-
This article was downloaded by: [McGill University Library]On:
02 April 2015, At: 13:55Publisher: RoutledgeInforma Ltd Registered
in England and Wales Registered Number:1072954 Registered office:
Mortimer House, 37-41 Mortimer Street,London W1T 3JH, UK
Discourse ProcessesPublication details, including instructions
forauthors and subscription
information:http://www.tandfonline.com/loi/hdsp20
The voices within narratives:The development ofintertextuality
in youngchildren's storiesDennie Wolf a & Deborah Hicks ba
Project Zero , Harvard Graduate School ofEducation , Longfellow
Hall, 13 Appian Way,Cambridge, MA, 02138b Harvard Graduate School
of EducationPublished online: 11 Nov 2009.
To cite this article: Dennie Wolf & Deborah Hicks (1989) The
voices withinnarratives: The development of intertextuality in
young children's stories,Discourse Processes, 12:3, 329-351, DOI:
10.1080/01638538909544734
To link to this article:
http://dx.doi.org/10.1080/01638538909544734
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy
of allthe information (the Content) contained in the publications
on ourplatform. However, Taylor & Francis, our agents, and our
licensorsmake no representations or warranties whatsoever as to the
accuracy,completeness, or suitability for any purpose of the
Content. Any opinionsand views expressed in this publication are
the opinions and views ofthe authors, and are not the views of or
endorsed by Taylor & Francis.The accuracy of the Content should
not be relied upon and should beindependently verified with primary
sources of information. Taylor andFrancis shall not be liable for
any losses, actions, claims, proceedings,demands, costs, expenses,
damages, and other liabilities whatsoever
-
or howsoever caused arising directly or indirectly in connection
with, inrelation to or arising out of the use of the Content.
This article may be used for research, teaching, and private
studypurposes. Any substantial or systematic reproduction,
redistribution,reselling, loan, sub-licensing, systematic supply,
or distribution in anyform to anyone is expressly forbidden. Terms
& Conditions of accessand use can be found at
http://www.tandfonline.com/page/terms-and-conditions
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
DISCOURSE PROCESSES 12, 329-351 (1989)
The Voices within Narratives: TheDevelopment of Intertextuality
in
Young Children's StoriesDENNIE WOLF
DEBORAH HICKSHarvard Graduate School of Education
In recent years we've come to understand how fully even young
speakers acquire morethan grammar, learning, in addition, how
language is used variously across differentregisters and genres.
However, young speakers are learning more than the rules of
babytalk or story telling. They are discovering that naturally
occurring speech is a rich mix ofvoices and forms, where the moves
between perspectives and kinds of text conveymeaning as certainly
as the words do. In the following paper, we use longitudinal
observa-tions of children's narratives to describe how this
occurs.
INTRODUCTION: INTERTEXTUALITY
In the study of discourse perhaps no concept is more familiarand
still in-completely understoodthan that of variation. There is the
variation we recog-nize as the signature of different communities
of speakers and writers. This is thekind of variation that makes
some speakers address a listener as an intimate,while others assume
he is a stranger, or which leads the members of one commu-nity to
expect silence from its children while, not far away, other parents
primetheir children in the art of storytelling. Even to describe
the discourse of languageusers from the same community, we need the
notion of variation. Without it, wecould not capture the subtle
ways in which speakers and writers tune theirlanguage to changes in
audience or task. We vary the phonological contours ofour speech
depending on how formal or informal we perceive our circumstancesto
be (Labov, 1972a; Wolfram & Fasold, 1974). We vary the
complexity of whatwe say depending on whether we are talking to a
baby, a child, or an adult. Oursensitivity to genre allows us to
reformat an account of "what happened lastnight" depending on
whether we give an eyewitness report to Sargent Smith or aslightly
embroidered first person narrative to Jake from downstairs
(Bruner,1986; Fowler, 1982; Langer, 1986).
This paper is the result of equal work between both authors.The
research reported in this paper was supported by funding from the
Carnegie Corporation of
New York.Correspondence and requests for reprints should be sent
to Dennie Wolf, Project Zero, HGSE,
Longfellow Hall, 13 Appian Way, Cambridge, MA 02138.
329
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
330 WOLF AND HICKS
Yet there is still another sort of variation in language use of
which we seem tobe less aware: This is a kind of variation within
the performances of speakers andwriters which we will call
intertextuality. When we look closely at sophisticatedtexts,
whether they come out of novels or someone's vivid oral rendition
ofexperience, what we find is that such texts are inevitably
compounds of a numberof different strands or voices. Consider this
segment taken from Henry James'novel, The Awkward Age:
. . . she went on before he had spoken. "I know how well you
knew my grand-mother. Mother has told meand I'm so glad. She told
me to say to you that shewants you to tell me." Just a shade, at
this, might over the old man's face, haveappeared to drop, but who
was there to detect whether the girl observed it? It didn'tprevent,
at any rate, her completing her statement. "That's why, today, she
wishedme to come alone. She wished you to have me, she said, all to
yourself." (James,1984, p. 98)
We would miss something vital about narrative in general, and
this particularnarrative specifically, if we were to describe the
"workings" of this episodesimply in terms of its cohesion or how it
subscribes to our expectations for theway that stories unfoldeither
in terms of sequentiality or suspense. What wewould miss is the
fact that James has given us many different lines of text,
orvoices, which interwoven, lend the narrative its texture and
dimensionality,never mind its essential ambiguity. While much of
the telling is carried outthrough the direct speech of characters,
other information we have as indirect orreported speech ("She
wished you to have me, she said, all to yourself.").These
quotations are complemented by what occurs in the narrative
voice(". . . she went on before he had spoken.") where James
reports what might beobserved by any onlooker. Then, from a stance
somewhere at the rim of thefiction, James speaks, not as
transparent narrator, but as author, proffering a lineof
commentary"but who was there to detect whether the girl observed
it?".
The workings of these voices within the text are complex: First
there is theinterplay of several basic stances or attitudes. We can
follow this interplaybecause each voice has its own linguistic
signature which we hear as signalling adistinctive kind of
reliability and insight. The involved or empathetic presenta-tion
of character through dialogue is marked, not just by quotation
marks, but bythe present tense, the I's and you's of conversation,
and by the presence ofpredicates such as "know" and "am glad" with
which characters make theirinner lives and visions public. By
contrast, the narrator's voice, speaking fromoutside the action, is
marked by its own characteristic pattern of past tense,
thirdperson, and predicates denoting observed actions ("she went on
before he hadspoken"). We are alert to the voice of the author
challenging us as readers whenJames crosscuts the recognizable
narrator's voice with his own peculiar set ofmodals and direct
questions ("Just a shade at this, might over the old man's
face,have appeared to drop, but who was there to detect whether the
girl observed
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 331
it?"). Thus, through recognizable patterns of linguistic cues,
and kinds of infor-mation, James reveals the world, or the event,
as seen, at that moment, byseveral intelligences. James' text is
not just a single story unfolding. It is whatGeertz might call "the
interplay of a disorderly crowd of not wholly commen-surable
visions" (1983, p. 161). The voices are far from iterative. Each is
offsetsome from the other, creating a complex murmur of
possibilities.
