Posted: November 21st, 2022

DATA EXERCISE-3: CRITICAL DISCOURSE ANALYSIS

This data exercise requires you to choose an animated Disney movie and analyze it utilizing critical discourse analysis. Your learning from Module on Discourse Analysis and Critical Discourse Analysis will help you to conduct necessary analysis.You will primarily examine the way societal power relations are reflected and reinforced through language use (verbal or nonverbal means) in your chosen animated movie. You will pay special attention to issues of power asymmetries, and structural inequalities that has been explicitly or implicitly portrayed in the movie.

When watching the movie pay attention to the following:

-Language choices concerning accent use

-More specifically the use accents to express the nature of the characters

-Language-based stereotypes

-Gender roles and/or stereotypes

-Ethnic or racial-based stereotypes

-Pay attention to word use, nonverbal cues (clothing, colors, body postures, gestures, etc.), who is the main character, the ones that holds power, the ones that represent kindness, etc.

Your research report will be 3-4 pages at minimum. 11 or 12-inch, 1 inch margins, double-spaced. Report will include the following:

-Introduction (You can use the articles assigned for this module here)-1 paragraph

-Analysis and discussion of the findings (use excerpts from the movie to support your arguments and findings, cite articles assigned for this module in these paragraphs, each source needs to be cited one time at minimum)- Multiple paragraphs

-Conclusion- 1 paragraph

Part I Language:
Some Basic

Questions

Living Language: An Introduction to Lir;{lu1′.,ticAnthropology, First Edition. Laura M. Ahearn.
© 2012 Laura M.Ahearn. Published 2012 by Blackwell Publishing Ltd.

The Socially Charged Life
of Language

All words have the ‘taste’ of a profession, a genre, a tendency, a party, a

particular work, a particular person, a generation, an age group, the day

and hour. Each word tastes of the context and contexts in which it has

lived its socially charged life …

Bakhtin 1981 :293

Words do live socially charged lives, as Bakhtin observes in the epigraph

that opens this chapter. Language is not a neutral n1edium for con,

munication but rather a set of socially embedded practices. The reverse

of Bakhtin’s statement is also true: social interactions live linguisti

cally charged lives. That is, every social interaction is mediated by

language – whether spoken or written, verbal or nonverbal. Consider

the following three examples.

Example 1: Getting Stoned in San Francisco
During the 1995-1996 school year, a special anti-drug class was run

as an elective in a large high school in the San Francisco Bay Area. 1

Students were trained as peer educators in preparation for visiting

other classes to perform skits about the danger of drugs and tobacco.

The class was unusually diverse, with boys as well as girls and with

students from many different class ranks, ethnicities, and racial groups.

On the day that the students were preparing to perform their skits in

front of an audience for the first time, they asked the teacher, Priscilla,

Living Language: An Introduction to Lir;{lu1′.,ticAnthropology, First Edition. Laura M. Ahearn.
© 2012 Laura M.Ahearn. Published 2012 by Blackwell Publishing Ltd.

6 A� IN 11,Mo �10 “L 1\1\t-%.,c, fl\/�

The Socially Charged Life of Language

N6 l<'.ON6, Al'l'lTt11N6 Gf'atJ5 oiJ Tile W”1T,
,1(, !’o0f{£ Mlour t;o � St.</ITCH6"0 1o

? ii-‘ �A!ZWN’ AtlO f� E
-‘1:’:’��.–1

Figure 1.1 Cartoon demonstrating how certain styles of speech can both
reflect and shape social identities.
Source: Jump Start (\’:) 1999 United Feature Syndicate, Inc.

what they should say if someone in the audience asked whether they

themselves smoked marijuana. Priscilla recommended that they say

they did not. Then the following exchange took place between

Priscilla and the students:

Priscilla: Remember, you’re role models.
Al Capone: You want us to lie?
Priscilla: Since you’re not coming to school stoned – (students

laugh)

Calvin: (mockingly) Stoned?
Priscilla: What do you say?
Calvin: I say high. Bombed. Blitzed.
Brand One: Weeded.
Kerry: Justified.
Brand One: That’s kinda tight.

Example 2: Losing a Language in Papua New Guinea
In 1987, the residents of the tiny village of Gapun in Papua New

Guinea (a country north of Australia) were son1e of the last speakers

of a language called Taiap, which at the tin1e had at most 89 reniaining

speakers.2 Adult villagers were almost all bilingual in Taiap and in Tok

Pisin, one of the three national languages of Papua New Guinea, and

all children were exposed to rich amounts of both Taiap and Tok Pisin

in their early years. By 1987, however, no child under the age of ten

actively spoke Taiap, and many under the age of eight did not even

possess a good passive knowledge of the language. The usual theories

5 I The Socially Charged Life of Language

about how and why so many of the world’s languages are becoming

extinct did not seem to apply to Taiap. Material and economic factors

such as industrialization and urbanization were not sufficiently

in1portant in the ren1ote village of Gapun to explain the language

shift away from Taiap. Why, then, was Taiap becoming extinct? Accor

ding to linguistic anthropologist Don Kulick, the adults in Gapun

clain1ed that the shift was occurring because of the actions of their

( often preverbal) children. Kulick writes:” ‘We haven’t done anything,’

one village man explained when I asked him why village children

don’t speak the vernacular, ‘We try to get them to speak it, we want

them to. But they won’t . . . They’re bikhed [big-headed, strong

willed] “‘ (Kulick 1992:16) .

Example 3: The Pounded Rice Ritual in Nepal
On a warn, February afternoon in 1993, a wedding procession made

its way down a steep hill in Junigau, Nepal. Several men carefully

maneuvered the bride’s sedan chair around the hairpin turns. At the

foot of the hill, under a large banyan tree, the wedding party settled

down to rest and to conduct the Pounded Rice Ritual. 3 The bride,

Indrani Kumari, remained in her palanquin, while some members of

the wedding party, including the groom, Khim Prasad, approached

her. Taking out a leafplate full of pounded rice, a popular snack in

Nepal, lndrani Kumari’s bridal attendant placed it in her lap. Khim

Prasad, coached by his senior male kin, tentatively began the ritual,

holding out a handkerchief and asking his new wife to give him the

pounded rice snack. He used the most polite, honorific form of”you”

in Nepali (tapai), and so his remark translated roughly as a polite

request to someone of higher social status:” Please bring the pounded

rice, Wife; our wedding party has gotten hungry.”

But this first request was not very effective. lndrani Kumari and her

bridal attendant poured just a few kernels of the pounded rice into

the handkerchief Khim Prasad was holding. Upon further coaching

from his elders, Khim Prasad asked a second time for the rice, this

time in a more informal manner using “timi,” a form of “you” in

Nepali that is considered appropriate for close relatives and/ or famil

iar equals.This time, Khim Prasad’s request could be translated roughly

as a matter-of-fact statement to someone of equal social status: “Bring

the pounded rice, Wife; our wedding party has gotten hungry.” But

6 The Socially Charged Life of Language

Figure 1.2 Khim Prasad (left) during the Pounded Rice Ritual, with the

bride, Indrani Kumari (seated at the right, completely covered by a shawl),

and the bridal attendant (standing in the center).

Source: Laura M.Ahearn, Invitations to Love: Literacy, Love Letters, and Social

Change in Nepal. Ann Arbor: University of Michigan Press, 2001.

again, the bridal attendant and Indrani Kun1ari poured only a few

kernels of pounded rice into Khin1 Prasad’s waiting handkerchief.

One last time Khim Prasad’s senior male kin instructed him to ask for

the rice, but this time he was told to use “ta,” the lowest form of”you”

in Nepali – a forn1 most con1n1only used ineJunigau to address young

children, animals, and wives. Khim_ Prasad complied, but his words

were halting and barely audible, indicating his deeply n1-ixed feelings

7 I The Socially Charged Life of Language

about using such a disrespectful term to address his new wife. This

third request translated roughly as a peremptory command to some

one of greatly inferior social status: “Bring the pounded rice, Wife!

Our wedding party has gotten hungry!” Hearing this, Indrani Kun1ari

and her attendant finally proceeded obediently to dump all the

remaining rice into the groom’s handkerchief, after which he handed

out portions of the snack to all members of the wedding party.

As different as these three examples are, they all describe situations in

which neither a linguistic analysis alone nor a sociocultural analysis

alone would come close to providing a satisfying explanation of the

significance of the events. The purpose of this book is to show how the

perspectives and tools oflinguistic anthropology, when applied to events

as wide-ranging as an anti-drug class in a San Francisco high school,

language shift in Papua New Guinea, or a ritual in Nepal, can shed light

on broader social and cultural issues as well as deepen our understand

ing of language – and ourselves. As we move through the chapters that

follow, we will be addressing a nun1ber of questions, including:

• What can such situations tell us about the ways in which language

is enn1eshed with cultural values and social power?
• How do dimensions of difference or inequality along lines such as

gender, ethnicity, race, age, or wealth get created, reproduced, or

challenged through language?
• How can language illun1inate the ways in which we are all the

same by virtue of being human as well as the ways in which we

are incredibly diverse linguistically and culturally?
• How, if at all, do linguistic forn1s, such as the three different words

in Nepali for “you” or the various slang words for “stoned,” influ

ence people’s thought patterns or worldviews?
• How might people’s ideas about language (for example, what

“good” language is and who can speak it – in other words, their

“language ideologies”) affect their perceptions of others as well as

themselves?
• How does the language used in public rituals and performances

both differ fron1 and resen1ble everyday, mundane conversations?
• What methods of data collection and analysis can we use to deter

n1ine the significance of events such as those described above?

8 The Socially Charged Life of Language I

The starting point in the search for answers to all of these questions

within linguistic anthropology is this fundamental principle: language

is inherently social. It is not just a means through which we act upon

the social world; speaking is itself a form of social action, and language

is a cultural resource available for people to use (Duranti 1997:2) .We

do things with words, as the philosopher J.L. Austin (1962) reminded

us decades ago. Even when we speak or write to ourselves, our very

choices of words, as well as our underlying intentions and desires, are

influenced by the social contexts in which we have seen, heard, or

experienced those words, intentions, and desires before. Linguistic

anthropologists therefore niaintain that the essence of language can

not be understood without reference to the particular social contexts

in which it is used. But those contexts do not stand apart from lin

guistic practices or somehow “contain” them, as a soup bowl would

contain soup. 4 Rather, social contexts and linguistic practices nmtu

ally constitute each other. For this reason, language should be studied,

Alessandro Duranti writes, “not only as a mode of thinking but, above

all, as a cultural practice, that is, as a form of action that both presup

poses and at the same time brings about ways of being in the world”

(1997:1) .

This approach to language differs from the popular view of lan

guage as an empty vehicle that conveys pre-existing meanings about

the world. Language, according to this view, which is held by many

n1ernbers of the general public as well as many linguists and other

scholars, is largely a set of labels that can be placed on pre-existing

concepts, objects, or relationships. In this mistaken way of thinking,

language is defined as a conduit that merely conveys information

without adding or changing anything of substance (Reddy 1979) .

Within the field of linguistics, a similar approach to language is

dominant: one in which language is reduced to a set of formal rules.

Such reductionism extends back hundreds of years but was made the

don1inant approach of the field oflinguistics by Ferdinand de Saussure,

a famous Swiss linguist who lived a century ago. De Saussure main

tained that it was not only possible but necessary to decontextualize

the study of language: “A science which studies linguistic structure

is not only able to dispense with other elements of language, but is

possible only if those other elements are kept separate” (Saussure

1986 [1916] :14) . 5 This perspective was reinforced by Noam Chomsky,

https://1997:2).We

9 I The Socially Charged Life of Language

an American linguist who revolutionized the field and has dominated

it for the past 50 years. Chomsky and his followers are interested in

discovering Universal Grammar (UG), which they define as: “The

basic design underlying the gran1n1ars of all hun1an languages; [it] also

refers to the circuitry in children’s brains that allows them to learn the

grammar of their parents’ language” (Pinker 1994:483) .

This is not to say that linguistic anthropologists are uninterested

in gran1n1ar or believe that linguistic forms cannot be studied

systematically – on the contrary, many build upon the “considerable

progress in the understanding of formal properties oflanguages” made

by scholars in the field of linguistics (Duranti 1997:7) , but they ask

very different kinds of questions that explore the intersections between

grammar and social relations, politics, or emotion. Even linguistic

anthropologists who value the work done by linguists believe that in

order to acquire a comprehensive understanding of language, it nrnst

be studied in real-life contexts ( cf. Hanks 1996) . Gran1n1ar, according

to linguistic anthropologists, is just one part of language’s “socially

charged life” (Bakhtin 1981:293) . 6

So, What Do You Need to Know in Order
to “Know” a Language?

In order to understand what it means to study language as a linguistic

anthropologist would, it is helpful to ask what it n1eans to “know” a

language (Cipollone et al.1998) . Linguists generally use the Chomskyan

distinction between “competence,” the abstract and usually uncon

scious knowledge that one has about the rules of a language, and

“performance,” the putting into practice – son1etimes in1perfectly –

of those rules. De Saussure made a sin1ilar distinction between langue

(the language system in the abstract) and parole (everyday speech) .This

distinction is partly analogous to the way a person might have abstract

knowledge about how to knit a sweater but in the actual knitting of

it might drop a stitch here or there or perhaps make the arms a bit

shorter than necessary. In both the Chomskyan and Saussurean

approaches, it is the abstract knowledge of a language system ( con1pe

tence or langue) that is of primary, or even sole, interest for a science

of language; perforn1ance or parole is irrelevant.

1 O The Socially Charged Life of Language I

To take the knitting analogy further, if Chomsky were a knittist

instead of a linguist, he would be interested only in the abstract rules

of Knitting (capitalizing the word, as he does with Language) such as

the following: Row 20: P 1, (k 1, p 1) 11 (13-15) times, k 5, TR 2, k 4,
TR 2, k 1,p 12, k 1, TL 2, k 4, TL 2, k 5,p 1, (k 1,p 1) 11(13-15)
times. 7 Chomsky the knittist would posit the existence of a Knitting

Acquisition Device (KAD, rather than LAD, a Language Acquisition

Device) , a specialized module ofthe brain that allows people to acquire

knitting skills.While he would acknowledge that people require expo

sure to knitting in their social environments in order to learn how to

knit, he would be completely uninterested in the following:

• How or why people learn to knit in various cultures and com

munities.
• How knitting practices have changed over time.
• The gendered nature of knitting and other handicrafts in many

societies (although knitting is often associated with girls and

women in this society, for example, handicrafts such as weaving

were until recently conventionally produced by lower-caste men

in Nepal) .
• The role of Madame Defarge in A Tale �[Two Cities, by Charles

Dickens, as she secretly encodes the names of counterrevolution

aries into her knitting. 8

• The global economics involved in the many different yarns people

use to knit – anything from yak wool fron, Nepal to Icelandic

wool to synthetic mohair.
• The many different kinds ofproducts ofeconomic, social, or emo

tional value that are made by knitters to be worn by themselves,

given to loved ones, donated to charity, or sold to tourists.
• The ways in which knitting is viewed by different groups in the

society – as a hip, in-group practice by some, as an old, fuddy

duddy practice by others, as a useful, money-making skill by yet

others.
• How one’s individual and social identities can be reflected in and

shaped by whether, how, what, and with whom one knits.

While this analogy with knitting is not by any means a perfect one,

it does nevertheless demonstrate how narrowly Chomsky and most

I The Socially Charged Life of Language 1 1

other linguists view language. Other practices such as playing music,

dancing, or painting would work equally well in the analogy I set up

above because knitting and all these other practices are – like

language – socially embedded and culturally influenced. Of course

there are abstract cognitive and biological dimensions to anything that

we as humans do, including language, but to reduce language solely to

these din1ensions, as Chon1sky and others do when they clain1 they

are interested only in con1petence and not in performance, is to miss

the richness and complexity of one of the most fundamental aspects

of human existence.

Linguistic anthropologists therefore reject the Chomskyan/Saus

surean distinction between competence (langue) and performance

(parole) , though they do so in various ways. Some deny the existence

of any distinction at all between competence and performance (langue

and parole) , while others give primacy to performance (parole) . Still

others either expand the definition of competence to include the

ability to use language skillfully and appropriately in particular social

contexts (cf. Hymes 2001[1972] ) , and many view competence and

performance (langue and parole) as equally important.What all linguis

tic anthropologists agree upon, however, is that to know a language,

one nmst know far n1ore than an abstract set of gran1matical rules.

What else must one know in order to know a language, then, aside

from grammatical rules? According to Cipollone et al. (1998:8-11) ,

there are five basic components of a language that can be studied, and

one nmst n1aster all five of these areas in order to know a language:

• Phonology. The study of sound in language. In order to know a

language, one must be able to recognize and produce the sounds

that are n1eaningful in that language. In the case of sign languages,

instead of sounds, one must be able to recognize and produce the

appropriate gestures.
• Morphology.The study of the internal structure of words. In order to

know a language, one nmst be able to use suffixes, prefixes, or infixes

( depending on the language) . In English, for example, one must

know how to create plurals by placing an “-s” on the end of most

(but not all) words, and nmst know what adding “un-” to the begin

ning of a word does to its n1eaning. In n1any Native American lan

guages, these sorts of affixes are placed inside a word to create infixes,

12 The Socially Charged Life of Language I

while in Chinese languages, each morpheme, or unit of meaning, is

a separate word, including morphemes indicating tense or plurality.
• Syntax. The study of the structure of sentences, including the con

struction of phrases, clauses, and the order of words. In order to

know a language, one must be able to combine subjects, verbs, and

objects in a grammatically correct way.
• Semantics. The study of meaning in language, including analysis

of the meanings of words and sentences. In order to know a lan

guage, one must know how to construct and interpret meanings.
• Pragmatics. The study ofelanguage use, of actual utterances, of how

n1eanings emerge in actual social contexts. This includes culturally

and linguistically specific ways of structuring narratives, perforn1-

ances, or everyday conversations. In order to know a language,

one must be able to use language in socially and culturally appro

priate ways.

Most linguists focus prin1arily or solely on one or n1ore of the first

three components (phonology, n1orphology, or syntax) , with syntax

being accorded primacy ever since Chomsky became dominant in the

field. In contrast, n1ost linguistic anthropologists (as well as some schol

ars in related fields such as sociolinguistics or discourse analysis) study

the final two components (semantics and pragmatics) in ways that

integrate these two components with the first three. Indeed, linguistic

anthropologists consider phonology, n1orphology, and syntax to be so

fundan1entally affected by the social contexts in which these aspects of

language are acquired and used that to consider them in isolation from

these contexts is at best artificial and at worst inaccurate. For the lin

guistic anthropologist, every aspect of language is socially influenced

and culturally meaningful. To use language, therefore, is to engage in a

form of social action laden with cultural values.

So, How Do Linguistic Anthropologists Study
Language as Social Action?

While linguistic anthropologists hold in con1n1on the view that

language is a form_ of social action, there is nevertheless great diversity

in topic choice and research n1ethods within the field. Chapter 2 will

The Socially Charged Life of Language 13

Figure 1.3 “Zits” cartoon about the varying cultural meanings associated

with language use.
Source: Reproduced with kind permission of Dan Piraro and Bizarro.com.

Distributed by King Features Syndicate.

examine the various research methods used by linguistic anthropolo

gists, so what I present here are son1e exan1ples of the topics scholars

have chosen and an explanation of how these topics contribute to our

understanding of language as a forn1 of social action. These studies

illustrate but by no means exhaust the wide-ranging diversity of con

temporary linguistic anthropology.

Keith Basso

Keith Basso’s (1996) ethnography, Wisdom Sits in Places: Landscape and

Language Among the Western Apache, explores “place-nuking” as a lin

guistic and cultural activity.eThis book was written after Ronnie Lupe,

chairman of the White Mountain Apache tribe, asked Basso to help

make some maps: “Not whitemen’s maps, we’ve got plenty of them,

but Apache maps with Apache places and names. We could use then,.

Find out something about how we know our country. You should

have done this before” (Basso 1996:xv) . When Basso took up this

suggestion and traveled with Apache horsemen to hundreds ofeloca

tions in the region, he began to notice how place names were used

in everyday Apache conversations in ways that were very new to him.

He also spoke with consultants, asking about the stories associated

with various places. Through entertaining vignettes and engrossing

storytelling, Basso explains how the richly descriptive Western Apache

uses of language and place names (such as “Whiteness Spreads Out

Descending to Water,” “She Carries Her Brother on her Back,” and

https://Bizarro.com

1 4 The Socially Charged Life of Language I

“Shades of Shit”) help reinforce important Apache cultural values.

For example, Western Apache speakers invoke these place names in

conversations to allude indirectly to cautionary tales from recent or

ancient history that may be relevant to the current speakers’ dilen1-

mas. This practice, called “speaking with names,” is a verbal routine

that “allows those who engage in it to register claims about their own

n1oral worth, about aspects of their social relationships with other

people on hand, and about a particular way of attending to the local

landscape that is avowed to produce a beneficial form of heightened

self-awareness” (Basso 1996:81) . In this book, then,eBasso shows how

the physical environment is filtered through language to solidify

social relations and strengthen Western Apache notions of wisdon1

and morality.

Marjorie Harness Goodwin

In her book, He-Said-She-Said: Talk As Social Organization Among
Black Children, Marjorie Harness Goodwin (1990) chooses a very

different focus: that of a mixed-age and mixed-gender neighbor

hood group of peers in a Philadelphia neighborhood. By analyzing

“situated activities” such as argun1ents, storytelling, and gossip,

Goodwin shows how the children’s relationships and values are

reflected in and shaped by their conversations. Her meticulously

transcribed conversations (over 200 hours of tape recordings) pro

vide evidence for the complexity of children’s social worlds. They

also demonstrate the necessity of situating any analysis of language

and gender ( or any other social dimension of difference) in actual

contexts, for when this sort of study is undertaken, Goodwin notes,

stereotypes about so-called “female” speech patterns fall apart

(Goodwin 1990:9) . Boys and girls do not use language in two

completely different ways, Goodwin discovered, but rather interact

in san1e-sex and n1ixed-sex groups using complex, overlapping sets

of linguistic practices. In studying phenon1ena such as gender dif

ferences, therefore, Goodwin argues, it is essential to look closely

at actual conversations, for “talk itself is a form of social action, so

that any rigorous account of hun1an interaction nrnst pay close

attention to the detailed structure of talk that occurs within it”

(Goodwin 1990:2) .

The Socially Charged Life of Language 15

Bonnie Urciuo l i

The focus of Bonnie Urciuoli’s (1996) ethnography, Exposing Preju

dice: Puerto Rican Experiences �f Language, Race, and Class, is “language

prejudice” – the ways in which Puerto Ricans in New York City’s

Lower East Side experience, accept, or resist the judgments that they

and others make about what constitutes “good” and “bad” language,

whether Spanish, English, or a mixture. There is a “political economy”

of language, Urciuoli argues, the workings of which she explains as

follows:” [T] he ways in which people formulate, value, and use words,

sounds, phrases, and codes are constituted through power relations:

bureaucratic, econon1ic, racial, and any combination thereof” (1996:4) .

The boundaries between Spanish and English can be clearly den1ar

cated or fuzzy, depending on the context. When the socioeconomic

class of the speakers is similar, as when Lower East Side Puerto Rican

men are playing basketball with their English-speaking African

American neighbors, shifting between Spanish and English (” code

switching”) occurs n1ore fluidly and comfortably, for exan1ple, though

the ways in which this happens differs according to gender, Urciuoli

finds. In contrast, when there is a stark difference in socioeconomic

class, race, or ethnicity between speakers, Urciuoli notes, the bounda

ries between Spanish and English are strictly enforced, so little if any

code-switching occurs, for example, in interactions between Puerto

Ricans and white social workers, even when those social workers may

speak son1e Spanish. Language use is therefore an important part of

unequal social relations, Urciuoli maintains, as it both reflects and

sometimes reinforces differences in status.

Alessandro Duranti

Alessandro Duran ti ( 1994) explores language use in a very different

part of the world. His ethnography, From Grammar to Politics: Linguistic

Anthropology in a J;l;estern Samoan Village, analyzes political rhetoric in

the local village council (fono) and shows how speechmakers’ seemingly

apolitical, technical choices of grammatical markers can have impor

tant political ramifications. Duranti argues persuasively that a close look

at the n1icro level of gran1mar – at one tiny Samoan grammatical par

ticle in particular – offers important insights into how “the choice of

1 6 The Socially Charged Life of Language I

specific linguistic framings for people’s actions, beliefs, and feelings does

not simply reflect existing power relations, it also constitutes them”

(1994:139) . In other words, how people describe their actions, beliefs,

and feelings – how they fran1e them linguistically – both influences and

is influenced by the power dynamics of the community.Just as the title

of Duranti’s book indicates, a grammatical analysis, when situated in

actual social contexts, can lead to a better understanding of both gram

niar and politics.

