Linguistics 187 / Cultural Anthropology 187 / English 187 / ICS 151C
Variety in Language: English in the United States
Duke University
Erin Callahan-Price
Spring 2011
Design Features of Language
LF 2.2
Language Is Unique to Humans/
Language Makes Us Uniquely Human
• …It is a very remarkable fact
that there are none so
depraved and stupid, without
even excepting idiots, that
they cannot arrange different
words together, forming of
them a statement by which
they make known their
thoughts; while, on the other
hand, there is no animal,
however fortunate and
circumstanced it may be,
which can do the same.
• –Descartes, “Discourse on
What makes human language unique?:
The search for Universals
• 1. A Mode of Communication: Communication needs a vehicle.
• We use our mouths to talk and ears to listen/understand (i.e. Vocal mode is most
common for production and auditory for perception), though certainly linguistic
systems can function visually (ASL).
• Basic representational systems: chemical, visual, digital, kinesthetic…
What makes human language unique?:
The search for Universals
• 2. Semanticity: the cues or signals in a communication system have meaning
• Sounds are meaningful as constrained by conventional usage in a community of
users (Sassure 1983 [1916]), producing signs.
• When meaning is causally related to the sign, we say the meaning is transparent or
iconic. The sign ceases to be a sign; it is merely a signal or cue.
• Iconicity, transparency, conventionalization
What makes human language unique?:
The search for Universals
• 3. Pragmatic Function: Communication is purposeful
• Communication has goals, whether it’s (i.e. for animals) to stay alive, keep their
nest clean, facilitate reproduction…. or deliver an inauguration speech .
What makes human language unique?:
The search for Universals
• 4. Interchangeability: Users can both send and receive/broadcast the same
• .
What makes human language unique?:
The search for Universals
• 5. Cultural transmission: Language is transmitted culturally rather than, say,
strictly genetically*:
• a baby born in Iceland will learn Icelandic words; a baby born in Iceland and
adopted in Bombay will speak Hindi… or Telugu/Kannada/etc. vs.
• “In most organisms, the signal code is innate…so an individual can no more learn a
code different from the one it is programmed for than it can grow an extra eye.” (LF:
• * note that while the ability to speak human language is innate, the child’s exposure to a particular
language in a particular place determines which language she will acquire.
What makes human language unique?:
The search for Universals
• 6. Arbitrariness: Form of the signal is not logically attached to its meaning
• Nothing inherent in the word daffodil that points to a delicate yellow early-Spring-
appearing flower. When I produce the sounds [dæf] [o] [dIl], there’s no way you
could guess, if, for instance, you were an Icelandic speaker, that I was talking about
the flower.
• Most animals* use fairly iconic communication systems: the signals do or look like
what they say: think of a dog who pants, bares his teeth,
• * including us: a noticeable amount of human language is iconic…
Iconic? How? Why? Why Not?
What makes human language unique?:
The search for Universals
• 7. Discreteness: Complex messages are built out of smaller, isolated
(discrete) parts
• In animal communication systems, each message is an indivisible unit. Even when a
parrot reproduces “Polly Want a Cracker,” it’s actually just imitating the unanalyzed
string of sound paliwanacraka.
• Conveniently, these isolated parts are infinitely combinable…
Other parametrized, discretecombinatorial systems
What makes human language unique?:
The search for Universals
• 8. Displacement: the ability to talk about things which are not immediately
present in space or time.
• Thus, we can talk about our cousin Larry being born yesterday in Cleaveland. We
can rave about a beautiful celebrity we’ve never met (or may be long dead, into the
ranks of legend). We can refer to the “house on the next block.” Animals can’t do
What makes human language unique?:
The search for Universals
• 8. Productivity: the ability to produce and understand novel sentences.
• Human language is an open-ended system which is “pre-engineered” to churn out a potentially
infinite number of sentences. You can utter a stream of words and syllables which has never
been uttered or understood before.
• Animals, by contrast, don’t have a factory-installed mechanism for systematically combining
discrete units to produce new signals.
• Neat trick (Mad-Lib style): the rutabega located in the west foyer whimpers in slovakian tagalog
Definitions: What is Linguistics? Today, tomorrow,
What is linguistics?
A polemical issue…
• At the heart of linguistics is the search for the unconscious knowledge that
humans have about language(s), an understanding of the structure of language,
and knowledge about how languages differ from each other (“Linguistic Society of
• For the last 50 years or so, has orbited around the work of NOAM CHOMSKY
(Syntactic Structures, 1957)
• Traditionally, divided into the subfields of FORMAL LINGUISTICS:
• Syntax: study of the grammatical structure of sentences— why a native English speaker would never
say “Bone dog the ate”
• Semantics: study of meaning– or why the well known phrase “Colorless green ideas sleep furiously” is
grammatically OK but semantically empty.
