2

The Critic as Reader

EVERY CRITIC, whether intralingual or interlingual, plays a dual role, as reader and as author: vis-à-vis the original author, the critic is a reader, but vis-à-vis the critic’s own reader, he assumes the position of the author and his critical work replaces the original work for the time being, while the world created by the original author in the work replaces the real world. Actually, the interrelations between world, author, work, and reader are more complex than suggested by the circular diagram, which has been offered merely for the sake of convenience. If we enquire further into these interrelations, we shall find that the real world (including both the natural world and the cultural world), the author’s lived world, the world created by the author in the work, the world re-created by the reader, and the reader’s lived world form a daisy chain of intersecting circles, which can never be proven to be identical yet must intersect each other in order for any communication to take place. The critic’s task as reader is to try to increase the area of intersection between the created world of the work and the world that he is re-creating from the linguistic structure of the work. How far he will succeed depends, among other things, on how much he knows about the author’s cultural world and how far his own lived world resembles that of the author’s. This does not mean that only someone who has lived in the same cultural world as the author and has had the same experiences can successfully read a poem, since obviously even two brothers who had the same background may react differently to the same environment, but it remains true that a knowledge of the author’s cultural world and a sympathetic understanding of his created world are both necessary for a successful reading of a poetic work.

Hence, to be a competent reader of Chinese poetry requires not only linguistic competence (knowledge of the vocabulary, syntax, phonology, and prosody of a given period and a given genre) and historical knowledge, but also the capacity to make the imaginative leap across temporal and spatial barriers into the world created by the author. Linguistic competence and historical knowledge are necessary but not sufficient conditions for a successful reading of poetry as something more than historical documents or linguistic data. (I am not repeating the old opposition between “reading literature as literature” and “reading literature as social documents,” which John M. Ellis has shown to be a false opposition,¹ but asserting that poetry is something more than historical documents or linguistic data.) Moreover, just as a connoisseur of wine may choose to talk about the bouquet, color, body, and taste of a vintage wine rather than the soil and climate where the vine grew and the kind of fertilizer used, so may a critic choose to talk about the linguistic structure and aesthetic qualities of a poem rather than the circumstances in which the poet lived and wrote. One more consideration may be brought forward: the distinction between a literary historian and a literary critic. Although a literary historian should also have critical discernment and a critic should also have historical awareness, the concerns and methods of the two are not identical. The former is by definition committed to a diachronic view of literature; the latter may, and sometimes must, take a synchronic view, for otherwise it would be impossible to establish any critical standards that are not historically relative. We shall pursue the question of historical relativism further in chapter 4.

The question “Who is a competent reader?” is not as easy to answer as “Who is a competent speaker?” Whereas any normal adult speaker of a language can be presumed to be a competent speaker of that language, no simple criterion exists for competence as a reader of poetry. The question is a particularly sensitive one with regard to classical Chinese poetry, since many who write in English about Chinese poetry are not native speakers of Chinese, and some do not even speak any Chinese at all. Such people will probably argue that no one is a “native speaker” of classical Chinese. Although I do not deny this, I think it is legitimate and useful to distinguish “native readers” from “non-native readers.” By a “native reader” of classical Chinese poetry I mean someone who was born and brought up in China, speaks some modern Chinese dialect as his native language, and has been reading and writing classical Chinese since childhood; by a “non-native reader” I mean someone whose native language is not Chinese and who has learned to read classical Chinese as an adult. (Those who fall into neither category may be considered “near-native” or “semi-native” readers, as the case may be.) Both kinds of readers can become competent readers of Chinese poetry, but each has certain advantages as well as disadvantages. Let us consider these.

