Relation of Chinese Language to Thought

Relation of Chinese Language to Thought       The Chinese language has propensities toward treatment of the world as a continuum, and to the denial of separate, internal ideas or external, abstract propositions.  The lines between poetry and prose, art and science are blurred by the ideographic nature of Chinese writing.  The first and last of these characteristics of Chinese, especially, make a holistic approach to reality easier than in Indo-European languages.       The early twentieth century work of Durkheim and Mauss concerning the linguistic shaping of thought, including Chinese schemes of classification, was not systematically followed up until the mid-twentieth century.  Benjamin Lee Whorf, whose name is attached to the thesis that language shapes thought, never mentioned Durkheim.  The popularity of the "Whorf Hypothesis" in the 1950s was replaced by an equally fashionable dismissal by the 1980s.  Nonetheless, Chad Hansen has made a number illuminating points about philosophical implications of differences between Chinese and Indo-European languages. (2) The Whorf Hypothesis      Whorf investigated languages of the indigenous American peoples, especially of the Hopi.  He claimed that the structure of Hopi language was so different from Indo-European languages that it produced a totally different worldview, leading the Hopi to conceive of the universe in terms of process.  Edward Sapir formulated a weaker hypothesis, that grammatical differences make it easier or more difficult to formulate certain concepts, and the general claim is often called the "Sapir-Whorf hypothesis."      The Whorf Hypothesis has become a less frequent point of reference in philosophy and much of linguistics, in part because of Whorf’s often naive ways of formulating his claims.  Two philosophical developments have led many philosophers and linguists to side-step or denigrate the Whorf Hypothesis.  The first is Noam Chomsky’s development of formal, transformational grammar and the claim that there is a “universal” grammar underlying all languages.  In the formalization grammars of various languages are claimed to be surface variations or transformations of this supposedly universal, deep structure, and it follows that there are no ultimate, root differences.  Chomsky claims that recent work (in terms of his Minimal Program) shows that Chinese is very close to English.  Those who used Chinese as a counterexample the supposed universal grammar thesis took their examples primarily from written, classical Chinese.  Against critics of the universal grammar hypothesis, Rosemont has questioned whether written classical Chinese is a direct transcription of the grammatical structure of ancient spoken language.  If classical, written Chinese is not a transcription of the spoken language, these counterexamples fail, because written, classical Chinese is not a "language."  However, even if at the deepest level of structure all languages are the same, there are strong divergences in surface structure that could bias conceptual schemes by making some formulations easier in one language than in another.      An example of a Chomsky-based attack on the Whorf hypothesis based upon this new consensus is Steve Pinker's.  He identifies Sapir's more modest hypothesis with Whorf's.  Pinker then concentrates on the claims made by Whorf about vocabulary, emphasizing the recent debunking of the claim that Eskimos have many words for snow.  This debunking itself focuses not so much on Whorf's text as on the exaggerated versions of Whorf's claim in popularizations and elementary textbook accounts.  These exaggerated claims do not deal with the effects of grammatical structure as opposed to vocabulary.  Pinker also suggests that the Whorf hypothesis, as commonly applied to Chinese, is racist.  Of course, claims that different cultures have language-shaped differences in frequency of certain patterns of thinking need not show the superiority of Western thought to Chinese.  Rather, the emphases might be advantageous in some fields and disadvantageous in others.  If  Chinese organicism is enhanced by the nature of Chinese grammar, the superiority of the early Chinese understanding of magnetism to that of Europeans is a case in point, as are precocious biological advances, such as biological pest control and hormone therapy.  If process and relational thinking is aided by Chinese grammar, early Chinese recognition of evolution in the heavens (sunspots, supernovae) would be examples of its advantages over Western reification.     Another philosophical idea that seemed to undermine the Whorf Hypothesis is W.V.O. Quine's claim of the indeterminacy of translation.  Quine argues that the precise specification of sameness of meaning or of the containment of one meaning in another is prevented by the indeterminate nature of meaning.  He argues that  radical translation is impossible because of our inability to distinguish between the different means of classification involved in pointing to an object or entity.  