language and thought

Question 1

The context:

Answer 1

• Unlike other animals, when humans think they often think in the language (or one of the languages) they speak.
  
  • This observation raises the question of how, very generally, language and thought are related to one another.
  • Last time we talked about concepts, mainly those that have a single corresponding word.
  • But the question this time is a broader one – it is not just about words.

Question 2

Categories, concepts and the world - 1:

Answer 2

• Linking back to last time…
  
  • If concepts reflect categories that are objectively out there in the world, independent of the languages used to express those concepts, then how we categorise (“think about”, in one sense) the world shapes what language must be like if we are to talk (sensibly) about the world

Question 3

Categories, concepts and the world - 2:

Answer 3

• But if some, or all, categories are socially constructed, different cultures (typically speaking different languages) may construct different concepts, and (maybe) the language we speak plays a role in shaping how we think.
  
  • The idea of different cultures/different concepts is more plausible for abstract concepts, but it can be applied more widely.

Question 4

The traditional view:

Answer 4

• The first of these views is the traditional view that thought has priority over language and that languages are tailored to express the thoughts we have. People who have promoted versions of this view include: Aristotle (ancient Greek philosopher), Jean Piaget (key figure in developmental psychology), Noam Chomsky (key figure in modern linguistics), Roger Schank (key figure in GOFAI – good old-fashioned AI – as applied to language)

Question 5

The traditional view - why?:

Answer 5

• Some adherents of the traditional view argue that there is a LANGUAGE OF THOUGHT (or “Mentalese”)
  
  • For example, Jerry Fodor in his book “The Language of Thought” (1980)
  • It has much in common with Natural Languages
  • And Natural Languages are the way they are so that we can externally express what we are thinking in Mentalese.

Question 6

The linguistic relativity hypothesis (Sapir-Whorf hypothesis):

Answer 6

• The second view, that language determines thought, is the Sapir-Whorf hypothesis, anticipate by Wilhelm von Humboldt and Johann Gottfried Herder in 18th/19th century Germany
  
  • “Language shapes the way we think, and determines what we can think about” - Benjamin Lee Whorf

Question 7

Some well-known claims attributed to Whorf:

Answer 7

• The Hopi don’t have a “linear” concept of time
  
  • The Inuit have vastly more words for types of snow than English speakers
  • Not having a word for a concept makes it hard/impossible to understand
  • German:
  • Schadenfreude
  • Anstandsstückchen

Question 8

But he also made observations about differences in sentence structure and claims about how they affected thought:

Answer 8

· English:
- It is a dripping spring
· Apache:
- To no go
- Water move down be clear

Question 9

Versions of the Sapir-Whorf hypothesis:

Answer 9

· STRONG – the language someone speaks determines how they think
· WEAK – the language someone speaks makes certain types of thought easy for them and other types difficult
· VERY WEAK – the language someone speaks affects how easily information can be encoded and remembered
· With these different versions, it becomes difficult to devise a test of the hypothesis

Question 10

Whorf’s story:

Answer 10

· Whorf was a fire prevention engineer who worked in the insurance industry, and who studied linguistics in his spare time (mainly with Sapir, an academic linguist who worked at Yale). Geoff Pullum called him a “Connecticut fire prevention inspector and weekend language-fancier” (in “The great Eskimo vocabulary hoax” – about the “snow” claim).
· He studied Amerindian languages at a time when they were falling into disuse and were in danger of disappearing
· His work on linguistic relativity became influential in Psychology in the 1950s after the posthumous publication of a set of his papers.
· Some initial results (by Lenneberg and colleagues) appeared to support Whorf
· Fairly quickly, (1960s onwards) results were found that seemed to go against his ideas, and he was criticized for being unscholarly and imprecise.
· More recently (c.2000 on) his ideas have been revisited and re-investigated.

Question 11

General criticisms of Whorf:

Answer 11

· He used a simplistic, word-by-word, approach to translation
· He assumed that every aspect of language and language structure is reflected in thought
· He ignores the fact that languages can express concepts that they do not have single words for (e.g. by using more complex expressions)
- “American Psychologist” – from the previous lecture
- Schadenfreude = “pleasure derived by someone from another person’s misfortune”
· He ignores the fact that language differences almost always go together with cultural differences and that cultural differences may be more important than language differences in bringing about different ways of thinking.

Question 12

Psychology against Whorf:

Answer 12

· The classic finding against Whorf was that speakers of languages with very different colour vocabularies (some languages only have two words that are purely colour terms – corresponding roughly to light/warm and dark/cool) see colours in similar ways (though there has been much recent work refining this view)
- Some of this work was by Eleanor Rosch, who we met last time, and who did anthropological work with her then husband Fritz Heider (among the Dani, in Papua New Guinea, who have just two basic colour terms: mili = light/warm and mola = dark/cold) [some published under her married name, Heider]
· This line of research led to the classic linguistic work “Basic Color Terms” by Brent Berlin and Paul Kay (1960), which was seen as cementing an anti-Whorfian position
· Note that perceiving colours is a very small corner of human thinking, if it is thinking at all.

