NONREFERENTIALITY IN LITERARY TRANSLATION

2013-12-03XIAOHUAJIANG

当代外语研究 2013年3期

XIAOHUA JIANG

Nonreferentiality is an important concept in semiotics.It is a kind of “extrasemantic use of language”, i.e.the signifier refers to the signified, and also refers to another or a series of related signifieds.It takes on some features of fuzziness.Based on nonreferentiality, this paper attempts a new approach to fuzziness in literary translation, arguing that (a) nonreferentiality is a thought-provoking concept for us to identify semantic traps in translation; (b) being faithful to the original often means betrayal of the original when translating nonreferential terms; and (c) a way out is to semantically fuzz up or generalize the nonreferential terms when caught in the dilemma or paradox.

Nonreferentiality is a kind of “extrasemantic use of language” (Seung 1982: 85).It takes on some features of fuzziness.

Fuzziness, or rather, semantic fuzziness implies semiosic semi-transparency in literary semiotics.Fuzziness is a superordinate concept or heading under which are lumped together many subcategories (cf.Kempson 1977: chapter 8; Heinz & Termini 1984), such as referential vagueness (e.g.“neighbor”), indeterminacy of meaning (e.g.“John’s book”), lack of specification (e.g.“fly a flag at halfmast”) and so on.Explorations have been made of the fuzziness (or vagueness) involved in literary language (cf.Leech 1971).William Empson’sSevenTypesofAmbiguity(2008), for instance, constitutes a noteworthy attempt, where seven kinds of literary vagueness are examined:

(a) A word, a syntax, or a grammatical structure, while making only one statement, is effective in several ways at once;

(b) Two or more meanings all add to the single meaning of the author; meanings all add to the single meaning of the author;

(c) An ambiguity of the third type, considered as a verbal matter, occurs when two ideas, which are connected only by being both relevant in the context, can be given in one word simultaneously (e.g., pun);

(d) Two or more meanings of a statement do not agree among themselves, but combine to make clear a more complicated state of mind of the author;

(e) The author is discovering his idea in the act of writing, or not holding it all in his mind at once, so that, for instance, here is a simile which applies to nothing but lies half-way between two things when the author is moving from one to the other;

(f) A statement says nothing, by tautology, by contradiction, or by irrelevant statements, if any; so that the reader is forced to invent statements of his own and they are liable to conflict with one another (e.g., nonsense poem);

(g) The two meanings of the word, the two values of the ambiguity, are the two opposite meanings defined by the context, so that the total effect is to show a fundamental division in the writer’s mind.

Some Chinese scholars have applied the theory of fuzziness to the quality assessment of translation (see Fan 1987; Chen 1990).

Lexical meanings are usually disambiguated in ordinary language to avoid misunderstanding (cf.Leech 1981: 66-7).Nevertheless, in literary language, especially poetry, lexical meaning is often fuzzed up, and the semiosis, i.e.semiosic process, is made semi-transparent to achieve a kind of literary or poetic effect.In what follows we will examine the transference of lexical meanings or lexical motivations involved in nonreferentiality that is taken as a subcategory of fuzziness.Generally speaking, lexical meaning(s) related to nonreferentiality is semantically clear, but pragmatically hazy in a sense.Nonreferentiality①is related to the paradigmatic relation of lexis, e.g.,

(1) 千山鸟飞绝,万径人踪灭。(柳宗元《江雪》)

(A) A hundred mountains and no bird,

A thousand paths without a footprint.(Tr.Bynner)

(B) Myriad mountains—not a bird flying

Endless roads—not a trace of men.(Tr.Wu Ching-hsiung)

From path to path no man in sight.(Tr.Xu Yuanchong)

Apparently, the original “千” and “万” do not really or precisely refer to “a thousand” and “ten thousand” respectively; in reality, they refer hazily to “many”, or rather, “all the mountains and paths within the author’s visual field”—that is why in Bynner’s translation, “千” becomes “a hundred (百)” while “万” turns out to be “a thousand (千)”.In other words, there is no essential difference between “千” and “万”.As noted by Even-Zohar and Toury (cf.Gentzler 2001:195), here the very concept of “meaning” is altered.What becomes visible instead is an unstable entity, cohering to a degree in the relation between the implicit and the explicit.The semiosic process can be diagrammatically shown as follows (taking “万” for instance):

In this light, it is justifiable to say that “万” is but an imaginary number or in Derrida’s terms a “trace” (1968) , or rather, it is a combination of “identitie” and “alteritie” (ibid).This subtlety can be illustrated in terms of the paradigmatic relation of lexis, thus:

万→ ←径→ ← 人→ ← 踪→ ← 灭 (syntagmatic relation)

↓

↑

百

↓

↑

千

↓

↑

无数

(paradigmatic relation)

This diagram implies that lexical items such as “万” interact semiotically with other text elements, present or absent, so that our perception of their meanings is closely bound up with the intended goal of the interaction.In other words, in addition to the syntagmatic relations, there is another kind of association with other signs that could have been used but in fact not (i.e., paradigmatic associations).This double-association is of great significance to translators in weighing up the value of signs in interaction in order to capture the elusive shades of meaning and render sense across semiotic boundaries.

