1

GENRE AND FIELD: SOCIAL PROCESSES AND KNOWLEDGE

STRUCTURES IN SYSTEMIC FUNCTIONAL SEMIOTICS

J R MARTIN
(University of Sydney)
1. Modelling field

Martin (2002) interprets field is a set of activity sequences oriented to

some global institutional purpose, alongside the taxonomies of
participants involved in these sequences (organised by both
classification and composition). The Australian Government's Bureau of
Meteorology's webpagei on clouds reflects these concerns in their
discussion of Cloud formation and Cloud classification. Cloud
formation focusses dynamically on activity, from the scientific
perspective of meteorology:
[1] Clouds have their origins in the water that covers 70 per cent of the earth's
surface. Millions of tons of water vapour are evaporated into the air daily from
oceans, lakes and rivers, and by transpiration from trees, crops and other plant
life.

As this moist air rises it encounters lower pressures, expands as a result, and in
doing so becomes cooler. As the air cools it can hold less water vapour and
eventually will become saturated. It is from this point that some of the water
vapour will condense into tiny water droplets to form cloud (about one million
cloud droplets are contained in one rain-drop). Thus, whenever clouds appear
they provide visual evidence of the presence of water in the atmosphere.

Cloud classification focuses statically on relationships, in this case a

meteorological taxonomy of cloud types:
[2] There are ten main cloud types, which are further divided into 27 sub-types
according to their height shape, colour and associated weather, Clouds are
categorised as low (from the earth's surface to 2.5 km), middle (2.5 to 6 km),
or high (above 6 km). They are given Latin names which describe their
characteristics, e.g. cirrus (a hair), cumulus (a heap), stratus (a layer) and
nimbus (rain-bearing). It's an interesting fact that all clouds are white, but
when viewed from the ground some appear grey or dark grey according to
their depth and shading from higher cloud.

Both orientations to clouds are in addition supported by images. Four

different ways in which clouds are formed are outlined in diagrams
involving dynamic vectors (Fig. 1).

Proceedings
33rd International Systemic Functional Congress
2006
 2

Fig. 1: Four Ways in which Moist Air can be Lifted to Form Clouds

And the ten major cloud types are illustrated with photographs, arranged
iconically according to cloud height as high, middle and low (Fig. 2).
The captions on each photo include further information about shape and
colour, and associated precipitation (if any). As Unsworth 2001 surveys
for educational contexts, the field is realised multimodally, with image
complementing verbiage. But the complementarity of activity
sequencing and taxonomising holds true across modalities - since fields
are about both what is going on, and who or what is involved in these
activities.
Proceedings
33rd International Systemic Functional Congress
2006
 3

Proceedings
33rd International Systemic Functional Congress
2006
 4

Fig. 2: Types of Clouds Illustrated

Martin 2002 also includes some speculative discussion about

relationships among fields, by way of exploring Bernstein's notion of
common and uncommon sense. A provisional mapping of field-types is
outlined in Fig. 3, taking into account everyday domestic, recreation and
trades, administration, humanities, social science and science.

Fig. 3: Martin's 2002 Mapping of Fields in Relation to Uncommon Sense

He further speculates about the nature of the activity sequences and

taxonomies associated with different positions on the cline (Fig. 4),
taking into account the way they tend to be documented in writing (or
Proceedings
33rd International Systemic Functional Congress
2006
 5

not) and the nature of the phenomena explored. These suggestions

anticipate exciting developments by Bernstein and his colleagues with
respect to the sociology of knowledge, which we will take up below.

Fig. 4: Kinds of Field in Relation to Activity and Taxonomy

2. Common and uncommon sense (science)

The uncommon sense discourse of science was used to illustrate field as

activity and field as taxonomy in section 1 above. Text 1 presented a
scientific understanding of cloud formation, and text 2 a scientific
understanding of cloud classification. Text 1 reconstrues the everyday
word cloud as a scientific concept, and drew on related scientific
understandings to do so (evaporation, transpiration, low pressure, water
vapour, saturation, condensation). Text 2 reworks our everyday
experience of clouds into a categorisation based on height, shape, colour
and associated weather - introducing scientific categories to do so
(cirrus, cumulus, stratus, numbus). The technical terminology encoding
these understandings is perhaps the most striking 'surface' feature of
scientific discourse.

This very technicality has the function of linking activity to taxonomy -

of linking cloud formation to cloud classification for example. Messel
1963.7.7 describes warm fronts as follows, and then uses the image in
Fig. 5 to label cloud types in relation to uplift.

Proceedings
33rd International Systemic Functional Congress
2006
 6

[3] Warm fronts. When a warm air stream meets a colder air mass, the warm
air, being less dense, slides up over the cold air and the temperature falls.
Condensation generally ensues. The surface between the two air masses is
inclined at a smaller angle than is the case for a cold front. See Fig. 7.8.
Warm fronts are rare in Australia.

The approach of a warm front is heralded by the appearance of high, white,

wispy clouds, known as cirrus cloud. As the front approaches, the clouds
become lower and thicker, culminating in masses of heavy rain clouds. The
weather usually clears quickly after the front has passed. However, a warm
front is commonly followed, after an interval which may be anything up to a
day or more, by a cold front. [Mesel 1963. 7.7]

Scientific explanations have been explored in detail by Unsworth

(1997a, b, c, 1999, 2001, 2004) and we won't develop this aspect of
scientific fields further here. As far as taxonomising is concerned we
should emphasise that scientific taxonomies are much deeper than
everyday ones. Where everyday discourse gives us perhaps clouds and
rain clouds, Wikpedia (http://en.wikipedia.org/wiki/Cloud_types) refers
to the following main cloud types:

Fig. 5: Warm Front Activity in Relation to Cloud Type [Messel 1963]

Proceedings
33rd International Systemic Functional Congress
2006
 7

1 High-level clouds
o 1.1 Cirrus
o 1.2 Cirrocumulus
o 1.3 Cirrostratus
o 1.4 Contrail
2 Medium-level clouds
o 2.1 Altostratus
o 2.2 Altocumulus
o 2.3 Nimbostratus
3 Low-level clouds
o 3.1 Stratocumulus
o 3.2 Stratus
o 3.3 Cumulus
4 Vertically developed clouds
o 4.1 Cumulonimbus

And any one of these can be broken down into further sub-types -
altostratus clouds, which form when "a large air mass is condensed,
usually from a frontal system, and can bring rain or snow", for example,
are subclassified as follows:
altostratus duplicatus
altostratus lenticularis
altostratus mammatus
altostratus opacus
altostratus praecipitatio
altostratus radiatus
altostratus translucidus
altostratus undulatus