The native reader naturally enjoys some advantages over the non-native reader, both linguistically and culturally. Linguistically, classical Chinese is, after all, not a totally different language from modern Chinese in any of its dialectal forms. I imagine no one would deny that a native speaker of Italian or French has a natural advantage over a native speaker of Chinese or Japanese when it comes to reading Latin, and I see no reason to pretend that a native speaker of modern Chinese does not enjoy a comparable advantage over a native speaker of an Indo-European language when it comes to reading classical Chinese. Since classical Chinese shares with modern Chinese the same script and, to some extent, the same vocabulary, though differing in syntax, a native speaker of modern Chinese is unlikely to make the kind of elementary mistake that a non-native reader may, such as taking Chijiao Daxian to mean “The Red-legged Immortal” instead of “The Bare-footed Immortal,” as a famous English translator of Chinese literature did. Also, a native reader tends to respond intuitively to a poem as a whole, rather than trying to decode it word by word. Furthermore, a native reader is likely to be more sensitive to the rhythm and tonal patterns of Chinese poetry than a non-native reader, despite the differences between classical and modern Chinese in actual pronunciation. As for writing classical Chinese, some may argue that the kind of “classical Chinese” that a modern Chinese may write is not the same language that Du Fu wrote, but then Du Fu’s classical Chinese is not the same language as that of the Book of Poetry, just as Milton’s Latin is not the same as Vergil’s. As long as a piece of writing follows the same basic semantic, syntactic, and, in the case of verse, prosodic rules as classical Chinese, it may be considered a work in classical Chinese, of whatever quality. The advantage of being used to writing classical Chinese and not merely reading it is that one acquires a feel for the medium and can appreciate the subtleties and difficulties of writing classical Chinese poetry, or what we call in Chinese the “sweetness and bitterness therein” (qizhong ganku), from the inside, as it were, just as someone who has been singing all his life, no matter what an indifferent singer he may be, is in a better position to appreciate good singing than someone who has never sung a note in his life, no matter how much he may know about music. Culturally, the native reader has the natural advantage of having lived in a cultural world somewhat closer to that of the classical Chinese poet than a non-native reader. Of course, modern China is not the same as ancient China, but the two still have more in common with each other than either with the Western world, past or present. For instance, to someone who grew up in Peking before the city walls were torn down, the word cheng would immediately convey the double meaning of “city wall” and “walled city,” with definite visual connotations, which, to someone who grew up in Manhattan, may not be readily perceptible. The same is true of traditional beliefs, assumptions, legends, folklore, and so forth, which a native reader has assimilated unconsciously from childhood, but a non-native reader has to learn consciously.

The native reader of Chinese poetry also has certain disadvantages. To begin with, as a native speaker of modern Chinese he may easily assume that a word or phrase in classical Chinese means the same as in modern Chinese, which is often not the case. Therefore, he has to be constantly on guard against such assumptions and remind himself of the differences between classical and modern Chinese. Second, having been brought up on standard authors and works, his taste can easily be shaped by traditional standards, and it takes a conscious effort to form independent tastes and judgments. Finally, since he more or less knows “what there is” in Chinese poetry, there is little likelihood of discovering a poet or a work that he has never heard of before. Reading poetry for such a reader is a gradual process of maturation, of rereading familiar works in the hope of refining one’s taste, deepening one’s understanding, and modifying one’s judgments, rather than an adventure in intellectual and imaginative exploration.

The non-native reader has corresponding disadvantages and advantages, compared with the native reader. His linguistic and cultural disadvantages are obvious and need to be overcome by effort. However, basically, the difference between reading poetry in one’s native language and reading poetry in a foreign language is one of degree rather than kind: even in the former case one still has to make an imaginative leap into a world different from one’s own. To a twentieth-century American, Shakespeare’s England and Shakespeare’s English are less remote than Tang China and classical Chinese, but they still require both consciously obtained knowledge and imaginative effort to be wholly accessible. As for the advantages, the non-native reader is less bound by the Chinese tradition and therefore more free to develop his own tastes and judgments, and he is able to discover new poets and poems for himself, in the same way that I “discovered” Blake for myself. Reading Chinese poetry for a non-native reader can be an exciting experience, which may open up new worlds and new modes of expression. Moreover, since no one uninterested in poetry in his native language is likely to develop an interest in poetry in another language, non-native readers of Chinese poetry may be presumed to be familiar with poetry in English and/or some other language or languages, and should be able to view Chinese poetry in a comparative perspective.

