Introduction to Contemporary Linguistics January 6, 1998 Writing OVERVIEW: 1. Writing and "real language" 2. The history of writing 3. Writing English 4. Writing Chinese 5. Summary ============================================================= 1. Writing and "real language" >Why is writing our LAST topic? Because in a sense, writing (orthography) is not "real language": >Babies learn to talk without a lot of effort, but children do have trouble learning to write, and many adults are never able to learn. >All people use spoken or signed language freely, but many languages have no written form. >Spoken language was not consciously invented by anybody, but writing was. >Unlike sign language, all forms of writing are parasitic on spoken language. They don't represent meaning directly, but instead represent linguistic elements like morphemes, syllables and phonemes: Like this: Not like this: ORTHOGRAPHY ORTHOGRAPHY | | V | SPOKEN LANGUAGE | | | V V MEANING MEANING >This is true for ALL writing systems in the world, including Chinese writing [OVERHEAD]. >Morphemic writing: writing that represents morphemes as the basic unit. >Chinese: >Characters usually represent morphemes, not sounds. >But some morphemes (usually borrowings) are written with more than one character (e.g. ¸²µå). >Moreover, some characters are now usually used just for their sound: ¥d³q¡B¥d¤ù¡B¥d¨®¡B¥d´µ¬fº¸ ("Kasperl", a German name) >Ancient Egyptian: [OVERHEAD] >The hieroglyphics (Greek for "sacred carvings") often represented morphemes, but also often represented sounds. >Ancient Mayan (Central America) [OVERHEAD] >The glyphs also represented a mixture of morphemes and sounds. >Syllabic writing: writing that represents syllables as the basic unit; such a writing system is called a syllabary. >Yi (a minority group in southwest China): The only "pure" syllabary in the world [OVERHEAD] >Japanese: a more typical "syllabary", since kana actually represent parts of syllables: [OVERHEAD] >Of course, Japanese also uses Chinese characters (kanji, from an earlier Chinese pronunciation of º~¦r): >Many content words are written with kanji >All function words and grammatical morphemes are written with kana. >Phonemic writing: writing that represents phonemes as the basic unit; such a system is called alphabetic, and the written units are called letters. >Russian is close to a perfect phonemic system: Cyrillic writing: x o p o w o ("w" = a symbol I Phonemic form: /x o r o S o/ can't print!) Phonetic form: [x@raSo] >Traditional Hebrew and Arabic writing only represents consonants, probably because in these languages, vowels act more like predictable infixes: Writing Meaning Pronunciation KTB "to write" [katab] KTB "I write" [aktib] KTB "a book" [kitab] >Korean writing is arranged in syllables like Chinese, but the basic units actually represent phonemes: [OVERHEAD] >In fact, the Korean writing system has been called "the best" because the shapes of the letters actually show how they are supposed to be pronounced! [OVERHEAD] >English writing is also mostly phonemic, but of course not always: "knight" = [najt], not *[knight] 2. The history of writing >Was writing only invented once? Family tree: [OVERHEAD] >In any case, the history of writing systems is always the same: (1) First, pictures are used to represent things: pictograms. (2) Pictures are also used to represent ideas: ideograms. (3) Actually, these pictures represent words, and so are more properly called logograms ("logo"="word"). (4) It soon becomes crucial to represent sounds, and not just morphemes or words >The history of the Roman alphabet (used for Latin, English, French, German, etc) shows this change from pictures to sounds: >By 1000 BC, the Semitic people of ancient Phoenicia (modern Lebanon) had invented a consonantal alphabet. >The Phoenician letters were pictures of objects whose names started with that sound. [OVERHEAD] >As the alphabet moved from Phoenicia to Greece to Rome, the various letters changed shape and function. >The Greek letter for "Z" came early in the alphabet, but the Romans didn't need it, so it was dropped. Later they realized they needed it (for foreign words), and so it was added again -- at the end. >The letter "J" was invented in the Middle Ages (after the fall of Rome in 