The depth and precision of classifications of this kind distinguishes

taxonomising in science from taxonomising in everyday life. And
comparable precision and delicacy is found for decomposition. We learn
from text 1 that clouds are made of water droplets, and we can pursue
this further into the realm of physics and atomic structure. There we
learn that water is a V-shaped molecule, known chemically as H2O
(meaning two hydrogen and one oxygen atom bonded together into a
molecule). Pushing further we might find that water molecules are
symmetric, with two mirror planes of symmetry and a 2-fold rotation
axis (http://www.lsbu.ac.uk/water/molecule.html); and this is probably
already further than a high school physics challenged mortal like myself
would like to go (cf. O'Halloran 2006 for a rich account of science and
mathematics from someone not challenged in this way). I can't help,

Proceedings
33rd International Systemic Functional Congress
2006
 8

however, finding the images modelling this uncommon sense

composition intriguing (from the website just noted).

Fig. 6: Uncommon Sense Composition of the Water Molecule

Alongside depth and precision, we also need to keep in mind the criteria
on which uncommon sense classifications and decompositions are based
- namely technologically augmented perception, over various depths of
time, in relation to meticulous record keeping (i.e. writing). Everyday
taxonomising is based more simply on what we sense going on around
us (by seeing, hearing, tasting, feeling, smelling); so we have clouds
(which block the sun) and clouds that look like they might rain or snow
(rain clouds, snow clouds), with no certain division between the two -
and that's pretty much all that city-dwellers need to get through everyday
life (alongside of course the unreliable fortune telling we tune in to
during weather broadcasts, in print or electronic media of one kind or
another, as we plan our recreational activities).

The evolution of the uncommon sense discourses of science in English

and Chinese is outlined in Halliday 2004. Recontextualisations of these
Proceedings
33rd International Systemic Functional Congress
2006
 9

in pedagogic discourse are explored in Eggins et al.1993, Halliday &

Martin 1993, Martin 1993a, b, Martin & Veel 1998.

3. Common and uncommon sense (history)

One of the first things we notice when we move from the field of science
to the field of history is the relative absence of technicality. We do find
some borrowed terminology, from say Marxism (e.g. capitulationist
tendencies, material aspirations, ideologues of the emerging elite,
demands of the masses) or critical theory (e.g. discourse, narrative,
subject, Law, desire, interrogation, power). And specialised
bureaucratic classifications are regularly drawn upon (South Australian
police, Australian Federal Police, High Court of Australia, Federal
Magistrates Courts), often in the form of acronyms:

DIMIA Dept of Immigration, Mutlicultural & Indigenous Affairs

DFAT Dept of Foreign affairs & Trade
DEET Dept of Employment, Education and Training
HREOC Human Rights & Equal Opportunity Commission

For the most part however, historical classification is a matter of

instantial classification in texts. In text 4 for example, Mares interprets
Ruddock, the then Australian Minister for Foreign Affairs, as presenting
three models of immigration. These scenarios are clearly outlined in the
text, but are not technicalised. Mares does not feel the need to introduce
and define terms for each, which means that Ruddock's vision will not
transcend this particular presentation and move into general
consciousness. The classification is left to reside in the text itself, from
which it will have to be recovered by future readers.

[4] The minister moves on to outline three competing visions for Australia’s
population in the century ahead. The first scenario is the high-immigration
model favoured by some business groups, which call for Australia’s net
migration intake to be set at 1 per cent of existing population per year... The
second scenario is net zero migration, the model pushed by sections of the
environmental movement and by groups such as One Nation, which say that
Australia should take just enough migrants to replace the number of people
who permanently depart the country each year... The minister’s final forecast
is reassuring – according to him, if we hold fast to the current government
policy, Australia’s population will increase gradually for the next forty years
before settling comfortably at around 23 million... [Mares, P 2001 Borderline:
Proceedings
33rd International Systemic Functional Congress
2006
 10

Australia’s treatment of refugees and asylum seekers. Sydney: UNSW Press

(Reportage Series) 141-142]

Historical composition is different story. As far as people, places and

things are concerned, we find the same pattern as with classification -
historians borrow technicality (e.g. geographical locations, the structure
of government agencies), but they don't create it. It is for activity
however that uncommmon sense composition comes to the fore (Martin
2002a, 2003).

Obviously history is concerned with what has happened in the world.

And there are certainly occasions when the temporal sequencing of
specific events has to be chronicled in some detail. In text 5 for
example, Brennan presents part of the sequence of events whereby the
Australian government prevented a cohort of asylum seekers from
landing on Australian soil. His reason for doing this is to document for
the historical record the shameful mendacity of the Australian
government of the time. In [5] the forward pointing arrows point to
events which follow in time, and backward pointing arrows to events
which precede.

[5] On 29 August the Tampa entered into Australian territorial waters

approaching Christmas Island. → The prime minister told parliament that ←
the captain had decided on this course of action because ← a spokesman for
the asylum seekers ‘had indicated that they would begin jumping overboard if
medical assistance was not provided quickly’. → Captain Rinnan gave a
different reason for this decision: ‘We weren’t seaworthy to sail to Indonesia.
There were lifejackets for only 40 people. The sanitary conditions were
terrible.’ → The SAS came aboard and took over Tampa. → An Australian
Defence Force doctor was given 43 minutes to make a medical assessment of
the 433 asylum seekers. → He reported, ‘Four persons required IV (2 urgent
including 1 woman 8 months pregnant).’ → Captain Rinnan was surprised at
the prompt medical assessment, because ← his crew had already identified ten
people who were barely conscious lying in the sun on the deck of the ship. →
The prime minister then made a finely timed ministerial statement to
parliament insisting that ← ‘nobody – and I repeat nobody – has presented as
being in need of urgent medical assistance as would require their removal to
the Australian mainland or to Christmas Island’. → One hundred and thirty-
one fortunate asylum seekers were granted immediate asylum by the New
Zealand government. → The rest, ← having been transported to Nauru, waited
processing under the evolving Pacific Solution. [Brennan 2003: 42-43]

Proceedings
33rd International Systemic Functional Congress
2006
 11

Texts featuring this kind of detailed sequence in time are relatively rare
in history discourse, since historians are responsible for a multitude of
overlapping activities unfolding through long passage of time. The
Tampa episode, referred to in [5], is just one moment in the history of
people arriving in Australia by boat to seek political asylum. Brennan
outlines a much longer phase of this activity as follows:

[6] The first wave of 2,077 Indochinese boat people came to Australia in 54
boats between 1976 and 1981. In that time, Australia was to resettle another
56,000 Indochinese through regular migration channels. The first boatload of
asylum seekers arrived in Darwin harbour on 28 July 1976. The five
Vietnamese had made the 6,500-kilometre journey in a small boat. At the end
of that year another two boats arrived carrying 106 people who were screened
for health reasons and then flown to Wacol migrant hostel outside Brisbane.
When the third Vietnamese boat of the first wave arrived, there was some
media agitation about the threatened invasion by boat people. One Melbourne
newspaper reported that ‘today’s trickle of unannounced visitors to our lonely
northern coastline could well become a tide of human flotsam’. The paper
asked how the nation would respond to ‘the coming invasion of its far north
by hundreds, thousands and even tens of thousands of Asian refugees”. The
invasion never occurred.

