PSC132: Language & Cognition

Language Science
The scien)fic study of language.

©2010 by Ma;hew J. Traxler

 The Research Agenda
• What does one know when one knows a
language?

• How did we get this way?

• How does an individual use their knowledge

when producing or understanding language?

• How do neural systems support language

func)on?
 I. What does one know when one knows a
language?
• Hocke;’s Design Features
– Seman/city
– Arbitrariness
– Discreteness
– Displacement
– Genera/vity
– Duality of Pa;erning

(Hocke;, 1960, The Origin of Speech, Scien/fic American, 203, 88-96)

• Components of Languages
– Phonemes Stored
– Syllables
– Morphemes
– Words
– Phrases/Clauses
– Sentences Generated
• Components of Languages
– Phonemes Lexicon
– Syllables
– Morphemes
– Words
– Phrases/Clauses
– Sentences Grammar
“The rules aren’t the ones we were taught in school.”
--Ivan Sag

• Prescrip)ve Grammar: Completely arbitrary

rules that you can follow if you wish to sound
like a proper English gentleman.

• Descrip)ve Grammar: Systema)c rules that

determine how people actually speak.
Examples of descrip)ve rules of English:

“Each clause can have only one main verb.”**

*My Grammar teachers liked gave rules.

“Verbs go in the middle.”

*Liked grammar teachers rules.
Examples of descrip)ve rules of English:

“Each clause can have only one main verb.”**

*My Grammar teachers liked gave rules.

“Verbs go in the middle.”

*Liked grammar teachers rules.

**Not true for Nicaraguan Sign Language**

“A theory of grammar is a theory about the
mental representa)on of language.”
--Ivan Sag et al., (2003, p. 42)

Three func)ons of grammars*:

1. Order
2. Agreement
3. Case Marking

*(Pinker & Jackendoff, 2005, The faculty of language: What’s special about
it? Cogni/on, 95, 201-236)
• 1. Order: red wine vs. vin rouge (*wine red)

• 2. Agreement:

– Number: girls eat vs. *girls eats

– Gender: el toro vs. *la toro

• 3. Case marking:

– I like him vs. *Me like him. *I like he.

– Я пил водку vs. *Я пил водка

/ Ya pil votkoo/ */ Ya pil votka/
(I drank vodkaaccusa)ve case vodkanomina)ve case )
• Three Ideas About Recursion

– It’s a Core Property of All Languages, Unique to

Language (Hauser, Fitch, Chomsky, 2002)
– It’s a Core Property of All Languages, Shared with
other systems (Pinker & Jackendoff, 2005)
– None of the Above (Evere;, 2008)
• What is recursion?
– “The ability to place a component inside another
component of the same type.”

– E.g., a sentence inside another sentence

• z = Tom likes beans.

• y = Susan thinks (z).

• x = Susan thinks Tom likes beans.

• x = (y(z))
• What effects does recursion have?

– It allows the expression of very complex concepts.

– It removes any upper limit on the length of a
sentence.

Dave knows Margaret feels Bob believes…Susan

thinks Tom likes beans.

(a(b(c(d(e(…(y(z))…))))))
• Is recursion universal?

– Perhaps not in Pirahã (Evere;, 2002, 2005, 2008)

– English sentence with recursion: “Hand me the nails Dan bought.”

– Pirahã equivalent expression without recursion.

• Hand me the nails.

• Dan bought those very nails.
• They are the same.

(Evere;, 2008, p. 227)

Photo by E. Gibson of MIT

 II. How did we get this way?

• 3-Factor Model:
– Biological characteris)cs of the species
(phylogeny)
– Matura)onal characteris)cs of the individual
(ontogeny)
– Culture/Environment

– All three contribute to language ability.

 II. How did we get this way?
• Compara)ve Method: How do humans compare to
evolu)onary ancestors and living biological rela)ves?

• Con/nuity: Modern human language results from

quan/ta/ve changes to more primi/ve communica/on
systems.

• Discon/nuity: Modern human language is qualita/vely

different from more primi/ve communica/on systems.

