University of Melbourne

This paper reports some findings from a five-year case study of bilingual literacy development (pre-school through the early and mid-primary years) in Victoria. Factual writing by a bilingual child allowed her to explore and negotiate factual (including "uncommonsense") information using both her languages. It also assisted in the child's developing control of written registers in both her tongues. We demonstrate this by examining the child's factual texts written in the two languages during her first four years in primary school (Grades Prep.-3).

Texts on factual topics have been selected out of all her written products which have been comprehensively collected over the five year period. The study utilises systemic functional analysis of the written texts (Halliday 1994), as well as genre and register theory (Martin 1992).

This paper analyses bilingual factual writing development in a primary-aged child. It is argued that factual writing allowed the bilingual child to explore and negotiate factual (including "uncommonsense" - Bernstein 1973: 99) information using both her languages. It also assisted in the child's developing control of written registers in both her tongues.

Until fairly recently, in Australia, the educational focus has been primarily on children's development of imaginative writing, whereas factual writing has traditionally been conceived of as unnecessarily cognitively demanding of young children, and therefore not introduced into the junior primary curriculum. Indeed, the studies carried out in the 1980s by Martin and Rothery (1981, 1984) found very few factual texts in a large body of primary students' writing in schools around Sydney. In his later work, Martin (1990) documents a situation where a factual text was in fact frowned upon by the teacher, as being not sufficiently "story-like". Martin argues that the factual piece was not judged in its own right, according to the linguistic criteria for factual writing. Martin (1985/1989) stresses that in the English-speaking culture, fictional and factual writing are different in their social purposes and linguistic realisation. Thus, fictional stories encourage the use of imaginative experience and creativity, whereas

The educational implication to follow is that exposure to story books and familiarity with narrative genres, often a feature of early childhood experiences, does not ensure development of an ability to handle factual writing. This awareness led to much of genre and register theorists' attention becoming focused on research into children's factual writing (Martin & Rothery 1984; Rothery 1989; Martin 1985/1989, 1990; Christie 1989; Christie et al. 1990; Veel 1997; Unsworth 1995) and exploring avenues along which teaching of factual genres can be most effective (Stead 1995; Sandiford 1997).

This paper seeks to highlight the significance of factual writing in the bilingual child's majority and minority tongues for her literacy and academic development. The questions I asked when setting out to investigate the bilingual child's engagement with writing on factual topics were as follows. What are the patterns of factual writing development in both of a bilingual's tongues? In what ways does exploration of factual information enhance the child's control of the written registers of language? Does the minority tongue play a role in the child's negotiating factual information? And if this is so, what is this role?

I found that the bilingual child used her both languages to negotiate factual information in school and family contexts. The different purposes for which the factual writing in the two languages was used obviously influenced the patterns of text types development across these tongues. In the majority tongue of English, the child's factual writing was developing to meet the demands of participation in the majority literate practices. Thus, there emerged a range of factual genres which are characteristic of the English speaking culture, and of the academic contexts of schooling in particular. Sometimes the minority tongue was used in parental scaffolding of English factual written genres. This is discussed elsewhere (Aidman, in preparation).

In the minority tongue, somewhat differently, the genres were much less articulate than the English text types. Minority factual writing was largely developing as a means of the child's exploration of factual reality and negotiating it in familial settings.

I also found that in writing on factual issues the child was learning to control the written registers in both her languages.

In section 2, I set out the methodological considerations which shaped this study of early bilingual literacy development. Some aspects of systemic functional text analysis are discussed. In sections 3 and 4, I will demonstrate the child's developing control of written registers in both her tongues by way of examination of her choices within the grammatical system of Transitivity. Section 3 will provide an overview of the major factual text types which the child wrote in English. It will be argued that, although some genres are embryonic, they nonetheless are examples of text types characteristic of the English-speaking literate culture. In section 4, I will look at the types of factual writing which the child wrote in her minority tongue. It will be shown that minority factual writing came to be used by the child as a tool for learning factual information. I will conclude in section 5 by comparing the patterns of factual writing development across the two of the bilingual's languages. Some implications for literacy pedagogy will be discussed.