Theoretically speaking, “万” may be replaced by any of the paradigmatically-related lexical items (which are in a quasi-synonymous relationship with it) and the literary function would remain quite the same (cf.Scholes 1982: 74).Thus “万” is namely a “trace” of all the “items” behind it.Since “万” does not necessarily refer to “ten thousand”, but hazily to “many”, this kind of linguistic or semiotic phenomenon is called nonreferentiality.Generally speaking, nonreferentiality functions to (a) diminish the semiosic transparency (viz.to make the semiosis semi-transparent to reinforce self-reflexivity of the signifier), and (b) give rise to euphemism or some associative meanings.Nonreferentiality also exists in English.For example, “a hundred”, or “a thousand” may also be “used vaguely or hyperbolically” (cf.Davidsonetal.1985: 1036) to refer to “many”, as in “I have a hundred things to do”, etc.In this light, (A) seems to be a paragon translation as far as “千” and “万” are concerned, where ST and TT not only correspond with each other in number (i.e.the signifieds), but in motivation (i.e., nonreferentiality) and function (i.e., literary significance)—here what matters is “the paramountcy of the equivalent-effect principle” (Newmark 1981:132).Nevertheless, it must be noted that translations (B) and (C) are also acceptable②, though they may not be as good as (A), for they convey the original fuzziness, but not exactly the nonreferentiality.

Nonreferentiality may be subcategorized into four kinds: (a) nonreferentiality of common terms, (b) that of numerals, (c) of proper names, and (d) of place names (see example 1, 2, 3, 4).Nonreferentiality exists so widely in classical Chinese poetry that it may constitute an important area of research in translation studies.In fact, merely a few explorations, often parenthetical, have been made into this kind of linguistic phenomenon (cf.Turner 1978; for instance).Yet, most of them are impressionistic and devoid of theorization and, more often than not, fail to capture the essence of this phenomenon.In this light, a theorized approach to nonreferentiality would be significant to both translators and translation commentators.

Nonreferential terms may be contradistinguished from allusions in two ways: (a) the latter is sometimes metaphorical, whereas the former is always non-metaphorical; (b) allusions come from classical stories (including historical events), proverbs, set phrases, maxims, etc.In other words, obvious intertextuality can be sensed in allusions.Nonreferentiality is involved in common words or phrases in which no intertextuality can be detected, though the word or phrase concerned may sometimes be culture-loaded.However, it must be noted that this contradistinction is not absolute for, when a word or phrase is culture-loaded, the border-line between nonreferential terms and allusions may sometimes become blurred (cf.“貂锦” in example 5).The significance of distinguishing the former from the latter is that, in literary translation, the ethnological coloring involved in allusions (or any culture-loaded idioms) is what we are normally committed to conveying, whereas the ethnological coloring, if any, involved in nonreferential terms is not what we generally intend to transfer for (a) the information carried by nonreferential terms is obviously false (and not used in a metaphorical way) and, thus, (b) the full preservation of it is often misleading (see example 2, 3, 4).Let’s look at the underlined characters in the following examples, where the originals involve nonreferentiality:

(2) 汉皇重色思倾国,御宇多年求不得。(白居易《长恨歌》)

(A) The Lord of Han loved beauty;

In love’s desire he pined.

For years within his palace

Such love he could not find.(Tr.Fletcher)

(B) The beauty-loving monarch longed year after year

To find a beautiful lady without a peer.(Tr.Xu Yuanchong)

(3) 秦时明月汉时关,万里长征人未还。(王昌龄《出塞》)

(A) The moon goes back to the time of Qin, the wall to the time of Han.

And the road our troops are travelling goes back three hundred miles...(Tr.Bynner)

(B) The age-old moon still shines o’ver the ancient Great Wall,

But our frontier guardsmen have not come back at all.(Tr.Xu Yuanchong)

(4) 驾长车踏破贺兰山缺。壮志饥餐胡虏肉,笑谈渴饮匈奴血。(岳飞《满江红》)

(A) I’d ride on a chariot as a he-man,

And make a break through the gap at Helanshan.

I could feed on the Hun’s flesh, forsooth,

In hunger—in the spirit of youth.

I could drink the Hun’s blood in a jovial mood.(Tr.Xu Zhongjie.The translator’s footnote: Helanshan—a mountain in today’s Cixian County, Hebei Province)

(B) Driving our chariots of war, we’d go

To break through our relentless foe.

