Chapter 3

Analysing Academic Texts

Chapter Overview
This chapter introduces a theoretical framework and metalanguage that
researchers and teachers can use for analysing how language is used in
academic contexts, in particular the variation of language according to dif-
ferent subject domains and the recurrent genres in these domains. The special
features of genres specific to different academic subjects will be examined
and how text analysis can be conducted by teachers and researchers to inform
teaching will also be discussed.

3.1 A Functional View of Language

Academic language can be analysed using different linguistic theories. For instance,
traditional school grammar books which draw on a structuralist linguistic theory
usually teach sentence grammatical structures such as the ‘passive voice’, and the
passive voice is said to be a feature of academic language (e.g. science laboratory
reports). However, we do not get a lot of mileage if we analyse academic texts only
focusing on linguistic structures without seeing these structural patterns as resources
for achieving communicative purposes. A functional view of language will enable us
to both ask and answer questions such as Why is the passive voice used frequently in
academic genres such as laboratory reports or academic theses? What commu-
nicative function(s) does this linguistic structure realize in these types of texts? Can
the function(s) be realized or achieved by using other kinds of linguistic structures?
Are there disciplinary variations in these patterns? Are there some functions which
are more important in science than in History, for instance? Is a similar function
realized by different linguistic features in different subjects at different levels (and in
different languages and cultures)? Furthermore, how can language teachers and

© Springer Science+Business Media Singapore 2016 29

A.M.Y. Lin, Language Across the Curriculum & CLIL in English
as an Additional Language (EAL) Contexts, DOI 10.1007/978-981-10-1802-2_3
30 3 Analysing Academic Texts

different content subject teachers develop metalinguistic awareness about these

questions and help their students identify and appreciate linguistic and functional
variations across subject domains (and across different languages such as L1 and L2)
so that they can make useful connections and comparisons of different academic
language styles across different subjects (and languages)?
The approach to linguistic analysis of academic texts adopted in this chapter
draws mainly on the seminal work by Michael Halliday, Raquia Hassan, Clare
Painter, Jim Martin and David Rose in systemic functional linguistics (SFL) and the
Sydney School of genre analysis and genre-based pedagogy. However, other lin-
guistic and genre theories will also be drawn upon when they are relevant to the
topic in focus. The difficulty created by the technical terminology of SFL will also
be mitigated by drawing on some of the traditional pedagogical grammar termi-
nology that most teachers and students are familiar with.
A functional approach to language analysis ‘looks at how language enables us to
do things in our daily lives’ (Derewianka 2011, p. 3) or how we mobilize language
as a resource to understand and construct (or ‘construe’—i.e. construct using
semiotic resources) the world around us, our social relationships, as well as our
texts in both spoken and written modes.
A succinct summary of the SFL assumptions about language is presented by
Derewianka (2011):
• Language is a dynamic, complex system of resources for making meaning.
• Language reflects the culture in which it has evolved. It is not a neutral medium,
but expresses certain world views, values, beliefs and attitudes.
• Our language choices change from situation to situation, depending on the
social purpose for which language is being used, the subject matter, who is
involved and whether the language is spoken or written.
• The emphasis in language study is on how people use authentic language in
various contexts in real life to achieve their purposes… [e.g.] on the language
needed for successful participation in school contexts.
• A knowledge of grammar can help us to critically evaluate our own texts and
those of others (e.g. identifying point of view; examining how language can be
manipulated to achieve certain effects and position the reader in different ways;
knowing how language can be used to construct various identities or a particular
way of viewing the world) (Derewianka 2011, p. 3; words in square brackets
added).
We can add to the above list the importance of multimodal and new media texts
as increasingly we are immersed in not just spoken/written linguistic texts but also
linguistic texts that are ‘meshed with’ visuals, (moving) images, hyperlinks [to
other texts/images, music and sounds—in short, multimodalities (see Kress and van
Leeuwen 2006)]. Also, language use in both primary/secondary and tertiary aca-
demic contexts will be discussed in this book.
3.2 The ‘Genre Egg’: A Metalanguage for Dissecting the Language Learning Task 31

3.2 The ‘Genre Egg’: A Metalanguage for Dissecting

the Language Learning Task

A functional view of language focuses on analysing language use in context (i.e. as

text-in-context) rather than on analysing language as abstract patterns and elements
detached from real people using language as a resource to achieve their social
purposes in real-life situations. In order to do this, an analytical framework is needed.
Martin (2010) provides a very good visual summary (Fig. 3.1) of the SFL model
of how language is structured as hierarchical patterns at different layers (called
strata). Linguistic analysis starts at the most macro-stratum of genre (e.g. analysis of
the primary social goals of a genre) to the stratum of register (e.g. analysis of how
the social goals of the genre interact with and shape the three dimensions—field,
tenor and mode—of the register that affects language choices). The analysis then
proceeds to the stratum of discourse semantics (e.g. analysis of how a text is
schematically structured into stages, phases and messages to achieve its primary
social goal) and to the stratum of lexico-grammar (e.g. analysis of how morphemes1
combine to form words and how words combine to form groups, clauses and
sentences) and ultimately to the microstratum of phonology/graphology (e.g.
analysis of how phonemes combine to form syllables and tone groups).
In the above linguistic analysis, we do not start with the most microlevel
(phonology/graphology) working up to the macro-levels of register and genre. That
is, we do not start our linguistic analysis using a bottom-up approach, which is the
way in which linguistic students and language teachers in education courses are
usually taught (i.e. a bottom-up approach to linguistic analysis of the target lan-
guage). As Rose (2015) points out, this bottom-up approach needs to be changed in
order to help language learners to be able to apply bottom-level linguistic knowl-
edge in authentic contexts of language use (i.e. to connect bottom-level patterns to
register and discourse patterns):
A similar [bottom-up] assumption underlies traditional language pedagogies–namely
that by teaching the grammatical structures of classical and modern languages, linguistic
analysis skills transfer to other learning tasks. Although grammar is explicitly taught,
transference depends on students intuitively applying these skills to register and dis-
course patterns (Rose 2015, p. 5; words in square brackets added).