The subject of this study is an English-Russian simultaneously bilingual child, Anna, who was born in Russia, and arrived in Australia at the age of three. Anna began mainstream schooling at age 4 in a fairly conservative Australian regional centre, which involved an almost exclusively monolingual environment, whereas communication in the family occurred in the minority tongue. Thus, the child's language situation can be characterised as minority home language maintenance without community support (Romaine, 1989: 167-168). This involved the linguistic support from the family coming predominantly in the minority language.

The researcher, who was the child's mother, had the advantage of knowing the broader context of the child's cognitive and emotional development. Anna was an active child who, when left to her own devices, would much rather prefer to dress up and role play as well as work with paint, glue and scissors than engage in literacy activities. However, she always willingly participated in the study, up to a point of offering her texts for analysis without being asked to. The interactions between the child and the parent usually occurred in relaxed home settings.

All texts which the child wrote in both her languages over the period of five years from pre-school through Grade 3 in school were collected and analysed using the Systemic Functional grammar (SFG - Halliday, 1994), and the genre and register theory (Martin, 1992). Register is understood here as being determined by the three features of the context of situation, such as field, tenor and mode.

Field refers to "what is going on, where what is interpreted institutionally, in terms of some culturally recognised activity" (Martin, 1992: 23), and it is commonly understood as "topic". Tenor reflects the negotiation of social relationships among the participants. Martin (1992:513) refers to mode as to "the role language is playing in realising social action", or "the extent to which a text constructs or accompanies its field."

The significance of the social purpose of text for its overall structure has been long recognised in linguistics (e.g. Gregory, 1967; Benson & Greaves, 1973; Halliday, 1985b; Hasan, 1984). In this paper we will refer to genres as conceptualised by Martin (1992) and Eggins (1996) who argue that genres are culturally determined; genres are realised through language; and that this process of realising genres in language is mediated through the realisation of register (Eggins 1996:34).

The way that genres are organised is not the result of arbitrary rules, but a reflection of the way that a particular language is organised to fulfil specific purposes. Each type of text, or genre, has its own particular language features which a writer must understand to produce a successful piece of writing. These include the overall structure, or organisation, with a particular ordering of the parts and specific grammatical features to realise the three language metafunctions (Martin 1992).

As has been stated, I aim to show, firstly, that factual writing stimulated the child's developing control over written language by providing real purposes and challenges, and, secondly, that the bilingual child's biliteracy development allowed her a flexibility in exploration of factual reality by using her both tongues.

In my analysis of the child's growing control of the grammar of written language, I will focus on her choices in the Transitivity system, as the primary grammatical resource of language that builds its experiential metafunction. I choose to focus on the child's learning to construct experiential meanings due to the primary purpose of factual writing, which is providing experiential information.

The linguistic system of Transitivity specifies the different kinds of processes in which people engage to get things done, as well as participants and associated circumstances (Halliday 1994: 106). Participants are realised in nominal groups, Processes are realised in verbal groups, whereas Circumstances are constructed in adverbial groups (adverbs and prepositional phrases). For example, in the following clauses the Processes (underlined), Participants (in bold), and Circumstances (in italics) are labelled (examples are taken from the child's texts):

I found them beyond the boundary of the fence.

Actor Process:material Goal Circumstance:location:place

Antephonte argues that all people are equal.

Sayer Process:verbal Projected clause (locution)

The mother loved her child.

Senser Process:affect Phenomenon

I think that a meeting should be held for the parents...

Senser Process:cognition Projected clause (idea)

The eggs were delicious.

Attribute Carrier Process:attributive Attribute

Communication is a way of contact.

Token Process:identifying Value

In the world there are stars.

Circumstance:location:place Process:existential Existent

The child's growing control of written language will be revealed in her expanding repertoire of Transitivity choices. Thus, by examining the child's choices of Processes, Participants and Circumstances, I will demonstrate that the range of process types and Circumstances used expands with the expansion of the factual fields explored by the child, and also depending on the text type which she constructs. It will be shown that Participants in factual texts tend from the start to be general rather than specific, with the emergence of abstraction, as the child grows older and develops as a writer. Technical, or subject specific, language emerges.

