228 Language and Interaction

social organization, social activities, symbolic and material resources, and interpre-
tive practices characteristic o a particular group o people.’ Ethnographers ask
themselves what is happening and they try to provide accounts which show how
the behavior that is being observed makes sense within the community that is being
observed. As Johnstone (2004, 76) says, ethnography ‘presupposes . . . that the best
explanations o human behavior are particular and culturally relative’ rather than
general and universal. Such studies are also qualitative rather than quantitative. In
ethnographies o speaking the ocus is on the language the participants are using
and the cultural practices such language re ects.
Canagarajah (2006, 155) observes that: ‘Ethnographers expect to live or an
extensive period o time in the community they are studying in order to capture
frst-hand its language patterns and attitudes. As much as possible, they try not to
alter the “natural” ow o li e and social relationships o the community, but under-
stand how language works in everyday li e.’ Tey are participant-observers and must
deal with the basic conundrum o participant observation, which rusting and
Maybin (2007, 578–9) explain as ollows: ‘Ethnographic work normally requires the
researcher to be actively involved in the social action under study, suggesting that
this generates insights which cannot be achieved in any other way. But the involve-
ment o the researcher in social action inevitably changes the language practices
under study.’ Tis issue may also become more and more important as di erences
increase between the linguistic and cultural backgrounds o the observer and the
observed. It is certainly one that must be con ronted by both those who publish
ethnographies and those who read them. Mendoza-Denton (2008, 48) addresses
this issue directly near the beginning o her ethnographic study o teenage Latina
girl groups in a Cali ornia high school:

No ethnographer is a blank notepad just as no linguist is a tape recorder. Te percep-

tual flters that we bring to feldwork situations are power ul indeed, and not always
conscious. You will read in the ollowing chapters an account that is my interpretation
o years o feldwork and research with a group o young people who allowed me into
their lives, and I will invite you to draw your own conclusions. I have been and will
be providing guideposts to show where my ethnographic interpretation might be
guided by actors such as my background, social class, and my own subjective and
a ective reactions to people around me and to events at the time.

She constantly reminds us in her report o the circumstances in which she collected
her data and o her involvement in the process.
Tree illustrative book-length ethnographic studies are those o Sherzer (1983),
Hill and Hill (1986), and Mendoza-Denton (2008). Sherzer describes how the Kuna
o Panama use language: their public language o the gathering house, and their use
o language in curing and music, in rites and estivities, and in everyday conversa-
tion. He points out that the Kuna wait very patiently to take their turns in speaking
so that interruptions and overlaps in conversation are rare events.
Hill and Hill describe how the Malinche o Central Mexico use language in their
daily lives and in their continuing struggle to preserve their linguistic and cultural
 Ethnographic Approaches in Sociolinguistics 229

