both Szu-ma Hsiang-ju 司馬相如 (179–117 bc) and Chang Heng 張衡 (ad 78–139) to describe the dragons37 or dragon-steeds38 that drew the Son of Heaven’s carriage, the second in a Han sacrificial song of 113 bc to laud a “Heavenly Horse” captured in Kansu and presented to the court.39 This symbolic identification of Hsüan Tsung’s dancing horses with supernatural or draconic steeds is further emphasized in the remaining pair of Chang Yüeh’s quatrains, where they are seen as modern avatars of Purple Swallow 紫燕,40 Glaucous Dragon 蒼龍,41 and the legendary dragon-horse of the Yellow River whose significantly spotted back suggested to the sage Fu Hsi 伏羲 the idea of the I ching’s eight fundamental trigrams.42 It was common in this way for T’ang writers to see in the performance of the dancing horses shadows of equine paragons celebrated in earlier days: the horses recalled to their audience an inevitable cluster of historical and legendary associations. In order fully to appreciate the literary works written in praise of these wonderful creatures, it is important for us now to become acquainted with some of their kindred antecedents and with some items of what might be called cultural hippology.

37  In Hsiang-ju’s “Ta-jen fu” 大人賦, quoted in Shih chi 史記 (Peking, 1972), 117.3057. 38  In Chang Heng’s “Tung ching fu” 東京賦, wh, 3.59. 39   Han shu 漢書 (Peking, 1975), 22.1060. The date noted here, viz. 元狩三年 (corresponding to 120 bc) should be corrected, following the annals of Wu Ti at Han shu, 6.184—which give a detailed account of the horse’s capture—to 元鼎四年 (113 bc). 40  “Wu-ma tz’u,” #5, cts, 890.10050. For “Purple Swallow,” a superior horse of legend, see especially Yen Yen-chih’s 顏延之 (384–456) “Che-pai ma fu” 赭白馬賦, wh, 14.288; also Shen Yüeh’s 沈約 (441–513) “San-yüeh san-jih shuai-erh ch’eng p’ien 三月三曰率爾 成篇, wh, 30.675. 41  “Wu-ma tz’u,” #5, cts, 890.10050. For “Glaucous Dragon,” the horse deemed ritually correct to pull the sovereign’s carriage in springtime, see Li chi chu-shu 禮記注疏, 14.13a/b (ssccs). 42  “Wu-ma tz’u,” #6, cts, 890.10050. On the dragon-horse of the Ho, see, among many other classical references, Shang shu chu-shu 尙書注疏, 17.23b (ssccs) and Li chi chu-shu, 22.27a–28b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1487

The “Heavenly Horses” of Han, twice mentioned above, are perhaps the best-known historical prototypes of all horses extraordinaire enjoyed by Chinese emperors. These well-studied steeds43 include equally those sent from Ferghana as tribute in 101 bc (for the acquisition of whom Han Wu Ti had been so avid) and a rare water-spawned horse entrapped at the Wo-wa 渥洼 River (near Tun-huang) twelve years previously. Two songs extolling these beasts were fashioned for use in Han Wu Ti’s “suburban sacrifices” (郊祀).44 The descriptions of the Heavenly Horses contained in these songs were plundered repeatedly by later writers to characterize subsequent but comparably marvelous steeds. We have already noted in Chang Yüeh’s lyrics for the dancing horses of Hsüan Tsung the phrase “unwonted and uncommon”45 and his designation of them as “the Dragon’s Mediators”,46 drawing his diction in both cases from these Han songs. Special attention should be paid to the latter expression. The Heavenly Horses were regarded in the sacred songs of Han as draconic agnates—they are sky-treaders and “mediators” or, in another verse, “friends” (yu 友) of the divine dragon itself. (Indeed, the horse born of the Wo-wa River was patently a dragon-horse.) The fundamental affinity of dragon and horse was well established in ancient and medieval China. It is very clearly illustrated, for instance, in the account in the Lieh-hsien chuan 列仙傳 of the legendary Ma-shih Huang 馬師皇 (Horse-Master Illustrious). This worthy was so celebrated an hippiatrist, invariably curing the steeds he treated, that ailing dragons likewise descended to avail themselves of his skill;47 obviously they recognized that a capable horse-doctor was also qualified to practice on the divine congeners of terrestrial steeds. For dragon and horse simply roamed in different mediums; each in 43  See Homer H. Dubs, “The Blood-Sweating Horses of Ferghana,” in his The History of the Former Han Dynasty, Vol. 2 (Baltimore, 1944), 132–35; Richard Edwards’ neglected but stimulating article “The Cave Reliefs at Ma Hao,” Artibus Asiae 17 (1954), esp. 13–25; and Arthur Waley, “The Heavenly Horses of Ferghana: A New View,” History Today 5.2 (1955). 95–103. 44  For the texts of the songs, written in three-word lines, see Han shu, 22.1060–61. Together they counted as the tenth in the group of nineteen sacrificial songs. See also Ch’üan Han shih 全漢詩, 1.7a (p. 57), in Ch’üan Han San-kuo Chin Nan-pei-ch’ao shih 全漢三國晉南 北朝詩 (Taipei, 1968). The songs have been translated by both Edwards and Waley, more literally by the former. 45  From a couplet in the first song which pictures the horses as “Able and careless of will;/ Unwonted, uncommon in essence.” 志俶儻, 精權奇. 46  From the second of the Han songs. 47   Lieh-hsien chuan, in Cheng-t’ung tao tsang 正統道藏 (Peking, 1977), hy 294, p. 2a. Ma-shih Huang eventually left this lower world, borne away by a grateful dragon patient.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1488

Kroll

his own sphere was supreme. As Ma Yüan 馬援(14 bc–ad 49), the great general and horseman of Han times, put it: “For moving through the heavens, nothing compares with the dragon; for moving over the earth, nothing compares with the horse”.48 It is no wonder, then, that exceptional steeds—like the dancing horses of T’ang—should invariably be endowed by their panegyrists with the sublime attributes of dragon nature. And from here it was an easy step to consider them, symbolically if not verily, dragon-horses. The earliest recorded dragon-horse was not the Wo-wa River prodigy of Han times but rather the one that emerged, bearing its spotted dorsal “chart,” from the great Ho in fabled antiquity. But all dragon-horses were deemed “humane” (jen 仁) creatures, appearing only in eras of rare universal concord. Here is an account of the discovery of one during the golden days of Hsüan Tsung’s reign in the T’ang: In the region between the sea and Tai 岱 (i.e., Mount T’ai) there came forth stones dusky and yellow. Some said, “If one consumes these, he may protract his life.” The illustrious thearch Hsüan Tsung decreed that the [Grand] Warden ([t’ai-]shou) of Lin-tzu 臨淄 should gather them each year and send them [to the court] as tribute. In the twenty-seventh year of “Opened Prime” (i.e., 739), Li Yung 李邕 (678–747) of Chiang-hsia was serving as Warden of Lin-tzu. In the autumn of this year he duly went into the mountains to gather the dusky and yellow stones, when suddenly he happened upon an old gaffer. Quite wonderful in appearance, the old man’s robust lineaments were hale and clear, and his moustaches and beard were exceptionally sleek. Of hodden were the clothes he wore. Emerging from off to the left of the road, he stayed Li Yung’s horse and declared, “Surely you, my lord, are personally gathering medicaments that will serve to extend the years of our sagely ruler?” “It is so,” said Yung. “It were meet,” said the gaffer, “if our sagely ruler procured a dragon-horse. In that case, he would for a myriad years enjoy the state and that without the trouble of gathering medicaments.” “And where,” said Yung, “might a dragon-horse be found?” “One may be expected on the moors of Ch’i and Lu,” replied the gaffer; “if it be procured, that would be a very talisman of Greatest Tranquility. Even unicorn, phoenix, tortoise, or dragon are not worthy to be paired with such a 48   Hou Han shu 後漢書 (Peking, 1974), 24.840, in a memorial offering up to his sovereign a bronze effigy of a horse, cast by his artisans in Chiao-chih 交阯 (modern Hanoi) from a bronze drum that came into Ma’s possession during his conquest of what is now northern Vietnam.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1489

token.” Yung then ordered [the old man] to mount and ride behind him, but all at once the fellow vanished from sight. Yung marveled greatly at it. Turning around, he addressed his attendant followers, “Have we not encountered a divine person?” And he forthwith ordered his bailiff, one Wang Ch’ien-chen 王乾真, to seek for the dragon-horse in Ch’i and Lu. When the fifth month, in summer, of the twenty-ninth year of “Opened Prime” (741) arrived, Ch’ien-chen actually obtained the steed at the household of Ma Hui-en 馬會恩 of Pei-hai county 北海郡 (modern I-tu district, in central Shantung). Its coloring was a mixed blue-black and white; its two flanks had fish-scale plating; its mane and tail were like the tufted bristles of a dragon; its whinnying call had truly the tone of a hollow flute; and in one day it could gallop three hundred li. When Ch’ien-chen inquired from whence it had come, Hui-en said, “I had but a solitary mare. Upon bathing in the waters of the Tzu 淄, she conceived and delivered. Accordingly, I called this horse by the name ‘Dragonling’ ” [since it must have been sired by the resident dragon of the Tzu River]. Ch’ien-chen reported this immediately to Yung, who, quite delighted, memorialized the matter and presented the horse [to the sovereign]. His Highness was greatly pleased. He proclaimed that the Inner Stable should take especial care with the beast’s food and fodder. And he commanded painters to depict its form, to be published and revealed both within [the court] and without).49 Although the preceding narrative comes to us from a ninth-century text, Li Yung’s own memorial (mentioned in the story) is still extant to verify the acquisition of the dragon-horse and its presentation to the emperor.50 We must note here that dragon-horses did not always resemble in appearance the T’ang exemplar described above. In the monograph on “Talismans and Auspicious Tokens” (符瑞志) in the Sung shu 宋書, written near the beginning of the sixth century, the dragon-horse, identified as “the quintessence of river water,” is portrayed as follows: “Eight feet, five inches tall, it is long-necked and possesses 49   Hsüan-shih chih 宣室志, 1a–2a (ttts); also quoted in T’ai-p’ing kuang-chi, 435.915b. As noted by Edwards, Chinese Prose Literature, i, 113, the attribution of this work to Chang Wei 張謂 (chin-shih 743) should be corrected to read Chang Tu 張讀 (fl. 860). A more complete edition of the Hsüan-shih chih than that included in ttts has been reprinted in Pi-chi hsiao-shuo ta-kuan 筆記小說大觀 (Taipei, 1960), i, 1–33. 50  Li Yung, “Chin wen-ma piao,” ctw, 261.17b–18a. The presentation of this animal is also reported, briefly, in hts, 36.953.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1490

Kroll

wings, with dangling hair on its flanks”.51 But we may assume that individual specimens might properly exhibit slightly differing features from one another, if only to accord with ontogenetic necessity. When the mind of a T’ang dynasty writer, dazzled by the performance of Hsüan Tsung’s dancing horses, sought comparisons with similarly superlative steeds recorded in earlier literature, it would commonly recollect not only the Heavenly Horses of Han and the unpredictably recurrent dragon-horse, but would also summon up a veritable stableful of legendary and somewhat less familiar equines, such as Purple Swallow and Glaucous Dragon (whom we have already encountered). Chief among these other equine literary eidola that materialize most often in works on the dancing horses are three—by name Battened White, Flying Yellow, and Benign Patrician. Battened White (Tzu-pai 玆白) was reputed to have been sent as tribute in elder days to the Chou from the country of I-ch’ü 義渠 in Chinese Turkestan; Li Shan 李善 (?–689), the T’ang scholiast of the Wen hsüan, quotes an ancient text which says that “The one called Battened White resembles a horse, has serrated teeth, and feeds on tigers and leopards”.52 Flying Yellow (Fei-huang 飛黄) was a marvelous horse, come—like most of his kind—from the Northwest, slightly vulpine in appearance and with a horn protruding from his back; the Huai-nan tzu 淮南子 notes that, in epochs of ideal government, he is “submissive in the fold,” and it was said that any cavalier lucky enough to gain a ride on him was assured a thousand years of life.53 Benign Patrician (Chi-liang 吉良) was reported to be a white-bodied, vermilion-maned steed, with eyes of yellow gold; bred in boreal lands, he too invested his jockey with a millennium of existence.54 It was a 51   Sung shu (Peking, 1974), 28.802. 52  See Li Shan’s commentary to Wang Jung 王融 (467–93), “San-yüeh san-jih ch’ü-shui shih hsü i-shou,” wh, 46.1016. Li Shan is quoting from the Chi-chung Chou shu 汲冡周書, a pre-Ch’in text purportedly discovered along with numerous other early manuscripts ca. ad 280 in a grave-mound in Chi county, Honan. Wang Jung names “Battened White” in his composition in a long list of rare objects that fill the treasuries, stables, and parks of his emperor, Ch’i Wu Ti. 53   Huai-nan tzu chu 淮南子注 (Taipei, 1962), 6.95. Flying Yellow is also identified as “Mounting Yellow” (Ch’eng-huang 乘黃), as remarked in the annotation of Kao Yu 高誘; see also Shan-hai ching chien-shu 山海經箋疏 (Taipei, n.d.), 7.308. There is some disagreement about whether a ride on this beast conferred a mere millennium of life; some texts say two thousand or even three thousand years. It was believed by some that this was the animal that the Yellow Thearch mounted, bearing him away from this world when he transcended on high. 54   Shan-hai ching chien-shu, 12.360–61. His name is also given sometimes as Benign Measure (Chi-liang 吉量) or Benign Yellow (Chi-huang 吉黃), but in medieval texts it appears most often as Benign Patrician. It is likely that the names of all three of the horses Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1491

natural matter for men of T’ang to filiate the dancing horses with such fantastic steeds of the past as these.55 There is no suggestion, however, that these earlier wonders of equinity knew how to dance; in this, the T’ang representatives surpassed them all. We are, indeed, accustomed to think of the T’ang dancing horses as unique in the annals of Chinese court amusements. It may therefore come as something of a surprise to learn that they had had predecessors at various times in prior centuries. One could cite, for example, as our poets sometimes did, the horses called the Nine Tai 九代 that were supposed to have danced to music at the enthronement of the venerable sage-king Yü 禹;56 but these performers are concealed from our view in the mythic mists of pre-history. More visible, and better documented, is a deep bay horse from Ferghana, acquired by Ts’ao Chih 曹植 (192–232). This great writer, and unhappy prince, trained the horse to bow and to move in time with a cadence of drums. He then bestowed the animal in a burst of fraternal generosity on his brother Ts’ao P’i 曹不 (187–226), reigning emperor of the Wei dynasty.57 We meet with more dancing horses in 458, when an entire team of them was sent to the Sung court at Chien-k’ang 建康 from Shih-yin 拾寅, ruler of the T’u-yü-hun 吐谷渾 (reg. 452–81) and King of Ho-nan 河南王.58 These are the first true precursors of the T’ang troupe. The Sung monarch, Liu Chün 劉駿 (pht. Hsiao-wu Ti 孝武帝, reg. 453–64), ordered his servitors to compose rhapsodies in celebration of this unusual gift. One of these works is still preserved mentioned in the preceding sentences (which may originally have represented transliterations of foreign terms or dimidiations of archaic single phonemes) underwent some modification and semanticization to reach the forms in which we know them. 55  The names Flying Yellow and Benign Patrician were also attached to two of the six “pleasances” 閑 or parks for imperial horses at Ch’ang-an in T’ang times. Ta T’ang liu-tien 大唐 六典 (Tokyo, 1973), 11.25a; thy, 65.1126. 56   Shan-hai ching chien-shu, 7.299–300. As remarked in the commentaries to this passage, the chiu tai 九代 may not have referred originally to horses at all but rather to the music performed at the ceremony, being perhaps a miswriting of chiu ch’eng 九成. But to most medieval writers, especially those who wrote of the T’ang dancing horses, there was no doubt that the phrase signified a performing horse troupe of Yü’s. 57  Ts’ao Chih’s memorial of presentation is preserved in the seventh-century encyclopedia I-wen lei-chü. See I-wen lei-chü; fu so-yin, lei-shu shih chung 藝文類聚附索引類書十種 (Taipei, 1974), 93.1623. The text may also be found in T’ai-p’ing yü-lan 太平御覽 (Taipei, 1968), 894.6b; and Ch’üan San-kuo wen 全三國文, 15.4b, in Ch’üan shang-ku San-tai Ch’in Han San-kuo Liu-ch’ao wen 全上古三代秦漢三國六朝文 (1893 ed.). 58  The rulers of the T’u-yü-hun, whose dominions included much of western and north China, had from the early fifth century been designated “Kings of Ho-nan.” Thomas D. Carroll, Account of the T’u-yü-hun in the History of the Chin Dynasty (Berkeley, 1953; Chinese Dynastic Histories Translation, No. 4), p. 32, n. 85. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1492

Kroll

for us—“A Rhapsody on Dancing Horses” by Hsieh Chuang 謝莊 (421–66).59 It is filled with pious allusions to the far-flung imperial virtue that the proffered horses betoken, but it furnishes us no notable description of the animals’ performance. Hsieh Chuang was also enjoined to write a “song” (ko 歌) in honor of the horses, which was then set to music; but this lyric has not survived. There are two other reports in the Sung shu of dancing horses being sent to court from the “Western Regions” and the “[King of] Ho-nan” at the end of 459 and during 461, but these reports are probably misdated references to the very horses celebrated by Hsieh Chuang.60 Half a century afterward, in 505, the Liang court was favored with a similar troupe, again provided by the T’u-yü-hun.61 A protracted fu on these horses exists, in the hand of Chang Shuai 張率 (475–527).62 It is a splendid composition, but far too long to quote in full here. The steeds are described in Chang’s preface as “red dragon foals, possessed of rare aspect, agile-footed, able to make obeisance and excellent at the dance.” We cannot fail to remark at least part of Chang Shuai’s poetic depiction of their performance: Having tipped their heads in concert with the pitch-pipes, Next they stamp their feet to the flurry of the drums. They toss their dragon heads, Twist round their cervine torsos; Their glances [like] double mirrors, Their kickings [like] wild-ducks paired. Once they have reached the arena, they bow in courtly fashion; Attending timely to the tune, they move sedately or in haste. It was to be two hundred years before a Chinese emperor and his court were to look on such an exhibition again. It was not Hsüan Tsung’s dancing horses, however, that were the first such group recorded in T’ang times. A similar team had performed some years previously at the court of his uncle Li Hsien 李顯 (pht. Chung Tsung 中宗, reg. 59   Sung shu, 85.2175–76; also in Ch’üan Sung wen 全宋文, 34.7a–8a. 60   Sung shu, 6.125, 96.2373. It is impossible, with the evidence at our disposal, to tell which date—458, 459, or 461—is the correct one for the horses’ presentation. It is most unlikely that groups of dancing horses would have been sent three times in a space of only four years—familiarity breeds contempt, especially in exotic gifts. 61   Liang shu 梁書 (Peking, 1973), 33.475. 62   Liang shu, 33.475–78; also Ch’üan Liang wen 全梁文, 54.3a–4b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1493

705–10). Schafer has translated a lively extract from the Ching-lung wen-kuan chi 景龍文館記 which describes their routine.63 Here I should like to call attention to another description of this troupe, done in verse by Hsüeh Yao 薛曜.64 Hsüeh’s poem, translated in its entirety below, is an elaborate and masterly evocation which not only presents a vivid picture of these steeds (the remembrance of whom surely had a large effect in inspiring Hsüan Tsung to select and train a like troupe, when he ascended the throne) but also breathes fresh literary life into many traditional equine motifs and allusions established in former eras. The poem begins in heptasyllabic meter but modulates about halfway through to a mixed prosodic scheme employing lines in six, five, or three syllables as well (indentation will indicate these metrical shifts in the translation). It consists of seven stanzas of varying length, each of which reflects a change in rhyme; within each stanza the rhyme adheres strictly to one chosen tone (Hsüeh does not avail himself of the looser rhymes allowable in the binary p’ing-tse 平仄 classification of tones), and each of the four tones ot Middle Chinese is utilized as a rhyming tonic at least once. A commentary will follow the translation. Of stars’ essence and dragon’s seed, they skitter and dance side by side; The yellow gold of their paired eyes glows against their purple gloss. When one dawn chances to meet with an age ascendant in tranquility, Submissive in the fold, a chart in muzzle, these then serve thearch and king.