Second, there is the plurifunctionality of each voice. Each one
can be used toportray speech, to describe events, to offer
commentarywith subtle, but tellingresults, for instance, when Nanda
says "She told me to say to you that she wantsyou to tell me."
James makes dialogue do the usual work of narrationrecallingpast
events. But, by delivering her mother's words through Nanda's
(rather thanthe original speaker's or a narrator's) mouth, James
shows usimplicitly, butpowerfully, how ingenuously the young woman
sees what others behold as hermother's clever social manipulations.
Thus, through the variety of voices andtheir functions, James
creates a story that is more than a monolithic telling, it is
anetwork of texts within texts.
But is James an unfair or extraordinary example? Possibly this
kind of jug-gling and weaving together of voices is the result of
what is often argued to be thecomplexity made possible by writing
and a literate tradition (Olson, 1979; Tan-nen, 1982) or James'
particular absorption in and skill at portraying psychologi-cal
complexity. However, the following excerpt, taken from the
spontaneousplay of a 3-year-old suggests that even oral and
relatively simple narratives arepotentially, or even typically,
multivoiced:
Heather (3 years, 5 months)narrative dialogue stage-managing
(H. plays with a king, queen, and a princess doll, walking each
along a table.)once upon a timethe baby and the mommy and the
daddythey walked through the forest to find a houseand said
there's a porch
(H. puts the king in the porch, but has trouble fitting the
queen in) and then thebaby said
there's not room enough
(H. looks to the adult)there's not enoughroom in this house/can
you make the porchbigger/people won't fitin the porch
(The adult enlarges the porch with blocks. "Tell me some more of
the story.")
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
332 WOLF AND HICKS
then they found a porch!(H. places the figures in the porch one
by one.)She found a porchand then he found a porch(high voice) and
then theywalked through the forest
(H. walks the figures around, then holds up the baby)hey,
where's my mommy and daddy?
and then the mommy goes
hihi, mommyI want to go for a walk
says the baby
(H. makes sound effects as she walks the baby figure. She looks
to the adult.)that's their house/and that's theirporch
At 3 years, 5 months, Heather may not have the subtlety or
plurifunctionalityof James; nevertheless, she juggles at least
three distinct voices in her playnarrative. She uses a narrative
voice to relate the main-line events (Polanyi,1986). Using
character dialogue, she conveys what characters say aloud.
Step-ping outside the story world, Heather engages in
stage-managing whenever shewants to negotiate about the literal
conduct of her narrative with her audience.Each of these voices is
marked by its own distinctive constellation of linguisticfeatures,
carries certain kinds of information, and suggests a particular
stance onthe unfolding events. For instance, Heather signals her
use of the narrative voicewith the past tense, the third person, a
predominance of declaratives, and con-nectives. In these narrative
utterances, she depicts the main events of the nar-rative, as if an
onlooker or spectator. She is equally clear in signalling
hermovement to dialogue or stage-managing and back to
narrative.
Both Heather and Henry James tell narratives across several
different voices,yet neither story is incoherent. This suggests
that the achievement of unified orcohesive texts may be more
complicated than is sometimes suggested by textgrammars (de
Beaugrand & Dressier, 1980; Halliday & Hasan, 1976) or
bydescriptions of the schemes that underlie different types of
texts such as stories(Kintsch & Greene, 1978; Mandler &
Johnson, 1978; Peterson & McCabe,1983; Stein & Glenn, 1979)
or reports (Eisenberg, 1985; Langer, 1986). Whatthe research on
cohesion acknowledges is the way new information echoes oramplifies
earlier occurring portions of the text. The work on underlying
schemasemphasizes how the different portions of a text make sense
of fulfilling ourexpectations for orientations, high points, or
denouements. But in both cases, the
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 333
accent falls on how a speaker binds textual elements into a
whole. But even theshort samples from Heather or The Awkward Age
point out that a text can be ajuggling act in which the teller or
writer orchestrates a choir of voices, whichonly as an ensemble
convey the texture of information and the author's fluctuat-ing
stance towards the information being shared. What we need, as a
comple-ment to models for coherence or schema, is something much
closer to whatLabov (1972b), Polanyi (1986), Reisman (1987), and
Schegloff (1971) describe:a portrait of a speaker who builds a
coherent narrative by interweaving manyfunctions and attitudes as
they are carried by a number of voices. This is what weterm
intertextuality, what writers like Kundera (1987) call "polyphony"
ortheorists like Bakhtin (1981; Todorov, 1981) call
"dialoguism."
Heather carries this point still further. Her play narrative
insists that thecapacity for intertextuality is not unique to
adulthood, writing, or literature. Herjuggling of narrative,
dialogue, and stage-managing suggests that the under-standing of
texts as multivoiced emerges early enough to be considered a
funda-mental aspect of what young speakers learn about discourse.
Together the novel-ist and the child question our conception of
text as linear or simply sequential.They hint that, even in
streamlining our experience into texts, we preserve aselemental the
social aspect of language, both the interplay of inner
perspectivesand overheard voices.
In what follows, we will examine young children's play
narratives in order tofind out what stories told by 2- to
7-year-olds can teach us about the nature andthe development of the
capacity for intertextuality. We will look chiefly at twoquestions.
First, we ask a developmental question: How do young speakers froma
particular language communitymiddle-class, literate familieslearn
tomark and juggle several kinds of text, or voices, within a single
performance? Todo this, we take a longitudinal look at the ways in
which children distinguish anduse several voices in play narratives
performed between the ages of 2 and 6.Second, we ask what appears
to be universal and what seems to be variable in theway that young
speakers from different language communities select, mark, anduse
different voices within their narrative texts. Here we compare play
narrativesof two 7-year-old girls. The first child comes from a
white middle-class familywhere even spoken narratives are modeled
on what are often called literate,autonomous, or written texts
(Olson, 1979; Tannen, 1982). The second childcomes from a black
working-class family where spoken narratives typicallymake rich use
of the options of oral performance: intonation, stress, dramatic
useof different patterns of character speech, gesture, and facial
expression (Gee,1989; Heath, 1983; Michaels, 1981).