Alexandra Jaffe

Alexandra Jaffe’s (1999) ethnography, Ideologies in Action: Language

Politics on Corsica, also investigates the intersections between language

and politics, though she takes a more macro-level focus than Duranti

in her research on the activities and attitudes of language activists and

ordinary residents on the Mediterranean island of Corsica. Jaffe looks

at the different statuses of the two n1ain languages spoken on the

island, French and Corsican, and shows how attitudes toward them are

intertwined with issues of cultural identity and economic and politi

cal power. She argues that “language planning/revitalization is an

immensely complex process . . . there are no neutral or purely linguis

tic choices or policies. Language choices and language form are heav

ily invested, in Corsica, with social and political significance” Qaffe

1999:7) .

James M. Wilce

James M. Wilce’s (1998) ethnography, Eloquence in Trouble: The Poetics

and Politics �f Complaint in Rural Bangladesh, looks closely at “troubles

talk,” or complaints, including the special genre of laments (impro

vised crying songs) in Bangladesh. The “eloquence in trouble” of

Wilce’s title has two n1eanings: Bangladeshis who resort to laments

to describe their suffering are often quite eloquent; and these sorts

of laments are becoming less and less common, and therefore repre

sent a genre in trouble – that is, in danger of disappearing. Wilce’s

interest in n1edical and psychological anthropology leads hin1 to pay

special attention to the lan1ents of people others label “crazy.” In so

doing, Wike den1onstrates how laments are n10re than just lengthy,

I The Socially Charged Life of Language 17

monologic complaints; instead, they are aesthetic performances and

social interactions during which labels can be both attached and

resisted by the perforn1er and the audience men1bers, and realities

can be “officialized” (1998:201) . A focus on linguistic practices such

as laments sheds light not only on the experiences of particular indi

viduals’ sufferings, Wike argues, but also on broader cultural ideas

about appropriate and inappropriate ways to speak and act, especially

for Bangladeshi won1en.

What these six very different ethnographies have in common is

their insistence that (1) language must not be studied in isolation from

social practices or cultural meanings, and (2) questions about social

relations and cultural n1eanings can best be answered by paying close

attention to language. The remainder of this book presents a detailed

case for each of these assertions.

Key Terms in Linguistic Anthropology

In order to provide readers with some tools they can use to approach

linguistic anthropology, I have chosen four key tern1s that provide

insight into the socially embedded nature oflanguage and the linguis

tically mediated nature of social life: multijunctionality, language ideologies,

practice, and indexicality. These terms draw upon an array of theoretical

approaches from within the field of linguistic anthropology and

beyond. As a rule in this book I try to avoid jargon, but linguistic

anthropology is no different from other fields such as chemistry or art

in having developed a set of specialized terms in order to refer effi

ciently and accurately to important concepts. The tern1s that I have

chosen here are “key” in two ways: first, they are central to the main

areas of research in the discipline, and second, they can provide read

ers with important keys to understanding the social nature of lan

guage because they con1e fron, the social and linguistic theories that

have had the greatest influence on current scholarship in the field.

Like the terms that are defined in Duranti’s (2001) edited volume, Key

Terms in Linguistic Anthropology, the four terms defined below illustrate

some of the features that unify the discipline and will therefore pro

vide common points of reference as we consider specific topics and

areas of study within the field.

1 8 The Socially Charged Life of Language I

3. context
(Referential function)

4. message
(Poetic function) 1 . speaker 2 . addressee

(Expressive function) � ——————————-➔ (Conative function)

5 . contact
(Phatic function)

6 . code
(Metal ingu istic function)

Figure 1.4 Jakobson’s model of the multifunctionality of language.

Source: Thomas A. Sebeok, Sty l e in Language, pp. 150, 154, 350-377, © 1960
Massachusetts Institute ofTechnology, by permission ofThe MIT Press.

Mu ltifu nct iona l ity

In the mainstream. view of language that is very con1 .n1 .on in the

United States, language is thought to be a way to describe events or to

label objects or concepts. Language is much more than this, however –

people accomplish many things with words. Linguistic anthropolo

gists use the term “multifunctional” to refer to all the different kinds

of work that language does. One of the first scholars to analyze the

various functions of language was Roman Jakobson, a Russian lin

guist who helped form what became known as the “Prague School”

of linguistic theory. Jakobson (1960) identifies six “constitutive fac

tors” in any speech event, and then attaches a corresponding function

to each of these constitutive factors. All functions are always present in

each speech event, Jakobson argues, but in certain cases, one function

may predominate over the others. Figure 1.4 is a slightly modified

version ofJakobson’s own model (1960:150, 154) .

Jakobson’s multifunctional model can be understood in the follow

mg way:

1 If an utterance ( or what Jakobson calls the “message”) is primarily

oriented toward the speaker, the predominant function is expressive.

Examples include “Ouch!” when someone stubs a toe, or “I feel

so embarrassed!” Of course, these kinds of speech events also

function in other ways, but to the degree that they mainly express

the speaker’s feelings or opinions,Jakobson considers the predom

inant function to be expressive.

https://con1.n1.on

I The Socially Charged Life of Language 19

2 If an utterance is primarily oriented toward the addressee,Jakobson

states that the predominant function is conative, an uncommon

word Jakobson used to denote “addressee-oriented.” Examples in

which this function is the principal one would be questions or

commands, as they are focused mainly on the addressee, or voca

tives (“Hey, Susie!”) .

3 When the utterance is oriented largely toward a third person,

toward the context, or toward events, Jakobson states that the

primary function is referential. Examples include, “The Dow

Jones plummeted 500 points today,” or, “Nepal is sandwiched

between India and China.” These types of utterances forn1 the

core of the folk n10del of language mentioned at the outset of

this section; for many people, the referential function is the

assumed main or even sole function oflanguage.Jakobson argues,

however, that referentiality is but one of six functions oflanguage,

and other scholars (e.g., Rosaldo 1982) have shown that it is not

even considered to be the default mode of conmrnnication in

son1e cultures.

4 When the utterance is oriented primarily toward itself – when it

son1ehow calls attention to the very sounds and patterns that are

used in its articulation – this makes the poetic function predomi

nate,Jakobson asserts. By “poetic,” Jakobson does not mean poetry

per se; instead, he is referring to occurrences in everyday speech

that involve rhyn1e, alliteration, repetition, parallelism, or other

sorts of playing around with the sound or structure of words.

Examples of the poetic function being evident outside of poetry

include political slogans such as, “I like Ike” (i.e. , Eisenhower) , and

grammatical parallelisn1 such as that which occurs in a staten1ent

such as, “I don’t want to hear you, I don’t want to see you, and

I don’t want to know you!” Jakobson has a great deal more to say

about the poetic function, but this will have to suffice for our

purposes.

5 If the utterance is oriented priniarily toward the channel that carries

it, whether the channel is social or physical, the associated function

is phatic, according to Jakobson. 9 An example of this type of utter

ance would be, “Testing, 1, 2, 3 . . . ,” as it focuses mainly on the

physical channel or mode of contact (a n1icrophone) between the

20 The Socially Charged Life of Language I

speaker and the addressee(s). When the channel is a more abstractly

conceived social connection rather than a physical one – a relation

ship of friendship or kinship, for example – an utterance that orients

itself prin1arily to this connection would also be considered niainly

phatic in function. An example of this would be the common

exchange, “Hi – how are you?” and the reply, “Fine, thanks.” In most

instances, the niain function of this question and its answer is to

draw attention to (and thereby reinforce) the social connection

between the two speakers. Sometimes the predominant function of

entire conversations can be said to be phatic – that is, mainly serving

the function of maintaining or solidifying a social connection.

6 If the utterance is oriented prin1arily toward language itself, the

predominant function is metalinguistic. Examples include, “Do

you understand what I just said?”, “How do you spell ‘relief’ ?”,

or, ‘”Metalinguistic’ n1eans ‘language that is about language.”‘

Son1e n1etalinguistic comments can be about language use (what

Silverstein [1993] calls “n1etapragn1atic” discourse) rather than

language structure, as in “It’s never appropriate to tell a joke on

the first date.” Many of these kinds of utterances are also exam

ples oflanguage ideologies, which will be discussed further below.

For some linguistic anthropologists, the n1etalinguistic function

of language represents the quintessentially human ability to be

reflexive about one’s own language use – in other words, the

ability to use language in reported speech or to reflect upon lin

guistic practice, structures, and contexts of use (cf. Agha 2007;

Lucy 1993) .

So, language is multifunctional; it accomplishes nmch more than sin,

ply referring to or labeling items or events. Through language, people

convey nuanced emotions, display or hide judgmental attitudes about

others, reinforce or sever social bonds, and talk about language itself.

It is to this latter function ofelanguage that we now turn.

Language ideo logies

Language ideologies1 0 are the attitudes, opinions, beliefs, or theories

that we all have about language. We may or may not be conscious of

them, and they may or may not square with scholars’ views about

The Socially Charged Life of Language 21

Figure 1.5 Cartoon playing off the language ideology that considers French a

romantic language.

Source: www.CartoonStock.com

language (which are also, of course, language ideologies) . Language

ideologies can be about language as a whole (e.g. , “Language is what

separates humans from other species”) , particular languages (e.g.,

“French is such a romantic language!”) , particular linguistic structures

(e.g., “Spanish is con1plicated as it has two forms of the verb ‘to be”‘ ) ,

language use (e.g., “Never end a sentence with a preposition”) – or

about the people who employ specific languages or usages (e.g.,

“People who say ‘ain’t’ are ignorant,” or, “People who live in the

United States should speak English,” or, “Won1en are n1ore talkative

than men”) . Scholars working within this fast-growing, exciting area

of language ideology study, for example, how socially and politically

influenced attitudes toward an endangered language can affect the

likelihood of its survival (e.g., Jaffe 1999; Kulick 1998) , or how teen

agers and adults embrace or reject ways of speaking that link then,

with various racial, ethnic, or gendered identities (e.g., Briggs 1998;

Bucholtz 2001; Cameron 1997; Cutler 2003; Gaudio 2001; Kroskrity

2000a) .

In almost all cases, language ideologies turn out to be about much

more than just language. As Judith Irvine notes, language ideologies

are “the cultural ( or subcultural) system of ideas about social and lin

guistic relationships, together with their loading of moral and political

www.CartoonStock.com

22 The Socially Charged Life of Language I

interests” (1989:255) . Language ideology as a concept therefore allows

scholars to connect micro- with macro-level social interactions and to

analyze questions of cultural identity, n1orality, power, inequality, and

social stereotypes. Paul Kroskrity (20006:8-23) lists four features that

characterize language ideologies.

1 Language ideologies almost always serve the interests of a specific

social or cultural group. In other words, in the uneven social ter

rain that exists in all con1nmnities, language ideologies come to

express the judgments and stereotypes of particular segn1ents of

each community. There are benefits to be had from certain lan

guage variants being deemed “standard” while others are labeled

“sub-standard dialects” or “slang.”

2 Language ideologies in any given society are best conceived of as

multiple because all societies consist of many different divisions and

subgroupings. There will therefore be niany different ideas about

language in any single community. Moreover, people can belong to

niany different social groups simultaneously and may therefore

hold nmltiple (sometin1es contradictory) language ideologies.

3 People may be more or less aware of their own or others’ language

ideologies. Certain types of ideas about language use or linguistic

structures tend to be more accessible to people, while others are

less so (Silverstein 1979, 2001) . At tin1es, language ideologies

become the subject of public debate, as happened during the

1996-1997 Ebonics controversy, and these occurrences can be

very illuminating to study. Just as interesting and potentially even

n1ore powerfully influential, however, are the language ideologies

that people do not realize that they hold.

4 People’s language ideologies mediate between social structures

and forms of talk. This bridging of micro-level speech and n1acro

level social structures is one of the most in1portant contributions

a study of language ideologies can make.

In many respects, therefore, attention to language ideologies can help

scholars in linguistic and cultural anthropology (and beyond) under

stand how both language and culture shape and are shaped by human

actions. To understand this recursive relationship further, we turn

now to the concept of practice.e1 1

The Socially Charged Life of Language 23

Practice

Consider Marx’s famous words in “The Eighteenth Brumaire of Louis

Bonaparte”: “Men nuke their own history, but they do not make it

just as they please; they do not make it under circun1stances chosen by

themselves, but under circumstances directly found, given and trans

mitted from the past” (Marx 1978[1852] :595) . In place of the word

“history” in this ren1ark, one could easily substitute “language,” “soci

ety,” or “culture,” and the statement would ren1ain equally insightful.

At the core of what is known as “practice theory” is this seeming

paradox: that language, culture, and society all apparently have a pre

existing reality but at the same time are very nrnch the products of

individual hunians’ words and actions. 12 Many linguistic anthropolo

gists explicitly or implicitly draw upon practice theory in their work.

The basic idea underlying practice theory is that structures (both

linguistic and social) at the same time constrain and give rise to hun1an

actions, which in turn create, recreate, or reconfigure those same

structures – and so on, with structures and actions successively giving

rise to one another.This kind ofhun1an action – that which is en1bed

ded within social and linguistic structures, that which both reflects

and shapes such structures – is known as “practice” or “agency.” Many

practice theorists define practice further as being in1bued with din1en

sions of inequality. Sherry Ortner, for exan1ple, considers any forn1 of

human action or interaction to be practice “insofar as the analyst rec

ognized it as reverberating with features of inequality, domination,

and the like in its particular historical and cultural setting” (1989:11-12;

see also Ortner 1984) . “Practice,” Ortner goes on to assert, “emerges

from structure, it reproduces structure, and it has the capacity to trans

form structure” (Ortner 1989:12) .

Practice theorists are interested in questions of social reproduction

and social transformation – why, in other words, things sometimes

change and sometimes remain the same. One concept practice theorists

have used to explain this process is Bourdieu’s notion of habitus, which

he uses to refer to a set of predispositions that produce practices and

representations conditioned by the structures from which they emerge.

These practices and their outcomes – whether people intend them to

do so or not – then reproduce or transform the habitus (Bourdieu

1977:78) . Habitus is a difficult concept but one that is potentially very

24 The Socially Charged Life of Language I

illuminating, for it can be used to describe how people socialized in a

certain way will often share many perspectives and values, as well as

styles of eating, talking, or behaving. To sin1plify, habi tus refers to how

we are predisposed (though not required) to think and act in certain

ways because of how we have been socialized. And usually, once we act

upon these predispositions, we end up reproducing the very conditions

and social structures that shaped our thoughts and actions to start with.

Not always, however. Because of the tensions and contradictions inher

ent in the habitus, actors are neither free agents nor completely socially

determined products. Instead, Ortner (1989: 198) suggests that they are

“loosely structured.” The central question for practice theorists, then, is

determining how such loosely structured actors manage at tin1es to

transform the systems that produce them.

Such loose structuring can occur linguistically as well as sociocul

turally. Speakers of a given language are constrained to some degree

by the grammatical structures of their particular language, but they are

still capable of producing an infinite number of grammatically well

forn1ed utterances within those constraints. Moreover, languages, like

cultures, change over time through drift and contact despite their sup

posedly self-reproducing structures. It is therefore helpful to look

closely at language (both its gran1n1atical structures and its patterns of

use) in order to gain a more thorough understanding of how people

reproduce and transform both language and culture.

Social systen1s – languages, habitus, structures, cultures, etc. – are

created and recreated, reinforced, reshaped, and reconfigured by the

actions and words of particular individuals, groups, and institutions

acting in socioculturally conditioned ways. In other words, languages

and cultures en1erge dialogically in a continuous manner through the

social and linguistic interactions of individuals “always already situated

in a social, political, and historical moment” (Mannheim and Tedlock

1995:9) . Neither structure nor practice, therefore, should be seen as

analytically prior to the other. Instead, each should be seen as being

en1bedded in the other. Social and linguistic structures en1erge fron,

the everyday actions of real people, and vice versa.

The concept of emergence as it is used here originated in biology,

and it goes beyond the simple everyday sense in which one thing

gives rise to another. In addition to this sense, emergence as it is used

in linguistic anthropology (as well as other fields) also refers to instances

I The Socially Charged Life of Language 25

when the whole is more than the sum of the parts. Ernst Mayr, the

famous biologist, writes of inorganic as well as organic systems that

they “aln1ost always have the peculiarity that the characteristics of the

whole cannot (not even in theory) be deduced fron, the most complete

knowledge of the components, taken separately or in other partial

combinations. This appearance of new characteristics in wholes has

been designated as emei;gence” (1982:63, emphasis in the original) .

Mayr is quick to point out that there is nothing mystical about such a

view of emergence; in fact, the characteristics (for example, its liquid

ity) of a system as “simple” as water cannot, according to Mayr, be

deduced from a study of its hydrogen and oxygen atoms. Similarly,

language, culture, and social structures emerge from social practice on

the part of individuals but cannot be understood with reference only

to those individuals.

Nevertheless, emergence does not imply absolute, unconstrained

unpredictability. On the contrary, Mannheim and Tedlock (1995:18)

emphasize that cultures have their own organizing principles that

emerge through the linguistic and social interactions of individuals who

themselves embody and enact social structures and cultural patterns,

just as practice theorists maintain. Take, for example, the actions of indi

viduals who are protesting something in their society by engaging

in street demonstrations. Their underlying assumptions, methods, and

principles are very likely to have been deeply influenced by the very

norms that they are protesting, even if the individuals work extremely

hard to counter such influences. What emerges from such formal pro

tests, as well as from informal, everyday activities, is shaped and con

strained by these influences – but not totally determined. Understanding

the constrained yet at least partially indeterminate outcomes of human

actions can help explain how social and linguistic structures that usually

reproduce themselves nevertheless always change over time. Whether

reproduction or transformation results, all languages and cultures can be

said to be emergent from social and linguistic practice.

lndexical ity

Identifying the precise ways in which language and social relations

intersect is one of the most pressing issues in linguistic anthropology.

A key concept that assists scholars in pinpointing these intersections is

26 The Socially Charged Life of Language

Object l nterp retant

correspondence

(b)
(a)

S ign

Figure 1.6 Semiosis as a relation between relations.

Source: From Kockelman (2007:377). Reproduced by permission of Paul

Kockelman. Current Anthropol oJiy, a journal published by University of

Chicago Press.

“indexicality” (Hanks 1999) , which, as it is used here, stems from

Charles Sanders Peirce’s sen1iotics (Peirce 1955; cf. Mertz 2007b) .

Semiotics, the study of signs, can be prohibitively difficult to grasp but

it is well worth going over some of the essentials in order to obtain a

fuller understanding of the term “indexicality.” Semiotics starts with

the definition of the linguistic sign. Perhaps the best-known defini

tion is de Saussure’s: a sign is the link between a concept (the “signi

fied”) and a sound pattern (the “signifier”) (Saussure 1986:66) . Thus,

in de Saussure’s famous example, the word “tree” is a sign because it

links the n1ental concept of a tree with the pattern of sounds that

comprises the word. For Peirce, however, sen1iosis, or meaning-nuking

through signs, involves a concept of the linguistic sign that is quite

different from de Saussure’s, for it is a process that “involves three

components: signs (whatever stands for son1ething else) , objects (what

ever a sign stands for) , and interpretants (whatever a sign creates insofar

as it stands for an object)” (Kockelman 2007:376; see Figure 1.6) . In

other words, meaning-making involves a sign such as the word “tree,”

the object that is represented, such as the actual tree – so far, these two

aspects could be said to be fairly similar to de Saussure’s “signifier” and

“signified” – but then there is in Peirce’s n1odel the extren1ely

I The Socially Charged Life of Language 27

important interpretant – the effect or outcome of the semiotic rela

tionship between the sign and the o�ject, such as a feeling of apprecia

tion for the beauty of a tree or the act of running away fron, smoke

for fear of a fire. Peirce’s tripartite signs do not reside solely in one

person’s head, therefore, as de Saussure’s signs do, but extend out into

the physical and social world.

There are three ways in which a sign can be related to its object,

according to Peirce, and it is the second of these ways that leads us to

the important concept of indexicality. These three types of signs – icon,

index, and symbole- are defined as follows (Peirce 1955:102-115): 13

• Icon. A sign that refers to its object by means of similarity. Exam

ples include photographs, diagrams, or sketches. Onomatopoeic

words (e.g. ,”choo choo train,””meow”) have an iconic dimension

because of the similarity in sound to that which they represent.
• Index. A sign that refers to its object “because it is in dynamical

(including spatial) connection both with the individual object, on

the one hand, and with the senses or memory of the person for

whom it serves as a sign, on the other hand” (Peirce 1955:107) . In

other words,ejust as an index finger points to an object, an indexi

cal sign “points to” its object through some connection or conti

guity, that is, a co-occurrence in the same context. Examples of

indexical signs include the classic one of smoke, which indexes

fire; a rolling gait, which indexes the profession of sailor; and a

clock, which indexes the time of day. Other indexical signs include

pronouns and words such as “here” or “now” because they are con

nected to (indeed, cannot be understood without knowledge of )

particular elements of the context. More will be said about this

property of indexicality below.
• Symbol. A sign that refers to its object by virtue of convention or

habit. Most words fall primarily into this category (though words

can have iconic, indexical, and/ or symbolic aspects simultane

ously) . The word “bird,” for example, does not represent its object

by virtue of similarity or any sort of”dynamical connection”; it is

simply conventional in English to call most flying animals with

wings “birds.” Some signs combine iconic or indexical features

with conventional ones. For example, it is conventional in English

to use the word” chickadee” to label a small black, white, and grey

28 The Socially Charged Life of Language I

bird – but this symbol also has an iconic aspect to it because the

name of the bird resembles the bird’s call, which sounds like

“chick-a-dee-dee-dee.”

While all three of these types oflinguistic signs have been employed by

linguistic anthropologists in their analyses, Peirce’s concept of the

indexical sign has drawn a great deal of attention in recent decades

because of its potential for showing how and where linguistic forms

“point to” aspects of social or cultural contexts. Certain categories of

words have been closely studied because they are completely context

dependent in that they inherently refer to particular mon1ents in time

or places in space (“here,” “then,” “now,” “there”) or social actors

(“you,” “I,” “that person;’ “such individuals”) . In order to understand

to whom “you” refers, for example, one must know the specific context

of the conversation or text in question. And these sorts of references

can shift; the person referred to as “you” can easily becon1e “I” ( or vice

versa) , and in reported speech a staten1ent such as, ‘Tn1 already here,”

can be reported using different words and verb tenses – for exan1ple,

“You said that you were already there.” 14

In addition to indexicals that refer to specific tin1es, places, indi

viduals, objects, or concepts, there are also more general ways in which

language can be indexical. In other words, as Jakobson has already

informed us, language can “point to” something social or contextual

without functioning in a referential way.eAspects ofelanguage use such

as regional or ethnic “accents” or “dialects,” for instance, “point to”

the speaker’s origins and are therefore examples of nonreferential or

“pure” indexicality (Silverstein 1976:29) .

Other indexicals have both referential and nonreferential functions.

The Nepali pronouns and verb forn1s used in the Pounded Rice

Ritual described at the outset of this chapter, for example, index not

just the particular addressee (the bride) but also her social position as

it plun1n1ets from the relatively high status of daughter to the lowly

status of daughter-in-law. Silverstein n1aintains that such indexes can

call into being the very social relations that they are indexing (1976:34) .

Similarly, the various words the San Francisco high school students

used for “stoned” index their youth status and most likely men1ber

ship in various social groups as well. Indexicality is also an in1portant

concept for understanding the disappearance of the language of Taiap

I The Socially Charged Life of Language 29

in Papua New Guinea, as it indexed certain social identities the villag

ers had come to devalue. Much more will be said about these sorts of

situations, as well as n1any others, throughout the rest of the book. For

our purposes here, it is important to realize the centrality of the con

cept of indexicality. Duranti writes,

To say that words are indexically related to some “object” or aspect
of the world out there n1eans to recognize that words carry with
them a power that goes beyond the description and identification
of people, obj ects, properties, and events. It means to work at iden
tifying how language becon1es a tool through which our social and
cultural world is constantly described, evaluated, and reproduced.
(1997 :e19)

The concept of indexicality is powerful but also extremely complex,

culturally and linguistically specific, and, therefore, quite challenging

to study (Hanks 1999:125) . Nevertheless, acknowledging the socio

culturally en1bedded nature of language is the first step toward being

able to shed further light on how indexicality works. Here are just a

few examples of the subtle ways in which language can index social

relations, identities, or values, “pointing to” such important aspects of

the sociocultural world and even creating, reinforcing, or challenging

those very relations, identities, or values:

• A college student min1ics the voice of a character on a television

comedy show, thereby indirectly referencing not only that charac

ter and that show but also indicating that she is the sort of cool,

hip, in-group sort of person who watches such a show.
• Labeling son1eone as an “enemy con1batant,” a “freedom fighter,”

a “terrorist,” or an “insurgent” can index the speaker’s political

views about the conflict in question and can also sometimes estab

lish, strengthen, or transform legal, n1ilitary, or political under

standings, thereby having real effects in the social world.
• Code-switching between two languages, dialects, or social regis

ters can index different processes involved in a person’s ethnic,

racial, gender, and/ or socioeconon1ic identity formation and can

have different social or even n1oral connotations, depending on

the situation.