63  E. H. Schafer, The Golden Peaches of Samarkand, 67. 64  There is some biographical confusion surrounding this writer in various sources, which should here be cleared up. The note on Hsüeh Yao, preceding his five extant poems, in cts, 80.869, is largely in error. It is taken primarily from the twelfth-century T’ang shih chi-shih 唐詩記事 (Hong Kong, 1972), 13.189. In its statement that Hsüeh married the Princess Ch’eng-yang 城陽公主 (daughter of T’ang T’ai Tsung) and the remarks following about his supposed sons, one of whom wed the influential Princess T’ai-p’ing 太平公主, it has confounded Hsüeh Yao 薛曜 with Hsüeh Kuan 薛瓘. This latter did indeed marry Princess Ch’eng-yang but he died sometime during the years 670–74 (see hts, 83.3647– 48). The only reliable information we have regarding Hsüeh Yao, the author of the poem below, are the one-line biographies of him in ChTS, 73.2591, and hts, 98.3893, from which we learn that he was the son of Hsüeh Yüan-ch’ao 薛元超 (622–83) who had held ministerial status at court, that he took part in the compilation during the years 698–700 of the San-chiao chu-ying 三教珠英, and that he was politically attached to Chang I-chih 張易之, one of the notorious brothers Chang who were the favorites of “Empress” Wu from 699 to 704.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1494

Kroll

The consummate virtue of our Illustrious One burgeons out to the Six Confines; The vulgar, the majestic are in accord at this time, as the stone of Yü is struck. From yesteryear one hears of the Nine Tai who held so much renown; In our present day the hundredfold beasts come ahead to dance. By the Hooked Array and the Guarded Enceinte—banners and standards for reverence; From bell and gong, gourd and earthenware vessels—sounds to make the earth thunder. The racketing clamor of “Receiving the Clouds” startles the sun’s numen; The lilting tinkle of “Harmonious Dew” sets Heaven’s team-of-four in motion. Racing the dust, they are arrows in flight—like to unicorn or dragonet; Treading the sunlight and chasing the wind—indifferent to our vision and ken. Chewing and champing on the “pulling iron,” they are equally unwonted, uncommon; Cloaked and caparisoned in tooled blazonry, very sightly and stunning! Clinking trinkets of purple jade dangle over their jewelled stirrups; Striated halters of bice-blue silk gird their golden bridles. Following drum and song, they bolt away fulgurously; Mete with grelot and sword, they gallop tempestuously. Demeanor collected, their light step remains nimble; Mettle concentrated, their rapid gait does not waver. Their brilliance rivals the white sun descending, Their breath envelops the verdant haze lowering. Now they give way, limping, but in fact have not yet done— For, suspended in air, their pace quickens, and red dust flies! Startled ducks, soaring egrets, are not fit to be set beside them; Sublime phoenix, swirling simurgh, do not suffice for comparison. As asarum-scent hangs and cinnamon wafts—fragrance full and favoring— With long neighs, they sweat blood, floating utterly with the clouds. But not shrinking from bitter misery, they have taken the eastern road,

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1495

The Dancing Horses of T ’ ang

And all due to the hallowed hymnody heard here dawn to dusk. Amid Ch’ang-ho And nigh the Jade Terrace, They take in the warmth of graciousness, oh! producing a glorious hue. With jingling of simurgh-bells, Winged glide of carriages, Completing the countenance of our state, oh! they are boisterous ornaments. Crowding the clouds, they soar, oh! in the Son of Heaven’s courtyard; Shouldering the sun, they engage, oh! our affections limitlessly. Benign Patrician now is ridden, oh! and a thousand years are won; Divinity here is attained, oh! to the tenure of heaven and earth. The prognostic from the Great Change sayeth: “the longevity of South Mountain”; And careering, cavorting, they share in the glee of this season of sagely brilliance.65 The poem is a virtual compendium of caballine lore. Many of the images should, however, be familiar to us now. Thus the dancing horses are Heavenly creatures, akin to stars and dragons. Their eyes are colored yellow gold, like those of Benign Patrician (who will be mentioned more explicitly near the end of the work). In the present rare age of earthly amity, they are avatars of Flying Yellow—“submissive in the fold”—and of the chart-bearing dragon-horse of the Ho.66 “Our Illustrious One” (我皇), in the second stanza, is of course the emperor whose excellent moral power reaches everywhere, to the “Six Confines” 六宇 (the four cardinal directions, as well as upward and down), uniting all the people. The “stone of Yü” 虞石 that is “struck” (fu 拊) is the lithophone once played by K’uei 夔, minister of music to the sage-king Shun 舜 (who hailed from Yü 虞 and is often identified by that toponym); the tones produced by this stone were so compellingly harmonious that “the hundredfold beasts led one another to dance”.67 It is thus an image of the unanimity enjoyed under the 65  “Wu-ma p’ien” cts, 80.870. 66  Note especially Sung shu, 28.796: “When the Kingly One’s virtue reaches to the abysmal springs, the Ho gives up the Dragon Chart.” 67   Shang shu chu-shu, 2.29b. We also find the “stone of Yü” in the fifth of Yü Hsin’s 庾信 (513–81) series of “Lyrics for Taoist Adepts Pacing the Void” 道士步虚詞. There it is used to “convene the luminous Realized Ones,” i.e., the resplendent inhabitants of the Taoist heavens. Yü Tzu-shan chi chu 庾子山集注 (Taipei, 1959), 5.4b. The Ch’ing commentator

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1496

Kroll

current monarch. The horses called the Nine Tai who danced, as we have seen, at the accession of the Great Yü 禹, are now reincarnated in our present steeds; and it seems that a multitude of other beasts as well are moved to join in the dance, just as happened in Shun’s time. “Hooked Array” (鉤陳) refers celestially to the potent constellation that includes the North Star.68 It corresponded to the Rear Palace within the imperial compound on earth69 and was joined to the sovereign’s own “Guarded Enceinte” (周衛). This is where the ceremonial ensigns are ranged on this occasion, where the court musicians play and the horses perform. The melodies include the piece “Receiving the Clouds” (承雲), once enjoyed by the Yellow Thearch himself,70 and “Harmonious Dew” (調露) which stirs the steeds to movement. The horses are characterized here as “Heaven’s team-offour” (天駟), again stressing their sidereal essence, for there was a star-group by that name which was the especial astral emblem of horses, particularly divine steeds belonging to the Son of Heaven.71 With this, at the end of the third stanza, all the necessary scene-setting has now been accomplished by Hsüeh Yao, the music has commenced, and from here to the end of the poem our attention will be centered solely on the actual spectacle of the dancing horses themselves. The long fourth stanza depicts, in shifting rhythm, the swift movements and bright trappings of these hooved Terpsichores. They are “arrows in flight”— like the Heavenly Horse described by the Sung writer Yen Yen-chih72—keeping on Yü’s works, Ni Fan 倪璠 (fl. 1711), states his puzzlement over the meaning of “the stone of Yü,” but we may now supply the correct interpretation. 68  Edward H. Schafer, Pacing the Void: T’ang Approaches to the Stars (Berkeley, 1978), 45–47. 69   Chin shu 晉書 (Peking, 1974), 11.289. 70  See Lieh tzu 列子 (Taipei, 1972), 3.1a/b, Chang Chan’s 張湛 commentary. 71   T’ien-ssu 天駟 is an alternate name ot the lunar lodging “Chamber” (Fang 房). See Erh ya shu 爾雅疏, 6.9a (Ssu-pu ts’ung-k’an ed.). This was one of the seven lodgings in the sidereal “Palace” of the “Blue Dragon” (靑龍宮) which was associated in T’ang times with the residence of the emperor and referred specifically to the dragon horses kept in his stables. See Schafer, Pacing the Void, 76–77, 82. Note also the comment of Kuo P’u 郭璞 (276–324) quoted in the early T’ang encyclopedia I-wen lei-chü, 93.1622, that “the horse… is descended from T’ien-ssu.” 72  See Yen’s “T’ien-ma chuang,” quoted in I-wen lei-chü, 93.1623 (also Ch’üan Sung wen, 37.1b), where the Horse of Heaven is seen as “Outracing the waters—a startled duck;/ Overrunning the land—an arrow in flight.//Happening upon hills, he is as the wind;/ Encountering clouds, he becomes the lightning.” The phrasing of these lines affects Hsüeh Yao’s diction in at least two other places later in his poem—where the dancing horses “bolt away fulgurously (i.e., like lightning)” and where they are compared with “startled ducks.”

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1497

pace with the sunlight.73 Champing on their bits, they are “unwonted, uncommon”—the familiar epithet first applied to the Han Heavenly Horses. Their fleet dartings are coordinated with the sound of “drum and song” and also with the harness-bells and rhythmically waving swords (wan chien 丸劍) that add extra music and pageantry to their performance.74 As radiant as the sun they seem, and their panting exhalations shimmer in the air like green mist—this last a wonderful image. The next two stanzas revert to the steady heptasyllabic meter with which the piece began. The horses appear to break off their dance, but it is merely an interval andante before their supreme finale. For now, amid the heavy scents of asarum and cinnamon, they resume and, like creatures of the sky but surpassing in perfection even duck, egret, phoenix, and simurgh, they vault aloft, “floating utterly with the clouds.” They have truly become horses of heaven! They are seen to sweat blood, as did their Han dynasty precursors and, again like them, have taken the “eastern road” (the Han horses came from the West) to grace us with their presence, although the journey here from their real dwelling place is so very arduous.75 And it is the “hallowed hymnody” (hsiaoshao 蕭韶) of our lord’s court, identical to the sacred music of legendary Shun,76 which has drawn them here. In the closing stanza, again in a varied rhythm, we are swept up with the horses to the starry realms of Ch’ang-ho 閶闔 and Jade Terrace 玉臺, respectively the Gate of Heaven and the seat of Shang Ti 上帝, the Thearch on High, celestial regions which are but the counterparts up above of the emperor’s residence below.77 The horses are the vehicles to take His Highness there. And these steeds, the crowning adornments of the T’ang state, moreover forebode 73  The version of Hsüeh’s poem preserved in the Wen-yüan ying-hua 文苑英華 (Peking, 1966), 344.3a–4a, gives these opening two lines of the fourth stanza as containing only five syllables each: “Like to unicorn or dragonet, they tread the sunlight;/ Chasing the wind, they are indifferent to our vision and ken.” 74  For a similar reference to “grelot and sword,” on which Hsüeh Yao’s is probably based, see Pao Chao’s 鮑照 (405–66) “Rhapsody on Dancing Cranes,” “Wu-ho fu,” wh, 14.292. 75  In one of the Han sacrificial songs, we read “The Heavenly Horses have come,/ Having crossed the grassless places.// Passing over a thousand li,/ They have followed the eastern road!” In the other we find that they are “Wetted with red sweat,/ Lathered with flowing sienna.” Han shu, 22.1060. 76  The “hallowed hymnody” of Shun was reputed to have attracted phoenixes to come dance before him. Shang shu chu-shu, 4.16b; and Shih chi, 2.81. 77  Again echoing one of the Han songs: “The Heavenly Horses have come,/ Mediators of the dragon,// To wander to Ch’ang-ho,/ And surview the Jade Terrace.” Han shu, 22.1061. On the asterism Ch’ang-ho, see Huai-nan tzu chu, 1.3, and Schafer, Pacing the Void, 47.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1498

Kroll

the everlasting longevity of our ruler, whose far-reaching virtue and percipience (“sagely brilliance” 聖明) they here so marvelously attest.78 The horses of whom Hsüeh Yao sang and their more famous successors, the hundred dancing horses of Hsüan Tsung, were unquestionably the elite steeds of the empire, kept merely for the joy their cleverness afforded, not to be used as mounts for courtiers, bureaucrats, soldiers, or post-riders. During Hsüan Tsung’s first dozen years on the throne, from 712 to 725, he oversaw an increase in the number of state-owned horses from 240,000 to 430,000.79 The best horses of this stock were of course reserved for service at the capital. We know quite a bit about the care and feeding, the pastures, the branding, even the dental chronology and the common diseases of these “imperial” beasts.80 It is easy to imagine the even more particular attention that must have been accorded the monarch’s personal performing troupe. The cruel fate of that part of Hsüan Tsung’s team that later passed into the hands of the uncomprehending warlord T’ien Ch’eng-ssu (recounted above) is rendered even more pathetic, when we recall that severe penalties were mandated by the T’ang legal code for the misuse of any government steeds, much less those of the emperor himself.81

78  The “prognostic” from the “Great Change” 大易, i.e., the I ching 易經, is not included in the current edition of the text, but it was a common classical idiom connoting long life, especially that of the sovereign. See, for example, Shih ching #166. 79   t ctc, 212.6767. According to this source, the T’ang state possessed a mere three thousand steeds at its founding in 618, a century before. 80  See especially Ta T’ang liu-tien, 11.22b–27b; also thy, 72.1305. Part of this information was copied into hts, 17.1217–18, 1220; for a translation, see Robert des Rotours, Traité des fonctionnaires et traité de l’armée (Leiden, 1948), 218–26, 232–37. Some of it is also extracted in Schafer, The Golden Peaches of Samarkand, 66; but Schafer’s whole section on “Horses” (pp. 58–70) furnishes important information on the general topic, particularly on the Central Asian provenance of many of the state horses. A good survey of the important role of the horse in the T’ang military is Li Shu-t’ung 李樹桐, “T’ang-tai chih chün-shih yü ma” 唐代之軍事與馬, in Chung-kuo shih-hsüeh lun-wen hsüan-chi 中國史學論文選集, Series 2 (Taipei, 1977), 353–406. 81  I count at least nine articles pertaining to the misuse of government horses in the T’ang lü shu-i 唐律疏議 (Shanghai, 1939; bss ed.): 10.91, overloading a post-horse; 10.92, riding a post-horse off the highway or passing a post-station without changing horses; 10.92, making a personal present of a post-horse temporarily in one’s possession; 15.12, concealing private goods in a post-horse’s baggage; 15.14, riding an unbroken government horse; 15.14, willfully slaughtering a government horse; 19.63, stealing and then slaughtering a government horse; 25.25, fabricating credentials to use a post-horse; 25.25, unauthorized use of a post-horse. The most serious of these offenses—stealing and then slaughtering a state steed—called for a sentence of two and a half years at hard labor.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1499

Hsüan Tsung’s dancing horses are of course best remembered for their appearances at the Thousand-Autumn Festival. However, they did sometimes perform on occasions other than the emperor’s birthday. We know of one such occasion that took place in the ninth month of a year late in Hsüan Tsung’s reign, with female entertainers from the “Instruction Quarter” (chiao-fang 敎坊), the court establishment that housed and trained chanteuses of secular music,82 engaging in an equestrienne regalement. This event, which constituted the divertissement at a court banquet, was celebrated in two rhapsodies written at the time, one by Ching Kua 敬括,83 the other by Li Cho 李濯.84 It is clear from these compositions that the dancing horses played a large part in the amusement. Two other rhapsodies on Hsüan Tsung’s troupe have been preserved, in honor of a non-birthday performance given sometime between 739 and 742.85 The names of the authors of this pair of fu have not come down to us, and the works themselves are, alas, not particularly inspiring. A rather obvious but still easily overlooked fact should be remarked here. Since we know that dancing horses performed for Hsüan Tsung from at least 729 (the year of the first Thousand-Autumn Festival) to 756 (when some are reported to have been appropriated by An Lu-shan), and since horses seldom live beyond twenty years, there must continually, over a span of three decades, have been fresh steeds admitted to the troupe and newly trained, to fill the places of those that died or became too old to perform satisfactorily. This indicates a concern for what might be called the institutional survival of the team and for the perpetuation of the entertainment they provided. They were a living symbol of the glory of Hsüan Tsung’s court and dominion. 82  Kishibe, i, 46–78, gathers all the pertinent information on the chiao-fang. 83  “Chi-ch’iu ch’ao-yen, kuan nei-jen ma-chi fu,” ctw, 354.11b–12b. Ching, an interesting if somewhat unambitious fellow—a fact that stirred some of his more energetic colleagues to criticism of him (see ChTS, 115.3376)—lived through the great rebellion and died in 771. Among numerous other fu penned by him, one finds compositions on “The Loft of Blossom and Calyx,” and also on “The Eight Trigrams,” “The Divine Milfoil,” on “The Camphor-laurel (豫章) Tree,” “Rush-reeds” (蒲盧), “The Tree Lotus” (木蓮, or Manglietia), on “Contemplating Musical Instruments,” and, perhaps most curiously, on “The Spider.” 84  “Nei-jen ma-chi fu,” ctw, 536.18a/b. Li’s rhapsody and that of Ching Kua both make use of an identical rhyme-sequence. 85  “Wu-ma fu,” ctw, 961.13b–14b; “Wu-ma fu,” ctw, 961.14b–15a. Both use the same rhymepattern. The preface of the first of these works refers to the emperor as “The Sagely and Cultured, Divinely Militant Illustrious Thearch of the Opened Prime” 開元聖文神武 皇帝, the title he used between 739 and 742. On Hsüan Tsung’s various titles and the dates of their adoptions and modifications, see thy, 1.6.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1500

Kroll

No one has sung the praises of these splendid animals more beautifully than Chang Yüeh. Earlier I quoted three of his six lyrical quatrains for the horses. He also wrote a set of three heptasyllabic lü-shih in appreciation of them. No study of the dancing horses of T’ang would be complete without Chang’s triple encomium, written in tribute of their performance at the Thousand-Autumn Festival. #1. For the paragon born of the Heaven of Metal—a Holiday for One  Thousand Autumns! The jade ale is shared, circulated—a toast for a myriad years of life! We have given ear to “The Purple Bayard,” a song from the Archive of Music; How may it compare, though, with these prime steeds as they dance on their floriate mound? Curvetting in company, their conformation evokes transfigured dragons and fishes; In promenade prancing, their vigor concentrates a whole column of birds and beasts. Year upon year, in accord with tradition, on the day one pointed to the tree, Light and lightly, they come as companions, soaring on felicitous clouds.86 Some of Chang Yüeh’s diction in this poem may be obscure enough to require explication. The “paragon born of the Heaven of Metal” is of course Hsüan Tsung, born in autumn, the season ruled by the “activity” of metal. “The Purple Bayard” (Tzu liu 紫騮) is an old yüeh-fu (“Archive of Music”) title, to which songs on the subject of superlative horses had been composed from at least the early sixth century.87 But for Chang Yüeh such songs do not compare with the reality of the extraordinary horses now before his eyes (“prime steeds” paraphrases 騄驥, a compound alluding to Lu-erh 騄駬, one of the eight legendary chargers of King Mu of Chou 周穆王, and Chi 驥, a thoroughbred courser of antiquity who could gallop a thousand li in a day). As we know, the performance is taking place on the emperor’s birthday. But why should this be referred to as “the day one pointed to the tree”? To 86  “Wu-ma ch’ien-ch’iu wan-sui yüeh-fu tz’u san-shou,” #1, cts, 87.962. 87  For examples of poems on “The Purple Bayard,” see Sung-pen Yüeh-fu shih chi 宋本樂府 詩集 (Taipei, 1961), 24.3a–5b).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

1501

understand this locution, we must remember that the T’ang dynasts, surnamed Li 李, claimed descent from Lao Tzu, whose surname was reputed also to be Li. According to the fourth-century Shen-hsien chuan 神仙傳, Lao Tzu’s mother gave birth to him beneath a plum (li 李) tree; the infant Lao Tzu, who was able to speak immediately upon passing into this world, pointed to the tree and declared he would take its name as his surname.88 “The day one pointed to the tree” is thus Lao Tzu’s birthday, and since the emperor is Lao Tzu’s descendant and representative, it is understandable that his own birthday should be regarded as symbolically equivalent to that of his putative ancestor. Finally, the “felicitous clouds” (ch’ing yün 慶雲) of the poem’s concluding line, upon which the horses come soaring (like celestial steeds), are rare atmospheric vapors of five colors which signal an era of universal tranquility.89 The second poem of the set focuses on the horses’ dance: #2. The Illustrious Paragon’s perfect virtue is equal to that of Heaven, And Heavenly Horses come for this ceremony, from far west of the sea. Sedately striding with pasterns flexed, now they bow down on both knees; Though full of high mettle, they do not advance, but stamp with a thousand hooves. Hispid whiskers and straining manes steadily sweep in time; As drums grow riotous, the gambading bodies now rear abruptly upward! And then to the end of the festal song they have winecups clenched in their mouths, And loll their heads and flicker their tails, getting drunk as mud!90 This is a charming depiction that needs no commentary. If one had to select a single poem to commemorate Hsüan Tsung’s dancing horses, I think one could choose no better composition than this. The concluding poem of the set is in the nature of a final survey of the scene and the occasion:

88   Shen-hsien chuan, in Tao tsang ching-hua lu 道藏精華錄 (Shanghai, 1922), Vol. 11, 1.1b. 89   Sung shu, 29.836. 90  “Wu-ma ch’ien-ch’iu wan-sui yüeh-fu tz’u san-shou,” #2, cts, 87.962.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1502

Kroll

#3. Afar it is rumored, our Illuminate Sovereign loves unconventional  talent; And by jade whip and golden wings the Dragon’s Mediators have been led in. If it not be accounted Battened White that now is here amongst us, Surely that must be Flying Yellow come from above the sky! Their silhouettes play with blossoms of sunlight, in mutual glimmer and shimmer; Snorting breath frames cloudy hues, as they pace round to and fro. Let it not be said that beneath the pylons peach flowers are dancing— Instead, by the river, there are leaves of orchid now opening out!91 None of the allusions here—to the “Dragon’s Mediators,” “Battened White,” “Flying Yellow”—should pose a problem for us any more. In fact, we may even take some pleasure in recognizing in Chang Yüeh’s image of the horses’ snorting cloud-breaths a pale shadow of the misty green huffings of Hsüeh Yao’s steeds of two decades earlier. The poem’s final couplet is a last gentle reminder of the season in which this spectacle is being enjoyed—not spring, when peach blossoms come forth, but autumn, when the orchid unfolds, autumn when we and all things celebrate a holiday for the emperor. We end, then, where we began, at the Thousand-Autumn Festival. And, as Chang Yüeh first proposed the festival itself, it will be fitting to let these three poems of his stand as a final appreciation of the most outstanding performers of that celebration, the hundred dancing horses of Hsüan Tsung. No later ruler of the T’ang was to revel in such a sight as this. The dancing horses lived to men of after times—as they do for us—only as a fable, part of the noble epoch that was swept away by the great rebellion of 756.