Since in this chapter we are concerned with the critic as reader, we need to consider what reading involves. Some literary theorists and critics, such as Roman Ingarden, Georges Poulet, Wolfgang Iser, and Paul Ricoeur, have dealt with the phenomenology of reading in detail. It is not my intention either to offer an entirely new theory of reading or to repeat or summarize what others have written. I shall simply describe what I think reading, especially reading poetry, entails. As suggested in the previous chapter, reading poetry is not a passive experience of receiving a message or being affected by the work, but an active experience of re-creation. At this point I would like to voice my agreement with those linguists, philosophers of language, and literary theorists who conceive of “meaning” not as an object, not even an intentional object in the phenomenological sense, but as an act. Therefore, instead of asking, “What does this poem mean?” we should rather ask, “What does the author mean by this poem?” To answer this question, we should then ask, “What would I mean by this poem if I were writing it?” This, of course, leads to the question of identification with the author, which we shall consider again in a moment. Meanwhile, let us clarify the interrelations between “world,” “meaning,” “text,” and “poem,” with regard to the author in the process of writing, and to the reader in the process of reading.

In the preceding chapter I said that a poem is both referential and self-referential. In its referential aspect, a poem “refers” to a world, by which I do not mean that it merely points to a world, but that it embodies or incarnates a world in its linguistic structure, which, when written down or printed, is the “text.” From the author’s point of view, “meaning” is an intentional act, which has as its object the poem as a work of art, whereas the text or linguistic structure is only the medium, not the intentional object, as sometimes thought. For the benefit of readers not familiar with phenomenology, it should be pointed out that the terms “intend” and “intentional” as used by phenomenologists do not have their usual meanings in English: to “intend” is to direct one’s consciousness toward an object, and the object of such an act is an “intentional object.” From the reader’s point of view, what he confronts is the text or linguistic structure, which enables him to “re-mean” (at least to some extent) what the author meant, thereby to perceive the “poem” as an aesthetic object, and to re-create the world embodied in the poem.

As Wolfgang Iser has realized, the “poem” (which is both a “work of art” and an “aesthetic object”) is the result of an intèraction between the “text” and the reader’s imagination.² This does not mean that there are as many “poems” as there are readers, but that there are as many “concretizations” of the same poem as there are readings (for each reading by the same reader may be different), yet there must be a common core among all the different readings.

So far we have been considering the process of reading poetry with regard to its referential aspect. However, we should not forget its self-referential aspect. In referring to itself, the linguistic structure of a poem enables the reader to satisfy his own creative impulse. This brings us back to the question of identification with the author. I think that the reader identifies with the author in two ways. First, the reader identifies with the author qua speaker of the poem, by “bracketing” (in the Husserlian sense, namely, suspending judgment on) his own lived world and entering the world created by the author. Of course, a reader, being human, cannot totally escape from his own historicity, but it is possible as well as necessary to suspend one’s normal beliefs and assumptions and adopt those of the author (or, more strictly speaking, of the speaker). For instance, we need to suspend our belief that the earth is round when reading a poem in which the earth is assumed to be square. It is through the willing suspension of our beliefs (to give Coleridge’s phrase a twist) that we can identify with the author qua speaker, for to write is to execute a fiat, and to read is to accept a covenant. The author decrees, “Let there be such-and-such,” and the reader has to accept such-and-such as true for the time being, whatever he may believe in normal life. Second, the reader identifies with the author qua author (that is, meaning-giver and creator), by repeating the process of writing, meaning, and creating, as if the reader were now writing, meaning, and creating. It is through this second kind of identification that the reader can share the author’s joy of creativity.