Rose (2015) continues to point out that this approach might work for some
students, but other students might not benefit from it. In fact, many students might
be turned off by the boredom of this bottom-up approach. Or, if they can bear with
it to pass the tests and exams, it is very likely that they cannot apply the
bottom-level linguistic knowledge (e.g. grammar knowledge) in authentic contexts
of language use (i.e. in real-life registers and genres to achieve authentic commu-
nicative goals). The Reading to Learn (R2L) genre-based pedagogy (Rose 2010,
2015; Rose and Martin 2012) was developed to recontextualize this bottom-up
approach to language analysis by starting from the level of genre and register
patterns and helping students to connect these macro-discourse patterns to
lexico-grammatical patterns.
32 3 Analysing Academic Texts

Language as text
in social context

genre register discourse lexico- grapho-

contexts of contexts of semantics grammar phonics
culture: situation: texts, text clauses, letter patterns,
purpose & field, tenor segments & groups/phrases , sounds,
staging & mode cohesion words punctuation

Recognising genre
& predicting
potential staging
Interpreting register
in terms of own existing
experience

Following semantic sequence

recognising semantic segments
and cohesive relations

Processing lexicogrammar
realisational relationships
Reading language between semantics & grammar

strata & spelling

of letter patterns
in words

Fig. 3.1 Stratal and rank hierarchies in the linguistic model of SFL (image from Martin 2010,
Slide 40; reproduced here by permission of Prof. Jim Martin)

The SFL model of language strata and instantiation of meanings provides a

theoretical framework to understand and design studies on learning, curricular and
pedagogical issues. It answers questions such as How is language structured and
organized? How are these linguistic patterns hierarchically related? How do they
simultaneously instantiate (i.e. exemplify) social meanings in context? From the
SFL perspective, every time we produce or comprehend an instance of language,
we are at once immersed in two contexts: (i) the context of the multiple levels of
3.2 The ‘Genre Egg’: A Metalanguage for Dissecting the Language Learning Task 33

contrasts (or options) in the language system (i.e. paradigmatic contrasts: What
could go instead of what—choices about selection), and (ii) the context of the
immediate contrasts in the unfolding text (i.e. syntagmatic contrasts: What goes
together with what—choices about combination). The following family interaction
reported in Painter (1993) helps to illustrate this:
Father: This car can’t go as fast as ours.
Child: I thought–I thought all cars could–all cars could go the same–all cars could go
the same (pause) fast…
Mother: The same speed.
Child: Yes, same speed.
(Painter 1993, cited in Rose 2012a, p. 3)

In this example, the child is guided through the mother–child interaction in the
context of shared experience (both the mother and child are in the car sharing the
here-and-now context) to develop mastery of the linguistic contrast between ‘fast’
and ‘speed’ within the linguistic system of lexico-grammar (i.e. the contrast
between an adjective and a noun). At the same time, the child is also immersed in
the shared social context of interaction (i.e. the unfolding conversation text). Prior
to the mother’s provision of the right word (‘speed’), the child seems to be
struggling to find the appropriate linguistic item (from his fledging language sys-
tem) to express his meaning, hence the pause before his coming up with the word
‘fast’, which has got the semantic meaning right but not the lexico-grammatical
contrast (permitted by the language system) right (i.e. it is an adjective instead of a
noun). This struggling effort seems to be reflected in his shifting extra conscious
attention to finding the right linguistic structure from the linguistic system (of
English) in order to instantiate a social meaning that he wants to contribute to the
ongoing conversation (that all cars can go the same speed—that the father’s
statement needs to be corrected or qualified).
L2 learners, likewise, also often have this experience of struggling to find the
right linguistic structure or contrast (from their fledging mastery of the L2 system)
to instantiate a meaning which is often important in the context of ongoing social
interaction. This linguistic struggle is one that many English language learners
(ELLs) can resonate with: they feel that they have something important to say in
this matter (e.g. in the ongoing academic argument) but only that they cannot find
the right linguistic means to do so. In the same vein, ‘focus on form/focus on
meaning’ is the researcher’s analytic term to capture these quick moments of
shifting extra conscious effort/attention between the twin contexts that every
speaker, writer or user of language seems to be experiencing all the time (whether it
is in one’s L1, L2, L3 …). Notice that the mother’s linguistic scaffolding (provision
of the right linguistic structure) is just in time and just in need (Gee 2003). In second
language acquisition (SLA) theoretical terms, it seems to be a focus-on-form (FonF)
technique (Lyster and Ranta 1997) that the mother is using (a recast: ‘the same
speed’) which has resulted in the child’s noticing and subsequent uptake (i.e. using
34 3 Analysing Academic Texts

the correct form ‘speed’ instead of ‘fast’) without interrupting the conversation
flow. This focus on form would not be perceived by the child as equally helpful if
the mother were to give the child a preconversation drill on the conversion between
adjectives and nouns (e.g. fast $ speed), not to mention the fact that the mother
could not have anticipated all the specific linguistic needs of the child as they arise
moment-to-moment in everyday conversations. Also, chances are that the child will
remember this linguistic feature better, and more importantly, how to use it in the
appropriate context, as it is provided to him just when he is struggling to put his
meaning into words (notice that it is his meaning and not the mother’s meaning).
All these will have important implications in our discussion on how to integrate
content and language learning in Chap. 7.
How does this linguistic theory help us conceptualize the nature of the language
learning task confronting the student? With a series of schematic representations of
what I call the ‘Genre Egg’, Rose illustrates the different aspects of the language
learning task based on the notion of text-in-context, which is delineated in
(Fig. 3.2a–c).
Let us first look at Fig. 3.2a. The diagram conceptualizes the language learning
task as one of learning to understand and produce not just a text but a text-in-
context. That is to say, a text (whether spoken or written) is always a text produced
and understood in context. The learner’s task of understanding a text-in-context
involves first understanding the primary social goal of the text (genre) and the three
dimensions of the context—field, tenor and mode (register):
• What it’s about—its subject matter (field);
• Who is involved—such as writer and readers, teacher and students, parent and
child; and
• The social purpose of the text—what the speakers, writers and readers are trying
to achieve (i.e. the social goal of the genre which the text instantiates).
Figure 3.1a shows that the linguistic text(-in-context) is crafted out at different
linguistic levels: the levels of discourse (text), grammar (sentence) and spelling
(word). In other words, the learner needs to simultaneously understand the text’s
contextual aspects (genre goals and register dimensions) as well as its linguistic
aspects (e.g. linguistic choices made at the levels of discourse, grammar and
spelling). As Rose (2010) delineates:
This model of language as ‘text-in-context’ is derived from the theory of systemic
functional linguistics (SFL). It seems like common sense because SFL is a theory of
how people make meaning in language (Halliday 1994; Martin and Rose 2007), so it is
very useful for investigating how language works, and how it is learnt, and then for
designing effective language teaching strategies. (Rose 2010, p. 8)

So, how does the SFL theoretical framework inform us when we design lan-
guage teaching and learning strategies? Figure 3.2b, c shows two different ways of
approaching the language teaching/learning task: (i) the disintegrating approach and
(ii) the integrated approach. In Fig. 3.2b, under the disintegrating approach, the
language learning task is disintegrated into separate tasks such as reading and
3.2 The ‘Genre Egg’: A Metalanguage for Dissecting the Language Learning Task 35