• Phonetics: the study of speech sounds themselves– or how the vocal articulators (tongue, teeth, soft
palate…) move together and coordinate like members of a symphony to produce a single, perfect [p h]
• Phonology: the study of the combination of speech sounds, especially across languages– or how and
why we have [ph]otato in English and [p]apa in Spanish.
What is linguistics?
A polemical issue…
• We are born with the ability to learn languages. However, the contexts in which we use them
is culturally mediated…the world of social action, where words are embedded in and
constituative of specific cultural activities such as telling as story, asking for a favor,
greeting, showing respect, praying, giving directions, reading, insulting, praising, arguing in
court, making a toast, or explaining a political agenda. (Duranti 2009, p. 1)
• For the last 50 years or so, has orbited around the work of people like FRANZ BOAS, ROMAN
• Dynamically dividing into the subfields of SOCIOLINGUISTICS:
Language Variation and Change: study of how linguistic varieties (in some cases called “dialects”) correlate with social
variables like region, ethnic identity, gender, class, style, etc.— why your grandfather distinguished the words “dew” and “do”
and you don’t.
Linguistic Anthropology: study of language’s role(s) in (individual) social life– like how, say, language plays into “FOB”
Punjabi California hip hop identity construction (Shankar 2008)
Sociology of Language: study of language use in broader socio(political) behavior– like investigating, say, “A woman's role
in constructing status hierarchies…” by “…using honorific language in Pohnpei, Micronesia” (Keating, 1998)
(Critical) Discourse Analysis: the study of how language structures (“text” and “talk”) reproduce processes of social
domination and institutional power
Finally! Phonetics: the sounds of human language
Speech vs. Writing
• Basic assumption of modern linguistics:
speech is primary; writing is
a cultural invention that encodes speech. Thus, linguists take spoken
language as their (best) source of data. Why?
• Writing is a more recent historical development than speech.
• Writing probably began somewhere in Sumer 6000 years ago, whereas spoken language
has been around for hundreds of thousands of years.
• There are many communities in which writing does not exist.
• According to Ethnologue (2004), 57% of the 6,900 languages of the world are unwritten.
• No society uses only a written form and no spoken form.
• Even in societies who have an established orthography, many individuals fail to learn
written language.
• A majority of the earth’s inhabitants are illiterate.
• Writing must be taught, whereas spoken language is acquired automatically
• All normally-developing children naturally learn the language of their community, whereas
they may or may not learn to write when, for instance, they enter school.
Speech vs. Writing
• So why does writing seem
“more perfect” (or at least more prestigious) than
• Writing happens slower in time
• When we write, we can take long periods of time to deliberately organize, revise, reword, etc.
• Spontaneous speech is a work in progress: some of its hallmarks are hesitations, restarts,
errors, corrections, misformulations, reframings, rephrasings…
• Writing is associated with education
• The speech of the educated is set up as the “standard language” and writing is associated
with the types of language people view as “correct.”
• The type of speech that gets to get written down is educated speech: who has the
money/education/resources/connections to build (or send their work to) a printing press? To
submit a manuscipt to Random House? (i.e. in the first place)
• Writing is more permanent
• Spoken language depends on sound waves propagating through air; it’s transient,
• Because of its physical medium (clay, flax, bone), writing tends to last.
• Spelling, because it is taught, does not vary as much form individual to individual as speech
does: therefore written forms (orthographies) are more stable.
Sound, not spelling!
• Crucially, there is not a 1:1 correspondence between
sounds and letter(s) in the English spelling system:
• Same sound/different letters: thief/amoeba
• Same letters/different sounds: charter/character
• Single sound/combination of letters: lock/book/shop/that
• Single letter/more than one sound: exit/use
• Letter(s)/no sound at all: know/doubt/moose
• In working with linguistic data, we need a transcription
system where each symbol represents one sound only,
and there is only one symbol for each sound:
The International Phonetic Alphabet [IPA]
• Makes phonetic transcription consistent and
unambiguous, (even across languages!)
• Consonant set of English:
The International Phonetic Alphabet [IPA]
• Vowel set of English:
Lots of Questions about this thing’s Engineering
What’s under the hood? Design Principles
• So what kind(s) of information goes into all those
• How are they arranged, and why?
• What patterns emerge when they’re arranged that way?
• Could we posit a limited set of parameters that
produces all those sounds (and just those sounds?)
• …and how come we have one chart for vowels and one
for consonants?