In 1978 the Communist government in Vietnam outlawed private business

ventures. Tens of thousands, mainly ethnic Chinese, then fled by boat. The
outflow of Vietnamese boat people throughout the region gave rise to great
moral dilemmas in the implementation of government policies. Countries such
as Malaysia would periodically declare that their camps were full and they
could take no more boat people. They would even threaten to shoot new
arrivals on sight. Alternatively, they would provide them with food, fuel and
repairs so they could set off for another country. Meanwhile Vietnamese
officials were profiting by charging the boat people high departure fees.

Camps were filling around Southeast Asia. There was no let-up in the
departures from Vietnam. In the end there was a negotiated agreement
involving Vietnam, the countries of first asylum such as Thailand and
Malaysia, and the resettlement countries, chiefly the United States, Canada
and Australia. In 1982 the Australian government announced that the
Vietnamese government had agreed to an Orderly Departure Program.
Australian immigration ministers Michael MacKellar and Ian MacPhee were
able to set up procedures for the reception of Vietnamese from camps in
Southeast Asia as well as those coming directly from Vietnam under a special
migration program. With careful management, they were able to have the
public accept up to 15,000 Vietnamese refugees a year when the annual
migration intake was as low as 70,000. [Brennan: 29-30]

Proceedings
33rd International Systemic Functional Congress
2006
 12

In texts of this kind, instead of moving sequentially from one specific

event to another as in [5], we hop selectively from one significant phase
of activity to another. This is scaffolded by circumstances of location in
space, typically in clause initial position, which tell us where we are (e.g.
between 1976 and 1981, in that time, on 28 July 1976, at the end of that
year). And as we can see from Brennan's reference to these activities as
the first wave of boat people, there are more waves to come. He moves
on in fact to consider four waves, each of which is broken down into
phases of time, as outlined in Fig. 7 (from Martin & Rose 2006).

The first wave of 2,077 Indochinese boat people came to Australia in 54 boats É
The first boatload of asylum seekers arrived in Darwin harbour on 28 July 1976.

At the end of that year another two boats arrived carrying 106 people É

When the third Vietnamese boat of the first wave arrived, there was some media agitationÉ

In 1978 the Communist government in Vietnam outlawed private business venturesÉ

In the end there was a negotiated agreementÉ

In 1982 the Australian government announced that the Vietnamese government had agreedÉ

In 1978 the government set up a Determination of Refugee Status (DORS) CommitteeÉ

In the early 1980s the committee considered fewer that 200 applications a year...

In 1982 the government decided that even offshore cases would be decided on a case by case basisÉ

At the same time the government set up a Special Humanitarian ProgramÉ

The In the first year, there were 20,216 offshore refugeesÉ

Within eight years there were only 1,537 under the offshore refugee categoryÉ

Initially it was assumed that there would be only a few hundred of such onshore cases a yearÉ
In 1985 the High CourtÉ decided that ministerial decisionsÉw ere reviewable by the courtsÉ

The second wave of boat people commenced with the arrival of a Cambodian boatÉ

The third wave of boat people arrived between 1994 and 1998É

The fourth and biggest wave of boat people in modern Australian history could not be
so readily categorisedÉ
Fig. 7: Brennan's Phases of Historical Activity

Proceedings
33rd International Systemic Functional Congress
2006
 13

The critical point here is that although Brennan doesn't name phases
within his four waves, he does name the waves themselves. This in
effect turns a lot of activity into a thing - four things in fact, which taken
together make up his history of the boat people. As with Ruddock's
classification of scenarios discussed above, Brennan's composition
remains an instantial one, recoverable from his book but not beyond.
But naming affords the possibility of technicalisation, and this is
certainly what has happened for the Tampa episode, a part of which was
sequenced in text 5. That particular phase of Australian immigration
history has transcended the texts in which it is construed, Comparable
technicalisations of phases of history with which many are familiar
would include the Middle Age, the Rennaissance, World War I, the
Treaty of Versailles, the Long March, the Cultural Revolution, or more
specifically, for Australian readers, the Dismissal, Mabo, Bodyline,
Kokoda, Lone Pine and so on.

Compositionally speaking then, uncommon sense history turns a

multitude of overlapping and successive events into a thing divided into
phases, which may be named (e.g. the first wave), and possibly
technicalised (e.g. Tampa). This reification processes allows for phases
to be grouped and divided, until a requisite compositional hierarchy is
construed.

Of course history has to do more than simply fashion the past as an

episodic chronicle; it also has to explain. To this end, causality is by and
large handled inside rather than between clauses, taking advantage of
experiential resources realised through transitivity structures (rather than
logical resources realised through conjunctive links in clause
complexes). The follow pattern is typical in that is involves two
nominalised events (outflow and implementation) causally related by the
process give rise to.

The outflow of Vietnamese boat people throughout the region

gave rise to
great moral dilemmas in the implementation of government policies

Realising cause inside the clause in this way enables historians to fine
tune causality by deploying verbs which can finely differentiated types
of cause and effect relations (e.g. argue, act, attract, spark). Manne

Proceedings
33rd International Systemic Functional Congress
2006
 14

begins text [7] in just these terms, with unwillingness related to response
via a strong causal process of 'determination'.

The Howard government’s unwillingness to apologise

determined
the nature of its response to other recommendations contained in Bringing
them home.

Since he is writing popular history, he then unpacks this intra-clause

causality as three inter-clausal causal connections realised by because.
This makes the reasoning more like the reasoning we are all familiar
with from spoken discourse.

[7] The Howard government’s unwillingness to apologise determined the

nature of its response to other recommendations contained in Bringing them
home.

