Guest Post: 'Ocean' Philology
Please extend a warm welcome to our first guest contributor, Gareth Wilson. My thanks to Gareth for his following contribution:
Ocean #4 (of 6)
Warren Ellis, Writer
Chris Sprouse, Pencils
Warren Ellis' Ocean is part of a genre that's surprisingly rare in comics: science fiction. Sure, the X-Men and JLA play around with time machines, alien empires and what-not, but few of their stories would pass muster in the pages of Analog or the schedule of Baen Books. As a fan of both comics and science fiction, I'm happy to see a serious effort to combine the two, to make a comic that aspires to be as good as a prose SF story. Ocean for the most part pulls it off, and I've bought and enjoyed all five of the issues so far. But, just like a lot of prose SF, Ocean contains some serious mistakes.
The mistakes I'll talk about, from issue #4, all involve language. This is one of the hardest topics to deal with convincingly in science fiction: whenever any SF story talks about languages it always seems to contain some kind of error. A partial exception is Ted Chiang's "The Story of Your Life", but even its realistic description of learning an alien language eventually dissolves into mysticism.
The first mistake I'll talk about is the simplest. A scientist is analysing the language of a dead race of humanoids, found floating in the ocean of Europa. He's discovered something troubling about their language, and searches for an analogy:
"You can tell a lot about a culture from its language.... I mean, if we were aliens looking at Inuit text here, we'd see that they've got fifty- some different words for snow."
This is not only a mistake, but a very old and well known mistake. Inuit or Eskimo languages don't have any more words for snow than English. Think of the different words for snow or things related to snow in English: snow, sleet, slush, flurry, blizzard and so on. The Inuit languages are radically different from English in many ways, but not in snow vocabulary. This might be put down to simple ignorance on the part of the character, but almost every serious linguist knows about the old Eskimo-snow myth. It's even in the sci.lang FAQ at http://www.zompist.com/lang16.html.
The second mistake is a little more interesting. The scientist goes on to describe the language of the humanoids:
"So far I've logged a hundred and sixty-three different words for murder."
The implication (made explicit in the following issue) is that having 163 words for murder means the humanoids must have a violent culture in which murder is common. This is also wrong, for the same reason that people in snowy places don't necessarily have more words for snow. The links between culture and language are real, but they're seldom simple. Turks don't distinguish between "he" and "she", but that doesn't make them liberal feminists. You could easily imagine a peaceful, orderly culture in which murder was rare - but when it did occur they could distinguish between (murder of a child) and (murder as part of a criminal conspiracy) and (murder due to extreme provocation) and hundreds of other variations, all expressed with a single word.
The third mistake I'll talk about is rather mysterious. The scientist has found translating the alien language easier than expected and starts to talk about human language:
"Human language comes from twelve root sounds."
As Wolfgang Pauli might say, this isn't even wrong. It's not that there are actually 13 root sounds and someone's made a typo. The problem is that linguists don't talk about "root sounds" at all. There are two types of sounds in human languages - consonants and vowels. Vowels are produced by using the mouth to modify a sound produced in the vocal cords. Different positions of the tongue and jaw make different vowels. Linguists often draw a diagram of tongue position versus jaw position and mark the vowels of a language on it. There are eight "primary cardinal vowels", which are used as reference points on the diagram: the vowels in some English words like "bead" are close to a cardinal vowel, depending the accent. But even simple English words like "kit", "book", "bird", and "sofa" contain vowels that are nowhere near the cardinal vowels. Consonants are produced by lips or the tongue interrupting a flow of air through the mouth. They're classified by place of articulation, where they happen, and manner of articulation, what they do to the airstream. "Stops", as their name suggests, completely stop the flow of air. English has stops which occur at the lips, as with "p", at the alveolar ridge, as with "t", and at the back of the roof of the mouth as with "k". Other languages stop the airstream further back in the throat, as with Arabic "q", or use intermediate positions. Other sounds are produced by letting air hiss past an obstruction, as in "s", or diverting it through the nose as in "m". The airstream can even go backwards, as in clicks and implosive consonants. None of these sounds has any claim to being a "root sound" or being the basis of all human language. And there's certainly not twelve of them, either.