Factual writing in English was on many occasions child-initiated. Even in cases of teacher-organised activity, the choices of text type and linguistic means were made by the child. Over the first four years in school Anna wrote the following factual genres (many of which are of course rather embryonic instances of those): explanation, judgment, procedure, procedural recount, description, report and definition.

These text types, apart from description (which has no generic structure of its own), have been described in their more complete expression by a number of SF theorists including Martin and Rothery (1981, 1984), Martin (1985, 1990, 1993), Christie et al. (1990), Unsworth (1995), Rose (1997), Veel (1997), and Sandiford (1997). In my examination of Anna's texts as instances of particular text types, I will rely upon their descriptions of written genres. It is noted here that the process-oriented approach to writing development (Graves, 1983, among others) in the child's classroom resulted in the absence of explicit instruction in writing factual genres. Given this fact, and also the young age of the child, her texts are very simple, and often embryonic, examples of these.

In the examples of Anna's factual genres, which we will now consider, the elements of the schematic structure of her texts are underlined, with commentary on the linguistic choices made by the child provided after each example.

The genre of explanation has been described by Unsworth (1995) and Veel (1997) who analysed junior secondary writing. The child's early example of a sequential explanation (Text Efa 1), though of course very immature, has some features which make it possible to consider it an attempt at an explanatory text type.

Text Efa 1 "Water"(age 5:11)

The water comes to our taps.

The rain does that.

The water comes through the pipes.

The water comes through the cleaning system.

The water is always good.

It's lovely.

The water is just wonderful!!

The Explanation element intends (although not quite successfully) to construct a sequence of steps in an explanatory order. It uses material processes (underlined) to build material activities, as well as the simple present tense, which is also expected of this text type.

First factual judgments, of which Text Efa 2 is an example, were written by Anna at home on her own initiative.

Text Efa 2 "Water"(age 6:1)

Water is good for us

because it is fresh

and more than half of our body is made of water.

Structurally, Text Efa 2 consists of the Statement of opinion and Reasons elements, where the causal conjunction because is employed to build reasoning. As is characteristic of factual judgment (Martin 1985/1989), there are general Participants and the simple present tense choices, where an attributive process (underlined) is used to build opinion. Significantly, in the final clause, there is an attempt at construction of an "uncommonsense" field of knowledge, that of the human body composition.

Procedure writing in English was initiated by the teacher, although no modelling of the genre took place. Schematically, the child's text consists of Steps of the Procedure:

Text Efa 3 "Anzac Biscuits" (age 7:7)

(1) wash your hands.

(2) collect the ingredients.

(3) preheat the oven.

(4) make the dough.

(5) spray the trays.

(6) put them on the tray.

(7) put it in the oven.

(8) eat them!

As would be expected in the procedure text (Martin 1985/1989), there are imperative mood choices used in topical theme position (underlined). Temporal sequencing of actions is signalled by use of numbering. The processes employed are material, which is also characteristic of the text type.

The genre of procedural recount was described by Veel (1997) in his analysis of the junior secondary writing. Structurally, Anna constructs Text Efa 4 in accordance with the conventions of the genre, which has the elements of Orientation, Record of Events, and Conclusion.

Text Efa 4 "Science Works" (age 8:1)

On Monday 29th April grade 3V went to Science works.

We went to a science show.

The lady [[that put on the show]] showed us experiments with liquid nitrogen.

One of them was with a flower.

She put the flower in the liquid nitrogen

and then put it on the table.

The flower died.

She did the same thing to a squash ball

but <<when she warmed it up>>

it went back to normal.

And that shows that non-living things can go back to normal

but living things cannot

after being put in liquid nitrogen.

In her reconstruction of scientific experiences, the child makes the following linguistic choices which Veel (1997) finds characteristic of a procedural recount. Thus, the Participants being constructed are of an "uncommonsense" nature. Technical language is used to build the field: liquid nitrogen, living things, non-living things. There are past tense choices which construct the events building the experiment. The temporal sequencing of the events is foregrounded by textual and marked topical Themes (underlined)

In the Conclusion element, typically of a procedural recount, the scientific significance of the experiment is foregrounded. This is achieved by the use of anaphoric reference "that"(in bold) to refer back to the whole of preceding clause complex (Halliday 1994: 315) and by the emergence of verbal process of signification shows (that)... The simple present tense choices are also expected in the Concluding stage of a procedural recount.

