Debates and discussions





Dr M Durie 7/7/94
Dept of Linguistics
University of Melbourne
Parkville, Vic 3052



Dear Mark,

Fran Christie passed on to me a copy of a draft of your article for The Age. I wanted to make a few comments on your article, and pass on to you some papers and references which you might find relevant. I do hope we have time for a chat in Melbourne (I'm down for ALI/AALA, from July 8 to 17) - and I'll look for you at the meetings. I don't believe we've met, but I'll get someone to point you out.

To begin, let me say that I welcome your interest in grammar for schools. Since I began working in the field some 15 years ago, I've not encountered much interest among non-systemic linguists - other than Huddleston's review of the decadent traditional grammars in use (ALAA Occasional Papers). Certainly linguists are thin on the ground at Australian Reading Association, Egnlish Teaching Association or Primary English Teaching Association conferences; and the several Language in Education conferences organised by Fran Christie, the three LERN conferences and the annual Australian Systemic Workshops and attendant 'summer' schools have not drawn many subscribers from among the many concerned professional linguists around Australia you suggest you represent (on the other hand, they have involved critical linguists like Hodge and Kress, and sociolinguists like Horvath and Clyne).

As for your article itself, I take it that you wish to argue that getting some grammar back into schools is a good idea, but that systemic grammar is the wrong grammar to put back. I'm not very clear about what you are offering in its place, or what exactly you expect your Australian Academy of Humanities or your linguistics departments in Melbourne to come up with. For purposes of this letter, I'll assume you mean something like the grammar represented by Huddleston in his more recent work on English grammar. I should perhaps say at the outset that I don't think your article will achieve what you have in mind. Rather, it will be used by the 'progressive' education lobby in Victoria to argue that linguists are hopelessly divided and that systemic grammar has no credibility within the discipline of linguistics - so why should educators pay any attention to linguists (including educational linguistics like Fran Christie). I think that serving reactionary interests of this kind is a step backward, and I regret the damage you may cause - both in terms of improving literacy outcomes for students in school and for the value the community at large might otherwise come to place on professional linguists for what they can give in a practical applied sense.

Now let me pick up a number of points in your article:

You suggest that grammar is back, but in a "particularly narrow form"; I suppose you mean 'narrow' in the sense that SFL is the main influence, rather than a range of mainstream theories. But do teachers really want an eclectic approach? Won't that just make the terminological load that concerns you even worse? And in any case, from another perspective, SFL is not narrow - as far as I am aware, it's coverage of strata (including context) and metafunctions is a lot more extensive than anything attempted in other models, which prefer to carve off a piece of the puzzle at a time, setting aside other aspects of language for future research or other disciplines. I also think some sense of history is also important here. SFL won its first purchase in education through its genre theory and then attempted to draw educators into grammar from there; this means that the grammar to be consumed needed to be a broad semantically oriented one, to help in genre analysis. I would suggest that the kind of formal grammar you are promoting would not be so valuable in this respect.

You describe SFL as a "theoretical model...for describing English texts"; this is a very misleading characterisation. Historically, the model was first developed and used for Chinese, then adapted to English - in the general context in the 50s of work on prosodic phonology across a range of languages. There is an unbroken history of descriptions of other languages, beginning with Halliday on Chinese in the 50s, Huddleston on French, Barnwell on Mbembe and Hudson on Beja in the 60s (including some prosodic analysis of an Australian language by Dixon). Last semester at Sydney I continued supervising an SFL PhD on Pitjantjatjara grammar (which I believe develops our understanding of the language in ways that complement Goddard), marked an MPhil thesis on conjunction in Chinese, participated for several weeks in post-grad research seminars concerned with French, Chinese, Japanese and Tagalog (theme and transitivity), read Mark Harvey's review of McGregor's book on Gooniyandi (which Harvey comments is "of interest... challenges commonly held analyses and assumptions... at its strongest and most stimulating... illuminating analysis... stimulating account... it seems likely that functional analyses of clausal and textual organisation, similar to that of McGregor, will prove fruitful more generally within Australianist linguistics... many interesting challenges...In overall terms McGregor's description of Gooniyandi is a stimulating and well-presented contribution to Australianist studies. It raises many issues for general consideration, including for those who do not share its theoretical orientation" Australian Journal of Linguistics 12.2 329-335), consulted with Matthiessen's multilingual text generation project (currently involving English, Chinese, Japanese with planned extensions to a range of other languages), consulted with a student from a linguistics department in Melbourne desperate for help with SFL work on Indonesian... I could go on (not to mention a long list of the many theses written on languages other than English in my department, some supervised by linguists like Jane Simpson and Michael Walsh with whom you might find it productive to discuss the contribution that SFL can make to descriptions of various languages). For your point to have any substance, you have to demonstrate that SFL can only be applied to English (which I know you know not to be the case) or that it can only be applied unfortunately - which I would challenge you to demonstrate to me (beginning perhaps with my own work on Tagalog?).