Because it refused to consider the present generation of Australians legally or

morally responsible for the mistakes of the past,
it refused altogether Bringing them home’s recommendation for
financial compensation for members of the stolen generations.

Because it thought the policies of child removal had been lawful and well-
intentioned,
it treated almost with contempt the arguments in Bringing them home
which suggested that in removing Aboriginal children from their
families by force previous Australian governments had committed
serious violations of the human rights treaties they had signed or even
acts of genocide.

Because, nonetheless, it accepted that the Aboriginal children who had been
taken from their families had suffered serious harm
it was willing to allocate modest sums to assist members of the stolen
generations with psychological counselling, family reunion, cultural
projects, oral histories and so on.

The price Manne pays for his popularisation is that he has to keep
repeated the same causal connection over and over again (because,
because, because). There are certainly moments in history when a
simple causal relation such as this is enough, although even there
historians prefer to explain within rather than between clauses.

The wide reporting of the violence

had caused
concern to grow among the politically powerful missionary societies in Britain
Proceedings
33rd International Systemic Functional Congress
2006
 15

The narrow selection of sources

results in
a profound ignorance of the basics of Van Diemanian economy, society and...

which in turn
leads
to a series of elementary errors.

The wide reporting of the violence

had caused
concern to grow among the politically powerful missionary societies in Britain

Elsewhere, variations on roughly the same structural configuration

afford a range of nuanced cause and effect relations.

The over-reliance on the government’s own records

grossly distorts
Windschuttle’s understanding of the realities of frontier life for two reasons.

‘very considerable difficulties

arise from
the insufficiency of stationery’.

Government record keeping

improves somewhat
with the arrival of Sorrell in 1817

the detention in remote places

were contributing to
more regular bad decision making at the primary stage

the savings from not holding unlawful arrivals in protracted detention…

could be devoted to
increased surveillance of all overstayers in the community

This (= increased surveillance of all overstayers in the community)

would facilitate
the orderly departure from Australia of overstayers

Managing causality along these lines depends on control over

grammatical metaphor, so that events can be nominalised as things and
related verbally to one another (Coffin 2006, Halliday & Matthiseen
1999, Martin 1993a, b, 2002a, 2003, 2007). It is this process which
creates the resources which historians depend on to explain.

Proceedings
33rd International Systemic Functional Congress
2006
 16

4. Knowledge structure: a sociological perspective

Recently the SFL tradition of field differentiating research has been

brought into dialogue with the rich sociological perspective on
knowledge structure inspired by Bernstein 1996/2000 and developed by
Muller 2000. Christie & Martin 2007 document significant moments in
this exchange. Bernstein is himself developing his earlier common vs
uncommon sense opposition which inspired much of the SFL research.
He begins by distinguishing between everyday horizontal discourse and
the vertical discourses of the humanities, social science and science.

A Horizontal discourse entails a set of strategies which are local, segmentally

organised, context specific and dependent, for maximising encounters with
persons and habitats....This form has a group of well-known features: it is
likely to be oral, local, context dependent and specific, tacit, multi-layered and
contradictory across but not within contexts. [Bernstein 2000:157]

...a Vertical discourse takes the form of a coherent, explicit and

systematically principled structure, hierarchically organised as in the sciences,
or it takes the form of a series of specialised languages with specialised modes
of interrogation and specialised criteria for the production and circulation of
texts as in the social sciences and humanities. [Bernstein 2000:157]

Then, within vertical discourse, he distinguishes between the

hierarchical knowledge structures characteristic of science and the
horizontal knowledge structures of the humanities. Bernstein uses the
image of a triangle below to symbolise hierarchical knowledge structure
(definitions from Maton & Muller 2007).

A hierarchical knowledge structure is "a coherent, explicit and

systematically principled structure, hierarchically organised’ which ‘attempts
to create very general propositions and theories, which integrate knowledge at
lower levels, and in this way shows underlying uniformities across an
expanding range of apparently different phenomena" (1999: 161, 162).

Proceedings
33rd International Systemic Functional Congress
2006
 17

A horizontal knowledge structure is defined as "a series of specialised

languages with specialised modes of interrogation and criteria for the
construction and circulation of texts" (1999: 162).

L1L2L3L4L5L6L7...Ln

Bernstein's indefinitely extendable listing of Languages is meant to

characterise the proliferation of theories in a field like linguistics, where
a theory such as SFL co-exists with related 'functional' models such as
Lexical Functional Grammar (LFG), Role and Reference Grammar
(RRG), Functional Grammar (FG) and Cognitive Linguistics, alongside
various formalist paradigms.

Wignell 2007 argues social sciences are better characterised as warring

triangles, since each language models itself in some respect on science
and tends not to get along very well with alternatives:

This distinguishes the social sciences from the humanities where

technicality and the drive to integration via general models and
propositions is much less strong (but where similarly, competing
segments are relatively intolerant of each other).

Muller 2007 proposes the term 'verticality' to describe progression in

the development of theories via ever more integrative or general
propositions (extending Bernstein's notion of strong vs weak internal
grammars of description). In his terms, hierarchical knowledge
structures (canonically physics) would exhibit more verticality than the
more aspirational languages of social science - a process we might image
as follows:

Proceedings
33rd International Systemic Functional Congress
2006
 18

Proliferating segments in the humanities on the other hand would exhibit

very little verticality at all, since they progress via the introduction of a
new language which constructs a ‘fresh perspective, a new set of
questions, a new set of connections, and an apparently new problematic,
and most importantly a new set of speakers’ (Bernstein: 1999: 162).

L1 L2 L3 L4 L5 L6 L7 Ln

In addition Muller proposes 'grammaticality' to describe how

theoretical statements in knowledge structures deal with their empirical
predicates (extending Bernstein's conception of strong vs weak external
grammars of description). The stronger the (external) grammaticality of
a language in these terms, the more stably it is able to generate empirical
correlates and the more unambiguous its claims because of a more
restricted field of referents.

data

For Muller segments of horizontal knowledge structures would have a

much weaker relation to data, which might, as in the case of cultural
studies or literary criticism, afford divergent readings no one of which
can be shown to be empirically more adequate than another.

L texts
As far as Bernstein's distinction of horizontal from vertical discourse is
concerned, SFL's main contribution to date has been to identify
grammatical metaphor as the key linguistic resource deployed to
construe vertical discourse. As far as taxonomy is concerned, it enables
classification by packaging up relevant information in nominal form so
that terms can be defined and related to one another. Note for example
the key packaging nominalisations, highlighted below, in Bernstein's
definitions of the concepts we are exploring here.