91  “Wu-ma…,” #3, cts, 87.962.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

The Dancing Horses of T ’ ang

plate i

1503

Silver wine-flask from Hsüan Tsung’s reign, decorated with dancing horse in gold relief. From Wen-wu, 1972.1

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T’ang Times Edward H. Schafer 1 Introduction The place of origin of the art of hawking is unknown. It is generally supposed to have been somewhere in “Turkestan” or “Central Asia”. From this conjectural center the craft is thought to have spread eastward into China, southward into India, and westward through Persia into the Mediterranean world and Europe. But Mongolia, for instance, might, with equal probability, have been the region where the first hawk was tamed, especially it we rely on the intensity of the practice of falconry as a criterion. Or perhaps it was Manchuria.1 Archeological evidence, however, gives the palm for antiquity to the Assyrians of the eighth century BC.2 Although Aristotle, in his Historia Animalium, observed the practice of hawking in Thrace,3 the earliest references in Western Europe are no earlier than the fifth century AD,4 after which hunting with birds-of-prey grew in popularity in the West to a peak in the twelfth and thirteenth centuries, attaining its greatest prestige among the upper classes after the example of the Emperor Frederick of Hohenstaufen pointed the way, especially after the publication of that learned monarch’s masterpiece De arte venandi cum avibus, containing the newest information on Oriental methods. Since then, this true “sport of kings” has continued in high esteem almost until modern times, no less in the countries of the East than in Europe. Source: “Falconry in T’ang Times,” T’oung Pao 46 (1959): 293–338. 1  B. Laufer, in Chinese Pottery of the Han Dynasty (Leiden, 1909), p. 231, favors Turkestan, but Hans J. Epstein, in “The Origin and Earliest History of Falconry”, Isis, 34 (1943), 497–500, states his belief that, “only a wealth of leisure, great patience, sensitivity and ingenuity, not ordinarily shown with regard to animals by primitive people, will make a successful falconer”. Falconry is, he says, “…. definitely the product of an advanced civilization”. While recognizing as necessary the qualities he demands of a trainer of hawks, I cannot agree with Epstein that these would probably be absent among the nomadic peoples of Asia. 2  Epstein, “Origin”, p. 498. T. G. Pinches, in his “Antiquity of Eastern Falconry”, Journal of the Royal Asiatic Society, 1897, pp. 117–8, thinks the evidence from the Assyrian sculptures inconclusive, but still gives Assyria priority in the West on the basis of “omen tablets”. In my lay opinion, his evidence is inconclusive. 3  Epstein, “Origin”, p. 501. 4  Epstein, “Origin”, p. 505.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_046

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1505

As for the Far East, the literary evidence has heretofore pointed to the introduction of hawking to the Chinese during the Han period, more particularly during the first or second century of the Christian era.5 There is, however, a tale, seemingly well known in medieval China, about the “Cultivated King” (Wen Wang) of the State of Ch’u in the seventh century bc, which tells how that prince hunted at Yün-meng with a goshawk.6 But there is no contemporary evidence for the truth of the story. More convincing evidence points to the last half of the third century bc, that is, to the Ch’in Dynasty. The Sung encyclopaedia of quotations, T’ai p’ing yü lan, gives the following passage as from the Shih chi: “When Li Szu’s punishment was imminent, he called to mind how he had gone forth from the East Gate of Shang-ts’ai, with yellow dog on leash and gray hawk on arm, but this was now unattainable7”. This passage appears, with minor differences, in modern editions of the Shih chi (ch. 87), except that the words “and gray hawk on arm” (pi ts’ang ying 臂蒼鷹) are always omitted. The same story of the great statesman and his last words to his son, but including his recollection of his hunting hawk, again as from the Shih chi, is also quoted in the thirteenth century encyclopaedia Ho pi shih lei pel yao 合璧事類備要.8 This same version of the story of Li Szu’s execution was also current in T’ang times, as we may judge from the language of an anonymous “Rhapsody on the Gray Goshawk” (Ts’ang ying fu 蒼鷹賦). whicn says that. the hawk “… will not return with Li Szu to the gate of Shang-ts’ai9”. All of this points to the existence in T’ang and Sung times of a manuscript edition of the Shih chi, then recognized as authentic, but

5  Non-specialists have, however, not hesitated to give a gloriously remote date to the knowledge of falconry in China; for instance, George Dock, Jr., author of the article “Falconry” in the Encyclopaedia Britannica (1956 ed.), states: “It appears to have been known in China some 2,000 years B.C.” His informant must have been a very patriotic Chinese. Eduard Erkes notes a tradition that a Chou king of the 9/8 cent. bc kept eagles and ospreys in his gardens, but there is no evidence of their use in hunting. See his “Vogelzucht im alten China”, T’oung Pao, 37 (1943), 19. 6  In Yu ming lu 幽明錄, attributed to Liu I-ch’ing 劉義慶 of the fifth century AD (pp. 9a–b in the Lin-lang pi shih ts’ung shu 琳琅祕室叢書 edition), and in the K’ung shih chih kuai 孔氏志怪, a book of the Six Dynasties period, but of uncertain date and authorship. Epstein, “Origin”, p. 499, treats this story as apocryphal. 7  T’ai p’ing yü lan, 926, 2a. 8  Hsieh Wei-hsin 謝維新, Ho pi shih lei pei yao (pich chi 別集), 65, 1b. 9  Quoted in T’u shu chi ch’eng, “ying pu” (鷹部), i-wen, 1, 5b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1506

Schafer

different from the one on which later printed versions were based. This lost edition contained the three critical words.10 Falconry, then, was practised in China in the 3rd century bc. It is interesting that Shang-ts’ai, the home-town of Li Szu, was in the old domain of Ch’u. It may even be, then, that the early medieval tradition about the hawking of King Wen rested on a solid foundation, though it would be surprising to find the first Chinese falconers in a region so remote from the steppes, the supposed homeland of the art. The only reference to falconry in the Early Han period (last two pre-Christian centuries) which I have been able to find is an anecdote in the I lei chuan 異類傳,11 a book of indeterminate date, about the gift of a black goshawk from the “Western Regions” to Wu Ti, who reigned in the 2nd century bc. But here again, though the event seems probable, the source is of uncertain reliability. It is only with the second century of our era that we begin to get an abundance of reliable information about the popularity of hunting with goshawks, in conjunction with dogs, among the princes and aristocrats of China.12 The earliest pictorial representation of a tamed hunting hawk in China which can be dated with some certainty is in a bas-relief of the second century ad, found at Hsiao T’ang Shan in Shantung by Edouard Chavannes, who photographed and studied it.13 This bird, shown on the fist of the hunter, is plainly a goshawk, the favorite hunting bird of the Chinese. 10  I had imagined myself to be the discoverer of this lost passage, but have learned that the T’ai p’ing yü lan version was noticed by the eminent bibliographer Wang Chung-min, whose discovery was published by Dr. A. W. Hummel, in George Sarton, Introduction to the History of Science, 3 (Carnegie Institution of Washington, Baltimore, 1947), 234. 11  Quoted in Ko chih ching yüan 格致鏡源, 79, 20b. 12  See the biography of Yüan Shu 袁術 in Hou Han shu, 105, 0867d; biography of Liang Chi 梁冀 in Hou Han shu, 64, 0772b; story of the dowager empress Teng 鄧 in ‘Tung kuan Han chi 東觀漢記 (Ts’ung shu chi ch’eng ed.), 6, 47–8. These and other Han references have been discussed by other scholars, especially by W. Eberhard, Lokalkulturen im alten China, 1 (Suppl. to T’oung Pao, vol. 37 [1942]), 67, and Epstein, “Origin”, pp. 499–500. The Hsi ching tsa chi 西京雜紀, ordinarily regarded as an unreliable source, describes both goshawks and sparrowhawks with accuracy (Szu pu ts’ung k’an ed., 4, 6b). 13  E. Chavannes, La sculpture sur pierre en Chine au temps des deux dynasties Han (Paris, 1893), p. 77 and plate 38. Early representations of hunting hawks, some said to be as early as the third century bc, are reproduced in W. C. White, Tomb Tile Pictures of Ancient China (Toronto, 1939), p. 59, plates 28, 108, 109. These are discussed in Epstein, “Origins”, p. 500. Duyvendak, in his review of Tomb Tile Pictures (in T’oung Pao, 35 [1939], 373), states his belief that Bishop White’s tiles are not as early as alleged, and thinks that Chavannes’ example remains the earliest.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1507

In the four centuries intervening between the Han and the Tang, falconry continued popular in North China, probably even more so than before under the dominion of “Tatar” rulers. The lords of the Northern Ch’i state, as we shall see later, were particularly devoted to this form of hunting, and it is even reported that the last monarch of that dynasty bestowed a patent of enfeoffment on a favorite goshawk.14 But the history of hawking in the Far East during these centuries remains to be written. That work, when it is finished, should prove to be of special interest, in that tradition assigns the introduction of falconry to Japan by way of the Korean kingdom of Paikche to the year ad 355.15 2

The Hawks of T’ang

When, in the autumn of AD 608, all of the professional austringers of the realm—they are called “Masters of the Goshawk” (ying shih 鷹師) in Chinese— were ordered to assemble in Loyang, the Eastern Capital, more than ten thousand persons responded to the summons.16 This was only a few years before the founding of the great T’ang empire. It may reasonably be assumed that this number increased during the prosperous century between the reigns of T’ai Tsung and Hsüan Tsung, in part due to the example of these two illustrious princes, who were themselves enthusiastic falconers. We may also assume, though I have not yet found contemporary quotations to support the assumption, that the markets of the great cities of the north thronged with vendors of hunting hawks, many of them Turks and other foreigners from the northern frontier,17 cheek by jowl with Persian gem dealers, Annamcse vendors of aloeswood, and Hindus selling sacred relics. The birds offered for sale to the nobles and franklins of T’ang were often native to Korea, Manchuria, Mongolia and Turkestan, but many had been taken 14  According to Tung Szu-chang 董斯張, Kuang po wu chih 廣博物志, 44, 35a. This Ming encyclopaedia gives no source for this story, but it ordinarily quotes T’ang and Sung books. The hawk’s title was “Lord of Ling-hsiao-chün” (凌霄郡君). I find no chün of this name, and suspect that it was imaginary, created for the hawk. Ling-hsiao means “empyrean scandent”. 15   Nihon shoki; discussed in Epstein, “Origin”, p. 500. 16   Pei shih, 12, 2782d. 17  There was a “Goshawk Bazaar” (ying tien 鷹店) in K’ai-feng in the early part of Sung. There merchants (called k’o 客 “strangers; visitors”, i.e. non-natives) sold hawks and falcons. See Meng Yüan-lao 孟元老, Tung ching meng hua lu 東京夢華錄 (Ts’ung shu chi ch’eng ed.), 2, 41.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1508

Schafer

in the northernmost Chinese provinces, that is, in those now named Hopei, Shansi and Shensi. Particularly famous were the hawks and eagles which bred north of Tai 代 in Shansi, among them the “noble goshawks whose young are born with feathers entirely red”, mentioned by the poet Tu Fu,18 and the eagles whose white feathers were sent as annual “tribute” to the imperial court,19 Prime eagles and falcons were taken at Ling-chou 靈州 on the Yellow River, in what is now eastern Ning-hsia,20 sparrowhawks and raven-black falcons at Hua-chou 華州 in southeastern Shensi,21 and sparrowhawks at Chin-chou 晉州 on the Fen River in southern Shansi.22 Pei-t’ing 北廷, on the Dzungarian frontier, was a celebrated source of eagles, falcons and accipiters.23 To these regions, Tuan Ch’eng-shih 段成式, the author of the treatise on hawks whose translation is the important part of the present essay, adds the mountains of northwestern Hopei. In these arid, stony, and almost inaccessible places, the Chinese austringers took young hawks from their nests and trained them as eyases. Further south, they were captured with deeoys as passage hawks, during the fall migration.24 3

Hawks in Tang Literature

Both wild hawks and trained hawks are abundantly represented in Tang poetry. The great poets Tu Fu and Yüan Chen 元稹 in particular, to judge from their frequent descriptions of the raptors, were interested in falconry. But many lesser writers treated this subject in fine verses. I venture here a translation of a quatrain about a hunting hawk bound to her perch, forgotten by her master, as an example of these.

18  In the poem “Sung Li Chiao-shu erh shih liu yün shih” 送李校書二十六韻詩. 19   T’ang shu, 39, 3723b–c. 20   T’ang shu, 37, 3720c. 21   T’ang shu, 37, 3719d. 22   T’ang shu, 106, 3925c. 23  Wang Ming-ch’ing 王明淸, Hui chu ch’ien lu 揮麈前錄 (Ts’ung shu chi ch’eng ed.), 4, 136. This is a Sung source. 24  Li Shih-chen, Pen ts’ao kang mu, ch. 49. This Ming authority states only what is obvious. Tuan Ch’eng-shih makes it clear that haggards (hawks taken and trained as adults) were much preferred to eyases (hawks taken as fledglings) in China, a preference which agreees with the opinion of the rest of the world.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1509

The Lay of the Hungry Hawk by Chang Hsiao-p’iao She imagines the level plain afar where the hares are plump just now; She turns her honed bill a thousand times and shakes her feather coat; Just let her peek loose this knot in her silken cord— ! But unless she got the call of a man she would not dare to fly.25 Little need be said about the role of the hawk in metaphor and simile: the flashing eye, the lightning attack, the paralyzing blow, are as familiar in Chinese verse as in Western. So also is the image of the hawk as a ruthless predator, liable to sententious treatment in verse. One convention, however, is peculiarly Chinese: by tradition, the hawk, like the tiger, is a symbol of the West, and of Autumn, and allusions to this artificial association are common in T’ang literature. The underlying metaphysical assumptions need not concern us here; suffice it to note that the association was justified by the observation that autumn (therefore necessarily “West”, as spring is “East”) was equally the time when the young hawk was sufficiently grown to hunt for itself, and the time when the hawks left their boreal eyries to migrate southward over the plains of China. This conception is registered in the Hou Han shu: “At the onset of autumn, the hawks and falcons strike”.26 4

Hawks in Tang Art

Hunting hawks were favorite subjects for the painters of the Tang era.27 By virtue of rank, at least, the most eminent falcon painter of that age was Li 25  Chang Hsiao-p’iao 章孝標 (fl. AD 826), “Chi ying tz’u” 飢鷹詞 (in Ch’üan T’ang shih, han 8, ts’e 4, p. 4b). 26   Li ch’iu erh ying chun chi 立秋而鷹隼撃. Han shu, 27a, 0406a. 27  The earliest Chinese representations of hawks known to me are the Han bas-reliefs already mentioned. There must have been painted pictures of them made by the bird painters of the Six Dynasties period, but only one such artist is definitely known to have

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1510

Schafer

Yüan-ch’ang 李元昌, son of the first emperor of the dynasty and younger brother of Li Shih-min, canonized “Grand Ancestor” (T’ai tsung), a monarch devoted to the hawking art, as we shall see. Li Yüan-ch’ang painted goshawks and falcons in a style said to have been superior even to that of the illustrious brothers Yen Li-pen and Yen Li-te. His paintings, though still extant in the ninth century, when the Li tai ming hua chi, the book which describes his artistry, was written, were apparently lost belore Sung times.28 Later, during the reign of Hsüan Tsung, another patron of falconry and of painting, there were several eminent painters of hunting birds, including the masters Feng Shao-cheng 瑪紹正, Wei Wu-t’ien 韋无忝, and especially Chiang Chiao 姜皎, a favorite of the sovereign. This latter craftsman, ennobled as “Commonlord of Ch’u” (Ch’u kung 楚公), was fortunate enough to have his representation of a “horned goshawk” on the arm of a hunter celebrated in a poem by Tu Fu himself.29 Judging from the listings in early Sung catalogues of painters and their works, no falcon paintings done in T’ang times survived to influence the great Sung bird-painters. The earliest great depictor of raptorial birds whose works remained then was Kuo Ch’ien-hui 郭乾暉, a native of the tenth century state of Southern T’ang, who specialized in painting sparrowhawks. Among his pictures were “Sparrowhawklet Chasing Birds” (Chu ch’in yao-tzu 逐禽鷂子), and

specialized in the depiction of hawks. Characteristically, this was a prince of the Northern Ch’i, a dynasty whose sovereigns were passionate devotees of falconry, as Tuan Ch’engshih indicates in his anecdotes, translated below. This lord, Kao Hsiao-heng 高孝珩, was a talented artist, and is said to have painted a goshawk on the walls of an audience hall so realistically that “spectators thought that it might be real, while doves and sparrows did not dare to come near” (Chang Yen-yüan 張彥遠, Li tai ming hua chi 歷代名畫記, Ts’ung shu chi ch’eng ed., 8, 249; Pei Ch’i shu, 11, 2215a). Unfortunately no pre-Sung paintings of falcons and hawks survive. B. Laufer, in Chinese Pottery of the Han Dynasty, p. 233, refers to a picture of a falconer, said to be based on an illustrated Erh ya, in G. Pauthier, Chine (Pt, I, Pl. 11), and states rather carelessly that this “… may lay just claim to the honor of being the oldest graphic book-illustration of falconry in the world …”. He is certainly mistaken: the picture is patently modern, and Pauthier (pp. 55–6) is himself cautious about attributing an age to it. 28   Li tai ming hua chi, 9, 267–8. 29   Li tai ming hua chi, 9, 289–90, 295; Tu Fu, “Chiang Ch’u Kung hua chiao ying ko”, 姜楚 公畫角鷹歌, see E. von Zach’s translation of the poem in Tu Fu’s Gedichte (Harvard University Press, 1952), 1, 304. Other painters of hawks in the T’ang era were Pei Chün 貝俊, Li Shao 李韶, Wei Chin-sun 魏晉孫, K’uai Lien 蒯廉, and Po Min 白旻. See Li tai ming hua chi, 10, 311–12. For the “horned goshawk”, see below.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1511

“Sparrowhawklet on its Perch” (Chia shang yao-tzu 架上鷂子), but he also did goshawks, and at least one falcon.30 5

Royal Hawkers He has also a great multitude of eagles which are very well trained to hunt; for they take wolves and foxes and buck and roe-deer, hares and other small animals, and take plenty of them. But those which are trained to take wolves are very extremely large and of great strength, for you may know that there is no wolf so large as to escape before those eagles without being taken … And he takes with him quite ten thousand falconers riding, and carries quite five hundred gerfalcons, and peregrine falcons and saker falcons and other kinds of birds in very great abundance, for such creatures are infinite and good in his domains; and they also carry goshawks in great quantity to catch birds on rivers.31

Marco Polo’s famous description of the hunt of the Great Khan cannot be matched with any similar passage in medieval Chinese literature, but the hunts of the Tang emperors must have been no less magnificent than those of the Yüan overlords. But it was not all beer and skittles for these flamboyant monarchs. Royal lovers of the chase had always to cope with the chilly morals and dour adages of their advisors, steeped in Confucian tradition. Hawking, in short, was frivolous: He took no pleasure in reading books, His only joy was in hawks and horses.32 The first representative of the T’ang dynasty was hardly seated firmly on his throne, in the second deeade of the seventh century, when he was sternly admonished by one of his austere magistrates for his acceptance of the gift of a young sparrowhawk. “A depraved practice of the preceding generation,” said that counsellor, referring to Sui Yang Ti, for whom the myriad of austringers

30  Kuo Jo-hsü 郭若虛 (11th cent.), T’u hua chien wen chih 圖畫見聞誌 (Ts’ung shu chi ch’eng ed.), 2, 73; Hsüan ho hua p’u 宣和畫譜 (Ts’ung shu chi cheng ed.), 15, 409–142. 31  A. C. Moule and Paul Pelliot, Marco Polo; The Description of the World, 1 (London, 1938), 92–94. 32   Chin shu, 122, 1390b.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1512