BACKGROUND ON THE STUDIESThe narratives that figure in the
remainder of the paper come from two longitudi-nal studies of
children's language development. In both of these studies,
wefollowed how children formatted narratives in different contexts
such as oral
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
334 WOLF AND HICKS
tellings, dictations, recitations based on picture books, and
episodes of symbolicplay with small replica-sized figures. In their
replica-play narratives, childrenplayed out sequences of events
using small toys as actors and turning floors orkitchen tables into
theatres. They created landscapes out of available props andmade
the figures run, hide, or sleep, all the while providing a running
commen-tary and sound effects. These narratives, on which we will
focus, are of twokinds: spontaneous play narratives in which
children invent the story-line as theygo and reenactments in which
children use the figures to play out events orthemes taken from a
book, television, or movie. In the first study we collectedplay
narratives from nine middle-class, Caucasian, English-speaking
childrenfrom families who encouraged their children to "speak text"
from early on(Wolf & Pusch, 1985). We followed these children
longitudinally, visiting themin their homes weekly between the ages
of 1 and 3, bimonthly between the agesof 3 and 7 (Shotwell, Wolf,
& Gardner, 1980; Wolf & Grollman, 1982). In asecond study,
we followed a group of 50 children from their kindergartenthrough
second grade years, looking at the transition to literacy (Wolf et
al.,1988). The children in this study included working-class white,
Hispanic, andblack children, along with children similar to those
in the first study. All thesechildren were seen individually, but
in the context of school.
In all cases, children's narratives were audio-recorded, with
extensive fieldnotes being made by an observer. These notes
included information about gaze,the figures or props being used,
the actions performed with those props, and theway in which the
child used his or her face and voice along with speech.
CODING THE DATA
The data from the replica play sessions was coded in three ways:
for clause units,voices, and the linguistic features characterizing
the clauses occurring withineach voice. Drawing on Berman and
Slobin (1984), we defined a clause as anyunit containing a
predicate. Clause units may be syntactically independent, orthey
may be subordinate or relative clause constructions. Thus, each of
thefollowing examples represents a single clause unit:
he went to the cavehe sees the Mommy Bear'cause she was sad
In the replica play narratives which constitute our sample,
subordinate and rela-tive clause constructions are quite rare. This
is perhaps due to the nature of thetask: Particularly, young
children tend to tell their replica play narratives in akind of
to-the-point, "sportscasting" diction which allows them to provide
on-line descriptions of characters' actions as they are performed.
Thus, they are lesslikely to recall events or to present
information in subordinate constructions.
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 335
Nevertheless, the segmentation of narrative clause into units
based upon verbphrase constructions is justified on theoretical
grounds. Namely, such segmenta-tion enables us to examine whether
children adopt and sustain characteristicpatterns of linguistic
features, including the temporal system, to mark
functionaldistinctions between voices.
Once children's narratives were broken into clauses, each of
these units wascoded for the voice in which it occurred. We defined
voice as a stance taken bythe speaker relative to the unfolding
narrative events. On the basis of earliercross-sectional studies
(Wolf, Goldfield, & Beeghly, 1984), we assigned indi-vidual
clauses to one of three of the most prominent voices we discerned
inchildren's play narratives: stage-managing, character dialogue,
or narrative. Wemade these assignments on the basis of pragmatic
information and the child's useof pronounsboth of which signal the
perspective the child is taking on theevents in the story
world.
When children focused their attention away from the table or
play area (thescene of the story), fixing it instead on the
experimenter or objects in theirsurroundings, and marked their
utterances with the "I-you" forms characteristicof joint
conversation, we coded these clauses as occurring in the
stage-managingvoice. In these cases, we saw children as taking up
the stance of a collaboratorinvolved in joint negotiations about
the actual performance or interpretation ofnarrative events or the
nature and use of the props.
We considered children's utterances as dialogue when children
focused theirattention on play figures and simultaneously provided
speech encoded primarilyin the first person and pronominal forms
characteristic of direct speech by aparticipant in an ongoing
event. These utterances were also often speciallymarked with
distinctive pitches and dynamics. Finally, utterances were scored
asnarrative when children focused their attention on characters,
using the thirdperson and depicting story events as an observer.
While this voice was typicallyperformed at normal pitch levels, it
sometimes carried the deliberate rhythms ofstory-telling or a
heightened sing-song tone.
Thus, the segmentation of clauses into stage-managing,
narrative, or dialoguestrands was the result of both linguistic
distinctions (prosodic and pronominal) aswell as visual cues
available to the experimenter and transcriber. In this
way,utterances which would be otherwise ambiguous could be
distinguished. Anutterance, for example, such as: "this is the
daddy bear" was scored as anarrative clause if the child's
attention was focused upon the toy bear. If, on theother hand, the
child raised up the toy bear for the experimenter to see
andannounced, "this is the daddy bear," the clause was scored as
stage-managing.
The linguistic features we examined consisted of the different
linguistic sub-systems which speakers and writers have available to
them to distinguish amongthe voices occurring within their texts.
This type of coding builds on recentresearch in child language and
discourse analysis which has shown that func-tionally distinct
types of narrative are often distinguished through the use of
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
336 WOLF AND HICKS
particular constellations of linguistic features rather than the
exclusive use ofindividual features. Thus, when telling
highly-scripted narratives, children typ-ically mark their accounts
by use of the present tense, second person pro-nominals, and
sequencing connectives used in a generalized sense (Nelson,1986).
By contrast, when engaged in story-telling, children typically
markedtheir utterances with patterns of first or third person
pronouns, predominant useof the past tense, and more specific
sequential connectives (Applebee, 1978;Eisenberg, 1985; Peterson
& McCabe, 1983). When children provide runningcommentary on
ongoing events they may also make use of first or third
personpronominals and sequential connectives, but in those "on-line
replays" childrencombine these features with distinctive patterns
of temporal features, such as thepredominant use of the English
progressive form (Gee, 1986).
Based on this work, we coded each clause within a child's play
narrative forthe total set of linguistic features shown in Table 1
below.