You write that SFL has "made inroads", is the "dominant paradigm", is a "new educational orthodoxy", simply because it has "filled a vacuum"; that's not my reading of the history. Rather what I recall is 15 years fighting against the grain to get anyone to pay any attention to language at all - and gradually winning people over by drawing their attention to generic structure, and then on to the ways in which grammar construes the social function of genres, stage by stage. In all this, the linguists you presume to represent played a negligible role, so it's hardly surprising that their work is not now on the agenda.

I would also object to your use of the term 'orthodoxy' (I will take up its ideological implications below). SFL comprises a range of voices theoretically (with considerable differences among say Sydney, Cardiff and Toronto); and a range of voices is also present in applications - the Queensland English syllabus differs in many ways from the NSW one, and the influence of SFL on NT and WA is markedly different again. It seems to me that SFL has thus demonstrated that it can adapt flexibly to a variety of consumer's concerns, including a kind of blending with less systemic linguistics as in Queensland and various degrees of detail as far as terminology goes.

You suggest that SFL involves "proposals adopted with little critical analysis". Again, that's not my reading of the history. The work in general has been subjected to intense critique from i. proponents of whole language pedagogy (see Reid 1987), and ii. from contemporary critical theory (feminist and post-structural varieties in particular - the Christie report represents a synthesis of the latter negotiations). SFL was only accepted because it demonstrated it could do a politically responsible job as far as redistributing literacy resources is concerned. My experience is that educators have thought very deeply and very critically about what is going on.

One very good example of careful critical examination is Queensland where the key players had studied both with Huddleston and Halliday. You will find the influence of both linguists in their syllabus, as I noted above. Steady evolution in the direction of SFL in Queensland had to do with a range of very practical considerations, not with filling a vacuum.

You state that "far from renaming or replacing traditional grammatical terminology...it simply adds a complicated and arcane terminology...such as 'fractalization, ideational metafunction, rank, syndrome' - on top of older forms". To begin, all linguistics terminology is in some sense 'arcane' to most teachers; there's nothing special about systemics in this respect. Second, in which materials for teachers have you run across the terms 'fractalization' or 'syndrome'? In my experience, terminology has been very judiciously edited and adapted before being used with teachers (consider for example HBJ Language - A Resource for Meaning series, or the NSW Board of Studies Handbook of Grammar`).

I wouldn't myself agree with your use of arcane ('secret, hidden') to describe SFL terminology. First of all, its purpose is not to exclude, but to provide tools. And second, people have taken great pains over the years to introduce the terminology gradually, step by step in accessible material (see again the HBJ Language - A Resource for Meaning series, the Derewianka PETA materials, the NSW Board of Studies materials). In addition, annual week long summer courses for teachers have been in operation since 1989, moving around Australia, precisely to facilitate access to the model. Perhaps this is just a question of reading position. When I think of 'arcane' I think of examples such as the use of 'theme' (following Gruber) by MIT linguists (contra SFL usage and contra the Prague School usage for describing information flow) to refer to a case relation, and subsequently to refer to all case relations as thematic relations (contra Fillmore, Halliday, Starosta or anyone else working on case). This kind of ethnocentric scholarship (sic) does seem designed to exclude.

You say that SFL "doesn't replace verb with action process...it uses both sets of terminology at the same time". Well, what are the alternatives, if we agree that we need more than traditional word class labels? My understanding is that we either need more labels or more structure. If you don't want to label the Subject of a clause, for example, then you need a more complicated tree that will allow it to be defined as the NP immediately dominated by S (or however the relevant trees now happen to be labelled and designed); similarly if you want to show the difference between the Epithet red in a red glass and the Classifier red in red wine, then either you label the functions differently, or introduce structural differentiation (or do nothing structurally and place the difference somewhere else in the model - features in the lexicon or whatever). I take it you prefer more structure (and possibly multiple levels of analysis) to more labels. If so, you will have to train teachers to do much more complex structural analysis than that involved in SFL - a problem you have not addressed. In addition, you have to consider whether this kind of complex tree drawing is more efficient than labelling as far as text analysis is concerned - which is the main thing teachers want their grammar to do for them (as part of assessment, evaluation, explaining text structure to students etc.). In our experience labelling functions is quicker and more transparent semantically.

