Most linguists who analyze discourse adopt, at least partially, a func- tional approach to language. This is not surprising: observing and analyzing what people dowith language leads naturally to an interest in the “work” that language can do – the functions it enables people to perform.
Structural approach Functional approach
Focuses on structure of language Focuses on structure of speech (as (a code) as a grammar. acts, events) as ways of speaking.
Analyzes language structure Analyzes language use before analysis before any (optional) analysis of of language structure. Assumes that language use. Assumes that language structure and use are language use derives from integrated; organization of language language structure. use reveals additional structural
features.
Assumes that the most important Assumes that language has a range of function of language is referential, functions, including referential, stylistic, i.e. the use of language to and social functions.
describe the world through propositions.
Studies the elements and Studies the elements and structures of structures of language separately language within their contexts of use;
from contexts of use; ignores the attends to the culture (ways of acting, culture (ways of acting, thinking, thinking, and being) of those using the and being) of those using the language.
language.
Assumes that language structure Assumes that languages, varieties and is independent of social functions styles can be adapted to different and uses. Any language can situations, functions, and uses, and gain (potentially) serve any social, different social values for their users.
cultural, or stylistic purpose.
Assumes that language is a single Assumes that language comprises a code within a homogeneous repertoire of speech styles within a community: each speaker diverse community: each speaker adds replicates a uniform structure. to an organized matrix of diversity.
Assumes the uniformity of Seeks to investigate the diversity of speakers, hearers, actions, events, speakers, hearers, actions, events, and and communities across world communities within world languages.
languages.
of people interacting with each other (by either speaking or writing) in everyday situations. They believe that the structure of discourse can be discovered not from peoples’ intuitions about what they might, could, or wouldsay, but primarily from analyses of what people do say. Discourse analysis focuses on the patterns in which sentences (and other units such as acts and turns) appear in the texts that are constructed as people inter- act with one another in social contexts.
Like other linguists, discourse analysts believe that the form of lan- guage is governed by abstract linguistic rules that are part of speakers’
competence. But added to linguistic rules are principles that guide per- formance, the useof language. Knowledge about discourse is part of what Hymes (1974) has called communicative competence– our tacit cultural knowledge about how to use language in different speech situations, how to interact with different people engaged together in different speech events, and how to use language to perform different acts.
We can see how our knowledge about language intersects with our knowledge about social and cultural life by taking a look at the discourse examples below, drawn from a collection of routine speech events:
(1) Gail: Hello?
Debby: Oh hi, you’re home already!
We can infer a great deal about what is going on in this brief exchange from features of the interchange. Gail and Debby seem to be talking on the phone; hellowith rising intonation (represented by the ?) is typical of the way Americans answer the phone. Notice that Debby doesn’t ask for Gail (for example, Hi, is Gail there?), nor does she identify herself (Hi, this is Debby).
What Debby doesn’t say, then, allows us to infer that Debby and Gail know each other pretty well; Debby recognizes Gail’s voice and seems to assume that Gail will recognize hers. We can also infer (from the exclamation oh) that Gail’s presence is surprising to Debby. And the statement you’re home already, shows that Debby knows something about Gail’s intended schedule.
Snippets of discourse from our daily lives show us how much we – as
“after-the-fact” analysts and as real-time language users – can infer about
“what is going on” from routine uses of language. Likewise, gathering examples of routine speech acts (such as requests, compliments, apolo- gies) and speech events (such as face-to-face greetings or telephone open- ings) can reveal both their similarities and their differences. However, although collecting examples of discourse from our own everyday lives is valuable, it has several limitations. First, when we hear an interesting bit of discourse and then jot it down, we usually cannot capture stretches of discourse that are longer than a few sentences or turns at talk. (Aturn at talkis the period of time in which someone is granted and/or takes the opportunity to speak.) Second, it is difficult to reconstruct the nuances of speech, particularly when several people are interacting with each other.
These nuances are especially helpful when we try to figure out how it is that interlocutors (people talking to one another) interpret what is going on; crucial information can reside in a pause, a sigh, a downward intona- tion, a simple ohor well, or the arrangement of words in a sentence (for example, I want the cakeversus The cake is what I want). It is impossible to
The ethnography of communication Dell Hymes (a lin- guistic anthropolo- gist) developed a subfield of linguistics and anthropology called “ethnography of communication.”
He persuaded lin- guists and anthro- pologists to analyze the social, cultural, and linguistic prop- erties of three units embedded in one another:
Speech act: an action performed by one person through speech. It can be labeled by a noun that names the act.