The first set of features is the use of various types of
pronominal forms. Thesepatterns of .pronominal use were used in
conjunction with pragmatic informationto assign clauses to
particular voices. Strands which are recounted in the
stage-managing voice or in a dialogue voice for the most part
contain first and secondperson pronominals, since the characters
are involved in face-to-face interac-tions. Strands which are
recounted in the narrative voice, however, generally
TABLE 1Coding Features for Replica Play Narratives
A. Referential System1) first person pronominals (I, we)2)
second person pronominals (you)3) third person pronominals (she,
he, it, they, someone, everyone)4) nominals (the baby bear, Hippo,
the pond)
B. Utterance Form1) declarative (he went to the forest)2)
imperative (come to my cave)3) interrogative (where are you
going?)
C. Temporal System1) verb inflections
present (-s) and past (-d, -ed) tenseEnglish progressive
(-ing)
2) verb semantic typephysical state (have, be, seem)internal
state (think, feel, wonder)process (play, swim, learn)event (take,
get, scare)
3) connectivessequencers (and, and-then)temporal connectives
(after, before, while)causal connectives (because)
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 337
contain third person pronominals since events are narrated from
the outsideperspective of the narrator. Since these pronominal
features were used to codethe voice of each clause, what is at
issue is the way in which children usepronouns in conjunction with
additional kinds of linguistic features to mark theirmovement
between several types of text within the same play narrative.
The second set of linguistic features we examined was children's
differentialuse of various types of utterance forms. One can
predict a priori that utterances inthe narrative voice will be
predominantly declarative in nature and that dialogueutterances
will contain more of a mix of utterance types. Dialogue and
stage-managing utterances reflect more of an interactive stance, so
that questions andcommands may be used as well as declarative
utterances. As was the case withthe pronominal system, we pursued
an examination of the differential use ofutterance forms as a part
of asking how various linguistic subsystems are used inconjunction
with one another.
The third and final set of linguistic features we coded included
children's useof the temporal system to encode various facets of
the narrative timeline. Welooked at their use of verb inflections,
verb semantic types (aktionsart), andsequential connectives across
each of the voices in order to capture the possibledifferential use
of these temporal forms within one narrative text. The use of
verbinflections, such as past versus present tense, is a
self-evident temporal dimen-sion, involving simply differential use
of one or the other verb inflection. Theuse of the English
progressive form is an aspectual distinction entailing durationor
the ongoing status of events. An examination of the differential
use of verbsemantic types within separate voices is based upon the
distinction noted inParsons (1986) and also Vendler (1967) between
verbs which encode events andthose which encode states and
processes. Event verbs (take, get, come) entailactions having a
punctual endpoint or terminus, whereas process verbs (play,swim,
run) entail activities with no specific endpoint. Physical state
(be, have,seem) and internal state (like, feel, want) verbs entail
actions which, by defini-tion, extend over an interval of time.
Differential use of verb semantic types mayreflect functional
distinctions in the types of events encoded in each voice.Within
the narrative voice, for example, one might expect to find a large
propor-tion of event verbs since narrative mainline clauses are
most typically punctate innature (Labov, 1972b; Polanyi, 1986). In
dialogue one might expect to find agreater proportion of verbs
representing the internal states (thoughts, feelings,wishes) of
characters.
Finally, an examination of the use of sequential connectives
(and, and then) isintended to capture children's awareness of the
temporal ordering of narrativeevents. If children are indeed
sensitive to the functional distinction betweenvoices in replica
play narrative, one should find a disproportionate use of
suchconnectors in the various strands. Narrative strands are
normally temporallysequenced, so that their order in time is
nonreversible (Labov, 1972b). Dialogueand stage-managing
utterances, however, need not be temporally sequenced
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
338 WOLF AND HICKS
since what is being encoded is often the characters' thoughts
and feelings. Thus,an examination of the use of sequential
connectors within each discourse strandmight also provide some
information about children's skill in differentiatingvoices.
HOW EARLY INTERTEXTUALITY EMERGES: A LINGUISTICANALYSIS OF THE
LONGITUDINAL DATA
Heather is no aberration. In all of our longitudinal subjects,
the capacity forintertextuality appears as early as the third year
and continues to develop rapidly.By age 3 all the children
consistently mark their movement between three voices,or types of
narrative text: narrative, character dialogue, and
stage-managing,using features such as performance qualities,
pronouns, and utterance types.Between 3 and 5, children add
temporal features to the constellations of lin-guistic
characteristics they use to signal their use of a particular voice.
Beginningat 5, children's use of the separate voices becomes
substantially more subtle asthey exploit the plurifunctionality of
the several voices, for example, reportingspeech as a narrator or
recounting events via character speech.
The Appearance of Voices Within the TextAs early as children can
produce the relevant linguistic features, their playnarratives
contain both dialogue and narrative voices. Our data also suggests
thatchildren at this very early age can produce stage-managing
strands. The replicaplay narrative from Heather below
illustrates:
Heather (2 years, 11 months)
narrative dialogue stage-managing
and (actions)(Heather manipulates toy animals and people who go
into a house, only to find alion inside)a person will go in the
houseand the lion will comeand eat the person
(Heather walks a toy lion along)here's the lion comingthe lion
ate Pierre [a toy human figure] up(Heather has a toy girl run to a
giraffe)
run, run, run!
the giraffe's scared of the lion
(as giraffe)go back in the houseand go to sleep
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 339
(experimenter toHeather: the girl's goingback to her
house?)yeah! and we'regonna get anotherperson scared
(Heather finds another human figure)two people are gonna go in
the houseand the lion's gonna comeand get them
Even before she is 3, Heather distinguishes separate voices in
her replicanarrative on the basis of pragmatic as well as
linguistic features within eachvoice. Stage-managing utterances are
distinctive in that they are a forum forinteractions with the
experimenter about characters in the story. In this sense,these
utterances represent an excursion from the story world itself, not
unlike theasides or departures that adult speakers index with
changes of tone, "push-popmarkers," or shifts in topic (Polanyi,
1986; Schriffrin, 1982). Such utterancesare nevertheless to be
distinguished from wholly external comments in that theyare
directly concerned with the events taking place in the narrative.
In the shortexcerpt shown above, stage-managing utterances are also
linguistically dis-tinguished from narrative and dialogue
utterances by the use of the second personproform (we're).
Heather differentiates narrative and dialogue strands in her
story on the basisof a still different constellation of features.
Segments recounted in the narrativevoice contain full nominal forms
(the lion, a person) along with third personpronominal forms
(them). In addition, narrative clauses are cast in
declarativesentence forms. In this particular example (although
this varies from child tochild), dialogue segments consist solely
of imperatives containing no overt pro-nominal forms. The
predominant tense in narrative strands is the present,
usedfrequently with gonna constructions as Heather engages in a
kind of predictivestyle of narration (and the lion's gonna come and
get them). Dialogue utterances,being imperative in nature, are not
marked for tense, although one could hypoth-esize that any
nonimperative utterances would also be marked with present
tense.