30 The Socially Charged Life of Language I

As Silverstein notes: “Some of us have long since concluded that such

phenomena are indexical all the way down” (2006:276) .

The Inseparabi l ity of Language, Culture,
and Social Relations

The rest of this book will provide concrete examples of how these

four concepts – nrnltifunctionality, language ideologies, practice, and

indexicality – are being applied in the field of linguistic anthropology.

In the process, the following chapters will also attempt to reach two

specific kinds of readers of this book: those who believe that language

should be studied in a technical way, isolated fron1 any actual instance

of its use, and those who believe that social relations and cultural val

ues should be studied without a close analysis of linguistic practices.

To these readers, and indeed to all other readers as well, I hope to

demonstrate in the following pages that language, culture, and social

relations are so thoroughly intertwined that they must be studied in

connection with one another. The field of linguistic anthropology

provides son1e of the necessary tools for arriving at a deeper under

standing of such linguistic, cultural, and social phenomena.

Structure Bookmarks

Part I Language:

Some Basic Questions

Artifact

1 The Socially Charged Life of Language

All words have the ‘taste’ of a profession, a genre, a tendency, a party, a

particular work, a particular person, a generation, an age group, the d

and hour. Each word tastes of the context and contexts in which it has lived its socially charged life … Bakhtin 1981 :293

Words do live socially charged lives, as Bakhtin observes in the epigraph that opens this chapter. Language is not a neutral n1edium for con,munication but rather a set of socially embedded practices. The reverse of Bakhtin’s statement is also true: social interactions live linguistically charged lives. That is, every social interaction is mediated by language -whether spoken or written, verbal or nonverbal. Consider the following three examples.

Example 1: Getting Stoned in San Francisco

During the 1995-1996 school year, a special anti-drug class was run as an elective in a large high school in the San Francisco BArea. Students were trained as peer educators in preparation for visiting other classes to perform skits about the danger of drugs and tobacco. The class was unusually diverse, with bs as well as girls and with students from many different class ranks, ethnicities, and racial groups. On the dthat the students were preparing to perform their skits in front of an audience for the fir

Figure

Figure 1.1 Cartoon demonstrating how certain styles of speech can both reflect and shape social identities. Source: Jump Start (\’:) 1999 United Feature Syndicate, Inc.

The Socially Charged Life of Language

N6 l<'.ON6,

Al'l'lTt11N6 Gf'atJ5 oiJ Tile W”1T,

,1(, !’o0f{£

Mlour

t;o Ł St.</ITCH6"0 1o ? ii-' ŁA!ZWN' AtlO fŁ E

-'1:':'ŁŁ.–1

what they should say if someone in the audience asked whether they themselves smoked marijuana. Priscilla recommended that they say they did not. Then the following exchange took place between Priscilla and the students:

Priscilla: Remember, you're role models.

Al Capone: You want us to lie?

Priscilla: Since you're not coming to school stoned -(students

laugh)

Calvin: (mockingly) Stoned?

Priscilla: What do you say?

Calvin: I say high. Bombed. Blitzed.

Brand One: Weeded.

Kerry: Justified.

Brand One: That's kinda tight.

Example 2: Losia Language in Papua New Guinea

In 1987, the residents of the tivillage of Gapun in Papua New Guinea (a country north of Australia) were son1e of the last speakers of a language called Taiap, which at the tin1e had at most 89 reniaining speakers.Adult villagers were almost all bilingual in Taiap and in Tok Pisin, one of the three national languages of Papua New Guinea, and all children were exposed to rich amounts of both Taiap and Tok Pisin in their early years. By 1987, however, no child under the age of ten actively spoke Taiap, and ma

The Socially Charged Life of Language

about how and why so many of the world's languages are becoming extinct did not seem to apply to Taiap. Material and economic factors such as industrialization and urbanization were not sufficiently in1portant in the ren1ote village of Gapun to explain the language shift away from Taiap. Why, then, was Taiap becoming extinct? According to linguistic anthropologist Don Kulick, the adults in Gapun clain1ed that the shift was occurring because of the actions of their ( often preverbal) children. Kulick writes

Example 3: The Pounded Rice Ritual in Nepal

On a warn, February afternoon in 1993, a wedding procession made its way down a steep hill in Junigau, Nepal. Several men carefully maneuvered the bride's sedan chair around the hairpin turns. At the foot of the hill, under a large banyan tree, the wedding party settled down to rest and to conduct the Pounded Rice Ritual. The bride, Indrani Kumari, remained in her palanquin, while some members of the wedding party, including the groom, Khim Prasad, approached her. Taking out a leafplate full of pounded rice

But this first request was not very effective. lndrani Kumari and her bridal attendant poured just a few kernels of the pounded rice into the handkerchief Khim Prasad was holding. Upon further coaching from his elders, Khim Prasad asked a second time for the rice, this time in a more informal manner using "timi," a form of "you" in Nepali that is considered appropriate for close relatives and/ or familiar equals.This time, Khim Prasad's request could be translated roughly as a matter-of-fact statement to s

The Socially Charged Life of Language

Figure

Figure 1.2 Khim Prasad (left) during the Pounded Rice Ritual, with the bride, Indrani Kumari (seated at the right, completely covered by a shawl), and the bridal attendant (standing in the center). Source: Laura M.Ahearn, Invitations to Love: Literacy, Love Letters, and Social Change in Nepal. Ann Arbor: University of Michigan Press, 2001.

again, the bridal attendant and Indrani Kun1ari poured only a few kernels of pounded rice into Khin1 Prasad's waiting handkerchief. One last time Khim Prasad's senior male kin instructed him to ask for the rice, but this time he was told to use "ta," the lowest form of"you" in Nepali -a forn1 most con1n1only used ineJunigau to address young children, animals, and wives. Khim_ Prasad complied, but his words were halting and barely audible, indicating his deeply n1-ixed feelings

The Socially Charged Life of Language

about using such a disrespectful term to address his new wife. This third request translated roughly as a peremptory command to someone of greatly inferior social status: "Bring the pounded rice, Wife! Our wedding party has gotten hungry!" Hearing this, Indrani Kun1ari and her attendant finally proceeded obediently to dump all the remaining rice into the groom's handkerchief, after which he handed out portions of the snack to all members of the wedding party.

As different as these three examples are, they all describe situations in which neither a linguistic analysis alone nor a sociocultural analysis alone would come close to providing a satisfying explanation of the significance of the events. The purpose of this book is to show how the perspectives and tools oflinguistic anthropology, when applied to events as wide-ranging as an anti-drug class in a San Francisco high school, language shift in Papua New Guinea, or a ritual in Nepal, can shed light on broader

•

What can such situations tell us about the ways in which language is enn1eshed with cultural values and social power?

•

How do dimensions of difference or inequality along lines such as gender, ethnicity, race, age, or wealth get created, reproduced, or challenged through language?

•

How can language illun1inate the ways in which we are all the same by virtue of being human as well as the ways in which we are incredibly diverse linguistically and culturally?

•

How, if at all, do linguistic forn1s, such as the three different words in Nepali for "you" or the various slang words for "stoned," influence people's thought patterns or worldviews?

•

How might people's ideas about language (for example, what "good" language is and who can speak it -in other words, their "language ideologies") affect their perceptions of others as well as themselves?

•

How does the language used in public rituals and performances both differ fron1 and resen1ble everyday, mundane conversations?

•

What methods of data collection and analysis can we use to detern1ine the significance of events such as those described above?

The Socially Charged Life of Language I

The starting point in the search for answers to all of these questions within linguistic anthropology is this fundamental principle: language is inherently social. It is not just a means through which we act upon the social world; speaking is itself a form of social action, and language is a cultural resource available for people to use (Durantido things with words, as the philosopher J.L. Austin (1962) reminded us decades ago. Even when we speak or write to ourselves, our very choices of words, as well as

1997:2).We

contain soup.

This approach to language differs from the popular view of language as an empty vehicle that conveys pre-existing meanings about the world. Language, according to this view, which is held by many n1ernbers of the general public as well as many linguists and other scholars, is largely a set of labels that can be placed on pre-existing concepts, objects, or relationships. In this mistaken way of thinking, language is defined as a conduit that merely conveys information without adding or changing anything of

Within the field of linguistics, a similar approach to language is dominant: one in which language is reduced to a set of formal rules. Such reductionism extends back hundreds of years but was made the don1inant approach of the field oflinguistics by Ferdinand de Saussure, a famous Swiss linguist who lived a century ago. De Saussure maintained that it was not only possible but necessary to decontextualize the study of language: "A science which studies linguistic structure is not only able to dispense with

The Socially Charged Life of Language

an American linguist who revolutionized the field and has dominated it for the past 50 years. Chomsky and his followers are interested in discovering Universal Grammar (UG), which they define as: "The basic design underlying the gran1n1ars of all hun1an languages; [it] also refers to the circuitry in children's brains that allows them to learn the grammar of their parents' language" (Pinker 1994:483).

This is not to sthat linguistic anthropologists are uninterested in gran1n1ar or believe that linguistic forms cannot be studied systematically -on the contrary, many build upon the "considerable progress in the understanding of formal properties oflanguages" made by scholars in the field of linguistics (Duranti 1997:7), but they ask very different kinds of questions that explore the intersections between grammar and social relations, politics, or emotion. Even linguistic anthropologists who value the work

So, What Do You Need to Know in Order to "Know" a Language?

In order to understand what it means to study language as a linguistic anthropologist would, it is helpful to ask what it n1eans to "know" a language (Cipollone et al.1998) .Linguists generally use the Chomskyan distinction between "competence," the abstract and usually unconscious knowledge that one has about the rules of a language, and "performance," the putting into practice -son1etimes in1perfectly of those rules. De Saussure made a sin1ilar distinction between langue (the language system in the abstr

–

1 O The Socially Charged Life of Language I

To take the knitting analogy further, if Chomsky were a knittist instead of a linguist, he would be interested only in the abstract rules of Knitting (capitalizing the word, as he does with Language) such as the following: Row 20: P 1, (k 1, p 1) 11 (13-15) times, k 5, TR 2, k 4, TR 2, k 1,p 12, k 1, TL 2, k 4, TL 2, k 5,p 1, (k 1,p 1) 11(13-15) times.Chomsky the knittist would posit the existence of a Knitting Acquisition Device (KAD, rather than LAD, a Language Acquisition Device), a specialized module of

•

How or why people learn to knit in various cultures and communities.

•

How knitting practices have changed over time.

•

The gendered nature of knitting and other handicrafts in many societies (although knitting is often associated with girls and women in this society, for example, handicrafts such as weaving were until recently conventionally produced by lower-caste men in Nepal).

•

The role of Madame Defarge in A Tale �[Two Cities, by Charles Dickens, as she secretly encodes the names of counterrevolutionaries into her knitting.

•

The global economics involved in the many different yarns people use to knit -anything from yak wool fron, Nepal to Icelandic wool to synthetic mohair.

•

The many different kinds ofproducts ofeconomic, social, or emotional value that are made by knitters to be worn by themselves, given to loved ones, donated to charity, or sold to tourists.

•

The ways in which knitting is viewed by different groups in the society -as a hip, in-group practice by some, as an old, fuddyduddy practice by others, as a useful, money-making skill by yet others.

•

How one's individual and social identities can be reflected in and shaped by whether, how, what, and with whom one knits.

While this analogy with knitting is not by any means a perfect one, it does nevertheless demonstrate how narrowly Chomsky and most

The Socially Charged Life of Language 11

other linguists view language. Other practices such as pling music, dancing, or painting would work equally well in the analogy I set up above because knitting and all these other practices are -like language -socially embedded and culturally influenced. Of course there are abstract cognitive and biological dimensions to anything that we as humans do, including language, but to reduce language solely to these din1ensions, as Chon1sky and others do when they clain1 they are interested only in con1petence and

Linguistic anthropologists therefore reject the Chomskyan/Saussurean distinction between competence (langue) and performance (parole), though they do so in various ws. Some deny the existence of any distinction at all between competence and performance (langue and parole), while others give primacy to performance (parole). Still others either expand the definition of competence to include the ability to use language skillfully and appropriately in particular social contexts (cf. Hymes 2001[1972]), and many

What else must one know in order to know a language, then, aside from grammatical rules? According to Cipollone et al. (1998:8-11), there are five basic components of a language that can be studied, and one nmst n1aster all five of these areas in order to know a language:

•

Phonology. The study of sound in language. In order to know a language, one must be able to recognize and produce the sounds that are n1eaningful in that language. In the case of sign languages, instead of sounds, one must be able to recognize and produce the appropriate gestures.

•

Morphology.The study of the internal structure of words. In order to know a language, one nmst be able to use suffixes, prefixes, or infixes ( depending on the language). In English, for example, one must know how to create plurals by placing an "-s" on the end of most (but not all) words, and nmst know what adding "un-" to the beginning of a word does to its n1eaning. In n1any Native American languages, these sorts of affixes are placed inside a word to create infixes,

12 The Socially Charged Life of Language I

while in Chinese languages, each morpheme, or unit of meaning, is a separate word, including morphemes indicating tense or plurality.

•

Syntax.The study of the structure of sentences, including the construction of phrases, clauses, and the order of words. In order to know a language, one must be able to combine subjects, verbs, and objects in a grammatically correct way.

•

Semantics. The study of meaning in language, including analysis of the meanings of words and sentences. In order to know a language, one must know how to construct and interpret meanings.

•

Pragmatics. The study ofelanguage use, of actual utterances, of how n1eanings emerge in actual social contexts. This includes culturally and linguistically specific ws of structuring narratives, perforn1ances, or everydconversations. In order to know a language, one must be able to use language in socially and culturally appropriate ws.

–

Most linguists focus prin1arily or solely on one or n1ore of the first three components (phonology, n1orphology, or syntax), with syntax being accorded primacy ever since Chomsky became dominant in the field. In contrast, n1ost linguistic anthropologists (as well as some scholars in related fields such as sociolinguistics or discourse analysis) study the final two components (semantics and pragmatics) in ws that integrate these two components with the first three. Indeed, linguistic anthropologists conside

So, How Do Linguistic Anthropologists Study Language as Social Action?

While linguistic anthropologists hold in con1n1on the view that language is a form_ of social action, there is nevertheless great diversity in topic choice and research n1ethods within the field. Chapter 2 will

The Socially Charged Life of Language 13

Figure

Figure 1.3 "Zits" cartoon about the varying cultural meanings associated with language use. Source: Distributed by King Features Syndicate.

Reproduced with kind permission of Dan Piraro and Bizarro.com.

examine the various research methods used by linguistic anthropologists, so what I present here are son1e exan1ples of the topics scholars have chosen and an explanation of how these topics contribute to our understanding of language as a forn1 of social action. These studies illustrate but by no means exhaust the wide-ranging diversity of contemporary linguistic anthropology.

Keith Basso

Keith Basso's (1996) ethnography, Wisdom Sits in Places: Landscape and Language Among the Western Apache, explores "place-nuking" as a linguistic and cultural activity.eThis book was written after Ronnie Lupe, chairman of the White Mountain Apache tribe, asked Basso to help make some maps: "Not whitemen's maps, we've got plenty of them, but Apache maps with Apache places and names. We could use then,. Find out something about how we know our country. You should have done this before" (Basso 1996:xv). When

14 The Socially Charged Life of Language I

"Shades of Shit") help reinforce important Apache cultural values. For example, Western Apache speakers invoke these place names in conversations to allude indirectly to cautionary tales from recent or ancient history that may be relevant to the current speakers' dilen1mas. This practice, called "speaking with names," is a verbal routine that "allows those who engage in it to register claims about their own n1oral worth, about aspects of their social relationships with other people on hand, and about a part

–

Marjorie Harness Goodwin

In her book, He-Said-She-Said: Talk As Social Organization Among Black Children, Marjorie Harness Goodwin (1990) chooses a very different focus: that of a mixed-age and mixed-gender neighborhood group of peers in a Philadelphia neighborhood. By analyzing "situated activities" such as argun1ents, storytelling, and gossip, Goodwin shows how the children's relationships and values are reflected in and shaped by their conversations. Her meticulously transcribed conversations (over 200 hours of tape recordings)

The Socially Charged Life of Language 15

Bonnie Urciuoli

The focus of Bonnie Urciuoli's (1996) ethnography, Exposing Prejudice: Puerto Rican Experiences �f Langue, Race, and Class, is "language prejudice" -the ws in which Puerto Ricans in New York City's Lower East Side experience, accept, or resist the judgments that they and others make about what constitutes "good" and "bad" language, whether Spanish, English, or a mixture. There is a "political economy" of language, Urciuoli argues, the workings of which she explains as follows:" [T] he ws in which people fo

Alessandro Duranti

Alessandro Duran ti ( 1994) explores language use in a very different part of the world. His ethnography, From Grammar to Politics: Linguistic Anthropology in a J;l;estern Samoan Ville, analyzes political rhetoric in the local village council (fono) and shows how speechmakers' seemingly apolitical, technical choices of grammatical markers can have important political ramifications. Duranti argues persuasively that a close look at the n1icro level of gran1mar -at one tiSamoan grammatical particle in partic

16 The Socially Charged Life of Language I

specific linguistic framings for people's actions, beliefs, and feelings does not simply reflect existing power relations, it also constitutes them" (1994:139). In other words, how people describe their actions, beliefs, and feelings -how thfran1e them linguistically -both influences and is influenced by the power dynamics of the community.Just as the title of Duranti's book indicates, a grammatical analysis, when situated in actual social contexts, can lead to a better understanding of both gramniar and p

Alexandra Jaffe

Alexandra Jaffe's (1999) ethnography, Ideologies in Action: Language Politics on Corsica, also investigates the intersections between language and politics, though she takes a more macro-level focus than Duranti in her research on the activities and attitudes oflanguage activists and ordinary residents on the Mediterranean island of Corsica. Jaffe looks at the different statuses of the two n1ain languages spoken on the island, French and Corsican, and shows how attitudes toward them are intertwined with iss

James M. Wilce

James M. Wilce's (1998) ethnography, Eloquence in Trouble: The Poetics and Politics �f Complaint in Rural Bangladesh, looks closely at "troubles talk," or complaints, including the special genre of laments (improvised crying songs) in Bangladesh. The "eloquence in trouble" of Wilce's title has two n1eanings: Bangladeshis who resort to laments to describe their suffering are often quite eloquent; and these sorts of laments are becoming less and less common, and therefore represent a genre in trouble -that

The Socially Charged Life of Language 17

monologic complaints; instead, they are aesthetic performances and social interactions during which labels can be both attached and resisted by the perforn1er and the audience men1bers, and realities can be "officialized" (1998:201). A focus on linguistic practices such as laments sheds light not only on the experiences of particular individuals' sufferings, Wike argues, but also on broader cultural ideas about appropriate and inappropriate ways to speak and act, especially for Bangladeshi won1en.

What these six very different ethnographies have in common is their insistence that (1) language must not be studied in isolation from social practices or cultural meanings, and (2) questions about social relations and cultural n1eanings can best be answered by paying close attention to language. The remainder of this book presents a detailed case for each of these assertions.

Key Terms in Linguistic Anthropology

In order to provide readers with some tools they can use to approach linguistic anthropology, I have chosen four key tern1s that provide insight into the socially embedded nature oflanguage and the linguistically mediated nature of social life: multijunctionality, language ideologies, practice, and indexicality. These terms draw upon an array of theoretical approaches from within the field of linguistic anthropology and beyond. As a rule in this book I try to avoid jargon, but linguistic anthropology is no

18 The Socially Charged Life of Language I

3. context (Referential function)

4. message (Poetic function)

1. speaker 2. addressee (Expressive function) � ——————————-➔ (Conative function)

5. contact (Phatic function)

6. code (Metalinguistic function)

Figure 1.4 Jakobson's model of the multifunctionality oflanguage. Source: Thomas A. Sebeok, Style in Language, pp. 150, 154, 350-377, © 1960 Massachusetts Institute ofTechnology, by permission ofThe MIT Press.

Multifunctiona lity

In the mainstream. view of language that is very in the United States, language is thought to be a way to describe events or to label objects or concepts. Language is much more than this, however people accomplish many things with words. Linguistic anthropologists use the term "multifunctional" to refer to all the different kinds of work that language does. One of the first scholars to analyze the various functions of language was Roman Jakobson, a Russian linguist who helped form what became known as the

con1.n1.on

–

Jakobson's multifunctional model can be understood in the followmg way:

1 If an utterance ( or what Jakobson calls the "message") is primarily oriented toward the speaker, the predominant function is expressive. Examples include "Ouch!" when someone stubs a toe, or "I feel so embarrassed!" Of course, these kinds of speech events also function in other ways, but to the degree that they mainly express the speaker's feelings or opinions,Jakobson considers the predominant function to be expressive.

The Socially Charged Life of Language 19

2 If an utterance is primarily oriented toward the addressee,Jakobson states that the predominant function is conative, an uncommon word Jakobson used to denote "addressee-oriented." Examples in which this function is the principal one would be questions or commands, as they are focused mainly on the addressee, or vocatives ("Hey, Susie!").

3 When the utterance is oriented largely toward a third person, toward the context, or toward events, Jakobson states that the primary function is referential. Examples include, "The Dow Jones plummeted 500 points today," or, "Nepal is sandwiched between India and China." These types of utterances forn1 the core of the folk n10del of language mentioned at the outset of this section; for many people, the referential function is the assumed main or even sole function oflanguage.Jakobson argues, however, that

4 When the utterance is oriented primarily toward itself -when it son1ehow calls attention to the very sounds and patterns that are used in its articulation -this makes the poetic function predominate,Jakobson asserts. By "poetic," Jakobson does not mean poetry per se; instead, he is referring to occurrences in everydspeech that involve rhyn1e, alliteration, repetition, parallelism, or other sorts of pling around with the sound or structure of words. Examples of the poetic function being evident outside of

5 If the utterance is oriented priniarily toward the channel that carries it, whether the channel is social or physical, the associated function is phatic, according to Jakobson. An example of this type of utterance would be, "Testing, 1, 2, 3 … ," as it focuses mainly on the physical channel or mode of contact (a n1icrophone) between the

20 The Socially Charged Life of Language I

speaker and the addressee(s). When the channel is a more abstractly conceived social connection rather than a physical one -a relationship of friendship or kinship, for example -an utterance that orients itself prin1arily to this connection would also be considered niainly phatic in function. An example of this would be the common exchange, "Hi -how are you?" and the reply, "Fine, thanks." In most instances, the niain function of this question and its answer is to draw attention to (and thereby reinforce)

6 If the utterance is oriented prin1arily toward language itself, the predominant function is metaliuistic. Examples include, "Do you understand what I just said?", "How do you spell 'relief'?", or, '"Metalinguistic' n1eans 'language that is about language."' Son1e n1etalinguistic comments can be about language use (what Silverstein [1993] calls "n1etapragn1atic" discourse) rather than language structure, as in "It's never appropriate to tell a joke on the first date." Many of these kinds of utterances are

So, language is multifunctional; it accomplishes nmch more than sin,ply referring to or labeling items or events. Through language, people convey nuanced emotions, displor hide judgmental attitudes about others, reinforce or sever social bonds, and talk about language itself. It is to this latter function ofelanguage that we now turn.

Language ideologies

Language ideologiesare the attitudes, opinions, beliefs, or theories that we all have about language. We may or may not be conscious of them, and they mor mnot square with scholars' views about

The Socially Charged Life of Language 21

Figure

Figure 1.5 Cartoon playing off the language ideology that considers French a romantic language. Source:

www.CartoonStock.com

language (which are also, of course, language ideologies). Language ideologies can be about language as a whole (e.g., "Language is what separates humans from other species"), particular languages (e.g., "French is such a romantic language!"), particular linguistic structures (e.g., "Spanish is con1plicated as it has two forms of the verb 'to be"'), language use (e.g., "Never end a sentence with a preposition") -or about the people who employ specific languages or usages (e.g., "People who say 'ain't' are i

In almost all cases, language ideologies turn out to be about much more than just language. As Judith Irvine notes, language ideologies are "the cultural ( or subcultural) system of ideas about social and linguistic relationships, together with their loading of moral and political

22 The Socially Charged Life of Language I

interests" (1989:255). Language ideology as a concept therefore allows scholars to connect micro-with macro-level social interactions and to analyze questions of cultural identity, n1orality, power, inequality, and social stereotypes. Paul Kroskrity (20006:8-23) lists four features that characterize language ideologies.