Fig. 3.2 a–c Illustrating (a) The language learning task–text-in-context

different conceptualizations of
the language learning task
(images from Rose 2013,
context genre/register
Slide 3, 7 and 8; reproduced
here by permission of text discourse
Dr. David Rose)
sentence grammar

word spelling

(b) Disintegrating the language learning task

context
reading and
text
listening
paragraph
tasks
grammar sentence
exercises
word group
vocabulary word
activities syllable

pronunciation letter pattern

(c) An integrated approach

context
text
paragraph
sentence
word group
word
syllable
letter pattern
36 3 Analysing Academic Texts

listening to texts, grammatical exercises, vocabulary activities and pronunciation

practice. These tasks might be done separately and might not need to follow any
particular sequence. In contrast, in Fig. 3.2c, under the integrated approach, the
language learning task is always approached by reading the text-in-context while at
the same time drawing students’ attention to the linguistic choices that the author
made at different linguistic levels (e.g. paragraph, sentence, word group, word…) to
achieve the overall communicative purpose of the text-in-context.
To understand the theoretical basis of these two approaches as discussed by
Rose, let us revisit in more detail the SFL theory of the stratified organization of
language as text-in-context and the relation between language systems and instances
in texts, i.e. stratification and instantiation. As mentioned in Sect. 3.1, SFL theo-
rizes language as a hierarchical system of different stratified layers (i.e. strata) of
patterns of different combinations of elements, which together instantiate meanings
(see Fig. 3.1). Rose (2010) delineates stratification and instantiation as follows:

Stratification refers to the organisation of language and its social contexts as a

hierarchy of levels or strata. The relation between strata is modeled in SFL as
realisation. Thus patterns of meaning in texts (or discourse semantics) are realised
(manifested/symbolized/expressed) by function of words in clauses
(lexico-grammar), which are realised by patterns of sounds or letters (phonology or
graphology). Looking up the hierarchy to social context, language enacts social
relations between speakers (tenor), construes the activities they are involved in
(field), and plays various roles in doing so (mode). Collectively, field, tenor and
mode are referred to as register … and together realise the global social purpose of a
cultural context or genre. …
… Instantiation refers to the relation between features in language systems and
instances of meaning in actual texts. Thus each genre and its attendant register
variables (field, tenor, mode) is a specific instance of the language system as a
whole. Within each genre we can then distinguish more variable sub-types, and each
text is recognizable as an instance of one of these types. Instantiation occurs at all
language levels, for example, … sequences of phonemes in a word instantiate
phonological systems. (Rose 2010, pp. 1 and 3; italics added)

The disintegrating approach (Fig. 3.2b) is one that many of us are familiar with,
e.g. the Chinese practice of teaching children to write by starting with tracing the
pattern of the strokes to form a Chinese character. The phonics approach is also an
example of explicit teaching of bottom-level linguistic (e.g. phonological,
graphological) patterns by helping students to form letter–sound relationships early
on so that they can have the skills to decode or ‘sound out’ new words. The
disintegrating or bottom-up approaches are criticized by top-down approaches such
as the whole language approach (Goodman 2005) which emphasizes literacy
learning in holistic meaningful contexts and de-emphasizes explicit teaching of
bottom-up patterns and skills. In L2 learning, the top-down approach is manifested
3.2 The ‘Genre Egg’: A Metalanguage for Dissecting the Language Learning Task 37

as the communicative language teaching (CLT) approach (Littlewood 1981).

Migrant children, linguistic minorities and L2 learners (e.g. EAL learners), how-
ever, might need to be explicitly taught some of the bottom-level skills as they often
do not have enough naturalistic experience with the L2 to infer these
patterns/relationships themselves without explicit teaching. How to resolve the
tension between bottom-up approaches (often criticized as decontextualized) and
top-down approaches (often criticized as neglecting the development of basic
language skills) remains a key question in the literature (e.g. No Phonics against
Whole Language). Rose (2010) summarizes this situation well:

Different approaches to literacy try to handle the complexity of learning to read and
write in different ways, depending on the particular theory of language they come
from.
• Phonics, phonemic awareness and basal reading book programs start at the
bottom, with sounds and letter patterns, then words, then phrases, then
sentences.
• ‘Sight word’ approaches and spelling lists focus on recognizing words and their
letter patterns.
• Grammar activities in school and ESL programs focus on rules for word groups
and sentences.
• Traditional composition writing focused on sentences in paragraphs.
• Whole language and critical literacy approaches focus at the top, on what the
text is about. This also includes shared big book reading in the early years.
• Genre writing (text types) starts with the context, then focuses on the staging of
texts, as well as various language features.

Most teachers use a ‘balanced approach’ that addresses the various parts of reading
and writing tasks with a combination of strategies. However, each activity may be
done in a separate part of the day’s program, using different texts, sentences, words
and letter patterns. For children with rich experience of reading in the home, each of
these activities is meaningful, so they can put them together and develop as readers
and writers. But children without such experience often struggle to understand and
synthesize all these activities, and so develop more slowly. (Rose 2010, p. 11)

Up to now, the reader might think that the integrated approach is similar to the
top-down approach. However, Rose’s notion of the integrated approach is actually
very different from the top-down approach. To Rose, the top-down approach errs in
not providing enough scaffolding to the learner in acquiring the bottom strata
patterns. To Rose, in the extreme form of top-down approaches,
all explicit teaching of language features was rejected from both the classroom and
teacher training, leading to generations of students and teachers without the rudimentary
knowledge of language afforded by traditional school grammars (2012a, p. 4)

Rose’s integrative approach refers to the Sydney School genre-based pedagogy,

which seeks to integrate both bottom-up and top-down approaches by proposing a
teaching/learning cycle (TLC); as Rose (2012a) explains:
38 3 Analysing Academic Texts

…[genre] pedagogy begins not with low level language features nor with a gener-
alized notion of communicative contexts but with the specific social purposes and
staging of written genres. Furthermore, its starting point is not with decontextualised
language systems but with instances of actual texts. In the teaching/learning cycle
designed by Joan Rothery and colleagues (Rothery 1994), an instance of a genre is
‘deconstructed’ by the teacher and students by reading it together and guiding
students to recognise its stages and key relevant language features. After decon-
structing the model text, teacher and students then jointly construct a new text, using
similar organisation and key language features, but writing about a field that they
have built up together. (Rose 2012a, p. 4)

These different layers of pedagogical activity—building content (field), analys-

ing (deconstructing) the genre and jointly constructing a text—all prepare students
for the task of constructing a new text of their own. The teaching/learning cycle
(Rothery 1994/2008) is schematized in Fig. 3.3. The Sydney School of genre-based
pedagogy will be discussed in more detail in Chap. 5. In this chapter, we shall
mainly look at the first layer of the teaching/learning cycle: deconstructing or
analysing the text.
Our discussion has so far focused on different conceptualizations of and
approaches to the language learning task. The content teacher might be asking this
question: What has the language learning task got to do with my content teaching?
Since one fundamental principle underlying this book is the assumption that lan-
guage and content cannot be separated, the learner’s task of learning content cannot
be separated from the task of learning the kinds of linguistic resources that are

Fig. 3.3 The

teaching/learning cycle
(image from Martin and
Matthiessen 2014, Fig. 9.6,
p. 149; reproduced by
permission of Springer)
3.2 The ‘Genre Egg’: A Metalanguage for Dissecting the Language Learning Task 39

essential to construing (or constructing) the content in a specific field or domain.