...a Vertical discourse (technical term)

takes the form of (=)
a coherent, explicit and systematically principled structure, hierarchically
organised as in the sciences, or it takes the form of a series of specialised
Proceedings
33rd International Systemic Functional Congress
2006
 19

languages with specialised modes of interrogation and specialised criteria for

the production and circulation of texts as in the social sciences and
humanities. (definition)

A Horizontal discourse (technical term)

entails (=)
a set of strategies which are local, segmentally organised, context specific and
dependent, for maximising encounters with persons and habitats. (definition)

And for explanation, as we have seen for history discourse, grammatical

metaphor is the resource which enables 'cause in the clause' realisations
of cause and effect relations. A full range of these is deployed in text 8
below, which complements the scientific explanation of warm fronts in
text 3 above (setting up a taxonomic opposition between cold and warm
frons as it does so):

[8] Cold fronts. A stream of comparatively cold, dense air tends to move
along close to the ground as it flows towards regions in which warmer, less
dense, air is rising. This rising air becomes cooler for the reasons mentioned
earlier, and if it is humid condensation of water vapour will take place. The
resulting clouds are usually of the cumulous type. The front edge of the cold
air mass is known as a cold front. Much of the rain that falls in Australia
occurs as a result of cold front conditions.

Fig. 7.7 shows how a cold front causes uplift and condensation in a warmer,
humid, air mass.

The arrival of a cold front is marked by a sharp drop in temperature and a

sudden change of the wind direction. [Messel 7.6]

The contingent implications are mapped out through causal relations

realised by a noun (reasons), a conjunction (if), a verbal modifier
(resulting), a preposition (as a result of) and two verbs (causes, is
marked) - and only one of these, if, functions between clauses.

For its part social science explains in comparable terms. Bernstein for
example explains distributive rules as follows:

[9] Consider a situation where a small holder meets another and complains
that what he/she had done every year with great success, this year failed
completely. The other says that when this happened he/she finds that this
'works'. He/she then outlines the successful strategy. Now any restriction to
circulation and exchange reduces effectiveness. Any restriction specialises,
classifies and privatises knowledge. Stratification procedures produce
distributive rules which control the flow of procedures from reservoir to
Proceedings
33rd International Systemic Functional Congress
2006
 20

repertoire. This both Vertical and Horizontal discourses are likely to operate
with distributive rules which set up positions of defence and challenge.
(Bernstein 1996/2000: 158).

As for historians, there is more than one kind of cause - which in text 9
he verbalises as reduces, specialises, classifies, privatises, produce,
control and set up.
Now any restriction to circulation and exchange (Agent)
reduces
effectiveness (Medium)

Any restriction (Agent)

specialises, classifies and privatises
knowledge (Medium)

Stratification procedures (Agent)

produce
distributive rules… (Medium)

distributive rules which (Agent)

control
the flow of procedures from reservoir to repertoire (Medium)

...distributive rules which (Agent)

set up
positions of defence and challenge (Medium)

As far as Muller's conception of verticality is concerned, the main

insight SFL has to offer has to do with the distilling impact of
technicality in relation to grammatical metaphor. In effect what happens
is that definitions kill off grammatical metaphor by generating technical
things. We've seen several examples of this in relation to scientific and
social scientific concepts such as clouds, warm fronts, colds fronts,
vertical discourse, horizontal discourse, hierarchical knowledge
structure, horizontal knowledge structure, verticality and
grammaticality. Once established, these terms can be assembled into
taxonomies (e.g. warm/cold fronts, vertical/horizontal discourses and
then within vertical discourses, hierarchical/horizontal knowledge
structures, featuring verticality and grammaticality). Without the
distillation, such classification would be a cumbersome process indeed -
one which would be unlikely to transcend residence in a specific text (cf.
Ruddock's instantial taxonomy in text 3 above) and not one which would

Proceedings
33rd International Systemic Functional Congress
2006
 21

facilitate verticality by lightening up the discourse so that it can move

on.

And part of moving on of course involves deploying technical terms in

explanations. Bernstein used several technicalisations in text 9 above, in
causal relation to one another: stratification procedures, distributive
rules, reservoir, repertoire, vertical and horizontal discourses,
distributive rules. Technical distillation means that 'cause in the clause'
can relate large scale condensations of meaning to one another; and this
makes it easier for the discourse to develop ever more general and
integrative propositions.

In short then we are suggesting from an SFL perspective that the

evolution of vertical discourse in any culture depends on grammatical
metaphor, and that the degree of verticality a knowledge structure
exhibits correlates strongly with its technicality.

5. SFL as a knowledge structure

We are now in a position to consider from a sociology of knowledge

perspective what kind of knowledge structure SFL enacts. As suggested
above, is looks to be a prototypical warring triangle - as predicted by
Wignell for segments in social science knowledge structures. In saying
this we are emphasising that it is just one segment in a horizontal
knowledge structure (i.e. the discipline of linguistics), since it is unable
to subsume rival theories. And we are highlighting the fact that SFL
exhibits more of Muller's verticality and grammaticality than comparable
segments in the humanities, as reflected in its high level of technicality
and its concern with relating its propositions to varied instances of
language use.

As far as verticality is concerned, SFL deploys a number of hierarchies

which resonate with Bernstein and Muller's notion of ever more
integrative propositions. Martin & Rose 2007 for example set up an
abstract level of genre coordinating combinations of ideational,
interpersonal and textual meaning at the more concrete levels of register
and language (see Fig. 8 below).

Proceedings
33rd International Systemic Functional Congress
2006
 22

g e nre

fi e ld
mod e

t ex t ua l

ide a t ion a l

t e nor
int e r p er s on a l

Fig. 8: Genre as a Generalisation of Register Variation

And at the level of genre relations among conventionalised text types are
modelled in system networks, with the more integrative generalising
features to the left and subclasses to the right:
te x t st ru c tu red n ew s st o ry

e x pec tant re c o un t

time st ru c tu red reso lve d n arr a tiv e

c ounte re x pect a nt reac ted to a n ec do te

un re s ol v ed in ter p ret e d exe mp lum

c ommente d o b s erva tio n

Fig. 9: Classification of Story Genres

Proceedings
33rd International Systemic Functional Congress
2006
 23

As Figures 8 and 9 indicate, however, SFL makes use of more than one
hierarchy (i.e. a realisation hierarchy in Fig. 8 and a classification
hierarchy in Fig. 9). And such complementary hierarchiesii reflect the
multi-nocular perspective on linguistic phenomena adopted by SFL.
Beyond this there is a range of linguistic patterns which SFL models not
as hierarchies but as complementarities - the ideational, interpersonal
and textual meaning modelled as simultaneous wedgesiii in Fig. 8 for
example. In physics, the well known complementarity of light as wave
and light as particle illustrates the modelling issue here - at times, for
certain phenomena, integration under a single generalising proposition is
not possible. For a full explanation, we have to theorise a dual or trial
perspective. This suggests that verticality needs to be interpreted as
accommodating complementarity alongside apical integration, to allow
for the complementarities just reviewed, and in the case of SFL to allow
for for complementary hierarchies.