Schafer

had assembled at Lo-yang. It is not recorded whether the High Forefather of the dynasty purged his mews as a result of this advice. His successor, the Grand Ancestor (T’ai Tsung), had a similar problem. This virile prince, the renowned conqueror of the Turks, was subjected to the formal censure of his minister Wei Cheng, who took pains to points out that while the first years of his lord’s reign had been unblemished by such giddy sports as hunting, it eventually transpired that “… since his resolution could not be stiffened, tribute in hawks and dogs came from as far as the Four Barbarians”.33 During these early years, the emperor stood in some awe of his mentor, like a schoolboy before a strict but revered teacher.34 The story is told how he was hunting one day with a favorite sparrowhawk, when he saw Lord Wei approaching, whereupon he hid the bird in the bosom of his garment. Wei Cheng, who had observed this act, was tactless and unimaginative enough to conceive that he might teach the monarch a lesson by stringing out a rambling discourse on political affairs, with the result that the hawk died of suffocation.35 This effort must have failed, since when Kao Tsung, the successor of T’ai Tsung, moved into the palace, he found the old hunting park well stocked with hawks and hounds. He gave them their freedom forthwith.36 Moreover, the new ruler, in the second year of his reign (AD 651), issued an edict terminating the annual tribute in hawks and falcons from the states of P’o-hai and Silla in the northeast.37 The royal sport experienced a revival with the accession of Hsüan Tsung early in the eighth century,38 but after the troubles of mid-century, the old splendors never returned to the court, and this kingly art had various fortunes thereafter, mostly, it seems, dismal. We can detect some traces of these changes in the historical records: in the closing years of the century, Te Tsung 德宗 33   T’ang shu, 103, 3919d (biography of Sun Fu-ch’ieh 孫伏伽). 34  T’ang shu, 97, 3907d (biography of Wei Cheng 魏徵). Compare T’ang shu, 51, 3753c, where it is stated that in the first years of T’ai Tsung’s reign, presents of horses, hounds and hawks along with certain other trifles, might not be offered to the young sovereign without a special permissive edict. 35   Tzu chih t’ung chien, 193, for AD 628. 36  T’ang shu, as quoted in T’ai p’ing yü lan, 926, 3b. This quotation, as well as the ones of p. 1513, n. 40 and 1517, n. 64, are presumably from the original T’ang shu, whose reconstructed text is now called the “old” T’ang shu (see R. des Rotours, La Traité des Examens, Paris 1932, 69–70). 37   T’ang shu, 3, 3638. 38  See, for instance, K’ai yüan t’ien pao i shih 開元天寳遺事) T’ang tai ts’ung shu ed.), 3, 68a, for the anecdote about the two princes who carried “a red goshawk from Koryo” and a “yellow falcon from the Northern Mountains” on the royal hunts. The sovereign named these peerless birds his “Cloud Bursters” (決雲兒). Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1513

released the royal hawks along with the royal actors;39 though Hsien Tsung 憲宗, reigning early in the ninth century, held annual winter hunts with hawks and hounds,40 he is reported to have returned a gift of twelve falcons from an imperial legate in Shantung,41 perhaps from motives of economy or ostensible incorruptibilitv, rather than puritanism; early in AD 823 his successor, Mu Tsung 穆宗, after an accident in a ball game, suddenly decreed the freedom of all hawks in the imperial mews and the burning of all imperial hunting gear.42 To sum up, the great periods of T’ang falconry, at least as far as imperial example and encouragement are concerned, were the first half of the seventh and the first half of the eighth centuries. During these good years, falcons of patrician breeding poured into the forbidden city in the trains of embassies from friendly and tributary states. Outstanding among these donors were the nations of Manchuria and Korea: the Mo-ho “Tatars” (靺鞨), Silla, and above all P’o-hai, straddling eastern Manchuria and northern Korea. This last nation sent at least six missions bearing hawks and falcons as royal gifts during the reign of Hsüan Tsung.43 We may imagine that the trade in less regal birds was enormous. Such presents to the throne, whether “goshawks, falcons, dogs, or leopards”,44 were formally received at the Hung-lu Office 鴻臚寺 and suitable gifts made in return.45 They were then turned over to the palace officers who governed the emperor’s mews and kennels.46 These latter were the “five quarters” (wu fang 五坊). One of them was for the hounds, while the names of the other four corresponded to the names of the four great classes of hunting birds 39  Chiu T’ang shu, 13, 3105b. 40   T’ang shu, as quoted in T’ai p’ing yü lan, 936, 4a; See p. 1512, n. 36, above. 41   Chiu T’ang shu, 15, 3110b. 42   Chiu T’ang shu, 16, 3116d. 43   Ts’e fu yüan kuei, chs. 971–2 (the old names of Manchurian peoples, such as Su-shen 肅愼 and Fu-yü 扶餘, appear even in the T’ang tribute lists); T’ang shu, 204, 4107a; Tou Kung 竇鞏, “Hsin-lo chin po ying” 新羅進白鷹, Ch’üan T’ang shih, han 4, ts’e 10, p. 23a; Su T’ing 蘇頲, “Preface to ‘Shuang po ying tsan’” 雙白鷹讃, Ch’üan T’ang wen, 256, 12b. 44  Hunting leopards, to judge from the context. 45   T’ang shu, 18, 3746a. 46  These officials were not much liked by the commons. The T’ang history reports the great resentment against the hordes of courtiers who descended upon the countryside like locusts during the annual autumn trials of the imperial hawks and hounds (an ying ch’üan 按鷹犬) held in anticipation of the winter hunts. See Chiu T’ang shu, 170, 3519c, for the early ninth century. For further discussion of this subject see S. Naitō 内藤雋轉, “Kōrai jidai no hōyō ni tsuite”, 高麗時代の放鷹について, Chōsen gakuhō, 8 (October, 1955), 65, 70 (note 1). This article sketches the history of the imperial mews through T’ang, Liao, and Chin, to the climax in Yüan, then their development in Koryo from the thirteenth century. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1514

Schafer

recognized by the medieval Chinese.47 These were eagles, falcons, sparrowhawks and goshawks, the same classes of raptors which have proved best for the chase everywhere in the world.48 6

Types of Hunting Hawks

1. Eagles (tiao 鵰). These handsome birds have been used and are still used for hunting large mammals, such as wolves, foxes and gazelles, by the peoples of Central Asia, including the Mongols and Turks.49 The T’ang emperors also used them. Probably the Golden Eagle (Aquila chrysaëtos) was the species most commonly trained, possibly also the Imperial Eagle (Aquila heliaca), 47   T’ang shu, 47, 3743a. Cf. Robert des Rotours, Traité des fonctionnaires et Traite de l’armée, 1 (Leiden, 1947), 222. 48  Thus A. von Le Coq, Von Land und Leuten in Ostturkistan (Leipzig, 1928), p. 80, observes that the four kinds of birds used by the hunters of Kucha in 1913 were “Sperber”, “Habicht”, “Edelfalke”, and “Steinadler”, that is, the sparrowhawk, goshawk, falcon, and eagle. Modern usage differs in many respects from medieval in regard to the Chinese names of hawks, and it has not always been easy to arrive at their meaning with certainty. Modern references, for instance, identify yao 鷂 variously as “harriers” (gen. Circus), “kites” (gen. Milvus), and “sparrowhawks” (gen. Accipiter). I have relied on the following modern authorities, in conjunction with medieval Chinese descriptions, to arrive at my identifications. Everyone interested in Far Eastern falconry is advised to consult them: Sten Bergman, Zur Kenntnis nordostasiatischer Vögel (Stockholm, 1935); Cheng Tso-hsiu 鄭作新, Chung-kuo niao lei fen-pu mu-lu (Peking, 1955); Armand David and E. Oustalet, Les Oiseaux de la Chine (Paris, 1877); Hans J. Epstein, “The Origin and Earliest History of Falconry”, Isis, 34 (1943), 497– 509; Louis Agassiz Fuertes, “Falconry, the Sport of Kings”, National Geographic Magazine, 38 (1920), 429–460; J. E. Harting, Bibliotheca Accipitraria; a catalogue of books ancient and modern relating to falconry with notes, glossary, and vocabulary (London, 1891): A. von Le Coq, “Bemerkungen über türkische Falknerei”, Baessler-Archiv, 4 (1913), 1–13; D. C. Phillott, The Bāz-nāma-yi Nāṣirī; a Persian Treatise on Falconry (London, 1908); Bernard E. Read, “Chinese Materia Medica, VI, Avian Drugs”, Bulletin, Peking Society of Natural History, 6 (1932), 1–112; E. Denison Ross, “Polyglot List of Birds—Turki, Manchu and Chinese”, Memoirs, Asiatic Society of Bengal, 2 (1907–1910), 253–340; Tsen-Hwang Shaw, “The Birds of Hopei Province”, Zoologica Sinica, Ser. B., 15/1 (Peking, 1936); Arthur de Carle Sowerby, “Shooting and Fishing”, China Journal, 7 (1927), 105; Casey A. Wood and F. Marjorie Fyfe, The Art of Falconry being the De Arte Venandi cum Avibus of Frederick II of Hohenstaufen (Stanford University, 1943). These have proved basic to me, but a great many other titles might be added. See Harting, Bibliotheca, for bibliography of the nineteenth century and earlier; for later publications see the bibliography in Wood and Fyfe, Art of Falconry. 49  See David and Oustalet, Oiseaux, p. 8; Le Coq, “Bemerkungen”, p. 4; Marco Polo, as cited above, et al.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1515

the Steppe Eagle (Aquila nipalensis), and the Spotted Eagle (Aquila clanga).50 Among the important hunting eagles of the Chinese must be numbered the “Hawk Eagle” or “Crested Eagle” (Spizaëtus [= Nisaëtus] nipalensis), though the Chinese named it “Horned Goshawk” (chiao ying 角鷹). This species preys chiefly on pheasants and squirrels.51 This, according to Phillott, may also have been the “Royal Goshawk” (Shāh-bāz) of the Persian falconers’ handbooks; tradition tells that one of those was brought from Chīn (China? Mongolia?) to King Bahrām-i Gūr of the Sassanian Dynasty.52 However, judging from the infrequency of references to the use of hunting eagles by the Chinese—indeed I have observed them only as inhabitants of the imperial mews—it must be assumed that they were a type of imperial gift, but not widely employed by them, as they were among the nomads. Their feathers, however, were in demand for the court fletchers.53 2. Falcons (hu 鶻).54 These are the long-winged hawks, the dark-eyed hawks, the “hawks of the lure”, which stoop to strike their prey in mid-air with clenched fist. They were the aristocratic hunting birds of medieval Europe. 50  Other genera of eagles and eagle-like birds recognized by various authorities as occurring in the Far East are Hieraaëtus, Circaëtus, Haliaeëtus, Thalassoaëtus, Spilornis (Haematornis), Ictinaëtus, and Ichthyophaga (Polioaëtus). Many of these, especially the Sea Eagles, would be useless to most hunters. Bonelli’s Eagle (Hieraaëtus fasciatus) has been described, however, as “… the best eagle in which the falconer can invest” (Wood and Fyfe, Art of Falconry, p. 525). Possibly it was used in China. As to the possibility of the use of the “White-tailed Sea-Eagle”, see discussion of the hai-tung-ch’ing falcon, note 62, p. 1517 below. 51  David and Oustalet, Oiseaux, p. 11; Kiyosu Yukiyasa 清棲幸保, Nihon chōrui daizukan 日本鳥類大図鑑(Tokyo, 1952), 2, 464. Nihon dōbutsu zukan 日本動物図鑑 (2nd ed., Tokyo, 1949), p. 149, states that Spizaëtus has been used for falconry in Japan. There is actually a “Crested Goshawk” (Accipiter [Astur] trivirgatus) in southern Asia. It sometimes reaches southwestern China, but can hardly have been the crested “hawk” so well known to the Northern Chinese, though, according to Phillott, it was formerly trained in India. 52  Phillott, Bāz-nāma, pp. 1–2. 53  Eagle feathers, especially white ones, are listed in several places in the geographical section of the T’ang shu as token tribute from the northern provinces. Li Shih-chen, Pen ts’ao kang mu, 49, takes particular notice of their use in feathering arrows. 54  This character is sometimes read ku. In the sense of “falcon” it should be read hu (*ɣuət); see Kuang yün. The word is also written 鴙 by Yen Shih-ku in his gloss on chun 隼 in Han shu, 27ca, 0415b, and by Tuan Ch’eng-shih. Rotours’ translation “vautours” would astonish any falconer; “faucons” would have been correct. Hu was the T’ang equivalent of classical chun, as Yen Shih-ku points out, though the latter word was still used, apparently rather poetically, in medieval literature. Wang Mao 王楙 (1151–1213) states, in his Yeh k’o ts’ung shu 野客叢書 (Ts’ung shu chi ch’eng ed.), 20, 201–2, that many of his contemporaries

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1516

Schafer

Their prestige was equally great among the Eastern Turks, for in the year 809 the Uigurs adopted, as the official Chinese version of their tribal name, hui-hu (*ɣuậi-ɣuǝt) 廻鶻 “whirling falcons55”. In China, however, though appreciated as kingly birds, they were not as highly valued as goshawks.56 Among the falcons, it was the saker (Falco cherrug), a hunter of herons and other large gamebirds, which was most favored in China, as among orientals generally,57 but the peregrine was also flown, especially at ducks and such lesser fowl.58 But it was the white falcon that was regarded as the most beautiful and noble of hunting birds—a perfectly suitable gift to a royal person. The white falcon of the medieval Chinese texts must have been the gerfalcon.59 The Far Eastern variety of this magnificent bird is a race of the Greenland Falcon, the most esteemed of all hunting hawks in medieval Europe, and venerated equally by the medieval Turks.60 T’ang T’ai Tsung himself possessed a white falcon—a gerfalcon as I suppose—which he named chiang chün 將軍, “Leader of the Army”, which was so well trained that it habitually struck down game-birds in front of one of the palace buildings, which was therefore named “Basilica of the Fallen Geese” (Lo thought that hu was a new word, absent in ancient literature, but Wang was able to find it (possibly in a different sense) in several classical books. 55  In place of the earlier hui-ho 廻紇. T’ang shu, 217a, 4141a. 56  So also elsewhere in eastern Asia. Thus Le Coq, in “Bemerkungen”, p. 6, states that when he was in Kucha a peregrine falcon cost 1/10 the price of a goshawk on the open market. 57  Among the Arabs, the saker falcon and the goshawk are styled ḥurr “free-born; noble”. Phillott, Bāz-nāma, p. 52. 58  The hobby (Falco subbuteo) and the merlin (Falco columbarius) also occur in China, but seem not to have been much used in falconry. The word chan 鸇, which occurs frequently in early literature, though only rarely for a hawk trained for hunting, is not now current. B. Read, in his Avian Drugs, takes it to refer to the kestrel, but since the literature describes the chan as similar to the sparrowhawk, and a swift attacker of pigeons and smaller birds, I feel fairly confident that is is the merlin, our “pigeon hawk”. Other falcons recorded for China are the Red-legged Falcon (Erythropus amurensis = Falco vespertinus), the Kestrel (Falco tinnunculus = Ccrchneis tinninculus), the Lesser Kestrel (Falco naumanni = Cerchneis naumanni), and the Falconet (Microhierax melanoleucus). 59   Falco gyrfalco grebnitzkii (= Falco rusticolus candicans = Falco rusticolus obsoletus), found in Manchuria and elsewhere in northeastern Asia. See Cheng Tso-hsin, Chung-kuo niaolei, pp. 66–67, Bergman, Kenntnis, pp. 90–2, and Kiyosu, Nihon chōrui, 2, 446–7. 60  Le Coq believes that modern Turki tīɣūn, which he renders “white goshawk”, is cognate to the toɣan of the Blue Turk inscriptions, and to the form tuiɣun which Radloff translates “white falcon”. Indeed the old Turkish nobles delighted to name themselves toɣan (Le Coq, “Bemerkungen”, pp. 2, 10). Although white goshawks occur occasionally (see below), I believe that the po hu 白鶻 of the Chinese and the toɣun of the Turks was the famous gerfalcon.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1517

yen tien 落雁殿).61 A kind of falcon held in the highest regard during Sung and Yüan was called “East of the Sea Blue” (hai tung ch’ing 海東靑, or simply “Sea Blue” (hai ch’ing 海靑); this too seems to have been a variety of the Manchurian gerfalcon.62 The “Frosted Falcon” (shuang hu 霜鶻), often referred to in T’ang and Sung literature, must also have belonged to this species. 3. Sparrowhawks (yao 鷂). These are the smaller of the accipiters, which are the short-winged hawks, the yellow-eyed hawks, the “hawks of the fist”. Accipiters kill their prey, not by a blow as the falcon does, but in the clutch of their great needlelike talons, usually pinning it to the ground. The sparrowhawks are used for hunting small birds, such as quail, in wooded places.63 In China there are two species, Accipiter nisus and Accipiter virgatus (= Accipiter gularis). Perhaps the small “goshawk” (Accipiter [= Astur] soloensis) was also grouped with them. Apparently white individuals occur, since it is reported that in ad 798 a white sparrowhawk taken at Ts’o-ê 嵯峨 Mountain in Shensi was presented to the court.64 4. Goshawks (ying 鷹).65 These are the large accipiters, in China limited to one species, Accipiter [Astur] gentilis.66 This was the hunting hawk par excellence of the Chinese, hence the typical Chinese hunter was not a falconer 61   Ch’ao yeh ch’ien tsai 朝野僉載 (T’ang tai ts’ung shu ed.), 53b. 62  Cf. Ross, “Polyglot List”, pp. 273–4. Sung shih, 5, 4505b, speaks of tribute of AD 992 in the form of a “falcon called East of the Sea Blue”. Despite this clear definition, which is verified in other medieval sources, it has been alleged that the hai tung ch’ing was actually no falcon, but a White-tailed Sea-eagle (Haliaeëtus albicilla); see the remarks of P. A. Pichot, translated by J. E. Harting, in The Zoologist, 3rd ser., 9 (Dec, 1885), 447. This latter bird is, it is true, used as a decoy by hawk-catchers, and its tail feathers are much desired for fans (La Touche, Handbook, 1, 168; Shaw, Birds of Hopei, p. 264). It may even have been trained occasionally. But I cannot believe that it was the great falcon which delighted the Jurchen and Mongol kings. 63  The sparrowhawk was evidently popular in China, and indeed everywhere in Asia, in contrast to the low esteem in which it was held in Europe. Epstein’s opinion that this preference for sparrowhawks is “a Persian peculiarity” is too narrow. H. J. Epstein, “Review of Hakan Tjerneld, Moamin et Ghatrif, Traités de fauconnerie et des chiens de chasse”, Isis, 37 (1947), 103. 64   T’ang shu, as quoted in T’ai p’ing yü lan, 926, 8a. See note 36, page 1512 above. 65  The name huang ying “yellow goshawk” is sometimes applied to the young saker falcon in modern Peking. Rotours has “faucons” for ying; “vautours” would be correct. “Goshawk” appears to be the correct translation of skt. ṡyena, to judge from the metathetic transcription of Li Shih-chen: Szu-na-yeh 斯那夜. See also Mahāvyutpatti, Sakaki ed. (Kyoto, 1916), 4903: Chinese t’u-hu 兎鶻 Tibetan sre(ste). 66  The Crested Goshawk (Accipiter [Astur] trivirgatus) and the Annamite Shikra (Accipiter [Astur] badius) have been found in southwestern China.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1518

Schafer

but an austringer. Unlike the falcon, the goshawk makes no spectacular kill in the upper air, but it is incomparable for taking the traditional game of the Chinese, rabbits and pheasants.67 It resembles the sparrowhawk in everything but size—it is much larger. This was also the practical man’s hawk in the West, “… the fiercest and most competent killers of all, and therefore used principally by the ‘yeomanry’ as meat getters68”. In western Asia, too, the goshawk, not the aristocratic falcon, was the type of the hunting bird, so much so that Persian books on falconry were called Bāz-nāma “Goshawk Books”, just as their Chinese analogues were called Ying ching “Goshawk Canons69”. The goshawk is sometimes called ch’ing-ch’iao 靑骹 “Blue Shins”, a term which goes back to the Han dynasty.70 White goshawks, like white falcons, were particularly valued, but there is no species of goshawk whose ordinary color is white, as is the case with the gerfalcon. But a white color variation does occur in northeast Asia; this white phase of the common goshawk has been styled Accipiter gentilis albidus.71 The Manchurian nations were known to the T’ang Chinese as the best source of white goshawks, especially the country of the Mo-ho, which abounded in them.72

67  In Chou times, before the introduction of falconry, rabbits and birds were taken in nets. According to Shaw, Birds of Hopei, p. 236, the larger female goshawk is used for hares, the male for pheasants and partridges. 68  Fuertes, “Falconry”, p. 444. 69  See Harting, Bibliotheca, pp. 198–202, The Turkish word yaraq/yarak “accipiter” even reached England, and was incorporated into the technical language of the hawkers: a hawk in keen hunting condition was said to be “in yaraq”. Le Coq, “Bemerkungen”, p. 2. 70  See J. J. L. Duyvendak, “Review of W. C. White, Tomb Tile Pictures of Ancient China”, T’oung Pao, 35 (1940), 373–5. Duyvendak regards the expression as a hunter’s euphemism, of magical import. The extended form “blue shinned goshawk” (ch’ing ch’iao ying 靑骹鷹) appears in a notice of a gift of these birds to the emperor from a legate stationed at Shachou (Tun-huang) in AD 866 (Chiu T’ang shu, 19a, 3135a). Another old name, obsolete long before T’ang, was shuang-chiu 鷞鳩. The reduced form chiu 鳩 occurs in the Li chi in passages which are incomprehensible if we take chiu to mean “dove”, as Couvreur does in his translation”… quand la colombe était changée en épervier”, for chiu hua wei ying 鳩化爲鷹. The word seems actually to describe a colorchange in the hawk, related to the molting periods. See Li chi, Wang chih 王制 and Yüeh ling 月令; also Ta Tai Li chi, Hsia hsiao cheng 夏小正. Compare the use of ko “pigeon” as a name for a goshawk in its first plumage in the Jou chüeh pu, translated below. 71  Bergman, Kenntnis, p. 97. Phillott observes that “albino” goshawks were known in Persia, where they were called kāfūrī “camphors” (i.e., camphor-colored); he says that in Turkestan these were named lāziqī, after a white flower. Bāz-nāma, pp. 3–6. 72   T’ang shu, 219, 4146d.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