Below, we shall focus our discussion on how we can develop a metalanguage for
both content teachers and language teachers to talk about and analyse academic
texts in different content fields.

3.3 Analysing Academic Texts in Content Subject

Domains

The ‘Genre Egg’ framework is useful in providing a metalanguage (or a common

vocabulary) for both content teachers and language teachers to work together to
analyse academic texts found in content subject areas. Without a common vocab-
ulary, it is almost impossible to foster collaboration between the language spe-
cialists and content specialists as they are typically trained in different disciplines
with different theories and concepts underpinning their pedagogical practices. For
instance, a math teacher once said to me, ‘In math lessons we focus on commu-
nication, not language’. At that time, I found it hard to make sense of her sentence,
precisely because she seems to be making a semantic contrast between ‘commu-
nication’ and ‘language’ as if the two are not related, or perhaps, what she wanted to
say is that math teachers do not focus on highlighting language aspects but just
focus on getting messages across. Many language teachers might have a similar
experience when trying to communicate with content teachers about language
matters (and how language matters in content learning and teaching). On the other
hand, content teachers might find it hard to communicate with language teachers as
they are often put off by the language teacher’s use of technical linguistic terms
(e.g. gerunds, imperatives and type I/II conditionals).
How would the ‘Genre Egg’ framework provide an accessible common
vocabulary for both content specialists and language specialists to communicate
with each other about the language demands of academic texts and genres found in
content subject domains? I have developed an adapted version of the SFL Genre
Egg (Fig. 3.4) to present to both content teachers and language teachers in seminars

Fig. 3.4 The ‘Genre Egg’

(from Lin 2010)
40 3 Analysing Academic Texts

on language across the curriculum. Usually, a practical concern of teachers is how

they can provide language support to students learning content subjects in an L2
and how content teachers and language teachers can collaborate in providing this
support. With the Genre Egg as a common analytical framework, both content
teachers and language teachers can conduct analysis of the linguistic demands at
different levels (e.g. vocabulary, sentence patterns, language functions, genre
conventions) of key academic texts in a subject domain and then to collaborate on
designing tasks that would provide language support to their students (more on task
design in Chaps. 5 and 6). In the following, I shall demonstrate how the adapted
‘Genre Egg’ can assist teachers and researchers in conducting analysis of academic
texts and how the analysis can inform our teaching.

3.3.1 Analysing Academic Vocabulary

The research literature speaks of three general types of academic vocabulary

(Mercuri 2010). The first type is field-specific, technical vocabulary, for instance, in
the field of science, e.g. water cycle, pollination, antioxidant, partition coefficient
and photodiode. The second type is general academic vocabulary which is found in
academic texts across a range of subjects. For instance, Coxhead (2000) has col-
lated a list of 570 high-utility academic word families. The word family of ‘analyse’
will contain words such as analysis, analyser, analytical and analytically. The third
type is linking words or logical connectors such as however, in contrast, firstly and
secondly, which indicate the logical relationships between different parts of the text.

Application Scenario 3.1

In Text 3.1, can you find examples of the three types of academic vocabulary?
Use a different colour to highlight the three different types of academic
vocabulary.

Text 3.1
In the process of photosynthesis, carbon dioxide is taken in by plants through
the stomata of their leaves. Simultaneously, the plants release oxygen and
excess water through the stomata, providing us with fresh air.

Application Scenario 3.1 represents a simple focus-on-form exercise that

teachers can use to raise both their own and their students’ academic language
awareness. Examples of field-specific technical vocabulary are photosynthesis,
carbon dioxide and stomata. Examples of general academic vocabulary include
release and excess. An example of logical connectors is Simultaneously. It is
3.3 Analysing Academic Texts in Content Subject Domains 41

important to notice that the boundaries between field-specific vocabulary can be

blurred as more and more technical words have spilled into everyday life through
repeated exposure in the mass media, e.g. QE (quantitative easing, subconscious-
ness, antioxidant, high-maintenance). Likewise, the boundary between field-specific
vocabulary and general academic vocabulary can be porous, e.g. ‘the water cycle’ is
a technical name given to a process in science and yet the word ‘cycle’ is found in
many other academic texts as a productive element in the formation of names of
field-specific processes (e.g. the teaching/learning cycle; the recession cycle).
Sometimes, the field-specific technical vocabulary looks like everyday vocabulary
and can lead to misunderstanding of academic concepts. For instance, words such
as ‘force’ and ‘pressure’ in physics have specialized definitions, and if students
interpret them using their everyday life understanding of these words, confusion can
arise.
The aim of this kind of simple vocabulary analysis is to gauge the language
demands of a text at the vocabulary level. For instance, if there is too high a
concentration of academic vocabulary, the text might need to be adapted to suit the
proficiency level of students in a particular class. For instance, ‘release’ can be
replaced by ‘give out’; ‘excess’ can be omitted without considerably changing the
meaning of the text. While this will be very useful for EAL students, this will also be
relevant to L1 speakers of English especially those students coming from disad-
vantaged backgrounds or those who speak a local variety of English as a home
language. On the other hand, if the texts used in a subject curriculum are all sim-
plified texts that do not provide students with enough exposure to field-specific
vocabulary, then some intervention needs to take place (more on this in Chaps. 4–6).
Content teachers and language teachers can also discuss how they can design
enrichment tasks and coordinate their efforts in helping students master these dif-
ferent kinds of vocabulary across the curriculum (more on this in Chap. 7).

3.3.2 Analysing Sentence Patterns that Realize Language

Functions

Lexico-grammatical patterns (or ‘sentence patterns’, which is a term that can be

more easily understood by both content and language teachers) realize a range of
language functions that are commonly found in academic texts, e.g. comparing and
contrasting, exemplifying, defining, classifying, interpreting, hypothesizing, pre-
dicting, giving evidence and expressing conditional or causal relationships. More or
less similar lists of functions are given different names under different theoretical
frameworks, e.g. knowledge structures (Mohan 1986; Kong 2009); aspects of the
scientific method (Zimmerman 1989); rhetorical functions (Hirvela 2004); and
language functions (ELDAC). I have chosen to call them ‘language functions’
following ELDAC as this term can be easily understood by teachers. An example of
42 3 Analysing Academic Texts

an important language function in academic texts is defining. Let us analyse Text

3.2 to illustrate how this function is realized in lexcio-grammatical (sentence)
patterns.