"Report-like" texts emerged in Anna's Grade 2 writing. It is significant that from her earliest attempts at the genre, she learns to construct the Classificatory Statement element, which is the first stage of the mature report (Christie, et al. 1990).

Text Efa 5 "Dogs" (age 7:5)

Dogs are warm blooded mammals.

Dogs eat meat.

A dog is a pet.

Pups drink milk from their mothers.

Dogs can jump high.

Some people keep dogs to guard their homes.

[Speech bubble]

Roof!

The Classificatory Statement uses the relational attributive process to build classification. Participants in such a process construct technical terms (in italics), thus building the "uncommonsense" field of living species taxonomies. There is notably hardly any Circumstance used in Grade 2 reports, whereas in Grade 3 report-like texts, the Circumstance comes to contribute to constructing the field. There emerge such types as the Circumstance of comparison and Circumstance of manner:

Like all whales, it breathes in air with a special blow hole;

Circumstance:comparison Circumstance:manner

as well as the Circumstance of accompaniment:

Albatrosses cannot fly without a breeze.

Circumstance:accompaniment

Definition (Derewianka, 1995; Rose, 1997), emerging in the child's writing in Grade 2, came in Grade 3 to deal with "uncommonsense" fields of experience. Thus, in Text Efa 6, the field constructed is an abstract notion. The Participants constructing abstract notions are realised in abstract nouns (underlined). Some such nouns are technical terms.

Text Efa 6 "Erosion" (age 8:1)

Erosion is the gradual wearing away of the soil by wind, running water, waves and temperature.

Rabbits cause a lot of damage to the land.

[[Removing all the trees]] causes

the soil to become loose

and it is easily blown

or washed away.

To present processes as things, the child uses nominalisation (underlined italics). Importantly, Halliday considers nominalisation to be one of the features which characterises written language as different from speech (Halliday 1985/1994: 94).

There is use of relational identifying process is and relational causative processes cause, causes (in bold), to construct the field of "uncommonsense" knowledge.

The above text types constructed by the child in the majority language are examples (albeit embryonic) of written genres characteristic of the English-speaking culture, and not typically occurring in speech. The analysis of Anna's factual writing in English also reveals an increase in the range of process types, Participants, and Circumstances in her texts. There is a tendency towards the use of more abstract language which constructs increasingly less "commonsense" fields in her texts. These are all signs of the child's learning to control some of the written registers of language.

Anna's factual writing in Russian extended her developing control of formal registers of the minority language. It also allowed the child to explore "uncommonsense" fields of knowledge in informal contexts of the child's everyday life in the family. This may have promoted her interest in the fields of science and history and enhanced the child's academic learning in school. In this section I will present an overview of the principal types of Anna's factual writing in Russian between age 5:11 (when the first factual text was written) and 9:0. I will then demonstrate Anna's developing control of written registers on examples from her writing on the "uncommonsense" fields of science and history. The fields were chosen for a discussion here because Anna usually willingly wrote on these. I will display the original texts on the left hand side of tables, whereas translations will be represented on the right hand side.

The first factual text in Russian (a report-like piece on the field of science) was written early in Grade 1. At the beginning of Grade 2, the child was encouraged by the parent to write procedure texts, outlines and summaries of videos that she watched, and also a range of judgmental texts.

In the second half of the child's Grade 2, reading and writing on mathematics, science and history in the home language were introduced by the researcher. This was done on seeing the child's interest in these "uncommonsense" subjects. Writing on science and history usually occurred in the form of answers to questions posed by the parent, or selected by her from text books. The child usually enthusiastically responded to such written tasks. Indeed, once she started to deal with the fields of science and history in Russian writing in Grade 2, the child willingly continued to write on related issues in Grade 3 and beyond (in Grades 4 and 5, not discussed in this paper). Thus, between ages 7:5 and 9:0, the child sat down for at least 45 times, with about 257 clauses written on history, maths and science.