(Lenneberg, 1967, Lieberman, 2000; Penn et al., 2008)

 There are no data which prove that other apes are unable to communicate
linguistically.
--E. Sue Savage-Rumbaugh

• Studies of Primates
1. Diana Monkeys (Cercopithecus diana)
• Separate “eagle” and “jaguar” alarm calls.
• Experiment
– Play “eagle” alarm call
– Follow with either real eagle noises, or real jaguar noises
– Response to “eagle” call is muted following “eagle” alarm
– Response to “jaguar” growl is extreme agita)on

– Seman)city?

(Zuberbühler, 2003; Zuberbühler, et al., 1999)

• h;ps://youtu.be/MJHb04VM1t8
• 2. Kanzi the Bonobo (Pan paniscus)
– Makes different vocaliza)ons in the context of
different objects (e.g., juice vs. bananas vs.
grapes)

(Tagliatela et al., 2003)

• 3. Chimpanzees (Pan troglodytes)
– Nim Chimpsky: learned to make gestures (see
also Washoe; see also Viki)
• 4. Mountain gorillas (Gorilla beringei beringei)
– Koko learned to make gestures

(Gardner & Gardner, 1975; Jensvold & Gardner, 2000)

Koko Fred Rogers, D.Div.
 Top 10 3-gesture sequences, Nim Chimpsky

• 10. me Nim eat • 5. grape eat Nim

• 9. eat me eat • 4. )ckle me Nim
• 8. banana eat Nim • 3. eat Nim eat
• 7. Nim me eat • 2. eat me Nim
• 6. banana Nim eat • 1. play me Nim

(Terrace et al., 1979)

Genera)vity?
• E.g., Washoe “water” + “bird”

Grammar? Word Order

• E.g., Nim: “more x” > “x more”
“give x” > “x give”
• Nurture: “Immersion” May produce greater
language ability in non-human primates.

• Nature: But biological differences are s)ll

important.

(Brakke & Savage-Rumbaugh, 1996a, b)

• E.g., bonobo vs. chimp (Brakke & Savage-Rumbaugh, 1996a, b)
– 4-year study
– Apes raised in enriched environment
– “Caregivers communicated to (the apes) with
spoken English and visuographic symbols called
‘lexigrams.’”
• Results:
– Bonobo produc)on and comprehension >
chimpanzee
– Bonobo lexigram use > chimpanzee
– “immersion”-reared apes appear to learn more
than “operantly”-reared apes.
• Differences between apes and humans
– universal acquisi)on in children; variable
acquisi)on in apes
– children experiment and innovate; apes copy
– children babble; apes don’t

“Keep your stinking paws off me, you damned,

dirty ape!”
--Charlton Heston, Planet of the Apes
• Differences between apes and humans
– As u;erances grow longer:
• In children: grammar becomes more complex
• In apes: signs are repeated

e.g., “eat Nim eat Nim,” "banana me eat banana,”

“give orange me give eat orange me eat orange
give me eat orange give me you”

(Terrace et al., 1979)

• Differences between apes and humans
– Humans use words to comment and express
inten/ons
– Apes use gestures as tools to get things* (up to
96% of gestures are requests)
(Rivas, 2005; Tomasello, 2007; Greenfield & Savage-Rumbaugh, 1990)
• Differences between apes and humans
– Humans apply gramma)cal rules consistently;
Apes apply gramma)cal rules inconsistently

– Apes interrupt far more than humans

 II. How did we get this way?
Language Origins
• Any language abili)es present in modern apes
likely existed in the common ancestor ~5-8
million years ago.

• Speech is an adapta)on.
– E.g., our vocal tracts differ from other apes
– They make us vulnerable to death by choking
– They are ideally suited to producing speech
• Producing speech requires
– The right kind of vocal tract.
• Equal distance from the larynx to the top of the throat
and from the top of the throat to the mouth opening
• Rudimentary speech pre-dates the modern physical
configura)on (Lieberman, 2000)
• Producing speech requires
– The ability to control the speech apparatus
– Rapid changes in air flow
• Human ancestors (and apes) lack the neural systems
necessary for fine breathing control
• Therefore, speech does not pre-date modern humans
(homo sapiens)

(MacLarnon & Hewi;, 1999)

• Physical change may pre-date language
– Increase in brain-to-body mass

– But selec)on would work against an idle large

brain.