Application Scenario 3.2

Can you find the sentence that realizes the function of defining? Can you
analyse the lexico-grammatical (sentence) pattern of such sentences?

Text 3.2
Preservatives are additives that maintain the freshness and quality of food.
They prevent food from spoilage caused by mould, bacteria and yeast and
from flavour and colour changes due to exposure to oxygen.
Many countries have laws that ensure that manufacturers list all preservatives
used together with the amounts on the ingredient part of the label. Chemical
names such as sodium nitrate and sodium benzoate are often found on the
labels of food products.

You would notice that the author of this Grade 4 science text chooses to define
‘preservatives’ right at the beginning of the text just as this term is introduced:
‘preservatives are additives that maintain the freshness and quality of food’.
Similarly, Text 2.2 in Chap. 2 has a similar pattern: ‘flowering plants are classified
as high-class plants’. The field-specific term (‘flowering plants’ and ‘preservatives’)
is bolded to highlight its key term status, and it is immediately defined by first
classifying it into a category of entities (e.g. ‘high-class plants’ and ‘additives’).
Then, some unique features are provided (although this part is omitted in the
sentence in the flowering plants text).
A sentence pattern that is useful in realizing the language function, defining, can
thus be outlined (Table 3.1).
Here, we minimize the linguistic terminology to make this sentence pattern
easily grasped by content teachers who do not have a linguistic background. When
content teachers read this text with students, it is useful to highlight at some point
how useful language functions such as defining can be realized by sentence patterns
like this. As teachers guide the students to experience instances of defining like this

Table 3.1 A sentence-analysis/sentence-making table to analyse and generate useful sentences to

do defining
X ¼ Y
Preservatives are additives that maintain the freshness and quality of
food.
Technical Relating General class Clause/phrase giving specific
term verb word characteristics
3.3 Analysing Academic Texts in Content Subject Domains 43

in repeated encounters with them in different academic texts, the academic language
awareness of students will be raised. They will start to become not just information
readers or form readers (Cai 2014), but also rhetorical readers (Hirvela 2004) or
writerly readers, i.e. they will now have an eye for noticing the lexico-grammatical
resources (e.g. words and sentence patterns) useful in achieving different rhetorical
or language functions such as defining, which they can use themselves when they
are constructing an academic text of their own (e.g. in assignments, projects, pre-
sentations or examinations). It is important to highlight to students that there are
usually many more diverse ways of achieving a rhetorical or language function
although some basic linguistic resources (such as the sentence pattern outlined
above) can be useful to start with. Students can be encouraged to keep a ‘writerly
reader’s notebook’ where they jot down instances of different recurrent functions
which they come across in different texts in different subject areas.
My research associate Emily He and I have analysed the Sarasas Science Corpus
(which we have built from the grades 1–6 science textbooks used in and published
by the Sarasas Affiliated Bilingual Schools, Thailand) and have found many more
instances of the ‘defining’ function in the corpus (Table 3.2).
The key point in deconstructing/analysing academic texts is thus to heighten the
academic language awareness of both (content/language) teachers and students so
that each individual experience (or encounter) with a curriculum text becomes a
learning opportunity to infer the linguistic resources (e.g. vocabulary and sentence
patterns) useful for achieving functions, and these resources can come in useful
when students are constructing texts of/on their own (i.e. they become writerly
readers—reading with an eye to becoming a writer themselves). It is important that
students are encouraged to discover these patterns from the texts they read in their
subject domains (initially under the teacher’s guidance), and they can keep a
writerly reader’s notebook on these patterns, instead of just teaching them a list of

Table 3.2 Instances of the language function defining in the Sarasas Primary Science Corpus
(from Lin and He 2014)
Terms ! General class Specific details
Fertilizers are compounds made to support plants’ growth.
Vitamins are organic found in fruits, vegetables, also in meat, eggs, milk
compounds and animals’ internal organs.
Minerals are organic found in vegetables, fruits, milk, meat, egg yolks
chemical and all kinds of seafood.
elements
Calorie is a unit of energy used as a measurement for the amount of energy a
particular food provides.
Flowers are the structures where reproduction takes place.
Fruits are ripened ovary that contain seeds.
walls of flowers
Fertilization is the process where the male’s pollen grains fuse together with
the female’s ovules inside the ovary and become
one new cell.
44 3 Analysing Academic Texts

decontextualized sentences outside of the curriculum context. In other words, these

instances of language functions need to be experienced and noticed in a meaningful
text-in-context. And this ‘noticing’ process (or ‘focus on form’) must not impede
content learning (i.e. not turning a content lesson into a language lesson), and this
requires skilful ‘shifting’ between focus on form and focus on content on the part of
the teacher. We shall discuss this in more detail in Chap. 5.
There are many other useful language functions such as exemplifying. In
Text 3.2, can you find the sentence pattern(s) used to achieve this academic
function? Notice that there are often diverse ways of achieving a similar function in
different texts and genres. Different researchers have come up with different cate-
gories of functions. For instance, researchers of the English Language Development
Across the Curriculum (ELDAC) Project have come up with a list of 19 language
functions (ELDAC Functions Index, see Department of Education Queensland,
1989). Kong and Hoare (2008), following the knowledge structure framework of
Mohan (1986) and Tang (1992), have come up with a list of knowledge structures,
which resemble what other researchers call functions. Cutting-edge research is
being conducted by Dalton-Puffer (2013) on cognitive discourse function
(CDF) which seeks to provide a comprehensive and yet teacher-friendly list of
CDFs (more on this in Chap. 9).
Whatever theoretical traditions or functional taxonomies we choose to follow, it
is important to recognize the need to allow students the opportunity to discover how
these functions are realized in texts that are meaningful to them. A pitfall exists for
teachers to organize their lessons simply according to a list of ‘functions—sentence
patterns’ that are presented to students in decontextualized ways. It is very
important to help students to see how these functions contribute to achieving the
overall communicative purpose of a text in a specific genre (e.g. a descriptive
report) rather than as a set of isolated functions standing on their own.
Academic texts in tertiary education are usually much more complex, and
functions are generally realized with much more sophisticated lexico-grammatical
patterns that can be outlined in a few basic sentence patterns. They are also tightly
related to the generic structuring or organization of the text to achieve the overall
purpose of the text. Teachers can encourage students to read with a ‘writer’s eyes’
to see how these functions are typically realized in context. Teachers can continue
to raise students’ academic language awareness to a point when students can see
these patterns on their own. We shall discuss more of this in Chap. 5.