Paradoxically, although the reader identifies with the author in the ways suggested above, at the same time he is conscious of the author as an Other speaking. For example, when I am reading Du Fu, I do not literally think that I am Du Fu but only feel as if I were Du Fu speaking, while at the same time I am also conscious of the voice as Du Fu’s. Now, if we compare the situation of reading with that of conversation, we seem to encounter a problem. As Husserl observed, in conversation we can only understand the meaning of the word “I” from the living utterance and the circumstances that surround it, and if we read the word without knowing who wrote it, it is perhaps not meaningless but at least estranged from its normal sense.³ He also pointed out that the conceptual meaning of “I” as “whatever speaker is designating himself” is not the same as what we take the word to mean in an actual utterance.⁴ In other words, when we hear the first person pronoun, we take it to mean the person who is speaking, not the abstract concept of “whatever speaker is designating himself.” When we turn to Chinese poetry, although the first person pronoun is often omitted, there is still an implied speaker in every poem. Does this mean that we can only understand a poem when we know the identity of the author and the circumstances in which he wrote it? Some scholars who adhere to the historico-biographical approach might answer in the affirmative, but the plain fact that we do understand anonymous poems makes such an answer untenable. To solve this problem we need to recognize three distinctions: first, between the author and the speaker; second, between the specified or intended reader and the implied reader; third, between the actual context in which a poem is written and what we may call the dramatic context in which the speaker is imagined as speaking.

Of course, an author may speak in his own person or adopt a different persona. When he is speaking in his own person, knowledge about his life can affect our understanding of a poem. For instance, the following poem by Li Shangyin (813?-858) may at first sight appear to be a celebration of the simple rustic life:

However, when one reads in a letter that Li wrote in 849 to Lu Hongzhi, military governor of Wuning, who had invited Li to join his staff, that the poet was “being cramped in a rented snail-cottage and living precariously in a swallow’s nest,” one cannot help suspecting that the title of the poem is ironic: instead of congratulating himself, the poet is really mocking himself and complaining about the straitened circumstances in which he is living. Even if the tone of the poem is not wholly ironic, it is at least ambivalent. On the other hand, the poet seems to enjoy the simple pleasures of a rustic existence and the freedom from political dangers (suggested by the trap); on the other hand, he is far from being content with his living conditions. Without reading the letter, in which the same expressions “snail-cottage” and “swallow’s nest” appear as in the poem, we might have missed the irony and the ambivalent attitude of the poet.⁷

Sometimes, even though the author is the speaker, he may simply act as the narrator, such as in Du Fu’s famous “Stone-moat Village Officer”:

It is not necessary to know anything about Du Fu’s own life in order to understand this poem, since he is acting as the narrator and serving as a bridge between the world of the poem and the world of the reader. Having introduced us to the world of the poem, the poet withdraws and lets the old woman take over the role of the speaker, so that we can learn her pathetic story from her own lips. At the end of the poem, the poet reappears to lead us out of the world of the poem again. The poet’s sympathy for the old woman and her family is implicit in the poem, and we need no extraneous knowledge about Du Fu’s attitude toward war.

When a poet assumes a dramatic persona, such as in the following poem by Li Bo (701-762), it is not necessary to know about his own life:

Chang’an in a swath of moonlight;

From ten thousand houses, the sound of pounding clothes.

What the autumn wind cannot fully blow away

Are all feelings about the Jade Pass.

When will the barbarians be suppressed?

No more distant campaigns for my husband then.

All that the reader needs to know is that the speaker is a woman in the imperial capital Chang’an, thinking of her husband far away guarding the frontier, represented by the Jade Pass (Yuguan). Li Bo’s own biography is irrelevant to our understanding of this poem.

Indeed, it is not always necessary even to identify the sex of the speaker of a poem, such as this anonymous poem from the Book of Poetry (Shijing, ca. 1100-600 B.C.):

The original gives no clear indications whether the speaker is a man or a woman, but this does not affect the way in which the poem successfully conveys the lover’s joy and desire to be with the loved one. The lack of specific details is but one example of what Roman Ingarden called “spots of indeterminacy,”¹³ which are found in all literary works of art and which each reader can fill in according to his imagination.

The question whether a poet is speaking in his own person or assuming a dramatic persona is not always easy to answer. When there are explicit references to persons, events, and places that can be corroborated from other sources, we may assume that the author is speaking in his own person. Nonetheless, some allowance must be made for poetic license, and it would be unwise to identify the created world of the poem with the lived world of the poet. In contrast, when the speaker of a poem is a woman but the author is known to be a man, it is obvious that he is assuming a dramatic persona. Whether he is doing so for allegorical purposes or not must be judged on an individual basis, and the only generalization that can be made is that when a poet assumes a dramatic persona, it is not necessarily for an allegorical purpose but may be simply to dramatize the situation and present it in a more effective way.

