National Council of Teachers of English

Teaching Science Writing to First Graders: "Genre Learning and Recontextualization"
Author(s): Julie E. Wollman-Bonilla

Source: Research in the Teaching of English, Vol. 35, No. 1 (Aug., 2000), pp. 35-65
Published by: National Council of Teachers of English
Stable URL: http://www.jstor.org/stable/40171506 .
Accessed: 25/06/2014 04:49
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
National Council of Teachers of English is collaborating with JSTOR to digitize, preserve and extend access to
Research in the Teaching of English.
http://www.jstor.org
This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

All use subject to JSTOR Terms and Conditions
Teaching Science Writing to First Graders:
GenreLearningandRecontextualization
Julie E. Wollman-Bonilla
Rhode Island College
Sciencewriting is integralto the work of scientists and to learning to do science. Science

texts composed by emergentand beginning writers in Family Message Journals provide
insights related to debates about genre instruction,writing as social practice,and writing
processprinciples of voice and empowerment.This qualitative study of four first-graders'
messages to family membersexplores whether the children'stexts can be characterizedas
sciencewriting when analyzedfor conventionalstructuraland lexicogrammatical features. It
also explores how the Family MessageJournal context, with its systematicinstructionand
familiar, responsiveaudience might influencechildren'sunderstandingof the conventionsof
sciencewritingand theirability to use themflexibly. The data demonstratethat the casestudy
participantsconsistentlycomposedtexts in which they appropriatedthe linguistic conven-
tions of scienceand that they seemed able to use these conventionsflexibly, recontextualizing
the genres tofit the task of a written dialogue with theirfamilies.
DearMom Dad andRosa writing. The topics of Family Message

we chiride[tried]ana exsperument do you Journal dialogues are learning experi-
knowthe difrinsbetweena hardboiledegg ences from acrossthe curriculum, like
anda ro [raw]egg well Mrs.Carolancract
a hardboild egg on MichaTshaed . . . and the science inquiry Sara described.
the hardboildegg spinsandthe ro egg dunt Their teacher challenged the firstgrad-
rememberthatinsiklupeedya[Encyclope- ers to figure out how to tell if an egg is
dia] brownboy dutektevwell remember raw or hard boiled without having to
thatwell I do LoveSara
crack it and risk the mess the children
Sara,a first grader,composed this expected to cover Michaels head.The
message recounting a science experi- teacher knew it was hardboiled before
ment in her Family Message Journal. she crackedit, but how? All the eggs in
FamilyMessageJournalsarenotebooks her carton felt cold and looked alike.
that travel from school to home and Working in small groups the chil-
back againdaily,with a messageby each dren weighed, measured,carefullyob-
child that family members reply to in served shell color and texture, and
Research in the Teaching of English • Volume 35 • August 2000 35

manipulated eggs. They found that an familiarityof the audience encouraged
egg spins only if hard boiled. Sara the deliberate recontextualization of
recalled that the literary characterEn- conventions.
cyclopedia Brown (Sobol, 1963) had
solved a case using this information, Theoretical Background
and she reminded her family of reading Writing as Central to Learning
his story together. and Doing Science
FamilyMessageJournalsarea con- Literacy in any field is as much a
text where content-area learning is sociocultural as a cognitive achieve-
linked with explicit instructionin writ- ment, involving particular frames of
ing for the purpose of communicating mind, conceptual understandings,ways
about school activities that the in- of knowing and inquiring,and forms of
tended audience did not experience. In communication developed in particu-
this article I look at the science writing lar social contexts (Gee, 1989; Lave,
first gradersproduced in Family Mes- 1988;Rogoff, 1984) . Developing scien-
sageJournals.Recent researchsuggests tific literacy involves adopting the
that children have relativelyfew expe- thought processes, interpretations of
riences with science writing and that experience, behaviors,and language of
primary-grade teachers may need to scientists.This learning process occurs
provide more guidance and practice in through social interaction with those
sciencegenres(Chapman,1995;Halliday who are members of the scientific
& Martin, 1993; Kamberelis, 1999; community, including teachers who
Rothery, 1989). This study took place guide children into the discourses of
in a context where teachers provided science by modeling and then scaffold-
such guidance and practice. ing students'attempts (Christie, 1998;
As Sara'smessageexemplifies,many Lemke, 1990; Martin, 1999).
of the children'sscience texts reflected National standardsand recent calls
conventional features of science writ- for reform in science education reflect
ing (e.g., the aim-method-resultsstruc- remarkableconsistencyin what it means
ture of an Experiment Recount, the to do science(AmericanAssociation for
universalityof "dhardboiled egg and a the Advancementof Science, 1996;Lee,
ro [raw] egg,"and the timeless,present 1999; National Center on Education
tense verbs"spins"and "doesn't [spin]" and the Economy, 1997) . Scientistswho
suggesting that these qualities are gen- adhere to Western views of science
eral characteristics).But their texts also interpretthe world, think, and write in
included components thatarenot char- discipline-defined ways. Learning to
acteristic of science writing, as when write like a scientist is central to
Saradirectly addressedspecific readers learning the processes of mainstream
with reference to a particular,shared science in U. S. society (Halliday &
story experience. I will explore the Martin, 1993; Lee, 1999; Lemke, 1990).
hypothesis that the dialogic nature of For scientists, writing is not only a
the journal and the specificity and mode of communicatingtheir findings,
36 Research in the Teaching of English • Volume 35 • August 2000

it is alsoa tool forreasoningandproblem- Perspectiveson Writing
solving, for organizing thoughts, and Writing can facilitate learning; com-
for engaging in and guiding inquiry. posing slows down thought processes,
The language of science models the inviting consideration of ideas and
discipline's conceptual structure- the questions.Writing also nudges students
taxonomies that are the foundation of to articulate,shape, and organize ideas
scientific knowledge. Moreover,learn- in their own words and to connect
ing the language of science is essential background knowledge and beliefs to
to being able to take a critical stance new information,makingpersonalsense
and perhaps open up new spaces and of it (Britton,1970;Emig, 1977;Martin,
practices for scientific communication D' Arcy,Newton, & Parker,1976).How-
(Bazerman, 1997). That is, students ever, writing is not just an inner-
must first learn the hegemonic dis- directed tool or "a means for personal
course of science in order to challenge growth and development" (Rothery,
the inequitable power structures it 1989, p. 203). It is a resource for
represents,whereby access to informa- constructing meaning that realizes so-
tion and participationin doing science cialpractices(Halliday& Martin,1993) .
is restrictedto those with the necessary Despite the focus on individualliteracy
cultural capital (Bourdieu & Passeron, development evidenced by the popu-
1977; Martin, 1989, 1998, 1999). larityof the writing processapproachin
Although there is more to doing the past 20 years,the power of writing
science than its language, scientific is grounded in socially constructed
reasoning is a linguistic process. Be- conventions.
cause the work and thinking of science Recognizing that texts accomplish
is grounded in specialized forms of social practices,some composition re-
communication and the language pat- searchershave focused on the impor-
ternsthey entail(Bazerman,1997, 1998; tance of introducing children to
Christie, 1989, 1998; Halliday & Mar- purposeful, audience-directed writing
tin, 1993; Lemke, 1990; Martin &Veel, tasks.When they write for functions of
1998), it is difficult to conceive of everyday life and work, even young
teachingscience without teachingwrit- childrenmay discoverwhat writing can
ing at the same time. Yet research accomplish in the world (Hall, 1998).
suggests that students are not often But writing s potential power relies on
taught that science revolves around successfullyanticipatingaudience per-
particular forms of writing that are spectivesand needs (Frank,1992;Kirsch
realized through conventional lexico- & Roen, 1990; Ryder, Vander Lei, &
grammaticalpatterns(Halliday& Mar- Roen, 1999).When children are writ-
tin, 1993;Martin, 1989;Rothery, 1989, ing about school activities,the teacher
1996) . In partthis problemgrows out of or peers may not constitute a real
dominant perspectiveson writing that audience.Families,however,are a more
profoundly influence teacher educa- authentic audience, in need of writing
tion and classroompractice. that is clear and engaging enough to
TeachingScienceWritingto First Graders 37

communicate what happened to those Chapman, 1995; Kamberelis; 1999).
not present.Sharingtheir writing with Although detractorsarguethatchildren's
families capitalizeson children'sdesire writing will be stripped of agency and
to communicate with others (King & voice, teachers who provide explicit
Rentel, 1979) by providing feedback instruction and practice in applying
on message clarity and effect within a genre characteristicsmake new forms,
purposefultask. appropriateto different disciplinesand
Another strand of composition functionsaccessibleto children(Christie,
research,rooted in the Australiangenre 1989, 1998; Lemke, 1990; Pappas &
movement, has focused on empower- Pettegrew, 1998; Rothery, 1996;
ing child writers by introducing them Schleppegrell,1998). In thissenselearn-
to socially valued genres. Within this ing genre conventions does not restrict
school of thought, genres are defined as children's agency and voice; rather,it
"social processes. . . for realizing pur- liberates children by developing their
poses or goals through language,"with power to engage in, contribute to, and
language characterizedby a particular critique a discipline (Martin,1999).
text structure and lexicogrammar Furthermore,some researchersas-
(Rothery, 1989, p. 221). Thus, textual sert that genres are not immutable
and social perspectives on genre are forms that limit self-expression (e.g.,
complementary(Bazerman,1998;Mar- Chapman,1995;Kamberelis,1999;Kress,
tin, 1998). Genres'social functions are 1999). A generativeview of genre por-
establishedthrough their structureand traysgenresnot as forms to be replicated
functional grammar (Bazerman,1997, but ratheras forms to be appropriated
1998; Cooper, 1999; Cope & Kalantzis, and reworked in light of text context,
1993; Halliday & Martin, 1993; function, and the individualwriter and
Kamberelis,1999; Kress,1999; Martin, intended audience. Such a view grows
1989). Knowing the right genre to use out of the Bakhtinian notion of re-
in a situation and knowing how to use voicing, which recognizes both the
it enhances children'spower to com- shaping power of socio-historical dis-
municate in society and participatein course forms and the agency and im-
academic disciplines (Christie, 1989; mediate socio-historical context of the
Martin, 1989; Rothery, 1989, 1996). text-producer and the eventual reader
Australianresearchershave decried (Bakhtin,1986;Bazerman,1997, 1998).
the child-centered, facilitative, non- Whereas researchers in North
directivestance of many teachersinflu- America have often framed genres as
enced by the whole language and flexible tools, Australianshave warned
writing process movements (Christie, that although knowing genres may
1989;Halliday& Martin,1993;Rothery, empower children to "manipulateand
1989, 1996). They argue that elemen- experiment with them,"they must first
tary school children are generally not be learned (Christie, 1989, p. 163; see
familiarwith many genres of power, in also Halliday& Martin, 1993; Rothery,
particular science writing (see also 1989). Without mastery of conven-