Text Rfa 1 begins with a general statement and then provides two simple taxonomising clauses using existential processes. Although the taxonomic information is very simple and "commonsense", the text reveals the child's awareness that one can use language to construct categories. The Participants are general, although concrete. There is no as yet Circumstance used.

With her emergent control of the factual fields of science and history, the child expanded her repertoire of process types, Participants, and accompanying Circumstances. The child's engaging with "uncommonsense" fields stimulated the emergence in her writing of such processes as those of identification and cognition, as well as verbal processes. The Participants in the processes building "uncommonsense" fields came to often be generalised, and later abstract, entities, similarly to Anna's texts on "uncommonsense" fields in English.

Identifying processes came to play an important role in building the fields of science and history in the child's writing, as will be seen from Text Rfa 2 below.

Text Rfa 2 (8:10)

A:

Matthiessen (1991) and Painter (1993, 1996, 1997) argue that the emergence of identifying processes in children's speech signifies their cognitive growth, as these are one of the three process types used to construct "symbolic processing" (Matthiessen 1991:84). The other two process types which Matthiessen and Painter identify here are verbal processes and processes of cognition. The term "symbolic processing" suggests the child's developing ability to use language for constructing experiences removed from her physical experiences of the world, or, as Painter (1996:102) would argue, to build "a second order representation of reality". According to Painter (1997), an ability to construct definitions is important for children's cognitive development and their academic success, as this appears to be linked to their further ability to deal with the increasingly abstract discourses of formal schooling. I have argued elsewhere (Aidman 1999) that constructing written definitions appears to be a challenge for many primary-aged students.

In Anna's factual texts, the Participants in identifying processes come to be constructed in nominalisation (Halliday 1994: 352), as in Text Rfa 3, where causal relationships are constructed in an abstract nominal group (underlined):

Text Rfa 3 (9:0)

A (a):

Congruently the idea could be constructed as follows (causal conjunction is in bold, whereas processes used instead of the originally employed abstract nouns are underlined):

A (b):

As is seen from example (b), the abstract nouns ïîðàæåíèå (defeat), and íåñîãëàñèå (disagreement) in the original example (a) would be more congruently constructed in (1) material and (2) attributive processes:

Abstract language in Anna's writing appears with the emergent exploration of the fields of science and history, coming to construct the fields of factual writing. This is to be expected in the exploration of "uncommonsense" information that is more demanding of the use of abstraction, which is a requirement for building the fields of science and the humanities (Halliday & Martin 1993). The child's attempts to meet this demand reveal her emergent understanding of the nature of the "uncommonsense" fields she engages with.

Anna's learning to define abstract notions in Russian reveals her developing ability to engage with abstraction using her minority tongue. This expands the child's opportunities to explore "uncommonsense" fields, to include both the classroom and familial contexts.

In factual writing in the minority language, cognition and verbal processes are found to play an important role in writing on fields of history. These processes emerge as part of an apparent process of the child's learning to attribute points of view to different people. The use of cognition and verbal processes significantly expands the child's linguistic potential in constructing experiential and interpersonal meanings, and thus in learning to control fields where both such meanings are foregrounded: that is, fields of the humanities.

The use of verbal processes and processes of cognition allows the writer to point to the source of information or opinion. In this function, cognition and verbal processes come to be used for projection of the child's, as well as third parties', ideas and locutions. Projection of the child's opinion by means of the cognition structure ß ñ÷èòàþ (I consider) emerges early in Grade 3 across the factual fields, including science and mathematics. The child's beginning to attribute opinions and sayings to other people emerges in writing on history. Here cognition and verbal processes come to include participants other than the child herself, and construe them as Sensors and Sayers. The child begins to simultaneously construct two fields, a field of factual experiences and that of interpreting them, with cognition and verbal processes contributing to the latter field construction. In Text Rfa 4 below, at one level there are opinions presented, whereas at the other level, there are constructed attitudes about those opinions.