1 Language ideologies almost always serve the interests of a specific social or cultural group. In other words, in the uneven social terrain that exists in all con1nmnities, language ideologies come to express the judgments and stereotypes of particular segn1ents of each community. There are benefits to be had from certain language variants being deemed "standard" while others are labeled "sub-standard dialects" or "slang."

2 Language ideologies in any given society are best conceived of as multiple because all societies consist of many different divisions and subgroupings. There will therefore be niany different ideas about language in any single community. Moreover, people can belong to niany different social groups simultaneously and may therefore hold nmltiple (sometin1es contradictory) language ideologies.

3 People may be more or less aware of their own or others' language ideologies. Certain types of ideas about language use or linguistic structures tend to be more accessible to people, while others are less so (Silverstein 1979, 2001). At tin1es, language ideologies become the subject of public debate, as happened during the 1996-1997 Ebonics controversy, and these occurrences can be very illuminating to study. Just as interesting and potentially even n1ore powerfully influential, however, are the language

4 People's language ideologies mediate between social structures and forms of talk. This bridging of micro-level speech and n1acrolevel social structures is one of the most in1portant contributions a study of language ideologies can make.

In many respects, therefore, attention to language ideologies can help scholars in linguistic and cultural anthropology (and beyond) understand how both language and culture shape and are shaped by human actions. To understand this recursive relationship further, we turn now to the concept of practice.e

The Socially Charged Life of Language 23

Practice

Consider Marx's famous words in "The Eighteenth Brumaire of Louis Bonaparte": "Men nuke their own history, but they do not make it just as they please; they do not make it under circun1stances chosen by themselves, but under circumstances directly found, given and transmitted from the past" (Marx 1978[1852]:595). In place of the word "history" in this ren1ark, one could easily substitute "language," "society," or "culture," and the statement would ren1ain equally insightful. At the core of what is known a

The basic idea underlying practice theory is that structures (both linguistic and social) at the same time constrain and give rise to hun1an actions, which in turn create, recreate, or reconfigure those same structures -and so on, with structures and actions successively giving rise to one another.This kind ofhun1an action -that which is en1bedded within social and linguistic structures, that which both reflects and shapes such structures -is known as "practice" or "agency." Many practice theorists define

Practice theorists are interested in questions of social reproduction and social transformation -why, in other words, things sometimes change and sometimes remain the same. One concept practice theorists have used to explain this process is Bourdieu's notion of habitus, which he uses to refer to a set of predispositions that produce practices and representations conditioned by the structures from which they emerge. These practices and their outcomes -whether people intend them to do so or not -then reproduc

24 The Socially Charged Life of Language I

illuminating, for it can be used to describe how people socialized in a certain way will often share many perspectives and values, as well as styles of eating, talking, or behaving. To sin1plify, habi tus refers to how we are predisposed (though not required) to think and act in certain ways because of how we have been socialized.And usually, once we act upon these predispositions, we end up reproducing the very conditions and social structures that shaped our thoughts and actions to start with. Not alws, h

Such loose structuring can occur linguistically as well as socioculturally. Speakers of a given language are constrained to some degree by the grammatical structures of their particular language, but they are still capable of producing an infinite number of grammatically wellforn1ed utterances within those constraints. Moreover, languages, like cultures, change over time through drift and contact despite their supposedly self-reproducing structures. It is therefore helpful to look closely at language (bo

Social systen1s -languages, habitus, structures, cultures, etc. -are created and recreated, reinforced, reshaped, and reconfigured by the actions and words of particular individuals, groups, and institutions acting in socioculturally conditioned ws. In other words, languages and cultures en1erge dialogically in a continuous manner through the social and linguistic interactions of individuals "alws already situated in a social, political, and historical moment" (Mannheim and Tedlock 1995:9). Neither structur

The concept of emergence as it is used here originated in biology, and it goes beyond the simple everydsense in which one thing gives rise to another. In addition to this sense, emergence as it is used in linguistic anthropology (as well as other fields) also refers to instances

The Socially Charged Life of Language 25

when the whole is more than the sum of the parts. Ernst Mr, the famous biologist, writes of inorganic as well as organic systems that they "aln1ost alws have the peculiarity that the characteristics of the whole cannot (not even in theory) be deduced fron, the most complete knowledge of the components, taken separately or in other partial combinations. This appearance of new characteristics in wholes has been designated as emei;gence" (1982:63, emphasis in the original). Mr is quick to point out that there

Nevertheless, emergence does not imply absolute, unconstrained unpredictability. On the contrary, Mannheim and Tedlock (1995:18) emphasize that cultures have their own organizing principles that emerge through the linguistic and social interactions of individuals who themselves embody and enact social structures and cultural patterns, just as practice theorists maintain. Take, for example, the actions of individuals who are protesting something in their society by engaging in street demonstrations. Their u

lndexicality

Identifying the precise ws in which language and social relations intersect is one of the most pressing issues in linguistic anthropology. A key concept that assists scholars in pinpointing these intersections is

26 The Socially Charged Life of Language

Object lnterpretant

correspondence

(b)

Figure

(a)

Sign

Figure 1.6 Semiosis as a relation between relations. Source: From Kockelman (2007:377). Reproduced by permission of Paul Kockelman. Current AnthropoloJiy, a journal published by University of Chicago Press.

"indexicality" (Hanks 1999), which, as it is used here, stems from Charles Sanders Peirce's sen1iotics (Peirce 1955; cf. Mertz 2007b). Semiotics, the study of signs, can be prohibitively difficult to grasp but it is well worth going over some of the essentials in order to obtain a fuller understanding of the term "indexicality." Semiotics starts with the definition of the linguistic sign. Perhaps the best-known definition is de Saussure's: a sign is the link between a concept (the "signified") and a sound

The Socially Charged Life of Language 27

important interpretant -the effect or outcome of the semiotic relationship between the sign and the o�ject, such as a feeling of appreciation for the beauty of a tree or the act of running away fron, smoke for fear of a fire. Peirce's tripartite signs do not reside solely in one person's head, therefore, as de Saussure's signs do, but extend out into the physical and social world.

There are three ways in which a sign can be related to its object, according to Peirce, and it is the second of these ways that leads us to the important concept of indexicality. These three types of signs -icon, index, and symbole-are defined as follows (Peirce 1955:102-115):

•

Icon. A sign that refers to its object by means of similarity. Examples include photographs, diagrams, or sketches. Onomatopoeic words (e.g.,"choo choo train,""meow") have an iconic dimension because of the similarity in sound to that which they represent.

•

Index. A sign that refers to its object "because it is in dynamical (including spatial) connection both with the individual object, on the one hand, and with the senses or memory of the person for whom it serves as a sign, on the other hand" (Peirce 1955:107). In other words,ejust as an index finger points to an object, an indexical sign "points to" its object through some connection or contiguity, that is, a co-occurrence in the same context. Examples of indexical signs include the classic one of smoke,

•

Symbol. A sign that refers to its object by virtue of convention or habit. Most words fall primarily into this category (though words can have iconic, indexical, and/ or symbolic aspects simultaneously). The word "bird," for example, does not represent its object by virtue of similarity or any sort of"dynamical connection"; it is simply conventional in English to call most flying animals with wings "birds." Some signs combine iconic or indexical features with conventional ones. For example, it is conventio

28 The Socially Charged Life of Language I

bird -but this symbol also has an iconic aspect to it because the name of the bird resembles the bird's call, which sounds like "chick-a-dee-dee-dee."

While all three of these types oflinguistic signs have been employed by linguistic anthropologists in their analyses, Peirce's concept of the indexical sign has drawn a great deal of attention in recent decades because of its potential for showing how and where linguistic forms "point to" aspects of social or cultural contexts. Certain categories of words have been closely studied because they are completely contextdependent in that they inherently refer to particular mon1ents in time or places in space ("

In addition to indexicals that refer to specific tin1es, places, individuals, objects, or concepts, there are also more general ways in which language can be indexical. In other words, as Jakobson has already informed us, language can "point to" something social or contextual without functioning in a referential way.eAspects ofelanguage use such as regional or ethnic "accents" or "dialects," for instance, "point to" the speaker's origins and are therefore examples of nonreferential or "pure" indexicality (

Other indexicals have both referential and nonreferential functions. The Nepali pronouns and verb forn1s used in the Pounded Rice Ritual described at the outset of this chapter, for example, index not just the particular addressee (the bride) but also her social position as it plun1n1ets from the relatively high status of daughter to the lowly status of daughter-in-law. Silverstein n1aintains that such indexes can call into being the very social relations that they are indexing (1976:34). Similarly, the var

The Socially Charged Life of Language 29

in Papua New Guinea, as it indexed certain social identities the villagers had come to devalue. Much more will be said about these sorts of situations, as well as n1any others, throughout the rest of the book. For our purposes here, it is important to realize the centrality of the concept of indexicality. Duranti writes,

To say that words are indexically related to some "object" or aspect of the world out there n1eans to recognize that words carry with them a power that goes beyond the description and identification of people, objects, properties, and events. It means to work at identifying how language becon1es a tool through which our social and cultural world is constantly described, evaluated, and reproduced. (1997:e19)

The concept of indexicality is powerful but also extremely complex, culturally and linguistically specific, and, therefore, quite challenging to study (Hanks 1999:125). Nevertheless, acknowledging the socioculturally en1bedded nature of language is the first step toward being able to shed further light on how indexicality works. Here are just a few examples of the subtle ways in which language can index social relations, identities, or values, "pointing to" such important aspects of the sociocultural world

•

A college student min1ics the voice of a character on a television comedy show, thereby indirectly referencing not only that character and that show but also indicating that she is the sort of cool, hip, in-group sort of person who watches such a show.

•

Labeling son1eone as an "enemy con1batant," a "freedom fighter," a "terrorist," or an "insurgent" can index the speaker's political views about the conflict in question and can also sometimes establish, strengthen, or transform legal, n1ilitary, or political understandings, thereby having real effects in the social world.

•

Code-switching between two languages, dialects, or social registers can index different processes involved in a person's ethnic, racial, gender, and/ or socioeconon1ic identity formation and can have different social or even n1oral connotations, depending on the situation.

30 The Socially Charged Life of Language I

As Silverstein notes: "Some of us have long since concluded that such phenomena are indexical all the way down" (2006:276).

The Inseparability of Language, Culture, and Social Relations

The rest of this book will provide concrete examples of how these four concepts -nrnltifunctionality, language ideologies, practice, and indexicality -are being applied in the field oflinguistic anthropology. In the process, the following chapters will also attempt to reach two specific kinds of readers of this book: those who believe that language should be studied in a technical way, isolated fron1 any actual instance of its use, and those who believe that social relations and cultural values should be s

Bernard, H. Russell, ed.
1998. Handbook of Methods

in Cultural Anthropology.
Walnut Creek: Altamira Press.

BRENDA FARNELL Æh
LAURA R. GRAHAM 88

Twelve

Discourse-Centered Methods

Discourse analysis comprises methods used by researchers across the social sciences
as well as in literary studies (see, for example, Josselson 1953 and Bernard and
Ryan in this volume). It is also used in clinical fields (see, for example, Labov and
Fanshel 1977; Cicourel 1981, 1982, 1992; West 1984). What unites this work is
analytic attention to the use of language in social contexts.

The term “discourse centered” (Sherzer 1987; Urban 1991) is an expedient label
with which to describe recent movements in linguistic anthropology. It does not,
however, constitute a single theoretical “school.”1 Discourse-centered approaches
draw on theoretical resources from several intellectual ancestries and build on earlier
work in sociolinguistics, the ethnography of speaking/communication, and perfor
mance approaches to language.2 These approaches distinguish themselves by
focusing on the dialogical processes through which persons, social institutions, and
cultural knowledge are socially constructed through spoken discourse and other
signifying acts/forms of expressive performance.

In current anthropological practice, discourse-centered methods are used in
participant observation, within the context of ethnographic fieldwork. Within this
general orientation, researchers pay close attention to how language is used in and
across social situations, focusing particularly on “naturally occurring discourse”—
that is, utterances that occur in the context of social interaction, in contrast to
utterances specifically elicited by a linguist or ethnographer.

We focus here on approaches in linguistic anthropology in which discursive
practices are seen as constitutive of culture. That is, culture and self are viewed as

411

412 FARNELL / GRAHAM
Discourse-Centered Methods 413

actually constituted by speech and other signifying acts. Discourse-centered work
emphasizes the heterogeneous, multifunctional, and dynamic character of language
use and the central place it occupies in the social construction of reality. A central
proposition of discourse-centered research is that “culture is localized in concrete
publicly accessible signs, the most important of which are actually occurring
instances of discourse” (Urban 1991:1). Edward Sapir (1949) prefigured discourse
centered approaches when he wrote: “The true locus of culture is in the interactions
of specific individuals and, on the subjective side, in the world of meaning which
each one of these individuals may unconsciously abstract for himself from his
participation in these interactions” (p. 515).

According to a discourse-centered framework, culture is an emergent dialogic
process, historically transmitted but continuously produced and revised through
dialogues among its members. It is constantly open to new associations and
interpretive moves. Such perspectives open up new discursive spaces within the
discipline of anthropology that embody Saussure’s original vision of a scientific
study of “the functioning of signs within social life” (Saussure 1916:33; see also
Mertz and Parmentier 1985; Thibault 1997).

New theoretical insights and questions lead to methodological challenges, and in
this chapter we examine methods of research in light of the theoretical developments
that gave rise to them. As Harré (1986c) said, “a theory determines where, in the
multiplicity of natural phenomena, we should seek for its evidence” (p. 83). In our
particular case, increasing theoretical attention to language in social contexts and an
emerging understanding of discursive practices as constitutive of culture, have
required, and are reinforced by, new ways of collecting and analyzing data.

For example, discourse-centered approaches contrast with the formalist logicism
and mentalism that define the notion of grammar characteristic of Chomskian
generative linguistics. In that paradigm, linguistic utterances are analyzed as if they
are generated by an idealized, individual, speaker-hearer independent of sociocul
tural context. Linguistic anthropologists have long been critical of this view of
language (see, for example, Hymes 1964, 1971, and, for a recent cogent critique,
see Ellis 1994). When discourse is considered from the point of view of language
use in social contexts, the assumption that there is a single underlying abstract gram
mar or that languages can be considered as discrete and monolithic codes becomes
questionable. Linguistic anthropologists find multiple linguistic varieties and styles
existing simultaneously within a culture and deem traditional notions of descriptive
grammar and linguistic structure inadequate (see Hymes 1971; Silverstein 1976;
Woodbury 1987; Urciuoli 1996).

Many researchers using discourse-centered approaches consider the rules of
grammar to be imminent in conversational and other expressive practices not located
in “mental structures” (Harré and Gillet 1994). A widely held position in discourse
centered research is succinctly captured in the statement, “People use rules to assess

the correctness of their actions; rules do not use people as the vehicles of their
causal efficacy to generate actions” (paraphrase of Wittgenstein in Mülhäusler and
Harré 1990:7).

The theoretical presuppositions that inform discourse-centered methods are a
shift away from approaches to ethnography that define “language” and “culture” as
transcendent realms or abstract systems that interact with or determine social prac
tices, or as preexistent patterns imperfectly expressed by each individual participant.
Instead, discourse-centered perspectives present a view of language use as social
action and cultural events as the scenes where culture emerges from interaction
among participants (Mannheim and Tedlock 1995:2). For example, the dialectic
between linguistic performances and wider sociocultural and political-economic
contexts is being explored (Hill and Hill 1986; Handler 1988; Irvine 1989; Woolard
1989; Urciuoli 1996). Investigators are also interested in how individuals gain rights
to particular modes of transforming discourse, or are denied such rights, and how
the signifying acts of those with status and power achieve performativity (that is,
are effective) while those of others fail. Investigating how these kinds of processes
operate places language in culture in new ways. Positioning discursive practices as
a central locus of cultural creativity opens up new possibilities for studying how
processes of cultural continuity and cultural change take place. Discussions of
“entextualization, decontextualization, and recontextualization” as processes that
connect discourse with wider sociocultural concerns can be found in Bauman and
Briggs (1990) and Silverstein and Urban (1996).

Discourse-centered approaches don’t focus on language forms alone, but on
forms and processes. “Meaning does not inhere to words or grammar alone, and
language function—the social meaning of what people say—goes well beyond the
semantic-referential function or dictionary sense of meaning” (Urciuoli 1996:5). In
discourse-centered approaches, the referential (word and grammar meaning,
independent of context) is but one aspect of linguistic function, such that attention
to language as “code” (that is, English, Spanish, Kwakiutl, etc.) is not the primary
or sole focus of analysis. The principle objectives are to discover the social mean
ings inhering in language forms and their relationship to social formations, identity,
relations of power, beliefs, and ideologies.

This doesn’t mean that discourse-centered methods eschew formalist concerns.
On the contrary, formal investigation has helped us understand exactly how dis
course structures a socialized world (Hill and Irvine 1993).

Quantitative approaches to linguistic variation in sociolinguistics have also
proved inadequate to handle discursive data. For example, ethnographic research
has shown that aspects of discourse such as accent and code switching play
important roles in the interactive performance of identity and so cannot be
understood adequately if reduced to quantifiable variable phonemes and variable
rules (Urciuoli 1996:5).

414 FARNELL / GRAHAM Discourse-Centered Methods 415

Below, we provide a brief overview of some of the shared features of discourse
centered approaches in linguistic anthropology and point out their utility for
addressing issues that concern anthropologists and social theorists. Later, we discuss
methods and methodological issues related to discourse-centered work.

Why Study “Discourse”?

Discourse-centered approaches in linguistic anthropology seek to dissolve the long
standing dilemma in Western social theory of how to connect “social structure”
and/or “culture” with individual human agency. The problem has been labeled vari
ously as a tension between structure/event (Rosaldo 1980:109), individual action/
collective representations (Sahlins 1985:108), structure/practice (Bourdieu 1977a),
and structure/agency (Giddens 1984). The enduring problem has been how to avoid
the twin errors of individualism and the reification of internal unconscious mental
structures, on the one hand (for example, Freud, Chomsky, Lévi-Strauss, and
Searle), and collectivism and the reification of external social structures, on the
other (for example, Durkheim and variations). These positions have been exposed
as violating the logic of “causal powers” and, as a result, mislocating human agency
(see Varela 1995; Varela and Harré 1996).

When such psychological and sociological determinisms are abandoned and
persons are viewed instead as causally empowered embodied agents with unique
powers and capacities for making meaning, discursive practices emerge as the means
by which social action, cultural knowledge, and social institutions are achieved and
enacted (Williams 1982; Farnell 1994, 1995b; Harré and Gillett 1994; Graham
1995;Urciuoli 1995; Silverstein and Urban 1996; Urban 1996b). Discourse-centered
approaches thus offer social theorists a number of principles that should be central
to their goal of achieving more dynamic views of culture and the relationships
between the individual and society (for example, Sahlins 1976, 1985; Bourdieu
1977a, 1977b, 1991; Giddens 1979, 1984).

Earlier generations of ethnographers necessarily acquired much of their knowl
edge through the discursive practices of their informants. Yet many ethnographers
tended to overlook the fact that the discursive forms in which this knowledge is
packaged are themselves unique cultural forms. In traditional ethnography, Sherzer
and Urban (1986a) note, “Discourse is invisible, a glass through which the eth
nographer comes to perceive the reality of social relations, of ecological practice,
and of belief. Little attention is given to the glass itself. We are rarely informed
about the structure of the discourse through which knowledge is produced,
conceived, transmitted and acquired by members of societies and by researchers”
(pp. 1-2).

Many discourse-centered researchers also take the position that persons constitute
themselves as a “self,” an embodied moral unit in the world through discursive

practices of reflexive talk. Basso (1979) offers a pioneering ethnographic example
of ways in which discourse is central to the cultural construction of self (see also
Rosaldo 1982; White and Kirkpatrick 1985; Harré 1986a; Lutz 1988; Mühlhäusler
and Harré 1990; Abu-Lughod 1993; Harré and Gillett 1994). The works of Brenneis
(1984, 1986), Du Bois (1993 [1987]), Duranti (1986), Keane (1991), and Graham
(1993) show how the “I” of discourse (Urban 1989) can project different selves and
the organization of talk can be linked to notions of the construction of “truth” and
individual accountability.

This reorients theories of person, self, and agency away from an ethnocentric,
individualistic psychologism toward sociocultural dimensions of communication,

! cross-cultural variability (Hill and Irvine 1993), and the enactment of indexical
dynamics (Urciuoli 1996). The assigned locus of “meaning” shifts correspondingly
from internal (private) mental structures (for example, the Freudian “unconscious”
or Searle’s “psychological states”) and the individual speaker/signer, and toward the
dialogic, interactional processes within which meanings are constructed.

Performativity and Indexicality

In addition to resources provided by sociolinguistics, the ethnography of
speaking, and performance approaches to language, a second set of resources used
in discourse-centered approaches draws on the “linguistic turn” in philosophy (see
Taylor 1985; Rorty 1992 [1967]) and the development of Peircian and Jakobsonian
ideas about indexical signs (see Jakobson 1971 [1957]; Silverstein 1976). Benveniste
recognized an important connection between indexicality in personal pronouns
(1971a [1956]) and performativity in certain verbs (1971b [1958]), thereby drawing
attention to the construction of person through discourse.

The linguistic turn contributed Wittgenstein’s post-Cartesian emphasis on the
social construction of mind, language, and emotion to linguistic anthropology (see
Rosaldo 1982; Harré 1986a; Lutz 1988; Lutz and Abu-Lughod 1990). Wittgenstein’s
insights have been complemented by the social constructionism of Vygotsky and
G. H. Mead in a relocation of agency toward a discursive view of mind and person
(see Harré 1984, 1986b, 1987; Harré and Gillett 1994). Austin’s speech act theory
(1962) introduced the notion of “performativity” into the study of language prac
tices. Ethnographically grounded critical discussions of speech act theory stimulated
important challenges to Western philosophical assumptions (see Rosaldo 1982;
Bauman and Briggs 1990). The central insight that language is “performative” (that
is, achieves social action) has remained important, however, and has been developed
further through the Peircian notion of indexicality and Jakobson’s discussion of
“shifters”—words that “shift” their reference depending on the context in which they
are used (for example, demonstratives such as this/that, here/there and pronouns)
(Jakobson 1960, 1971; Silverstein 1977).

416 FARNELL I GRAHAM Discourse-Centered Methods 417

Silverstein argues that the preoccupation with the “semantico-referential”
function of language (that is, where the meaning of words derives from their
naming things, and their relationships to other words, as in dictionary definitions)
provided the basis for a uniquely biased Western linguistic ideology. In this ideol
ogy, other functions of language, especially “nonreferential” indexical functions
(ways in which linguistic utterances also carry social/pragmatic meanings through
spatio-temporal contiguities to the social context) are accorded secondary impor
tance. Prague School linguists and the Russian formalists advanced theories of the
multifunctionality of language. Until Jakobson’s work became influential in Western
linguistic circles, however, these ideas were little known outside of the former
Soviet Union (see Hymes 1975).