7

1519

Hawking Equipment

The apparatus of hawking has been remarkably uniform in all times and places. This is a conservative art. Timetested methods and equipment are jealously maintained, and so appear little different in China from what they are in England. The chief articles needed by the Chinese hawker are listed below; their Chinese names may be found in the glossary. 1. Mew. This is a cage in which the hawk is kept during the molting period. It must often have been quite handsome, to judge the epithets given it in T’ang poetry: one poet speaks of it as ts’ai lung 綵籠 “polychrome (painted) cage”, another as tiao lung 雕籠 “carved cage”.73 2. Perch. The usual resting place of a hawk, this is a kind of rack or frame, whose principal member is a horizontal pole, which may be ‫ח‬-shaped (supported at both ends), or T-shaped (supported at the center).74 This too could be finely ornamented; the poet Chang Chü 張莒 refers to a hawk’s “golden frame” (chin chia 金架).75 It should be noted that, by virtue of their natural habitats, falcons, which nest on rocky ledges, are ordinarily perched on solid wooden blocks, while accipiters, which roost on the branches of trees, are perched on horizontal poles. That the latter are frequently referred to in Chinese literature, but the former never (to my knowledge) is strong evidence of the preeminence of the goshawk, as opposed to the falcon, in the Far East. 3. Leash. This is a long cord, one end of which is fastened to the jesses on the hawk’s legs, while the other is bound to the perch, to prevent the bird’s escape.76 4. Swivel. A device at the end of the leash, where it fastens to the jesses, to prevent twisting. It was evidently often made of precious substances; a poet, for instance, writes of a “jade swivel”.77 73  P’u-yang Kuan 濮陽瓘, “A Falcon comes from the Mew” (ch’u lung hu 出籠鶻), Ch’üan T’ang shih, han 11, ts’e 8, Yüan Nan-ming 員南溟 chüan, 3b; Chang Chü 張莒 (fl. AD 774), “Rhapsody on a Goshawk Freed from her Mew” ( fang lung ying fu 放籠鷹赋), in Yü ting li tai fu hui 御定歷代陚棄 132, 6b–7a. 74  The latter type is technically a “crutch”. A mural painting at Ïdïqutschähri, near Turfan, shows a pair of hawks leashed to a crutch. See A. Grünwedel, Altbuddistische Kultstätten in Chinesisch-Turkestan (Berlin, 1912), p. 33. The former kind is shown in part, with a white hawk leashed to it, in a painting attributed (with little probability, I am told), to the Emperor Hui Tsung of Sung in E. A. Strehlneek, Chinese Pictorial Art (Shanghai, 1941), p. 107. 75   Chia 架 “frame; scaffold”, is the usual word for a perch. See Chang Chü, “Rhapsody”. 76  Hawks sent as tribute, leashed to a perch or to a carrying pole, are counted with the classifier lien 聯 “binding; bond”, e.g. po hu erh lien 白鶻二聯 “two [bindings of] falcons”. 77  P’u-yang Kuan, “Falcon”.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1520

Schafer

5. Jesses. These are straps attached to the hawk’s feet. They remain there while the bird is in flight; on her return the leash is fastened to them. These are ordinarily made of leather,78 but more costly materials were also used, for we read of “jesses of pease-green silk-thread”,79 and “clouded brocade jesses”.80 6. Bell. Unlike the falcons, which hunt the sky over open plains, the accipiters (goshawks and sparrowhawks) hunt the woods. Since there is danger that their jesses will become entangled in branches or shubbery, they wear bells to help their masters find them.81 Unlike most matters pertaining to falconry, there is considerable variation in local practice as to where the bell is tied. In Europe it is normally attached to the leg;82 the Turks of Central Asia hang it from a neck-band;83 but the Chinese and Japanese uniformly bind it to the medial tail-feathers.84 We read of “gold bells”,85 “jade bells”,86 and “finely chased goshawk bells”.87 7. Collar. This neck-band, with a lead held between the falconer’s fingers, was to steady the bird as it prepared to take flight. Generally it was worn only by sparrowhawks.88 A sparrowhawk’s embroidered collar is mentioned in a poem by Yüan Chen.89 78  Wei Yen-shen 魏彥深 (Sui dyn.), “Rhapsody on the Goshawk” (ying fu 鷹賦 ), in Yü ting li tai fu hui, 132, 4a–5a: “they knot long thongs to both of her feet”. 79  Chang Hsiao-p’iao, “The Goshawk” (ying 鷹), Ch’üan T’ang shih, han 8, ts’e 4, p. 3b. 80  Yüan Chen 元稹, “Yin shan tao yüeh fu”, 陰山道樂府, Ch’üan T’ang shih, han 6, ts’e 10, ch. 24, 11a. 81  Su Shih 蘇軾, Ai tzu tsa shuo 艾子雜說 (in Shuo fu, han 36), 4b–5a. 82  Wood and Fyfe, Art of Falconry, p. 143. 83  Le Coq, “Bermerkungen”, p. 6. Only sparrowkawks normally wear a neck-band, but both Le Coq and Phillott (Bāz-nāma, p. 2) note their occasional use on goshawks among the Persians and Turks, presumably only for the bell, not as a steadying collar as in No. 7, below. 84   Ai tzu tsa shuo, 4b–5a; Kokon yōran kō 古今要覽稿, 185, 77 ff. The Emperor Frederick II knew of this custom, but disapproved of it. See Wood and Fyfe, Art of Falconry, p. 143. 85  P’u-yang Kuan, poem cited above. This poem is about a falcon, which does not need a bell; did the poet err, or was the custom not rigid? 86  Chao Hsia 趙嘏 (ca. 810–806), “Hua ch’ing kung ho Tu she jen” 華淸宮和杜舍人, Ch’üan T’ang shih, han 9, ts’e 1, ch. 2, p. 1b. 87  A courtly gift from the king of Silla. Ts’e fu yüan kuei, 971, 5a. 88  Phillott, Bāz-nāma, p. 2; Harting, Bibliotheca, p. 224. The use of the sparrowhawk collar was observed by William of Rubruck in Mongolia in the thirteenth century. See W. W. Rockhill, trans., The Journey of William of Rubruck to the Eastern Parts of the World, 1253–55, as narrated by himself (London, 1900), pp. 69–70. 89  Yüan Chen, “T’ai chung chü yü i K’ai yüan chiu shih” 臺中鞫獄憶開元舊事, in Yüan shih Ch’ang-ch’ing chi 元氏長慶集 (Szu pu ts’ung k’an ed.), 5, 5a.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1521

8. Hood. Since only falcons, not goshawks, are hooded, references to the hood are rare in Chinese literature.90 An actual falcon’s hood of leather, probably of the ninth century, has been found in Chinese Turkestan.91 9. Gauntlet. Falconers wear a heavy glove or wristlet to protect their arm from the sharp talons of the hawk. This is basically of leather,92 but in medieval China was often elegantly decorated, as indicated by such verses as “The barbarian goshawk with the green eyes treads the brocaded gauntlet”.93 The Chinese name (see Glossary) is an old word for an archer’s cuff. I cannot determine whether it had finger stalls in T’ang times. 8

Hawking Techniques

It is possible to glean from the early records a few facts about Chinese hawking practice which are not alluded to in Tuan Ch’eng-shih’s book. I note them here. 1. Seeling. This is sewing up the hawk’s eyelids during training, a practice well known in the West. It was also done in China: Wei Yen-shen 魏彥深 of the Sui Dynasty, in his “Rhapsody on the Goshawk” (ying fu 鷹賦), wrote, “they stitched light threads in her paired eyelids”. 2. Calling. The falconer must be able to call his hawk to the attack or to himself: “Cry havoc!” But sometimes he whistles instead of shouting, and may even use a manufactured whistle.94 In any case, the hawk must be trained to the appropriate sound. I have not found any T’ang references to hawk-whistling,95 but there are frequent references to calling with the voice. Tu Fu uses, in a poetic simile, the image of a falcon, which “hearing the call, dashes towards

90  Su Shih, Wu lei hsiang kan chih 物類相感志 (Ts’ung shu chi ch’eng ed.), p. 26, refers to a “crow-black falcon wearing its hoodlet”, (ya hu tai mao erh 鴉鶻帶帽兒). Laufer notes that the hood was first introduced into Europe by Frederick II, and adds that it was known in China “since times of old”, whatever that may mean. He gives the name p’i t’ao 皮套 to it. Chinese Pottery, p. 231. 91  Le Coq, “Bermerkungen”, p. 1. This was in the ruins of Chotscho (Ïdïqutschähri). 92  Lu Kuei-meng 陸龜蒙 (ninth century), “Water Bird” (shui niao 水鳥), Ch’aan T’ang shih, han 9, ts’e 10, ch. 5, p. 3b, has wei kou 韋韝 “leather gauntlet”. 93  Hsüeh Feng 薛逢 (fl. 853), “Hsia shao nien” 俠少年, Ch’üan T’ang shih, han 8, ts’e 10, p. 18b. 94  As the hawkers of Kublai Khan: “Every man of them is provided with a whistle and hood, so as to be able to call in a hawk and hold it in hand”. Henry Yule, The Book of Ser Marco Polo the Venetian (3rd ed. rev. by H. Cordier, London, 1903), 1, 402. 95  The art of attracting birds by whistling was certainly known in medieval China. See Hsiao chih 嘯旨, in T’ang tai ts’ung shu, 10.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1522

Schafer

the fowl96”; Yao Ch’ung 姚崇, in answer to an inquiry by the young ruler Hsüan Tsung, said, “When your vassal was twenty years old, he dwelt by the Kuangch’eng Mere, and found happiness in calling goshawks to chase the beasts97”. 3. Carrying. It is traditional in Europe to carry one’s hawk to the hunt on the left arm. In many Asiatic countries the right arm is preferred.98 Judging from what little evidence can be found in art and literature, the medieval Chinese and their immediate neighbors preferred the right: “on his right, his goshawk on arm: on his left, his hound on leash99”. 9

Chinese Books on Hawking

Over a decade ago, George Sarton wrote, “It is known that the Chinese have always practised hawking, but I am not aware of the existence of a single medieval book on falconry ( fang ying shu)100”. The Jou chüeh pu, translated just 96  “Sung shuai fu Ch’eng lu shih huan hsiang” 送率府程錄事還鄕 translated in von Zach, Tu Fu’s Gedichte, 1, 133. 97   T’ang shu, 124, 3959b. The eleventh century book Wu lei hsiang kan chih (see note 90, above) has this curious statement about a falcon which flies off still wearing her hood: “If you call to her standing, she will go away high up, but if you call to her crouching, she will come to you”. 98  Contrary examples can be found, and Frederick II was sublimely indifferent to both traditions. He recommended that the hawk be carried on whichever hand happened to be on the leeward side of the body. Wood and Fyfe, Art of Falconry, p. 143. 99  Written of a young man hawking in the fifth century, Nan shih, 31, 2622c (biography of Chang Ch’ung 張充). The Yüan painter Yen Hui 顔輝 shows hawks carried on the right arm in his painting reproduced in Yün hui chai tsang T’ang Sung i lai ming hua chi 韞輝齋 藏唐宋以來名畫集. Wiillam of Rubruck states that the Mongol falconers carried their birds on the right arm, but his translator, W. W. Rockhill, says that nowadays they prefer the left. See Journey of William of Rubruck, p. 70, n. 1. On the other hand, a painting of a “Mongol” horseman, supposed to be of Sung date, shows a hawk borne on the left. See O. Siren, Chinese Paintings in American Collections, (Paris and Brussels, 1928), 2, pl. 9. 100   Introduction to the History of Science, 3 (Carnegie Institution, Baltimore, 1947), 233. Laufer was rather careless in following Harting when he stated that the earliest Chinese book on falconry was the “Rhapsody 0n the Goshawk”, a descriptive poem written by Wei Yen-shen in the Sui dynasty (see note 78, above). Laufer, Chinese Pottery, p. 234; Harting, Bibliotheca, p. 207. The ultimate source for this mistaken opinion is a very rare and (it is said) beautifully illustrated book: Schlegel and Wulverhorst, Traité de fauconnerie (Leiden, 1844– 1853). I have not seen this work nor can I say what was the source of the authors’ mistake. The poem in question is not the earliest of its kind, nor is it a treatise in any sense (see addendum, below).

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1523

below, was such a book, but others existed in medieval times which are now lost. Indeed a “Goshawk Canon” (ying ching 鷹經), from its title a general account in prose of the goshawk, for the use of hunters, existed during the Han period.101 This may well be the same as the book of that name listed in the official bibliography of T’ang.102 The latter, whether written in Han or not, failed, it seems, to survive into Sung times, for it is missing from the bibliography of that dynasty, though the names of two treatises on veterinary medicine for the goshawk and sparrowhawk are preserved in the History of Sung.103 The most comprehensive extant treatise on hawking in the Chinese language which has come to my attention is, surprisingly, not about Chinese falconry at all. This is the “Discourses on Goshawks” (ying lun 鷹論), authored by one Li Lei-szu 利類思, and preserved in the encyclopaedia T’u shu chi ch’eng.104 The name Li Lei-szu disguises the eminent Jesuit missionary Ludovicus Buglio, a Sicilian who went to China in 1637, resided first in Ch’eng-tu, and later in Peking with Fathers Verbiest, Schall and Magalhaens. The Reverend Buglio knew the Chinese language well, and wrote many books in that tongue, mostly on religious subjects. The Ying lun was written in 1679 to inform the K’ang-hsi Emperor about European falconry.105 Of particular interest in this book are the Chinese phonetic transcriptions which the author has used for the names of favorite European hunting hawks.106 Po-le-ki-no 百勒基諾 Ital. Pellegrino or Peregrino (Peregrine Falcon). [Y]a-ki-la 雅基辣 Ital. Aquilla (Eagle).107 Ju-erh-fa-erh-küeh 入而發兒學 Ital. Girofalco, or Lat. Gyrfalco (Gerfalcon). When, in the future, the general history of falconry in the Far East is written, its author will be obliged to consult a considerable literature in Japanese. Texts 101   San Kuo chih (Wei), 9, 0947a. 102   T’ang shu, 59, 3769a; Chiu T’ang shu, 47, 3265d. 103   Sung shih, 206, 4999d. 104  “Ch’in ch’ung tien” 禽蟲典, ch. 12, “ying pu” 鷹鄧, hui k’ao 彙考, 12b–19a. 105  Louis Pfister, “Notices biographiques et bibliographiques”, 1 (Variétés sinologiques, No. 59; Shanghai, 1932), 230–243. In this reference, the book on falconry is entitled Chin ch’eng ying shuo 進呈鷹說, Both titles are respectably Chinese in containing the word ying “goshawk”, but the book itself shows an aristocratic European preoccupation with the true falcons. 106  Though he writes of eagles and falcons, Rev. Buglio does not use the old Chinese names tiao “eagle” and hu “falcon”. 107  Glossed as yü wang 羽王 “king of feathered (kind)”, and also styled shen ying 神鷹 “divine goshawk”.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1524

Schafer

of various date are reprinted in the Gunsho ruijū 群書類従, 12, 440–517, and in Zoku gunsho ruijū 続群書類従, 19, 590–907. The former of these contains the text of the Shinshū yōkyō 新修鷹経 “Newly Revised Goshawk Canon”, written in the Chinese language and bearing the date AD 818. Possibly this derives irom the lamented “Goshawk Canon” written in Han China, and seemingly lost in that country during T’ang.108 10

The Jou chüeh pu of Tuan Ch’eng-shih

Though Tuan Ch’eng-shih 段成式 is best known as a ninth century man of letters,109 his short study of hunting hawks, the Jou chüeh pu 肉攫部, is (unlike my own) not based on literary sources, but on direct observation and actual experience. He was, in short, extravagantly devoted to hunting, and especially to hunting pheasants and hares, the traditional quarry of the goshawk. In other words, he was an amateur austringer. His notes on the several varieties of goshawk known to him, along with his remarks on a few other matters pertaining to falconry, are here translated.110 108  Other important Japanese materials appear in Yashiro Hirokata 屋代弘賢, Kokon yōran kō 古今要覽考 (1821–1840), chs, 179–188 and 473–490. 109  Date of birth unknown; died AD 863. He has biographies in T’ang shu, 89, 3896a and in Chiu T’ang shu, 167, 3515a, the latter a very brief one. 110  The book is noted in the annotated bibliography in Wood and Fyfe, Art of Falconry, p. 607, as “Tuan Ch’êng-shih (9th century AD). Jou kuo pu. Description of birds of prey, such as eagles, and the way to catch them”. These authors were misinformed: the book is almost entirely devoted to goshawks, and far more with distinguishing the racial and individual varieties of that bird than with the manner of capturing them. A small section of the book has been translated by E. D. Edwards, in Chinese Prose Literature of the T’ang Period, 1 (London, 1937), 199–200. The book appears in modern editions in three forms: (1) as a separate work, with the title Jou chüeh pu 肉攫部, printed in T’ang tai ts’ung shu, Shuo fu (cheng hsü), and Wu ch’ao hsiao shuo. These versions are fundamentally the same; (2) as a separate work, with the title Ying p’in 鷹品 “Sorts of Goshawks”, in some versions of the (Ming) Ch’ung-chen (AD 1628–1643) edition of the collectanea Po ch’uan hsüeh hai. Unfortunately not all sets of the latter are alike: a Ch’ung-chen edition in the Library of Congress does not contain the text, though it apparently exists in the Tsinghua University Library’s copy of the Ch’ung-chen edition. There is also a copy of a Ch’ung-chen edition in the Jimbun kagaku kenkyūjo in Kyoto, but there the title is Jou chüeh pu. I have inspected all of these, and they are virtually identical, even in points requiring emendation, and therefore appear to descend equally from a late T’ang or early Sung exemplar. I am greatly indebted to Dr. Edwin G. Beal, Jr., of the Library of Congress, for helping me to untangle these matters; (3) as chüan 20 of the various modern versions of Tuan Ch’eng-shih’s Yu

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1525

Translation The Divisions of the Predators of Flesh111 1 The Manner of Taking Goshawks The twentieth day of the seventh month112 is the superior time—then the ones in our inner land are abundant, while those beyond the frontier are quite few. The upper decad of the eighth month is the second-rate time. The lower decad of the eighth month113 is the inferior time, when the goshawks from beyond the frontier come no longer. The eyes of a goshawk net are 1.8 inches square, with eighty eyes vertically and fifty eyes horizontally. It is dyed with yellow phellodendron114 mixed with oak liquor,115 which makes it like the color of earth.116 The locust insects like to eat the nets, but they are protected by the phellodendron. There are net poles, pickets and stakes,117 “centipedes”,118 and two split poles, one being the quail pole and one the pigeon pole.119

yang tsa tsu. I suspect that the Jou chüeh pu did not originally form part of this book, but was added editorially in post-T’ang times to round out the traditional twenty chapters. 111  The title of the book indicates that it purports to classify the several kinds of hunting hawks. 112  This might fall anywhere in the period from mid-August to the first week in September according to our calendar. This is the time when the fall migration of hawks southward from Mongolia and Manchuria is at its peak. 113  Any time from mid-September to mid-October. Then the fall migration comes to an end. 114   Huang po 黄蘖, Phellodendron amurense, whose inner bark was used both as a yellow dye and as an insect repellent. 115   Chu chih 杼汁, a black dye prepared from the acorn cups of Quercus serrata (= Q. acutissima) of North China. See Erh ya i shu 爾雅義疏. 116  So as to be invisible from the sky. 117   Tu i 都杙. Tu was a dialect word for chüeh 橛 east of the lower Yangtze. See Fang yen chu 方言注. 118   Wu-kung 吳厺. Clearly a device connected with the decoy-and-net arrangement, but I have been unable to identify it. Kimura Kōkyō 木村孔恭, San-kai mei-san zu-e 山海名 產圖會 (Osaka, 1799), II, describes the use of a painted artificial snake, made of jointed wood, to keep a decoy starling agitated enough to draw the attention of a hawk overhead. The “centipede” may have played a similar role. 119  Apparently the quail was the decoy bird. The pigeon was the sentry bird, as the next sentences show. However, Richard F. Burton reported that “In India … a pigeon is generally the bait for falcons or long-winged hawks, a quail for the short-winged birds”. See his Falconry in the Valley of the Indus, (London, 1852), pp. 7–8. A variety of birds have been used in different parts of the world to lure hawks to the net. In Persia, for instance, trained

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1526

Schafer

A pigeon in flight can discern things from afar, and regularly sees a goshawk before a man does; so when it quivers its body and moves its eyeballs, watch for [the hawk] in the direction of its gaze.120 2 Taking Sparrowhawks for Wood Fowl and Wood Sparrows121 The eyes of this net ate two inches square, with thirty eyes vertically and eighteen eyes horizontally. 3 [Taking Eyases]122 All raptorial birds are kindly disposed towards the chicks at their birth. After these have emerged from their shells, they are immediately free of the nest and will go outside this den. The great raptors, fearful lest they fall down, or lest they incur injury from siriasis in the violent heat of the sun, take the branches of leafy trees and bring them to insert in the parapet of the nest, which keeps them from falling down, and also makes it shady and cool. Now if you wish to assay whether the chicks are large or small, you should attend to these inserted leaves. Should it be one day or two days, the leaves will still carry some green color, though wilted. In six or seven days, the leaves will be slightly yellow. After ten days they will be withered and spent. By this time the chicks have grown big enough to be taken. 4 [Protective Coloration] All birds and beasts must, by their similarity to various kinds of objects, conceal and cover shadow and shape. Thus the color of a snake goes with the earth; the grass123-rabbit is sure to be red; the color of the goshawk takes after the trees.