3.3.3 Analysing Academic Genres in a Curriculum Context

In the Genre Egg (Fig. 3.5) that guides our analysis of academic texts, the layer
embedding language functions and vocabulary is text type or ‘genre’. While genre
is defined slightly differently under different theoretical traditions (see review of the
three traditions by Hyon 1996), the Sydney School’s definition seems to be most
useful to teachers:
3.3 Analysing Academic Texts in Content Subject Domains 45

Fig. 3.5 Genres in the school curriculum (from Rose 2012b, Slide 12; reproduced here by
permission of Dr. David Rose)

The Sydney School approach starts with a broad definition of genres as ‘staged
goal-oriented processes’: they are goal-oriented because a text unfolds towards its social
purpose, and staged because it usually takes more than one step to reach the goal.
Genres evolve in a culture to achieve common social purposes that are recognised by
members of the culture so that the stages they go through are generally predictable for
members of the culture. (Rose 2012a, p. 1)

Genres are thus patterned ways of organizing our speaking and writing for
specific communicative purposes in specific contexts. To succeed in school or
university, a student needs to master a number of key academic genres for different
academic subjects, e.g. to write a book review for the English class, to write an
expository essay for a social studies assignment, to write a descriptive report on
endangered species or to write a research proposal or research report for the science
or engineering project. Genres are usually introduced to students as ‘text types’,
though we must remind students of the fluid and evolving nature of genres so that
students see text types as helpful tools rather than static, set-in-stone templates for
speaking and writing.
Different theoretical traditions have approached genre analysis using slightly
different terminologies, but they basically follow the same procedure of identifying
46 3 Analysing Academic Texts

Table 3.3 Genre analysis of a Grade 4 science text—flowering plants

Introduction
Classifying Flowering plants are classified as high-class plants.
Description 1: adult At the adult stage, they produce flowers which develop into fruits and
stage seeds after being pollinated and fertilized.
Description 2: Tulips, water lilies, mangoes and bananas are examples of flowering
examples plants.

stages and phases in a text (called ‘moves and steps’ in John Swales’ genre analysis
tradition) as the text unfolds to achieve its primary communicative purpose.
If we revisit Text 2.2 in Chap. 2 (which is reproduced in Table 3.3 with genre
structure analysis in the left margin), we shall notice that this short Grade 4 science text
is an instance of a descriptive report under the taxonomy of school genres developed
by the Sydney School of genre analysis (Martin and Rose 2008, 2012). A descriptive
report usually has two stages: Introduction ^ Description (the symbol ^ is used to
indicate ‘followed by’). In the Introduction stage, the topic is introduced, usually by
defining or classifying it. Then, the text unfolds into the Description stage, where more
descriptive details are given. While the stages are quite predictable across different
instances of the genre, the phases under each stage can be quite variable, and so instead
of prescribing a template for writing a descriptive report, students can be encouraged
or guided to discover both the predictable stages and the variable phases across
different texts in different curricular contexts.
The school genres identified by the Sydney School researchers are divided into
three main types depending on their global communicative purpose: Informing,
Engaging, Persuading (Fig. 3.6). David Rose, in particular, has written a series of
booklets entitled Reading to Learn (http://www.readingtolearn.com.au/) which
presents the Sydney School genre-based pedagogy in teacher-friendly language
with many practical examples drawn from genre analysis of the Australian school
curriculum texts.

Fig. 3.6 Grammatical metaphor: shifts in grammatical class and functional status
3.3 Analysing Academic Texts in Content Subject Domains 47

The Sydney School genre researchers have mostly worked on analysing school
genres and have made great contribution to the teaching of academic literacies in
school settings. As for genre analysis of academic texts in university settings, it is
the English for Academic Purposes (EAP) and English for Specific Purposes
(ESP) studies which are influential in the literature. John Swales and his colleagues
have conducted genre analysis on research writing genres, catering for the L2 EAP
needs of international students in universities in the US Swales’ CARS (Creating A
Research Space) model (1990) for writing in research genres which is classic now
and frequently drawn upon in academic writing courses in universities.

3.4 Technicality and Abstraction

It is generally rather easy to distinguish academic writing from other kinds of

writing, as academic language is usually characterized by a high level of technicality
and abstraction, two notions introduced by Halliday and Martin (1993). Halliday
(2004) describes the ‘history’ of the evolution of language forms on the individual
plane (ontogenesis), on the societal plane (phylogenesis), as well as on the textual
plane (logogenesis). ‘History’ is here understood as having three dimensions:
(1) the history of the child’s language development,
(2) the history of the evolution of language, and
(3) the unfolding of the text (or information flow of the text).
On the ontogenetic plane, the child goes through three stages:
• from protolanguage to language,
• from everyday spoken grammar to the grammar of written language (or liter-
acy), and
• from the grammar of written language to that of the language of the subject
disciplines (e.g. science, mathematics, geography, history).
Speaking in terms of knowledge development, the critical moments are when the
child learns to develop additional layers of knowledge:
• Common-sense knowledge (age 1–2),
• literate educational knowledge, (age 4–6),
• technical knowledge (age 9–13) and
• theoretical knowledge (age 18 onwards).
Developing the additional layers of educational and technical knowledge, the
child needs additional language resources to construe and configure these new
layers of knowledge. The language for these additional layers of knowledge is
characterized by increasing lexical density but decreasing grammatical delicacy (or
clausal complexity). Let us illustrate these ideas with the hypothetical ‘repacking’
examples provided by Halliday (2004: 31–32):
48 3 Analysing Academic Texts

1. Look—it must be raining! People have their umbrellas open.

The above sentence might very naturally be said by a 3-year-olds, and Halliday
shows how this sentence can be ‘repacked’ step by step going up the age range (age
is included in brackets at the end of each sentence):
2. How can you tell that it’s raining? You can see that people have got their umbrellas
open. (6)
3. We can prove that there’s rain falling by seeing that people’s umbrellas are open. (9)
4. What best proves that it’s rainy weather is the fact that the umbrellas have been
extended. (12)
5. The best proof that the weather is pluvious is the fact that the umbrellas are extended.
(15)
6. The truest confirmation of the pluviosity of the weather is the extendedness of the
umbrellas. (18 up)

To successfully participate in school work, the child needs to learn how to

repack sentence 1 into sentences 2 and 3 in primary school and further into sen-
tences 4–6 in secondary school. As the child gets increasingly apprenticed into
school ways of writing, the sentences that they produce are marked by increasingly
complex nominal groups but decreasing grammatical delicacy. Grammatical deli-
cacy or intricacy refers to the complexity of clause structure. For instance, sentences
2–3 are dominated by verbal clauses and their structure can be schematically rep-
resented as follows:
Sentence 2: How can you tell A? You can see X.
(A = that it’s raining) (X = that people have got their umbrellas open)

Sentence 3: We can prove A by seeing X.