Text Rfa 4

Q:

A:

In Text Rfa 4, cognition (underlined) and verbal processes (in italics) allow Anna to bring into discussion opinions of at least four participants (in bold): of the child herself, two Greek philosophers, and the generalised participant referred to as "those who want slaves". It is significant that verbal and cognition processes allow the writer to bridge the enormous distance in time and space that separates the writer from the other participants.

To sum up, the child's coming to use cognition and verbal processes in her writing on history allows her to instantiate her own and other people's sayings and opinions, as well as to express judgement of those opinions. This is important not only in texts on history, but in writing in the humanities more generally, and is in fact an expectation of much of academic writing.

As has been shown earlier, the child's early factual writing (Text Rfa 1) makes no use of Circumstances. In Grade 2 texts, there emerge Circumstances of location: place (underlined in examples from Text Rfa 5 below) and of location:time (Text Rfa 6), but no other types of Circumstance are found:

Text Rfa 5 (age 7:5)

Q:

A:

Text Rfa 6 (age 7:5)

Q:

A:

The experiential information constructed in the above Circumstances allows the child to build the fields of her inquiry as she discusses planets and history.

Circumstance of manner (instrument) emerges only in Grade 3 texts, and is realised in an adjunct as well as in a prepositional phrase:

Text Rfa 7 (8:0)

In texts on history, there emerge some choices of Circumstance which are more characteristic of the written register than speech:

Text Rfa 8 (8:11)

Q:

A:

In Text Rfa 8, the choice of the Circumstance of manner realised in a prepositional phrase íå ïî ñâîåìó æåëàíèþ (through no will of theirs) is a typically written register choice. The more congruent realisation of the meaning constructed by this Circumstance would be the mental process of volition îíè (ýòîãî) íå õîòåëè (they didn't want (it)). The use of an abstract noun in a prepositional phrase constructing the Circumstance of location: place â òàêîì ïîëîæåíèè (in such a situation) is also the child's attempt to use a more "written" register choice. Thus the whole hypotactic clause

would more congruently be constructed as follows:

Circumstance of condition emerges in Grade 3 writing where the child constructs it in an abstract nominal group:

Text Rfa 9 (8:11)

(a)

The idea constructed in the Circumstance in the above example would be congruently realised in a hypotactic clause (underlined in version (b) below):

(b)

The Russian spoken language would gravitate towards the more congruent choices (example (b)). I will thus conclude that the child's choice of abstract nominal groups to construct historical realities and their interpretation reflects her growing awareness of registers of the written language which tends to deal in abstraction to a much greater degree than every day conversation.

Overall, I have demonstrated that the child was successfully using her minority tongue to explore some "uncommonsense" subject areas. It has also been shown that engagement in literacy activities on the fields of science and history enhanced the child's exploration of the written registers of her minority language.

In conclusion, I hope that I have demonstrated at least two things. Firstly, the range of factual genres and their articulation is greater in the majority language, most likely due to expectations of factual text types in school and in the English-speaking society into which the child is being socialised. Factual text types constructed by the child in the majority language are examples (albeit embryonic) of written genres characteristic of the English-speaking literate culture.

Somewhat differently, factual writing in the minority tongue came to be used by the bilingual child as a means of exploring factual fields, particularly in familial contexts.

Secondly, in both tongues, factual writing enhanced the child's learning to control the written registers of language. Thus, in English, the texts types which she constructed are characteristically written language text types, not typically occurring in speech. In both languages, the child was stimulated to expand the repertoire in the system of Transitivity as the major linguistic resource used to construct the field in texts. I have shown that many choices which Anna made were formal, or written register choices, different from those occurring in everyday conversation. I have commented upon the significance on the emergence of relational processes to build classificatory and identifying relationships between Participants in the clause. I have highlighted the use of verbal processes for signification, and also of verbal and cognition processes to construct people's opinions. There was noticeable movement from concrete and specific Participants towards generalised and more abstract ones across the two languages. With her growing maturity as a writer, the information typically dealt with in factual writing tended to be increasingly of a less "commonsense" character than the child would usually use in speech in either tongue.