owls have been used as decoys (Phillott, Bāz-nāma, pp. 18–20). See Wood and Fyfe, Art of Falconry, pp. 439 ff. for various methods, including the use of pigeon poles. 120  By its agitation, the pigeon warns the netter of the approach of the hawk. In the West, this sentry role is most often performed by a shrike. 121   Mu chi mu ch’üeh yao 木雞木雀鷂 i.e. “Wood-fowl Sparrowhawks and Wood-sparrow Sparrowhawks”, non-specific names for sparrowhawks trained to catch medium gamebirds (quail, partridge, etc.) and small game-birds (sparrows, thrushes, etc.) respectively. 122  Subject titles in brackets have been supplied by the translator to facilitate reading and reference. Although passage hawks netted during the fall migration were preferred as hunting birds, eyases (i.e. hawks raised in captivity from the nestling state) were also used. 123   Mao 茅 “floss-grass”, esp. gen. Imperata.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1527

5 [Technical Terms] The goshawk’s nest is also named the goshawk’s “eyrie”,124 and what we call an “eyas”125 is the chick goshawk. 6 [Molting and Mewing] On the first day of the fourth month,126 casting the goshawk is suspended, and during the upper decad of the fifth month127 she is put in her cage to molt her feathers.128 This molting of feathers starts first from the head, and will, by the coming of a dawn, have gone from crown to belly.129 (A quail, on the other hand, goes only from the nape down past the “soaring feathers”,130 stopping when it reaches the tail. The feathers under the root of the tail are named “soaring feathers”). The feathers of the back, along with the great pinions131 and covering quills132 in the pair of wings, and also the twelve feathers of the tail—she molts them all. The great feathers of the pair of wings combined are forty-four, while the covering quills and pinions are also forty-four. In the middle decad of the eighth month,133 she leaves the cage. Casting of eagles, horned goshawks,134 and the like, is suspended on the first day of the third month;135 these are placed in their cages in the upper decad of the fourth month.136 124  Emending ch’u ying 菆鷹 to ying ch’u 鷹菆. 菆 is normally tsou (*ṭṣi̯ə̭u), but Kuang yün gives also the reading *ṭṣ’i̯u, defined as “bird’s den”. This would be Mand. ch’u. Clearly cognate are 芻 ch’u (*ṭṣ’i̯u) “hay” and 雛 ch’u (*ḍẓ’i̯u) “chick”. 125   Ch’u tzu 菆子. “Eyas” is ultimately from Latin nidus “nest” (French “niais”). 126  In late April or early May. 127  Late in May or during the first two weeks of June. 128  The word mao 毛, also used of a mammal’s body-hair, means, with reference to birds, the small contour feathers as opposed to the large flight feathers, though it is occasionally used more broadly. 129  “Belly” is conjectural. The text has ju 伏, but no suitable meaning is registered in the dictionaries. The common meanings “brood; incubate”, however, suggest that it refers to the underside of the body. 130   Yang mao 颺毛. A name for the under tail coverts, as explained by Tuan. 131   Ling 翎, i.e. the primaries and secondaries. 132   Ho 翮, i.e. the wing coverts. 133  Mid- or late September. 134  The listing of this crested “hawk” here confirms the identification of chüeh ying as the eagle Spizaetus nipalensis. 135  Late March or early April. 136  Late April or early May.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1528

Schafer

If the casting of falcons is suspended when the north-ward-returning goshawks have all passed,137 and they are put in their cages in the upper decad of the fourth month, they will not molt their feathers. Casting of falcons should be suspended in the upper decad of the fifth month,138 and they are put in their cages to molt their feathers in the upper decad of the sixth month.139 7 [Plumage Phases] All raptorial and aggressive classes are “dove-hawks” in their first phase, and “sore-hawks” in their second phase, when they evolve into “gray-hawks”; in their third phase they are true “gray-hawks”. After this, having achieved the sum of their phases, they are always true “gray-hawks”.140 8 The White “Dove-Hawk”141 If they have white beak and claws, when, in the next phase, they are sorehawks and then achieve the sum of their phases, they will be wholly fixed in their white color. There will be no further alteration or change. If beak and claws should be black, there will be the slightest trace of yellow color in the vertical patterns142 on the front of the breast and in the spotted sections of the pinions and tail. When, in the next phase, they become 137  I.e., the spring migration is over. 138  Late May or early June. 139  End of June to mid-July. The falcons molt, our author says, a month later than the goshawks. 140  The three plumage phases are, in Chinese, (1) ko 鴿, lit. “pigeon”, (2) p’ien 鴘” reddish plumage”, and (3) ts’ang 鶬 “slate-gray or blue-gray hawk”. As for my translations, ko evidently refers to the dove- or pigeon-gray color of the goshawk chick; for this I have adopted English “dove-hawk”, which is properly a dialect word for the hen harrier, a very different kind of hawk. “Sore-hawk” is the traditional falconer’s name for a young hawk in sorrel plumage, and suits the Chinese perfectly; it is used properly even if the hawk should not turn reddish in its first change, as is the case with albino hawks. “Gray-hawk” is a coined word in the sense I have given it (the term is sometimes used of a plumage phase of the peregrine falcon in English) as equivalent to Chinese ts’ang 鶬, which is cognate to ts’ang 蒼 “slate-gray; blue-gray; glaucous”. Mature goshawks have slate-gray backs, and are called ts’ang ying 蒼鷹. To the phase-names employed by Tuan Ch’eug-shih, those used in Kuang chih 廣志 (4th century? As quoted in Ko chih ching yüan, 79, 20a–b) should be compared. This earlier text calls the bird in its first phase “yellow goshawk”, in its second “fondled(?) goshawk” ( fu ying 撫鷹), and in its third “blue goshawk” (ch’ing ying 靑鷹). 141  This section deals with white and other light-colored nestlings, and their subsequent color changes. 142  The immature plumage (red stage) of the goshawk shows vertical bands of breast spots. In the mature plumage (slate-gray stage), the breast is horizontally banded.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1529

sore-hawks, their feathers will appear purplish white in the region extending from the top of the hump of both wings as far as both thighs. Elsewhere the white color will be unaltered.143 Kao Wei, Prince of Ch’i, in the sixth year of “Martial Tranquillity”,144 obtained a white dove-hawk, sent him by P’an Tzu-kuang of Ho-tung,145 ServantToxophilite of the Circuit Tribunal in Yu-chou.146 The whole of its body was like the color of snow. It showed imperceptible vertical patterns of “white” spots on the front of the breast; the color of these patterns was indistinctly pinkish. The color of the base of the beak was faintly tinged with blue-white, but toward the tip it tended to become raven-black. The claws were just the same as the bill. Both cere147 and shanks were yellow and whitish red. Such is the supreme quality. If it is yellow hemp-colored, there will not be much change or alteration in the next phase, when it becomes a sore-hawk, though the vertical spottings on the front of the breast will tend to become broader and shorter. After its evolution and emergence from the sore-hawk state, when it has achieved the sum of its phases, a blue color is faintly added on the back, while the vertical patterns on the front of the breast turn shorter and finer, and a clear white tends to be added above the knee. Such is the second-rate quality. If it is blue hemp-colored, the colors of its phases are altogether the same as those of the yellow hemp sore-hawk. Such is the inferior quality.148 (We also have “netted raven-black falcons” and “netted hemp-colored falcons”).149

143  Cf. Bergman, Kenntnis, p. 98, of a variety of white goshawk, “Die Schulterfedern sind schwach braun marmoriert”. Chinese tzu “purple” includes some shades of dull red and brown. 144  This was the last ruler of “Northern Ch’i”. The year was AD 575. 145  Ho-tung P’an Tzu-kuang 河東潘子光. He was son of P’an Le 潘樂. His name appears as Tzu-huang 子晃 in his father’s biography in Pei Ch’i shu, 15, 2220c. 146   Yu-chou hsing t’ai p’u yeh 幽州行臺僕射 second ranking officer at the provincial magistracy in Yu-chou, region of modern Peking. 147  The soft yellow area at the base of the bill of a bird-of-prey, into which the nostrils open. Curiously, the cere (“wax”) is called la 臘 “wax” in Chinese also. 148  Emending the “color” of the Szu pu ts’ung k’an edition to “quality”, conformably to the earlier analogues. 149  These are color varieties of falcons, taken in nets. Cf. T’ang shu, 37, 3719d, for tribute of “sparrowhawks and raven-black falcons” from Hua-chou 華州. Bergman, Kenntnis, p. 90, reports that he saw a peregrine falcon from Kamchatka whose upper parts were almost black.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1530

Schafer

9 The White Rabbit Goshawk150 If they have white beak and claws, when, in the next phase, they are sorehawks and then achieve the sum of their phases, they will be wholly fixed in their white color. There will be no further alteration or change. If beak and claws are black, tinged faintly with a bluish white color, there will be a faint yellow color in the vertical patterns on the front of the breast and in the spotted sections of the pinions and tail. In the next phase a faint ash color will appear on the back, the pinions, and the tail, while the vertical patterns on the front of the breast will be transformed into horizontal patterns. The color in this phase is faint and scant, almost non-existent. The region of the shanks will remain white. After it has achieved its evolution from the sore-hawk state, this ashy color will become faintly brown, tending gradually towards white. If beak and claws are black in the extreme, it will be spotted on the top of the body like a yellow magpie, the color being rather deep. In the next phase it will become a bluish white sore-hawk, and after its evolution from sore-hawk, when it has achieved the sum of its phases, the horizontal patterns on the front of the breast will turn finer, and it will gradually take on the color of a gray-hawk. Kao Yang, Prince of Ch’i, in the third year of “Heavenly Conservation”,151 acquired a152 white rabbit goshawk. The place where it was taken is unknown. Feathers and plumage over the entire body were like snow. The color of the eyes was purple. The claws were white at the base, but became a pale raven color towards the tips. Both cere and shanks were yellow. In that time it was called “Gold Feet”. Moreover, at the beginning of the Ch’i Theocrat’s “Martial Tranquillity”,153 Chao Yeh, Army Captain Commanding an Army,154 also presented a white rabbit goshawk. Viewed from afar its head was entirely white up to the crown, but studied thoroughly from close by, there were traces of purple in the hearts of the feathers. On its back, the hearts of the feathers were dotted with purple on a white ground. The purple was encompassed and girded on the outside with 150  Our author notes later that “pheasant goshawks” are normally males (tiercels), while “rabbit goshawks” are normally females. I take it that this section is devoted to the larger white females. 151  Wen Hsüan Ti 文宣帝 of Northern Ch’i; the year was AD 552. 152  The translation here does not reveal the use of the classifier lien 聯 “binding; bond”, used to count hawks, pictured as tied to a perch. The phrase “white rabbit goshawk, one binding” (白兎鷹一聯) occurs in the next paragraph. 153  This would be about AD 570. 154   Ling chün chiang chün 領軍將軍, at this time a commander of the imperial household guards.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1531

whitish red, and beyond the white there was a border of black. The wing feathers also had a white ground, sectioned in purple color. There was a ground of white on the front of the breast, with imperceptible vertical patterns in pinkish red. The yellow of the eyes was like true gold. The color at the base of the bill was rather white, tending to raven-black towards the tip. The cere was a pale yellow color, and the color of the shanks and toes was also yellow. The color of the claws was the same as that of the beak. 10 The “Scattered Flower White” If they have black beak and claws, faintly tinged with a blue-white color, they become purple-patterned white sore-hawks in the next phase. After the evolution from sore-hawk, when they have achieved the sum of their phases, the horizontal patterns turn finer, and the purple on the front of the breast tends to disappear and become white. If beak and claws are black in the extreme, it will become a blue-white sore-hawk in the next phase. After the evolution from sore-hawk, when it has achieved the sum of its phases, the horizontal patterns will turn finer, and the front of the breast will tend to take on an ashy white color. 11 The “Red Colored” As a sore-hawk in the next phase, the color is tinged with black. After the evolution from sore-hawk, when it has achieved the sum of its phases, the horizontal patterns turn finer, and the front of the breast tends imperceptibly to whiten. The color of the back is unaltered. These are superior colors.155 12 The “While Engouted”156 When, in the next phase, it becomes a sore-hawk, it is faintly tinged with ash color. After evolving from sore-hawk, and achieving the sum of its phases, the horizontal patterns turn finer, and the front of the breast tends imperceptibly to whiten.

155  Referring to the two varieties just above. 156  “Engout” is an old falconer’s word: “to show black spots on the feathers” (see Wood and Fyfe, Art of Falconry, p. 618). The Chinese word is t’ang 唐, explained later in the text as meaning “black-blotched”. This variety of hawk, then, is basically while, with black gouts.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1532

Schafer

13 The “Skylark Yellow”157 As a sore-hawk in the next phase, it is colored like the plumes of an adjutant.158 After the evolution from sore-hawk, when it has achieved the sum of its phases, the horizontal patterns turn finer, and the front of the breast tends gradually and imperceptibly to whiten. 14 The “Yellow Colored” After the next phase, and when it has achieved the sum of its phases, its color suggests the plumes of an adjutant, but the color is slightly deeper. By and large, it is the same as the phase-color of the “Skylark Yellow”. 15 The “Blue Spoiled”159 It is a blue …160 sore-hawk in the next phase. After the evolution from sorehawk, when it has achieved the sum of its phases, the horizontal patterns turn finer, and the front of the breast tends imperceptibly to whiten. These are second-rate colors.161 16 The “White [Spot]162 Engouted” (“Engouted” is a black color). This means that there is a black color on the spots. In the next phase it is a blue-white sore-hawk, tinged at random with black color. After the evolution from sore-hawk, when it has achieved the sum of its phases, the horizontal patterns turn finer, and the front of the breast tends gradually and imperceptibly to whiten.

157   Yen-lan-tui’ (*ân-lân-tuâi) 鷃爛堆, thought to mean “skylark”, appears in other literature as yen-lan-tui (*an-lâm-tuâi) 鷃濫堆 and e-lan-tui (*â-lâm-tuâi) 阿濫堆. Since both 爛 and 濫 show archaic *GL-, we may surmise an early form like *ARGLANTOI, evidently a loan-word from a non-Chinese tongue. 158   Ch’iu ch’ang 鶖氅. Compare ho ch’ang 鶴氅 “crane plumes”, especially as used in medieval China for feather capes and ceremonial pennants. The ch’iu 鶖 (or t’u ch’iu 禿鶖) appears to be Leptoptilos javanicus, the adjutant of South China, with “… parties supérieures du corps et plumes occipitales d’un noir foncé à reflets verts métalliques …” (David and Oustalet, Oiseaux, pp. 449–450). 159  “Blue” is here, as elsewhere, ch’ing 靑. 160  The text has fu 父 “father”, but no emendation suggests itself, except possibly wen 文 “patterned”. 161  The four varieties above. 162  The text has only “white engouted”, but clearly there is a lapsus of pan 斑 “spot” here; cf. article 12 above, and 17 and 18 below.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1533

17 The “Red Spot Engouted” This means that there is a black color on the spots. When, in the next phase, it is a sore-hawk, its color is tinged with bluish black. After the evolution from sore-hawk, when it has achieved the sum of its phases, the horizontal patterns turn finer, and though the black on the front of the breast tends to become brown, still men commonly name it “black gray-hawk”. 18 The “Blue Spot Engouted” This means that there is black color on the spots. When it is a sore-hawk in the next phase, its color is tinged with bluish black. After the evolution from sore-hawk, when it has achieved the sum of its phases, although the horizontal patterns are finer, the color of the front of the breast remains smudged with night-black. These are inferior colors163. 19 [Male and Female Goshawks] The female and the male of the goshawk may be distinguished only by greater or smaller size. They cannot, in the end, be set apart or discriminated by anything else in form or aspect. Thus, though the “pheasant goshawk” is small, yet it is the male goshawk. Since from the first and through its phases, the several colors of the feathers in its coat are the same as those of the “rabbit goshawk”, there is nothing further to be told by way way of discriminating them.164 When the “pheasant goshawk” has broad vertical patterns on the front of its breast in the first year, its common name is “Falcon Spotted”.165 When it has achieved the later phases of sore-hawk and gray-hawk, and the vertical patterns on the front of the breast become horizontal patterns, they will still be broad and large. If the vertical patterns on the front of the breast are fine from the first, when it has achieved the later phases of sore-hawk and gray-hawk, the horizontal patterns on the front of the breast will still be fine.

163  The three above. 164  It appears from this that the “pheasant goshawk” and the “rabbit goshawk”, named for their normal prey, are the male and female respectively, in technical language the “tiercel” and the “falcon” (the latter word was formerly reserved for a female hunting hawk). 165   Hu pan 班. The young male goshawk is sometimes, it seems, patterned like a saker or peregrine.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1534

Schafer

20 The “Redbud Den White”166 It is large, though short of body, being five catties and over. It is swift, and suited to birds.167 Its alternate name is “Sandy Interior White”.168 It breeds in the interior of the sandy waste, north of Tai,169 in its den in a redbud. It flies towards Goose Gate and Horse City.170 21 The “Tai Metropolis Red”171 It is purple of back, black-bearded, white of eyeball, and white feathered. It is three and a half catties and over, to four catties and under.172 It is suited to rabbits. It breeds within the red cliffs on the Tai River,173 and flies toward Empty Hill,174 Chung-shan and Po-chien.175 22 The “North of the Waste White”176 Its body is long and large, being five catties and over. In fineness of spotting and shortness of shank, it is foremost among the goshawks. It breeds north of

166   Ching k’o po 荆窠白. K’o is the “den” of a bird or mammal, having wider connotations that ch’ao 巢 “nest”. This variety of hawk nests in the redbud tree (ching 荆). Most plants included under the name ching belong to the genus Vitex, which is restricted to Central and South China. However, the tzu ching 紫荆, the redbud or Judas Tree (genus Cercis) grows in the arid north, the region here described. There is even a “Purple Cercis Mountain” [tzu ching shan 紫荆山), named for the many redbuds on it, in Tai-chou. See Ch’ing i t’ung chih 淸一統志, 114, 1b. The preceding list of goshawk varieties is a classification based on color and pattern. With the present item, a new list begins, classifying the goshawks native to North China, especially to northern Shansi and Hopei, with notes on their size, normal prey, and breeding and hunting ranges. 167  I.e. best used as a hunter of birds. 168   Sha li po 沙裏白, i.e. the white hawk which dwells in the interior of the desert. 169  Tai-chou 代州 in northern Shansi. The breeding range of the hawks described in this and the immediately following sections is north of about 390 N, latitude, across Shansi and Hopei. 170   Yen men Ma i 雁門馬邑, both being towns in northern Shansi, west of Tai. 171   Tai tu ch’ih 代都赤. 172  About five or six pounds. 173   Tai ch’uan 代川, apparently the modern Hu-t’o 滹沱, River, which flows from the eastern mountains past the city of Tai towards the southwest. 174   Hsü ch’iu 虛丘, perhaps a misprint for Ling ch’iu 靈丘. 175   白 中山. Po-chien is probably the “White Gorge Mountains” (白澗山) south of Tai in Shansi; Chung-shan “Central Mountains”, is eastward, across the mountains in Hopei. 176   Mo pei po 漠北白.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1535

the sandy waste, but I do not know whether far or near. It flies towards the Tai River and Chung-shan. Its alternate name is “Western Road White”.177 23 The “Fang-shan White”178 It is purple of back and finely spotted. It is three catties and over, to four catties and under.179 It is suited to rabbits. It breeds east of Tai on the white lindenpoplar trees180 of Fang-shan, and flies towards Fan-yang181 and Chung-shan. 24 The “Yü-yang White”182 Both belly and back are white. Large ones are five catties.183 It is suitable for rabbits. It breeds from Hsü-wu to the East and West Bights,184 which are alternately named “Great Bight” and “Small Bight”. It lives on the white-leafed tree,185 and flies towards Chang-wu, Ho-k’ou, and Po-hai.186 25 The “Eastern Road White” Both belly and back are white. Large ones are six catties or more.187 This is the largest among the goshawks. It breeds north of Lu-lung and Ho-lung,188 but I do not know whether far or near. It flies towards Huan-lin, Chü-hei, Chang-wu,

177  Cf. item 25 below, the “Eastern Road White”. 178   Fang shan po 房山白. Fang-shan is modern P’ing-shan 平山, across the eastern mountains in Hopei. This and the previous variety of hawk, then, inhabit the vicinity of the Wu-t’ai Mountains on the Shansi-Hopei border. 179  Five or six pounds. 180   Po yang tuan shu 白楊椴樹. Yang tuan should be tuan yang “linden-poplar”, a local name for the Hopei poplar (Populus hopeiensis), and has been so emended. The prefixed “white” may mean, however, that the author had in mind the white poplar (Populus alba). 181   范陽, also in northwestern Hopei. 182   Yü yang po 漁陽白. Yü-yang is in northern Hopei, near the Great Wall. 183  About seven and a half pounds. 184   Hsü wu chi tung hsi ch’ü 徐無及東西曲. Hsü-wu is the name of a mountain and town in northen Hopei. The Bights remain unidentified. 185  Unidentified. 186  Chang-wu 章武 and Ho-k’ou 合口 are in eastern Hopei, near the coast. I cannot find the name Po-hai 博海; it may be an error for Po-ling 博陵, Po-p’ing 博平, or P’o-hai 渤海. 187  About nine pounds. 188  Lu-lung 盧龍 is in northeastern Hopei, near the Manchurian border. I have not been able to find Ho-lung 和龍.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1536