Consonants and Vowels
Consonants and vowels represent the evolution of a
maximally contrastive (and thus efficient) system :
Consonants are sounds in which the air flow through the vocal
tract is either mostly or totally obstructed
Vowels are sounds made with the vocal tract open
Beyond this basic “setting,” there are several variable
parameters for consonants and vowels that allow us to
make different sounds.
Beyond Basic OPEN-CLOSED
Just like a musical instrument (clarinet, tuba, whatever), which varies the
size, shape, and character of sound by changing the size of its internal
cavities, the vocal tract changes the size, shape, and character of the
resonant chambers of our throats and heads in order to make distinct
speech sounds. Just like……..a pipe organ!
Vocal Tract Anatomy
A first look
Parameterization in Speech Sounds:
How do we vary Cs and Vs?
For consonants, the “levers” are
1. VOICING: whether or not the vocal
cords are vibrating.
Only 2 levers*:
voiced vs.
2. PLACE: where the airstream is most
obstructed by the articulators.
11 levers, e.g.:
bilabial (consonants made with both lips in contact) vs.
labiodental (consonants made with contact between
the lower lip and upper teeth); etc.
the particular way
the airstream is obstucted:
8 levers, e.g.:
plosive (or stop) consonants (air-flow is stopped
abruptly by the articulators) vs.
fricatives (friction is created by the air passing
through lightly touching articulators, etc.)
• An
Vocal Tract Anatomy
Slices of a real human
Voicing Demo
Put two fingers on your voice box
Alternately say the sounds /s/ and /z/ a few times.
Where is the (almost) obstruction occurring?
These two sounds “happen” at the same place, but they
are voiced differently, so they're distinct sounds.
Manner Demo
Try /s/ and /z/ again. The sound is made roughly at the
alveolar ridge.
Now, instead of squeezing everything through a small
opening, let's completely obstruct the air flow.
What's the resulting sound when it's unvoiced? Voiced?
When we change from partial to complete obstruction of
the air flow, that's a change of manner.
Place Demo
Now let's try /d/ and /g/
Both made by completely obstructing air flow. Both
What's the difference?
Make each sound a few times and see where your
tongue touches. Try to make it as natural as possible.
Paramaterization in Action!
Classes of Consonants
So what happens when we pull each individual “lever” (PLACE
A bunch of kinds of fresh-off-the-circuit-board consonants:
Stops: flow of air through the vocal tract is completely obstructed.
Fricatives: air flows through a narrowed area in the vocal tract
Affricates: stops followed by fricatives at the same place of
Nasals: consonants where part of the air flow is redirected through
the nose
Approximants: situated between vowels (no turbulence) and
fricatives (high turbulence)
What levers do you have to pull to make
the sounds of human language(s)?
• Basic “levers” are the
• 1. rounding of lips (round
vs. unround)
• 2. height of tongue (highness vs. low-ness of
• 3. front-ness vs. backness of tongue
1962 X-Ray
Speech Organs in Action
Quick examples
Broad Transcription
• Guessing game!
• Notes:
• The slashes indicate that this is a phonemic, or idealized, transcription– it’s the way
we think of the word, not the way it “really” sounds coming out of someone’s mouth.
• Stress is marked with what looks like a straight apostrophe that comes before the
stressed syllable.
A Course in Phonetics
• Peter Ladefoged
Transcription  English orthography Exercise I:
Plantersville, AK
=‘dialect’ feature
Speech Accent Archive Project
[English] C, V, and CC distribution
Speech Accent Archive Project
Transcription  English orthography Exercise I:
Brooklyn, NY
OK, so she has an accent. But what about him…?
Transcription  English orthography Exercise I:
Brooklyn, NC
unstressed 
Phonology: “Sounds in the Mind” (Finegan 2004)
Notes on Blog Responses
• Remember, it’s a “mock assignment.” You get the grade
you would have gotten.
• Attention to prompt
• Organization
• “Tightening up” sentence structure (edit/write multiple drafts)
• Use of linguistic terminology (NOT ‘good/bad’ but
‘descriptive/prescriptive’). Use terms wherever you can.
• Proofreading
• Take note of notes; come to office hours
Phonetics vs. Phonology
• We’ve just finished studying phonetics, which was an enterprise of
looking at the “bricks and mortar” or sounds– how they’re built.
• If linguists were not linguists–
phoneticians would be the mechanics
…phonologists would be the physicists.
• Phonology is this study of which sounds are distinct enough to
produce different meanings– how they are stored in the mind, as
evidenced by how they are organized in words.
Baby’s-eye-view on phonemes
• We learned before that there’s seldom a 1:1 correspondence between
the number of letters in a word and the number of sounds.