In this way, Heather already performs stories in which she moves
fluentlybetween narrative, dialogue, and stage-managing, marking
each strand using adistinct constellation of types of pronominal
forms and sentence types. In somerespects, one also sees a nascent
distinction between strands with regards to thetypes of temporal
linguistic forms used. Note, for example, the frequent use ofand as
a type of sequencer in Heather's narrative strands. In general,
however,narrative and dialogue strands are not distinguishable in
Heather's narrative interms of either the types of verbs used to
encode events (most of the verbs usedare event verbs), the
predominant tense (the present), or the use of
sequentialconnectives (and is found in both narrative and dialogue
strands).
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
340 WOLF AND HICKS
The excerpt from Heather's narrative illuminates two separate,
though relat-ed, linguistic skills available to 3-year-old
children. First, even children thisyoung have the ability to
represent and to encode separate voices: Even after shehas exited
from the story world to plan what is coming up, when she resumes
heruse of the narrative voice, it is with the same constellation of
characteristics sheused earlier. Second, Heather's narrative points
out that the manipulation ofseparate voices in narrative also
involves the ability to move back and forthwithout disrupting the
overall task of making a coherent narrative. The theme ofcharacters
being scared by a lion is maintained and the narrative events
advancewithout disruption, even though she moves back and forth
across three voices.The fact that Heather makes cohesive ties
across voices is equally important.When, in a stage-managing
clause, she says "and we're gonna get anotherperson is scared,"
"another" may refer back not only to the preceding commentby the
experimenter, but to her own earlier mention of people whom the
lionfrightens. This means that young children represent to
themselves the demandsfor separate voices, and then utilize
linguistic skills in creating distinct voices atthe same time that
they understand themselves to be telling an
integratednarrative.
The Temporal Marking of VoicesAn examination of data from
subjects between the ages of 3 and 6 suggests thatdevelopmental
changes in the marking of separate voices occur mainly in
chil-dren's use of the temporal system. The excerpts below further
suggest that by age5 the differential use of temporal forms within
voices is well in hand. The replicaplay narrative from Jeannie at
age 4 illustrates:
Jeannie (4 years, 1 month)
narrative dialogue stage-managingand actions(Jeannie is playing
with large and small elephants as well as some other
animalreplicas. She pushes over a small elephant.)
this is going to beher baby
and her baby followed him(experimenter: and the big elephant
said:
"oh, how beautiful I am!")(in high voice) me too, Mommy(as
hippo) he sure is ugly
(experimenter: the elephant was very sad)and the baby too
(as baby) let's go to another jungle(Jeannie walks elephant off
to the "city")
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 341
they cameand heard himand they were his friends(she turns the
figures to face elephant)they gave them foodand gave them a
barn
Jeannie's narrative points out several changes with respect to
the use of thetemporal features (semantic verb types, verb
inflections, and sequential connec-tives) used to mark the
different voices within a text. She recounts narrativestrands
entirely in the past, whereas she casts dialogue (as well as stage
manag-ing strands) in the present. In addition, narrative strands
contain a large numberof sequencers (and), but dialogue strands
consist of interchanges between char-acters which are not
explicitly marked for their temporal sequence. Finally, whenshe
narrates segments one finds a large number of event type verbs
{gave,followed, came) representing actions having a set terminus.
In contrast, when shespeaks for characters one finds the use of
state verbs (is-ugly, is-beautiful) inaddition to event verbs (go).
Thus, in her narrative at age 4, Jeannie uses abroader
constellation of features to distinguish among voices. Not only
does shemake use of pronominal and nominal forms and sentence types
but she also usesfeatures of the temporal system to signal whether
events are being portrayed fromthe perspective of a spectator or a
participant.
In addition to the changes exemplified in Jeannie's narrative,
both dialogueand narrative segments become more lengthy between
ages 3 and 5, so thatcharacters may have exchanges extending over
several turns and the narrator mayengage in lengthy discussions
about events. The excerpt below from Jonathan'snarrative at age 5
illustrates;
Jonathan (4 years, 11 months)
narrative dialogue
(Jonathan is constructing a story centered on a "strange
creature" living in a "funnyland")(Jonathan holds a boy figure)and
then he would go outand restand then he found a snakebut he picked
it up
it's really a snakehe saidthen he went over to his sisterbut she
was scaredso she ran away
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
342 WOLF AND HICKS
and she got lostthis was the land of funny menand then the guy
[the strange creature] cameand he said
what's the matter?
(girl figure) I got lostwell, what's this land called?
(strange creature) funny landwhat do you think it's called?
(girl) well, where were you?(strange creature) in that
tree(girl) what tree? over there?(strange creature) yeah(girl)
bring me over
so he didhe [the strange creature] was madthis one (Jonathan
points to the strange creature)because he liked that little girland
he decided to take her away to funny land
In the context of these longer segments, children recast
information acrossvoices. For example, in the excerpt shown above,
Jonathan, in his voice asnarrator, announces that the boy character
in the story found a snake ("and thenhe found a snake"). Then, this
same information is recast in the form of thereplica character
announcing, "it's really a snake." Additionally, one finds atthe
end of the series of interchanges between characters an imperative
from thegirl character in the story ("bring me over"). This same
information is thenrecast in the narrative voice as Jonathan
narrates, "so he did." These recastingsprefigure what children
achieve between 5 and 6, as they work out the ways inwhich any
particular voice can be made to serve a variety of narrative
functions.
Differentiations within Voices: The Development
ofPlurifunctionalityBetween ages 5 and 6, children elaborate their
understanding that any voice cancarry any type of information in a
new direction. Each of the several voices theyuse comes to exhibit
increasing and more diversified plurifunctionality. By thetime they
are 5, children begin to have characters speak about
eventsrecollect-ing, forecasting, and sportscasting ongoing
actions. In a complementary waythey use the narrative voice to
present speech indirectly, offering as a matter ofobservation that
"the bear told him . . . " or "the pirate shouted that . . . " .
Theexcerpt below from Heather's narrative shown below
illustrates:
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 343
Heather 4 years, 7 months
(Heather is creating a story surrounding a giant living in a
strange forest with"regular" animals)dialogue stage-managing
(Heather holds a replica giraffe)oh, I'll eat some grassyum,
yumlet me eat some leaves from the earthquake last nightdid you
knowthat there was an earthquake last night?