You suggest that "functional grammar is a very misleading label to use for" for SFL. But who owns the name? Halliday deploys the label from the late 60s to bring out the continuity between his perspective and that of the Prague school, which he has acknowledged in several articles; he calls the model SFL throughout the 70s and 80s and published his grammar as an Introduction to Functional Grammar , not Introduction to SFL, because it didn't include systemic representations. It's thus hardly surprising that SFL has termed its approach to grammar in schools functional grammar in opposition to traditional school grammar on the one hand and the formal grammar you represent on the other. I don't find the label misleading in an educational context (nor a linguistics one for that matter).

You suggest that SFL "appropriates traditional grammar terminology wholesale to describe parts of speech". If so, why doesn't SFL use participle or gerund or coordination or subordination? In any case, if we consider Halliday 1985:191, is it really the case that traditional grammar classifies prepositions as verbal or determiners as nominal?

You suggest that SFL has a more "unquestioning acceptance of traditional grammar and its terms" than other models. I don't agree that there is a significant different in this respect. Class labels in the tradition you speak for come from Harris 'From morpheme to utterance' Language 22. 1946. In spite of the abbreviations, we know that N stands for noun, V for verb, A for adjective, P for preposition, Ad for adverb and so on. The frames Harris uses to define many of these classes are quite compatible with SFL usage and indeed the paradigmatic reasoning behind them.

You speak of "traditional grammar's terminological overkill"; again here, you're not addressing the real issue, which is how information is represented in analysis - lots of features, lots of functions, lots of structure, lots of levels or what? You have to represent information somehow, and in education one primary goal is representational tools which can be used efficiently for text analysis (in which case I find complex structures or complex feature lists are not on).

You repeat that SFL has "relatively little application beyond English" and that SFL practitioners are "...among the most Anglo-centric group of linguists". I've commented on this above; let me add here that history is also important. Many SFL people follow Halliday in treating linguistics as an ideologically committed form of social action; in this light, the reason so much work has been done on English is because this is the culture they are trying to change. Again, it remains for you to demonstrate that the model cannot be used in similar ways in other cultures. Just because a model hasn't been so used to do something doesn't mean that it can't be.

You suggest that the "complex terminologies of Systemics will be of little use to school students learning other languages". The issue here is difficult to arbitrate. The more specifically English one's description is, the more use it will be for English and presumably the less useful for other languages, to the extent that they are different from English. This is simply a dilemma, not a reason for rejecting a rich and finely tuned description of any one language. In any case, our experience working across languages at Sydney is that there are always areas where comparable descriptive terminology can be retained (e.g. Theme for Tagalog ang phrase and Japanese wa phrase). In general, I think I prefer a model which uses a general theoretical apparatus to describe languages in their own terms, even where this means different labelling (the Whorfian in me, I suppose).

You forecast that "when students go on to study linguistics at university they will also find that these terminologies have little currency". We're talking now about the year 2000 onwards for students currently in primary or secondary school and you certainly seem to be more confident than I am about the future of linguistics teaching. But, surely it would be fair to hazard a guess that with Chomsky's retirement around the turn of the millennium, things will open up as far as linguistics is concerned, in particular to make room for theories that have immediate and short-term pay-offs for the wider community of interests universities are now expected to serve. Systemics has a growing currency as far as linguistics teaching in universities is concerned, a great deal of which goes on outside of linguistics departments, and is now taught at Edith Cowan University, Murdoch, Wollongong, Macquarie, Sydney, UTS, UNSW, UWS, QUT and NTU - and isn't it the case that in Victoria students will study linguistics with Kamler (in Education at Deakin), Threadgold (in English at Monash) or Christie (in Education at Melbourne)? In any case, is it really your position that no grammar at all is better than systemic grammar - given that you are unlikely to find a ready market for the kind of grammar you want to sell?

You say that "many professional linguists...are concerned about this trend"; if so, I certainly hope they will speak out and publish their concerns so that they can be addressed. There's lots to be negotiated, and more interest would be welcome.