• We take it for granted that words are made up of a series of discrete
[acoustic] sounds. Babies don’t know this.
• What is it like to hear a radio broadcast in a foreign language? Can
you tell where word boundaries are?
The Spectrogram: A (Continuous)Picture
of Sound
• When we look at the acoustic speech signal, this lack of separation for
individual sounds becomes clear.
• One important reason for this is (for example) coarticulation: a “single
sound”’s features (voicing/nasalization) do not all neatly start and end at the
• Instead, individual features of one sound continue into the next sound, are
anticipated and show up too early one sound back, or are deleted entirely…
The Language-ing Mind as Sieve
• So– what kind of critical information does a child need just to recognize a
• What noise must be filtered “out” (e.g. passing truck/cough/false start)
• What generalities must be filtered “in”? What similarities between speakers
of a common language (who all talk different) must be in place before we can
“understand” someone new?
• Speed
• Pitch
• Volume, etc.
Coarticulation Demo
• Take the words <cop> and <keep>
• Where does your tongue touch for each word?
• Position your tongue like you’re going to say <keep.> Once your tongue is in
this initial position, say <cop> instead.
Introducing Ladefoged…
Vowel Variation Plotted for a speaker of
British English
F1 (openess) vs. F2 (front/backness)
What are formants?
Coarticulation in English:
The Acoustic View
• Articulation of a vowel depends on what consonant comes
before it
Effect of Cumulative (English) Formants
on Comprehension
Word-initial aspiration in English
• Now try <spot> vs. <pot>
Skipping to the Punch Line
Lexical Item
Phonetic Transcription
• It turns out that aspirated [ph] occurs only word-initially in English and
unaspirated [p] occurs word-medially.
• They can’t occur in the same place in a word (we call this “complementary
distribution”) and therefore can’t distinguish one word’s meaning from
• Thus, we say that [ph] and [p] are not distinctive sounds in English. They are
part of one “master unit” in English– the phoneme.
Definitions Herein
• Phoneme: structural element in the sound system of a language. CAN create
a meaning difference (“contrastive”).
• Allophone: different phonetic realizations of a single phoneme. CAN’T
create a meaning difference (non-contrastive).
• So! In English, despite the physical differences….
• …we perceive each gesture as “the same sound.”
• You can whether a sound is phonemic in a given language by inserting it to
see if it creates a meaning difference (i.e. in a fixed environment):
cop vs. keep
pop vs. mop
Back to the PTE
Duranti, Alessandro. 2009. “Linguistic Anthropology: History, Ideas, and Issues.” In Linguistic Anthropology: A Reader.
Malden, MA: Blackwell.
Finegan, Edward. 2004. Langauge: Its structure and use, 4th ed. Boston: Wadsworth.
Hockett, Charles F. and Altmann, Stuart. 1968. “A note on design features.” In Sebeok, Thomas A., (ed.), Animal
communication; techniques of study and results of research, pp. 61–72. Bloomington: Indiana University Press.
Jackendoff, Ray and Fred Lerdahl . 1981. "Generative Music Theory and Its Relation to Psychology." Journal of Music
Theory 25(1): 45-90.
Keating, Elizabeth. 1998. “A woman's role in constructing status hierarchies: using honorific language in Pohnpei,
Micronesia.” International Journal of the Sociology of Language 129: 103-115.
Krauss, Robert M. 2002. “The Psychology of Verbal Communication.” In the International Encyclopedia of the Social
and Behavioral Sciences (edited by N. Smelser & P. Baltes). Oxford: Elsevier.
Linguistic Society of America. “What is linguistics?” From LINGUIST List Student Portal. <> retrieved 1/19/2011
Ladefoged, Peter. A Course in Phonetics.
Pinker, Steven. 1994. The Language Instinct. New York: William Morrow and Company
References, cont.
Shankar, Shalini. 2008. “Speaking like a Model Minority: ‘FOB’ Styles, Gender, and Racial Meanings among Desi Teens
in Silicon Valley.” Journal of Linguistic Anthropology 18(2): 268-289.
Tserdanelis, Georgios, & Wai Yi Peggy Wong. 2004. Language Files: Materials for an Introduction to Language and Linguistics. Columbus: The
Ohio State University Press.
Weinburger, Steven H (administrator). George Mason University Department of English and Program in Linguistics. THE SPEECH ACCENT
ARCHIVE. Plantersville, AK Speaker (ID 107?) accessed at on 1/18/2011.
Brooklyn, NY speaker (ID 121?) accessed at on 1/18/2011.

Linguistic Preliminaries Part I: Phonetics & Phonolgoy