identity. Spanish is constantly encroaching on their own language so they have

deliberately tried to maintain certain o its eatures in an almost ‘purist’ way.
Mendoza-Denton (2008) o ers an account o Latina gangs in a Cali ornia high
school. She calls the school, which is located in the San Francisco Bay area, Sor
Juana High School. She describes the students as a mixture o well-to-do Euro-
Americans, A rican Americans, Pacifc Islanders, Asians, Asian-Americans, and
Latinas/Latinos. She was particularly interested in this last group, especially the
girls. She ocused her research on the Norteñas and the Sureñas, two rival Latina
gangs. She studied these groups in depth, having become over a period o time the
confdant o members o both groups. She ound a strong ideological divide between
the groups. Te Norteñas were ‘northern’-oriented, pre erred to speak English, wore
red accessories and red lipstick, ‘ eathered’ their hair, avored Motown Oldie music
and the numbers XIV, 14, and 4, and, though Hispanic, were mainly US-born. In
contrast, the Sureñas were ‘southern’ (i.e., more Mexican)-oriented, pre erred to
speak Spanish, wore blue accessories and brown lipstick, ponytailed their hair,
avored Mexican bands, pop music, and the numbers XIII, 13, and 3, and were
mainly recent immigrants. Mendoza-Denton shows how the members o each
group express and rein orce their identities through their various practices and
some o the linguistic consequences o such behavior. For example, she ound that
the pre erred use o English or Spanish sometimes concealed a very good knowledge
o the dispre erred other language, and that certain linguistic eatures o Spanish
varied according to strength o commitment to the gangs. Mendoza-Denton’s study
ranges over a wide variety o issues and is a mine o suggestions and insights.
It is important to remember, however, that these studies represent the results o
lengthy and time-consuming ethnographic projects. It is also possible to do smaller-
scale studies using participant observation, ocusing on very specifc types o inter-
actions in a group and particular linguistic eatures. For example, a seminal work
by Frake (1964) ocuses on how to ask or a drink; while this study makes uses o
the author’s extensive knowledge o the culture, it is illustrative o how a narrowly
ocused question about linguistic behavior can lead to an insight ul analysis o
cultural norms. Another important early study by Mitchell-Kernan (1972) discusses
particular ways o speaking among some A rican Americans re erred to as ‘marking’
and ‘signi ying,’ ocusing on how cultural knowledge is needed to interpret certain
types o implied meanings. A third study which shows this specifc ocus is Basso
(1970), who discusses the meanings o silence in Western Apache. Students wishing
to do ethnographic research should note that although a deep understanding o the
cultures is necessary or the interpretation o the data in all cases, ocusing on very
specifc elements o communication helps to constrain the scope o these projects.
In the rest o this chapter, we will outline three ethnographic approaches which
have been part o the feld o sociolinguistics. Te frst, ethnography o communica-
tion, is the main ocus o this chapter, as it is by ar the most in uential and long-
standing use o ethnographic concepts and methodologies in the discipline o
sociolinguistics. We will also brie y cover ethnomethodology, which we will then
take up again in chapter 11 when we delve more deeply into a type o discourse
230 Language and Interaction

analysis called conversation analysis, which is derived rom ethnomethodology.

Finally, we will outline the approach called linguistic ethnography, which is a more
recently introduced approach in sociolinguistics.

Exploration 9.1: Cultural Norms in Idioms

In English, we have sayings about how people use language that reveal
certain attitudes about language and particular types of speakers. For
instance, we say ‘Children should be seen and not heard’ and ‘Loose lips
sink ships.’ There is a clear trend of sayings which value silence as well as
discretion, as in ‘Speech is silver, silence is golden’ and ‘Still waters run deep’
(however, linguist Rick Hallett has reported that his grandmother would
extend this idiom as follows: ‘ . . . and are damn dirty, and the devil lies at
the bottom.’) What idioms do you know about language, in English or any
other language? What do they imply about the role of language in society
or the desired linguistic behavior of (particular groups of) speakers?

The Ethnography of Communication

As discussed in chapter 1, the study o language involves more than just describing
the syntactic composition o sentences or speci ying their propositional content.
Sociolinguists are interested in the various things that people do with that
language.

Communicative competence

Te term communicative competence (introduced in chapter 1) is sometimes used

to describe the knowledge o how to use language in culturally appropriate ways.
Tis term was suggested by Hymes (1972) as a counter-concept to Chomsky’s lin-
guistic competence, which ocused on an ideal hearer-speakers’ knowledge o gram-
maticality o sentences in their native language. Hymes maintained that knowledge
o a language involved much more than that. Gumperz (1972, 205) explains the term
as ollows: ‘Whereas linguistic competence covers the speaker’s ability to produce
grammatically correct sentences, communicative competence describes his ability
to select, rom the totality o grammatically correct expressions available to him,
orms which appropriately re ect the social norms governing behavior in specifc
encounters.’
Working with an ethnographic or unctional approach, we may attempt to speci y
just what it means to be a competent speaker o a particular language. It is one thing
 Ethnographic Approaches in Sociolinguistics 231

to learn the language o the Subanun, but quite another to learn how to ask or a
drink in Subanun (see Frake 1964, mentioned above and discussed in more detail
below). o do the frst you need a certain linguistic competence; to do the latter you
need communicative competence. As Saville- roike (1996, 363) says:

Communicative competence extends to both knowledge and expectation o who may

or may not speak in certain settings, when to speak and when to remain silent, whom
one may speak to, how one may talk to persons o di erent statuses and roles, what
nonverbal behaviors are appropriate in various contexts, what the routines or turn-
taking are in conversation, how to ask or and give in ormation, how to request, how
to o er or decline assistance or cooperation, how to give commands, how to en orce
discipline, and the like – in short, everything involving the use o language and other
communicative dimensions in particular social settings.

