The study of narrative extends over a broad range of human activities: novels,
short stories, poetic and prose epic, film, folk tale, interviews, oral memoirs, chronicles,
histories, comic strips, graphic novels and other visual media. These forms of
communication may draw upon the fundamental human capacity to transfer experience
from one person to another through oral narratives of personal experience.
A focus on spontaneous recounting of experience was greatly stimulated by the
development of sociolinguistic research in the 1960s, designed to capture the closest
approximation to the vernacular of unmonitored speech. Narratives of personal
experience were found to reduce the effects of observation to a minimum (Labov 2001).
Since then it has appeared that such narratives are delivered with a similar organization in
a wide variety of societies and cultures as for example, in the Portuguese of fishermen in
northeastern Brazil (Maranhão 1984). The discussion of oral narratives to follow is based
on the initial analysis of Labov and Waletzky 1967, as developed further in the suggested
reading.
The discussion to follow will first treat the structural organization of narrative
(temporal organization, orientation, coda), then turn to the uative component, and
finally to the construction of narrative as a folk theory of causality instrumental to the
assignment of praise and blame.
Structural organization
A narrative is defined here as one way of recounting past events, in which the
order of narrative clauses matches the order of events as they occurred. (1) is a minimal
narrative organized in this way.
(1) a Well, this man had a little too much to drink
b and he attacked me
c and a friend came in
d and she stopped it.
The same events could have been reported in the non-narrative order c,d,a,b, as in (2),
which employs a variety of grammatical devices within a single clause.
(2) A friend of mine came in just in time to stop this person who had had a little
too much to drink from attacking me.
Narrative structure is established by the existence of temporal juncture between
two independent clauses. Temporal juncture is said to exist between two such clauses
when a change in the order of the clauses produces a change in the interpretation of the
order of the referenced events in past time. These are narrative clauses. Narrative clauses
respond to a potential question, “what happened then?” and form the complicating action
of the narrative.
A narrative normally begins with an Orientation, introducing and identifying the
participants in the action: the time, the place, and the initial behavior. The orientation
section provides answers to the potential questions, “who? when? where? what were they
doing?” In the minimal narrative (1), the first clause (a) is the orientation. More
information is usually provided.
(3) a my son has a--well, it was a fairly new one then.
b It's a 60 cc Yamaha.
c and it could move pretty good.
d This fella and I were going down the road together
The end of a narrative is frequently signaled by a Coda, a statement that returns
the temporal setting to the present, precluding the question, “and what happened then?”.
(4) a And you know the man who picked me out of the water?
b He’s a detective in Union City,
c and I see him every now and again.
Evaluation.
Most adult narratives are more than a simple reporting of events. A variety of
uative devices are used to establish the uative point of the story (Polanyi 1989),
Thus we find that narratives, which are basically an account of events that happened,
frequently contain irrealis clauses—negatives, conditionals, futures,—which refer to
events that did not happen or might have happened or had not yet happened.
(5) And the doctor just says, “Just that much more,” he says, “and you’d a
been dead.”
(6) I’ll tell you if I had ever walloped that dog I’d have felt some bad.
(7) a And he didn’t come back.
b And he didn’t come back.
Irrealis clauses serve to uate the events that actually did occur in the narrative by
comparing them with an alternate stream of reality: potential events or outcomes that
were not in fact realized. Frequently such uative clauses are concentrated in an
uation section, suspending the action before a critical event, and establishing that
event as the point of the narrative.
Evaluative clauses vary along a dimension of objectivity. At one extreme, narrators
may interrupt the narrative subjectively by describing how they felt at the time.
(8) a I couldn’t handle any of it
b I was hysterical for about an hour and a half
In a more objective direction, narrators may quote themselves (“I said to myself, ‘This is
it’”), or with more credibility, cite a third party witness, as in (5). At the other extreme,
objective events speak for themselves, as in the account of a plane developing motor
trouble over Mexico City.
(9) And you could hear the prayer beads going in the back of the plane.
Evaluation provides justification for the narrative’s claim on a greater portion of
conversational time than most turns of talk, requiring an extended return of speakership
to the narrator until it is finished (Sacks 1989). Evaluation thus provides a response to the
potential question, “So what?” (Spanish “Y que?”, French “Et alors?”).