Quine suggests that an informant pointing to a rabbit might be identifying the substance rabbit, or the parts of a rabbit, or an event, or a rabbit-appearance.  If we cannot distinguish between these different metaphysical schemes implicit in object identification then neither can we assume that the radically foreign language works the same or differently from ours.  In these cases, Whorf  cannot sensibly make claims about different metaphysical schemes of various languages.      While denying the extreme cases of  incommensurable conceptual schemes that are not mutually translatable, we can still grant Sapir's weaker form of the  Sapir-Whorf Hypothesis.  It is quite possible to formulate in English the sorts of insights that Hopi makes it easier to present.  Critics of Whorf are fond of pointing out that the process view of the world that Whorf attributes to the Hopi language is found in some Western philosophers such as Henri Bergson and Whitehead.  However, process philosophy is a minority view that has arisen only recently and is notable for its obscurity of expression, precisely because of the substance bias of European languages.  Whorf himself attempts to convey what he claims to be the Hopi world view in articles written in English, showing that the Hopi concepts are, at least partially, expressible in English.  It does not show, however, that they are naturally or easily so formulated.  Donald Davidson criticized Whorf by noting that Whorf presented Hopi sentences in English translation,but Graham notes that "there is no paradox here; Whorf would hardly have denied that bilingual speakers would be clearer about the divergence with an equally sophisticated Hopi account to compare with his." (2) Ideographs, Kircher and Leibniz      One of the major areas of dispute concerning the relationship of the Chinese language to thought concerns the “ideographic” nature of the written Chinese character.  The Westerners who originally came into contact with Chinese claimed that Chinese characters represented ideas rather than sounds.  Athanasius Kircher, an early Jesuit polymath, scientist and linguist, claimed that Egyptian hieroglyphics and Chinese ideographs directly represented concepts while Western languages represent sounds.  Kircher produced large and sumptuously illustrated Baroque texts full of symbolic diagrams that supposedly explained things by virtue of the symbolic imagery itself.  He also published a large text concerning Chinese ideograms and their relation to Egyptian and other picture writing.The philosopher-scientist Leibniz was interested in Chinese as something that he thought approximated his notion of an ideal language of thought where each symbol precisely represented a fundamental concept.      Both Kircher and Leibniz, who emphasized the ideographic nature of the Chinese language and who considered it superior to a phonetic language, are important also in the development of concepts of magnetism and field theory, respectively.  Kircher concerned himself with the nature of the magnet, with the magnetism of the Earth, and with magnetism in astronomy.  Leibniz formulated an organic view of the universe that influenced later Romanticism, an important proponent of the continuum view of nature and of the relational nature of space and time.     As more was learned about Chinese, and as twentieth-century linguists have attempted to integrate their accounts of Chinese with that of other languages, there has been a strong reaction against the view of Chinese as ideographic.  In fact, Michael D. Coe makes Kircher into the villain in his story of the decoding of Mayan glyphs.  For Coe, Kircher is the creator of “the myth of the ideograph” that misled generations of linguists concerning the nature of Egyptian, Chinese, and Mayan languages.  John DeFrancis, the author of major textbooks of Chinese, wrote polemics in several works against the “myth of the ideograph” claiming that all languages are speech-based.  In the case of Mayan language, Coe claims that decoding was delayed for up to century by the refusal to relate Mayan inscriptions to the spoken language of the indigenous peoples of the area.  DeFrancis is motivated by his admirable concern for the alphabetization of Chinese in order to increase literacy among the Chinese.  However, these concerns can be pursued without denying ideographic features of Chinese.       Coe ridicules “the dead hand of Kircherian thought" basing himself on an elementary, secondary account.  Kircher has been unjustly made the whipping-boy of Egyptologists, rather than their honorable founder by many who have not read him.  Kircher correctly hypothesized that ancient Egyptian was related to modern Coptic (a conjecture that later scholars, to their detriment, rejected) but was the first to determine the phonetic value of an hieroglyph resembling waves of water that Kircher related to the Coptic for water, mu, as the sound “m.”  Kircher’s compilations of inscriptions were valuable down through the eighteenth century as reference material for other workers.  The source for Kircher was the work of Horapollus in late antiquity, who claimed to reveal the meaning of the hieroglyphs, but is now mistakenly claimed to have simply made up his meanings.  