(A = that there’s rain falling) (X = that people’s umbrellas are open.)
On the other hand, sentences 4–6 have a simpler sentence structure but
increasingly more generalized and abstract nominal (i.e. noun) groups:
Sentences 4–6: B is Y.
(B = What best proves A’) is (Y = the fact that X’),
(B = The best proof that A’’) is (Y = the fact that X’’),
(B = The truest confirmation of A’’’) is (Y = X’’’),
Wherein:
A’ = that it’s rainy weather,
A’’ = that the weather is pluvious,
A’’’ = the pluviosity of the weather,
X’ = the umbrellas have been extended,
X’’ = the umbrellas are extended, and
X’’’ = the extendedness of the umbrellas.
We can notice that the nominal groups (B, Y) in sentences 4–6 are becoming
increasingly abstract, and they function to re-present a dynamic process into a static
nominal entity. This abstraction process takes away the specificities of the ‘here and
3.4 Technicality and Abstraction 49

now’ of what is happening (it’s raining) and turns it into a general, impersonal,
atemporal, static and abstract concept (the pluviosity of the weather). However, the
clause structure is a relatively simple one, i.e. a relational clause (B = Y). It can be
said that the child initially lives in a world of rich clause complexes (e.g. If you
don’t give me …. I’ll tell Mum about it…!), and upon entering the school, the child
starts to encounter both technical and abstract nominalizations; e.g., preservatives
are additives that help maintain the freshness and quality of food. In this sentence,
preservatives and additives are both technical terms. Preservative is a nominalized
entity; i.e., the verb preserve is turned into the noun preservative (to refer to the
chemicals that function to preserve food) and becomes further technicalized—it is a
technical term. The same process has taken place with the verb add which is
converted into the noun additive (chemicals that are added to food) and is given
field-specific, technical meaning in the discipline of science. The adjective fresh is
turned into the noun freshness; however, it has not gained the status of technical
term and can readily be unpacked back into everyday language and its meaning is
not field-specific (i.e. not technical). We can see that technicality and abstraction are
the result of linguistic transformation processes, which are called nominalization
and grammatical metaphor. Both are technical terms themselves in the discipline of
linguistics and need to be unpacked with further explanations below.

3.4.1 Nominalization and Grammatical Metaphor: The

Linguistic Engine for Constructing Technicality
and Abstraction

Consider the adjective, ‘hot’. When it is used in everyday language, one can say,
‘Be careful, the water is hot!’ However, in a science textbook, the adjective ‘hot’
becomes nominalized (i.e. turned into a noun) as ‘heat’, which is then turned into a
technical term that can be classified into different types, e.g. latent heat and radiant
heat. Scientists can also talk about ‘heat transfer’ (e.g. it would be difficult to talk
about ‘heat transfer’ if there is only the word ‘hot’ without the technical term ‘heat’
in the language of science). Sometimes, the L1 of the students might not encode or
construe technicality in the same way as English. For instance, the Chinese word for
‘heat’ (technical term, a noun) and ‘hot’ (everyday word, an adjective) is the same:
熱 and this has an impact on Chinese students’ learning of the concept of ‘heat
transfer’ (Fung and Yip 2014). Another example is the verb, ‘move’. When it is
used in everyday language, it is easy for an EAL learner to pick it up in conver-
sations, e.g. ‘Move on! Quick!’ However, in a science textbook, the verb ‘move’
becomes nominalized as ‘motion’ and becomes a technical term, as Halliday
explains:
So, for example, when we turn move into motion we can say things like all motion is
relative to some fixed point; we can set up laws of motion, and discuss problems like that
of perpetual motion; we can classify motion as linear, rotary, periodic, parabolic,
50 3 Analysing Academic Texts

contrary, parallel, and the like. Not because the word motion is a noun, but because in
making it a noun we have transformed ‘moving’ from a happening into a phenomenon
of a different kind: one that is at once both a happening and a thing. … By calling
‘move’ motion, we have not changed anything in the real world; but we have changed
the nature of our experience of the world. (Halliday 2004, pp. 15–16; italics original)

There is no mystery in academic language as we can actually trace the origins of

academic language (and their technicality and abstraction) in everyday interper-
sonal conversations. For instance, Halliday mentions the example of his son, Nigel,
when he was 3:
When my son was small, he used to play with the neighbour’s cat, which was friendly
but rather wary, as cats are with small children. On one occasion he turned to me and
said ‘Cats have no other way to stop children from hitting them; so they bite’. He was
just under three and a half years old. Some years later, in primary school, he was reading
his Science textbook. One page was headed: ‘Animal Protection’; and underneath this
heading it said ‘Most animals have natural enemies that prey upon them. … Animals
protect themselves in many ways. Some animals … protect themselves with bites and
stings.’ (Halliday 2004, pp. 12–13)

So, in repackaging a verbal process (so they bite) into a nominal entity (some
animals protect themselves with bites), the primary science textbook has
re-presented the child’s common-sense knowledge as school knowledge or edu-
cational knowledge. If one needs to help an L1 child to go through these linguistic
transformation processes in order to succeed in school, helping L2 learners (e.g.
EAL students) to unpack academic language into everyday language as well as to
repack everyday language into academic language (e.g. in writing assignments and
examinations in schools or writing research reports or papers in university) becomes
an even more important curricular and pedagogical design question when L2 is
used as the medium of instruction (MOI) in schools. In order to do this, it is worth
spending some more effort in understanding the lexico-grammatical resources that
have evolved in the English language (and in many other languages as well, e.g.
Chinese) to construct technicality and abstraction in different academic disciplines
through nominalization and the use of grammatical metaphor.
Technicality ‘refers to the use of terms or expressions … with a specialised
field-specific meaning’ (Halliday and Martin 1993, p. 144). In the example above,
the word ‘bites’ has not been turned into a technical term. It functions mainly to
make school language more abstract and to package information in a more compact
manner (e.g. with higher lexical density). However, the disciplines of physical
sciences (physics, chemistry, biology, earth science) have employed a process of
technicalizing which involves two steps: (i) naming the phenomenon and
(ii) making that name technical (i.e. with field-specific meaning) (Halliday and
Martin 1993).
To support students in tackling technical academic texts, this two-step process
can be highlighted to students to show how a term has become technicalized in a
specific discipline. This explicit discussion can heighten students’ awareness of
how everyday words are transformed into technical terms (e.g. add ! additives;
preserve ! preservatives). Likewise, students can be explicitly engaged in
3.4 Technicality and Abstraction 51