The defining properties of meaningful action are precisely those not visible in a
grammatical-semantic model, the units and rules of which are essentially divorced
from social and historical contexts. Urciuoli (1995) synthesizes insights from
Jakobson, Hymes, and Silverstein to describe exactly how indexical language is
constitutive of self and social action:

The property that language shares with all sign systems is its indexical nature: its
maintenance and creation of social connections, anchored in experience and the sense
of the real. Linguistic indexes may be grammaticalized or lexicalized as “shifters”—
devices that locate actions in time and space: personal pronouns, verb tenses, demon
stratives, and time and space adverbs. These are deictic in that they point outward
from the actor’s location. The structure of action fans out from the center, the locus
of I and you, to delineate where and when everything happens relative to the central
actors: he and she versus I and You; there versus here; then versus now; present versus
non-present (past or future)…. The indexes that embody discourse extend beyond
pronouns, adverbs and verbal categories both to the sounds and shapes of speech that
identify the actor with a particular group and to the speech acts marking the actor’s
intent as others recognize it. In short indexes make the social person, (p. 190)

This interest has prompted ethnographic studies of spatial and temporal deixis as
well as the social functions of pronouns. (For ethnographic studies of deixis, see
Hanks 1990; Haviland 1993; Famell 1995a.’) Fundamental connections between
indexicality and performativity are at work because:

Indexicality, a sine qua non of language, may be more or less creative: the more an
index creates the who, what, when and where of the action, the more performative it
is. The more an index maintains the status quo of action, the more presupposed (and
less performative it is). Performativity may be thought of as a process that sometimes
surfaces as an explicit formula (commands, promises etc.) but it is more often implicit.
Any index can be performative, depending on the dynamics of the context. (Urciuoli
1995:190; see also Silverstein 1977)

Formal discourse features frequently cany indexical value. For example, speakers
may use reported speech as a means to distance themselves from the content of their

utterances, or it may be used to add the moral weight of other voices to the
speaker’s own (see, for example, Bakhtin 1986; Hill and Irvine 1993:6-7). Speakers
can use evidential particles to index their orientation to a message (see papers in
Hill and Irvine 1993; also Basso 1995). The analysis of person markers and other
deicticshas been particularly fruitful to the study of issues in the social construction
and representation of identity and personhood. Formal pragmatic devices such as
honorifics frequently convey significant information about social relationships
between individuals (see Brown and Gillman 1960; Irvine 1992; Agah 1996;
Pressman 1997). Ethnographic studies that use these resources have focused atten
tion on such topics as political economies of language (see Irvine 1989), language
ideologies (Silverstein 1976; Woolard 1989,1992; also Kroskrity et al. 1992), con
flict resolution (Murphy 1990; Watson-Gegeoand White 1990), and the discursive
construction of asymétries of power (for example, Hill and Hill 1986; Handler 1988;
Urciuoli 1996; see also Bauman and Briggs [1990] for further discussion).

A third important source of theoretical insights has been the symbolic inter
actionism of G. H. Mead, Goffman’s microsociology (1972 [1967], 1974, 1981),
and the sociological study of conversation (sociolinguistics, ethnomethodology), with
its emphasis on the negotiated and emergent quality of meanings and the inter
actional construction of social structure and institutions. Researchers who specialize
in conversational analysis, a tradition that developed largely within sociology with
the work of Sacks and Schegloff, focus especially on cultural strategies that make
conversations function effectively or not (see, for example, Schegloff 1972;
Schegloff and Sacks 1973; Sacks et al. 1974; see also Fishman 1968, 1970;
Garfinkel 1967, 1972; Goodwin 1981; Gumperz 1982; Tannen 1982, 1984, 1994;
Goodwin and Heritage 1990; Grimshaw 1990, 1994).

These studies note strategies for turn-taking, conversational overlap, utterance
length (holding the floor), patterns of interruption, and conversational “repair,” for
example. This line of inquiry has been fruitful in analyzing how gender roles are
created and enacted through talk (see, for example, M. Goodwin 1990 and papers
in Tannen 1994). Although a great deal of work in conversational analysis focuses
on patterns of communication among English speakers, this approach has also been
fruitful in cross-cultural contexts (for example, Moerman 1988; also Testa 1988).

Reflexive Dimensions

Recognizing that language ideologies are as much at work in our own disciplinary
discourse as in the discursive practices of the peoples we study, many discourse
centered researchers explore reflexive dimensions of their research and seek to
deconstruct and challenge dominant Western conceptions of language and social life.
(For discussion of language ideologies, see Lucy [1990] and papers in Kroskrity et
al. [1992].) Several ethnographic studies challenge Western “personalist”

418 FARNELL / GRAHAM Discourse-Centered Methods 419

(individualist) notions of discourse (Holquist 1983), in which the individual is
understood to be solely responsible for utterance production (for example, Searle
1969). These studies reveal language ideologies in which discourse is thought to be
the product of multiple speakers (Rosaldo 1982; Brenneisl984,1986; Duranti 1986,
1993; Keane 1991; Chafe 1993; Du Bois 1993 [1987]; Graham 1993; Hill and
Zepada 1993; Irvine 1996). Reflexive studies thus seek to expand traditional defini
tions of what counts as discourse, discourse production, and interpretive processes
in ways that allow a better understanding of indigenous classifications.

Studies that include consideration of complementary vocal acts such as wailing,
chanting, laughter, the use of sound symbolism, the blurring of boundaries between
music and speech, and the coproduction of utterances expand the traditional
boundary within the modality of socially produced and meaningful sound (List
1963; Brenneis 1986; Duranti 1986; Urban 1988; Feld 1990 [1982]; Nuckolls 1992;
Graham 1993; Feld and Fox 1994). Ethnographic studies of sign languages and the
coproduction of speech and gesture seek to make problematic the reductionism
inherent in the conflation of discourse with speech. Including the modality of bodily
action into what counts as discourse challenges the traditional Western division of
communicative practices into “verbal” and “nonverbal” (Farnell 1995a) and recon
ceives discourse as the simultaneous production of vocal signs and action signs
(Famell 1995a; see also Heath 1986; Kendon 1989; McNeill 1985, 1992; and the
section Embodiment of Discourse).

In this critically reflexive arena, meta-level discourses on language and social life
provided by native speakers are no longer simply sources of data for the presenta
tion of a problem-free “indigenous perspective.” Speakers become intellectual
collaborators who can make substantial theoretical contributions. Greater reflexive
awareness of the very processes of recording and analysis have been called for and
demonstrated (Briggs 1986; Ardener 1989 [1978]; Bauman and Briggs 1990; Famell
1994; Haviland 1996; Urban 1996a).

Context

Expanding what counts as discourse beyond “naturally occurring speech” to
“naturally occurring speech and bodily action” prompts reflection about what an
adequate theory of context might look like. Clearly, a theory of context must take
into account the copresenceof other semiotic practices. Conceptions of context must
also be elaborated to account for events taking place elsewhere in space or time that
give meaning to novel events and actions. Emergent performances are necessarily
situated in cultural histories and long-term patterns of action. Performances are
situated within the context of histories of performance and cultural knowledge. (For
further discussion of context in discourse-focused work, see Introduction and papers
in Goodwin and Duranti [1992].)

Actors also bring individual perspectives to the situations in which they operate,
and a context analysis must account for the strategies actors use “to constitute the
culturally and historically organized social worlds that they inhabit” (Duranti and
Goodwin 1992:5). This means defining context “from the inside” rather than, as in
traditional approaches to folklore, considering context “from the outside in”; context
is not some kind of “surround” that exerts influence on expressive forms and texts
(Bauman 1992:142). It is agent-centered,emergent, and may vary across participants
in a given event. Social actors position themselves in conversations according to
their personal histories and goals and are simultaneously positioned by others
according to factors such as class, ethnicity, and gender (Davies and Harré 1990).

Previous discourse is a further dimension of context because discourse itself is
the source of much presupposed knowledge. Discourse is highly intertextual: Any
present utterance, as Bahktin observed, is “replete with echoes, allusions, para
phrases, and outright quotations of prior discourse” (Mannheim and Tedlock
1995:7). Utterances and narratives typically reference other texts, and pieces of pre
vious discourse are frequently inserted into newly created texts (see Bauman 1992).
Often references to prior discourse take the form of direct or indirect quotations
(see, for example, Bakhtin 1981; Briggs 1992; Lucy 1990; Silverstein 1985; Tannen

I 1995; Urban 1984). These “entextualized”piecesof discourse import meanings from
other situations and contribute to the creation of emergent meanings in novel
contexts (see Bauman and Briggs 1990; also Kuipers 1990; Urban 1996a; Silverstein

1 and Urban 1996; Graham In press).

Methods: Text Collection and

Elicitation

The ethnographic focus on socially situated discourse contrasts with earlier methods
for language study in American anthropology, as well as those adhered to by con
temporary descriptive linguists. Two traditional methods for studying language that
became central to the Boasian tradition of the study of North American Indian lang
uages are text collection and elicitation (see, for example, Boas 1940; Radin 1949).

Texts

In this tradition, linguists take written texts as the starting point for linguistic
analyses (see Bernard and Ryan, this volume). These texts are typically elicited from
native speakers by a freíd worker who analyzes grammatical structure based on the
patterns that emerge in the texts. Before the use of tape recorders, this process
necessarily involved a stop-start process of dictation, which destroyed the natural

, flow of a storytelling performance or other speech event, and texts were usually
• collected outside of normal cultural contexts of narration.4

Discourse-Centered Methods 421420 FARNELL I GRAHAM

Such limited technological means precluded any detailed attention to the role of
naturally occurring discourse within social life. Analytical goals centered on gram*
matical structure, the mythological, and other ethnographic content of narratives. As
Tedlock (1993:38-39) notes, only after new recording technologies emerged could
linguistic anthropologists and folklorists turn their theoretical attention to features
of oral performance that could not be captured through dictation or elicitation.
Throughout the late 1970s and early ’80s, the emphasis on narrative performance
of verbal art directed attention away from the study of formal patterning and
symbolic content of texts typical of the Boasian tradition and an earlier poetics.
The focus instead was on the emergence of verbal art in social interaction and
fine-grained analyses of speech events. This interest connects with a long tradition
of thinking about language and society that emerges clearly in the writings of
Vico, Herder, and von Humboldt, who argue that verbal art is a central dynamic
force in shaping linguistic structure and study. Attention to poetics from Sapir, the
Russian formalists, and members of the Prague school contributed to the
development of performance and poetics in the 1970s and 1980s (Bauman and
Briggs 1990:59).

Elicitation

In this second method, a linguist elicits words and sentences from a native
speaker and transcribes the sounds into a written form. Usually, the linguist works
with one or a small number of informants. Each utterance produced by the native
speaker is seen as an instantiation, a token, of the abstract code or language.
Through elicitation of words and sentences, the linguist tries to determine the struc-
ture of the grammar underlying the spoken utterances. In this way, it’s possible to
determine and describe such things as the sounds of a language (its phonetic inven
tory), the way these sounds combine (the phonology), and syntactical patterns (word
order, grammatical particles). Descriptive linguists, who aim to describe language
structure, use this methodology.

Although discourse-centered research incorporates both elicitation and text-
focused methods of analysis, it differs from earlier approaches by taking
contextually situated discourse as the starting point for analysis. Elicitation and text-
focused analysis are often incorporated at later stages, as specific problems emerge.

Identifying, Collecting, and Recording Situated Discourse

There is no single best method of collecting information on language use within a
community. The starting point, though, is to seek, or discover, through participant
observation, those forms of discourse that will form the data for analysis. (For an

excellent discussion of ethnographic methods and transcription see Duranti 1997:
84-161). Appropriate procedures will depend on the theoretical focus of the
research, the relationship of the ethnographer to the community, and the specific
situation.

One method advocated as a way to make “ethnographies of communication” (for
example, Gumperz and Hymes 1964; Hymes 1967,1974; Sherzerand Darnell 1972;
Bauman and Sherzer 1989; Saville-Troike 1989) involves making an inventory of
distinct spoken forms, both marked (not typical) and unmarked (typical), and the
contexts in which they occur. Recording a broad range of spoken discourse provides
data from which to compare and contrast distinct expressive forms and styles along
formal and functional lines. A researcher may want to learn if speakers change how
they talk when addressing certain individuals (such as elders, relatives, or those of
certain social roles or genders) or in distinct social settings (for example,
ceremonial contexts), or if there are special oratorical styles or styles of interacting
when telling stories or myths. If special forms or styles of speech do exist, then
the researcher would attempt to map their social distribution (that is, find out who
has the capacity and authority to use these forms) and try to discover how the
forms are learned and in what situations they can occur.

Depending on the researcher’s theoretical focus, it might also be important to
note whether some expressive modes of speaking stimulate other forms of expres
sion (see Feld 1990 [1982]) and to determine the relationships between speech,
gesture, facial expression, body position, and spatial orientation (for example,
Goodwin 1984; Wiget 1987; Kendon 1989; Hanks 1990; Duranti 1992; Streeck
1993; Farnell 1995a, 1995c; Kaeppler 1995) and other sensory modalities.

The notion of a “soundscape” (Schafer 1997) has directed researchers’ attempts
to record and describe the full range of socially meaningful sounds used within a
community (see Feld 1990 [1982]; Graham 1995). A soundscape describes the
sounds that people make and use over a 24-hour period, together with explanations
of their meanings. It includes talk, musical forms, silences (Basso 1972), and

‘ attention to groups that are silenced (Lederman 1984; Gal 1991). Description of a
soundscape might include variations in the social use of sound that occur due to
seasonal changes, in response to (or to provoke) natural events, and in conjunction
with ceremonial cycles. The concept of soundscape draws attention to the fact that
speech is just one modality within an entire arena of socially constructed and
meaningful sound production. Distinct genres, such as lament, song, and oratory,
may have important interconnections yet contrast in significant ways (see Graham
1984,1986). Moreover, communities vary in the way that they divide the continuum
between speech and song (see List 1963). In a soundscape, expressive forms that
sound like speech (or song) to Western ears may be classified in ways that challenge
ethnographers’ assumptions (see Seeger 1986). The notion of soundscape has proven

: useful in calling researchers’ attention to unmarked forms of speech, which tended
to be overlooked in earlier ethnographies of communication/

422 FARNELL ! GRAHAM Discourse-Centered Methods 423

Everyday or unmarked forms of discourse constitute important data for analysis
because they contain information about language ideology, socialization, and ways
in which gender, age, ethnic identity, and power relationships are linguistically
constructed and conceived. These unmarked forms provide a basis against which
more salient forms (oratory, storytelling, poetry, etc.) can be compared and con
trasted. Moreover, marked forms derive some of their meaning through their
contrasts with everyday forms of expression (see, for example, Urban 1985).

If the research focus is more specifically defined at the outset, as in studies of
language and gender (see papers in Philips et al. 1987; Hall and Bucholtz 1995) or
language socialization, (for example, Schieffelin and Ochs 1986a 1986b; Ochs 1988;
Schieffelin 1990), then the researcher would select those social contexts that facil
itate observation of the linguistic features targeted for investigation. For example,
in her study of language socialization among Kaluli children, Schieffelin (1990)
used ethnographic observation to determine the activities and times of day when
caregivers and children interact most intensively. She found that opportunities for
lively interaction occurred in the early morning when families awaken and begin to
prepare the first meal. Interactions during the late afternoon when mothers return
from the garden and begin preparations for the evening meal and during trips to the
stream for bathing were other opportunities for lively interaction and were par
ticularly fruitful occasions for studying socialization practices (p. 25). Like time,
space may be another dimension that permits access to distinct types of social inter
action. Thus, once Ochs (1988) broke through the “honored guest” status in the
Samoan household, she was able to enter the domestic space at the back of the
house where previously inaccessble forms of interaction between caregivers and
children take place. Garvey (1984) and Ochs (1979b) show how the study of
children’s talk is an area with its own methodological problems.

A crucial element of the research process is to make recordings of discourse as
it unfolds in social settings. Typically, this involves using a high-quality portable
audiotape recorder and microphones. (Internal microphones are generally of poor
quality. For practical advice regarding recording techniques in natural settings, see
Goodwin 1993; Ives 1995). In recent years, as video-recording equipment has
become more affordable and portable, researchershave incorporated this technology
into their fieldwork so that visual-kinesthetic aspects of discursive practices can be
incorporated into analyses (see Kendon 1989; Streeck 1993; Farnell 1995a).

Carrying a recording device in the field as much as possible can provide a
researcher with many samples and a variety of types of naturally occurring dis
course for later analysis. Recording devices may present problems, however, and a
researcher must take into account ethical concerns about privacy, access to restricted
information, and rights to ownership, as well as the ways in which, like the presence
of the ethnographer, the presence of recording devices may significantly alter the
events themselves. In many communities, expressive forms such as songs are con
sidered intellectual property (Seeger 1991, 1997), and intellectual property issues

associated with field recordings are increasingly likely to confront researchers. (See
Shuy [1986] and Murray and Ross Murray [1992] for discussion of the legality of
surreptitious recording. For discussion of ethical dimensions of fieldwork, see
Fluehr-Lobban, this volume.)

The advent of affordable recording technologies has in some cases enabled our
research subjects to own and operate their own recording devices. Situations where
outside researchers no longer have a monopoly over recording technologies
challenge us to think of the power vested in our control of technologies. In some
cases, new technologies have stimulated changes in the form or use of expressive
genres (see Abu-Lughod 1989; Caton 1990; Manuel 1993; also Turner 1991.
Spitulnik [1993] provides an excellent overview of anthropological studies of mass
media). Local control of recording technologies may also open up new possibilities
for collaboration (see Michaels 1994). Ways in which such ownership affects
discourse-focused research is an area for further ethnographic investigation.

Participants may be self-conscious or conscious of addressing audiences beyond
those immediately present when they know they are being recorded, and their
language use can reflect this. Alternatively, people may forget that they are being
recorded and after several minutes of self-conscious interaction, they may settle
into familiar communicative patterns and activities will proceed as usual (Hymes
1981; Tedlock 1983:18, 302-311; Briggs 1986; Darnell 1989 [1974]; Graham
1995; Haviland 1996; Urban 1996a). Members of a community can become
accustomed to the constant presence of electronic technology in much the same
way that people often become accustomed to seeing ethnographers writing things
in notebooks. Individuals have also been known to comment on the tape recorder
or camera’s absence when it’s not present and wonder why an event is not being
recorded.

In situations where participants are sensitive to intrusive technologies, a small
tape recorder, which is less obtrusive than a camcorder, may be allowed. In some
cases, however, participants may refuse to speak or perform when a recording

; device is present, or the ethnographer may be excluded from an event if carrying
} such devices. One option in such cases would be to explore the possibility of
■ recording an elicited or “staged” performance at a later time. This will, of course,
. be different in many ways, but is not thereby rendered inauthentic nor is it without
. considerable analytic value. The presence of the ethnographer and recording devices
■ will almost certainly affect people and their interactions, but reflexive consideration
: of these factors, rather than being denied, can be incorporated into analyses with
: interesting results. Labov (1972) recognized the problem of how to observe people’s
: interaction without their being self-conscious as an “observer’s paradox” in

sociolinguistic research and developed various fieldwork techniques to attempt to
transcend this dilemma.

Certainly, good field recordings of contextually situated interaction either audio
or audiovisual, are essential to discourse-focused work. This doesn’t negate the

424 FARNELL / GRAHAM Discourse-Centered Methods 425

usefulness of field notes, however, which are often essential companions to recorded,
data. For example, field notes can be used to document important contextual:
observations that can later be incorporated into transcriptions. Alternately, comments
can be spoken into the tape recorder/video camera itself. A useful technique is to i
correlate numbers on a tape/video recorder counter with notes written in a notebook
(for example, Schieffelin 1990:30).

In and of themselves, however, field notes are insufficient as records because
they are only fragments of discourse or descriptions of it at best; they don’t provide
records adequate to analyze the richness and nuances of discourse as performed.
Tedlock (1983:38-39), for example, observed that the Zuni texts he recorded using
a tape recorder, were nearly twice as long as those recorded by Benedict. Among ‘
the reasons for this is that narratives given for dictation tended to be condensations
of what a performer would tell in a normal, spontaneous situation and because a
responsive audience was lacking. At the time field recordings are made, a researcher
may be unaware of components that subsequent analysis of audio/videotape reveal
as significant carriers of social meaning (see discussion of the two tape recorder/two
video method below).

Processes of Translation and Transcription

The extensive use of tape recorders and, more recently, video cameras, has neither
replaced nor reduced the need for adequate methods of written transcription. On the
contrary, innovations in electronic recording technologies have facilitated the devel
opment of written transcription systems more adequate to the analytic tasks at hand.
The processes of transcription and translation are themselves interpretive analytic
tasks; they are re-presentations of performance that are necessarily informed by
explicit and implicit theoretical assumptions (see Ochs 1979a; Hymes 1981; Tedlock
1983; Woodbury 1985; Sherzer and Urban 1986a: 11-12; Sherzer 1992; Swann
1992). As Ardener (1989) reminds us, particular events that are registered depend
on our modes of registration and specification—that is, the means by which they are
apperceived. He advises us to know as much as possible about these modes because
“our definition of… the events depends upon the modes of registration available
to us” (p. 87). The techniques we use to complete the tasks of transcription and
translation influence the interpretation of texts and the ways in which meanings are
understood to emerge from socially situated discourse. Similarly, the phonological
patterns or grammatical categories of the translator’s language may affect the
transcription and translation of texts (see Whorf 1956; also Sherzer 1987).

MacMahon (1996:831-837) provides an excellent history of the interest in and
development of phonetic transcription systems that can be traced back, in England
and North America, to the late sixteenth century. Interest in Asian languages and
cultures at the end of the eighteenth century first led scholars to question the

. adequacy of characters from the Roman alphabet to represent spoken pronunciation
in foreign tongues. In other contexts, the need to incorporate information about

: pronunciation into dictionaries, interest in transcribing local dialects of European
languages, and intensive Christian missionary activity in Africa and elsewhere also
led to specific methods of transcribing speech (MacMahon 1996).

A group of modem language teachers in the International Phonetic Asso
ciation incorporated many of these efforts into the International Phonetic
Alphabet (IPA) in 1888. Although it has undergone subsequent changes and
additions, the IPA remains the major phonetic alphabet in use today. It com
prises an extensive set of symbols and diacritics derived from the Roman alpha
bet and has symbols for consonants, vowels, and some prosodic features such
as stress and intonation.6

It is worth remembering that in the era of dictation (an era stretching back to the
making of Homeric texts and lasting until well into the Boasian era) the only
language-recording technology was writing. Before the invention of the tape record
er, important collections of spoken language and other vocal expressive forms were
recorded on wax cylinders, glass acetate, and aluminum phonograph discs.7 Materi
als recorded with these technologies are deposited in the Archives of Traditional
Music at Indiana University and in the Smithsonian Institution. In addition to
collections made by Boas and his students such as Radin and Fletcher, Haugen’s
study of Norwegian language change in the Midwest (1969) and Turner’s collection
of Sea Island Gullah (1969) included samples of situated discourse.*

As mentioned above, the advent of the tape recorder transformed the field
worker’s ability to record oral discourse (Tedlock 1983:38-39), and only sub
sequently did the examination of native poetic structure (ethnopoetics) and the
adequate translation of verbal art emerge as central concerns.

At this point, analytical limitations of conventional transcriptions began to
emerge. Ironically perhaps, the very success of phonetic notation systems had led
investigators to limit their definitions of “language” to those segmentable com
ponents that could be represented with the Roman alphabet and its extensions. New
technologiesand innovative forms of transcription have exploded limiting theoretical
assumptions about language and enabled anthropologically oriented linguists to
expand notions of linguistic functionality.

With the explicit goal of exploring native poetic structures, Hymes (1981),
Tedlock (1983), and others began to expand the notational repertoire to include
conventional poetic and dramatic devices. Figure 1 illustrates how Tedlock
(1983:84) used line breaks to indicate short pauses and dots to indicate longer
pauses, thereby inviting the reader to listen more closely to the rhythms of speech
and silence in performance. Upper-case letters mark louder speech and marginal
parentheses guide the reader’s attention to voice quality and volume. A vowel
followed by a long dash indicates vowel lengthening; repeated letters indicate

426 FARNELL / GRAHAM Discourse-Centered Methods 427

drawn-out sounds; letters spilling downward on the page indicate descending 2
glissando; and split lines indicate tonal changes in chants. 1

Now there’s a voice 2
and she hears it. 9
“Perhaps 1
you are $
someone from the village who’s lost the way 4
and you’re calling out 5
Well, 1*11 answer you, and fl
perhaps 1 could depend on you for the night.” fl
That’s how the poor girl felt. I
She quickly put on her snow boots, then fl
she went out in the clearing to the edge of the firelight. fl
(as if from å (Ostanet, and very hi^h) fl
hooooooooooohaaaaaaaaaaaaaaaaaaaaaaa 2• fl* 1a 4a fl

“y it said. 4
(ailing out) “Come over here, I’m spending the night here,” the girl 1

said. fl

And then fl
there was another call then: 1
hooooooooooooooooooohaaaaaaaaaaaaaaaaaaaaaaaaaaa 3

a Ia fl
“y it said. 1

(alting eut) “Come over here. I’m spending the night here,” the girl 1
said again. 5

The girl went inside and PUT MORE WOOD ON, the fire was really I
blazing, then it came CLOSER. :

It came doser 1
calling 3
hooooooooooooooooooooooooohaaaaaaaaaaaaaaaaaaa 3

a a
“y it said. 1

The girl heard it vety clearly now. J
(imping) “Eeee! this is why you 1
warned me. 4
Why did I answer? 1
So this is why you told me i
that I shouldn’t have a fire late at night. |
Well, I asked for it,” she said, and the girl |

Figure I. Transcription designed to convey features of oral performance such as volume, |
pause structure, utterance length, and pitch (from The Spoken Word and the Work of Inter- 3
pre talion by Dennis Tedlock. Copyright © 1983 University of Pennsly vania Press. Reprinted 1
by permission of the publisher.). ì

Tedlock called for punctuating according to rising and falling contours of oratory, |
and shaped lines and stanzas according to the stops and starts of dramatic timing. 1
His explicit aim was to create “open or performable texts” that “open the ear” to the I
sounds of vocal performance (1983:62). Students of “the ethnography of speaking” 1
contributed to these developments, choosing from these transcription options accord- 1
ing to their analytic purposes. Papers in Edwards and Lampert (1993) discuss a ‘

number of transcription and coding systems for discourse analysis from contrasting
theoretical perspectives, highlighting the point that a transcription format should be
related to the goals of analysis.