(duck) oh, no(giraffe) eat some leaves of this tree standing
up(as forest animals) what have you done?(as giant) took your
bathand I made a pond for you
this is really Santa Claus
(as giant) I gave you a nice present to the little animalsthey
lived in the forest once long agothey came this year
Both the brief recall at the beginning of the segment and the
longer narrativerecounted by the giant in the last dialogue segment
represent the plurifunc-tionality of a particular voice within the
larger discourse. Here, Heather demon-strates that she has skill
not only in manipulating separate voices but also inembedding a
variety of discourse functions within separate voices. She
hasbroken out of a strict one-to-one match between voices and
functions. She canpresent speech as dialogue, she can report it as
an event, she can duck out intostage-managing and try out, edit,
and reperform a bit. She has, if not all, many ofJames'
options.
SUMMARY OF THE LINGUISTIC ANALYSISThis look at children's
differential use of constellations of linguistic featuresshows that
already in their third year, children tell narratives that are
anything butsimply sequential. Already their stories are
multivocal, involving the weavingtogether of at least three
distinct voices: stage-managing, character dialogue, andnarrative.
By age 3, separate voices are distinguished in children's replica
playnarratives on the basis of use of pronominals and of sentence
types. Betweenages 3 and 5, children use their emerging
understanding of the temporal nature of
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
344 WOLF AND HICKS
different types of narratives and their increasing control of
the temporal systemsof their language to mark distinctions between
the several voices in their playnarratives still more emphatically.
Between 5 and 6, children showed an increasein the
plurifunctionality of voices. By age 5, the children in the study
showed anincreasing tendency to embed multiple functions within
voices, so that narrativesegments could be recounted by story
characters and character speech could berecounted from the
perspective of the narrator. Thus, although the children in
thelatter 2 years of the study generally marked voices with
consistent use of lin-guistic forms, they also demonstrated an
increasing amount of skill in the manip-ulation of these linguistic
forms.
One way of thinking about the early and elegant emergence of
several voicesand multiple functions within children's narratives
is to suggest that the linguisticdistinctions "piggyback" on a
fundamental understanding of the differentstances or perspectives
that a person can takeor find themselves inwithrespect to events.
The role or voice of a narrator with its more distant stance
withrespect to events, may encode the experience of looking-on, or
observing, whatBritton (1982) terms a spectator stance. When
children engage in characterdialogue, however, their perspective on
events is much more intimate in nature,so that this voice may be
representative of what Britton terms a participantstance, a stance
which may encode what it is like to be in the midst of an event.The
role and voice of the stage-manager is a kind of executive stance,
it comesfrom the position of someone who must shepherd events
towards their end,without wanting to observe and without being in
the midst of what is going on.Perhaps it has its earliest roots in
the nurturing, nudging voices of mothers andfathers. The increase
in the latter 2 years of the study of the plurifunctionalitywithin
voices may reflect recognition that more than words convey the
meaningof stories: The vantage point from which we offer
information hints broadly athow much involvement, empathy, and
intimacy is entailed.
SOCIOCULTURAL DIFFERENCES IN THE MARKING OFVOICE
DISTINCTIONS
The previous discussions were based upon data from middle-class
children, alargely homogeneous population in terms of home language
socialization (Snow,1983). In our discussions of the developmental
data, we have seen changesbetween ages 3 and 6 in how voice
distinctions are marked. Yet, the ability tomark separate voices in
play narrative is not a linguistic given but forms part
ofchildren's larger socialization into language skills. We know
that adult speakersfrom different sociocultural backgrounds utilize
different linguistic means toencode narrative events (Gumperz,
Kaltman, & O'Connor, 1983; Labov,1972b), and that children also
adopt different narrative styles as part of theirprimary language
socialization. Indeed, sociolinguistic analyses of children's
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 345
narratives has shown that children of different cultural
backgrounds have quitedistinctive ways of talking about events
(Gee, 1985; Heath, 1983; Michaels &Collins, 1984).
The immediate question then becomes "How is the way in which
childrenmark and use voices sensitive to the differing notions of
narrative prevalentamong speakers and writers in different language
communities?" The narrativesof two 7-year-old girls from different
language backgrounds provide at least aninitial answer. In looking
at these narratives, we shall see that while both girlsmark the
differences in separate voices using similar constellations of
linguisticfeatures, the overall pattern of voices within their
narratives is markedlydifferent.
Malka, the first of our two subjects, comes from a home
environment which isupper middle-class and in which a great deal of
emphasis is placed upon literacy-related activities. Malka attends
a private school where children are exposed tolong, sophisticated
written texts and expected to be active "inventors" when itcomes to
spelling and writing. In her classroom, Malka is considered a
"writ-er." In fact, she spent many free afternoons drafting a
"novel". Rene comesfrom a working-class home and attends a public
elementary school where literacyskills are taught chiefly by rote
and where correct code use is valued much morethan inventiveness.
In addition, Rene is black, and her narratives even at age 7contain
some of the characteristically rich storytelling style of speakers
of BlackEnglish (Gee, 1985; Heath, 1985; Labov, 1972b): rich use of
prosodic devices,active gestures and facial expressions, open
expression of emotions and evalua-tions. Rene is considered a
"talker" in her classroom. She knows countlessjump rope rhymes and
lyrics to radio songs. She can be a fierce tease. When shewinds up
to tell the full story of Martin Luther King, it includes
everythingright down to her grandmother's opinions.
Rene and Malka's stories are drawn from a larger corpus of
narratives ob-tained from 54 children. Each of the 54 subjects was
shown a short silent filmabout children playing with a magic yellow
hat and was asked to perform threetasks. Children were first asked
to narrate the film's events as a story without theuse of toy
figures, pictures, or other props. They were then asked to narrate
thefilm's events using small replica figures representing
characters (along with theyellow hat) in the film. Finally,
children were asked, according to their abilities,to either write
or dictate the story in the film. The replica play task constitutes
auseful base for comparison with our earlier sample of eight
children. It is,furthermore, of particular interest for our
purposes here since this task was basedupon the same stimulus, a
short silent film, modeled on Chafe's "Pear Film"(Chafe, 1980).
When Malka narrates the replica play task she renders the events
in a carefullysequenced form that is perhaps closer to an account
than to a story. She immedi-ately assumes the outside voice
(perspective) of the narrator and maintains this
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
346 WOLF AND HICKS
particular stance throughout. She is quite exact about what she
says, rarely goingbeyond the clearly observable events of the film.