You state that SFL is "a side-current...and one of the more arcane ones at that". This is not a reason to believe that mainstream linguistics will be more relevant to education, since mainstream linguistics has not evolved with educational concerns in mind. SFL has evolved with applied concerns in mind, so is likely to be relevant.

You suggest that "after earlier periods of discipleship under Halliday" Huddleston and Dixon "renounced the Systemics orthodoxy". First let me object to the religious discourse you are using to characterise SFL - with the implication that the people involved do so out of faith rather than reason or practical concerns. In my experience systemic linguists do not differ from other school as far as faith or reason are concerned; where some do differ is in their political commitment to designing their model as a tool for intervening in political processes. Is it this political commitment that you mistake for religious zeal? Since it is politics rather than religion which is at issue here, I think you should address it, and challenge it if you will, as such. But I take it from your remarks on deconstruction and political critique that you aren't opposed to the politics of SFL.

Now, as far as Huddleston and Dixon are concerned, there are obviously different readings of what went on. My understanding is that if you wanted your linguistics to be valued by linguists in general in the 60s then you did work that could be valued by mainstream linguistics, (i.e. consumable in some respect by MIT trained linguists). It is quite understandable that certain linguists, trained in alternative paradigms, would shift the direction of their work to allow it to be valued. I wonder whether this counts as 'renouncing' a particular model. Against these two linguists trained in the 60s it is important to place the hundreds of linguists trained by Halliday who have carried on his work and now run annual International Systemic Congresses around the world (ISFC 21 this July in Belgium), and annual national conferences in each of Europe, China, Japan and Australia. In any case, it is important to consider carefully what linguists like Huddleston actually say about Halliday - for example:

...I have a great admiration for him [= Halliday], and acknowledged in my review 'my own deep indebtedness to [him], with respect to both his influence on my thinking in linguistics theory and the grammar of English and also to the practical help he gave in my postgraduate and postdoctoral career' (1988:140). I emphasised at the outset my view that IFG 'contains innumerable original insights and valuable observations: it has a great deal to offer anyone interested in the grammar of English' (1988:140). Huddleston 1991:128

Dixon is, I think, a special case, whose recent book on English case (A New Approach to English Grammar on Semantic Principles ) is however profoundly Hallidayan (witness the notion of types of process with case frames assigned to each) - not that Dixon gives Halliday any credit for the inspiration or the affinities between his work and that in IFG.

You suggest that we "take time to evaluate alternative approaches", which is I think a good idea. Here's an agenda for negotiation:

1. For this kind of work you need a good theory of context, not just an open ended set of factors along the lines of your "depends on many factors, such as context, genre, style, voice and purpose". The model we are currently using has evolved into the form outlined in the ARAL article . What alternative are you proposing and how does it improve on field, mode, tenor, genre and ideology?

2. Beyond this, you need a good theory of structure beyond the sentence, since teachers are dealing with and teaching in texts, not clauses. What is your alternative to Halliday and Hasan 1976 and Martin 1992? What are the advantages of your alternative?

3. Context and discourse then need to be related to grammar. To do this effectively the grammar will have to provide descriptions of -

- case relations (transitivity), of particular use in deconstructing the uncommon sense knowledge of different disciplines; whose model of case are you promoting (Filmore, Starosta, Dixon, LFG, GB or what?); will it really do a better job than that demonstrated in Halliday and Martin 1993?

- mood and modality, of particular use in deconstructing classroom discourse and persuasive speaking or writing; whose model of conversational analysis are you promoting (ethnomethodology, speech act theory, Labov & Fanshel...?); how is it an improvement on the Birmingham to Nottingham to Sydney perspectives; whose model of modality are you proposing (formal semantics?); how do any of the models deal with metaphorical modalities (e.g. I suspect that, it is more than obvious that...)?

- information flow (Theme/Rheme & Given/New), of particular use in studying differences between speech and writing and the organisation of information in written texts; what alternative are you proposing (Chafe, Givon, Kuno...?); what are the advantages for teachers?

- clause combining (projection/expansion, hypotaxis/parataxis); are you proposing to distinguish hypotaxis from embedding or go with a general category of subordination; if you are going with subordination, how will this affect our understanding of differences between spoken and written text (cf. Halliday 1985 Spoken and Written Language )

Obviously we could go on to negotiate nominal groups (issues of nominalisation, modification), verbal groups (theory of tense to be deployed) and intonation to round off the picture. I think the NSW Board of Studies Handbook of Grammar roughly covers the relevant round.