Of course, no transcription can ever be so detailed and precise as to provide for
the full re-creation of the sound (or vision) of a tape: The hypothetical goal of total
notation is a positivist’s dream. In light of this, some researchers supplement their
written publications with audiotapes (Bowen 1991a,1991b; Feld 1990 [1982];
Sherzer and Urban 1986b; Seeger 1987) and CD-ROMs (Famell 1995c). Individuals
and communities can also exert influence over transcription conventions. For
example, Becker(1983) describeshow his Burmese consultant objected to phonetic
transcription and insisted on using the Burmese writing system because it related to
Buddhist understanding of the language. Elizabeth Brandt (1981) describes an
extreme case where members of a speech community entirely rejected the writing
of their languages.

Computer-Assisted Analyses

Recently, scholars have experimented with computer programs, such as
SoundEdit and Signalyze, to digitize sound for visual display.9 Such programs
provide graphic representations of wave forms that can be useful for analysis and
representation of patterning in amplitude, pause structure, and pitch (using spectro
graphs). Digitized images can be especially productive in analysis because they
enable researchers to capture microfeatures of performance and to perform analytic
tasks involving looping, splicing, and juxtaposition of instances of speech or other
sounds. For example, it is possible to create a “tape loop” of a recorded segment
and replay it repeatedly in order to listen—and see—aspects of sound such as
utterance or pause length and amplitude. Sound segments can also be juxtaposed to
discern the existence of regularities or microvariations across instances.

Digitized sound samples enabled Roseman (1991:171) to recognize an acoustic
iconicity between cicada buzzing, certain bird calls, and Temiar bamboo tube
drumming, which were essential to efficacious performance. The digitized rep
resentations shown in Figure 2 illustrate the striking resemblance between the
wave forms of sounds Temiar link with performance and associate with evocative
power. Urban uses digitized samples to represent the regularity of utterance
length and amplitude in an excerpt from the Shokleng origin myth to illustrate
the syllable-by-syllable dialogic nature of the performance (1991:109; see also
1996b: 201). Digitized figures also illustrate the regularities of Warão lament
performance in Briggs’s .analysis (1993). Sherzer uses digitized samples to display
visually the parallelism of amplitude (from high to low) that characterizes certain
genres of Kuna oral performance, as well as pitch and volume patterns (1992:
430, 435-438).

428 FARNELL ! GRAHAM Discourse-Centered Methods 429

Figure 2. Digitized samples of (1) cicadas, (2) the golden-throated barbet (Megalaima
franklini), and (3) the beat of bamboo tubes illustrate an ¡conicity that is essential to
efficacious Temiar performance (from Roseman 1991:170. Reproduced by permission of the
University of California Press from Healing Sounds from the Malaysian Rainforest: Temiar
Music and Medicine. Copyright © 1991.).

Researchers also experiment with transcriptions that depict pitch variations and
other features of speech, plus other musical or music-like vocal genres. Specto-
graphs may be used or graphs may best represent the salient features that a scholar
seeks to illustrate. Modified Western musical staffs can also be suitable for depicting
features of vocal performance. Linguistic anthropologists, often working in collab
oration with ethnomusicologists (for example, Sherzer and Wicks 1982; Graham
1984, 1994, 1995), or those with ethnomusicological training (for example, Seeger
1987; Feld 1990 [1982]), have developed innovative ways to represent melodic gen
res. Figure 3 (in two parts) shows a transcription of a Xavante musical performance
in which the musical staff has been reduced from five to three lines to highlight the
reduced pitch inventory of this musical genre. Notes with indistinct pitch appear as
“x,” and slanting lines before notes indicate a slide up or down to the indicated
pitch (transcription by Scruggs in Graham 1994:734-735, 1995:120-121).

By contrasting four representational modes of the same stretch of Kuna chanting,
Sherzer (1992:434-435) shows how a variety of transcription modes are suitable for

= tes

ju hi 2 2 2 7 22ju hi 22 72 2

Part 1

… e -T_ »

i 1

I J
hi fi fi* fi fi fi fi

n n n~j—y—i

A
E=

hi fi Ti z z zi zi

n n .H-j——
! .

•

ju hi fi fi fi fi fi fi ju hi fi fi fi fi fi fi

B_
——–———– J—-+ — n n n J { M>

ju hi ha ju hi hi ju hi fi fi fi fi fi fi

‘a. ————J—+ —+ n n n j .!1 ….. »
ju

1
hi ha ju hi ha ju hi fi fi fi fi fi fi

a. —L n n n ii . -i
|ke”1 – >

ju hi fi fi fi fi fi fi ju hi fi fi fi fi fi fi

Repeat three times
A
1 ‘ i- n , ——— FJ rj n-j——’«J —s

ju hi fi fi fi fi fi fi ju hi fi fi fi fi fi fi

430 FARNELL / GRAHAM
Discourse-Centered Methods 431

Transcription Key:

The reduced number of lines in the musical system
highlights the limited melodic range of the song.
The four-note pitch set shown on the left correspon

The pitch of the note GW is lowered slightly and
the pitch of the note A is lowered a quarter tone.

strong accent staccato

sustained slightly;
weak accent

slide up to pitch

9 less definite pitch slide down to pitch

Movement: The dance moves are in a consistent, regularized pattern that establishes the
basic meter. In this dance, a da-hipapi, performers remain in the same place,
bend slightly at the knees and then return to a more erect position. The
completion of these moves represents a basic beat and is represented in the
transcription as a quarter note value. Therefore, as shown in the transcription,
there are six beats per phase throughout the composition.

Duration: The duration of this song is one minute and six seconds.

kurkin ipekantinaye.

Owners of kurkin.

olopillise pupawalakan akkuekwiciye.
Your roots reach to the level of gold.

kurkin ipekantinaye.
Owners of kurkin.
olopillise pe maliwaskakan upoekwiciye.
Your small roots are placed into the level of gold.

kurkin ipekantinaye.
Owners of kurkin.
olopillise pe tnaliwaskakana pioklekekwiciye.
Your small roots are nailed into the level of gold.

kurkin ipekantinaye.
Owners of kurkin.
olopillipiiye apikaekwiciye kurkin ipekantinaye.
You are resisting within the very level of gold owners of kurkin.

olopilli aytikkimakkekwici kurkin ipekantinaye.
You weigh a great deal in the level of gold owners of kurkin.

olopilli kwamakkekwici kurkin ipekantinaye.
You are firmly placed in the level of gold owners of kurkin.

olopilli aytitimakkekwakwiciye kurkin ipekantinaye.
You are moving in the level of gold owners of kurkin.

Part 1

Part 2

Figure 3. This musical transcription of Xavante da-ño ‘re illustrates how modifications of the
conventional Western musical notation system may highlight salient features of non-Westem
musics. Here, the Western musical staff is reduced to call attention to the pitch inventory and
notation is designed to indicate sliding pitch and less definite pitch (transcription by T. M.
Scruggs in Graham 1994:734-735,1995:120-121. Reproducedby permission of the Ameri
can Anthropological Association from American Ethnologist, 1994. Not for sale or further
reproduction.).

distinct analytic ends (see Figure 4, in four parts). These alternative modes of
representation and transcription can be used to highlight or downplay the importance
of varied linguistic and stylistic features that serve various expressive and aesthetic
ends (see also Hymes 1981; McClendon 1981; Tedlock 1983,1987; Silverstein
1984; Woodbury 1985, 1987).

•2 • »lifhi upward glide in pitch
m ■ guttural quality

■ ■ breath intake

Part 2

432 FARNELL / GRAHAM
Discourse-Centered Methods 433

I”TT™T”™mI™”
lu MawaMluuMukuam

<2» <

la kurkin ipekantinaye.

Owners of kurkin.

lb olopillise pupawalakan akkuekwiciye.

Your roots reach t the level of gold.

2a kurkin ipekantinaye.

Owners of kurkin.

2b olopillise pe maliwaskakan upoekwiciye.

Your small roots art placed into the level of gold.

Part 3

Figure 4. Four representational modes of the same stretch of Kuna chanting illustrate a
variety of transcription modes uniquely suitable for distinct analytic ends. Part 1 provides a
segmental transcription of sounds, morphemes and words in which line and verse structure
is based on musical and pause patterns; Part 2 is a graph that captures pitch and tempo; Part
3 is a SoundEdit display that illustrates amplitude and pause patterns; and Part 4 is pitch
tracking (of formant structures) that shows the falling pitch and volume that is characteristic
of this genre, as well as long pauses between lines. Parts 1, 3, and 4 from Sherzer
1992:434-436 reproduced by permission of the Smithsonian Institution Press; Part 2
reproduced by permission of Joel Sherzer and Sammie Wicks and Latin American Music
Review, of the University of Texas Press. Copyright C 1982.).

The Embodiment of Discourse

The shift to performance and the inclusion of context in linguistic analyses also
opened up theoretical space within which to add the embodiment of social actors to
the notion of language-in-use (Farnell 1995a:9). Several investigators have recog
nized that the visual-kinesthetic components of discourse, such as gestures, postures,
gazes, facial expressions, and spatial orientation, are meaningful components of
linguistic utterances in social contexts (for example, Goffman 1961, 1974; Sherzer
1973; Kendon 1980,1983; Goodwin 1981; Tedlock 1983; Wiget 1987; Hanks 1990;
Duranti 1992; Haviland 1993; Keating 1998, In press). Until recently, however, a
way of making the analysis of body movements in spatial contexts part of the
normal ethnographer’stool kit had not been developed. In the 1970s, Birdwhistell’s
“kinesics” promised a solution, but failed because bodily movements were still seen
as “behaviors” rather than signifying acts—that is, as agentic, semiotic practices that
are shared expressive resources which require translation from one culture to
another. Kinesic analyses and attempts to create an adequate transcription system
dissolved under the weight of ever-increasing microscopic behavioral detail. Also
absent from earlier conceptions was an adequate analytic framework for dealing
with the multidimensional spaces within which humans move; spaces that are
simultaneously physical, conceptual, moral, and ethical (Williams 1995:52).

Williams’s “semasiological theory” of human action with its central concept of
the “action sign,” has provided investigators with new theoretical resources more
adequate to this task (see Williams 1975,1982,1991,1995,1996). From a semasio
logical perspective, discourse is necessarily embodied because persons are conceived
as subjects empowered to perform signifying acts with both vocal signs (speech) and
action signs (bodily movements). In addition, a new emphasis on acquiring a lit
eracy specific to the medium (that is, the ability to read and write body movement)
has been achieved through a movement transcription system called “Labanotation”
(Page 1990,1996; Williams and Farnell 1990; Farnell 1994). The introductory essay
in Farnell (1995b) offers an overview of recent theoretical developments in the
anthropology of human movement systems. Ethnographic studies can be found in
Farnell (1995b), Williams (1996) and in the Journal for the Anthropological Study
of Human Movement.

An alphabetic script for writing body movement must solve the problem of
how to represent all the parts and surfaces of the body with two-dimensional
graphic signs (see Figure 5). It must also make finite in some way the space(s)
in which the body moves. Figure 6 illustrates how the Laban script uses simple
set theory to represent spatial directions viewed from the actor’s perspective.
Locating this imaginary three-dimensional “cross of axes” at each joint of the
body achieves a remarkable economy of graphic signs: The same scheme
specifies the direction of individual limbs and smaller body parts and provides
a framework for indicating the direction of pathways for the whole body (as

Discourse-Centered Methods 435
434 FARNELL I GRAHAM

when a person moves from one location to another). These graphic signs for
body parts and spatial directions are arbitrary but ¡conically motivated to assist
reading fluency.

LIMBS

JOINT

4 • shoulder

7 8 elbow

3 s wrist

3 I hand

3 I fingers

left

I thumb I kra, I

SIGNS

1 r hip

* • knee

3 e ankle

1 c foot

I I toe»

left

knuckle | 2nd etc.
8 yount

22,2_2
a a

dla.

A double line 9 on the side of a joint sign indicates
the limb above that joint e.g., I upper arm. It lower
arm etc. The surfaces of limbs can be specified when
necessary as shown below:

1 a limb
c
u neck

Î both arm»

1 T whole arm

ï both legs

i r whole leg

H
SURFACES OF LIMBS

H i I-
outer or under or thumb little finger
top back or big or little toe

toe side side

The three-dimensional cross of
axes that organizes spatial
direction: the body is in the
center of this kinesphere.

middle down or low

• forward 0
P back Q

to the right p

< to the left d
0 up

I down

forward right diagonal

back right diagonal

back left diagonal

forward left diagonal

AREA SIGNS
□ basic area sign

BB shoulder area

E) chest

• pelvis

2 whole pelvis

E unit from knee to head, etc.

Cl area of hand or foot

fl back of hand or top of foot

fl palm of hand or sole oí foot

A fingertips or tips oí toes

U heel of hand or foot

f thumb or big toe side

n- little finger or little toe
side

stepping forward

) stepping to
\ the right

‘ stepping back
N right diagonal

Sides of an area can be specified using a set of minor directional pins I low,
1 middle, I high—e.g, • upper front side of chest,lower left back diagonal
side of pelvis, è front middle area of head, i.e., face. Signs for parts of the face
are also built out of these units, e.g., 8 eyes, E right ear, 8 mouth.

Figure 5. Labanotation: graphic signs for body parts and surfaces (from Farnell 1995a.
Reproduced by permission of the University of Texas Press.).

Gestural data: a smaller cross of

axes is imagined at the center of
each joint so that direction for each
part of a limb can be specified.

Track data: direction symbols for moving the

whole body from one place to another. These

would be placed in the central support column
on the staf

Figure 6. Labanotation: graphic signs for specifying spatial direction (from Farnell 1995a.
Reproduced by permission of the University of Texas Press.).

436 FARNELL I GRAHAM Discourse-Centered Methods 437

Movement necessarily takes place through time and the flow of action and
direction of reading in texts written with the Laban script is upward. Placing
symbols in different columns on a vertical staff organizes the flow of action; a
central dividing line separates left and right sides of the body. The Laban script can
be divided into bars like a musical score if metrical timing is required, or it can be
used without if not—the length of spatial symbols indicating relative timing instead.
Farnell’s ethnography of the simultaneous production of spoken Nakota and Plains
Sign Language in the storytelling practices of Assiniboine elders illustrates this
approach (1995a, 1995c). Figure 7 is an example of a transcription of speech and
gestural signs from a storytelling performance that was first videotaped and then
transcribed and translated in collaboration with the storyteller.

Concepts of the body, systems of spatial orientation, and concepts of time differ
considerably across languages and cultures, and the Laban script has proven flexible
enough to take these into account (see Williams and Farnell 1990; Farnell 1994,
1996; Williams 1995; Page 1996). For example, Farnell (1995a:141-154) found that
understanding Assiniboine conceptions of the cardinal directions was, among other
things, central to understanding deictic utterances in speech and gesture. (This
applied to unmarked forms of everyday discourse as well as storytelling per
formances.) Such a conception was incorporated into the transcription by placing a
“spatial orientation key” at the start of the movement score; this tells the reader that
this particular conception is in operation throughout, much like the key of C# minor
might operate at the start of a musical score.

We know as much as we do about the speech components of discursive practices
because we remove them from the flow of “real time” by writing them down.
Alphabetic scripts are a clear mode of specification for components of spoken
languages at a phonological level. A script designed to produce movement texts
establishes similar conditions for the specification and registration of the bodily,
spatial, and dynamic components of actions. This makes translation into words
unnecessary for creating ethnographically appropriate descriptions of action. In
addition, when actions and words integrate, as is most often the case in social
interaction, both can be given equal analytical weight.

In addition, embodied aesthetic sensibilities such as voice quality and melody can
underlie notions of propriety, correctness, appropriateness, and honorification in
language use. Laderman (1987), for example, illustrates how only “beautiful voices”
are efficacious in Malay ritual. Irvine (1992) demonstrates how Wolof speakers
ideologically associate vocally embodied affectivity with honorification.

Collaboration with Native Speakers

Since meanings of utterances are embedded within the context in which they
occur, it’s important to note not only what explicit propositions are made but other

NOT
(negative marker)

LIVE/STAY
(human)

(¡pHi-

nen

ec’ake

(İpi,

I
İ

HERE

0 ‘
6
—

– a N LIVE/STAY
2 ) (human)

F
Nak’ota

I NAKOTA

HERE

(juncture)

LONG AGO

1. W‘aną—kai, nen. Nak’ota ąpi, ec’ake nen úpis.
Long ago here Nakota they live/ be always here they live not
Lo—ng ago, the Indians that live here now, did not always live here.

Figure 7. Assiniboine storytelling with Plains Sign Talk and Nakota: p. 1 of the Labanotated
score (from Farnell 1995a:83. Reproduced by permission of the University of Texas Press.).

438 FARNELL / GRAHAM

features of the environment that might have some relationship to an utterance’s
meanings. Many studies show how meaning emerges from context; others show how
the internal structure of stories reflects embeddness in larger interactive processes
(M. Goodwin 1982a, 1982b; Heath 1983;Goodwin and Duranti 1992:11). Since con
sultants are always active participants, collaborating with them during transcription,
translation, and interpretation is crucial to discovering presupposed knowledge,
interpretive frames, and the relevanceof prior discourse. Basso (1979,1990), Briggs
(1986, 1988), and Hymes (1981), among others, illustrate the advantages of long
term collaborations with consultants. Farnell (1995a), Graham (1995), and Valentine
(1995) explicitly address how consultant and community concerns can influence the
shape and scope of research. The methods used will depend on research goals as
well as practical and political realities of the fieldwork situation.10 Collaboration
itself is a socially constructed relationship that develops throughout fieldwork.

To better understand native perspectives in instances of socially situated speech,
recent discourse-focused research has attempted to directly involve native speakers
in transcription and analysis. In transcribing and analyzing the speech of Kaluli
children, for example, Schieffelin worked with the mothers of children she had
recorded (1990:30-32). They helped with initial transcription by repeating back their
own and others’ speech as they listened to the tape-recorded samples. Schieffelin
notes that this was especially helpful in glossing special forms children used,
identifying speakers in multiparty talk, and specifying the history and ownership of
objects involved in the interactions.

The women also provided important metalinguistic commentary that made
explicit Kaluli notions about proper language use. Schieffelin took careful notes on
the women’s comments and incorporated them into the extensive contextual notes
she had made while recording. This metacommentary became part of extensive text
annotatations and Schieffelin used it to generate further questions about interactional
patterns, classifications of speech acts and events, and notions of what constituted
“good talk” (1990:30). She notes that for every hour of recorded speech, she spent
an additional 20 hours preparing the annotated and translated transcript. Given the
recognition that discourse is both saturated with and constitutive of social meaning,
this is neither surprising nor atypical for discourse-centered work.

To obtain further insights into the recorded interactions, Schieffelin incorporated
a “reliability check” into her work (1990:30-31). This involved relistening and
reviewing the transcriptions with a young man who hadn’t participated in the
interactions. Schieffelin originally conceived of this exercise as a means to check
her initial transcriptions, but she discovered that it was a way of “extending and
enriching the contextualizing ethnographic and linguistic information” (p.31) that
the mothers had provided. Work with the young man revealed that devices such as
prosody, voice quality, and affect-marked affixes convey significant affective infor
mation and that pragmatic uses of word order can disambiguate utterances and
clarify issues. To avoid influencing consultants’ interpretations of video-recorded

Discourse-Centered Methods 439

data, Erickson and Shultz (1982:59-63) developed a technique that allowed con
sultants to stop the tape wherever they found items worthy of attention and to
comment on them in interactions they were watching. They found that consultants
usually chose to stop at points where discourse topics or conversational routines had
just been completed.

Graham (1995:15-16) worked with Xavante assistants in ways similar to
Schieffelin but used two tape recorders as she was transcribing and translating field
recordings with native speakers. While one tape recorder was used to play back
recorded instances of contextually situated discourse, the other recorded the consul
tants’ explanations and responses to questions Graham posed about the interactions
and language use.

Using two tape recorders enables the investigator to return repeatedly to
informants’ responses to the original taped utterances. It also permits the researcher
to discover discrepancies between forms that appear in an initial recording and what
informants “repeat” for transcription (see Urban 1996a). Such discrepancies may be
due to a straightforward difference between citation form (pronunciations given in
elicitation or interview contexts) and those used in actual practice, but not neces
sarily. Factors such as a difference in age, class, or gender may be at work, and
comparisons may reveal social and pragmatic meanings associated with such
alternative forms. It was in using this technique, for example, that Graham discerned
significant pragmatic phenomena such as features associated with the speech of
elders and alternative patterns of person marking and verb morphology used in
speech to, or about, certain kin and affines. These forms, Graham (1990) ascer
tained, are not typically given in elicitation situations.

In a similar manner, Farnell made important discoveries about Assiniboine
conceptions of the integrated nature of speech and signing and the primacy of
movement in indigenous classifications by videotaping her own interaction with
informants during collaborative transcription and translation sessions as they
watched and listened to their own previously recorded storytelling performances
(Farnell 1995a:247). She had to change her methodological decision to transcribe
first the speech and then the action because one of her informants consistently failed
to repeat the words he had spoken on the videotape. It became apparent that he was
adding spoken material in order to ensure that meanings contained in the gestural
component of his utterances were also included. Only when both speech and action
were discussed and transcribed simultaneously was he satisfied that what he had
“said” was all there. As in most discourse centered research, consultants’ comments
on recorded texts provided rich contextualizing information and can be used to
generate further questions and elicitations. Briggs (1986:99-100) also notes that
participants commented extensively whem reviewing their videotaped interactions.

As these examples show, discourse-centered research incorporates elicitation as
a fieldwork technique. However, such elicitation never substitutes for socially
situated utterances: Elicitation and interviewing always follow and support the

440 FARNELL I GRAHAM

former. Elicitation is limited because, although it frequently enables researchers to
determine basic grammatical patterns, when and why social actors use certain forms
in specific situations for what kind of pragmatic ends remain to be investigated.

Interviews, too, can supplement direct observation by providing important infor
mation. (For an excellent discussion of problems in interview based research, see
Briggs [1986] and Urban [1996b]). Communities and individuals within com
munities vary in the extent to which they are able to provide metalinguistic
commentary (that is, talk about) any formal features of discourse or criteria for
judging performance (see Hymes 1981; Silverstein 1981). In cultures that don’t
engage in elaborate metalinguistic glossing, native speakers may not identify dis
tinctive styles or genres or may not be able to describe characteristics of speech or
criteria for evaluating performance. Observing how participants in an event react to
performances can shed light on such features. For example, noting aside com
mentary and what causes laughter, smiling, frowning, silence, changes in bodily
postures, or responsive gestures can be very instructive. Such visible responses also
require careful translation from one cultural context to another, however, and
potential meanings associated with such “action signs” or “body languages” must
also be ascertained ethnographically, rather than assumed.”

Transcription and translation themselves must be viewed as contextually situated,
emergent dialogic acts which may generate novel meanings. Accordingly, these tasks
have become the focus of productive and provocative analytic inquiry (see Tedlock
1983; Haviland 1996; Urban 1996a; also papers in Sammons and Sherzer In press).
Recent studies of roles of native informants in language-focused work challenge
researchers to reflect on the situated nature of producing transcriptions, translations,
and analyses. Anthropologists have long recognized the significance individual
informants and their social positions play in research endeavors. In discourse-
focused work, native consultants play important and unique roles. Not only can their
social position influence whether they will transcribe or translate the speech of
certain individuals (see Graham 1995:13-16), but relationships between speakers and
consultants may influence the meanings that are understood to be encoded in texts,
and even how speech itself is transcribed.

Haviland (1996) examines how Tzotzil Mayas from the highlands of Chiapas,
who have little experience with reading and writing, transform aspects of recorded
conversation in producing native written texts of multiparty talk. He notes that
natives change written texts in ways that “normalize” significant pragmatic features
of conversational situations. This includes erasing context-specific elements of
speech, such as quotative particles and evidentials that relate utterances to previous
conversation. Natives also smooth “interactional edges,” eliminate false starts,
disfluencies, asides, and Spanish loan words from their transcriptions.