Turning to the distinction between the specified reader and the implied reader, we should realize that countless Chinese poems are addressed to individuals, who may be called specified readers. On the other hand, just as every speech act implies a hearer, so does every poem imply a reader, even if it is the author himself, who cannot help being his own first reader. When someone reads a poem, he becomes the actualization of the implied reader. Two questions may be raised here. First, does the actual reader identify with the specified reader? Second, does the actual reader need to know anything about the specified reader? I am not sure about the answer to the first question. I myself always identify with the speaker of the poem, even when I know the author to be a woman. But one woman student told me that when she read Shakespeare’s sonnets as a young girl she imagined them to be addressed to her. In other words she identified with the specified reader rather than with the speaker. Perhaps individuals differ in this regard. As for the second question, I think it is sometimes, though not always, necessary to know who the specified reader was and what his relationship with the author was. For instance, in one of a pair of poems entitled “Dreaming of Li Bo,” Du Fu wrote,

If we did not know that these two men are generally considered the two greatest poets of China and that neither was successful in his official career, the lines would lose much of their poignancy and tragic irony, created by Du Fu’s awareness of his own and Li Bo’s greatness and his disappointment at the failures of both their lives.

As for the distinction between the actual context and the dramatic context of a poem, the former refers to the actual circumstances in which a poem was written, the latter to the dramatic situation in which the speaker is imagined as speaking. Chinese poets often do inform us, in the title or a note or a preface, of the actual context in which a poem was written. Such information can be very helpful to our understanding of a poem. For example, in a preface to a group of poems entitled “Willow Branch,” Li Shangyin tells us how he met a girl named Willow Branch (Liuzhi), who admired his poetry, how he missed a rendezvous with her, and how she became a nobleman’s concubine.¹⁶ Without this preface we would not have known that the title refers to a woman’s name, nor would we have understood the specific implications of some of the images used. Sometimes a preface can throw light on a poet’s creative process. The poet, composer, and critic Jiang Kui is particularly fond of attaching prefaces to his lyrics (ci). As Shuen-fu Lin has demonstrated in his book on Jiang, some of the prefaces can be considered artistic entities in themselves, and others only provide external references.¹⁷ The latter kind can inform us how the poet-composer proceeded: whether he first composed the music and then set it to words, or wrote words to fit existing music, as most Chinese lyricists did.

Of course, it is not always possible to know the actual context in which a poem was written, but it is not always necessary to know either. Often the poem itself provides sufficient indications of the dramatic context in which the speaker is imagined as speaking, such as in this poem by Cui Hao (ob. 754):

In the original, the speaker is clearly indicated to be a woman by the word qie, which means literally “handmaid” or “concubine” but was used conventionally as a way for a woman to refer to herself, tantamount to a first person pronoun. Once we realize that the speaker is a woman, it is easy to see the dramatic context in which she is speaking: she is addressing some man to whom she has obviously taken a fancy, and is using the question about where his home is as an excuse to make his acquaintance. When reading poems like this, there is no need to know what the actual context of the poem was.

In short, how much we need to know about the author, the specified reader, and the actual context of a poem depends on the nature of the poem itself. Although we may use biographical data to help us understand a poem, we should guard against the danger of falling into the vicious circle of deducing biographical data from the poet’s works and then reading his works in the light of the supposed biography. Further, it should be realized that knowing about an author’s life as a person is not the same as being able to recognize his voice as a poet. The latter means being familiar with his linguistic habits, such as his favorite expressions, his use of imagery, his handling of syntax and of allusions, and, most elusive of all, his typical tone. Just as in everyday conversation the more one knows about a speaker’s speech habits, the better one will understand what he says, so in reading poetry the more one knows about the author’s linguistic habits, the better one will understand what he writes. Only long immersion in a poet’s works can give one such knowledge.