tional features,what appearsto be genre children write by self-selection
experimentation may simply be failure (Chapman, 1995) or when asked by a
to have learned the conventions (Mar- researcher to produce (Kamberelis,
tin, 1989; Rothery, 1996). Despite dif- 1999) or recreate (Pappas,1991) par-
ferences, researchershave agreed that ticular genres. My study considers
genresprovide a structureto work from children'stexts as examples of science
and a set of linguistic resources that writing in a context characterizedby
must be mastered.They also agree that systematic instruction and authentic
learning conventions need not neces- readers. This is a naturalistic study
sarilyeradicateand may in fact enhance within a setting where genre instruc-
a writer's power, creativity,or voice. tion and writing assignments,an out-
Martin's (1999) recent research with of-class audience (families), and daily
11- to 12-year-olds demonstratesthat written feedback from that audience
once taught genre conventions, stu- were all provided. Guided by the con-
dents treat them as "tools that can be cept of genre as social action realized
adaptedand redeployedin relatedcon- through specific text structures and
texts" (p. 139). linguistic patterns,I focus on identify-
Looking at first graders' science ing the science genresproducedby four
writing may help to illuminate the case-studychildren.I analyzehow they
relationship between learning to use appropriatedconventional text struc-
specific languagepatternsand develop- tures and lexicogrammatical resources
ing one's voice and social power as a to achieve and recontextualize these
writer. As Lemke (1990) argues,"The genres to fit the task of a written
job of science education is, at the very dialogue with their families.
least, to teach students to use language In this article I addresstwo ques-
according to the thematic patterns of tions:
science, flexibly and for their own
purposes" (p. 100). The question re- 1. To what extent can the messages
mains:Can young children appropriate composed by first-grade,emergent,
the patternsof science writing and use and beginning writers be charac-
them flexibly for their own purposes? terized as science writing when
analyzedfor conventional structural
Purpose of the Study and lexicogrammaticalfeatures?
This article explores firstgraders'Fam- 2 How might children appropriate
ily MessageJournalentries produced in the conventions of science writing
response to assignments and related and recontextualize these for the
instruction intended to elicit science purpose of communicating with
writing. The messages discussed here families in the context of Family
were related to the science and health MessageJournals?
curriculain two first-gradeclassrooms.
Previousstudiesof young children's This is not a study of Family
genre writing have investigated what MessageJournals;ratherbecause of the

dual purpose of the science writing lies that their participationwas impor-
contained therein, the journals provide tant and that a brief nightly responseto
a context in which to explore genre the content of the child's message was
learning and recontextualization. By expected. They reassuredfamilies that
exploring how writing science genres masteryof conventional English wasn't
is integrated with writing for an au- necessary,and some wrote in dialects
thentic audience in a personaldialogue, other than standardEnglish.The teach-
this study contributes to the field's ers' clear expectation of involvement,
understanding of how genre conven- regardlessof families'educationalback-
tions may guide text production with- grounds,along with their willingnessto
out acting as rigid formulae. I thus talk with families uncertain about the
intend to disentangle some of the content or form of their replies, was
knotty issues of voice and flexibility effective. Only two families of 48 did
around which writing process advo- not reply regularlyin the journal;other
cates and genre theorists debate. correspondents were found for these
two children.
Method The two firstgradeteachersworked
Setting and Participants as a team, planning learning activities
I discoveredFamilyMessageJournalsin that included Family MessageJournal
two first-gradeclassroomsin a suburban procedures and assignments.Children
Boston elementary school enrolling in both classes,therefore, experienced
about 630 children.Though primarily similarinstruction and in general were
Anglo the studentbody included about assigned to write on the same topics
7% African Americans, 3% Asian and taught the same written forms.The
Americans, and 2% Hispanic Ameri- teachers demonstrated some personal
cans. The majority of students came differencesin teaching style,but due to
from middle-class backgrounds, but their common philosophy,joint plan-
there were many working- and upper- ning, and sharing of materials, the
middle-classchildren as well. similarities in their classrooms were
The two first grade teachers con- more remarkablethan any differences.
sidered families an essentialpart of the The teachers' instruction and their
classroom community, and family in- interview comments reflected a clear
volvement was part of the classroom belief that children learn to read and
culture.Parentsand guardiansserved as write by reading and writing, with
regular volunteers in the classroom, explicit skills instruction embedded in
homework usually involved families, meaningful literacy activities.Invented
and Family MessageJournals- a strat- spellingwas welcomed and encouraged
egy developed by these teachers- were (and explained clearly to families) so
central to the literacy curriculum.The that emergent and beginning writers
teachersexplained to families the value could express themselves freely. All
of writing back to children in these children were treated like readersand
journals and frequentlyremindedfami- writers from the first day of school, no

matter how unconventional their at- posed a weekly newsletter about class-
tempts.Integrationwithin the language room activities that was photocopied
arts and across the subject areas was for families;andsometimeswroteThank
central to the teachers' curricula and You letters.
instruction, with weekly and monthly Family Message Journal writing
themes of study as focal points. often directly followed the activity
Both classroomswere full of trade about which children were writing
books from concept books appropriate (e.g., the egg investigation).The teach-
for early emergent readers,to predict- ers began with an assignment such as
able books, to more linguisticallycom- "Write to your family about what we
plex picture books. Basal readers and just did."They discussed possible con-
textbooks were used selectively and tent for the message as a group prior to
infrequently.Children read to them- writing, with children brainstorming
selves; their classmates;their teacher, what they might write and the teacher
parent and senior citizen volunteers; composing a model message based on
and their families from books they their brainstormedideas and her own
selected with guidance as necessary. sense of the target genre and appropri-
Collections of books relatedto current ate content.
themes of study were availablein the The first grade teachers'modeling
classrooms.Many of these themes grew was similar to Rothery's (1989, 1996)
out of the science and health curricula, description of the joint construction of
but the books related to these themes a text. Such joint construction is
were generally fiction. These theme- grounded in the practice of scaffolding
focusedbooks were regularlyreadaloud children's writing by guiding them
by the teachers and discussed.Some of through interaction to participate in
children'sself-selectedreadingincluded writing a text that models the linguistic
information books, but generally they patternsof the intended genre (also see
chose fiction, which was the most Christie, 1998; Lemke, 1990; Martin,
commonly stocked genre in the class- 1999). However, the teachers in this
room collection.The children also read study did not explicitly discuss gram-
the classroom newspaper The Weekly matical choices such as the use of
Reader every week, which was their present tense verbs to suggest timeless-
most frequent and regularexposure to ness in science writing. Rather, they
publishedinformationaltext. modeled the use of such verbs and
Children usually wrote daily in systematicallydiscussed text structures
their Family Message Journals, in a such as "I think ... I proved . . ." or
variety of genres including poetry, fic- "Today we did an experiment. I ...
tion, informational text, and literary Then I ... I found ..." They also
response, all related to the classroom "reword[ed]some of the students'con-
curricula.Additionally,they occasion- tributions, . . . thus modeling for them
ally wrote, revised, and edited other the languageof written texts"(Rothery,
stories for publication; jointly com- 1996, p. 105). For example, when a

child suggested "snowflakes all look model messages often were begun but
different,"the teacher responded,"Yes, not completed as the teachersadjusted
good thought, no two snowflakes are their role to reflectthe children'sgrow-
alike," rewording to achieve greater ing ability to compose messages on
precision and to avoid the colloquial their own (Christie,1998).At the same
sound. Additionally, they encouraged time that the teachers provided less
the use of a scientific lexicon, remind- initial direction, they raised their ex-
ing students to use the specific terms pectationsfor children'smessages.They
they had learned such as"investigated" encouraged the first gradersto be sure
instead of "looked at" or "tadpole" to "explain everything your family
instead of "babyfrog." might need to know," or asked "Will
The teachers'science writing mod- they understand if you forget to start
els were very rarelydirectly copied by with*We did an experiment to find out
students,but the first gradersused the . . . ?'" In the teachers' instruction,
modeled structures and appropriated attention to audience awareness was
some of the modeled language (e.g., combined with attention to the appro-
"owls'eyes cannot turn in the sockets") priate structure and content for a
as a resource in composing their own message of the targetgenre.
messages(Christie,1989;Rothery,1989). Family replies also provided in-
At the sametime,becausestudentswere struction, albeit indirectly. Elsewhere
writing in a dialogue to familiarreaders (Wollman-Bonilla,2000)I describehow
and because science writing was quite families' replies provided instructional
new to them, the teachers did not scaffolding: feedback indicating mes-
always discourage the contribution of sages' clarity and impact, text models
ideassuch as"Did you know that when demonstratingconventional genre fea-
an owl wants to look aroundhe cannot tures,andoccasionallyrepliesrecontext-
turn his eyes in their holes?" They ualizing some of these conventions.
sometimes used such language in their Table 1 summarizes how replies may
models although it is non-scientific in have functioned as instruction,but it is
its reference to "his eyes" and wanting difficult to determine their actual ef-
to look around (personifying rather fect. The tabulated categories simply
than objectifyingthe owl), in its lexicon describe replies'instructionalpotential.
("holes"),and in that it speaks directly The teachers'assignments,instruction,
to a reader ("Did you know"). With and modeling had a clear impact on
very few other models of science writ- children'smessages(e.g., they immedi-
ing except some selections from The ately appropriatedspecific text struc-
WeeklyReader,the jointly composed tures, scientific terms, or grammatical
messagewas the most salient classroom constructions the teacher modeled).
model to which childrenwere exposed. The families' replies, however, came
As the year progressed, instruc- after a message was written and were
tional discussions grew shorter, and not framed as instruction,so any effect