Gareth Wilson
Christchurch
New Zealand
Ocean #4 (of 6)
Warren Ellis, Writer
Chris Sprouse, Pencils
Warren Ellis' Ocean is part of a genre that's surprisingly rare in comics: science fiction. Sure, the X-Men and JLA play around with time machines, alien empires and what-not, but few of their stories would pass muster in the pages of Analog or the schedule of Baen Books. As a fan of both comics and science fiction, I'm happy to see a serious effort to combine the two, to make a comic that aspires to be as good as a prose SF story. Ocean for the most part pulls it off, and I've bought and enjoyed all five of the issues so far. But, just like a lot of prose SF, Ocean contains some serious mistakes.
The mistakes I'll talk about, from issue #4, all involve language. This is one of the hardest topics to deal with convincingly in science fiction: whenever any SF story talks about languages it always seems to contain some kind of error. A partial exception is Ted Chiang's "The Story of Your Life", but even its realistic description of learning an alien language eventually dissolves into mysticism.
The first mistake I'll talk about is the simplest. A scientist is analysing the language of a dead race of humanoids, found floating in the ocean of Europa. He's discovered something troubling about their language, and searches for an analogy:
"You can tell a lot about a culture from its language.... I mean, if we were aliens looking at Inuit text here, we'd see that they've got fifty- some different words for snow."
This is not only a mistake, but a very old and well known mistake. Inuit or Eskimo languages don't have any more words for snow than English. Think of the different words for snow or things related to snow in English: snow, sleet, slush, flurry, blizzard and so on. The Inuit languages are radically different from English in many ways, but not in snow vocabulary. This might be put down to simple ignorance on the part of the character, but almost every serious linguist knows about the old Eskimo-snow myth. It's even in the sci.lang FAQ at http://www.zompist.com/lang16.html.
The second mistake is a little more interesting. The scientist goes on to describe the language of the humanoids:
"So far I've logged a hundred and sixty-three different words for murder."
The implication (made explicit in the following issue) is that having 163 words for murder means the humanoids must have a violent culture in which murder is common. This is also wrong, for the same reason that people in snowy places don't necessarily have more words for snow. The links between culture and language are real, but they're seldom simple. Turks don't distinguish between "he" and "she", but that doesn't make them liberal feminists. You could easily imagine a peaceful, orderly culture in which murder was rare - but when it did occur they could distinguish between (murder of a child) and (murder as part of a criminal conspiracy) and (murder due to extreme provocation) and hundreds of other variations, all expressed with a single word.
The third mistake I'll talk about is rather mysterious. The scientist has found translating the alien language easier than expected and starts to talk about human language:
"Human language comes from twelve root sounds."
As Wolfgang Pauli might say, this isn't even wrong. It's not that there are actually 13 root sounds and someone's made a typo. The problem is that linguists don't talk about "root sounds" at all. There are two types of sounds in human languages - consonants and vowels. Vowels are produced by using the mouth to modify a sound produced in the vocal cords. Different positions of the tongue and jaw make different vowels. Linguists often draw a diagram of tongue position versus jaw position and mark the vowels of a language on it. There are eight "primary cardinal vowels", which are used as reference points on the diagram: the vowels in some English words like "bead" are close to a cardinal vowel, depending the accent. But even simple English words like "kit", "book", "bird", and "sofa" contain vowels that are nowhere near the cardinal vowels. Consonants are produced by lips or the tongue interrupting a flow of air through the mouth. They're classified by place of articulation, where they happen, and manner of articulation, what they do to the airstream. "Stops", as their name suggests, completely stop the flow of air. English has stops which occur at the lips, as with "p", at the alveolar ridge, as with "t", and at the back of the roof of the mouth as with "k". Other languages stop the airstream further back in the throat, as with Arabic "q", or use intermediate positions. Other sounds are produced by letting air hiss past an obstruction, as in "s", or diverting it through the nose as in "m". The airstream can even go backwards, as in clicks and implosive consonants. None of these sounds has any claim to being a "root sound" or being the basis of all human language. And there's certainly not twelve of them, either.
Gareth Wilson
Christchurch
New Zealand
posted by Loren | 8:44 PM |
<< Home