Schafer

Ho-k’ou and Kuang-chou.189 Although it is rather weak, if you come by a swift one, it surpasses the previous goshawk.190 26 The “Earth Yellow” It occurs everywhere in the mountain valleys. It lives on the oak trees.191 Some are large, some are small. 27 The “Black Inky Black”192 Large ones are five catties. It breeds in the mountains of Yü-yang, on pine and fir trees.193 Most of them die.194 At times there is a swift one. They fly to Chang-wu. 28 The “White Inky Black”195 Large ones are five catties. It breeds at Yü-yang and Po-tao.196 It occurs everywhere on the sunside of the Ho197 and north of the waste. It lives on thuja trees.198 It is suitable for birds. It flies towards Ling-ch’iu,199 Chung-shan, Fanyang and Chang-wu. 29 The “Blue Spotted” Large ones are four catties. It breeds from north of Tai to the Tai River, on the white poplar trees. The finely spotted ones are swiftest. It flies towards Lingch’iu Mountains and Fan-yang. 189   渙林巨黑章武合口光州. Some edition have chü-mo 巨墨 for chü-hei, and kuang-ch’uan 光川 for kuang-chou. I cannot find them in any case, nor Huan-lin either, nor do any probable emendations occur to me. Possibly they were in southern Manchuria. 190  I.e. the “Yü-yang White”. 191   Tso li shu 柞櫟樹. This is a collective term for the deciduous oaks of North China, including Quercus mongolica, Quercus liaotungensis, and Quercus acutissima. 192   Hei tsao li 黑皂驪. Tsao is the black dye derived from oak-galls and other vegetable sources, hence “inky”. Li is a black horse, a “Black” in horseman’s language; the name is sometimes applied to other black animals. 193   Sung sha shu 松杉樹. Collective for needle-leafed conifers, excluding scale-leafed gymnosperms such as the arborvitae, juniper and cypress. Sha includes yews, hemlocks, larches, etc., as well as the firs. 194  This black variety is not normally viable. 195   Po tsao li 白皂驪. 196   白道, in northern Shansi. 197  Ho-yang 河陽, i.e. north of the Yellow River. 198  The po k’u shu 栢枯樹 of the text is incomprehensible. Perhaps k’u 枯 should be emended to sung 松. Po 栢 (= 柏) probably refers here to Thuja orientalis, the arborvitae of northeastern China. 199   靈丘, east of Tai, in northeastern Shansi. Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Falconry in T ’ ang Times

1537

30 The Sore-Goshawk “Perilla Seed”200 The bluish black ones are swiftest. 31 [Signs of Good and Ill Health] If the exuviae201 are clean and the eye is luminous, such a one has not been raised from a chick and is especially swift. But if the eye is much bleared and the exuviae are not clean, it will have been raised from a chick, and will not bear use; many such die. Furthermore, if the head of the mutes202 lacks flowers, and if they are congealed even though cast afar,203 or if they make a sound like “kou!” as they issue, the prognosis is a short destiny. But if the interior of the mouth is red, and if the heat of the reversed sole transfuses to a person through the partition of his garment, the prognosis is a long destiny. If, when it folds its tail, it shakes it out and furls it so as to strike the frame,204 and stands apart to preen the feathers of its face, and hides its head when sleeping, the prognosis is a long destiny. In all raptorial birds one dreads in particular the complication205 of a disease of the trachea entering the “fork”;206 not one in ten will survive. The “fork” is under the skin which overlies the trachea-bone207 in the throat. It is inside the “broken basins”-bone,208 and below the crop.

200   P’ien ying jen tzu 鴘鷹荏子 appears in all accessible editions. If this is not corrupt, it may refer to a variety of hawk with seedlike spots in the sore-hawk stage. But indeed the remainder of the book has a mangled look. 201   Shui 蛻, i.e. molted feathers. 202  The character t’iao 條 appears five times in the following paragraphs. The term ying t’iao 鷹條 is defined in ch. 9 of Pen ts’ao kang mu shih i 拾遺 as “hair (or feathers, mao 毛) in the feces, which have not been completely transformed (i.e. digested)”. From the context in our present text, however, I take this to be a falconer’s word for the hawk’s excrement, like English “mutes”. A cognate word would be hsiu 滫 “stinking liquor” (in Huai nan tzu), and “urine” (in Hsün tzu); related also is niao 溺 (see Chung hua ta tzu tien 中華大字典). 203  These clauses are unclear: 又條頭無花, 雖遠而聚. In a healthy hawk, “… the mutes should be cast clear away to a distance”, Phillott, Bāz-nāma, pp. 153–4. This seems to bear on the issue, but I can’t tell just how. 204   Ko 格, alternate for chia 架, in the sense of “perch”. 205  My translation of the first part of this sentence is somewhat conjectural. 206  Apparently the bifurcation of the trachea above the lungs. Here written 叉, in the next sentence written 汊. 207  The author uses “bone” also for cartilaginous structures. 208   Ch’üeh p’en ku 缺盆骨. The syrinx, or “lower larynx”? Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1538

Schafer

32 The Suction Pipe209 It is made of silver leaf,210 and is as large as the wing-tubes211 of the horned goshawk. For goshawks and below,212 it is large or small in proportion to the wing-tube. 33 [A Fatal Symptom] Mutings at night should not exceed five. When mutings are numerous, it shortens the destiny. If the mutes are like the juice of the red little beans,213 mixed with white, it will die. 34 [Defects to be Watched for] Injuries got in the net; wounds from preening;214 wounds from the treading of hares, and from the “weapon-talons” of cranes;215 all these are maladies. an4 chan1 che4 chi1 chia4 chih4 ch’ih4 ch’u2 ho2 hsieh4 hsüan4 hu2

Glossary of Falconer’s Terms 按 “try out; test a hunting hawk” 鸇 “merlin?”



“sparrowhawk’s collar”

羈 “leash” 架 “perch” 鷙 “raptorial; a raptor” 翅 “wing” 菆 “eyrie” (hence ch’u2 tzu3 菆子 “eyas”) 翮 “quill” 紲 “leash” 鏇 “swivel” (often misprinted 鏃 “arrowhead”) 鶻 “falcon”

209   Hsi t’ung 吸筒. This looks like a drinking tube, but I have not been able to ascertain its use. The Japanese falconers carry a tube storing the hawk’s drinking water; this is called 水筒 (Kokon yōrankō, 183, 60, has a picture). But “suction” seems inappropriate for such an object. 210   Yin yeh 銀鍱, i.e. a thin sheet of silver, rolled up. 211   Ch’ih kuan 翅管, apparently the primary quills. 212  Goshawks and other hawks smaller than the “horned goshawk”. 213   Ch’ih hsiao tou 赤小豆, the “red mung bean” (Phaseolus mungo var. radiatus). 214  ? pai shang 擺傷. 215   Ping chao 兵爪, a name for the first claw on the crane’s foot, See Yu yang tsa tsu, 16 (Ts’ung shu chi ch’eng ed.), 126.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1539

Falconry in T ’ ang Times

ko1 kou1 la4 ling2 ling2 mao2 pa2 pien4 p’ien3 t’ang2 t’ao1 tiao1 t’iao2 ts’ang1 yao4 ying1

鴿 1. “pigeon”; 2. “dove-hawk” [nestling hawk] 韝, 鞲 “gauntlet” 蠟 “cere” 鈴 “bell” 翎 “pinion; flight feather” 毛 “contour feathers” 拔 “molt” 變 “coat; phase of plumage” 鴘 “sore-hawk” [immature hawk] 唐 “engouted” 絛, 縚, 縧 “jess” 鵰 “eagle” 條 “mutes” 鶬 “gray-hawk” [mature hawk] 鷂 “sparrowhawk” 鷹 “goshawk; accipitcr; hunting hawk”

Addendum p. 1522, n. 100.—The Traité de fauconnerie by G. Schlegel and J, A. Verster van Wulverhorst (vi + 90 + viii pp., 5 + 12 plates—71.5 cm. high and 54 cm. wide) contains i.a. a brief description of falconry in China (p. 65), as well as the titles of four Chinese works on the subject on p. V of the appendix, entitled “catalogue raisonné des ouvrages de fauconnerie”. These were collected and (partly) translated for the authors by J. Hoffmann, professor for Japanese at Leiden University, who had found them quoted in the Japanese encyclopedia the Wa Kan san-zai zū-e of 1714. The Ying-fu is described as a “classification des faucons”, with the remark that this poetic effusion is only of little scientific interest. However, Hoffmann does not say that this Sui work is the earliest Chinese book on falconry.—A. Hulsewé.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Public Values in Calligraphy and Orthography in the Tang Dynasty Amy McNair Calligraphy and orthography were considered important means of assessing moral fitness for government office beginning in the Tang dynasty (618–907), due to the equation between style and personality. As handwriting was considered to reveal the moral condition of the candidate, “exemplary writing” (calligraphy) and “correct writing” (orthography) were necessary skills to acquire in order to create an impression of exemplary correctness. This situation raises certain questions for art historians. As the visual substance of written communication and moral quality, how do calligraphy and orthography embody values of public utterance and posture? Were these values imposed by the state and the demands of the examination system, or were they espoused independently by critics and connoisseurs? How are these values represented in the critical discourse of the time, and how do they affect the changes in critical reception of certain calligraphers over the course of the Tang? Although models for stylish calligraphy and for correctly written characters had existed since at least the Latter Han dynasty (25–220), only in the Tang did the importance of calligraphy and orthography to success in the examinations become explicit. As a result, in the case of calligraphy, the government sponsored the study of the style of the “Sage of Calligraphy,” the Daoist aristocrat Wang Xizhi 王羲之 (303–361), as the accepted standard. In the case of orthography, private scholars and government agencies compiled books of correct character-forms for potential examination candidates to follow. One of these is the Ganlu zishu 干祿字書 (Lexicon for Seeking a Salary), which was written by Yan Yuansun 顏元孫 (d. 732) in the early 8th century. Yan Yuansun was the uncle of the renowned Tang calligraphier Yan Zhenqing 顏眞卿 (709–785), who transcribed the Lexicon onto stone steles that were engraved and set up for public edification in the courtyard of the

Source: “Public Values in Calligraphy and Orthography in the Tang Dynasty,” Monumenta Serica 43 (1995): 263–78. © Monumenta Serica Institute, reprinted by permission of Taylor & Francis Ltd, www.tandfonline.com on behalf of Monumenta Serica Institute.

© koninklijke brill nv, leiden, ���9 | doi:��.��63/9789004380196_047

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Public Values in Calligraphy and Orthography

1541

prefectural office when Yan Zhenqing served as prefect in Huzhou 湖州 in 774.1 By virtue of its survival, the Lexicon is now the best known of the Tang manuals of orthography, but there were many others written around the same time.2 Yan Yuansun divided his selection of characters into three categories, which he called “vulgar” (su 俗), “common” (tong 通), and “correct” (zheng 正). Pairs of characters are listed for a particular word, and for each pair, the upper character is designated as “vulgar” or “common” and the lower character “correct.” One point to note is that the “correct” characters are usually the most complex and formal in structure. In his introduction, Yan Yuansun said of his classification scheme: The characters we call “vulgar” all have shallow and recent precedents. They are found only in account books, government correspondence, contracts, and prescriptions. They are not considered elegant words. The ones we call “common” have a long history of use. They may be used for memorials, notes, letters, and legal documents. However, if you wish to write literary prose for the examinations for government posts, it would be much better to choose the “correct” forms of characters. For the characters we call “correct,” there is proof. These may be used to write literature and stele inscriptions, and they will always be appropriate. You should always use “correct” forms for the jinshi 進士 exam.3 Yan Yuansun never stated what his “proof” was for the “correct” forms, but the modern scholar Shi Anchang 施安昌 has determined that his sources were the Shuowen jiezi 說文解字, the Zilin 字林, and the Stone Classics 石經.4 The Shuowen is the earliest extant dictionary of Chinese characters, which was written by Xu Shen 許慎 around the year 100 CE. The Zilin, or “Forest of Characters,” was written by Lü Chen 呂忧 of the Jin 晉 dynasty (265–420). It was considered as important as the Shuowen up through the Tang, but is now lost. The version of the Stone Classics known in the early Tang dynasty was commissioned in 241 by Emperor Shao 少帝 (r. 239–254) of the Wei 魏. He had the Confucian 1  Shi Anchang 施安昌 (ed.), Yan Zhenqing shu Ganlu zishu 顏眞卿書干祿字書 (Beijing: Zijincheng, 1990), p. 96. 2  Shi Anchang, “Tangdai zhengzi xuekao” 唐代正字學考 (On Tang-Dynasty Orthography), Gugong bowuyuan yuankan 故宮博物院院刊 3 (1982), p. 78. 3  Shi Anchang, “Guanyu Ganlu zishu ji qi keben” 關於干祿字書及其刻本 (On the Lexicon for Gainful Employment and its Engraved Editions), Gugong bowuyuan yuankan 1 (1980), p. 70. 4  “Tangdai zhengzi xuekao,” p. 80.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1542

McNair

Classics engraved in stone in three script types and set up in Luoyang 洛陽; this version is no longer extant.5 There are two points to note about the Lexicon. The first is the explicit statement that the material is to be used in preparation for the government examinations. The second is that the sources that constitute Yan’s authority exactly match the curriculum for students in the government’s school of calligraphy. In the early Tang, the government set up seven schools in the capital, under the Directorate of Education.6 One of them was the Calligraphy School, which was charged with teaching students who were the sons of low-ranking officials and commoners. The curriculum was the Shuowen, the Zilin, and the Stone Classics.7 In addition to this technical school, there was also the Hongwenguan 弘文館 (Institute for the Advancement of Literature), which later functioned as the imperial editorial office. Originally, however, it served as a school of calligraphy for sons of high-ranking officials. According to the Xin Tangshu 新唐書 (New Tang History), in 627, the throne summoned 24 sons of capital officials serving in posts of rank five and higher who loved calligraphy. They were attached to the Hongwenguan, where they practiced calligraphy. Calligraphy models were taken from the palace to use for their instruction.8 Emperor Taizong 太宗 of the Tang (r. 626–649) promoted the style of Wang Xizhi as the orthodoxy through a number of activities, one of which was having various types of copies made of the works of Wang Xizhi in the palace collection. One type was the tracing copies produced by the court calligraphers; another was the free-hand copies produced by his high officials Yu Shinan 虞世南 (558–638), Ouyang Xun 歐陽詢 (557–641), and Chu Suiliang 褚遂良 (596–658). Yu Shinan and Ouyang Xun also demonstrated the writing of regular script (kaishu 楷書) at the Hongwenguan.9 One reason why the study of calligraphy was sponsored by the government is that it was one of the four criteria for selection for government office. According to the Xin Tangshu, 5  For reproductions of surviving ink rubbings, see Shodō zenshū 書道全集, 25 vols. (Tokyo: Heibonsha, 1954–1961), vol. 3, pl. 65–67. 6  Ouyang Xiu 歐陽修 and Song Qi 宋祁, Xin Tangshu 新唐書 (rpt. Beijing: Zhonghua shuju, 1975), juan 48, p. 1265. 7  Da Tang liudian 大唐六典 (rpt. Taibei: Wenhai, 1962), juan 21, p. 16. 8  Xin Tangshu, juan 47, p. 1209. 9  Tang huiyao 唐會要 (rpt. Taibei: Shijie shuju, 1963), juan 64, p. 1115.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Public Values in Calligraphy and Orthography

1543

There are four [areas considered in] the method for choosing men [for office]. The first is stature: a physique and appearance that is handsome and imposing. The second is speech: the vocabulary and diction to debate the truth. The third is calligraphy: a regular script that is powerful and beautiful. The fourth is judgment: logical writing that is excellent and strong. Success in all four areas, then, is the prerequisite for virtuous conduct [in office].10 The philosophical foundation for this method of assessment is characterology, the belief that the style of the inner being and the outer man is unitary and that moral character can therefore be deduced from an examination of any one or more external manifestations, such as appearance, behavior, or aesthetic endeavor. The practice of characterology in China can be documented from the Han dynasty onward; its principal use has been to assess the virtue and ability of candidates for government office.11 The earliest conception of characterology was a simple equation of style and personality: “seeing the man in his writing.” Concomitant to this belief, the notion arose that any attempt to disguise the natural expression of the personality through imitation of another’s calligraphic style was wrong. Zhao Yi 趙壹 (fl. c. 168–190) wrote: Now of all men, each one has his particular humours and blood, and different sinews and bones. The mind may be coarse or fine, the hand may be skilled or clumsy. Hence when the beauty or ugliness of a piece of writing must depend both upon the mind and the hand, can there be any question of making (a beautiful writing) by force of effort? When there are (natural) degrees of difference in beauty and ugliness in people’s faces, how can one strive to look like someone else?12 The fifth-century calligraphy connoisseur Wang Sengqian 王僧虔 (426–485) wrote, “your mind must lose itself in the brush and your hand lose itself in the

10   Xin Tangshu, juan 45, p. 1171. 11  See John Timothy Wixted, “The Nature of Evaluation in the Shih-p’in (Gradings of Poets) by Chung Hung (AD 469–518),” in Theories of the Arts in China, ed. by Susan Bush and Christian Murck (Princeton: Princeton University Press, 1983), p. 228. 12  Translation from Fei cao shu 非草書 by William R. B. Acker, Some T’ang and Pre-T’ang Texts on Chinese Painting (Leiden: E.J. Brill, 1954), p. lvi.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1544

McNair

writing, so that your mind and hand will convey your feelings.”13 The implication is that it is wrong to obscure the reading of one’s personality through an unnatural or imitative style of calligraphy. Obviously, the fact that these men inveighed against the practice of obscuring or elevating one’s character by imitating the style of another meant that the practice was thriving. In terms of representing himself in an assessment for government office, the only quality the candidate could afford to demonstrate was virtue. By the Tang, characterological notions about the revelation of virtue through calligraphy had two aspects. The first was the old notion that a man’s virtue would automatically reveal itself in his calligraphy. The wellknown story involving the statesman and calligrapher Liu Gongquan 柳公權 (778–865) illustrates this notion that a man’s virtue will of necessity manifest itself through his calligraphy. According to the Tangshu 唐書 (Tang History), Emperor Muzong 穆宗 (r. 820–824) once asked Liu Gongquan about the proper method for the brush. Liu replied, “The use of the brush lies in the heart. If the heart is upright, the brush will be upright.”14 The second aspect of characterology in Tang calligraphy was the belief that virtuous people have a moral obligation to display their calligraphy, to broadcast their virtue as an example to others. Yan Zhenqing once wrote a memorial asking Emperor Suzong 肅宗 (r. 756–762) to transcribe a heading for a series of Buddhist steles commemorating the emperor’s pious release of living creatures. In the memorial, Yan Zhenqing said, In antiquity, the First August Emperor of Qin 秦始皇帝 [r. 221–210 BCE] was a cruel and tyrannical ruler and Li Si 李斯 [d. 208 BCE] an evil and fawning minister, yet the bronzes and stones they had engraved have been handed down to later generations. Emperor Wen 文帝 of Wei 魏 [r. 220–226] was a lord who gained his throne from another clan and Zhong You 鍾繇 [151–230] was an official who was partial to a particular clique, yet they also set up a stele extolling the emperor’s virtue. How much the worse that Your Majesty’s towering meritorious achievement should go unrecorded!15

13  Wang Sengqian 王僧虔, Bi yi zan 筆意贊, in Lidai shufa lunwenxuan 歷代書法論文選, ed. by Huang Jian 黃堅, 2 vols. (Shanghai: Shanghai shuhua, 1979), vol. 1, p. 62. 14  Liu Xun 劉珣 et al., Jiu Tangshu 舊唐書 (rpt. Beijing: Zhonghua shuju, 1975), juan 165, p. 4310. 15  “Ponds for the Release of Living Creatures throughout the Subcelestial Realm,” Yan Lugong ji 顏魯公集, ed. by Huang Benji 黃本驥 (rpt. Taibei: Zhonghua shuju, 1970), 1:6a. See also

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Public Values in Calligraphy and Orthography

1545

Yan’s request for the emperor’s handwriting makes clear that he believed the emperor’s virtue should not only be described in the inscription, but should also be displayed through his calligraphy. Yan’s contrasting of the bad emperors of the past with virtuous Emperor Suzong further implies a contrast between their bad ministers and virtuous Yan Zhenqing, suggesting that a display of Yan’s calligraphy would also be a force for good in the world. The trend toward the mandatory display of correctness in public calligraphy was echoed in the trends in orthography and calligraphy over the course of the Tang. According to Shi Anchang’s statistical research, the use in stele inscriptions and epitaphs of those character-forms that Yan Yuansun called “common” or “vulgar” dropped dramatically over the course of the seventh and eighth centuries. A typical example of late Sui dynasty orthography is found in the anonymous Epitaph for Lady Ji 太僕卿元公夫人姬氏墓誌銘, which is dated to 615. “Vulgar” and “common” characters account for 12 percent of the total. By contrast, an early Tang work such as the Epitaph for Huangfu Dan 皇甫誕碑, written out by Ouyang Xun around 637, which dates from only some 30 years after the Epitaph for Lady Ji, contains only eight percent “vulgar” and “common” characters. Even greater improvement is seen in mid-Tang orthography, as in Yan Zhenqing’s Duobaota ganyingbei Inscription 多寶塔感 應碑 (Prabhūtaratna Pagoda Gratitude for Prayers Answered), dated to 752. As one might expect, given Yan Zhenqing’s family background, this inscription contains only 2.4 percent “vulgar” and “common” characters (Figure 1). Liu Gongquan’s Xuanbi Pagoda Inscription 玄秘塔碑 of 841 contains only 3.8 percent “vulgar” and “common” characters, while his Epitaph for Li Sheng 李晟碑 of 829 contains even fewer. A rate of around three percent continued to hold for the remainder of the Tang and afterward.16 This dramatic plunge from 12 percent in the Sui to three percent by the mid-Tang, revealing the broad acceptance of orthography in Tang public monuments, is a vivid demonstration of Tang public values. The function of orthography was to confirm correct forms, based on established authorities. The values that orthography reveals are orthodoxy, formality, and the authority of antique models. Orthodoxy is seen in the rectification of correct forms, formality in the use of the most complex forms, and the authority of antique models in the reliance on the Shuowen, Zilin, and Stone Classics. These same values were also made manifest in public calligraphy from the early to mid-Tang. Most public inscriptions in the Tang dynasty were executed my dissertation, “The Politics of Calligraphic Style in China: Yan Zhenqing (709–785) and the Song Literati” (Ph.D. diss., University of Chicago, 1989), p. 118. 16  Shi Anchang, “Tangdai zhengzi xuekao,” p. 82.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1546