Fig. 10: Complementarity within and between Hierarchies

As far as technicality is concerned, SFL (and linguistics in general)

positions itself among the more technical of the social science
disciplines. The metaphor distilling power of technicality is recursively
deployed so that defining a category such as English Subject, for
example, may itself involve several technical terms:
The Subject is the interpersonal clause function which changes sequence with
the Finite to change MOOD between declarative and interrogative and is
referred to by an anaphoric pronoun in mood tags.

We can contrast this with the relative lack of recursive technicality in

Bernstein’s definitions, which tend to pack a number of grammatical

Proceedings
33rd International Systemic Functional Congress
2006
 24

metaphors into a single term but less commonly re-deploy this

technicality in the definition of further concepts.

A Horizontal discourse entails a set of strategies which are local, segmentally

organised, context specific and dependent, for maximising encounters with
persons and habitats....This form has a group of well-known features: it is
likely to be oral, local, context dependent and specific, tacit, multi-layered and
contradictory across but not within contexts. [Bernstein 2000:157]

Muller's grammaticality is a divisive issue amongst linguistic theories,

since linguists are deeply divided over whether their data is constituted
by instances of language in use on the one hand or intuitions about the
well-formedness of structures on the other. SFL takes language is use as
data and so devotes a great deal of modelling to the problem of
connecting theory, description and documented instances of language
use with one another. Alongside realisation (degrees of abstraction from
phonic substance) and delicacy (subclassification), illustrated in Figures
8 and 9 above), instantiation and individuation hierarchies are also
deployed. Instantiation relates climate to weather along a scale
concerned with the specification of meaning potential. Individuation
relates the reservoir of meanings in a culture to individual repertoires,
along a scale concerned with coding orientation. In a model of this kind,
all levels of abstraction (language, register and genre) instantiate - the
system to text relation; and all levels of abstraction also individuate - the
reservoir to repertoire relation. While this might seem from the outside
as a rather over-determined implementation of Muller's grammaticality,
from the perspective of SFL it is simply arguing that in order to make
linguistic data bear on theory we have to consider how far we are from
the noises coming out of people's mouths (realisation), how many
instances we are generalising across (instantiation) and just who and
how many people are speaking (individuation).

Proceedings
33rd International Systemic Functional Congress
2006
 25

Fig. 11: Overdetermined Grammaticality in SFL

6. SFL as a meta-language

Martin (1999, 2007) interprets the genre-based literacy programs of the

'Sydney School' in relation to Bernstein's notion of pedagogic discourse
(1996). Since these programs deploy explicit knowledge about language
and related modalities of communication in their pedagogy, they involve
a doubling of instructional discourse - with a social semiotic
instructional discourse (SSID) projecting the instructional discourse (ID)
of the various subject areas. Metadiscourse in other words can be read
as projecting instructional discourse, with both of these discourse in turn
projected by regulatory discourse (RD); Christie 2002 discusses the

Proceedings
33rd International Systemic Functional Congress
2006
 26

notion of projection used here, referred to as embedding by Bernstein

(e.g. 1996).

SSID ID
RD

This raises the question of what kind of recontextualisation process is

involved when SFL models the discourse of a discipline by way of
enabling its pedagogy and curriculum planning. History, sociology and
science have been subjected to this kind of metadiscursive analysis in
various sections of this paper. Let's pursue the case of history here, in
this instance by factoring it as a system of genres (Coffin 1997, 2000,
2006; Martin 2002a, 2003, Martin & Rose 2006). Work on secondary
school history is outlined in Fig. 12, which deals with a range of
chronologically unfolding (recounts and accounts), explanatory
(explanations) and argument genres - arranged with those closer to
common sense at the top of the diagram and those further away below.

Proceedings
33rd International Systemic Functional Congress
2006
 27

first person autobiographical common

recount sense
biographical

biographical
temporal third person
recount

field time historical historical recount

causal historical account

inputs factorial explanation

explain

outcomes consequential explanation

text time promote exposition

one-sided

argue rebut challenge

uncommon
multi-sided discussion sense

Fig. 12: Secondary School History as a System of Genres

This mainly typological perspective can be complemented in SFL with a

more topological one which uses vectors to establish a semantic region
in which genres can be related as more or less alike one another. This
kind of modelling in outlined in Fig. 13 for the recount genres, using the
difference between serial time ('and then') and episodic time ('in another
time') as one vector and the contrast between individual participants and
generic classes of participant as another.

Proceedings
33rd International Systemic Functional Congress
2006
 28

serial time

personal
recount

individual generic
participants participants
autobiography

biography
historical
recount

episodic time

Fig. 13: A Topological Perspective on the Recount Genres

Topology is a useful tool for evaluation, when teachers need to reason

about how close an approximation a student's text is to an ideal type in
the field. Topology also makes it easier to build genesis into the picture
and reason about effective learner pathways for apprenticeship into a
discipline. One such pathway is outlined in Fig. 13, which attempts to
optimise a series of steps taking students from the spoken genres they
bring to school through the written genres they need for assessment and
public examination purposes.

In Muller's terms it would probably be fair to say here that this kind of
analysis verticalises the discipline. It also grammaticalises history,
although we will not have time to develop this point here - since each
genre is conceived as a configuration of meanings (at the levels of
register, discourse semantics, lexicogrammar and
phonology/graphology) related to recurrent instances of language use
(cf. the discussion of time and cause in history discourse in section 3
above). The experience of the Sydney school is that when this kind of
deconstruction of a discipline is shared by teachers and students then

Proceedings
33rd International Systemic Functional Congress
2006
 29

access to the discourses of the discipline can be far more widely

distributed.

multi-sided

exposition through discussion

one sided internal

explanation to argument cause

external
cause rhetoric
account to explanation

chronology cause
recount to account

time setting
story to history in time

sequence
in time

Fig. 14: A Learner Pathway for Secondary School History Genres

The price that must be paid for this redistribution of discursive resources
is that history teachers have to get comfortable with a knowledge
structure from the social sciences which has much more verticality and
grammaticality than their own. By the same token, science teachers
have to get used to a knowledge structure with less verticality and
grammaticality, and allow in effect for a horizontalisation of their
knowledge structure since a new language of description (SFL) has
entered their discipline. The different contexts for the social semiotic
metadiscourse engender different stances of resistance. On the
humanities side, teachers worry that creativity and critique are being
Proceedings
33rd International Systemic Functional Congress
2006
 30

sacrificed in the interests of overly schematic and indoctrinating rules.