In closing let me suggest that an evolving grammar such as the functional grammar developed over the last 15 years for schools will be far more effective than the designed one you propose. So I'd suggest you get involved in its evolution rather than setting yourself up in an oppositional position that might simply end up discrediting us all. This means arguing with me and other linguists responsible for the ideas being applied. In case you are interested I'm passing on a slightly dated annotated bibliography of materials we've prepared for teachers. A slightly dated list of references backing this up is provided in the ARAL overview paper. I'm also including a few recent papers which will put you in touch with current SFL thinking on transitivity, modality, logical meaning and theme ; each contains a good set of relevant references; the transitivity paper is an attempt at some meta-theory, which might be a good jumping off point for our discussions; the theme paper is part of a long argument with Huddleston about the value of Halliday's IFG (references to that debate included in the paper).

Let me finish off by stressing once again the political commitment of the people you are attacking, as far as redistributing literacy resources is concerned. We've thought long and hard about what we are doing, and let me assure you that we are not on about promoting SFL, but rather are concerned with designing a linguistics that can act as a political tool. There is no need for closure around an enterprise of this kind; if you want to talk, I'll talk back.

In the meantime, I wonder if you would consider republishing your article in Network, an SFL newsletter of which I am co-editor; I would then include this letter by way of reply, offering you as chance to respond as you wish.

cheers,


Jim Martin

Annotated references

i. Recent materials from the Sydney Metropolitan East Region's Disadvantaged Schools Program (DSP Productions, Bridge & Swanson Streets, Erskineville, NSW 2043, Australia) include:

- video materials on writing in science

Teaching Factual Writing: secondary science - teaching scientific writing in a secondary classroom. [video] 1990.

Earthworms: teaching factual writing in the early years of school. [video] 1991.

The Action Pack: activities for teaching factual writing. 1992.


- a report on the literacy demands of science based industries (further reports on literacy in media and cultural production and on literacy in public administration are scheduled for 1994)

Rose, D., D. McInnes & H. Korner. 1992. Scientific Literacy (Literacy in Industry Research Project - Stage 1). Sydney: Metropolitan East Disadvantaged Schools Program.


- inservice materials for teaching literacy in junior secondary English and Science (materials on literacy in social science will be published in 1994)

Rothery, J. 1992. The Language of School English. Sydney: Metropolitan East Disadvantaged Schools Program.

Veel, R. 1992. The Language of School Science. Sydney: Metropolitan East Disadvantaged Schools Program.


ii. Harcourt Brace Jovanovich (30-52 Smidmore St, Marrickville, NSW 2204, Australia) has published a third strand of genre based literacy materials in 1992, for use in upper primary school:

Christie, F, B Gray, P Gray, M Macken, J R Martin & J Rothery. 1992a Exploring Explanations about Natural Disasters (level 1). Sydney: Harcourt Brace Jovanovich (HBJ Language: a resource for meaning).

Christie, F, B Gray, P Gray, M Macken, J R Martin & J Rothery. 1992b Exploring Explanations about Life Cycles. (level 2). Sydney: Harcourt Brace Jovanovich (HBJ Language: a resource for meaning).

Christie, F, B Gray, P Gray, M Macken, J R Martin & J Rothery. 1992c Exploring Explanations about Electricity (level 3). Sydney: Harcourt Brace Jovanovich (HBJ Language: a resource for meaning).

Christie, F, B Gray, P Gray, M Macken, J R Martin & J Rothery. 1992d Exploring Explanations about Astronomy (level 4). Sydney: Harcourt Brace Jovanovich (HBJ Language: a resource for meaning).

Christie, F, B Gray, P Gray, M Macken, J R Martin & J Rothery. 1992e Exploring Explanations: teachers book (levels 1-4). Sydney: Harcourt Brace Jovanovich (HBJ Language: a resource for meaning).


iii. The Primary English Teaching Association (PO Box 167, Rozelle, NSW 2039, Australia) has developed a new, genre-based video package:

Derewianka, B. 1991. Exploring How Texts Work: the video. Sydney: Primary English Teaching Association.


iv. LERN (the Literacy and Education Research Network) has made available collections of papers from its 1989 and 1991 Working with genre conferences (PO Box 721, Leichhardt, NSW 2040, Australia):

Working with Genre (Papers from the 1989 LERN Conference, University of Technology, Sydney, 25-26 November 1989). 1991. Sydney: Common Ground.