McNair

in regular script. As the model promoted in the capital was the writing of Wang Xizhi, we might expect to see a certain uniformity in the calligraphic styles of the generations following the institution of the Hongwenguan. To see this, however, we need a sense of what was considered the style of Wang Xizhi’s regular script in early Tang. Unfortunately, not many regular-script works attributed to Wang Xizhi survive. Chu Suiliang’s catalog of the imperial collection lists three that are extant in some form today: Yue Yi lun 樂毅論 (On General Yue Yi), the Huangtingjing 黃庭經 (Yellow Court Scripture), and the Dongfang Shuo hua zan bei 東方朔畫贊碑 (Encomium on a Portrait of Dongfang Shuo).17 On General Yue Yi may have been the most important of these works as a model; we know, for instance, that Emperor Taizong ordered the Hongwenguan official Feng Chengsu 馮承素 to make copies of it to be given to six of his high-ranking officials.18 The Yellow Court Scripture was also used as a model. There are copies attributed to the Sui Buddhist monk Zhiyong 智永, who was a descendant of the Wang clan and Yu Shinan’s teacher, and others attributed to Ouyang Xun, Yu Shinan, and Chu Suiliang.19 As for the Encomium on a Portrait of Dongfang Shuo, the original was supposedly interred in the tomb of Wang Xiu 王修 (334–357), a younger relative of Wang Xizhi. If this anecdote were true, the piece in the imperial collection could only have been a first-generation copy at best.20 Yet it seems scarcely less important than On General Yue Yi and the Yellow Court Scripture, for it survives today, as they do, in Tang ink-written copies and post-Tang ink rubbings. The other two scrolls of regular-script works recorded by Chu Suiliang, one of which included the famous Gaoshiwen 告誓文 (Announcement of an Oath), are now lost. Works of questionable authenticity such as these were probably all that could be seen by Tang times, yet they would have been the models brought out from the archives for the calligraphy students to study. If the Encomium (Figure 2), for example, represented the “Wang style” of regular script in Tang times, certain stylistic qualities it has in common with On General Yue Yi and the Yellow Court Scripture may be discerned: character compositions that fan out to the right, sharply pointed strokes made with the exposed brushtip 17  Chu Suiliang, Youjun shumu 右軍書目, in Fashu yaolu 法書要錄, ed. by Zhang Yanyuan 張彥遠 (rpt. Shanghai: Shanghai shuhua, 1986), 3, p. 70. For reproductions, see Shodō zenshū, vol. 4, pls. 1–9. 18  See Chu Suiliang, “Record on the Copies of On General Yue Yi,” in Fashu yaolu, 3, p. 105. 19  See Yang Zhenfang 楊震方, Bei tie xulu 碑帖敘錄 (Shanghai: Shanghai guji, 1988), pp. 151–152. 20  See Lothar Ledderose, Mi Fu and the Classical Tradition of Chinese Calligraphy (Princeton: Princeton University Press, 1979), p. 71.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1547

Public Values in Calligraphy and Orthography

technique, and highly-modulated, bottom-heavy final strokes. Those calligraphers who were educated in Chang’an 長安 in the mid-7th to early 8th centuries should demonstrate elements of this style, if the vigorous state promotion of the “Wang style” were at all effective. The epitaph for the esoteric Buddhist monk Amoghavajra (705–774), for example, was written out by the high official Xu Hao 徐浩 (703–782) in Chang’an in 781 (Figure 3). His style exhibits the hallmarks of the Wang style: the “left tight, right loose” character compositions, the sharp-tipped strokes, and the highly-modulated, bottom-heavy final strokes. The same “Wang-style” elements are revealed in the early works of Yan Zhenqing, such as the Prabhūtaratna Pagoda Inscription of 752 for the Qianfu Monastery 千福寺 in Chang’an (Figure 1). The use of the “Wang style” of regular script as a public standard reveals the same values that orthography does: orthodoxy, formality, and the authority of antique models. It is important to note that these values were accepted and promoted by both the state and private scholars. Are these same public values also manifested in calligraphy criticism of the mid-Tang? A case study may be made of the critical reception of the group of calligraphers known today as the “Four Masters of Early Tang”: Ouyang Xun, Yu Shinan, Chu Suiliang, and Xue Ji 薛稷 (649–713). In contrast to later opinion, which has canonized these four men, the connoisseurs of the mid-Tang did not see them as an automatic group, and those that did, did not see them as equals. For example, the court calligrapher Zhang Huaiguan 張懷瑾 (fl. c. 713– 742) thought the four prominent calligraphers during Emperor Taizong’s rule were Yu Shinan, Ouyang Xun, Chu Suiliang, and Lu Jianzhi 陸柬之 (a nephew of Yu Shinan).21 Cai Xizong 蔡希綜 (fl. c. 742–756) listed the ten important calligraphers of early Tang as the high officials Fang Xuanling 房玄齡, Du Ruhui 杜如晦, Yang Shidao 楊師道, Pei Xingjian 裴行儉, Gao Shilian 高士廉, Ouyang Xun, Yu Shinan, Lu Jianzhi, Chu Suiliang, and Xue Ji.22 The history of critical reception of the “Four Masters of Early Tang” points up how extra-artistic factors are as important as the aesthetic qualities of calligraphic styles. These factors include characterology and the public values of orthodoxy, formality, and the authority of antique models. In a recent article, Stephen J. Goldberg has stated that the styles of Ouyang, Yu, and Chu were promoted by the imperial family during the early Tang because “the classical ideal of a bilaterally symmetrical, hierarchical structure, particularized and concretized in the specific styles of early T’ang calligraphy” was the means by which

21   Pingshu yaoshi lun 評書藥石論, in Lidai shufa lunwenxuan, vol. 1, p. 231. 22   Fashu lun 法書論, in Lidai shufa lunwenxuan, vol. 1, p. 270.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1548

McNair

the ruling class expressed its will for “great harmony in government.”23 Regular script, no matter whose style it is in, does not manifest bilateral symmetry; true bilateral symmetry is found only in seal script (zhuan 篆). As is clear from the documentary record, the throne did not patronize imperious forms, but rather the Wang style, which had constituted the elite southern-gentry manner for over 200 years and had been imperially sponsored during the Six Dynasties and Sui periods. This is attested to in the public inscriptions in the hand of Emperor Taizong that have come down to us. His Hot Spring Inscription 溫泉銘, in running script (xingshu 行書), resembles the style of Yu Shinan’s Jishi Letter 積時帖 while his inscription at Jinci 晉祠 closely imitates the running script of Wang Xizhi as seen in the version of the Preface to the Buddhist Canon 聖教序, compiled from Wang characters by Huairen 懷仁 in 671.24 The throne achieved its desired legitimacy by sponsoring calligraphers who were said to follow the Wang style and who had clear political ties to the court; Ouyang, Yu, and Chu fit these criteria. Goldberg is much closer to the mark with his view that there is a political message implicit in the viewer’s assumption that he is seeing a normative standard validated by the calligrapher’s involvement in court affairs.25 This view allows for the effects of characterology, biography, and stylistic affiliations on the history of critical judgment. Both Ouyang Xun and Yu Shinan served as high court officials under the Sui dynasty and Emperors Gaozu 高祖 and Taizong of the Tang. Both were said to have followed the style of Wang Xizhi and served as connoisseurs of calligraphy for Emperor Taizong. Chu Suiliang, by contrast, was recommended to Tang Taizong as a connoisseur by Wei Zheng 魏徵 (580–643) only after the death of Yu Shinan. Although he subsequently enjoyed a long career as a high-ranking court official and close confidant of the emperor, he died in exile, accused of plotting to stage a revolt over the elevation of Wu Zhao 武瞾 to empress. According to the earliest calligraphy text that mentions him, Chu first studied the style of Yu Shinan and then the Sui dynasty regular-script specialist Shi Ling 史陵.26 Only as a mature man did he study the style of Wang

23  Stephen J. Goldberg, “Court Calligraphy of the Early T’ang Dynasty,” Artibus Asiae 49, no. 3/4 (1988–1989), p. 233. 24  For a reproduction of the Hot Spring Inscription, see Chinese Calligraphy, ed. by Nakata Yujiro (New York/Tokyo: Weatherhill & Tankosha, 1983), pl. 35; for the Jishi Letter, see Shodō zenshū, vol. 7, pl. 78–79; for the Jinci inscription, see Shodō zenshū, vol. 7, pl. 86–89; and for Huairen’s compilation, see Chinese Calligraphy, pl. 36. 25  “Court Calligraphy of the Early T’ang Dynasty,” p. 232. 26  Li Sizhen 李嗣眞 (d. 696), Shu hou pin 書後品, in Lidai shufa lunwenxuan, vol. 1, p. 134.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Public Values in Calligraphy and Orthography

1549

Xizhi.27 His follower, Xue Ji, was a grandson of Wei Zheng, who served as a court official under Empress Wu (r. 690–705) and Emperors Zhongzong 中宗 (r. 684, 705–710) and Ruizong 睿宗 (r. 684–690, 710–712), and who worked as a connoisseur for Imperial Son-in-law Wu Yanxiu 武延秀 and afterwards for Emperor Ruizong. But he was implicated in the attempted coup of the Taiping Princess 太平公主 (d. 713) and allowed to commit suicide at the beginning of Emperor Xuanzong’s 玄宗 reign. Xue Ji was said to have learned calligraphy from the works of Yu Shinan and Chu Suiliang in his grandfather’s collection, and his style has traditionally been considered most in debt to Chu Suiliang. Goldberg argues that the regular-script styles of Ouyang Xun, Yu Shinan, and Chu Suiliang together represent “the dominant paradigm of aesthetic norms, artistic conventions, and social values traditionally associated with the classical style of the T’ang dynasty.”28 The unique contribution of Chu Suiliang may be found in his Preface to the Buddhist Canon (Figure 4), a court-sponsored stele inscription set up in the Goose Pagoda 雁塔 of the Daci’en Monastery 大慈恩寺 in Chang’an, in which Chu synthesized the styles of Yu Shinan and Ouyang Xun, incorporating their “outer plasticity and inner tensile strength,” respectively.29 This ahistorical view of the reputation of Chu Suiliang and his calligraphic style gives the impression that his style was always considered “classical” and an improvement on the styles of Ouyang and Yu. However, the writings of the mid-Tang critics make it clear they considered his style neither classical nor even as good as the styles of Ouyang and Yu. Xu Hao had the following to say: In the recent past, Ouyang Xun and Yu Shinan have had their brush force handed down. But Chu Suiliang and Xue Ji have already fallen. The saying “Yu has the sinews, Chu has the flesh, and Ouyang has the bones,” is apt indeed. Hawks and falcons lack color, yet they fly up to heaven. Because their bones are strong, their spirit is fierce. Ringed pheasants and Tartar pheasants may show every hue, yet they hover but a hundred paces. Because their flesh is heavy, their strength is weighed down. Were there an ornate and brilliant bird that could also soar to the heights, such would be a phoenix of calligraphy! In calligraphy, Ouyang Xun and Yu Shinan are the hawk and falcon, and Chu Suiliang and Xue Ji are the ringed pheasant and Tartar pheasant.30 27  Zhang Huaiguan, Shuduan 書斷, in Fashu yaolu, p. 225. 28  “Court Calligraphy of the Early T’ang Dynasty,” p. 190. 29  Ibid., p. 230. 30   Lun shu 論書, in Fashu yaolu, 3, p. 92.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1550

McNair

By means of this elaborate metaphor (borrowed from the realm of literature31), Xu Hao disparaged the style of Chu Suiliang and his follower, Xue Ji, for its emphasis on a modulating brushstroke (that is, “flesh”) at the expense of the structure of the characters (that is, “sinews and bones”). See, for example, the fifth character in the third column of Figure 4 (mo 莫). Here, the lower element of the character, composed of a long horizontal stroke with two shorter strokes splayed unevenly to either side, is given a dramatic sense of centrifugal movement and modulation of line that sacrifices any cohesion at the core of the character. Zhang Huaiguan also remarked on this centrifugal quality to Chu Suiliang’s writing. He noted that Chu had once studied with Shi Ling. Of Shi’s style (and perhaps Chu’s by association) he wrote, “Shi had an antique straightness, but was harmed by his scattering of slender strokes.”32 Zhang also agreed with Xu Hao’s assessment of Xue Ji. He assigned Xue Ji to the “capable class,” the lowest of his three rankings, and remarked of him that he “obtained the fleshiness of Chu Suiliang’s style and thus could be called Chu Suiliang’s best pupil.”33 This was not a positive assessment. Elsewhere, Zhang wrote, “horses with much sinew and little flesh are considered superior, while those with much flesh and little sinew are inferior. Calligraphy is also like this.”34 He also noted that Xue Ji was executed for a crime, the coup attempt by the Taiping Princess against Zhang’s patron, Emperor Xuanzong. Another important mid-Tang critic, Dou Ji 竇臮 (d. c. 769), agreed with the assessments made by his contemporaries, Xu and Zhang. He said that of all early Tang calligraphers, Ouyang Xun and his son Ouyang Tong 歐陽通 (d. 691) stood at the head, while Yu Shinan was “transcendent and wielded the brush as if divine.” But of Chu Suiliang, he wrote, He humbly attended to [Wang Xizhi’s] Gaoshi, and vigorously applied himself to this beautiful writing. I am afraid, however, that he had no 31  Xu Hao borrowed this imagery from Liu Xie 劉勰 (465–522): “Ringed and Tartar pheasants show every hue, yet they hover but a hundred paces. Because their flesh is heavy, their strength is weighed down. Hawks and falcons lack color, yet they fly up to heaven. Because their bones are strong, their spirit is fierce…. Only when [a literary work] is ornate and brilliant and can soar to the heights do we have a singing phoenix in the world of literature” (adapted from The Literary Mind and the Carving of Dragons, by Liu Hsieh, translated and annotated by Vincent Yu-chung Shih [Taipei: Chung Hwa Book Co., 1970], p. 229). 32   Shuduan, 3, p. 226. 33  Ibid., 3, p. 238. 34   Pingshu yaoshi lun, in Lidai shufa lunwenxuan, vol. 1, p. 229.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Public Values in Calligraphy and Orthography

1551

more success than trying to paint a tiger [only to have it resemble a dog], the same sort of thing as [her ugly neighbor] frowning to imitate the beauty of Xi Shi 西施. Though he was thought much of by the gentry and his name was famous in China and beyond, his students had a degenerate manner, and must he not take responsibility for that?35 Dou Ji also derided Lu Jianzhi as a mere imitator of his uncle Yu Shinan and Xue Ji as one who had modeled himself on the style of Chu Suiliang, while “doubling him in showiness.” Elsewhere, he noted that “in his own day, it was allowed that Xue Ji’s calligraphy was lovely and plump.”36 He also stated that the styles of Chu and Xue were valued in the early Kaiyuan 開元 era (713–742), which suggests they were not as highly valued afterwards.37 Clearly the opinions of these three mid-eighth century critics indicate a severe distaste for the styles of Chu and Xue, even while acknowledging their historical standing. What are the public values these critics reveal? The regular script that Ouyang Xun and Yu Shinan used on public monuments was scrupulously true to script-type, they were said to have studied the classical tradition of Wang Xizhi, and they were high-ranking officials under a cultural standardizer, Emperor Taizong. The “bone and sinew” of Ouyang and Yu are metaphors not only for their calligraphic style, but also for their lives as men and their calligraphic lineage. In traditional Chinese physiology, bone and sinew are the concealed structures that organize movement and strength.38 Like the Confucian gentleman who contains his virtue in his person, they do not display themselves on the surface. Their regularity of structure and the inherent virtue of their affiliations accord with the public values of orthodoxy, formality, and the authority of antique models. By contrast, Chu Suiliang and Xue Ji mixed more informal running-script elements into their regular script and loosened the structure of their characters. They were not primarily influenced by the style of Wang Xizhi, since Chu Suiliang was a student of Yu Shinan’s writing, while Xue Ji was a student of Chu Suiliang’s. Xue Ji, moreover, had lived during the reign of Empress Wu 武后, who officially promoted certain unorthodox character forms, and he was executed for alleged involvement in a treasonous plot. The “flesh” of Chu Suiliang and Xue Ji, then, was not only a metaphor for their calligraphic style, but for 35   Shu shu fu 述書賦, in Lidai shufa lunwenxuan, vol. 1, p. 255. 36  Ibid., vol. 1, p. 257. 37  Ibid., vol. 1, p. 258. 38  See John Hay, “The Human Body as a Microcosmic Source of Macrocosmic Values in Calligraphy,” in Theories of the Arts in China, pp. 74–102.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1552

McNair

their status as epigones, their lesser standing as admirable political figures, their distance from the Wang style, and their violations of orthography. In traditional Chinese physiology, the flesh is what gives a body its outward shape or appearance. It is the source for superficial attractiveness, which in traditional thought was considered almost a guarantee of lack of virtue. The metaphors of “bone, sinew, and flesh” as critical terms illustrate the public values of the Tang that are also seen in calligraphy, orthography, and beliefs about characterology. Only later, when expressiveness, informality, and learning from contemporaries became acceptable again, did connoisseurs consider Chu Suiliang and Xue Ji as models of the same rank as Ouyang Xun and Yu Shinan. The canonization of the “Four Masters of Early Tang” appears to have taken place in the late Northern Song, among those critics who were least attached to Tang standards of judgment. The historian Zhu Changwen 朱長文 (1039–1098), in his history of calligraphy, did not list the four together, but quoted a “Tang saying” that “first there were Yu [Shinan] and Chu [Suiliang], then there were Xue [Ji] and Wei [Shuyu 魏叔瑜, a son of Wei Zheng].”39 This seems consistent with the varied groups chosen by Tang critics such as Zhang Huaiguan and Cai Xizong. But Ouyang, Yu, Chu, and Xue do appear together in a chapter given over to them in the engraved model-letters compendium Ru tie 汝帖, which was compiled by the innovative scholar-official Wang Cai 王寀 (1078–1118) in 1109.40 The group is then cited consistently in calligraphy texts from the early 14th century onward, beginning with Zheng Shao’s 鄭杓 Yan ji 衍極 and continuing with Ming dynasty texts such as Xie Jin’s 解縉 Chunyu zashu 春雨雜述, Feng Fang’s 豐坊 Shu jue 書訣, and Dong Qichang’s 董其昌 Huachanshi suibi 畫襌室隨筆.41 It seems that by the Northern Song enough time had passed to loosen the grip of the mid-Tang critics’ judgments concerning the relationship of calligraphy to public life. Su Shi 蘇軾 (1036–1101) said, in reference to Chu Suiliang, The critics of the past always joined [his calligraphy] with a discussion of his life. Were it not for his life as a man [being linked to the assessment of his calligraphy], however, they could have simply called him skilled but not noble.42 39   Xu Shuduan 續書斷, in Lidai shufa lunwenxuan, vol. 1, p. 339. 40  See Rong Geng 容庚, Cong tie mu 叢帖目, 4 vols. (Hong Kong: Zhonghua shuju, 1980– 1986), vol. 1. pp. 108–116. 41   Lidai shufa lunwenxuan, vol. 1, pp. 423, 500; vol. 2, pp. 504, 508, and 547, respectively. 42   Dongpo tiba 東坡題跋, in Yishu congbian 藝術叢編 (Taibei: Shijie shuju, 1962). vol. 4, p. 92.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Public Values in Calligraphy and Orthography

Figure 1

1553

Yan Zhenqing, Prabhūtaratna Pagoda Gratitude for Prayers Answered inscription, 752 CE, ink rubbing.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1554

Figure 2

McNair

Wang Xizhi, attributed, Encomium on a Portrait of Dongfang Shuo, 356 CE, ink rubbing.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

Public Values in Calligraphy and Orthography

Figure 3

1555

Xu Hao, Epitaph for Amoghavajra, 781 CE, ink rubbing.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access

1556

Figure 4

McNair

Chu Suiliang, Preface to the Buddhist Canon, 653 CE, ink rubbing.

Paul W. Kroll - 978-90-04-38019-6 Downloaded from Brill.com11/09/2020 12:21:35PM via free access