This week consisted of diving deep into the world of
near-synonyms, morphology, and typology. The reading from Atkins and Levin
studied how 3 different English dictionaries (OALD, LDOCE, Cobuild) defined 7
words similar to shake: quake, quiver, shake, shiver, shudder, tremble, and
vibrate. They are similar semantically, but they differ depending on the
context in which they are used. Using the corpus, we find that the words that
we think are exclusively intransitive or transitive can actually be used as the
other form, respectively. The 3 dictionaries define the shake words differently because they have different defining
policies. The article used a quote from Dr. Johnson, which I found very
interesting. “No dictionary of a living tongue ever can be perfect, since… a
whole life cannot be spent upon syntax and etymology, and even a whole life would
not be sufficient” (1755). Another point Atkins and Levin made is that small
differences in meaning can greatly affect syntactic behavior; this can be due
to the characterization of verbs as internally
or externally caused.
The Haspelmath articles introduce morphology and the 2 types
of a “word”: the dictionary word, the lexeme,
and the text word, the word-form. The
basic idea of morphology is that words consist of morphemes, and sentences can
be broken up into meaningful parts. The example they used, “Camilla met an
unfriendly chameleon”, could be broken up either by the individual words or
also by the affixes. A lexeme is a set of word-forms. Related lexemes for a word family, but they are listed
separately because they are less predictable than related word-forms. Compounds
have at least 2 base lexemes that can be joined in different combinations (N+N,
A+N, V+N, etc.). These compounds can be either endocentric (head-dependent) or exocentric.
Like last week’s article about syntactic structure, morphological trees can be
used to display hierarchical structure.
Finally, Slobin uses Talmy’s typography to show how
different languages translate a frog story. There are V-languages, verb-framed languages, that express path and manner
differently. A few of these languages include Spanish, Hebrew, French, and
Turkish. S-languages, or
satellite-framed languages, are much more elaborate and fancy. They are more
specific when describing the manner. His final, and one of the most important
points, is that although Talmy’s typology has provided immense information on
the diversity of language, there is much more that can be done. That requires
more samples of data, but not just written data; audio and video data will help
typologists and society to further understand the diversity and similarity
between different languages.
Reading Slobin’s article about how different languages
translate a simple story made me think about the text-to-text and text-to-voice
translators we worked with. Understanding the patterns between languages and
cultures could help to improve our translations between languages dramatically.
Still, we have to give Siri some credit.
I enjoy the quote you included- about how spending an entire life on parsing out the complexities of our language use would not be nearly enough. But I wonder why we would even want to pursue this end, if we could pursue it. At some point, I feel as though we should zoom out from the minute differences in syntactic behavior, and ask ourselves the big questions of "why does this inform us as people?" rather than just "as linguists."
ReplyDeleteI was struck by the quote you included, "No dictionary of a living tongue ever can be perfect, since… a whole life cannot be spent upon syntax and etymology, and even a whole life would not be sufficient." There seem to be two interesting readings - one, a more literal message of never being able to perfect a dictionary of the English language, given its constantly changing and growing nature. But the second is perhaps a bit more cynical, given that we cannot perfect our account of language, how do we find value in this pursuit of the impossible? And big picture, what does this all mean?
ReplyDelete