Hymes (1972, 279) has argued that, in learning a language, children not only
must learn how to construct sentences in that language but also must ‘acquire
knowledge o a set o ways in which sentences are used. From a fnite experience o
speech acts and their interdependence with sociocultural eatures, they develop a
general theory o the speaking appropriate in their community, which they employ,
like other orms o tacit cultural knowledge (competence), in conducting and inter-
preting social li e.’ Hymes provides some examples o the kinds o learning that are
involved:

Tey come to be able to recognize, or example, appropriate and inappropriate inter-

rogative behavior (e.g., among the Araucanians o Chile, that to repeat a question is
to insult; among the zeltal o Chiapas, Mexico, that a direct question is not properly
asked (and to be answered ‘nothing’); among the Cahinahua o Brazil, that a direct
answer to a frst question implies that the answerer has not time to talk, a vague answer,
that the question will be answered directly the second time, and that talk can
continue).

In learning to speak we are also learning to communicate in ways appropriate to

the group in which we are doing that learning; this is sometimes called language
socialization. Tese ways di er rom group to group; consequently, as we move
rom one group to another or rom one language to another, we must learn the new
ways i we are to ft into that new group or to use that new language properly. Com-
municative competence is there ore a key component o social competence.
A amous study which ocuses on communicative competence is ound in Frake
(1964); it outlines kinds o speech used in drinking encounters among the Subanun
o the Philippines. Such encounters are very important or gaining prestige and or
resolving disputes. Frake describes how talk, what he calls ‘drinking talk,’ proceeds
in such encounters, rom the initial invitation to partake o drink, to the selection
o the proper topics or discussion and problems or resolution as drinking proceeds
competitively, and fnally to the displays o verbal art that accompany heavy, ‘suc-
cess ul’ drinking. Each o these stages has its own characteristics. Tose who are the
232 Language and Interaction

most accomplished at drinking talk become the de acto leaders among the Subanun
because success ul talk during drinking may be used to claim or assert social leader-
ship. Success gives one a certain right to manipulate others, because it is during such
talk that important disputes are settled, or example, disputes which in other socie-
ties would have to be settled in the courts. Tus it is clearly not enough to merely
be adept at the grammar o the language; you also have to understand the social
appropriateness o di erent constructions. A ramework or the systematic study o
how talk is used in certain societies is presented in the next section.

SPEAKING

Hymes (1974) has proposed an ethnographic ramework which takes into account
the various actors that are involved in speaking. An ethnography o a communica-
tive event is a description o all the actors that are relevant in understanding how
that particular communicative event achieves its objectives. For convenience, Hymes
uses the word SPEAKING as an acronym or the various actors he deems to be
relevant. We will now consider these actors one by one (see also the link in our
companion website to a short video explaining this acronym).
Te setting and scene (S) o speech are important. Setting re ers to the time
and place, that is, the concrete physical circumstances in which speech takes place.
Scene re ers to the abstract psychological setting, or the cultural defnition o the
occasion. Te Queen o England’s Christmas message has its own unique setting
and scene, as has the President o the United States’ annual State o the Union
Address. A particular bit o speech may actually serve to defne a scene, whereas
another bit o speech may be deemed to be quite inappropriate in certain circum-
stances. Within a particular setting, o course, participants are ree to change scenes,
as they change the level o ormality (e.g., go rom serious to joy ul) or as they
change the kind o activity in which they are involved (e.g., begin to drink or to
recite poetry).
Te participants (P) include various combinations o speaker–listener, addressor–
addressee, or sender–receiver. Tey generally fll certain socially specifed roles. A
two-person conversation involves a speaker and hearer whose roles change; a ‘dress-
ing down’ involves a speaker and hearer with no role change; a political speech
involves an addressor and addressees (the audience); and a telephone message
involves a sender and a receiver. A prayer obviously makes a deity a participant. In
a classroom, a teacher’s question and a student’s response involve not just those two
as speaker and listener but also the rest o the class as audience, since they too are
expected to beneft rom the exchange.
Ends (E) re ers to the conventionally recognized and expected outcomes o an
exchange as well as to the personal goals that participants seek to accomplish on
particular occasions. A trial in a courtroom has a recognizable social end in view,
but the various participants, that is, the judge, jury, prosecution, de ense, accused,
and witnesses, have di erent personal goals. Likewise, a marriage ceremony serves
 Ethnographic Approaches in Sociolinguistics 233