Table 1
Categories Describing Families'Instructional Scaffolding Replies
INSTRUCTIONALFEEDBACKTO CHILDREN'SMESSAGES 52% (n=164)
AskingQuestions 33%(n=104)
AcknowledgingImpact 11%(n=35)
AcknowledgingLearning 8%(n=25)
MODELINGGENRES 48% (n=152)
InformationalText 13%(n=41)
Jokesand Riddles 13%(n=41)
Narrative 9%(n=29)
MoralLessons 7%(n=22)
PoeticText 6% (n=19)
TOTAL 100%(n=316)
may have been subtle and therefore philosophy, all four children experi-
difficult to detect. Nevertheless, replies enced very similar curriculum and
were part of the instructionalcontext. instruction.
In order to permit a close analysis The children ranged in age from
of science writing in Family Message five to seven over the course of the
Journals,four case study children were school year with a mean age of 6.3
selected from among the larger group when they began first grade.Two six-
of 46. These three girls and one boy- year-olds were emergent readers and
Kristen, Kyle, Maryanne, and Sara- writers (pre-communicative spellers,
were nominated by the classroom uncertain of conventional letter sym-
teachers at years end. I asked them to bols), and two (one five-year-old and
suggest students who, as a group, were one nearlyseven-year-old)were begin-
representativeof the entire firstgradein ning readers and writers (semi-pho-
terms of writing ability, sociocultural netic spellers)asthe school yearopened.
background,and families'attitudes to- All four children lived in families with
ward involvement in school learning. at least one sibling and two working
Based on my own familiaritywith the parents, in occupations ranging from
students and classrooms, I confirmed construction work to investmentbank-
that this group reflected the full spec- ing; they represented the socio-eco-
trum of first graders. Three of the nomic diversity of their classrooms.
children came from one classroom,the Maryanne'sfamily was bilingual; they
fourth from the other. Because of the immigrated from Poland prior to her
teachers'close cooperation and shared birth.

Data Gathering and Analysis nomena (hereafterreferredto asscience).
Data include field notes from one-hour This percentage matches the teachers'
periods of weekly participant-observa- articulated concern with giving stu-
tion in one classroom from October dents much experience writing about
through May;interviews with the two science (see Table 2).
teachers,four case study students, and
four sets of parents;the four students' Selecting Data
year-long corpus of journal messages Once messageson science topics were
and their families'replies (524 messages identified, I considered the teachers'
and 512 repliestotal);and relatedclass- assignment and observed instructional
room artifacts (letters sent home re- guidance related to those messages. I
garding Family Message Journals and eliminated messagesthat they intended
other activities and expectations, a to 1) addresspersonalfeelings (e.g.,why
weekly class newsletter, and materials I like winter), 2) simply state that a
used in the study of message topics). science activity had been done and
The children's journal messages could be repeated at home (e.g., "Sci-
were the key pieces of data and the ence is lik magic let me show you"), or
focus of this textual analysis.All mes- 3) create a literary experience for the
sages were decodable, despite invented reader (e.g., poetry or a story about
spelling and some persistent letter re- weather).Though these may have been
versals,because of my participant-ob- related to science writing messagesby
server familiarity with the classroom, topic, these messageswere deliberately
curriculum, students, and individual not framed as science writing by the
message assignments. For the present teachers or students and thus are not
analysisI firstsorted messagesbased on germane to the focus on science genres.
their assigned topic. Over one third Nor were they examples of recontext-
(n=48) of the year's141 messageassign- ualization;ratherthey showed children
ments were directly related to science otherways to write about topics studied
or to health topics that involved learn- in science, reflecting the teachers'con-
ing about and exploring naturalphe- cern with integration across the cur-
Table 2
Message Topics
TEACHER ASSIGNMENTS MESSAGES COLLECTED
ScienceTopics 34% (n=48) 22% (n=117)

Non-Science Topics 66% (n=93) 78% (n=407)
TOTAL 100% (n=141) 100% (n=524)

riculum through thematic units that other two had lost some, and all four
engaged childrenin a varietyof types of had been absent occasionally.The 82
writing. Although such writing may messages accounted for 70% of the
make science topics seem more acces- messageson science topics (seeTable3).
sible, it is not scientific in content or
genre.In fact,it sharesfeatureswith the Analyzing Data
writing down to children that charac- Young children'swriting is often viewed
terizes many juvenile science texts, from the developmental perspective of
which is based on the assumption that emergent literacy (Chapman, 1995),
because young children cannot under- but my analysis focuses on emergent
standscience concepts or writing, con- andbeginning writers'appropriationof
tent must be presentedin more familiar conventional text genres. Moreover,
but less conventional ways (Halliday& whereaspreviousstudieshavelooked at
Martin, 1993; Rothery, 1996). informational texts or science writing
I did not eliminate messages that as global categories, I provide a more
the teachers intended to be science fine-grained look at a range of types of
writing but that in fact broke text writing used by scientistsin their daily
conventions. These messages, as dis- work and emphasized in science in-
cussedbelow,provide a crucialwindow struction.
into issues of genre development and In order to analyze whether and
error, recontextualization in light of how students appropriated and re-
task, and whether there is space for worked science genres, I drew on
flexibilitywhen working with the rela- previous researchers'elaboratedescrip-
tively rigid conventions of scientific tions of the featuresof scientific writ-
discourse. ing. These featuresinclude text structure
The analysisbelow includes all 82 and lexicogrammar. Structure,embody-
messages from the four case study ing the specialized functions of scien-
children, growing out of 35 science tific discourse, is realized through
writing assignments.Two of the chil- lexicogrammar, the linguistic choices
dren had savedall of their messages,the reflecting the relations between terms
Table 3
Messages on Science Topics
TEACHERASSIGNMENTS MESSAGESCOLLECTED
ScienceWriting 73% (n=35) 70% (n=82)

Other Genres 27% (n=13) 30% (n=35)
TOTAL 100% (n=48) 100% (n=117)

and concepts that characterize scien- development,and expectationsincrease
tific meaning (Halliday& Martin,1993; with experience and education level
Lemke, 1990). (Rose, 1998). Young children in par-
However, the most extensive de- ticular tend to rely upon spoken pat-
scriptions of science writing are based terns of language and organize texts
on analysis of the historical develop- aroundpeople'sactinginsteadof around
ment of scientific English used by passive,generic participantsor objects
scientists (Halliday& Martin, 1993) or (Halliday& Martin, 1993; Kamberelis,
the oraldiscourseof science instruction 1999). It is difficult to imagine a first
in secondaryschool classrooms(Lemke, grader who might compose the com-
1990). The oral and written discourse plex nominal:"Distinguishedfrom raw
of school science is necessarilydifferent eggs by their spin,hardboiled eggs ..."
from that of adult science. In the in place of Sara'smessage at the begin-
pedagogicalsituationteachersmustcon- ning of this article:"we chiride [tried]
cern themselves with classroom man- an a exsperument do you know the
agement, instructionalapproaches,and difrinsbetween a hardboiled egg and a
students' developmental readiness as ro [raw]egg ... the hardboild egg spins
well aswith science content. Addition- and the ro egg dunt."
ally, adult science texts may be too However, young children have at
alienating and densely packed, con- least an unfolding understanding of
ceptually and for
linguistically, children some features of science writing
to comprehend, appreciate,or emulate (Kamberelis, 1999), and elementary
(Christie, 1998; Lemke, 1990). For ex- studentsin grade two and above can be
ample, adult and even high school taught the grammaticalresources that
science texts are characterizedby ab- achieve scientific modes of organizing
stractionrealized through grammatical ideas and reasoning (Christie, 1998;
metaphor,creatingnominal groupsthat Rothery, 1989, 1996). In this study I
condense concepts, construct inter- analyzedfirstgraders'writing in light of
locking definitions, and objectify "ac- the findings of previous research on
tions, events and qualities"(Halliday& science genres in education.
Martin, 1993, p. 52; cf. Rothery, 1989). First,I categorizedthe firstgraders'
Nominalization reflects the thematic messagesaccordingto the science genres
content, the logical arguments,and the identified as most common in school
complex concepts of science (Halliday science textbooks:Report, Experiment,
& Martin, 1993; Lemke, 1990). How- and Explanation (Halliday & Martin,
ever, this grammatical reflection of 1993). Report,the most common sci-
scientific modes of thinking, arguing, ence genre in schools,functions to store
organizing,and presentinginformation factsby classifying,describingthe com-
may be too complex for young chil- position of classesand sub-classes,and
dren. The ability to comprehend and sometimes explainingrelatedprocesses.
produce grammatical complexity in It is what Lemke (1990) terms a major
scientific texts is linked to cognitive genre.Experimentis also a major genre