Valiantly we’d cut off each head;

Laughing, we’d drink the blood they shed.(Tr.Xu Yuanchong)

(5) 誓扫匈奴不顾身,五千貂锦丧胡尘。

(可怜无定河边骨,犹是春闺梦里人。)(陈陶《陇西行》)

(A) They swore the Huns should perish:

they would die if needs they must...

And now five thousand, sable-clad,

have bit the Tartar dust.(Tr.Giles)

(B) They would lay down their lives to wipe away the Huns,

They’ve bit the dust, five thousand sable-clad dear ones.(Tr.Xu Yuanchong)

The underlined lexical items in the originals may be subsumedgrossomodounder nonreferentiality of

(a) common terms: 长征,汉皇,貂锦

(b) numeral: 万里

(d) place name: 贺兰山缺

Analysisofexample(2)

“汉皇” (Emperor of Han Dynasty), actually referring to Emperor Xuang Zong of Tang Dynasty (唐玄宗), is apparently nonreferential in the original.It seems to be a stereotype that poets of Tang Dynasty are inclined to use the term “汉” in place of “唐” in their poems , for instance:

(a) 日暮汉宫传蜡烛,轻烟散入五候家。(韩翊《寒食》)

(b) 边霜昨夜堕关榆,吹角当城汉月孤。(李益《听晓角》)

The nonreferentiality involved in “汉皇” may function as euphemism and, as a result, generate some poetical fuzziness, which fits in with the imaginary or artistic image of Xuan Zong (唐玄宗), who, as a historical figure, becomes fuzzy in the poem.In translation (A) “汉皇” is turned into “the Lord of Han”, which implies that the original nonreferentiality is turned into referentiality.Here, a problem has arisen: TT readers would, very likely, be disoriented to the assumption that the story the poem depicts happens at some time in the Han Dynasty (汉朝), and is related to a certain emperor of Han.In other words, referentiality may result in “contextual misplacement and distortion” (Seung 1982:172).The beautiful love story between Xuan Zong and Yang Guifei (杨贵妃), his imperial concubine, is so well-known to Chinese readers that they can easily decode the nonreferentiality and appreciate the euphemism.However, TT readers, being devoid of sufficient “socio-cultural parole” (Newmark 1981: 139), or “socio-cultural context” (Hatim & Mason 1990: 32; Scholes 1982: 28), would, almost inevitably, take “the false” to be true.Therefore, in cases like this, “fidelity” to the original often means “non-fidelity” to or betrayal of the original.This is a paradox, and “a trap of referentiality” involved in the transference of nonreferentiality.

In version (B) , the translator renders “汉皇” into a rather general term “the beauty-loving monarch”, which may be subsumed under the subcategory of “lacking of specification”.Thus, it is safe to say that the general term “the beauty-loving monarch” takes on some fuzziness.At this rate, (B) is successful to some extent, whereas (A) is unacceptable if without any footnote.But (B) is still questionable in a way for the original multi-valence has not, in the strict sense, been fully preserved—it conveys some fuzziness, yet, loses, as (A) does, the nonreferentiality.Owing to the great gap between the socio-cultural contexts of ST and TT, it seems that we have no better alternative unless a footnote is added to a literal translation like “the Emperor of Han”, if we try to make clear the cultural connotation③and rhetorical effect involved in the term (but this is, sometimes, all too clumsy).

From the above analyses we may draw an inference that, when we wrestle with nonreferential terms, (a) it would be the most satisfactory if we can fully transfer the original nonreferentiality (e.g., example 1A); (b) if it is too hard to preserve the nonreferentiality, we may try to keep some fuzziness (e.g., example 2B), since nonerferentiality belongs to fuzziness; (c) we should be on the alert against being “faithful” to the original semantic meaning (i.e.avoiding falling into the “trap of referentiality”).

Analysisofexample(3)