400 AD) by changing "I". >Small letters (lower-case letters) developed from handwriting styles in the Middle Ages: A ® a, etc. >The history of Chinese characters shows the same change from pictures to sounds: >All Chinese characters began as pictograms and ideograms, and for some characters, this origin is still relatively clear: >Pictograms: °¨¡B¤ì¡BÀt¡B¤H >Ideograms: ¤W¡B¤U¡B¤T¡B¨¥ >Compound pictograms/ideograms: «H (honest) = ¤H (person) next to ¨¥ (words) >However, even very "picture-like" characters are really logograms, since they represent spoken words, not ideas themselves: >¤ü began as a pictogram: [OVERHEAD] >¤ü represents the word [tShiEn214], not the word [kou214] (even though the idea of "dog" is the same). >Moreover, like English writing, modern Chinese characters almost always represent sound to some extent: >Most characters for content words are created by combining a semantic element (radical="root" ³¡­º) and a sound element: [OVERHEAD] >Since it's impossible to draw "pictures" of function words, the characters for function words in Chinese are usually taken from older characters for content words that had the same sound: Character Old meaning Picture of: ¯à a bear a bear standing up, with legs on right ¨º an ancient city a fur coat (left), in the west, with "capital city" perhaps Tibet (¨¶) on right ¨Ó kind of grain grain seeds hanging (compare ³Á) from a plant >Other changes have been due to the physical/psychological process of writing and reading: >Pictograms were sometimes reinterpreted: >©ú was originally X + ¤ë (moon through window) >The "wholes" of the old writing were broken into "features" that could be used again and again in many different characters: ¯à ¤ë ¥_ ¤½ 3. Writing English >Why is English spelling so weird? >Sometimes the weird spelling reflects older forms of English (e.g. before the Great English Vowel Shift): Spelling Older pronunciation "hate" [hate] "knight" [knixt] >English spelling was also influenced by conventions for spelling French and Dutch: Older spelling Modern spelling Influence "cween" "queen" French "gost" "ghost" Dutch >In the 1700s, English writers stupidly tried to return words to their "original forms" in Greek or Latin: Older spelling Latin word Modern spelling "det" debitum "debt" "samon" salmonem "salmon" "caitiff" captivus "captive" >NOTE: In the last case, the spelling change also caused a change in pronunciation! >Because of this confusing spelling system, many people have proposed reforms to make English spelling more phonemic: [Quotation from O'Grady and Dobrovolsky 1996:574 proposing a gradual change towards the new spelling] In Year 1 that useless letter "c" would be dropped to be replased either by "k" or "s", and likewise "x" would no longer be part of the alphabet. The only kase in which "c" would be retained would be the "ch" formation, which will be dealt with later. Year 2 might reform "w" spelling, so that "which" and "one" would take the same konsonant, wile Year 3 might well abolish "y" replasing it with "i" and Iear 4 might fiks the "g-j" anomali wonse and for all. Jenerally, then, the improvement would kontinue iear bai iear with Iear 5 doing awai with useless double konsonants, and Iears 6-12 or so modifaiing vowlz and the rimeining voist and unvoist konsonants. Bai Iear 15 or sou, it wud fainali be posibl tu meik ius ov thi ridandant leterz "c", "y" and "x" -- bai now jast a memori in the maindz of ould doderers -- tu replais "ch", "sh", and "th" rispektivli. Fainali, xen, after sam 20 iers ov orxogrephkl riform, we wud hev a lojikl, kohirnt speling in ius xrewawt xe Ingliy spiking werld.... >Why hasn't spelling reform worked? >Social reasons: >Writing skills are considered a test of socially valued abilities: Only if you can read, write, and spell can you be rewarded with a good job and respect. >Writing is a permanent record, so any change is obvious and controversial. >Linguistic reasons: English writing often spells allomorphs the same way, so readers can easily see that they are all the same morpheme: Morpheme Pronunciations Spellings Plural /s/ [khaets] "cats" [dOgz] "dogs" /elektrik/ [@lEktrIk] "electric" [@lEktrIsIti] "electricity" 4. Writing Chinese >Educated Americans only need to know 26 letters to recognize any real word (and pronounce an infinite number of invented or borrowed words like "fax" or "tofu"). >Educated Chinese, however, need to know up to 5000 different characters. How is this psychologically possible? >Psycholinguistic and neurolinguistic studies have shown that Chinese readers automatically look up the sounds of characters: they don't just go by the visual image alone! >There are often clues to the pronunciation, as noted above. >Characters are built up of smaller units, and these smaller units often follow patterns, which may help memory: X X X XX X X X XX X X X XXX XX X X ªB ¦h «~ ±_ ºó (none) >Many people at many different times have thought that Chinese characters are still too hard to use, and have invented simplified characters. >Many simplified characters have been used for centuries in China (and many have become standard Japanese kanji): >The simplified forms for many characters have been used since the Tang dynasty or even earlier: >By using simpler homophones: ¬Æ»ò -> ¤°»ò >By blending strokes: ¨Ó (simplified in º~´!) >The PRC government created committees starting in the 1950s to invent new simplified characters; unlike attempted spelling reforms in Europe, these changes were quickly adopted by all of the PRC (for obvious political reasons!) >The ROC government outlawed all PRC-invented language reforms until recently, but people in Taiwan have always used many of the "traditional" simplified characters. >Still, the extreme simplification used in the PRC seems ugly to most Taiwanese, and it's controversial whether it makes reading easier [OVERHEAD] >Sound-based writing of Mandarin: >The first sound-based writing system for Mandarin was invented around 1600 by an Italian, not a Chinese. >The most common romanization system was Wade-Giles, originally based on a French system. >ª`­µ²Å¸¹ : The first Chinese-invented sound-based system, originally developed in 1919, revised in 1932. >This system is sort of like a syllabary, since like Japanese kana it represents part of a syllable. >However, nobody uses it today outside Taiwan. >º~»y«÷­µ (Pinyin): Based on a system invented by a Russian which became popular among Chinese Communists in the 1940s; developed into the first version of Pinyin in 1956. >Spaces are placed between µü, not ¦r, which often makes tone marks unnecessary: "Wo yong pinyin xiele zhege juzi." >Pinyin is mostly a phonemic system, but it often makes different assumptions about what the phonemes are than does Zhuyin Fuhao (see Phonology handouts!) >Sound-based systems for Mandarin have advantages and disadvantages: >They help communication between Chinese people and the outside world. >The China News has decided to adopt Pinyin [OVERHEAD] >They may hurt communication within China: >Mandarin, Taiwanese, Hakka, etc are all pronounced very differently. >Even just in Mandarin, there are often too many homophones, and it's very difficult to know where word boundaries are: [OVERHEAD] >Inventing ways to write other languages of Taiwan >Many languages have never been written, and people are now trying to invent writing systems for them (e.g. Formosan languages like Tsou: OVERHEAD). >How should people write Taiwanese? >People try using Chinese characters, but there are many Taiwanese morphemes (sometimes borrowings from non-Sinitic languages) that have no characters. >Sometimes people try to use Zhuyin Fuhao to transcribe Taiwanese sounds: £x£¸ ¦Ñ®v £~£¸£« °Õ¡I [sign on campus last year] >But this often fails, since the phonemes and syllable structures of Taiwanese and Mandarin are not the same. >IPA would get all the sounds, but IPA is a phonetic system, not a phonemic system, and so it would have too much useless detail and would be hard to type! >As far as I know, there are at least three different romanization systems for Taiwanese that are floating around the island. I hope one gets chosen without too much fighting, but I doubt it will happen soon.... 5. Summary >There are no "pure" pictographic or ideographic systems; all writing is parasitic on spoken language. >There are no "pure" morphemic, syllabic, or phonemic systems; all systems try to respect both morphology and phonology. >Designing a new writing system (such as for Taiwanese) thus requires an understanding of morphology and phonology.