that includes Procedures
for and Recounts ports for 1) introduction or organizing
of inquiries into natural phenomena. statement and 2) description of at-
Experimentsfunction to exemplify and tributes and/or characteristic events,
test facts and are common in science activities,and processes.
textbooks and other instructionalma- Experiments, whether Procedures
terials.Explanationsare a minor genre or Recounts, are characterizedby four
or rhetorical structure,often included partsand were coded for these features:
in the major genre Report but also 1) aim, 2) method, 3) results, and 4)
regularly standing alone in science conclusions (Halliday& Martin, 1993).
textbooks to describe naturalprocesses. Explanationsmust be organized in
a particularlogical sequence reflecting
Text structure. In order to categorize a natural process. Like Reports they
the messages into genre categories, I include an introduction and a descrip-
analyzed each one for elements of tion of characteristicevents, activities,
conventional text structure.Reports in or processes.They were coded for 1)
science textbooks usuallybegin with an introduction, 2) description of charac-
organizing statement that may identify teristic events, activities, or processes,
where the topic fits in a class and/or and 3) logical sequence (Halliday &
what parts of the topic will be de- Martin, 1993).
scribed. The bulk of the Report is a
description of the topic's attributes Lexicogratnmar. Once messages were
such as appearance and behavior categorized for genre by their text
(Halliday& Martin, 1993). This struc- structure,I analyzed their lexicogram-
ture is similarto the more generalized maticalfeatures.Theselinguisticchoices
informational text structure identified work with text structure to realize a
by Pappas,Kiefer, and Levstik (1995): genre's intended function and charac-
Obligatory elements include an intro- terize writing as scientific. Previous
duction to the topic;a description of its researchhas looked at young children's
attributes;a discussion of characteristic science writing in comparison with
events,activities,or processesrelatedto other types of texts such as narrative
the topic;and a finalsummary.Optional andpoetry (Chapman,1995;Kamberelis,
elements arecategory comparisonsand 1999). In this study I focused insteadon
an afterwardof extrainformationabout how closely the first graders' texts
the topic. Science Reports in particular matched the conventional lexicogram-
maynot includea summaryor afterword, maticalfeaturesthat distinguishscience
but categorycomparisonsarefrequently genres(Halliday& Martin,1993;Lemke,
part of the classification function of 1990; Rothery, 1996). Reports and
Reports, often realizedin the introduc- Explanationsare found in other disci-
tory statement.Kamberelis(1999) found plines;however,not only their structure
that this text structure was likely to but also their content and lexicogram-
mark first graders' science reports as mar differ in science writing. The
opposed to other genres. I coded Re- lexicogrammar gives meaning to con-

cepts by indicating their logical rela- ing). Though I alone coded data, my
tionship and realizing the universal, year-long participant-observerfamil-
abstractdescriptions that characterize iarity with the students,classroomcur-
science. ricula,andmessageassignmentsuniquely
In general the lexicogrammar of positioned me to understandmessages
science, especially Reports, is charac- despite invented spellings,frequentlet-
terized by generic participants,timeless ter reversals,some reversalsin the order
verbs of abstract relation in simple of all lettersin a word creating"mirror-
present tense (shows, have), and a seri- writing,"and handwriting characteris-
ous linguistic register marked by spe- tic of beginners.Below I provide many
cialized, technical terms that represent examples of complete messages and
the lexicon of science. These qualities how I analyzed them, allowing readers
create a sense of universalityand cer- to judge the accuracy of my analytical
tainty and suggest that events and decisions.
processes are habitual. Writers must
avoid referringto individual people (I, Results
we, you) or animals and their material When analyzed for elements of con-
actions,avoidpersonification,and avoid ventional text structure,all 82 science
playfulor colloquial language (Halliday writing messagesclearlyfell into one of
& Martin, 1993; Lemke, 1990). the four genre categories:1) Report, 2)
Kamberelis (1999) found that young Experiment Recount, 3) Experiment
children'sscience reportshad a signifi- Procedure, or 4) Explanation.Table 4
cantly higher proportion of present summarizes the features of these cat-
tense verbs and specialized scientific egories and indicates how many mes-
terms than the narrativesand poetry sages fell into each one. It is important
they composed. to point out that the first-gradeteach-
Particulargenres have slight varia- ers and students did not label the
tions. For example,Experiment Proce- science genres as I do here;they simply
dure uses imperative verbs of action called all of the texts "science writing."
(place,mix, look) and Experiment Re- In thissection I discussthe structural
count includes specific rather than andlexicogrammatical featuresthatchar-
generic participantsin the method and acterized the children'stexts, revealing
results sections and past tense action how they consistentlyappropriatedcer-
verbs in the method section. Explana- tain conventionsbut not others.
tions, too, may include more action
verbs (replaced,eroded) than Reports Report- Text Structure
(Halliday& Martin, 1993).All messages The majority (60%)of the firstgraders'
within each genre category were coded science messagesfell into this category,
for these features:1) participants(ge- which is dominant in science educa-
neric or specific), 2) type of verbs, 3) tion. Nearly all of the children's Re-
lexicon and linguistic register (scien- ports included both elements of text
tific or colloquial, playful, personify- structure(seeTable 5).

Table 4
A Frameworkfor Analysis:Science Genres- Conventional Features
GENRE TEXT LEXICOGRAMMAR MESSAGES

STRUCTURE IN THIS CATEGORY*
Report Introduction Generic participants 60%

Description Simplepresenttense (n=50)
verbsof abstractrelation
Scientificlexicon
Experiment Aim Generic participantsin 20%

Recount Method Aim and Conclusion (n=16)
Results Specificparticipantsin
Conclusion Method and Results
Simplepresenttense verbs
but pasttense in Method
Scientificlexicon
Experiment Aim Generic participants 6%

Procedure Method Imperativeaction verbs (n=5)
Results Scientificlexicon
Conclusion
Explanation Introduction Generic participants 14%

Description Presenttense action verbs (n=H)
LogicalSequence Scientificlexicon
andnumberof firstgraders'
*Percent thatfit thiscategory.
82 sciencewritingmessages
Table 5
Percent of Reports with Conventional Features
TEXT STRUCTURE LEXICOGRAMMAR*
Introduction 98 Generic Participants 92

Description 100 PresentTenseVerbs 100
Science Lexicon 92
*Unconventionallexicogrammarwasfoundintheopeningsandclosingsofsomeof thesemessages,
butthese
sectionsweredistinctfromthesciencecontent.
Only one of 50 was missing an The Introductions, however, var-

Introductionto the topic andallincluded ied from very brief mention to state-
Descriptionof attributesand/or charac- ments that provided some organization
teristicevents,activities,and processes. for the text that followed. Some Intro-

ductions took the rudimentary form of snakeseat ather snakeslike gardinsnakesor
the message s greeting (e.g., "Grrrrrr" baby moles or sam snakes will eat baigs
to introduce "that bears can be ten feet [bugs]
love kyle
toll"). Most often the Introduction
directly addressed the reader: "Did you Less common were the most con-
know that Mars is cold," or referred to ventional messages that opened with
the individual writer, for example: Introductions providing clear organiz-
ing statements for the text, such as "the
February3, 1997 wind blows alawt of things" followed
Dear Mom + Dad
by a description of the many "things" it
I learned that whale sharksdon't eat peepel.
blows.
Sharks only atack when thay are mad or
The six Reports (out of 50) that
hungry.Baby sharksare called pups. Sharks
can see in the dark. Sharks eyes sumtims were composed in list form, delineating
glow in the dark. properties of the topic, also tended to
Love Maryanne include complete Introductions before
the list, for example:
The use of the message greeting as an
Introduction and the direct reference January 14, 1997
to the writer or audience are examples Brrrrr Mom and dad,
of how children appropriated a con- I can tell it is winter!
ventional feature but recontextualized l.The rabitsfiire turnes white. 2. It turnes
it to fit the Family Message Journal dark at 5:00 pm! 3. The birds fly south. 4.
You can go skaiting Love Maryanne
dialogue with a known, familiar audi-
ence. Also, as in Maryanne's message, As the examples above illustrate,
the first graders often introduced one most of the children's messages in-
aspect of their topic (what sharks eat cluded a Description that consisted of
and that they attack to get food) and attributes and characteristic events, ac-
then moved quickly to describing other tivities, or processes, though a very few
attributes. consisted of simply one attribute or a
More conventional Introductions single characteristic event as in, "Yo
were less common but accounted for mom did you know that baby teeth are
about one third of the texts in this les ten a inch tol?" Such messages
category. Some of these messages intro- tended to be written in the first half of
duced the topic through the Theme, or the school year, but variations in length,
opening clause, as in the following: conceptual and linguistic complexity,
and completeness of information oc-
December 4, 1996
curred throughout the year for all four
Pine needles are covered with wax and stay
children, depending upon the assigned
green in the winter.
topic, how much discussion and mod-
"Pine needles"identifies the topic around eling had taken place first, how much
which the message is organized, as does encouragement to say more occurred
"Snakes eat" in the following message: as the students were writing, and how