His work was mostly erroneous but was based on some correct information.  The later Egyptian priests, after the hieroglyphics died out in the culture beyond the temple, played with and manipulated the images for incantations.  Correct information Horapollus handed down concerning the meanings of some hieroglyphs was from scribes during this late period. (2) Criticism and Defense of the Ideographic Thesis      Those who completely reject the notion that Chinese is ideographic emphasize that one component of many Chinese graphs is phonetic.  There are 895 phonetic components that appear in Chinese graphs (mostly on the right).   Opponents of the ideograph claim these allow the reader of Chinese to correctly pronounce two-thirds of Chinese graphs without further knowledge.  DeFrancis and Coe are correct to reject the simplistic notion that Chinese is wholly ideographic, but they illegitimately move from the fact that Chinese has a major non-ideographic component to the claim that Chinese is not ideographic at all.  A third of the Chinese graphs still cannot be read in terms of sound even with full knowledge of the phonetic components.      Coe and DeFrancis are correct to emphasize that Chinese ideographs are not pictographs and often serve a phonetic function.  However, Western "phonetic" languages are not entirely phonetic, that is they do not, for the most part directly represent the sound structure of the words.  Jacques Derrida correctly points out that there is a "myth of the phonetic" (which is parallel to Coe and DeFrancis' "myth of the ideogram") in that the arbitrariness of language eliminates any direct tie between the spelling and the sound in a "phonetic" language, just as it eliminates a direct, intuitive, Kircherian tie between the ideograph and the object.  Derrida claims that genuine phoneticism is not achieved even in "phonetic spelling" because of spaces and punctuation marks.  Kircher's baroque claim of the superiority of the ideograph to the phonetic, was replaced by the claims of Rousseau, Hegel and later linguists of the superiority of the phonetic to the ideographic.  Rousseau correlated the sequence from pictographic to ideographic to phonetic with the sequence of  progress from savagery to barbarism to civilization.  The supposedly transparent revelation of thought by speech makes the phonetic writing system superior.  Coe's and DeFrancis' attack upon the ideographic interpretation is contains a tacit residue of belief in the cultural superiority of the phonetic--what Derrida calls phonocentrism.  To counter the privileging of the phonetic Derrida has appealed to hieroglyphic writing, called 'a new Egyptology.'  Derrida says of non-phonetic writing, and presumably of the philosophical approach he builds upon it, "It is to speech, what China is to Europe."      Another fact that goes against the close identification of written Chinese with spoken Chinese is that the dialects of spoken Chinese are as far apart from one another as are European languages from each other and are no more mutually comprehensible.  A speaker of Mandarin can no more understand Shanghai or Cantonese dialect than an English speaker can understand German.  However, the written language is common to all of the eight or more spoken languages or "dialects" that make up Chinese.  Chinese movies shown to Chinese audiences have Chinese subtitles, comprehensible to all dialects.      Though some two-thirds of Chinese characters can be read correctly in terms of sound, if one knows only phonetic elements in the graphs, these sounds can be radically different in the different dialects.  Written Chinese has an inter-language or translation function.  Written Chinese already performs the function with respect to the mutually incomprehensible spoken dialects of Chinese that ideas, meanings, or ideal propositions perform in Western philosophy. The theory of ideas was not needed.  The language that refers to mental entities rather than things, “mentalese” (to use Wilfrid Sellars’ term), was unnecessary for Chinese philosophers.      Those who oppose the notion that Chinese is ideographic often do so for the politically admirable reason of opposing racist theories of the inferiority of Chinese to European languages.  Coe ridicules the nineteenth-century evolutionary theory of language that claimed that language evolved from pictograms to ideographs to syllabic (one symbol per syllable, as in Japanese) to alphabetic symbols.  He calls it one of many pseudo-Darwinian theories that were used for the purpose of justifying European racial superiority and colonial rule.  However, Coe and DeFrancis tacitly grant the very evolutionary theory that they claim to be opposing.  They feel that to show that Chinese is not primitive or simple-minded they must show that it is not ideographic.  In fact, DeFrancis does assert that all languages arose as pictographic languages and that Chinese is “ abysmally bad” as a phonemic writing system.      Some features of classical Chinese suggest that the written language was more detached from the spoken language than are syllabic or alphabetic languages.  