One of the implications of this study for literacy pedagogy is the desirability of using both of a bilingual's languages in learning across the subject areas in school. By developing a bilingual's minority literacy, there increase opportunities for his or her drawing on the linguistic resources of the family, and, as a result, opportunities for his or her academic development across the contexts of both school and home.

Another implication for literacy pedagogy is that factual writing significantly expands the child's opportunities for developing control over an increasing range of registers of language, particularly its written registers.

Aidman, A. (in preparation) Early Bilingual Writing: Some influences of the mother tongue on the majority written genre learning.

Benson, J.D. & Greaves, W.S. (1973) The Language People Really Use. Agincourt, Canada: The Book Society of Canada.

Christie, F. (1989) Genres in writing. Writing in Schools: Study Guide. Geelong: Deakin University Press, 3-48.

Christie, F., Gray, B., Martin, J., Macken M., Gray, P. and Rothery, J. (1990a) Exploring Reports. Sydney: Harcourt, Brace, Jovanovich.

Christie, F., Gray, B., Martin, J., Macken M., Gray, P. and Rothery, J. (1990b) Exploring Reports: Teachers Manual. Sydney: Harcourt, Brace, Jovanovich.

Derewianka, B. (1995) Language development in the transition from childhood to adolescence: the role of grammatical metaphor. Unpublished Ph.D. dissertation. Sydney: Macquarie University.

Eggins, S. (1996) An Introduction to Systemic Functional Linguistics. London: Pinter.

Graves, D.H. (1983) Writing: Teachers and children at work. New Hampshire: Heinemann.

Gregory, M. (1967) Aspects of varieties differentiation. Journal of linguistics, 3, 177-198.

Halliday, M.A.K. (1985a) Spoken and written language. Deakin University, Victoria: Deakin University Press.

Halliday, M.A.K. (1985b) Language, context and text. In M.A.K. Halliday & R Hasan (1985) Language, context and text: Aspects of language in a social-semiotic perspective. Deakin University, Victoria: Deakin University Press.

Halliday, M.A.K. (1994) An introduction to functional grammar. London: Arnold.

Halliday, M.A.K. & Martin, L.R. (eds.) (1993) Writing Science: Literacy and Discursive Power. London: Falmer Press.

Martin, J. R. (1985/1989) Factual writing: Exploring and challenging social reality. Deakin University, Victoria: Deakin University Press.

Martin, J.R. (1990) Literacy in science: learning to handle text as technology. In F. Christie (ed.) Literacy for a changing world. Hawthorn, Victoria:ACER, 79-117.

Martin, J.R. (1992) English text: system and structure. Amsterdam: John Benjamins.

Martin, J. R. (1993) Technicality and abstraction: Language for the Creation of Specialised texts. In M. A. K. Halliday & J. R. Martin (eds.), Writing Science: Literacy and Discursive Power. London: The Falmer Press, 203-220.

Martin, J. & Rothery J. (1981, 1982) Writing Project Report , 1 and 2. Department of Linguistics. University of Sydney.

Martin, J. & Rothery J. (1984) Choice of genre in a suburban primary school. A paper given at the Annual Conference of the Applied Linguistics Association of Australia. Alice Springs, August 1984.

Romaine, S. (1989) Bilingualism. Oxford: Basil Blackwell.

Rose, D. (1997) Science, technology and technical literacies. In F. Christie & J.R. Martin (eds.), Genres and institutions. London: Cassell, 40-72.

Rothery, J. (1989) Exploring the written mode and the range of factual genres. Writing in Schools. Study guide. Geelong: Deakin University Press, 49-90.

Sandiford, C. (1997) The genre of explanation in the primary curriculum. An unpublished M.Ed. dissertation. Parkville, Australia: University of Melbourne.

Stead, A. (1995) Report genre writing in the primary curriculum. Unpublished M. Ed. project. Parkville: University of Melbourne.

Unsworth, L. (1995) How and why: Recontextualizing science explanations in school science books. Unpublished Ph.D. dissertation. Sydney: University of Sydney.

Veel, R. (1997) Learning how to mean - scientifically speaking: apprenticeship into scientific discourse in the secondary school. In F. Christie & J.R. Martin (eds.) Apprenticeship into scientific discourse. Cassell: London, 161-195.