much time the children had to write. to specific people, yet these greetings
This variation suggests the powerful are not usually part of the science text
impact of the instructional context as itself.Sometimes they are,aswhen Kyle
well as individualinterestin a topic and used specific reference to his family
motivation to write on a given day. ("you")and himself ("I")to describean
owl attribute:
Report- Lexkogrammar
2/25/97
As noted above, the lexicogrammar of
the first graders' Reports was often whoo whoo Mom andDad,
markedby referenceto specific people; Owls might eat mice, fish,snakes,rabbits
andevenskunks!Do you knowwhy anowl
however, such reference was usually doesnot mindThe smellof skunksI know
reserved for the opening and closing love kyle
clausesof a message.Within the core of
On the other hand, the science
the message, students maintained the
content of messages most often stood
generic quality of participants (e.g.,
"snakes,""sharks,"and "the wind" are apartfrom the specific people referred
to at the end:
referredto generically in the examples
above) and avoided the spoken forms 1/28/97
"I,""we,"and "you" by which writers DearFamily,
make themselves or others actors in a Shadowsare made when light is blocked
text (see Table 5). Sometimes students' Shadowscanget biggeror
froma surfisace.
personal referenceswere even marked smallerchangeshaPemov CanYoumakea
Shadow Circleone LoveSara Yes No
asa differentpartof the messageor asan
aside or addition: Saraconsistentlytreatedher topic,shad-
2/26/97 ows, as generic.Typically,the firstgrad-
ers were able to use genre conventions
Hoo hoo Mom Dad andRosa
owl Pelitskumfromowls andthey kawfit flexibly to write science texts that were
uP the dayafterthey eat it and the things also engaging, audience-directed mes-
they kawfuP areteeth fer boansand tails sages.
andit cankawfuP Partsav fishsnakesand Only four of the 50 Reports did
skunksLovesaraDearmom DadandRosa not use generic language at all to refer
I sawwut wasinsighda owl PelitI wawnt
to go to the wuds to findeone Lovesara to their topics. One example centered
the text around the writer ratherthan
After the generic discussion of "owls" the topic, spring:
or a typical owl ("it") and habitual
3/24/97
processes characterizing the animal,
Sara added a new greeting and then Dear Mom dad and Rosa
shifted to self-reference. In a sense I like wen the flowerscome up in the spring
and wen the birds come back and in the
nearly all of the messagesmake such a morning wen I woke up I hird soem birds
shift because with few exceptions they outside my window and I looked out my
open and close with greetingsreferring window and I saw three blue birdsLove sara

Others personified animals or objects allyappropriatedthe languagethat they
ratherthan treatingthem asgeneric and learned during science lessons and that
abstract,making them seem individual their teachers used in model messages.
and capable of feelings, speech, and For example,while studyingwind they
other human qualities,as in the follow- learned the word blusteryand three of
ing examples:"patatoplantsgrow 20 to the four case study students used this
10 inches and they like to grow in sandy term in their wind messages.Similarly,
soil" and "After a fater bull frog has they generallyfollowed the instructions
tooken very good carof his bayshe says to include terms indicating scientific
yum I'll eat one of my babys I'm taxonomy.For example:
hungry." It is important to reiterate,
however, that messages like these that 4/30/97
did not follow the convention of ge- Dear Family
neric participantswere unusual. ScienceSentencesSpidersarenot insects,
Sara'sshadow message above is a they arearachnids. All spidersspinsilkbut
somedo not makewebs.Spiderseatinsects.
good example of the use of abstract, Someeveneatsmallmiceor fish.LoveSara
present tense verbs (e.g., "shadows are
made . . . shadows can get"), as op- There were rareexceptions to the
posed to verbs of material action, to use of scientific terms in students'
create a sense of the timeless,incontro- messages.For example ratherthan be-
vertible nature of scientific informa- ing technical, Kyle playfullycalled spi-
tion. Other examples above include ders vampires:"Did you know that a
"pine needles are covered . . . and stay spiter is a vapier caus it suks the Boold
green"and"it turnes dark."At the same from inects ..." In a messageabout how
time many of the children's messages wind is "usefull," Maryanne began:
did use verbs of material action (e.g., "The wind isn't allway horrible. . . "
"eat,""kawf"),butthey were alwaysin Such playful, emotional language,
the presenttense,signalingthe timeless, though not serious or technical enough
universal, or habitual nature of this to be scientific, is what makes texts
action. Past tense verbs were common interesting to readers. In fact, Lemke
only in the opening and closing of (1990) argues that its absence from
some messages(e.g., "I learned"or "we science texts may make science seem
chiride [tried] an a exsperument").As more alienating.Perhapsthe first grad-
noted above, such openings and clos- ers who engaged in this playful lan-
ings were often distinct from the scien- guage understood this and treated
tific content of the message, instead conventions somewhat flexibly to en-
serving an interactionalpurpose inher- gage their audience; or perhaps they
ent in the journal dialogue. just hadn'tfully masteredthe linguistic
The first graders were also quite conventions of science. In any case,
adept at using a scientific or technical emotional, playfullanguage was rarein
lexicon in their Reports. They gener- the first graders'Reports.

Experiment- Text Structure 5/21/97
Twenty-six percent of the firstgraders' DearFamily
science messages functioned to de- I cut my appleopen and I had two seeds
scribe inquiry that exemplifies or tests in myn they were kind of sllipery and
brownandshapedlikea raindrop andsome
facts. Of these, 20% were Experiment kids didn'thave a froot one of them was
Recounts (see Table 6) and 6% were Rani so I shearedwith her and afterwe
Experiment Procedures (see Table 7). foundthe seedsRani and me ate some of
Both genres share the same text struc- the appleandwe cut it the rightwayandI
foundthe starI didn'tevenknowLoveSara
ture: Aim, Method, Results, Conclu-
sion. Nearly all of the first graders' Only 2 of the 21 Experiments did not
Experimentsincluded anAim, although include an Aim.
this was often implied rather than Although Sara'sapple text is not in
explicitly stated or was stated near the conventional order and mingles per-
end of the text rather than at the sonal information with that directly
beginning as is conventional. For ex- related to the experiment, it includes
ample,describingan inquiry where the both Method and Results. Only one of
Aim was to find the starinside an apple the first graders' texts included no
and then investigate the seeds, Sara Method. But, like the Aim, this compo-
wrote the following: nent was sometimes implied or stated
Table 6
Percent of Experiment Recounts with Conventional Features
TEXT STRUCTURE LEXICOGRAMMAR
Aim 95 Generic Participants-Aim 13

Method 92 SpecificParticipants-
Results 95 Method & Results 100
Conclusion 33 Generic Participants-Cone. 60
PresentTenseVerbs
except in Method 44
Science Lexicon 19
Table 7
Percent of Experiment Procedures with Conventional Features
Aim 95 Generic Participants 60

Method 100 PresentTense,Imperative
Results 0 ActionVerbs 100
Conclusion 0 Science Lexicon 0

partiallyor indirectly,as in the follow- Only one Experiment Recount
ing: did not include Results ("I'm going to
askyou about if some things flote") but
5/14/97
the few Experiment Procedure texts
DearFamily
(only 6% of the science writing) all
I'm going to askyou aboutif some things included only Aim and Method. Their
flotedosea woodenstickfloattheiris more
I will do it with you at home you haveto focus was to describehow to achieve an
ges is a wooden stickfloatsand if a paper Aim. Because of the nature of the
clip floatsandif a cottonballfloats. . . Procedures being described, Results
were implied in the Aim:
It was uncommon for the studentsto be
as explicit about the Method as Sara 3/18/87
was in the following example: Dearfamil
can Mrs. carolanpick up a cup without
5/13/97 tuchingit shecouldty a Big not inThe cup
DearFamily and putThe not andput the stringthrew
We took aparta babyseedbutfirstwe took The top andpullThe string
off the seed coat andwe sawthe food and If the Method worked, the implied
the realthingit is calleda cotelydonandwe
Result would be liftingthe cup.Children
openedit up andfoundthe plant
wrote these messagesbefore doing the
LoveSara
procedure,so they had only hypotheses
Perhapsthe typicallack of explicit- and no concrete resultsto report.
ness in Method grew out of the first Finally,with respect to text struc-
graders' recognition that either their ture, as the examples above demon-
families would not be doing the Ex- strate,relativelyfew (only 33%) of the
periment at all, or that if they tried it, Experiment Recounts had Conclu-
the first grader would be there to sions (e.g.,"thingsthat arelite float and
explain how to proceed.The messages things that arehevy sink"),and none of
as assignedwere primarilymeant to tell the Experiment Proceduresdid.Addi-
families about an inquiry in which the tionally,whereas many of the Reports
children had engaged but not to in- had added endings addressed to the
struct families in how to do it. In children'sfamilies, such closings were
contrast,Experiments in science text- rare in Experiments. Only a few texts
books and other instructionalmaterials ended with statements such as "Im
are intended to teach what happens going to test you when I get home" or
when a clearly specified procedure is "it was so mutch fun I think I will do it
carried out (Recount) or to guide a at home."Perhapsthe genre of Experi-
student through an inquiry process ment seemed more bound to concrete
(Procedure).The journal context may procedures and findings and less open
have encouraged students to to personaladditionsthatmight distract
recontextualizethe genrestructurebased from its function- clear description of
on the function of their messages. inquiry into facts.