Her narrative is reproduced inpart below:
Malka (age 7)
narrative dialogue
one day two boyswere hiding in thepark with a hat
let's surprise the boywho always walks through
one whispered to the otherthis boy walked into the parkhe looked
aroundand then one of the boys threw the hat inhe picked it upand
put it onthen he started twirling aroundone of the boys ran outand
grabbed the hatpulling it back inthe boy dove into the bushes after
his hathe's probably got a headache afterwards'cuz he dove in
headfirstfinally he came output his coat over his headand put on
his mittensand then walked behind a big bushto hidetwo boys looked
for himbut they couldn't see himsince he was hiding
(the two boys speaking)let's go look for him
they took the hat . . .they walked alongfinally/the/the other
kid who had hid behind the bushesjumped outthey walked off together
with their coats upand one of the kids wearing the hat . . . .
It is obvious from a global examination of Malka's narrative
that events arerecounted almost wholly from the narrator voice,
with the dialogue voice playing
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 347
only a minor role. Aside from the dominance of narrative
strands, the narrativevoice is quite sharply marked with the
distinctive use of sequencers like and thenand finally, and also
uses connectives which indicate precise temporal bound-aries and
causal relatedness (until, since). In addition, both of the
dialogueutterances in this particular example serve only as a
forecast of events to occur inthe narrative but do not provide the
kind of psychological or evaluative informa-tion which typically
supplements or enhances the depiction of an event chain(Labov,
1972b). Finally, it is the narrative strand that Malka has
differentiated:In the narrative voice she both depicts events ("the
boy dove into the bushesafter his hat") and comments on them ("he's
probably got a headache after-wards"). In sum, it seems that Malka
takes the stance of a spectator as shenarrates the events from the
film.
In her replica play version of the yellow hat film, Rene assumes
a dramat-ically different approach to encoding the film's events,
treating it, not as a report,but as a highly evaluated story,
perhaps even bordering on a dramatic representa-tion of the film
events. In Rene's story, the dialogue voice is given higher
status,and many events important to her story's development are
recounted from theperspective of one or the other character. In her
rendition, it is dialogue ratherthan narrative which is internally
differentiated. She creates two quite distinctiveroles in the
finder and his friend, whose two-part interactions move the
story,rather than merely echo what the narrator has already laid
out. As a result, herreplica play narrative as a whole is
representative of a more participatory stancetowards events:
Rene (age 7)narrative dialogueso those two guys in the bushesI
mean/the housethey are eating/um/some pie/grape pieand the
other/there's a boycomes
and throws a yellow hatand hides behind the bushesand then the
oth . . . /the boy goes/saw itand picked it upand put it onand
said
look what I found over there!and his friend said
that's a nice bright yellow hatand then
are you gonna keep it?
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
348 WOLF AND HICKS
and he saidyeahbut I'm gonna comb my hair a little
and so/his friend put it right downthen/the other one started
eatingthe other one was combing his hairthen they took the hat
awayand the friend said
where's my hat?where's my hat?it was right over there!
and then the other guy saiddo you/are you hiding it from me?
the other guy saidno I'm not
and the guy saidI believe you
then they started looking for it in the bushes . . .
The proportional use of dialogue segments is higher in Rene's
replica playnarrative than in Malka's, but even more interesting is
the way in which dialoguestrands are used. Recall that in Malka's
story, dialogue strands played a second-ary role in the overall
structure of the narrative. In Rene's story, however,dialogue
carries important information which is conveyed from the
perspectiveof the characters involved in the actions. The
projection of information fromcharacters' perspectives means that
one obtains additional information about howcharacters feel about
certain events (that's a nice bright yellow hat). It also meansthat
certain events in the story take on a quality of greater importance
andexcitement, as characters express their reactions to happenings.
Compare, forexample, Rene's rendition of one of the key events in
the storythe point atwhich the yellow hat is stolen from the
principal characterto Malka's renditionof the same events. From her
perspective as an outside spectator, Malka gives astraightforward
account of what happened (one of the boys ran out and grabbedthe
hat, pulling it back in). In contrast, through her use of the
dialogue voiceRene takes a radically different approach to the same
event (and the friend said,"Where's my hat? Where's my hat? It was
right over there!"). Seen through theeyes of the characters, these
key events are recounted in all their gravity andexcitement. Thus,
Rene in her replica narrative assumes much more of aparticipant
stance in terms of her extensive use of dialogue strands.
These differences, however, take place against a background of
shared capac-ities: Both Malka and Rene distinguish among voices by
employing largely
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 349
similar constellations of features. Both girls mark narrative
clauses through theuse of third person, declaratives, past tense, a
preponderance of event verbs, andsequential connectives. They
signal the onset of dialogue segments with first andsecond person
pronouns, the presence of questions and imperatives, presenttense,
a higher proportion of state verbs, and the absence of sequential
connec-tives. The way in which the two voices are usedthat is,
their function in thediscourseis, however, quite different. Malka
assumes greater distance fromevents occurring in her replica play,
narrating actions in the voice of a narrator.As a result, the
emphasis in her narrative is on the sequential and causal
linksbetween eventsin short, the when and why of what happened.
Rene takes asomewhat different stance, narrating events in part
from the perspective of anarrator but in part from the perspective
of the characters involved in the action.Some of the important
events in the story are recounted through the eyes of thecharacters
involved and thus take on a more emotional quality. In this way,
thedescriptive analyses of the two narratives above provide some
evidence thatintertextuality in children's play narratives is like
many other aspects of narrativecompetence, highly sensitive to
sociocultural variation in terms of how separatevoices are used to
encode events. Although both girls understand the power andthe
importance of portraying the "murmur of many voices" within their
nar-ratives, they choose different leading voices. Malka speaks as
an observer poisedat the rim of events. Rene speaks chiefly in the
voices of participants, from themiddle of the narrative.
CONCLUSIONS
Heather, Jonathan, Malka, and Rene should revise our notions of
what it is to tella story. Certainly, that work involves getting
the sequence of events down,subscribing to some notion of high
point or goal, and remembering to mind themoment-to-moment ties and
references which will make the performance awhole rather than a
collection of spoken bits. But this kind of portrait is tooshallow,
too simple. Milan Kundera, the Czech novelist, is probably
right:Stories are much more like pieces of music than we admit. The
best ones areanything but single tunes.
REFERENCESApplebee, A. (1978). The child's concept of story.
Chicago: University of Chicago Press.Bakhtin, M. (1981). The
dialogic imagination: Four essays. Austin, TX: University of Texas
Press.Berman, R., & Slobin, D. (1984). Coding manual:
Temporality in discourse. Nijmegen, Nether-
lands: Max-Planck-Institut.Britton, J. (1982). Spectator role
and the beginnings of writing. In M. Nystrand (Ed.), What
writers
know. New York: Academic.de Beaugrande, R., & Dressier, W.