For one thing, classical Chinese is extraordinarily terse and Modern Chinese seems relatively terse compared to modern European languages.  Hansen suggests that if the ancient Chinese spoke with the terseness of written classical Chinese they would be far more laconic than Gary Cooper.      The inter-language function of Chinese can be seen to the extent that its symbols were borrowed and used for other purposes in East Asian languages such as Japanese, Korean, and Vietnamese.  Japanese kambun texts present classical Chinese works in Chinese characters but with added markings to identify Japanese sentence order and grammar.  This shows that written Chinese functions on an even broader scale than as translator of Chinese dialects.  DeFrancis wrote a satire called “The Singlish Affair” in which he presented a supposed report of the transcripts of meetings of a group of Japanese, Korean, and Vietnamese scholars who were involved in a project to force Americans to write English in Chinese characters.  Many readers, including professional linguists, believed the story.  DeFrancis believed that this showed the ignorance and ideographic bias of experts in Chinese.  What it shows is that written Chinese could function in this manner.      Another feature of spoken Chinese that shows the separation of written from spoken Chinese is what Graham calls the “phonetic poverty and semantic wealth” of Chinese.  If one disregards tones, Chinese contains some three-hundred distinct syllables.  If one adds the four tones of Mandarin or the eight tones of Cantonese this leads to a thousand to two thousand syllables distinguishable by sound alone.  However, given that a great many Chinese words are monosyllabic with the proportion of monosyllabic words to the total number of words increasing greatly as we move further into the past, there are four thousand words in a minimal vocabulary and eight to fifteen thousand words in a more technical vocabulary.  This means that the average spoken monosyllabic word will average some four different meanings distinguished by context.  Written Chinese distinguishes these words by adding to the phonetic component a  radical that gives the class of things to which the ideograph refers.  If the English word “ bat” had a wood radical attached, we would know it referred to a baseball bat, not a flying mammal (whose symbol  would have an animal radical).(2) Derrida, Grammatology and the Defense of the Ideographic Nature of Chinese      Derrida criticizes of the presumption of priority of spoken to written language.  Part of the support for this presumption is an excessive focus on Western alphabetic and syllabic languages to the exclusion of ideographs, hieroglyphics.  In the criticism of this thesis held by linguists following Leonard Bloomfield, Derrida's point is neglected, even by A. C. Graham, David Hall, and R. T. Ames who discuss Derrida in considerable detail on other topics.  Graham refers to Derrida with respect to fundamental polarities underlying our syntactical or grammatical structures.  Polarities of light and dark, male and female, right and left, etc. that we find in Chinese cosmology as well as in other pre-modern thinking are claimed by Graham, following Derrida, to still underlie modern philosophical reasoning.  Graham does refer to the possibility of applying Derrida’s de-emphasis on the spoken in favor of the written, but claims that classical Chinese dispute was primarily oral, and so this does not apply to the early period.  Hall and Ames make use of Derrida’s ideas on deconstruction, but do not cite Derrida’s equally important critique of the assumption of spoken language over written language.      Derrida traces this assumption back to Socrates’ distrust of writing portrayed by Plato at the end of the dialogue The Phaedrus.  This relates to the transition from an oral to a written culture in which written words become detached from the intentions and answerability of the author.  Derrida also notes Rousseau’s claim that writing is a corruption of the authentic, primitive transparency of spoken language as part of the general corruption of civilization.  Derrida traces these attitudes into twentieth century theorists of language such as Ferdinand de Saussure and Claude Lévi-Strauss.  Bloomfield and his polemical follower DeFrancis exhibit a particularly naive version of this assumption of the priority of spoken to written language.      Derrida’s defense of writing as in a sense prior to spoken language depends on a broad highly metaphorical conception of “writing.”  This conception includes all those aspects of language that are undecidable and open to further indefinite sequences of interpretation.  On a more literal and historical level, a case has been made by the science reporter become anthropologist, Alexander Marshack, that very simple forms of written symbolic expression appear simultaneously with the earliest forms of spoken symbolic expression.  Marshack cites patterns of dots and sequences of drawings of what appear to be phases of the moon on early bone tools.  