Experiment - Lexicogrammar pushes the other marble that hits the other
The conventional grammar character- marble the last marblewill be the won that
is going to be the won that roles off of the
izing the two types of Experiment texts roler and the other one that dinit fall the it
differs: Recounts feature specific (not gave the push to the other marblesto push
generic) participants in the Method the last won and made it tip over.
and Results sections and past (not
Kristen identified herself at the start as
present) tense verbs in the Method; the person doing the experiment and
Procedures feature generic participants
then referred to a specific group of
and many imperative verbs. Both genres
marbles ("the marble," "the other
feature a technical, scientific lexicon.
marbles," "the other one," "it").
The lexicogrammar of each genre in
As is conventional, when the first
the first graders' writing will be dis-
graders did include a Conclusion, it
cussed in turn.
usually referred to generic participants
In their Experiment Recounts the
(e.g., "things that are lite float and
first graders always referred to specific
things that are hevy sink") and used
participants in their Method and Re-
present tense verbs to indicate the
sults and usually did so in their Aim as
timelessness and universality of the
well (see Table 6). Although using
Conclusion. The convention of using
specific participants in the Aim is un-
past tense verbs in the Method was not
conventional, it is another example of so closely followed. When the Experi-
how the children recontextualized this
ment was finished before the children
convention to fit the journal context in
wrote, they used past tense verbs:
which all participants were familiar to
the audience.The Theme or opening of January24, 1997
the first clause in each text was gener- Wow Mommy
ally something like, "I did an experi- Today in the morning we went out side
ment to see ... ." or "We investigated and investegaited snow flaicks. Eatch one
was difrenteatch one hade six sids it was so
how ..." or "Mrs. Carolan tried ..."
mutch fun I think I will do it at home
Usually if an Aim was included, it was in Love Maryanne
this clause (e.g., "I'm going to ask you
about if some things flote" or "can Mrs. However, when the children wrote the
carolan pick up a cup without tuching Method and predicted Results first, as
in Kristen's rolling marble text, the
it"). A typical example is Kristen's text
about a marble experiment: verbs were in present and future tense
with only the final Result in past tense.
I think Another example is Maryanne's mes-
when the marblehits the other marblesthe
marbleswill role off of the roler when the sage:
it hits them and it will role all over the looe
May 9, 1997
[floor] it will hit five or four marblesor so
it mite Science Mrs.Wilensky has a pail 1/3 full of
water. She will swing it over her head. I
I proved proved that the water did not spill,the paill
If won marble hits the other marble that moved to fast.

She used present and future tense until objects (e.g., an apple or a paper cup
she reached"I proved." and string) and events (e.g., cutting the
Like the Recounts, Experiment apple,lifting the cup), so there was less
Proceduresdid not alwaysinclude ge- opportunity to teach scientific terms.
neric participants.As discussed earlier, Perhapsin their effort to make science
many of them functioned to tell fami- familiar by doing simple experiments
lies about a specific Procedure under- with everyday objects that children
taken by specific people ("I" or "mrs could easily label (marbles,apples,rul-
carolan") rather than to instruct ge- ers,cups,string,knives), the teachersdid
neric others how to do the Procedure. not introduce technical terms associ-
The convention of generic participants ated with the tools of science (the ruler
was thus recontextualized to fit the created an incline, the marble was a
context and function of a Procedure sphere,the string and cup contraption
told to families in a personaljournal. was a simple machine).
On the other hand, Experiment
Procedures were highly conventional Explanation- Text Structure
in their use of imperative,present tense Fourteen percent of the first graders'
verbs:"poack a hoal in the cup and put science writing functioned to explain a
sum schring [string]in the cup . . . and naturalprocess.The structureof Expla-
pick the schring up" or "takewet snow nationsis characterizedby two features:
and make one big boll av snow." an optional Introduction to the topic
Finally,with few exceptions (e.g., and a Description of characteristic
"cotyledon" or "investegaited"), Ex- events in Logical Sequence.The inclu-
periment Procedures and Experiment sion of characteristicevents organized
Recounts tended not to include the sequentiallyis essentialto Explanations'
scientific lexicon that was common in function: to describe how processes
the first graders'Reports. Rather these occur. Eight of the first graders' 11
texts used more colloquial language. Explanations included Introductions,
Their use of everyday language is not but most of these were rudimentary,as
surprisingbecause the firstgradeteach- might be expected of an optional
ers viewed these messages (and the element (see Table 8). Some of these
activitieswhich gave rise to them) as a named the topic in the Theme of the
way to teach scientific inquiry; they first clause:
were not as concerned with scientific
12-3-96
terminology in these texts and did not
ScienceInfo
regularly encourage students to use
technical terms, as they did when Whenpine conesdry,the seedsfallout and
scatter.New pine tressmaygrow.
discussingReports. It is also important
to note that the nature of the experi- OtherExplanationsintroducedthe topic
mental activitiesalso reduced the num- in the opening clause after "I learned
ber of technical terms that might be that . . . ,"and even more rudimentary
used. Procedures involved everyday Introductions took the form of the

Table 8
Percent of Explanations with Conventional Features
Introduction 73 Generic Participants* 100

Description 100 Present Tense Action Verbs 91
Logical Sequence 100 Science Lexicon 91
*Eighteenpercenthad specificparticipantsbut only in the opening clause,not in the processdescriptions.
message greeting, as in the following "When pine cones dry"or "if you nees
message, where "wigl" alluded to the [sneeze] on Sum Buty thenYou wil get
topic: the coald."Another common structure
was before-after:
1/8/97
wigl Mom did you know thatBabyteeth 3/6/97
Beginto fallout whena childis 6 or 7 years owlslook wise B Th are not as smartas
old theyarereplacedwith permanentteeth crowsor blueJays.Beforeit catchs[hatches]
a babyowl hasan egg ToothThe Toothis
As noted above in the discussion of on its Beak, the BaBy usesThe Tooth t
Reports, the message context allowed crackThe shallof The egg.
childrento reworkconventions,using a
In this example the "after"is implied.
greeting as an Introduction instead of
the more conventionalorganizingstate-
ment. Explanation- Lexicogrammar
Only one of the children'sExpla- Like conventional Explanations the
nationsincluded a full-fledged organiz- first graders' texts in this category
ing statement as Introduction: included only generic participants in
5/19/97 the processdescriptions(e.g.,"trees,""a
DearFamily child,""ababyowl"), though two of the
11 had opening clauses that addressed
Treeshelp us breath.And if you put a bag
overyor faceyou cansuficateanddie.And specific readers ("Did you know").
did you know we help trees breath we Some messages referred to "us" and
breathout carbondunocksideandgiveit to "you" in explaining processes ("Trees
the treesandthe treesgive us airto
LoveSara help us breath,""if you put a bag over
yor face") but the context suggeststhat
As the examples above show, the these terms were used generically to
firstgradersconsistentlyorganizedtheir mean "any person."
texts in logical sequence to describe The conventional lexicogrammar
how processes occur, and they used of Explanations is marked by more
many relationalclausesto indicate cau- action verbsthan arefound in Reports,
sality or temporal sequence such as where abstractverbs predominate,but

like Reports, Explanations feature the factualgenres . . . which have tradition-
presenttense.Presenttense action verbs ally been considered beyond the abili-
were prevalentin the firstgraders'texts. ties of primary school children"
These included "fall,""scatter,""grow," (Rothery, 1996, p. 107).
"nees [sneeze]," "breath [breathe]," At the same time, this finding and
"crack,""floss,"and"brush"Therewere those of previous researcherssuggest,as
far more action verbs in this category might be expected, that the abilities
than in any other science genre in students demonstrate may be closely
which the first graderswrote. tied to what they are taught.And what
Finally,Explanations,like Reports, studentsaretaught is restrictedby most
did tend to include many technical, teachers' very limited knowledge of
scientific terms such as "suficate,""egg genres and how they are achieved
tooth," "replaced," and "permanent structurallyand grammatically(Martin,
teeth"in place of more colloquial terms 1989, 1993; Rothery, 1989, 1996).The
common in the children's speaking first grade teachers in this study dem-
vocabulary like "stop breathing,""big onstratedthe lexicogrammarof science
teeth,"or "grown-up teeth." genres in their jointly-constructed
model texts but did not discuss these
Discussion conventions explicitly, except for en-
This study complements and extends couraging the use of a scientific lexi-
the findings of previous research.First, con. They were explicit about text
it broadens current perspectives on structureand that was directly taught.
earlyliteracyto look at young children's Nevertheless, students did learn to use
writing as socioculturalpracticewithin most of the conventional lexicogram-
a specific discourse community. Sec- matical features of the science genres
ond, it focuses not on self-selected consistentlyin their writing, suggesting
writing or emergent knowledge but on that perhaps repeated modeling, with
the teaching and learning of particular student participation in constructing
forms of writing from the start of the models, may be a powerful instruc-
formal education. Results show that tional tool; explicit instructionmay not
young children who are emergent and be as essential as other researchers
beginning writers can learn the basic suggest.Of course,it is impossibleto tell
components of science genres in a how much more conventional the stu-
context where they are guided to dents'texts might havebeen with more
compose such texts through teacher- directinstructionin lexicogrammar.Or
led joint construction of models fol- perhapsthey had learned enough about
lowed by independent practice. This conventions to be able to appropriate
finding complements the work of and deliberately recontextualize these
Christie (1989, 1998), Martin (1989, features, at times writing unconven-
1998, 1999), and Rothery (1989, 1996) tionally on purpose.
who have "demonstrate[d] that, with Furthermore, the population of
good teaching studentsareable to write students in this study,though ranging