On the science side, teachers worry that crucial content will have to be
set aside in order to make room for intruding domains of technicality.
My concern here is not to address these anxieties (see Christie 1992,
Cope & Kalantzis 1993, Feez 2002, Giblett & O'Carroll 1990, Macken-
Horarik 2002, Martin 1991, 1993c, 1998, 1999b, 2000a, b, 2001 for
discussion), but rather to draw attention to their origin - namely the
intrusion of a distinctive social semiotic instructional discourse into the
pedagogic discourse of a foreign knowledge structure.

7. Envoi

In this paper I have presented a brief synopsis of SFL work on field,

contrasting science and history with respect to both activity sequence
and taxonomy. Bernstein and Muller's complementary sociological
tradition of research was then introduced, and key relations to SFL
research reviewed. In particular the relation of grammatical metaphor to
vertical discourse, of technicality to verticality and of hierarchies and
complementarities to grammaticality was highlighted. These and further
dimensions of this dialogue are developed in Christie and Martin 2007.

Bernstein and Muller's conception of knowledge structure was in

addition deployed to reflect on the nature of SFL itself as a type of
vertical discourse, including discussion of one challenge it presents for
current definitions of hierarchical discourse - namely its multi-nocular
vision in relation to its various complementarities (such as metafunction)
and complementary hierarchies (e.g. realisation, instantiation and
individuation). The precise nature of Muller's verticality and
grammaticality in SFL has important implications wherever SFL is
deployed as a metadiscourse for disciplines, since such interventions
inevitably involve a recontextualisation of the nature of disciplinary
knowledge. This raises crucial issues about democratisation of learning
and teacher reactions to SFL interventions, which can perhaps now be
pursued with deeper understanding and polemic defusing insight.

Proceedings
33rd International Systemic Functional Congress
2006
 31

REFERENCES

BERNSTEIN, B. 1996. Pedagogy, Symbolic Control and Identity:

theory, research, critique. London: Taylor & Francis. [Revised
Edition 2000].
_____. 1999 Vertical and horizontal discourse: an essay. British Journal
of Sociology of Education. 20(2): 157-173.
CHRISTIE, F. 1992. Literacy in Australia. ARAL 12. 142-155.
_____. [Ed.]. 1999. Pedagogy and the Shaping of Consciousness:
linguistic and social processes. London: Cassell (Open Linguistics
Series).
_____. 2002. Classroom Discourse Analysis: a functional perspective
(Open Linguistics Series). London: Continuum.
CHRISTIE, F. & J R MARTIN. 1997. Genre and Institutions: social
processes in the workplace and school. (Ed. with F Christie).
London: Cassell.
_____. [Eds.]. 2007. Knowledge Structure: functional linguistic and
sociological perspectives. London: Continuum.
COFFIN, C. 1997. Constructing and giving value to the past: an
investigation into secondary school history. Christie & Martin [Eds.]
196-230.
_____. 2000. Defending and challenging interpretations of the past.
Revista Canaria de Estudios Ingleses. 40. 135-154.
_____. 2006. Historical Discourse: the language of time, cause and
evaluation. London; Continuum.
COPE, W & M Kalantzis [Eds.]. 1993. The Powers of Literacy: a genre
approach to teaching literacy. London: Falmer (Critical
Perspectives on Literacy and Education) & Pittsburg: University of
Pittsburg Press (Pittsburg Series in Composition, Literacy, and
Culture).
EGGINS, S, J R MARTIN & P WIGNELL. 1993. The discourse of
history: distancing the recoverable past. M Ghadessy [Ed.] Register
Analysis: theory and practice. London: Pinter (Open Linguistics
Series). 1993. 75-109.
FEEZ, S. 2002. Heritage and innovation in second langauge education.
in Johns [Ed.] 43-69.
GHADESSY, M [Ed.]. 1993. Register Analysis: theory and practice.
London: Pinter.
_____ [Ed.]. 1999. Text and Context in Functional Linguistics.
Amsterdam: Benjamins (CILT Series IV).
Proceedings
33rd International Systemic Functional Congress
2006
 32

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]
HALLIDAY, M A K. 1998. Things and relations: regrammaticising
experience as technical knowledge. in Martin & Veel. 185-235.
_____. 2003. On Language and Linguistics (Vol 3 in the Collected
Works of M A K Halliday. J Webster Ed.). London: Continuum.
_____. 2004. The Language of Science. London: Continuum (Vol 5 in
the Collected Works of M A K Halliday J Webster Ed.). London:
Continuum.
HALLIDAY, M A K & J R MARTIN. 1993. Writing Science: literacy
and discursive power. London: Falmer (Critical Perspectives on
Literacy and Education).
HALLIDAY, M A K & C M I M MATTHIESSEN. 1999. Construing
Experience through Language: a language-based approach to
cognition. London: Cassell.
_____. 2004. An Introduction to Functional Grammar. (3rd Edition)
London: Arnold.
JOHNS, A M [Ed.]. 2002. Genre in the Classroom: applying theory and
research to practice. Mahwah, NJ : Lawrence Erlbaum.
KRESS, G & T van LEEUWEN. 1996/2006. Reading Images: the
grammar of visual design. London: Routledge.
MACKEN-HORARIK, M. 2002. 'Something to shoot for': a systemic
functional approach to teaching genre in secondary school science.
Johns [Ed.] 17-42.
MARTIN, J R. 1991. Critical literacy: the role of a functional model of
language. Australian Journal of Reading 14.2. 117-132. [Focus
Issue on Literacy Research in Australia edited by Peter Freebody &
Bruce Shortland-Jones]
_____. 1992. English Text: system & structure. Amsterdam: Benjamins.
_____.1993a. Technology, bureaucracy and schooling: discursive
resources and control. Cultural Dynamics 6.1. 84-130.
_____. 1993b. Life as a noun. in Halliday & Martin. 221-267.
_____. 1993c. Genre and literacy - modelling context in educational
linguistics. ARAL 13. 141-172.
_____. 1998. Practice into theory: catalysing change. S Hunston [Ed.]
Language at work: Clevedon: Multilingual Matters (British Studies
in Applied Linguistics 13 [BAAL 13]). 1998. 151-167.
Proceedings
33rd International Systemic Functional Congress
2006
 33