(1980). An introduction to text linguistics. London:
Longman.Bruner, J. (1986). Actual minds, possible words. Cambridge,
MA: Harvard University Press.Chafe, W. (Ed.) (1980). The pear
stories. Norwood, NJ: Ablex.
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
350 WOLF AND HICKS
Eisenberg, A. (1985). Learning to describe past experiences in
conversation. Discourse Processes,8, 177-204.
Fowler, A. (1982). Kinds of literature: An introduction to the
theory of genres and modes.Cambridge, MA". Harvard University
Press.
Gee, J.P. (1985). The narrativization of experience in the oral
style. Journal of Education, 167, 935.
Gee, J.P. (1989). Two styles of narrative construction and their
linguistic and educational implica-tions. Discourse Processes, 12,
287-307.
Gee, Julie. (1986). Beyond semantics: A discourse analysis of
the verb inflectional system in distinctnarrative-like and
communicative formats in the speech of a two-year-old. Paper
presented atthe Symposium on the Acquisition of Temporal Structures
in Discourse, Chicago LinguisticsSociety Meetings.
Geertz, C. (1983). The way we think now: Toward an ethnography
of modern thought. In Localknowledge: Further essays in
interpretive anthropology (pp. 146-163). New York: BasicBooks.
Gumperz, J., Kaltman, H., and O'Connor, M.C. (1983). Cohesion in
spoken and written discourse.In D. Tannen (Ed.), Coherence in
spoken and written discourse. Norwood, NJ: Ablex.
Halliday, M.A.K., & Hasen, R. (1976). Cohesion in English.
London: Longman.Heath, S.B. (1983). Ways with words: Language, life
and work in communities and classrooms.
Cambridge, England: Cambridge University Press.Heath, S.B.
(1985). The cross-cultural study of language acquisition. Keynote
address, Stanford
Child Language Research Forum, Stanford, CA.James, H. (1984).
The awkward age. Hammonds worth, England: Penguin.Kintsch, W.,
& Greene, E. (1978). The role of culture-specific schemata in
the comprehension and
recall of stories. Discourse Processes, 1, 1-13.Kundera, M.
(1987). An Interview with Milan Kundera. Salnagundi, Spring,
1-24.Labov, W. (1972a). Sociolinguistic patterns. Philadelphia:
University of Pennsylvania Press.Labov, W. (1972b). The
transformation of experience in narrative syntax. In Language in
the inner
city. Philadelphia: University of Pennsylvania Press.Langer, J.
(1986). Children reading and writing: Structures and strategies.
Norwood, NJ: Ablex.Mandler, J., & Johnson, N. (1978).
Remembrance of things parsed: Story structure and recall.
Cognitive Psychology, 9, 111-151.Michaels, S. (1981). "Sharing
time": Children's narrative styles and differential access to
literacy.
Language in Society, 10, 423-442.Michaels, S., & Collins, J.
(1984). Oral discourse styles: Classroom interaction and the
acquisition
of literacy. In D. Tannen (Ed.), Coherence in spoken and written
discourse. Norwood, NJ:Ablex.
Nelson, K. (1986). Event knowledge: Structure and function in
development. Hillsdale, NJ:Erlbaum.
Olson, D. (1979). From utterance to text: The bias of language
in speech and writing. HarvardEducational Review, 47, 257-281.
Parsons, D. (1986). Underlying events in the logical analysis of
English. In E. LePore (Ed.), Thephilosophy of Donald Davidson:
Perspectives on actions and events.
Peterson, C , & McCabe, A. (1983). Developmental
psycholinguistics: Three ways of looking at achild's narrative. New
York: Plenum.
Polanyi, L. (1986). Telling the American story: A cultural and
structural analysis. Norwood, NJ:Ablex.
Reisman, K. (1987). Men and women's divorce narratives.
Unpublished manuscript, Smith College,Northampton, MA.
Schegloff, E.A. (1971). Notes on conversational practice
formulating place. In Giglioli (Ed.),Language and social context.
Hammondsworth, England: Penguin.
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015
-
THE VOICES WITHIN NARRATIVES 351
Schriffrin, D. (1982). Discourse markers: Semantic resource for
the construction of conversation.Unpublished doctoral thesis,
University of Pennsylvania, Philadelphia.
Shotwell, J., Wolf, D., & Gardner, H. (1980). Styles of
achievement in early symbolization. In M.Foster & S. Brandes
(Eds.), Universals and constraints in symbol use. New York:
Academic.
Snow, C. (1983). Literacy and language: Relationships during the
preschool years. Harvard Educa-tional Review, 53, 165-189.
Stein, N., & Glenn, C. (1979). An analysis of story
comprehension in elementary schoolchildren. InR. Freedle (Ed.), New
directions in discourse processing. Hillsdale, NJ: Ablex.
Tannen, D. (Ed.). (1982). Spoken and written language: Exploring
orality and literacy. Norwood,NJ: Ablex.
Todorov, T. (1981). Mikhail Bakhtin: The dialoguicprinciple.
Minneapolis: University of MinnesotaPress.
Vendler, Z. (1967). Linguistics in philosophy. Ithaca, NY:
Cornell University Press.Wolf, D., Davidson, L., Davis, M., Scripp,
L., Walters, J., & Hodges, M. (1988). Beyond A, B,
and C: A broader and deeper view of literacy. In A. Pellegrini
(Ed.), The psychological basesof early education. London:
Wiley.
Wolf, D., Goldfield, B., & Beeghly, M. (1984, October
10-11). "There's not enough room," thebaby said. Paper presented at
the Boston University Language Conference, Boston.
Wolf, D., & Grollman, S. (1982). Ways of playing: Individual
differences in imaginative play. In K.Rubin & D. Pepler (Eds.),
The play of children: Current theory and research. New
York:Karger.
Wolf, D., & Pusch, J. (1985). The origins of autonomous
texts in play boundaries. In L. Galda & A.Pellegrini (Eds.),
Play, language, and stories: The development of children's literate
behav-ior (pp. 63-78). Norwood, NJ: Ablex.
Wolfram, W., & Fasold, R. (1974). The study of social
dialects in American English. Englewood,NJ: Prentice-Hall.
Dow
nloa
ded
by [M
cGill
Univ
ersity
Libr
ary] a
t 13:5
5 02 A
pril 2
015