Question 13

Basic colour terms:

Answer 13

· Terms whose primary meaning is just a colour
· Different languages have different numbers of these terms
· The order in which terms appear in languages is fixed

Question 14

more recent work on colour

Answer 14

• Roberson et al (2000)
• the Berinmo have 5 basic colour terms
• unlike Rosch, Robertson et al. found that across tasks categorical perception of colour was aligned with colour terms
• these results suggest that perception/thoguht is guided by language categories

Question 15

more recent work on colour 2

Answer 15

• Winawer et al (2007) noted that Russian has two basic colour terms for the area of colour space that is called blue in English.
• these terms are goluboy ( light blue) and siniy (dark blue)
• distinction makes it easier to for Russian speakers to discriminate two blues, if one is light and one is dark.
• but this effect is abolished if participants perform a verbal interference task at the same time
• the verbal interference task prevents them using verbal labels for the colours

Question 16

Linguistic relativity - what happened?:

Answer 16

· Task: watch videos of both kinds of events
- Study 1: “What happened?”
- Study 2: “Who did it?”
· Study 1: Differences in language
- For intentional events: no difference
- For accidental events: English speakers used more agentive descriptions (“She broke…”) than Spanish speakers (2 bars on right hand side)
- This finding reflects the fact that the non-agentive construction is common in Spanish for accidental events, whereas in English we tend to stick with “she broke…”, even though, in the strong sense, there is no agency (=deliberate action of breaking)

Question 17

Linguistic relativity - who dunnit?:

Answer 17

· Study 2: Differences in memory
- The memory task does not rely on language in the same way that the description task does.
- For intentional events: no difference
- For accidental events: English speakers remembered the correct actor more frequently than Spanish speakers (2 bars on right hand side)
· English speakers would typically have said “she broke…”, Spanish speakers “It broke…”
- As a control, an object orientation memory task showed no baseline language differences in memory ability

Question 18

Linguistic relativity - who dunnit? 2:

Answer 18

· Conclusion: differences in language influenced the encoding/memory of the event
· See also Boroditsky (2001): time metaphors in Mandarin and English
· English speakers tend to talk about time as if it were laid out along a horizontal time line, Mandarin speakers as if it were laid out vertically.
- Processing a vertical array primed time judgements in Mandarin
- Processing a horizontal array primed time judgements in English
· Chen (2007)
- Disputed the linguistic analysis – Mandarin speakers also use horizontal metaphors
- Failed to replicate the empirical findings
· Fuhrman et al. (2011)
- Confirmed horizontal metaphors in Mandarin
- But still found language differences – vertical time coding in Mandarin

Question 19

Other comparisons that could shed light on the relation between language and thought:

Answer 19

· Language in the deaf vs language in the hearing
· Language in aphasics vs language in speakers without brain damage
· Language in humans vs “language” (or communication systems) in animals
· Language in monolinguals vs language in bilinguals and multilinguals
· Use of different dialects or different registers of the same language, either by the same speaker or by different speakers
- There has been work on language in different social classes in the UK (Basil Bernstein)
- And on black vs white dialects in the US (William Labov)

Question 20

Is there a middle way?:

Answer 20

· The Russian psychologist Lev Vygotsky proposed, based on
· observations of children’s development, that language and thought are initially separate (with early language being mainly vocal imitation), but that “internal speech” gradually becomes the major form of thinking.
· His influence was curtailed by his early death, aged 37 (compare; Piaget, 84)

Question 21

Vygotsky’s Three Stages:

Answer 21

· STAGE ONE – language and thought are independent
- Speech is mainly imitative (“pre-linguistic” babbling)
- Thought is similar to thought in non-human animals
· STAGE TWO – Speech is connected to behaviour via overt spoken accompaniments (“egocentric speech” – not intended for others). Words direct attention, but their connection with action is not always clear to an observer.
- As this stage progresses, words begin to precede action more often
· STAGE THREE – after the age of about 7 speech becomes internalized. The distinction between form and meaning and the mappings between them are mastered. Inner speech becomes the main way of thinking.
· BUT – for Vygotsky Speech and Thinking are never identical

Question 22

finally

Answer 22

· Although much of our thinking appears to be language-based…….
· Different people give different subjective reports of how much of it is.
· So, what other kinds of thinking are possible? (e.g. purely spatial? Imagery-based?)
· And, how is non-language-based thinking in humans related to thinking in other animals?
- Do animals think in a Language of Thought, and is it the same as or similar to our Language of Thought?
· Wittgenstein: “If a lion could speak, we could not understand him”
· Something to think(!) about, probably in a language-based way