Marshack claims that these are early written lunar calendars. (2) Some Consequences of Ideographic Writing for the Relationships of Language and Painting, Logic and Poetry      Hansen notes that both the Chinese and the Westerner first exposed to Chinese are struck by the pictorial aspect of the Chinese language even if, in fact, it contains phonetic elements.  This ideographic aspect of the Chinese language probably accounts for the relatively closeness of writing and painting in Chinese art.  In China, calligraphy is a major art form and one’s level of culture is estimated at least as much by one’s handwriting (calligraphy) as by one’s grammar.  In the West, conversely, handwriting counts for little.      The fact that Chinese paintings often have a poem written upon them as part of the visual beauty of the painting as well as functioning as a verbal description or supplement shows how close poetry and painting are in China.  In China, later owners and connoisseurs of the painting write tributes or encomiums on the painting itself.  This may seem strange to Western connoisseurs of art who would consider such an action on a classic Western painting as a defilement and vandalism.      Poetry and prose are less separate in China as well.  Chinese spoken syllables are distinguished not only by their vowels and consonants but also by tones.  Given the relatively few syllables and hence monosyllabic words distinguishable solely by vowels and consonants,  spoken Chinese would be many times as ambiguous and context dependent as it is without differences in tone.  The standard example in Elementary Chinese is that the word “ma” can mean mother, horse, hemp or scold.  However, in spoken Chinese these words are distinct because each is given a different one of the four tones of Mandarin Chinese: high, rising, falling, and falling then rising clearly distinguish the spoken words.  Cantonese has eight such tones that allows for even greater disambiguation.  Despite the tones, Chinese spoken words still have multiple meanings.  However, the tones augment three-hundred or so basic syllables to twelve-hundred to twenty-four-hundred distinct syllables with tones.  This leads to an interesting contrast between the use of tone and its relationship to emotion.  In English, a rising tone is used for a question.  (California “Valley Girl” speech applies this rising tone universally, giving a hesitant tone to all statements.)  In English we use different tones of voice for doubt, commands, and flirting.  Since Chinese has already preempted certain uses of tones for the actual meanings of the words, explicit words must be added to designate functions such as doubting or questioning.  In English the statement of fact is the paradigm, while the emotive or persuasive uses of language are considered outside the formal grammar.  In contrast, in Chinese philosophy, as in western literary analysis, it is natural for emotive tones and attitudes to be just as much an object of formal analysis as the factual content of the statement.      Another feature of Chinese that brings poetry and logic closer together than does English is the extreme simplicity of the basic grammar.  Languages can make grammatical distinctions by: modifying the core words, as in inflected languages such as Latin, adding suffixes, infixes, or prefixes as in agglutinative languages such as Turkish, or, 3) by word order as in analytic or isolating languages such as Chinese.  English is the most isolating of Indo-European languages, but makes some use of other devices.  Chinese is the most isolating language in the world.  Unlike French or Latin, and even more so than English, any changes in word order in Chinese will change the grammar.  Furthermore,  Chinese puts adjectives, adverbs, and subordinate clauses before the word or phrase that they modify.  Modern Chinese allows a construction for possessives and subordinate clauses that allows for long complex sentences to be constructed.  However, if there are three or more levels of subordinate clauses and modifiers it becomes very difficult to follow such a sentence and the sentence itself sounds highly unnatural.  These complex constructions are themselves "a response to Western pressure."      Classical Chinese writers often used parallel constructions in order to give hints to the reader of the grammatical construction that might otherwise be ambiguous.  The less ambiguous sentence would supply the grammatical order and the parallel sentence would be interpreted as having the same grammatical order as the initial sentence.  This leads to the use of poetic devices, such as paradigms, as Roman Jakobson would call them, involving analogy and metaphor as a way of giving information about the grammatical structure of the sentence.  Poetical structure in classical Chinese philosophy is itself part of the grammatical structure.  Along with tones this makes emotive uses of languages and poetic metaphors in classical Chinese very much part of the syntax and logic of argument.  