The speaker intends to perform the act and that intention is recognized by the recipient. Examples:
a greeting, a request, a boast, a compliment.
Speech event: an interaction between two or more people in which more than one speech act occurs. Examples:
greetings, request, and compliance.
Speech situation:
a social occasion with more than one speech event. During the occasion, speech contributes to what happens, but it is not necessarily all that happens.
Examples: a class- room, a party.
recall all of the speech that appears throughout the course of a speech event (the type of interaction that participants assume is going on) or the speech situation (the type of occasion or encounter). And because our memories are fallible, we usually fill in details based on our prior knowl- edge of what typically happens.
Discourse analysts correct for the limitations of relying only upon what they hear in their everyday lives in several ways. The way that they do so depends partially upon the topic they are studying and partially upon their interest in generalizing their findings. For example, discourse ana- lysts who are interested in how groups of people use discourse to com- municate at work often do fieldwork in a workplace. There they observe activities (e.g. meetings, chats at the water cooler) and interview people who perform different tasks (e.g. managers, secretaries). They can then propose generalizations about that workplace and perhaps about other workplaces with similar characteristics.
Other discourse analysts may be interested in a particular aspect of dis- course: how do people apologize to one another? When, where, and why do people use the word like (as inI’m like, “Oh no!” orIt was like a crazy thing, like weird)? Then they may rely upon tape-recorded speech from a wide variety of settings and occasions, paying less attention to obtaining a sample that represents a subset of people and their activities in a particular social set- ting, and more attention to getting enough examples of the discourse phe- nomena in which they are interested. Still other discourse analysts might be interested not in the discourse of a particular setting, or one aspect of discourse, but in every aspect of only onediscourse. They might delve into all the details of several minutes of a single conversation, aiming to under- stand how it is that two people use many different facets of language to construct a discourse that makes sense to them at that time.
Regardless of their type of inquiry, most discourse analysts rely upon audio or video-recordings of interactions between people in which speech is the main medium of communication. Once speech has been recorded, analysts have to then produce a transcript– a written version of what was said that captures numerous aspects of language use, ranging from fea- tures of speech (such as intonation, volume, and nonfluencies) to aspects of interaction (such as overlaps between turns at talk) and, if possible, aspects of nonvocal behavior (such as gaze and gesture). Transcriptions of spoken discourse look quite different than other scripts with which we might be familiar. For example, unlike most scripts for dramatic produc- tions, linguists’ transcripts try to indicate features of speech production and interaction, often using notations like those in Box 5.2 on transcrip- tion conventions.
Transcribing spoken discourse is challenging and often frustrating, not to mention time-consuming: some linguists spend close to ten hours tran- scribing just one hour of speech. But fortunately, what results is a tran- script that they can analyze from different angles years after the original speech. The process of transcribing is also very instructive! By listening – again and again – and trying to fine tune one’s written record of what is said, linguists often end up doing preliminary analyses.
Box 5.2 Transcription conventions (adapted from Schiffrin 1987; Tannen 1989)
Discourse analysts use a variety of symbols to represent aspects of speech, including the following:
. sentence-final falling intonation clause-final intonation (“more to come”)
! exclamatory intonation
? final rise, as in a yes/no question
… pause of 1>2second or more
´ primary stress CAPS emphatic stress [ overlapping speech.
] no perceptible inter-turn pause : elongated vowel sound
- glottal stop: sound abruptly cut off
“ ” dialogue, quoted words
( ) “parenthetical” intonation: lower amplitude and pitch plus flattened intonation contour
hhh laughter (h one second)
at right of line indicates segment to be continued after another’s turn; at left of line indicates continuation of prior segment after another’s turn
/?/ inaudible utterance
{ } transcriber comment on what is said
You can see many of these symbols in the excerpts of discourse throughout this chapter.
We create discourse by speaking or writing. These two processes rely upon language, of course, but they do so in strikingly different ways. And not surprisingly, their products achieve coherence through very different means. We can see this briefly by comparing the following excerpts from a story that Gina tells her friend Sue (in 2a) and writes(in 2b). In both excerpts, Gina introduces a story about how her love for magnolia blos- soms got her into trouble when she tried to smell a blossom that then snapped off in her hand.
(2) a. Gina: Have you ever smelled a magnolia blossom?
Sue: Mmhmm.
Gina: Absolutely gorgeous.
Sue: Yeah, they’re great.
b. On one particular morning this summer, there was a certain fragrance that I recognized to be a glorious magnolia.