This may be the appropriate time to raise another question: is it necessary to know the traditional genre to which a Chinese poem supposedly belongs, in order to arrive at what E. D. Hirsch, Jr., calls the “intrinsic genre”?¹⁹ Before attempting to answer this question, we need to clarify what is meant by “genre.” Critics writing in English about Chinese poetry are far from being unanimous in their use of the term. Sometimes, different verse forms, such as guti shi (Ancient Style Verse) and jinti shi (Recent Style Verse) are referred to as “genres,” but at other times poems on different subjects, such as huaigu shi (“poems recalling antiquity”) and yongwu shi (“poems singing of objects”) are called “genres.” When we turn to traditional Chinese criticism, there is no clear and consistent definition of “genre” either. The word that corresponds to “genre” is ti, which, however, is sometimes closer to “style” in English. It may refer to the style of a period, such as Jian’an ti (the Jian’an style), or the style of an individual poet, such as Taibo ti (Li Bo’s style), or that of a school, such as Xikun ti (the style of the Xikun School, who imitated Li Shangyin), or that of poetry on a particular subject, such as xianglian ti (the style of boudoir verse). In anthologies and editions of the collected works of poets, poems may be classified on the basis of form, or subject, or chronology. Similarly, literary theorists who drew up lists of genres did not follow a single criterion but used formal, thematic, or historical criteria as they wished.²⁰ Modern Chinese literary historians and critics, influenced by the nineteenth-century European evolutionary view of literature, have attempted to trace the origin and evolution of each literary genre as if it were a biological species. Actually, literary genres are neither biological species nor a priori categories, but convenient labels applied post hoc to existing bodies of works. Hence, any attempt to define precisely the exact nature of a genre or the Platonic Idea behind it is foredoomed to failure. However, we need not go to the opposite extreme, as did Croce, who thought genres no more meaningful than the arrangement of books in a library.²¹ Diachronically, “genres” represent what poets and critics of given periods perceived to be the appropriate interrelations between forms, subjects, and styles. These interrelations may change when an innovative poet appears. For instance, the ci or “lyric” was at first considered suitable only for the expression of romantic love, but Su Shi (1037-1101) showed that it was possible to write on historical and philosophical themes in lyric meters.²² Synchronically, if we adopt the concept of poetry as the overlapping of linguistic structure and artistic function, then “genres” may be conceived of as the overlapping of specific linguistic structures and specific artistic functions. Faced with a traditional Chinese generic term, no matter whether it primarily denotes a form, a subject, or a style, we can ask what its distinctive linguistic features, thematic range, stylistic traits, and artistic effects are. This will guide us in our expectations of a given poem and help us arrive at the “intrinsic genre.” Thus, we have answered, at least partially, the question posed earlier. However, our expectations based on knowledge of traditional genres may be frustrated, for a poet may surprise us by breaking conventions and establishing new interrelations between form, subject, and style. Traditional generic labels are only rough guides, not infallible ones.

Furthermore, since poetic genres do not come into being or cease to exist at precise moments in history, generic distinctions cannot be clear-cut, and some borderline cases are bound to appear. In such cases, the correct assignment of a particular poem to a particular genre is a matter of interest to literary historians rather than critics, for our understanding of a poem need not be greatly affected by knowledge of the genre to which it is supposed to belong. For example, the following poem, entitled “Grievance on the Marble Steps” (Yujie yuan), by Xie Tiao (464-499) is considered a “Music Department” song (yuefu):²³

Yet it is practically indistinguishable from a Quatrain (jueju), a subdivision of Recent Style Verse, so much so that the critic Shen Deqian (1673-1769) commented: “This is virtually a Tang Quatrain; among Tang poets this would be considered of the highest quality.”²⁵ The only reason why Xie Tiao’s poem is not called a Quatrain is that in his time the Recent Style Verse had not yet been fully established as a major verse form. If a reader ignorant of this fact of literary history mistook the poem for a Quatrain, his understanding of it would not be seriously affected. Likewise, Li Bo’s poem bearing the same title is sometimes classified both as a yuefu and a jueju,²⁶ but one’s understanding of it would not be different whether one thought it was the former or the latter:

Both poems embody a similar world and both share certain formal features, so that the identification of the genre to which each belongs does not much affect our understanding of either. But when it comes to evaluating each poem in terms of its importance in literary history, it will then be relevant to ask how far each poet has contributed to the development of the genre as a whole.