– And small homo sapiens brains nonetheless

support fully developed language abili)es.
• Proto-word hypothesis
– Word-like units must be present before complex
sequences emerge.
– “Naming insight”

Possible proto-word types:

– Animal calls: “Moo”
– Exer)on noises: “Heave-ho”
– Lip-smacking: “Yum yum”
– Gree)ngs: “Hi there”

But there’s no data. (Aitchison, 2000)

• Proto-languages?
– Pidgins?
• Emerge in modern humans due to language-mixing in
adults (e.g., Hawaiian Creole, Tok Pisin).
• Simpler than full-blown languages.
• Restricted vocabulary
E.g., Tok Pisin Expressions:
– Han bilong diwai (tree branch)
– Han bilong pik (front legs of a pig)
– Han bilong pisin (bird’s wing)
• Rudimentary grammar.
– E.g., absence of determiners,
agreement, auxiliary verbs,
case, tense marking

(Bickerton, 1988; Aitchison, 2000)

• What caused the change from proto-speech to
proto-language to language?

The Gene)c Bioprogram Hypothesis

(Lenneberg, 1967; Bickerton, 1988; Pinker, 1994).

• Gene)c changes produced a “language organ”

• The language organ specializes in building complex,

hierarchical (mul)-level), symbolic structures that support
language produc)on and comprehension.
 Evidence for Genetic Bioprogram

• 1. Hawaiian Pidgin & Creole: (Bickerton, 1988)

Pidgin: “Building - high place - wall part - )me - now-)me -and

then - now temperature every
)me give you.”

Creole: “Get one [There is an] electric sign high up on da wall of

da building show you what )me an temperature get [it is]
right now.”
1. Hawaiian Pidgin & Creole: (Bickerton, 1988)

Pidgin: “Now days, ah, inside, washi clothes machine get, no?
Before )me, ah, no more, see? And then pipe no more, water
pipe no more.”

Creole: Those days bin get [there were] no more washing

machine, no more pipe water like get [there is] inside house
nowadays, ah?
 Evidence for Genetic Bioprogram

2. Nicaraguan Sign Language (LSN)

– 1977: central school for the deaf established
– “First-wave”: standardized vocabulary,
gramma)cally impoverished
– “Second-wave”: spontaneous addi)on of
gramma)cal features

(Emmorey, 2002; Senghas & Coppola, 2001)

 Evidence for Genetic Bioprogram

3. Selec)ve Language Impairment

– Normal IQ
– Gramma)cal Impairments, including
• Past-tense marking: Susan is cry in the church.
• Plural morphology: “zat” è “zacko”
• All affected individuals have abnormal form of FOXP2
gene.

(Lai et al., 2001; Enard et al., 2002; Gopnik, 1990, 1994; Gopnik & Crago,
1991; Vargha-Khadem et al., 1998; Watkins et al., 2002)
 I. What does one know when one knows a
language?
Language and Thought
• Are they the same?

– Watson, Skinner: Thought = sub-vocal speech

“Do we think only in words, that is, in verbal motor contrac)ons? My

own answer has been: Yes…’thinking’ is largely sub-vocal talking.”
--J.B. Watson, 1924

– If so: Elimina)ng speech eliminates thought

Curare experiment
– 34-year-old, healthy volunteer injected with curare
– Complete paralysis results
– Could perceive, remember, and answer ques)ons
________________________________________________________________
2:11 PM: Curare injec)on administered over 15 minutes.
2:20: Speech no longer possible. Can hear dis)nctly. S)ll able to nod head and move hands.
2:22: Subject reports by movement of head that the experience is not unpleasant.
2:26: Ability to comprehend and answer ques)ons accurately is indicated by correctness of replies
when inquiries are restated in the nega)ve or double nega)ve.
2:45: Subject now unable to signal response to inquiries due to complete paralysis.
2:48: Eyelids manually opened. Subject stated upon recovery that he was "clear as a bell" all this
period.
4:50 PM: Subject is able to sit up on edge of bed. Complete subjec)ve report dictated.
__________________________________________________________________
All events are direct quotes or paraphrases of the original report.