Narratives of personal experience normally show great variation in the length of
time covered by the clauses in orientation, complicating action and uation sections,
ranging from decades to minutes to seconds. Sequences of clauses of equal duration
may be termed chronicles; these are not designed to report and uate personal
experience.
Reportability and credibility.
A reportable event is one that itself justifies the delivery of the narrative and the
claim on social attention needed to deliver it. Some events are more reportable than
others. The concept of reportability or tellability (Norrick 2005) is relative to the
situation and the relations of the narrator with the audience. At one end of the scale, death
and the danger of death are highly reportable in almost every situation. At the other end,
the fact that a person ate a banana for lunch might be reportable only in the most relaxed
family setting. Most narratives are focused on a most reportable event. Yet reporting this
event alone does not make a narrative: it only forms the abstract of a narrative.
For a narrative to be successful, it cannot report only the most reportable event. It
must also be credible if the narrative is not to be rejected as a whole by the listener. There
is an inverse relationship between reportability and credibility: the more reportable, the
less credible. Narrators have available many resources to enhance credibility. In general,
the more objective the uation, the more credible the event.
Narrative preconstruction.
When a narrator has made the decision to tell a narrative, he or she must solve the
fundamental and universal problem: “Where should I begin?” The most reportable event,
which will be designated henceforth as e0, is most salient, but one cannot begin with it.
Given the marked reportability of e0 and the need to establish its credibility, the narrator
must answer the question, “How did this (remarkable) event come about?” The answer
requires a shift of focus backwards in time to a precursor event e-1, which is linked to e0
in the causal network in which events are represented in memory (Trabasso and van den
Broek 1985). In traversing this network in reverse, the causal links found may be eventto-
goal, goal-to-attempt, or attempt-to-outcome. The process will continue recursively to
e-2, e-3, etc. until an ordinary, mundane event e-n is reached, for which the question “why
did you do that?” is absurd, since en is exactly what we would expect the person to do in
the situation described. The event en is of course the Orientation. Thus a narrator telling
of a time he was on shore leave in Buenos Aires begins,
(10) a Oh, I was settin’ at a table drinkin’.
Triggering events. Given the mundane and non-reportable character of the
orientation, it follows that the first link in the causal chain is a triggering event, which
drives the narrative along the chain towards the most reportable event. Thus (10) is
followed by (11).
(11) b an’ this Norwegian sailor come over
c an’ kep’ givin’ me a bunch o’ junk about how I was sittin’ with his
woman.
How ordinary situations like (10) can give rise to the reportable and violent events
that followed is a mystery that narrative analysis can only contemplate, since they are
part and parcel of the contingent character of history.
The transformation of experience.
The participants in many narratives include protagonist, antagonist and third party
witnesses, of which the first is the most complex. Elaborating on Goffman (1981:144-5),
one can identify many egos present: the self as original author of the narrative and its
immediate animator; the self as actor; the self as generalized other (normally as “you”);
the anti-self as seen by others; and the principal, the self in whose interest the story is
told. That interest is normally advanced through a variety of techniques which do not
require any alteration in the truthfulness of the events reported. The re-creation of the
causal network involves the assignment of praise and blame for the critical events and
their outcomes. Most narratives of conflict involve linguistic devices that contribute to
the polarization of protagonist and antagonist, although within the family, other linguistic
forms lead to the integration of participants. The devices used to adjust praise and blame
include most prominently the deletion of events, an operation which can often be detected
by close reading. Key elements in further manipulation are the grammatical features of
voice: active vs. passive, but also zero causatives which assign agency (“He drove
through town with a chauffeur”) or verbs which imply the exertion of authority and
resistance to it (“My dad let me go with him”). Other narrative devices function to
increase the impression of agency: pseudo-events that may not correspond to any
physical event (“I turned to him and”, “I took this girl and”, “I started to hit him but”).
Narrative analysis can show how the PRIMA FACIE case is built to further the
interests of the principal. This involves detecting insertions of pseudo-events and
removing them; detecting deletions and replacing them; and exchanging excuses for the
action excused. It is then possible to approximate the original chain of events on which
the narrative is based. A useful exercise is to develop a complementary SUB-ROSA case in
the interests of the antagonist. The comparison of these two constructions deepens our
understanding of how narrative skills are enlisted to transform the social meaning of
events without violating our commitment to a faithful rendering of the past.