a certain social end, but each o the various participants may have his or her own
unique goals in getting married or in seeing a particular couple married.
Act sequence (A) re ers to the actual orm and content o what is said: the precise
words used, how they are used, and the relationship o what is said to the actual
topic at hand. Tis is one aspect o speaking in which linguists have long shown an
interest, particularly those who study discourse and conversation, and it is one about
which we will have more to say in the next two chapters. Public lectures, casual
conversations, and cocktail party chatter are all di erent orms o speaking; with
each go di erent kinds o language and things talked about.
Key (K), the f h term, re ers to the tone, manner, or spirit in which a particular
message is conveyed: light-hearted, serious, precise, pedantic, mocking, sarcastic,
pompous, and so on. Te key may also be marked non-verbally by certain kinds o
behavior, gesture, posture, or even deportment. When there is a lack o ft between
what a person is actually saying and the key that the person is using, listeners are
likely to pay more attention to the key than to the actual content, or example, to
the burlesque o a ritual rather than to the ritual itsel .
Instrumentalities (I) re ers to the choice o channel, or example, oral, written,
signed, or telegraphic, and to the actual orms o speech employed, such as the
language, dialect, code, or register that is chosen. Formal, written, legal language is
one instrumentality; spoken New oundland English is another, as is American Sign
Language; code-switching between English and Italian in oronto is a third; and the
use o Pig Latin is still another. In Suriname a high government o cial addresses a
Bush Negro chie in Dutch and has his words translated into the local tribal lan-
guage. Te chie does the opposite. Each speaks this way although both could use
a common instrumentality, Sranan. You may employ di erent instrumentalities in
the course o a single verbal exchange o some length: frst read something, then tell
a dialect joke, then quote Shakespeare, then use an expression rom another lan-
guage, and so on. You also need not necessarily change topic to do any o these.
Norms o interaction and interpretation (N) re ers to the specifc behaviors and
properties that attach to speaking and also to how these may be viewed by someone
who does not share them (e.g., loudness, silence, gaze return, and so on). For
example, there are certain norms o interaction with regard to church services and
conversing with strangers. However, these norms vary rom social group to social
group, so the kind o behavior expected in congregations that practice ‘talking in
tongues’ or the group encouragement o a preacher in others would be deemed
abnormal and unacceptable in a ‘high’ Anglican setting, where the congregation is
expected to sit quietly unless it is their time to participate in group prayer or singing.
Likewise, a Brazilian and an Anglo-Saxon meeting or the frst time are unlikely to
fnd a conversational distance that each fnds com ortable, as they may have di er-
ent ideas about how close one stands when conversing with a stranger.
Genre (G), the fnal term, re ers to clearly demarcated types o utterance; such
things as poems, proverbs, riddles, sermons, prayers, lectures, and editorials. Tese
are all marked in specifc ways in contrast to casual speech. O course, in the middle
o a prayer, a casual aside would be marked too. While particular genres seem more
234 Language and Interaction

appropriate on certain occasions than on others, or example, sermons inserted into

church services, they can be independent: we can ask someone to stop ‘sermoniz-
ing’; that is, we can recognize a genre o sermons when an instance o it, or some-
thing closely resembling an instance, occurs outside its usual setting.
What Hymes o ers us in his SPEAKING ormula is a very necessary reminder
that talk is a complex activity, and that any particular bit o talk is actually a piece
o ‘skilled work.’ It is skilled in the sense that, i it is to be success ul, the speaker
must reveal a sensitivity to and awareness o each o the eight actors outlined above.
Speakers and listeners must also work to see that nothing goes wrong. When speak-
ing does go wrong, as it sometimes does, that going-wrong is o en clearly describ-
able in terms o some neglect o one or more o the actors. O course, individuals
vary in their ability to manage and exploit the total array o actors; everyone in a
society will not manage talk in the same way. Nonetheless, conversations can be
analyzed in terms o how they ft with social norms or interaction.