(Smith et al., 1947)

• “Brother John”
– Epilep)c seizures
– Episodes up to several hours in length
– Severe language disrup)on, including “inner
speech”
– Intact: memory, object recogni)on, complex tool
use, instruc)on following, mathema)cal ability

– Thought without language

(Lecours & Joane;e, 1980)

Language with impaired cogni)on: 1. Williams
Syndrome
“I love listening to music. I like a li;le bit of Beethoven, but I especially like
Mozart and Chopin and Bach. I like the way they develop their music – it’s
very light, it’s very airy, and it’s very cheerful music. I find Beethoven
depressing.”

– Individual can’t do basic math, retrieve a small set of

objects

(Lightwood, 1952; Williams et al., 1961; Finn, 1991; Karmiloff-Smith et al., 1998; Reilly
et al., 1990; Thomas et al., 2001; Tyler et al., 1997; Losh et al., 2000)
Language with impaired cogni)on: 2. Language
Savant

– “Christopher”
– Learned 13 foreign languages
– Can’t look aˆer himself
Normal People “Brother John,”
Curare Subject

Williams Syndrome, Linguistic Down Syndrome

Savants
 Language and Thought:
Linguistic Determinism
• Benjamin Lee Whorf: “The language we speak
determines the way we perceive the world
and the way we categorize concepts.”
E.g., “Eskimos have many words for snow.”
 Linguistic Determinism
• But:
– Eskimo’s don’t have many words for snow.
• “qanik”: snow in the air
• “aput”: snow on the ground

– And: There’s no evidence that Eskimos perceive

the world differently from anyone else.

(Mar)n, 1986; Schultz-Lorentzen, 1927; Pullum, 1989)

• And:
– Emo)on and color percep)on similar worldwide
– Despite differences in languages, e.g., color
hierarchy:

• Black/white è red è yellow/green èblue, brown,

purple, pink, orange, gray

(Kay & Maffi, 1999; Hupka et al., 1999; Berlin & Kay, 1969; Ekman et al., 1969)
Recent Findings Resuscitate Whorf: Language
Affects Thought
• Chinese speakers:
– Advantage in number term acquisi)on and basic
arithme)c
– Disadvantage in counterfactual reasoning:

• If all circles are large, and this triangle Δ is a circle, is it

large?

(Chen et al., 2006; Hunt & Agnoli, 1991; Miller & S)gler, 1987)
• Pirahã number processing:
– Pirahã has three quan)fica)on
terms: hói, hoí, and baágiso
– Pirahã speakers
can match set sizes
– Pirahã speakers perform
poorly when they need to
remember quan))es
• E.g., “nuts in a can” task
(Evere;, 2008; Frank et al., 2008; Gelman & Gallistel, 2004; Gordon, 2004)

Photo by E. Gibson, MIT

• Russian color categoriza)on:
– Russian requires “light-blue” vs. “dark-blue”
categoriza)on
• голубой ("goluboy") vs. синий ("siniy“)
• Compared to English speakers, Russian speakers are
faster at color-judgment tasks involving shades of blue.
• The advantage is largest when the physical difference
between shades is smallest.

(Winawer et al., 2007)

• Deaf and hearing fluent signers
– Sign-language requires spa)al transforma)ons
(mental rota)on) and detailed face percep)on
– Fluent signers are be;er at
• Mental rota)on
• Face percep)on/judgment

– Fluent signers perceive apparent body mo)on

differently than non-signers

(Emmorey, 2002)
How does an individual use their knowledge
when producing or understanding language?
• The modularity hypothesis:
– Language is a module
– The language module contains sub-modules

• Domain specific
• Gene/cally determined
• Dis/nct neural structure
• Computa/onally autonomous/encapsulated

(J. Fodor, 1983)

 Modular Speech Production System

Hypothetical Mental Modules

Hypothetical Mental
Processes

Lexical Access

Positional
Processing
& Inflection

Phonological &
Prosodic
Encoding

Execution
 Review
• 1. Language has features that dis)nguish it
from other communica)on systems.
• 2. These include grammars that govern how
symbols combine.
• 3. Language and “thought” rely on separate,
but interac)ng systems.
• 4. Modern human language abili)es developed
from more primi)ve systems.
• 5. Modern language abili)es reflect the
interac)on of gene)c, developmental, and
cultural influences.