Urban (1996a) observes that there may be predictable regularities in how native
transcriptions of recorded speech differ from original utterances. He proposes that
when replication occurs in relatively deliberate contexts (as in transcription) the

Discourse-Centered Methods 441

copy may include segmentable forms that do not occur in the original but that
encode meanings that are only pragmatically inferable from the original. He also
notes that native transcribers may delete metadiscursive instructions, especially
indications of mistakes or deviations from an intended “text” (as in a myth), and
may “correct” mistakes indicated by such remarks. Urban proposes that power
relations between the speaker of the original text and the “copier” (or transcriber)
may influence the production of a written text. He says that when the relationship
between the originator and copier is more symmetrical and egalitarian, there will be
a greater divergence between the copy and the original; the copier is also more
likely to respond to the originator. In contrast, where there is greater power dif
ferential, transcribers attempt to render versions that are more faithful to the original
text. These observations have important consequences because the ways in which
a text is transcribed influences its subsequent translation and analysis.

Concluding Remarks

Discourse-centered approaches in linguistic anthropology position speech and other
expressive acts as the principle means by which social actors create and recreate
social life. They thus open up new ways to conceive of “culture.” This perspective
offers new resources for investigating some traditional problems in anthropology
and related disciplines, and for discovering how processses of cultural change and
continuity take place.

The focus of analysis in discourse-centered methods is on data collected in actual
social life. With a corpus of transcribed and translated texts in hand, researchers
may examine formal discourse features and grammatical patterning to discover ways
in which such devices are implicated in broader social and cultural phenomena.
Scholars have examined such features as evidentials, deictics, reported speech,
metadiscursive particles and phrases, semantic and grammatical patterning, as well
as formal features of performance such as the organization of talk.

In addition to such microconstituents, formal analysis may focus on macro
structures such as the formal organization of entire narrative texts (see Hymes 1981 ;
Urban 1986; also Bernard and Ryan, this volume) or on relationships between entire
genres (see Graham 1986; Seeger 1986). Work with native assistants in transcription
and translation, including metacommentary and elicitation, will help tease out
further significant features. Through detailed attention to such formal elements, as
well as examination of the ways in which creative indexical reference conveys addi
tional social meaning in communicative acts, researchers seek to discover how social
phenomena, such as relations of power, kinship, identity, genderrelations, aesthetics,
ideologies and affect, are discursively constituted and negotiated.

Ways in which wider social phenomena will be encoded in discourse can rarely
be anticipated before ethnographic fieldwork is well underway. They are usually

442 FARNELL I GRAHAM Discourse-Centered Methods 44J

discovered during fieldwork and necessarily vary according to such factors as
interactions between the ethnographer and the individuals and community with
whom she or he works, as well as formal elements of the texts themselves. More
over, just as meanings and interpretations are co-produced among interactants in any
ethnographic situation, anthropological understandings of discourse forms are inter-
subjectively achieved in field research. Outside of the context of fieldwork,
interpretations continue to be intersubjectively achieved in dialogue with academic
colleagues and the intellectual capital of linguistic anthropology.

Discourse-centered research also entails ongoing relationships between tech
nologies, method, and theory. In an ever-evolving dynamic process, new theoretical
questions and insights challenge researchers to devise new methods and adopt or
devise new technologies which, in turn, stimulate more theoretical questions and so
on. Since theory and method are closely intertwined and are part of the dialogical
processes through which we, as social scientists construct ethnographic knowledge,
we can expect that new methods of research will continue to arise.

NOTES

We thank Russ Bernard and Jane Hill for their helpful suggestions and the opportunity
for collaboration. We also thank Andy Orta, Jon Pressman, Joel Sherzer, Bambi Schieffelin,
Greg Urban, and Bonnie Urciuoli for their comments on an earlier version. We are grateful
to Tony Seeger, Marilyn Graf, and Judith Gray for helping us find information about early
recordings and technologies. We appreciate tire help of Jon Wolseth and Jerry Wever in
preparing the manuscript. Each author thanks the other for patience, good humor and intel
lectual companionship. Our contributions are shared; our names appear in alphabetical order.

I. We do not attempt an exhaustive survey of the literature on discourse analysis or
linguistic anthropology generally. See Gumperz and Hymes (1986 [1972]) for a survey of
work in sociolingistics and the ethnography of speaking up to that date. See Bauman and
Briggs (1990) for later contributions.

2. See Hymes (1971) for a historical overview of the emergence of sociolinguistics and
the ethnography of speaking as a contrast to the canon in mainstream linguistic theory of the
1960s and 1970s. For further background, see Hymes (1974), Bauman (1977), Bauman and
Sherzer (1989), and Bauman and Briggs (1990).

3. See papers in Järvellä and Klein (1982) and Pick and Acredolo (1983) for further
references on this topic. Agah (1996), Benveniste (1971 [1956]), Graham (1995:175-206),
Mühlhäusler and Harré (1990), and Urban (1989) provide studies of pronoun systemsand
uses.

4. Urban (1991:16) notes that among Boasian anthropologists, Paul Radin’s work stands
out because he discusses the methods by which texts in the native language were collected
prior to the invention of the tape recorder. These consist of standard informant elicitation and
dictation, the writing down of myths in the Winnebago syllabury by the Winnebego
themselves, and the use of the Edison phonograph to record elicited tales from which

transcriptions were made (Radin 1949). Radin’s work strikes us as remarkably contemporary
because he employed the discourse of native accounts in his “method of reconstruction from
internal evidence” (1965 [1933]:183-252).

5. In calling for ethnographies of communication, Gumperz and Hymes (1964) did not
intend to stimulate disproportionate attention to marked speech styles. This may have
occurred because researchers sought to call attention to the richness of verbal life cross-
culturally. The tendency parallels disproportionate attention to ritual activities and ceremonial
life on the part of sociocultural anthropologists generally. Recent ethnography, both socio
cultural and linguistic, seeks to remedy this through increased attention to “everyday” life.
See, for example, Abu-Lughod (1993), Schieffelin (1990), and Ochs (1988).

6. For a guide to IPA and other systems of phonetic representation, both linear and
parametric, see MacMahon (1996). Articulatory phoneticians concerned with the dynamics
of vocal production now use moving x-ray video images to study the production of “vocal
gestures” (see Studdert-Kennedy and Goodall 1992).

7. For discussion of early recording technologies, see Gelatt (1977) and Welch and Burt
(1994); see also Copeland (1991).

8. Turner recorded Sea Island Gullah dialects on 154 12-inch aluminum discs, recorded
at 78 rpm. The technology he used to make these field recordings was very cumbersome.

9. SoundEdit 16 is available from MacroMedia. Signalyze is available from Network
Technology Corporation, 91 Baldwin St, Charlestown, MA 02129. Other sound analysis
programs include Capmedia and Kay CSL (computerized speech lab).

10. The political entailments of field research are widely acknowledged in anthropology
(see Rabinów 1977; Fox 1991; Clifford and Marcus 1986). They have received less attention
in linguistics (see Aissen 1992). Briggs (1986) discusses assymetries of power in linguistic
anthropology with a focus on the interview.

11. Anthropologists of Human Movement use the term “body language” to mean systems
of action signs with distinct spatial grammars and multiple functions that constitute one of
the semiotic practices distinctive of human life. These are often closely integrated with
spoken language conceptions, often having given rise to spoken propositional forms, and are
certainly never independent of them. We do not intend to invoke popular misconceptions of
the term “body language” that connote one-to-one correlations between a movement and “its”
meaning. See Williams (1991) and Farnell (1995b) for further information on this subfield.

444 FARNELL / GRAHAM Discourse-Centered Methods 445

REFERENCES

Abu-Lughod, Lila. 1989. Bedouins, Cassettes, and Technologies of Public Culture. Middle
East Report 159.1-W.

Abu-Lughod, Lila. 1993. Writing Women’s Worlds: Bedouin Stories. Berkeley: University
of California Press.

Agah, Asif. 1996. Honorifics. Annual Review of Anthropology 22(4):473-488.

Aissen, Judith. 1992. Fieldwork and Linguistic Theory. International Encyclopedia of
Linguistics, Vol. 2. William Bright, ed. Pp. 9-10. New York: Oxford University
Press.

Ardener, Edwin. 1989 [1978]. Some Outstanding Problems in the Analysis of Events. In
Edwin Ardener: The Voice of Prophecy and Other Essays. M. Chapman, ed. Pp. 86-104.
Oxford: Basil Blackwell.

Austin, John L. 1962. How to Do Things with Words, 2d ed. J. 0. Urmson and M. Sbisa, eds.
Cambridge: Harvard University Press.

Bakhtin, M. M. 1981. The Dialogic Imagination. C. Emerson and M. Holquist, trans. Austin:
University of Texas Press.

Bakhtin, M. M. 1986. Speech Genres and Other Late Essays. V. W. McGee, trans. Austin:
University of Texas Press.

Basso, Ellen. 1995. The Last Cannibals: A South American Oral History. Austin: University
of Texas Press.

Basso, Keith. 1972. “To Give Up on Words”: Silence in Western Apache Culture. In
Language and Social Context. Pier Paolo Giglioli, ed. Pp. 67-86. Harmondsworth,
England: Penguin Books.

Basso, Keith. 1979. Portraits of the Whiteman: Linguistic Play and Cultural Symbols among
the Western Apache. New York: Cambridge University Press.

Basso, Keith. 1990. WesternApache Language and Culture. Tucson: University of Arizona
Press.

Bauman, Richard. 1977. Verbal Art as Performance. Rowley, MA: Newbury House.
Reprinted by Waveland Press, 1984.

Bauman, Richard. 1992. Contextual ization, Tradition and the Dialogue of Genres: Icelandic
Legends of the Kraftaskáld. In Rethinking Context: Language as an Interactive Phe
nomenon. C. Goodwin and A. Duranti, eds. Pp. 125-145. Cambridge: Cambridge
University Press.

Bauman, Richard, and Charles Briggs. 1990. Poetics and Performance as Critical Perspectives
on Language and Social Life. Annual Review of Anthropology 19:59-88.

Bauman, Richard, and Joel Sherzer, eds. 1989 [1974], Explorations in the Ethnograph of
Speaking, 2d ed.

Cambridge: Cambridge University Press.

Becker, A. L. 1983. Biography of a Sentence: A Burmese Proverb. In Text, Play, and Story:
The Construction and Reconstruction of Self and Society. Edward M. Bruner, ed. Pp.
185-210. Proceedings of the American Ethnological Society, American Anthropological

Association. Reprinted in 1995 in Beyond Translation: Essays Toward a Modern
Philology. Ann Arbor: University of Michigan Press.

Benveniste, Emile. 1971a [1956]. The Nature of Pronouns. In Problems in General
Linguistics. M. E. Meek, trans. Pp. 217-222. Coral Gables: University of Miami Press.
Originally published in 1956 in For Roman Jakobson. M. Halle, H. G. Lunt, H. McLean,
and C. H. van Schooneveld, eds., pp. 34-37. The Hague and Paris: Mouton.

Benveniste, Emile. 1971b. Subjectivity in Language. In Problems in General Linguistics.
M. E. Meek, trans. Pp. 223-230. Coral Gables: University of Miami Press.

Boas, Franz. 1940. Race, Language and Culture. New York: Macmillan.

Bourdieu, Pierre. 1977a. Outline of a Theory of Practice. R. Nice, trans. Cambridge:
Cambridge University Press.

Bourdieu, Pierre. 1977b. The Economics of Linguistic Exchanges. Social Science Infor
mation 16:645-668.

Bourdieu, Pierre. 1991. Language and Symbolic Power. Cambridge: Harvard University
Press.

Bowen, John. 1991a. Gayo Speech and Song. Audiotape. New Haven: Yale University
Press.

Bowen, John. 1991b. Sumatran Politics and Poetics: Gayo History, 1900-1989. New
Haven: Yale University Press.

Brandt, Elizabeth. 1981. Native American Attitudes Toward Literacy and Recording in the
Southwest. The Journal of the Linguistic Association of the Southwest 4(2): 185-195.

Brenneis, Donald. 1984. Straight Talk and SweetTalk: Political Discourse in an Occasionally
Egalitarian Community. In Dangerous Words: Language and Politics in the Pacific.
D. L. Brenneis and F. R, Myers, eds. Pp. 69-84. New York: New York University Press.

Brenneis, Donald. 1986. Shared Territory: Audience, Indirection, and Meaning. Text
6(3):339-347.

Briggs, Charles. 1986. Learning How to Ask? A Sociolingüístic Appraisal of the Role of the
Interview in Social Science Research. Cambridge: Cambridge University Press.

Briggs, Charles. 1988. Competence and Performance: The Creativity of Tradition in
Mexicano Verbal Art. Philadelphia: University of Pennsylvania Press.

Briggs, Charles. 1992. “Since I Am a Woman, I Will Chastize My Relatives”: Gender,
Reported Speech and the (Re)Production of Social Relations in Warao Ritual Wailing.
American Ethnologist 19(2):336-361.

Briggs, Charles. 1993. Personal Sentiments and Polyphonic Voices in Warao Women’s Ritual
Wailing: Music and Poetics in a Critical and Collective Discourse. American Anthro
pologist 95(4):929-957.

Brown, Roger, and Albert Gillman. 1960. The Pronouns of Power and Solidarity. In Style in
Language. T. Sebeok, ed. Pp. 253-276. Cambridge: M.I.T. Press.

Caton, Steve. 1990. “Peaks of Yemen I Summon”: Poetry as Cultural Practice in a North
Yemeni Tribe. Berkeley: University of California Press.

446 FARNELL / GRAHAM Discourse-Centered Methods 447

Chafe, Wallace. 1993. Seneca Speaking Styles and the Location of Authority. In Respon
sibility and Evidence in Oral Discourse. Jane Hill and Judith Irvine, eds. Pp. 72-87.
Cambridge: Cambridge University Press.

Cicourel, A. V. 1981. Language and Medicine. In Language in the USA. C. A. Ferguson,
and S. B. Heath, eds. Pp. 407-429. New York: Cambridge University Press.

Cicourel, A. V. 1982. Language and Belief in a Medical Setting. In Contemporary
Perceptions of Language: Interdisciplinary Dimensions. Heidi Byrnes, ed. Pp. 48-78.
Washington, DC: Georgetown University Press.

Cicourel, A. V. 1992. The Interpenetration of Communicative Contexts: Examples from
Medical Encounters. In Rethinking Context: Language as an Interactive Phenomenon. C.
Goodwin and A. Duranti, eds. Pp. 1-42. Cambridge: Cambridge University Press.

Clifford, James, and George E. Marcus. 1986. IPriting Culture: The Poetics and Politics of
Ethnography. Berkeley; University of California Press.

Copeland, Peter. 1991. Sound Recordings. London: The British Library.

Darnell, Regna. 1989 [1974], Correlates of Cree Performance. In Explorations in the Eth
nograph of Speaking, 2d ed. Richard Bauman and Joel Sherzer, eds. Pp. 315-336. New
York: Cambridge University Press.

Davies, Bronwyn, and Rom Harré. 1990. Positioning: The Discursive Production of Selves.
Journal for the Theory of Social Behavior 20(1):43-63.

Du Bois, John. 1993 [1987], Meaning Without Intention: Lessons from Divination. Papers
in Pragmatics 1(2):80-122. Reprinted in Responsibility and Evidence in Oral Discourse.
Jane Hill and Judith Irvine, eds. Pp. 48-71. Cambridge: Cambridge University Press.

Duranti, Alessandro. 1986. The Audience as Co-Author. Text 6(3):239-247.

Duranti, Alessandro. 1992. Language and Bodies in Social Space: Samoan Ceremonial,
Greetings. American Anthropologist 94(3):657-691.

Duranti, Alessandro. 1993. Intentions, Self, and Responsibility: An Essay in Samoan Ethno-‘
pragmatics. In Responsibility and Evidence in Oral Discourse. Jane Hill and Judith Irvine,
eds. Pp. 24-47. Cambridge: Cambridge University Press.

Duranti, Alessandro.
Press.

1997. Linguistic Anthropology. Cambridge: Cambridge University

Duranti, Alessandro, and Charles Goodwin, eds. 1992. Rethinking Context: Language as at.
Interactive Phenomenon. Cambridge: Cambridge University Press.

Edwards, Jane A., and Martin D. Lampert. 1993. Talking Data:
in Discourse Research. Hillsdale, NJ: Lawrence Erlbaum.

Transcription and Coding

Erickson, Frederick, and Jeffrey Shultz. 1982. The Counselor as Gatekeeper: Socit
Interaction in Interviews. New York: Academic Press.

Ellis, John. 1994. Language, Thought, and Logic. Evanston, IL: Northwestern Universi
Press.

Farnell, Brenda. 1994. Ethno-Graphics and the Moving Body. Man: Journal of the Roy.
Anthropological Institute 29(4):929-974. ■

Farnell, Brenda. 1995a. Do You See What I Mean?: Plain Indian Sign Talk and the
Í Embodiment of Action. Austin: University of Texas Press.

Farnell, Brenda. 1995b. Introduction. In Human Action Signs in Cultural Context: The
Í Visible and the Invisible in Movement and Dance. B. Farnell, ed. Pp. 1-28. Metuchen, NJ:

Scarecrow Press.

Farnell, Brenda. 1995c. Wiyuta: Assiniboine Storytelling with Signs. CD-ROM. Austin:
University of Texas Press.

Feld, Steven. 1990 [1982], Sound and Sentiment: Birds, Weeping, Poetics, andSonginKaluli
Expression, 2d ed. Philadelphia: University of Pennsylvania Press.

Wd, Steven, and Aaron A. Fox 1994. Musicand Language. Annual Review of Anthropology
23:25-53.

lishman, Joshua. 1968. Readings in the Sociology of Language. The Hague and Paris:
Mouton.

lishman, Joshua. 1970. Sociolinguistics: A Brief Introduction. Rowley, MA: Newbury House.

‘ox, Richard. 1991. Recapturing Anthropology: Workingin the Present. SantaFe: School of
American Research Press.

Jal, Susan. 1991. Between Speech and Silence: The Problematics of Research on Language
and Gender. In Gender at the Crossroads of Knowledge: Feminist Anthropology in the

f Postmodern Era. Michaela Di Leonardo, ed. Pp. 175-203. Berkeley and Los Angeles:
; University of California Press.
Garfinkel, Harold. 1967. Studies in Ethnomethodology. Englewood Cliffs, NJ: Prentice Hall.

Garfinkel, Harold. 1972. Remarks on Ethnomethodology. In Directions in Sociolinguistics.
! J. Gumperz, ed. Pp. 301-324. New York; Holt, Rinehart and

Winston.

Garvey, Catherine. 1984. Children’s Talk. Cambridge: Harvard University Press.

Gelatt, Roland. 1977. The Fabulous Phonograph 1877-1977, 2d rev. ed. New York:
, Macmillan.

Giddens, Anthony. 1979. Central Problems in Social Theory: Action, Structure and
Contradiction in Social Analysis. Berkeley: University of California Press.

iddens, Anthony. 1984. The Constitution of Society. Berkeley: University of California
Press.

oddard, Ives. 1996. The Description of the Native Languages of North America Before
“Boas. In Handbook of North American Indians. Vol. 17, Languages. William C.

Sturtevant, ed. Pp. 17-42. Washington DC: Smithsonian Institution Press.

offinan. Erving. 1961. Encounters: Two Studies in Social Interaction. Indianapolis: Bobbs-
Merrill.

ffman. Erving. 1972 [1967]. Interaction Ritual. Harmondsworth, England: Penguin Books.
ffman. Erving. 1974. Frame Analysis. New York: Harper and Row.

ffman. Erving. 1981. Forms of Talk. Philadelphia: University of Pennsylvania Press.
odwin, Charles. 1981. Conversational Organization: Interaction Between Speakers and
Hearers. New York: Academic Press.

448 FARNELL ! GRAHAM Discourse-Centered Methcds 449

Goodwin, Charles. 1984. Notes on Story Structure and the Organization of Participation. In
Structures of Social Action. M. Atkinson and J. Heritage, eds. Pp. 225-246. Cambridge:
Cambridge University Press.

Goodwin, Charles. 1993. Recording Human Interaction in Natural Settings. Pragmatics
3(2): 181-209.

Goodwin, Charles and John Heritage. 1990. Conversational Analysis. Annual Review of
Anthropology 19:283-307.

Goodwin, Charles, and Alessandro Duranti. 1992. Rethinking Context: An Introduction. In
Rethinking Context: Language as an Interactive Phenomenon. C. Goodwin and A.
Duranti, eds. Pp. 1-42. Cambridge: Cambridge University Press.

Goodwin, Majorie. 1982a. “Instigating”: Storytelling as a Social Process. American Ethnol
ogist 9:799-819.

Goodwin, Majorie. 1982b. Processes of Dispute Management among Urban Black Children.
American Ethnologist 9:76-96.

Goodwin, Majorie. 1990. He-Said-She-Said: Talk as Social Organization Among Black :
Children. Bloomington: Indiana University Press.

Graham, Laura R. 1984. Semanticity and Melody : Parameters of Contrast in Sbavante Vocal
Expression. Latin American Music Review 5(2): 161-185.

Graham, Laura R. 1986. Three Modes of Sbavante Vocal Expression: Wailing, Collective
Singing, and Political Oratory. In Native South American Discourse. J. Sherzer and G.
Urban, eds. Pp. 83-118. Berlin: Mouton.

Graham, Laura R. 1990. The Always Living: Discourse and the Male Lifecycle
Xavante Indians of Central Brazil. Ph.D. diss., Austin: University of Texas.

of the

Graham, Laura R. 1993 in Amazonia? TheA Public Sphere Depersonalized Collaborative
Construction of Discourse in Xavante. American Ethnologist 20(4):717-741.

Graham, Laura R. 1994. Dialogic Dreams: Creative Selves Coming into Life in the Flow of
Time. American Ethnologist 21(4):719-741.

Graham, Laura R. 1995. Performing Dreams: Discourses of Immortality Among the Xavante
of Central Brazil. Austin: University of Texas Press.

Graham, Laura R. In press. The One With the Long Hair: Tellings, Meanings, and Inter-
Textuality in the Translation of Xavante Narrative. In Translating Native Latin American
Verbal Art: Ethnopoetics and Ethnography. K. Sammons and J. Sherzer, eds. Washington,
DC: Smithsonian Institution Press.

Grimshaw, Allan D., cd. 1990. Conflict Talk: Sociolingüístic Investigations of Arguments in
Conversations. Cambridge: Cambridge University Press.

Grimshaw, Allan D. 1994. What’s Going on Here? Complementary Studies of Professional
Talk. Norwood, NJ: Ablex Press.

Gumperz, John J. 1982. Discourse Strategies. Cambridge: Cambridge University Press. ;

Gumperz, John J., and Dell Hymes, eds. 1964. The Ethnography of Communication.
American Anthropologist, Special Publication 66, Number 6, Part II.

Gumperz, John J., and Dell Hymes, eds. 1986 [1972], Directions in Sociolinguistics: The
Ethnography of Communication. Oxford and New York: Basil Blackwell.

Hall, Kira, and Mary BuchoItz. 1995. Gender Articulated: Language and the Socially
Constructed Self. New York: Routledge.

Handler, Richard. 1988. Nationalism and the Politics of Culture in Quebec. Madison:
University of Wisconsin Press.

Hanks, Wiliam. 1990. Referential Practice: Language and Lived Space among the Maya.
Chicago: University of Chicago Press.

Harré, Rom, ed. 1984. Personal Being. Cambridge: Harvard University Press.

Harré, Rom. 1986a The Social Construction of Emotions. Oxford: Basil Blackwell.
Harré, Rom. 1986b. Mind as a Social Formation. In Rationality, Relativism and the Human

Sciences. J. Margolis, M. Krausz, and R. M. Burien, eds. Pp. 91-106. Dortrecht, the
Netherlands: Martinus Nijhoff.

Harré, Rom. 1986c. Varieties of Realism. Oxford: Basil Blackwell.

Harré, Rom. 1987. The Social Construction of Selves. In Self and Identity: Psycho-social
Perspectives. K. Yardley and T. Honess, eds. Pp. 41-52. New York: John Wiley.

Harré, Rom, and Grant Gillett. 1994. The Discursive Mind. London: Sage Publications.

Haugen, Einar. 1969. The Norwegian Language in America. Bloomington: Indiana Uni
versity Press.

Haviland, John. 1993. Anchoring, Iconicity and Orientation in Guugu Yimithirr Pointing
Gestures. Journal of Linguistic Anthropology 3(l):3-45.