_____. 1999a. Mentoring semogenesis: 'genre-based' literacy pedagogy.

Christie [Ed.]. 123-155.
_____. 1999b. Linguistics and the consumer: theory in practice.
Linguistics and Education 9.3. 409-446.
_____. 2000a. Design and practice: enacting functional linguistics in
Australia. Annual Review of Applied Linguistics 20 (20th
Anniversary Volume ‘Applied Linguistics as an Emerging
Discipline’). 116-126.
_____. 2000b. Grammar meets genre – reflections on the ‘Sydney
School’. Arts: the journal of the Sydney University Arts Association
22. 47-95. [reprinted in Educational Research on Foreign
Languages & Arts Sun Yat Sen University, Guangzhou (Special
issue on Functional Linguistics & Applied Linguistics) 2. 2006. 28-
54.
_____. 2001. Giving the game away: explicitness, diversity and genre-
based literacy in Australia. R de Cilla, H Krumm & R Wodak et al.
[Eds.] Functional Il/literacy. Vienna: Verlag der Osterreichischen
Akadamie der Wissenschaften. 2001. 155-174.
_____. 2002a. Writing history: construing time and value in discourses
of the past. C Colombi & M Schleppergrell [Eds.] Developing
Advanced Literacy in First and Second Languages. Mahwah, N.J.:
Erlbaum. 87-118.
_____. 2002b. From little things big things grow: ecogenesis in school
geography. R Coe et al. [Eds.] The Rhetoric and Ideology of Genre:
strategies for stability and change. Cresskill, N.J.: Hampton Press.
2002. 243-271.
_____. 2003. Making history: grammar for explanation. J R Martin & R
Wodak (eds.) Re/reading the past: critical and functional
perspectives on time and value. Amsterdam: Benjamins. 2003. 19-
57.
_____. 2006. Metadiscourse: designing interaction in genre-based
literacy programs. R Whittaker, M O'Donnell & A McCabe [Eds.]
Language and Literacy: functional approaches. London:
Continuum. 90-117.
_____. 2007. Construing knowledge: a functional linguistic perspective.
in Christie & Martin. 34-64.
MARTIN, J R & D ROSE. 2003. Working with Discourse: meaning
beyond the clause. London: Continuum.
_____. 2005. Designing literacy pedagogy: scaffolding asymmetries (J R

Proceedings
33rd International Systemic Functional Congress
2006
 34

Martin & D Rose). R Hasan, C M I M Matthiessen & J Webster

[Eds.] Continuing Discourse on Language. London: Equinox. 2005.
251-280.
_____. 2006. Genre Relations: mapping culture. London: Equinox.
MARTIN, J R & R VEEL (eds.). 1998. Reading Science: critical and
functional perspectives on discourses of science. London:
Routledge.
MARTIN, J R & PR R WHITE. 2005. The Language of Evaluation:
appraisal in English. London: Palgrave
MARTIN, J R & R WODAK (eds.). 2004. Re/reading the past: critical
and functional perspectives on discourses of history. Amsterdam:
Benjamins.
MATON, K & J MULLER. 2007. A sociology for the transmisson of
knowledges. in Christie & Martin. 14-33.
MESSEL, H. 1963. Science for High School Students. Sydney: Nuclear
Research Foundation.
MULLER, J. 2000. Reclaiming Knowledge: social theory, curriculum
and education policy. London: Routledge (Knowledge, Identity and
School Life Series 8).
_____. 2007. On splitting hairs: hierarchy, knowledge and the school
curriculum. in Christie & Martin. 64-86.
O'HALLORAN, K. 2007. Mathematical and scientific forms of
knowledge: a systemic functional multimodal gramamtical
approach. in Christie & Martin. 205-236.
SCHLEPPEGRELL, M J. 2004. The Language of Schooling: a
functional linguistics perspective. Mahwah, N.J.: Erlbaum.
UNSWORTH, L. 1997a. "Sound" explanations in school science: A
functional linguistic perspective on effective apprenticing texts.
Linguistics and Education, 9.2, 199-226.
_____. 1997b. Scaffolding reading of science explanations: Accessing
the grammatical and visual forms of specialised knowledge .
Reading, 313:30-42.
_____. 1997c. Explaining explanations: Enhancing science learning and
literacy development. Australian Science Teachers Journal, 43.1,
34-49.
_____. 1999. Teaching about explanations: Talking out the grammar of
written language. In Watson, A. and Giorcelli, L. Eds. Accepting the
literacy challenge. Sydney: Scholastic.
_____. 2001. Evaluating the language of different types of explanations

Proceedings
33rd International Systemic Functional Congress
2006
 35

in junior high school science texts. International Journal of Science

Education 236:585-609.
_____. 2004. Comparing school science explanations in books and
computer-based formats: The role of images, image/text relations
and hyperlinks. International Journal of Instructional Media, Vol
31/3:283-301.
WHITE, P. 1998. Extended reality, proto-nouns and the vernacular:
distinguishing the technological from the scientific. Martin & Veel.
266-296.
van LEEUWEN, T & S HUMPHREY. 1996. On learning to look
through a geographer's eyes. in Hasan & Williams. 29-49.
VEEL, R. 1992. Engaging with scientific language: A functional
approach to the language of school science. Australian Science
Teachers Journal. 38. 4. 31-35.
WIGNELL, P, J R MARTIN & S EGGINS. 1990. The discourse of
geography: ordering and explaining the experiential world.
Linguistics and Education 1.4. 359-392 [republished in Halliday &
Martin 1993. 136-165.]
WIGNELL, P. 2007. Vertical and horizontal discourse and the social
sciences. in Christie & Martin. 184-204.

i
Web address: http://www.bom.gov.au/info/clouds/
ii
Additional hierarchies include rank (constituency), instantiation (system to instance),
individuation (reservoir to repertoire) and genesis (unfolding text, individual
development and cultural evolution).
iii
Additional complementarities include axis (system and structure), perspective
(synoptic or dynamic) and modality (language, image, music etc.).

Proceedings
33rd International Systemic Functional Congress
2006