from working-class to upper-middle- between the restrictions imposed by
class, may have learned more from genre instruction and the flexibility
models than children whose families within any genre. In fact, the teachers'
have less cultural and financial capital, instruction did not seem to turn the
limiting their access to the genres of studentsinto followers of rigid formu-
power in society.Additionally,despite lae. As in previous researchwith older
his or her family's financial situation, students (Martin, 1999), few of their
each of the four case study studentswas messageswere copied from the teacher's
being raisedin a highly attentive,rela- model. Over 90% of their texts were
tively school-oriented, comfortable original and each different from the
home. These children may have been texts written by classmateson the same
better positioned to learn from models topic, although some of the same
and use them flexibly simply because modeled language and structureswere
they felt entitled and supportedto learn taken up by multiple students.In many
and succeed in school and society. of their science texts, the children
Perhapsthe most significant con- treated the dialogue with families as
tribution of this study is the finding something separatefrom the science, as
that,basedon their learning of conven- when they included non-scientific
tions, children may establisha founda- opening or closing clausesaddressedto
tion from which to work in creating their audience within messages that
socially-valuedgenresbut may simulta- were otherwise quite conventional. In
neously develop the ability to impro- this way they wrappedtheir own voices
vise on thesegenreswith theirintentions andpurposesaroundtheirscience writ-
and their audience in mind. The first ing.
graders'not-quite-conventionalscience The firstgraderswere also adept at
messages may have reflected knowl- recontextualizinggenres in light of the
edge under construction but may also fact that they knew their families were
have demonstrated that the children not readingtheir messagesto learn how
were able to reworktheir genre knowl- to do an Experiment but ratherto find
edge to fit the context and audience. out what the students had done and
Family MessageJournalsprovided learned. Thus their Experiment texts
an interestingcontext for exploring the were the most unconventional of the
deliberaterecontextualizationof genre four genres in which they wrote. It
conventions to fit the context, purpose, seems that these young children were
and audience. The children were not capable both of using genre resources
only composing impersonal, abstract, flexibly to write texts that were recog-
and dispassionate science texts, they nizably conventional enough and of
were alsowriting these texts aslettersin recontextualizing genre in light of the
a dialogue with a familiaraudience that taskandits context.Thisfindingcomple-
they were trying to engage and impress ments and extends Martin's (1999)
with their knowledge. Thus, the Family discoverythatgenre recontextualization
MessageJournal embodied the tension is the norm in the writing of older

elementarystudents.It is difficultto tell child'stext to reflect all of the conven-
whether messages demonstrating in- tions of science texts written by adults,
complete or inconsistent knowledge of especially since the first gradersin this
genre features were intentionally un- study were concerned with communi-
conventional or simply reflective of cating in an interesting way to their
errorthat could be attributedto lack of audience; the stylistic conventions of
genre mastery.At the same time what science could potentially"impedecom-
may appearas deliberateapplicationof munication"in this context (Lemke,p.
genre knowledge may not always be 134).
intentional but rather serendipitous Finally,anotheraspectof education
(Martin, 1989; Rothery, 1989, 1996). also influences the learning of science
However,the deliberateaddressto their genres: the text models to which stu-
audience within their messagessuggests dents are exposed. The majority of
that the firstgraderswere intentionally science books the first graders read
reworking their basic genre knowl- were fiction related to science topics,
edge. though they also readnon-fiction peri-
Furthermore, as Christie (1998) odically.Moreover, the factual writing
points out, the act of teaching itself models children are exposed to are
involves some degree of recontext- often unconventional. For example,
ualization.The writing of school chil- popular science-related television pro-
dren should not necessarily look like gramsand books for childrenlike "The
the scientific texts of adult scientists Magic School Bus"series rely on many
because children'swriting occurs in a conventions from narrativenot factual
different institutional context where writing.Thosewho write science books
the primaryfocus is to learn a genre not and other materials for children (e.g.,
to use it to produce and share knowl- The WeeklyReaderclassroom newspa-
edge. Moreover,teachersmust focus on per) often hold a patronizing view of
teaching the content of science along their readers' inability to understand
with the text conventions for writing science genres (Halliday & Martin,
science. Additionally,teachers aware of 1993).Therefore,science texts for chil-
how students find science and its lan- dren often simplify the language and
guage alienatingoften try to humanize concepts and mix narrativeand factual
and personalizescience to make it more genres in an attempt to engage young
interestingand relateit to students'own readers (Rothery, 1989). Children's
lives. In the process of trying to make models, then, often are so oversimpli-
science less abstract,impersonal, and fied as to lack obligatory genre features
devoid of action, the language that and may in fact encourage highly
teachers use to representscience con- unconventional science writing in
cepts is inconsistent with genre con- which inanimate objects and animals
ventions (Lemke, 1990). Thus, because are personified, speak, and are given
of the nature of education, it may be human feelings, just as in some of the
unreasonable to expect any school first graders'messages that broke con-

ventions. In light of the models they ive of children's learning in general,
may have drawnupon, it is strikingthat they were also highly responsive to
the first graders'texts were as conven- their children's writing. Their replies
tional as the resultsshow. were consistently encouraging (e.g.,uI
liked your message"), and reflected
Limitations attention to the young writers' texts
This is a smallstudy,in two very similar (e.g.,"Thanksfor the information.I did
classrooms,with a focus on the writing not know that . . ."),and interestin their
of only four children. Therefore, al- school work (e.g., "That experiment
though it is suggestive,the resultscan- sounds very interesting!").Their re-
not necessarilybe generalized to other sponsivenessmotivated the firstgraders
classroomsand students.However, be- to write well (as the children reported
cause these resultsbuild on the work of in interviews and during observations).
other researchers,confirmingtheirfind- The fact that families replied regularly
ings about students'abilityto appropri- also showed that they valued writing.
ate genre conventions flexibly and Although researchshows that regardless
extending these findings to younger of income, ethnicity, or culture, most
children, this case study is significant. familiesare interestedin fostering their
Furtherstudies would help to establish children'seducational success (Barone,
the generalizabilityof the presentfind- 1999; Delgado-Gaitan, 1990; Paratore,
ings. Melzi, & Krol-Sinclair, 1999; Quint,
Because this study focused on the 1994; Taylor & Dorsey-Gaines, 1988),
children's texts and not the teachers' some familiesmay lack the resourcesin
instruction,there is little explicit atten- time, energy,and literacy skills to pro-
tion to precisely how the first grade vide such a responsive audience for
teachers taught science genres, aside children's writing. Because audience
from the fact that they consistently seemed to play a role in the children's
taught text structuresbefore children recontextualization of science genres,
wrote and modeled lexicogrammar these findings may not generalize to
through the jointly-constructed texts. contexts without an authentic audi-
The text focus is illuminating but the ence or where the audience is less
results would be more informative if responsive.
the children'swriting were more closely
linked to day-to-day data on exactly Conclusions
how the teachers instructed the stu- The findingsof this study suggestthat it
dents. Unfortunately, extremely fine- may be overly simplistic to fear that
grained data on daily instruction were genre instruction will limit young
not collected, but future studies are children's development of flexibility
needed that do revolve around the and power as writers. In general, the
instruction-composition link. firstgradersnot only appropriatedmost
Finally, the families of the case- of the basic features of four science
study students were not only support- genres, they also were able to rework

these genres flexibly in light of the task implied by dominant perspectives on
to situate their texts as letters in a emergent literacy.The belief that per-
dialogue to a familiar audience. As sonal ownership and self-expression
Hallidayand Martin (1993) argue,they mustbe foregroundedin teachingwrit-
treatedgenre conventions as"resources ing to children ignores the largersocial
for meaning" rather than "systems of context and functions for writing in
rules" (p. 22), and they balanced the society.Children, especiallythose from
dual social practicesthey were trying to non-mainstreamhomes, must learn the
realize through their texts: writing mainstream genres of power to gain
science and interactingwith their fami- access to culturalcapitalin our society.
lies. Whereassome genreswill be familiarto
Similarlyit may be overly simplis- children being raised in mainstream
tic to claim that young children cannot homes where conventional literacy is
develop their own voice as writers at valued and readingmaterialis plentiful,
the same time that they are systemati- science genresmay be so unfamiliarand
cally taught specific forms of writing. so rarelyencountered outside the class-
The first-grade teachers in this study room that unless explicitly introduced
had faith from the first day of school they will never be mastered.As this
that their students could deal with study demonstrates,ratherthan impos-
multiple demandsat once.They did not ing rigid formulae on children, learn-
assume that composing itself would be ing genres gives them the tools they
all-consuming or that young children need to act in and on the world
would be incapable of focusing atten- (Martin,1999). Children become criti-
tion beyond content and their personal cally literate when they realize that
intentions to consider simultaneously texts aresociallyconstructed,according
conventions of form. These teachers to genre conventions, to serve specific
assigned specific forms of writing to socialfunctions (Martin,1998;Rothery,
attain their curricular goals, and they 1996). Once students understandhow
modeled how to appropriate these text structure and grammar serve to
forms, preferring such teacher direc- "position readers"they have "the po-
tion to potentially haphazardexplora- tentialto challengetheirmeanings"and
tion.At the same time they ensuredthat the underlyingsocioculturalideologies
children were writing for a purpose:to that restrictaccess to power and recog-
communicate with a real audience. nize only certain modes of thinking
Results suggest that their teaching was and problem-solving (Rothery, 1996, p.
effective without restrictingflexibility. 118).
The children were able to appropriate Finally,since children seem quite
genre conventions for new purposes to willing and able to appropriate new
produce hybrid texts. genres, teachers must be aware of the
The earlyyearsmay be a time ripe sociocultural power relationshipsem-
for teaching and learning, not simply bedded in choices about what genresto
for self-directed experimentation, as introduce.The canonical texts of West-

ern science may empower children to (Martin, 1998). Teachers'and writers'
conform to social expectations, but choices reflectsocioculturalstancesthat
teachers may also choose to introduce are neither inevitable nor immutable.
other ways of thinking and writing As children are mastering forms of
about natural phenomena and help disciplinary writing, they can also be
children develop awareness that the developing the knowledge and power
way they perceive and expressideasand to question these dominant genres and
observations is only one way. Other generate new ones. This study shows
noncanonicaltypes of science discourse that even firstgraderscan treatconven-
might alsoprovidesocioculturally-con- tions as guides to be questioned, re-
gruent access to science for children worked, and redeployed for new
from nonmainstream backgrounds purposes.
Author Note
Direct correspondence to*Julie E. Wollman-Bonilla, Professor, Department of Elementary
Education, Rhode Island College, 600 Mount Pleasant Ave., Providence, RI 02908.
E-Mail: jwollman@ric.edu.
References
American Association for the Advancement of Science. (1996). Summarycomparisonof
contentbetween"National scienceeducationstandards"and "Benchmarks
for scienceliteracy"and
for all Americans."Washington, D.C.: Author.
"Science
Bakhtin, M. M. (1986). Speechgenres and otherlate essays.(C. Emerson & M. Holquist, Ed.;
V.W. McGee, Trans.). Austin, TX: University of Texas Press.
Barone, D. (1999). Resilient children:Storiesof poverty,drugexposure,and literacy.Chicago &

Newark, DE: National Reading Conference & International Reading Association.
Bazerman, C. (1997). Discursively structured activities. Mind, Culture,and Activity, 4, 296-308.
Bazerman, C. (1998). Emerging perspectives on the many dimensions of scientific discourse.