Exploration 9.2: De ning Gossip

How can you de ne the communicative event of gossiping? Use Hymes’

SPEAKING categories to discuss who participates in this type of communica-
tion with whom, the characteristic linguistic features, and the social goals.

Ethnography and beyond

In more recent studies, the description o underlying communicative competence

and actual language use are combined with critical perspectives and other orms o
discourse analysis. For example, Du (2002) looks at classroom interactions in a
multiethnic Canadian high school classroom through ethnography o communica-
tion research while also adopting critical and post-structuralist theoretical stances
in her analysis. She describes her work as ollows:

Tis study employed EC [ethnography o communication] to consider how students’

identities and interpersonal di erences are created and mani ested through interaction
patterns during classroom discussions. Unlike many past EC studies, I did not provide
an indepth structural analysis o the boundaries o the activities (beginning, middle
and end) or explicit instruction provided by the teacher about how to participate in
di erent phases o one activity, or explicit sanctions or non-compliance. Nor does the
analysis ocus on just one type o linguistic structure or raming device. Rather, I
combined content and interaction analyses o turn-taking in discussions as parallel
mani estations o how knowledge, identities, and di erences are established and main-
tained by members o a classroom ‘community.’ (Du 2002, 315)
 Ethnographic Approaches in Sociolinguistics 235

Te analysis, which also includes in ormation about student achievement and

attitudes stated in interviews with the researcher, provides a picture o the classroom
interaction which is much broader and more nuanced than a description o what
occurred in the interactions. For example, some ESL (English as a second language)
students, many o whom had Chinese parents or were themselves not born in
Canada, participated less in classroom discourse, but were nevertheless high achiev-
ing and in many cases per ormed better academically than their classmates who
were born in Canada and had English as their frst language. Te ‘locals’ were none-
theless the ones whose voices and perspectives were most o en heard in class dis-
cussion. Further, a simplistic analysis o turn-taking and the e orts o the teacher
to include di erent students in the discussion ell short o recognizing that despite
the good intent o this teacher, she had also contributed to the otherization o some
students by calling on them to comment on issues o discrimination and exclusion.
As Du (2002, 315) writes, ‘Everyday interactions such as these positioned students
within di erent communities – the very communities students may or may not have
wished to venture out o .’
Such an analysis thus draws both on ethnography o communication perspectives
and on other types o discourse and content analysis; we will continue to address
such issues in the last section o this chapter, on linguistic ethnography, and in the
sections on interactional sociolinguistics and critical discourse analysis in chapter
11. In the next section, we will look at what has been called ‘ethnomethodology,’
and one o its derivatives, conversation analysis.

Ethnomethodology

While it is possible to investigate talk, the various actors that enter into it, and the
variety o its unctions, and make many sound observations, this does not by any
means exhaust all we might want to say on the subject. As indicated at the beginning
o the chapter, talk itsel is also used to sustain reality and is itsel part o that reality.
We can there ore look at talk as a phenomenon in its own right. Ethnomethodology
is that branch o sociology which is concerned, among other things, with talk viewed
in this way. (See link in our companion website to an interactive overview o
ethnomethodology.)
Ethnomethodologists are interested in the processes and techniques that people
use to interpret the world around them and to interact with that world. Tey are
interested in trying to discover the categories and systems that people use in making
sense o the world. Tere ore, they do not conduct large-scale surveys o popula-
tions, devise sophisticated theoretical models o social organization, or hypothesize
that some social theory or other will adequately explain social organization. Instead,
they ocus on the phenomena o everyday existence, actually on various bits and
pieces o it, in an attempt to show how those who must deal with such bits and
pieces go about doing so. Teir methods are entirely inductive. As Leiter (1980, 5)
states, ‘the aim o ethnomethodology . . . is to study the processes o sense making