Haviland, John. 1996. Text from Talk in Tzotzil. In Natural Histories of Discourse. M.
Silverstein and G. Urban, eds. Pp. 45-78. Chicago: University of Chicago Press.

Heath, Christian. 1986. Body Movement and Speech in Medical Interaction. Cambridge:
Cambridge University Press.

Heath, Shirley Brice. 1983. Ways with Words: Language, Life and Work in Communitiesand
Classrooms. Cambridge: Cambridge University Press.

Hill, Jane, and Kenneth Hill. 1986. Speaking Mexicano: Dynamics of Syncretic Language in
Central Mexico. Tucson: University of Arizona Press.

Hill, Jane, and Judy Irvine. 1993. Responsibility and Evidence in Discourse. Cambridge:
Cambridge University Press.

Hill, Jane, and Olivia Zepeda. 1993. Mrs. Patricio’s Trouble: The Distribution of
Responsibility in an Account of Personal Evidence. In Responsibility and Evidence in
Oral Discourse. Jane Hill and Judith Irvine, eds. Pp. 197-225. Cambridge: Cambridge
University Press.

Holquist, Michael. 1983. The Politics of Representation. Quarterly Newsletter of the
Laboratory of Comparative Human Cognition 5(l):2-9.

Hymes, Dell. 1962. The Ethnography of Speaking. In Anthropology and Human Behavior.
T. Gladwin and William C. Sturtevant, eds. Pp. 13-53. Washington, DC: Anthropological
Society of Washington.

450 FARNELL I GRAHAM Discourse-Centered Methods 451

Hymes, Dell. 1964. Introduction: Toward Ethnographies of Communication. In The
Ethnograph of Communication. John J. Gumperz and Dell Hymes, eds. American
Anthropological Association Special Publication 66, Number 6, Part 11:1-34.

Hymes, Dell. 1967. Models of the Interaction of Language and Social Setting. Journal of
Social Issues 23(2):8-28.

Hymes, Dell. 1971. Sociolinguistics and the Ethnography of Speaking. In Social
Anthropology and Language. Edwin Ardener, cd. Pp. 47-94. London: Tavistock Press.

Hymes, Dell. 1974. Foundations in Sociolinguistics: An Ethnographic Approach. Phila
delphia: University of Pennsylvania Press.

Hymes, Dell. 1975. The Pre-War Prague School and Post War American Anthropological
Linguistics. In Transformational Generative Paradigm and Modern Linguistic Theory.
E.K.F. Koerner, ed. Pp. 359-381. Amsterdam: John Benjamins.

Hymes, Dell. 1981. “In Vain I Tried to Tell You Essays in Native American Ethnopoetics.
Philadephia: University of Pennsylvania Press.

Irvine, Judith. 1989. When Talk Isn’t Cheap: Language and Political Economy. American
Ethnologist 16:24^-261.

Irvine, Judith. 1992. Ideologies of Honorific Language. Pragmatics 2(3):251-262.

Irvine, Judith. 1996. Shadow Conversations: The Indeterminacy of Participant Roles. In
Natural Histories of Discourse. G. Urban and M. Silverstein, eds. Pp. 131-159. Chicago:
University of Chicago Press.

Ives, Edward D. 1995. The Tape Recorded Interview: A Manual for Field Workers in
Folklore and Oral History, 2d ed. Knoxville: University of Tennessee Press.

Jakobson, Roman. 1960. Linguistics and Poetics. In Style in Language. Thomas Sebeok, ed.
Pp. 350-377. Cambridge: M.I.T. Press.

J akobson, Roman. 1971 [1957], Shifters, Verbal Categories and the Russian Verb. In Selected
Writings of Roman Jakobson 2:130-147. The Hague and Paris: Mouton.

Järvellä, Robert J. and Wolfgang Klein, eds. 1982. Speech, Place and Action: Studies in
Deixis and Related Topics. New York: John Wiley.

Josselson, Harry Hirsh. 1953. The Russian Word Count and Frequency Analysis of Gram
matical Categories of Standard Literary Russian. Detroit: Wayne State University Press.

Kaeppler, Adrienne. 1995. Visible and Invisible in Hawaiian Dance. In Human Action Signs
in Cultural Context: The Visible and Invisible in Movement and Dance. B. Farnell, ed.
Pp. 31-43. Metuchen, NJ: Scarecrow Press.

Keane, Web. 1991. Delegated Voice: Ritual Speech, Risk, and the Making of Marriage
Alliances in Anakalang. American Ethnologist 18:311-330.

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.

Keating, Elizabeth. In press. Power Sharing: Language, Rank, Gender and Social Space in
Pohnpei, Micronesia. Oxford: Oxford University Press.

Kendon, Adam. 1980. Gesticulation and Speech: Two Aspects of the Process of Utterance.
In The Relationship of Verbal and Non-Verbal Communication. M. R. Key, ed. Pp.
207-227. The Hague and Paris: Mouton.

Kendon, Adam. 1983. Gesture and Speech: How They Interact. In Nonverbal Interaction. J.
M. Wieman and R. P. Harrison, eds. Pp. 13-45. Sage Annual Reviews of Communication
Research, Vol. 11. Beverly Hills, CA: Sage Publications.

Kendon, Adam. 1989. Sign Languages of Aboriginal Australia. Cambridge: Cambridge
University Press.

Kroskrity, Paul, Bambi Schieffelin, and Kathryn Woolard, eds. 1992. Language Ideologies.
Special issue of Pragmatics 2(3).

Kuipers, Joel. 1990. Power in Performance: The Creation of Textual Authority in Weyewa
Ritual Speech. Philadelphia: University of Pennsylvania Press.

Labov, William. 1972. Sociolingüístic Patterns. Philadelphia: University of Pennsylvania
Press.

Labov, William, and David Fanshel. 1977. Therapeutic Discourse: Psychotherapy as Con
versation. New York: Academic Press.

Laderman, Carol. 1987. The Ambiguity of Symbols in the Structure of Healing. Social
Science and Medicine 24(4):293-301.

Lederman, Rena. 1984. Who Speaks Here? In Dangerous Words: Language and Politics in
the Pacific. D. L. Brenneis and F. R. Myers, eds. Pp. 85-107. New York: New York
University Press.

List, George. 1963. The Boundaries of Speech and Song. Ethnomusicology 7:1-16.
Lucy, John, ed. 1990. Reflexive Language: Reported Speech and Metapragmatics.

Cambridge: Cambridge University Press.

Lutz, Catherine. 1988. Unnatural Emotions: Everyday Sentiments on a MicronesianAtoll and
Their Challenge to Western Theory. Chicago: University of Chicago Press.

Lutz, Catherine, and Lila Abu-Lughod, eds. 1990. Language and the Politics of Emotion.
Cambridge: Cambridge University Press.

MacMahon, Michael K. C. 1996. Phonetic Notation. In The World’s Writing Systems. Peter
Daniels and William Bright, eds. Pp. 821-846. Oxford and New York: Oxford
University Press.

Mannheim, Bruce, and Dennis Tedlock. 1995. Introduction. In The Dialogic Emergence of
Culture. Bruce Mannheim and Dennis Tedlock, eds. Pp. 1-32. Urbana: University of
Illinois Press.

Manuel, Peter. 1993. Cassette Culture: Popular Music and Technology in North India.
‘ Chicago: University of Chicago Press.

McClendon, Sally. 1981. Meaning, Rhetorical Structure, and Discourse Organization in
Myth. In Analysing Discourse: Text and Talk. Deborah Tannen, ed. Pp. 284-305.

I Georgetown University Roundtable on Language and Linguistics, Washington, DC:
t• Georgetown University Press.

452 FARNELL / GRAHAM Discourse-Centered Methods 453

McNeill, David. 1985. So You Think Gestures Are Non-Verbal? Psychological Review
92(3):350-371.

McNeill, David. 1992. Hand and Mind: What Gestures Reveal about Thought. Chicago:
University of Chicago Press.

Mertz, Elizabeth, and Richard Parmentier. 1985. Introduction. In Semiotic Mediation. E.
Mertz and R. Parmentier, eds. Pp. 1-19. New York: Academic Press.

Michaels, Eric. 1994. Bad Aboriginal Art: Tradition, Media and Technological Horizons.
Minneapolis: University of Minnesota Press.

Mithun, Marianne. 1996. The Description of the Native Languages of North America: Boas
and After. In Handbook of North American Indians. Vol. 17, Languages. William C.
Sturtevant, ed. Pp. 43-63. Washington DC: Smithsonian Institution Press.

Moerman, Michael. 1988. Talking Culture: Ethnography and Conversation Analysis.
Philadelphia: University of Pennsylvania Press.

Mühlhäusler.Peter.andRomHarré. 1990. Pronouns and People: The Linguistic Construction 1
of Social and Personal Identity. Oxford: Basil Blackwell. i

Murphy, William. 1990. Creating the Appearance of Consensus in Mende Political Dis- J
course. American Anthropologist 92:24-41. 3

Murray, Thomas E., and Carmin Ross Murray. 1992. Legal and Ethical Issues in Surrep- å
titious Recording. Publication of the American Dialect Society 76. Tuscaloosa: University 1
of Alabama Press. 1

Nuckolls, Janice. 1992. Sound Symbolic Involvement. Journal of Linguistic Anthropology |
2(1):51-80. I

Ochs,Elinor. 1979a. Transcription as Theory. In Developmental Pragmatics. Elinor Ochs and i
Bambi B. Schieffelin, eds. Pp. 43-72. New York: Academic Press. a

Ochs, Elinor, 1979b. Introduction: What Child Language Can Contribute to Pragmatics. In A
Developmental Pragmatics. Elinor Ochs and Bambi B. Schieffelin, eds. Pp. 1-17. New§
York: Academic Press. 1

Ochs, Elinor. 1988. Culture and Language Development: Language Acquisition andm
Language Socialization in a Samoan Village. New York: Cambridge University Press

Page, JoAnne. 1990. A Comparative Study of Two Movement Writing Systems: Laban anda
Benesh Notation. Master’s Thesis, University of Sydney. I

Page, JoAnne. 1996. Images for Understanding: Movement Notations and Visual Recordings^
Visual Anthropology 8(2-4): 171-196. #

Philips, S„ S. Steele, and C. Tanz, eds. 1987. Language, Gender and Sex in Comparatt^K
Perspective. Cambridge: Cambridge University Press. MB

Pick, H. L., and L. P. Acredolo, eds. 1983. Spatial Orientation: Theory, Research and
Application. New York and London: Plenum Press. 9

Pratt, M. L. 1987. The Ideology of Speech Act Theory. Centrum N.S. J:5-18. 3
Pressman, Jon. 1997. Honorification and Projection on Saint Barthélemy. Anthropolog^^

Linguistics 39(1 ): 111-150. 2

Rabinów, Paul. 1977. Reflections on Fieldwork in Morocco. Berkeley: University of
California Press.

Radin, Paul. 1949. The Culture of the Winnebago: As Described by Themselves. Indiana
University Publications in Anthropology and Linguistics, Memoir 2. Baltimore: Waverly
Press.

Radin, Paul. 1965 [1933]. The Method and Theory of Ethnology: An Essay in Criticism. New
York: Basic Books.

Rorty, Richard ed. 1992 [1967]. The Linguistic Turn. Chicago: University of Chicago Press.
Rosaldo, Michelle. 1982. The Things We Do with Words: Hongot Speech Acts and Speech

Act Theory in Philosophy. Language in Society /7:203-235.
Rosaldo, Renato. 1980. Hongot Headhunting 1883-1974: A Study in Society and History.

Stanford: Stanford University Press.
Roseman, Marina. 1991. Healing Sounds from the Malaysian Rainforest: Temiar Music and

Medicine. Berkeley: University of California Press.

Sacks, Harvey, Emanuel A. Schegloff, and Gail Jefferson. 1974. A Simplest Systematics for
the Organization of Turn-Taking Conversation. Language 50:696-735.

Sahlins, Marshall. 1976. Culture and Practical Reason. Chicago: University of Chicago
Press.

Sahlins, Marshall. 1985. Islands of History. Chicago: University of Chicago Press.
Sammons, Kay, and Joel Sherzer. In press. Translating Native American Verbal Art:

Ethnopoetics and Ethnography of Speaking. Washington, DC: Smithsonian Institution
Press.

Sapir, Edward. 1949. Selected Writings of Edward Sapir in Language Culture and
Personality. David Mandelbaum, ed. Berkeley: University of California Press.

Saussure, Ferdinand de. 1916. Course de linguistique générale. Charles Bally and Albert
( Sechehaye, eds. Paris: Payot.

-Troike, Muriel. 1989. The Ethnography of Communication: An Introduction, 2d ed.
New York: Basil Blackwell.

Schafer, Murray R. 1997. The Tuning of the World. New York: Knopf.
Schegloff) Emanuel A. 1972. Sequencing in Conversational Opening. In Directions in
1 Sociolinguistics. J. Gumpertz, ed. Pp. 346-380. New York: Holt, Reinhart and

Winston.

Schegloff, Emanuel A., and Harvey Sacks. 1973. Opening Up Closings. Semiotica
”p 7:289-327.

hieffelin, Bambi. 1990. The Give and Take of Everyday Life: Language Socialization of
{Kaluli Children. New York: Cambridge University Press.

hieffelin, Bambi, and Elinor Ochs. 1986a. Language Socialization Across Cultures. New
York: Cambridge University Press.

hieffelin, Bambi, and Elinor Ochs. 1986b. Language Socialization. Annual Review of
Anthropology /5:163-191.

454 FARNELL / GRAHAM Discourse-Centered Methods 455

Searle, John R. 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge:
Cambridge University Press.

Seeger, Anthony. 1986. Oratory Is Spoken, Myth Is Told, and Song Is Sung, But They Are
All Music to My Ears. In Native South American Discourse. Joel Sherzer and Greg Urban,
eds. Pp. 59-82. Berlin: Mouton.

Seeger, Anthony. 1987. Why Suyá Sing: A Musical Anthropology of an Amazonian People.
Cambridge: Cambridge University Press.

Seeger, Anthony. 1991. Singing Other Peoples’ Songs. Cultural Survival Quarterly 15(3):
36-39.

Seeger, Anthony. 1997. Ethnomusicology and Music Law. In Borrowed Power: Essays on
Cultural Appropriation. Bruce Ziff and Pratima V. Rao, eds. Pp. 52-67. New Brunswick,
NJ: Rutgers University Press.

Sherzer, Joel. 1973. Verbal and Nonverbal Deixis: The Pointed Lip Gesture among the San
Blas Kuna. Language in Society 2:117-131.

Sherzer, Joel. 1987. A Discourse-Centered Approach to Language and Culture. American
Ethnologist 89:295-309.

Sherzer, Joel. 1992. Modes of Representation and Translation of Native American Discourse:
Examples from the San Blas Kuna. In On the Translation of Native American Literatures.
Brian Swann, ed. Pp. 426-440. Washington, DC: Smithsonian Institution Press.

Sherzer, Joel, and Regna Darnell. 1972. Outline Guide for the Ethnographic Study of Speech
Use. In Directions in Sociolinguistics: The Ethnography of Communication. John J.
Gumperz and Dell Hymes, eds. Pp. 548-554. New York: Holt.

Sherzer, Joel, and Greg Urban. 1986a. Introduction. In Native South American Discourse. Joel
Sherzer and Greg Urban, eds. Pp. 1-14. Berlin: Mouton.

Sherzer, Joel, and Greg Urban, eds. 1986b. Native South American Discourse. Berlin:
Mouton.

Sherzer, Joel, and Sammie Wicks. 1982. The Intersection of Language and Music in Kuna
Discourse. Latin American Music Review 3:147-164.

Shuy, Roger. 1986. Ethical Issues in Analyzing FBI Surreptitious Tapes. International ì
Journal of the Sociology of Language 62:119-128.

Silverstein, Michael. 1976. Shifters, Linguistic Categories, and Cultural Description. In ,
Meaning in Anthropology. K. H. Basso and H. A. Selby, eds. Pp. 11-56. Albuquerque:
University of New Mexico Press.

Silverstein, Michael. 1977. Cultural Prerequisites to Grammatical Analysis. In Georgetown ,
University Round Table on Languages and Linguistics: Linguistics and Anthropology. M. ;
Saville-Troike, ed. Pp. 139-151. Washington, DC: Georgetown University Press. -,

Silverstein, Michael. 1981. The Limits of Awareness. Austin: Southwest Educational;
Laboratories. 14

Silverstein, Michael. 1984. On the Pragmatic “Poetry” of Prose. In Georgetown University,
Roundtable in Linguistics, Vol. 18. D. Schiffrin, ed. Pp. 181-199. Washington,
Georgetown University Press.

Silverstein, Michael. 1985. The Culture of Language in Chinookan Narrative Texts; or On
Saying That … in Chinook. In Grammar Inside and Outside the Clause: Some
Approaches to Theory from the Field. J. Nichols and A. C. Woodbury, eds. Pp. 132-171.
Cambridge: Cambridge University Press.

Silverstein, Michael, and Greg Urban. 1996. Introduction. In Natural Histories of Discourse.
Greg Urban and Michael Silverstein, eds. Pp. 1-20. Chicago: University of Chicago
Press.

Spitulnik, Debra. 1993. Anthropology and Mass Media. Annual Review of Anthropology
22:293-325.

Streeck, Jürgen. 1993. Gesture as Communication I: Its Coordination with Gaze and Speech.
Communication Monographs 60:275-299.

Studdert-Kennedy, Michael, and Elizabeth Whitney Goodall. 1992. Gestures, Features and
Segments in Early Child Speech. Haskins Laboratory Status Report on Research. SR
111/112. Pp. 89-102.

Swann, Brian, ed. 1992. On the Translation of Native American Literatures. Washington, DC:
Smithsonian Institution Press.

Tannen, Deborah, ed. 1982. Analyzing Discourse: Text and Talk. Washington, DC: George
town University Press.

Tannen, Deborah. 1984. Conversational Style: Analyzing Talk Among Friends. Norwood, NJ:
Ablex Press.

Tannen, Deborah. 1994. Gender and Discourse. New York: Oxford University Press.
Tannen, Deborah. 1995. Waiting for the Mouse: Constructed Dialogue in Conversation. In

The Dialogic Emergence of Culture. Bruce Mannheim and Dennis Tedlock, eds. Pp.
198-218. Urbana: University of Illinois Press.

Taylor, Charles. 1985. Human Agency and Language. Cambridge: Cambridge University
Press.

Tedlock, Dennis. 1983. The Spoken Word and the Work of Interpretation. Philadelphia:
University of Pennsylvania Press.

Tedlock, Dennis. 1987. Hearing a Voice in an Ancient Text: Quiché Maya Poetics in
Performance. In Native American Discourse. Joel Sherzer and Anthony Woodbury, eds.
Pp. 140-175. Cambridge: Cambridge University Press.

esta, Renata. 1988. Interruptive Strategies in English and Italian Conversation: Smooth
Versus Contrastive Linguistic Preferences. Multilingua 7(3):285—312.

ibault, Paul J. 1997. Re-Reading Saussure: The Dynamics of Signs in Social Life. London
and New York: Routledge.

mer, Lorenzo Dow. 1969. Africanisms in the Gullah Dialect. New York: Amo.

mer, Terence. 1991. Representing, Resisting, Rethinking: Historical Transformations of
Kayapo Culture and Anthropological Consciousness. In Colonial Situations. History of
,Anthropology. Vol. 7. George Stocking, ed. Pp. 285-313. Madison: University of
Wisconsin Press.

456 FARNELL / GRAHAM Discourse-Centered Methods 457

Urban, Greg. 1984. Speech about Speech and Speech about Action Journal of American
Folklore 97:310-328.

Urban, Greg. 1985. Semiotics of Two Speech Styles in Shokleng. In Semiotic Mediation. E.
Mertz and R. Parmentier, eds. Pp. 311-329. New York: Academic Press.

Urban, Greg. 1986. Semiotic Functions of Macro-Parallelism in the Shokleng Origin Myth.
In Native South American Discourse. Joel Sherzer and Greg Urban, eds. Pp. 15-58.
Berlin: Mouton.

Urban, Greg. 1988. Ritual Wailing in Amerindian Brazil. American Anthropologist 90(2):
385-400.

Urban, Greg. 1989. The “I” of Discourse. In Semiotics, Self, and Society. Greg Urban and
Benjamin Lee, eds. Pp. 27-51. New York: Mouton.

Urban, Greg. 1991. A Discourse-Centered Approach to Culture: Native South American
Myths and Rituals. Austin: University of Texas Press.

Urban, Greg. 1996a. Entextualization, Replication, and Power. In Natural Histories of
Discourse. G. Urban and M. Silverstein, eds. Pp. 21-44. Chicago: University of Chicago
Press.

Urban, Greg. 1996b. Metaphysical Community: The Interplay of the Senses and the Intellect.
Austin: University of Texas Press.

Urciuoli, Bonnie. 1995. The Indexical Structure of Visibility. In Human Action Signs in
Cultural Context: The Visible and Invisible in Movement and Dance. B. Farnell ed. Pp.
189-215. Metuchen NJ: Scarecrow Press.

Urciuoli, Bonnie. 1996. Exposing Prejudice: Puerto Rican Experiences of Language, Race
and Class. Boulder: Westview Press.

Valentine, Lisa. 1995. Making It Their Own: Severn Ojibwe Communicative Practices.
Toronto: University of Toronto Press.

Varela, Charles R. 1995. Ethogenic Theory and Psychoanalysis: The Unconscious as a Social
Construction and a Failed Explanatory Construct. Journal for the Theory of Social
Behavior 25(4):363-385.

Varela, Charles, and Rom Harré. 1996. Conflicting Varieties of Realism: Causal Powers and
the Problems of Social Structure. Journal for the Theory of Social Behavior 26(3):
313-325.

Verschueren, J. 1985. What People Say They Do with Words. Norwood, NJ: Ablex.

Watson-Gegeo, Karen Ann, and Geoffrey White, eds. 1990. Disentangling: Conflict Dis
course in Pacific Societies. Stanford: Stanford University Press.

Welch, Walter, and Leah Brodbeck Stenzel Burt. 1994. From Tinfoil to Stereo: The
Acoustic Years of the Recording Industry 1877-1929. Gainesville: University Press of
Florida.

West, Candace. 1984. Routine Complications. Bloomington: Indiana University Press.
White, Geoffrey M., and John Kirkpatrick, eds. 1985. Person, Self and Experience: Exploring

Pacific Ethnopsychologies. Berkeley: University of California Press.

Whorf, Benjamin Lee. 1956. Language Thought, and Reality. Cambridge: M.I.T. Press.
Wiget, A. 1987. Telling the Tale: A Performance Analysis of a Hopi Coyote Story. In

Recovering the Word: Essays in Native American Literature. B. Swann and A. Krupat,
eds. Pp. 297-338. Berkeley: University of California Press.

Williams, Drid. 1975. The Role of Movement in Selected Symbolic Systems. D. Phil.,
Oxford University.

Williams, Drid. 1982. Semasiology. In Semantic Anthropology, ASA Vol. 22. D. Parkin, ed.
Pp. 161-181. London: Academic Press.

Williams, Drid. 1991. Ten Lectures on Theories of the Dance. Metuchen, NJ: Scarecrow
Press.

Williams, Drid. 1995. Space, Intersubjectivity and the Conceptual Imperative: Three
Ethnographic Cases. In Human Action Signs in Cultural Context: The Visible and the
Invisible in Movement and Dance. B. Farnell, ed. Pp. 44-81. Metuchen, NJ: Scarecrow
Press.

Williams, Drid, ed. 1996. Special Issue: Human Action Signs. Visual Anthropology 6(3/4).

Williams, Drid, and Brenda Farnell. 1990. A Beginning Text on Movement Writing for Non
Dancers. Canberra: Australian Institute for Aboriginal Studies.

Woodbury, Anthony. 1985. The Functions of Rhetorical Structure: A Study of Central
Alaskan Yupik Eskimo Discourse. Language in Society /4:153-190.

Woodbury, Anthony. 1987. Rhetorical Structure in a Central Alaskan Yupik Eskimo
Traditional Narrative. In Native American Discourse. J. Sherzer and A. Woodbury, eds.
Pp. 176-239. Cambridge: Cambridge University Press.

Woolard, Katherine. 1989. Double Talk: Bilingualism and the Politics of Ethnicity in
Catalonia. Stanford: Stanford University Press.

Woolard, Katherine. 1992. Language Ideologies: Issues and Approaches. Pragmatics
2(3):235-249.

Expert paper writers are just a few clicks away

Place an order in 3 easy steps. Takes less than 5 mins.

Calculate the price of your order

Type of paper needed:

Pages:

You will get a personal manager and a discount.

Academic level:

We'll send you the first draft for approval by at

Total price:

$0.00