In J. R. Martin & R.Veel (Eds.), Reading science:Criticalandfunctional perspectiveson
discoursesof science(pp. 15-28). New York: Routledge.
Bourdieu, P., & Passeron, J. C. (1977). Reproductionin education,societyand culture.Thousand

Oaks, CA: Sage.
Britton, J. (1970). Languageand learning.Hammondsworth, U.K.: Penguin.
Chapman, M. L. (1995). The sociocognitive construction of written genres in first grade.

Researchin the Teachingof English, 29, 164-192.
Christie, F. (1989). Language development in education. In R. Hasan & J. R. Martin (Eds.),

Languagedevelopment:Learninglanguage,learningculture(pp. 152-198). Norwood, NJ: Ablex.
Christie, F. (1998). Science and apprenticeship: The pedagogic discourse. In J. R. Martin &
R.Veel (Eds.), Reading science:Criticalandfunctionalperspectiveson discoursesof science(pp.
152-177). New York: Routledge.

Cooper, C. R. (1999).What we know about genres,and how it can help us assignand
evaluatewriting. In C. R. Cooper & L. Odell (Eds.),Evaluating
writing(pp.23-52).
Urbana,IL:National Council of Teachersof English.
Cope, B., & Kalantzis, M. (1993). The powersof literacy:A genre approachto teachingwriting.
Pittsburgh,PA:Universityof PittsburghPress.
Delgado-Gaitan, C. (1990). Literacyfor empowerment:
The roleof parentsin children'seducation.
New York:Falmer.
Emig,J.(1977).Writing as a mode of learning.CollegeComposition

and Communication,
28,
122-128.
Frank, L.A. (1992).Writing to be read:Young writers'abilityto demonstrateaudience

awarenesswhen evaluatedby their readers.Research in theTeaching
of English,26, 277-298.
Gee,J. P.(1989). Literacy,discourse,and linguistics:Introduction.Journalof Education,171,
5-25.
Hall, N. (1998). Real literacyin a school setting:Five-year-oldstake on the world. The
ReadingTeacher,52, 8-17.
Halliday, M. A. K., & Martin, J. R. (1993). Writingscience:Literacyand discursivepower.

Pittsburgh,PA:Universityof PittsburghPress.
Kamberelis,G. (1999). Genre developmentand learning:Childrenwriting stories,science
reports, and poems. Researchin the Teachingof English, 33, 403-460.
King, M., & Rentel,V. (1979).Towarda theory of earlywriting development.Research

in the
Teachingof English, 13, 243-253.
Kirsch, G., & Roen, D. H. (Eds.) (1990). A sense of audiencein writtencommunication.Thou-

sand Oaks,CA: Sage.
Kress, G. (1999). Genre and the changingcontexts for Englishlanguagearts.Language

Arts,
76, 461-469.
Lave,J. (1988). Cognitionin practice.
New York:CambridgeUniversityPress.
Lee, O. (1999). Equity implicationsbasedon the conceptionsof science achievementin
major reform documents. Review of EducationalResearch,69, 83-115.
Lemke,J. L. (1990). Talkingscience:Language,learning,and values.Norwood, NJ: Ablex.
Martin, J. R. (1989). Factualwriting:Exploring and challengingsocial reality.New York: Oxford

UniversityPress.
Martin, J. R. (1993).A contextualtheory of language.In Cope, B., & Kalantzis,M. (Eds.),
The powersof literacy:A genre approachto teachingwriting (pp. 116-136). Pittsburgh, PA:
Universityof PittsburghPress.
Martin, J. R. (1998). Discoursesof science:Recontextualisation,genesis,intertexualityand
hegemony. In J. R. Martin & R.Veel (Eds.), Reading science:Criticalandfunctionalperspec-
tiveson discourses
of science(pp.3-14). New York:Routledge.
Martin, J. R. (1999). Mentoring semogenesis:"Genre-based"literacypedagogy.In F.Christie
(Ed.), Pedagogyand the shaping of consciousness(pp. 123-155). London: Cassell.

Martin, J. R., &Veel, R. (Eds). (1998). Reading science:Criticalandfunctionalperspectiveson
discoursesof science.New York: Routledge.
Martin, N., D'Arcy, P., Newton, B., & Parker, R. (1976). Writingand learningacrossthe
curriculum11-16. London: Ward Lock.
National Center on Education and the Economy. (1997). Performancestandards:English

languagearts,mathematics,science,appliedlearning:Vol. 1. Elementaryschool.Washington, D.C.:
Author
Pappas, C. C. (1991).Young children's strategies in learning the "book language" of informa-

tion books. DiscourseProcesses,14, 203-225.
Pappas, C. C, Kiefer, B. Z., & Levstik, L. S. (1995). An integratedlanguageperspectivein the

elementaryschool:Theory into action (2nd ed.). New York: Longman.
Pappas, C. C, & Pettegrew, B. S. (1998). The role of genre in the psycholinguistic guessing
game of reading. LanguageArts, 15, 36-44.
Paratore, J. R., Melzi, G., & Krol-Sinclair, B. (1999). What should we expectoffamily
literacy?Chicago & Newark, DE: National Reading Conference & International Reading
Association.
Quint, S. (1994). Schoolinghomelesschildren:A workingmodelfor Americas publicschools.New

York:Teachers College Press.
Rogoff, B. (1984). Introduction: Thinking and learning in social context. In B. Rogoff&

J. Lave (Eds.), Everydaycognition:Its developmentin socialcontext (pp. 1-8). Cambridge, MA:
Harvard University Press.
Rose, D. (1998). Science discourse and industrial hierarchy. In J. R. Martin & R.Veel (Eds.),
Reading science:Criticalandfunctionalperspectiveson discoursesof science(pp. 236-265). New
York: Routledge.
RotheryJ. (1989). Learning about language. In R. Hasan & J. R. Martin (Eds.), Language
development:Learninglanguage,learningculture(pp. 199-256). Norwood, NJ: Ablex.
Rothery, J. (1996). Making changes: Developing an educational linguistics. In R. Hasan &
G.Williams (Eds.), Literacyin society (pp. 86-123). New York: Longman.
Ryder, P. M.,Vander Lei, E. U, & Roen, D. H. (1999). Audience considerations for

evaluating writing. In C. R. Cooper & L. Odell (Eds.), Evaluatingwriting (pp. 93-113).
Urbana, IL: National Council of Teachers of English.
Schleppegrell, M. J. (1998). Grammar as resource:Writing a description. Researchin the

Teachingof English, 32, 182-211.
Sobol, D. J. (1963). EncyclopediaBrown, boy detective.New York: Bantam.
Taylor, D., & Dorsey-Gaines, C. (1988). Growingup literate:Learningfrom inner-cityfamilies.

Portsmouth, NH: Heinemann.
Wollman-Bonilla, J. E. (2000, April). What canfamily involvementcontributeto primary-grade

writinginstruction?Paper presented at annual meeting of the American Educational
Research Association, New Orleans.


National Council of Teachers of English

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

National Council of Teachers of English

Uploaded by

Copyright:

Available Formats

Teaching Science Writing to First Graders: "Genre Learning and Recontextualization"

Author(s): Julie E. Wollman-Bonilla

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

Sciencewriting is integralto the work of scientists and to learning to do science. Science

DearMom Dad andRosa writing. The topics of Family Message

Research in the Teaching of English • Volume 35 • August 2000 35

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

36 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

TeachingScienceWritingto First Graders 37

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

38 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

TeachingScienceWritingto First Graders 39

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

40 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

TeachingScienceWritingto First Graders 41

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

42 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

MODELINGGENRES 48% (n=152)

TeachingScienceWritingto First Graders 43

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

ScienceTopics 34% (n=48) 22% (n=117)

TOTAL 100% (n=141) 100% (n=524)

44 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

ScienceWriting 73% (n=35) 70% (n=82)

TOTAL 100% (n=48) 100% (n=117)

TeachingScienceWritingto First Graders 45

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

46 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

TeachingScienceWritingto First Graders 47

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

48 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

GENRE TEXT LEXICOGRAMMAR MESSAGES

Report Introduction Generic participants 60%

Experiment Aim Generic participantsin 20%

Experiment Aim Generic participants 6%

Explanation Introduction Generic participants 14%

TEXT STRUCTURE LEXICOGRAMMAR*

Introduction 98 Generic Participants 92

Only one of 50 was missing an The Introductions, however, var-

TeachingScienceWritingto First Graders 49

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

50 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

TeachingScienceWritingto First Graders 51

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

52 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

TEXT STRUCTURE LEXICOGRAMMAR

Aim 95 Generic Participants-Aim 13

TEXT STRUCTURE LEXICOGRAMMAR

Aim 95 Generic Participants 60

TeachingScienceWritingto First Graders 53

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

54 Research in the Teaching of English • Volume 35 • August 2000

This content downloaded from 188.72.96.115 on Wed, 25 Jun 2014 04:49:10 AM

TeachingScienceWritingto First Graders 55