Working with Genre II: literacy across the curriculum (Papers from the 1991 LERN Conference, University of Technology, Sydney, 23-24 November 1991). 1992. Sydney: Common Ground.

- in addition LERN has prepared an important collection of papers on genre and literacy teaching in Australia:

Cope, W. & M. Kalantzis [Eds.] 1993. The Literacies of Power and the Powers of Literacy.


v. Australia's federal Department of Employment, Education and Training's report on the preservice preparation of teachers for teaching English literacy is now available through the Centre for Studies of Language in Education, Northern Territory University, Casuarina, NT 0810, Australia). The 'Christie' report marshals contemporary critical theory and functional linguistics to produce a series of ground-breaking recommendations, and a three volume report (including a collection of 13 papers from Australia's leading literacy educators as Volume 2):

Christie, F., B. Devlin, P. Freebody, A. Luke, J. R. Martin, T. Threadgold & C. Walton. 1991. Teaching English Literacy: a project of national significance on the preservice preparation of teachers for teaching English literacy. Vols. 1, 2 & 3. Canberra: Department of Employment, Education and Training & Darwin: Centre for Studies of Language in Education, Northern Territory University.


vi. The NSW Adult Migrant English Service [Curriculum Support Unit, 85-86 Mary St., Surry Hills, NSW 2010, Australia] has recently published materials oriented to workplace literacy in manufacturing industry. Joyce introduces a range of operator level workplace texts to ESL classroom teachers; Prince focuses more on the needs of workplace based literacy teachers in the context of post-Fordist industrial restructuring.

Joyce, H. 1992. Workplace Texts in the Language Classroom. Curriculum Support Unit, NSW Adult Migrant English Service.

Prince, D. 1992. Literacy in the Workplace. Curriculum Support Unit, NSW Adult Migrant English Service.


vii. The following collections of papers track the development of genre-based literacy pedagogies from around 1985. Reid 1987, and Giblett and O'Carroll 1990, provide forums for an often intense debate. Christie 1991 extends the discussion in the direction of critical literacy.

Painter, C & J R Martin [Eds.] 1986 Writing to Mean: teaching genres across the curriculum. Applied Linguistics Association of Australia (Occasional Papers 9). [available from Brian McCarthy, Department of Modern Languages, University of Wollongong, P.O. Box 1144, Wollongong, NSW 2500, Australia]

Reid, I [Ed.] 1987 The Place of Genre in Learning: current debates. Geelong, Vic.: Centre for Studies in Literary Education, Deakin University (Typereader Publications 1). [available Deakin University, Geelong, Vic 3217, Australia]

Christie, F. [Ed.] 1989. Literacy for a Changing World: a Fresh Look at the Basics:. Hawthorn, Vic.: The Australian Council for Educational Research (Theme Monograph Series). [Radford House, Frederick Street, Hawthorn, Vic. 3122, Australia]

Walton, C. & W. Eggington [Eds.] 1990. Language: maintenance, power and education in Australian Aboriginal contexts. Darwin, N.T.: Northern Territory University Press. [Casuarina, NT 0810, Australia]

Giblett, R. & J. O'Carroll [Eds.] 1990. Discipline - Dialogue - Difference: proceedings of the Language in Education Conference, Murdoch University, December 1989. Perth: 4D Duration Publications, School of Humanities, Murdoch University. [Murdoch, WA 6150, Australia]

Christie, F. [Ed.] 1991. Literacy in Social Processes: papers from the inaugural Australian Systemic Linguistics Conference, held at Deakin University, January 1990. Darwin: Centre for Studies of Language in Education, Northern Territory University. [Casuarina, NT 0810, Australia]


viii. The theoretical foundations of the educational linguistics assumed here are further developed in recent publications by Hasan & Martin, by Martin (providing documentation of the discourse analyses used to deconstruct texts in genre-based literacy programs) and by Halliday and Martin (drawing together their deconstructions of written scientific English, as a complement to Lemke's 1990 work on the spoken mode).

Halliday, M. A. K. & J. R. Martin. 1993. Writing Science: literacy and discursive power. (with M A K Halliday) London: Falmer (Critical perspectives on literacy and education).

R. Hasan & J. R. Martin [Eds.] 1989. Language Development: learning language, learning culture. Norwood, N.J.: Ablex (Advances in Discourse Processes 27 - Meaning and Choice in Language: studies for Michael Halliday).

Martin, J. R. 1992. English Text: system and structure. Amsterdam: Benjamins.