“秦时” (the time of Qin) and “汉时” (the time of Han) are almost the same in implication, meaning “since ancient time”, and “万里长征” (ten thousandlilong march) implies that “soldiers have marched a long way to guard the frontier”.But, “秦时” and “汉时” are clarified in version (A) as “the time of Qin” and “the time of Han”, which falls into the “trap of referentiality” and somewhat distorts the original meaning.The second line “and the road our troops are travelling goes back three hundred miles...” is a mistranslation (cf.translation B).The translator of (B) turns “秦时”and “汉时” respectively into “age-old” and “ancient” (these two words convey some fuzziness), and “万里长征” into “our frontier (guardsmen)”.These renderings have avoided “the trap” and is commendable in a sense.“秦时” and “汉时” as terms related to China’s socio-cultural context, though nonreferential, they carry respectively some sort of associative meaning, for nonreferential words are, at any rate, a combination of “identitie” and “alteritie”—identitiedetermines itself, or rather, its own characteristics whilealteritiegives rise to haziness (or fuzziness) and associative meaning .Associative meaning of this kind is very hard to preserve: if we keep in TT the terms which may activate associative meaning (like “the time of Qin”, “the time of Han”), we would fall into the “trap of referentiality” and mislead the readers unless a footnote is added to make the intended meaning(s) clear; if we discard the terms, we will fail to convey the associative meaning—either way would lead us to a dilemma.The way out seems to abide by “the minimax principle”, since “translation is an unnatural, artificial and artistic activity, always in varying degrees”(Newmark 1981:138).Parenthetically, in classical Chinese poetry, most proper terms, like “扬州” (Yang Zhou), “洛阳” (Luo Yang), “长江” (The Yangtze River) etc., are not only culture-loaded (and with associative meanings), but also sound euphonious (which also adds some flavor to poems).The preservation of the terms, inpinyinor other forms, will be hardly of any help to the activation of TT readers’ imagination and, so much the worse, they sound harsh or cacophonous, or even funny in English (cf.Turner 1978).④

Analysisofexample(4)

“匈奴” (the Huns), as a nationality, disappeared during the period of the Northern and Southern Dynasties (420～518).In the source text, “匈奴” is used metaphorically to refer to Nuchen (or Nuzhen) Nationality (女真族) , which built a dynasty called “金” (Jin, 1115～1234).Nuchen Nationality (or rather, Nuchen invaders) was in confrontation with the Southern Song Dynasty (南宋, 1127～1279), where the poet was a famous general fighting against the Nuchen invaders.“贺兰山” (Helanshan or Helan Mountain, in today’s Hebei Province) in the poem refers to “长白山” (Changbaishan, or Changbai Mountain in today’s Heilongjiang Province), because the capital of Jin was in close vicinity to Changbaishan.“踏破贺兰山缺” (viz.“踏破长白山缺”) implies “to destroy the capital of Jin, or rather, to wipe out all the Nuchen invaders”.Some critics doubt the authenticity of this lyric because Helanshan was neither where the capital of Jin was, nor close to it, nor where the poet ever had any battle against his enemies.Viewed from the perspective of nonreferentiality, however, it is safe to say that the skeptics have fallen into “the trap of referentiality”.The translator of (A) puts “贺兰山缺” into “the gap of Helanshan”, and “匈奴” into “the Huns”—both are straitjacketed and, consequently, misleading.Furthermore, his footnote concerning “Helanshan” does not bring out the cultural connotation and even reinforces the misleadingness.

Version (B) is acceptable.It’s very likely that the translator has noticed the original nonreferentiality and artistically turns “贺兰山缺” and “匈奴” respectively into “our relentless foe” and “they” (i.e.the enemies), which not only avoid “the trap”, but also convey some fuzziness.

From the above analyses we may come to the following conclusions: (a) nonreferentiality is a thought-provoking concept for us to identify semantic traps in translation; (b) being faithful to the original often means betrayal of the original when translating nonreferential terms; and (c) a way out is to semantically fuzz up or generalize the nonreferential terms when caught in the dilemma or paradox.

NOTES

1 Nonreferentiality is still a hot issue among semioticians.Riffaterre (1985) argues that poetry is essentially antimimetic and nonreferential.However, what is in our particular concern here is that we should remember that Jacobson never excludes the referential function from poetic utterances; he only says that it is not dominant in poetry and that it is often complicated by ambiguity (cf.Scholes 1982:46).

2 Zhang Zeqian comes to a conclusion similar to this from an aesthetic perspective (1994: 176).

3 Unfortunately, no really adequate method has been found to measure the connotative values (including the associative meaning) of words.Perhaps the least inadequate developed to date is what suggested by Osgood, Suci, and Tannenbaum (cf.Nida 1982: 94-95).

4 When being confronted with proper terms, J.Turner usually tries to refrain from verbatim transference (cf.Turner 1978), for instance:

岱宗夫如何？齐鲁青未了。

To what shall I compare

The sacred Mount that stands,

A balk of green that hath no end,

Betwixt two lands.

The original proper terms are fuzzed up in “the Sacred Mount” and “two lands”, which convey some associative meaning(s) and avoid harshness in sound.

REFERENCES

Davidson, W.& E.Kirkpatrick.1985.ChambersConcise20thCenturyDictionary[M].Edinburgh: Chambers.

Derrida, J.1968.‘Plato’s pharmacy’ [J].TelQuel32/33: 3-49/18-59.

Empson, W.2008.SevenTypesofAmbiguity[M].London: Chatto, Windus.

Gentzler, E.2001.ContemporaryTranslationTheories(2ndedition) [M].London & New York: Routledge.