Lecture 8: Language Diversity I Flashcards
1
Q
Sapir-Whorf hypothesis (3)
A
- The words and structures of a language can affect how the speakers of that language conceptualize or think about the world.
- Logic is problematic: the number of words used isn’t evidence about how people think, only evidence of what they’re able to encode in their language.
- One has to show that linguistic categories affect non-linguistic thinking → experiments can’t prove this.
2
Q
linguistic relativity (3)
A
- “Language is the mold into which the mind is poured.”
- Differences among languages are mirrored in differences in thoughts of their speakers.
- People’s thoughts, perceptions, and memories are determined by the categories and structures made available by language.
3
Q
Kay and Kempton (1984) (7)
A
- Chose three closely neighboring color chips from the Munsell color chart.
- Subjects told 2 of the colors were very similar to each other, but that one was more distant from the other two; identified the “odd one out.”
- Would participants sort by wavelength or lexical divide?
- Researchers calculated a score of perceived distance between critical pairs of color chips, and compared these to measures of objective distance.
- English speakers went with color labels (picked the blue as odd one out) → judged the distance between a “blue” and a “green” chip to be bigger than the distance between two shades of “blue,” even if the objective distances were identical.
- Tarahumara speakers (who only have one word for blue and green) went with wavelength (picked the more green as odd one out) → didn’t exaggerate the distance between colors across the green/blue boundary.
- But perhaps this result doesn’t reflect differences in perception so much as availability of appropriate labels.
4
Q
Winawer et al. (2007) (15)
A
- Color naming study measuring response time to see the effects during color processing.
- Time constraint didn’t allow for use of colour names as a fall-back.
- English has one word for blue, but Russian speakers always differentiate between light blue (goluboy) and dark blue (siniy).
- Subjects saw three squares of colors: one square on top, two on the bottom. One of the bottom squares was identical in color to the one on the top.
- Told to quickly press a button to indicate if the left or right square was a match.
- Results: Both groups took longer to respond if the two bottom squares were very close in color than if they were further apart.
- Russian speakers: two colors that were very similar but sat on opposite sides of the siniy/goluboy fence → responses faster than if the two colors would both be classified as either siniy or goluboy.
- English speakers: no advantage for colors that straddled the light/dark blue divide.
- Suggests that Russian speakers are sensitized to this particular distinction in color.
- Verbal (linguistic) interference task: memorize a string of numbers before the color task and repeat them afterwards.
- Russian-speakers took longer when the colors straddled the divide versus when they belonged in the same category; i.e. interference removed this facilitation effect.
- Spatial interference task: remember a spatial pattern.
- The facilitation effect was seen again in this task.
- Conclusion: Categories in language affect performance on colour perception tasks.
- Language can draw attention to certain conceptual distinctions and give us another code for remembering and computing with certain concepts.
5
Q
grammaticalization of concepts (3)
A
- In English, expression of tense is obligatory.
- e.g. He walk-s vs. He walk-ed vs. *He walk
- However, in languages like Mandarin this isn’t necessary.
6
Q
lexicalization of concepts (3)
A
- Languages make systematic choices about which concepts they lexicalize (i.e. express in words).
- Most verbs in English (e.g. roll, slip, slide, limp, etc.) express motion and manner.
- However, motion verbs in Spanish express motion and path.
- e.g. The bottle rolled into the cave.
- *La botella rodo en la cueva.
- La botella entro en la cueva, rodando.
7
Q
Papagragou et al. (2008) (9)
A
- Do English speakers pay more attention to the manner of motion, and Greek speakers to path (since Greek verbs lexicalize motion and path)?
- Participants looked at an image of a girl skateboarding into a net.
- Measured their eye movements.
- English speakers focussed on the skateboard, i.e. manner details.
- Greek speakers focused on the net more, i.e. path endpoint.
- But this effect was only seen if participants knew they had to describe the event afterwards.
- This effect wasn’t seen for a non-verbal memory task.
- Conclusion: lexical features of language do affect visual attention, but only when the relevant task involves language.
- Compare this to Winawer et al. (2007); linguistic interference impacted performance on color perception.
8
Q
language universals (7)
A
- Properties hypothesized to be common to all human languages. These can be divided into three groups:
- Absolute universals (all languages);
- Vowels and consonants;
- Core lexical categories (e.g. nouns, verbs);
- Sentences with subjects (but different diagnostics as to what a subject is).
- Universal tendencies (most languages);
- Implicational universals (“if a language has X, then it has Y”).
9
Q
preferred argument structure (9)
A
- The preference to keep the object of and avoid lexicalizing the subject of a transitive verb.
- Regarded as universal; note that it’s merely a tendency, not an obligation.
- Transitive subjects are somehow more peripheral in the discourse profile (i.e. not attached to the verb, as the object is).
- Is this why they’re most salient in our mental representation? (Think back to pronoun resolution).
- Or is it because they’re most salient that they can be dropped?
- Objects (and subjects of intransitive verbs) tend to be lexicalized.
- About 25% of the world’s languages actually instantiate this in their grammar (i.e. mark the transitive subjects).
- These are known as ergative languages.
- But preferred argument structure is found in many languages, whether they’re ergative or not.