The use of analogy is central to much western thought concerning magnetism from the renaissance through Maxwell. (2). Mass terms, the Continuous Universe, and Holistic Metaphysics in Chinese      One of the most important suggestions concerning the influence of Chinese language on thought is Hansen's claim that Chinese nouns in general behave like  "mass nouns."  Most ordinary English nouns are "count nouns."  These nouns can be preceded by numbers, such as "one pencil, two pencils," "one woman, two women."  "Mass terms, on the other hand, such as water, sand or grass, are not properly preceded by numbers.  One cannot say "one water, two waters," unless one means kinds of waters or species of grasses.  One can say one glass of water, two glasses of water, or one bushel of grass, two bushels of grass.  Similarly although count nouns admit singular and plural, mass nouns do not admit singular or plural.  Again, two sands does not make sense (unless one is speaking of kinds of sand) but two buckets of sand makes sense.  In addition, the indefinite article does not apply to mass terms.  One does not say a water, but a glass of water, or a bucket of water.     The hypothesis that Chinese nouns are mass nouns has a number of interesting consequences with respect to Chinese metaphysics or the implicit worldview of the Chinese language.  Entities treated as mass terms would not be considered built up from discrete individuals, as in the count noun paradigm.  In Chinese the word for "a human" and for "humans" in general, either as a class or as an abstraction, is the same.  One has to say something like "two head of humans" the way one says "two head of cattle" in order (unnaturally) to get the count noun effect.      Hansen's claim accounts for the atomistic bias of Western thought (in which count nouns are typical and paradigmatic for nouns) and for the continuist or holistic bias of Chinese thought (in which mass nouns would be the standard nouns).  The world presupposed by Chinese is not an aggregate of things, but a continuous stuff (say, ch'i) that is divided up in various ways.  The principle of division is relative to our purposes, giving the sort of perspectivalism that one finds in Chuang Tzu.  To reject an absolute perspective and to recognize different ways of classifying things for different purposes need not be a nihilistic, self-refuting relativism, but pragmatism.      Hansen's thesis was criticized by one of the foremost grammarians of Chinese, Christoph Harbsmeier.  The latter argued that more recent studies of Chinese nouns such as his own have shown that there is a distinction between count nouns and mass nouns in Chinese.  Hansen somewhat modified his position in response to the criticism.  He still argues that Chinese nouns in general behave like mass nouns in English rather than like count nouns in English, even if Chinese has a subtle implicit distinction built into its usages.  Chinese lacks singular versus plural and lacks the indefinite article.  Abstractions in Chinese are generally denoted by the same word that denotes an individual.      Hansen's revised, weakened thesis does still effectively capture the tendency of Chinese to treat the world as a continuous stuff that is then divided in various ways for various purposes, as opposed to an aggregate of atoms or individuals.  Graham writes "The most interesting conclusion for Chinese philosophy drawn by Hansen from his mass noun hypothesis was that China tends to divide the world down into variously divisible and countable parts, the West tends to assemble it from individuals.  This still holds if we reclassify philosophical terms as generic rather than mass nouns (few would be count nouns)."  Graham thinks we that Davidson assumes a count noun-based model of the world as 'obviously' an aggregate of individuals in his critique of the Whorf hypothesis.  Davidson writes, "We cannot attach a clear meaning to the notion of organizing a single object (the world, nature, etc.) unless that object is understood to contain or consist in other objects"  Davidson's thought presupposes a language that sharply contrasts singular and plural, and that Davidson uses examples which show that he assumes wholes are made up out of prior, independent parts, rather than being continua (a closet full of objects but not the ocean).      In summary, despite recent reaction against traditional claims in this regard, the structure of Chinese language, particularly the written language, would seem to predispose speakers and writers to certain emphases different from those of speakers of Indo-European languages.  These include a closer unity of written language and picturing, such as painting, the role of written Chinese as an inter-language means of communication, eliminating the need for abstract propositions or mental "idea," disassociation of the written form from the phonetics of spoken language, and a tendency to treat the world as continuous rather than discrete, viewed from different relative perspectives and divided into different classifications for pragmatic purposes.  All these features of the Chinese language support or predispose to continuism, holism and field theory.