discussing the different technical (i.e. field-specific) meanings that different disci-
plines give to seemingly similar terms (e.g. the word ‘field’ has a very different
meaning in science, Mathematics and daily life, respectively). Once these
action-processes (verbs) are turned into entities (nouns), a lot of things can be done
with them, e.g. you can pluralize them (e.g. additives), qualify them (e.g. food
additives) and ‘tag’ more information onto them (e.g. modern-day food additives).
Technicality is closely linked to an important practice that the disciplines of
modern physical sciences have evolved to embrace. Modern sciences are basically
about naming, defining, describing, classifying phenomena and establishing hier-
archies of taxonomies of these phenomena (Halliday and Martin 1993). Mastering
the academic subject science is about mastering these taxonomies which consist of
technical terms that enter into different taxonomic oppositions (e.g. living things vs.
non-living things; flowering plants vs. non-flowering plants; vertebrates vs. inver-
tebrates; plant cells vs. animal cells). Naming, defining, classifying, describing,
exemplifying, comparing and contrasting and so on thus become important cog-
nitive discourse functions that students need to learn to master in relation to the
subject content of modern sciences. These functions are simultaneously cognitive
and linguistic/discursive as they require students to apply the technical terms and
taxonomies (embedded in the specialized discourses of the disciplines) to name,
define, describe, compare, contrast and classify different physical (and social)
phenomena. Learning content in the science subject is thus a semiotic process, i.e.
learning to use the technical terms and taxonomies (i.e. specialized discourses)
handed down from the traditions of the disciplines to see (or construe) the world
around them or to make technical sense of (or technical meaning out of) their
everyday experience (i.e. to turn or reconfigure their experience into technical
knowledge) (Lemke 1990).
However, technicality is only half of the story of the evolution of the academic
language in the past five hundred years (Halliday 2004). Analysing the science
writings of key figures (e.g. Bacon, Descartes, Newton) in Western science,
Halliday comes to the conclusion that starting from the sixteenth century and
increasingly so into the nineteenth centuries, science writings in the Western tra-
dition have gone through not just a technicalizing process but also an abstracting
process. Specifically, these writings have used the lexico-grammatical resources of
nominalization and grammatical metaphor to construe the technical and abstract
knowledge of their disciplines. We have explained nominalization above, and let us
explain grammatical metaphor below.
Grammatical metaphor is closely linked to nominalization. When a nominalized
word or group functions as if it were a grammatical participant (e.g. grammatical
subject or object in traditional grammar terminology), it is called a grammatical
metaphor. For instance, consider the following clause and its nominalized
counterpart:
clause: a planet moves in an elliptical orbit
nominal group: the elliptical orbital motion of a planet
52 3 Analysing Academic Texts

Figure 3.6 shows a schematic explanation of how the verb moves which func-
tions as a process (in the original clause: a planet moves in an elliptical orbit)
becomes nominalized as motion and functions as a thing in the nominal group (the
elliptical orbital motion of a planet). This analysis is modelled on the analysis
offered by Halliday and Martin (1993) in their example:
clause: an electron moves in an orbit
nominal group: the orbital motion of an electron (Halliday and Martin 1993, p. 128).

This nominal group (‘the orbital motion of an electron’) can in turn function as a
constituent further embedded in a more complex nominal group:
the combined motion of an electron resulting from the coincidence of the orbital with the
rotational motion [X] (Halliday and Martin 1993, p. 129; [X] is added by the author)

In principle, such further embedding can go on and on to create an increasingly

complex and compact nominal group [X] which can function as a participant (e.g.
as a grammatical subject or object) in a sentence that has a simple relational
structure: [X] is/is known as [Y] (where both [X] and [Y] are called participants in
Halliday’s systemic functional grammar), for instance:
The combined motion of an electron resulting from the coincidence of the orbital with
the rotational motion is known as…[Y]

Halliday calls this ‘grammatical metaphor’ to contrast with lexical metaphor. To

unpack the meaning of grammatical metaphor, let us start with an example of a
lexical metaphor, which is easier to understand, e.g. ‘Juliet is cold to her father’. We
know that ‘cold’ here is a metaphor because it is based on comparison or analogy
with temperature (e.g. the weather is cold ! she is cold to her father). But now let
us consider another kind of metaphor, e.g. the coldness of Juliet to her father is due
to her love for Romeo. This is an example of grammatical metaphor. What is
originally an adjective or a quality of things (‘cold’) gets nominalized into a noun or
a thing (‘coldness’), which now functions in another sentence as a grammatical
participant (as the grammatical subject) of the sentence—hence the name, gram-
matical metaphor.
We are, of course, not trying to turn Shakespeare’s play into a technicalized or
abstract academic text by writing modern-day ‘fan fiction’ on it. However, if we do
this experiment of taking a literary work and transforming the text into one full of
nominalizations and grammatical metaphors, we can see how a literary text can be
transformed into an academic text through mobilizing what Halliday calls the
lexico-grammatical resources (the linguistic powerhouse) of the English language.
In literary writing, accomplished writers use lexical metaphors to achieve the
purpose of engaging the audience by turning some abstract processes into concrete,
visualizable processes. For instance, consider the following sentence from Suzanne
Collins’ popular fiction, The Hunger Games—Catching Fire:
‘Just the sound of his voices twists my stomach into a knot of unpleasant emotions like
guilt, sadness and fear.’ (Collins 2013, p. 11)
3.4 Technicality and Abstraction 53

We know that ‘twists’ and ‘knot’ are lexical metaphors as their meanings here
are based on analogy with the concrete action of twisting (verb) something into a
knot (a physical entity). By employing these lexical metaphors, Collins visualizes
for the reader vividly the sudden invisible change of emotions in the female pro-
tagonist (Katniss Everdeen) upon hearing the voices of President Snow.
In academic writing, we use grammatical metaphor for just the opposite effect:
turning what is concrete and everyday into something abstract and technical. But
why do scientists do that? Halliday (2004) argues that the use of grammatical
metaphor in scientific writing enables the writer to construe not only technicality
but also rationality. To understand this, we need to turn to the next topic: thematic
progression and logical flow.

3.5 Thematic Progression and Logical Flow

Nominalization and grammatical metaphor play an important role in construing

rationality (Halliday 2004, p. 124) by enabling the writer to construct logical
semantic relations in the text. Logical semantic relations are not a privilege of
scientific or academic texts. Hasan (1992) shows the importance of reasoning in the
conversation of three-year-old children. However, what is special about scientific
discourse, according to Halliday, is:
(1) that it constructs an argument out of a long sequence of connected steps, and
(2) that at any one juncture a large number of previous steps may be marshalled together
as grounds for the next. (Halliday 2004, p. 124)

The language unit for construing one such step is a clause (e.g. ‘If you don’t take
the medicine’). A clause is both a unit of experience and a unit of information.
Clauses are the building blocks of an argument. Consider the following hypo-
thetical conversation (A child is coughing hard but refuses to take the medicine and
his mum tries to ‘reason’ him into taking it):
Mother (to Child): If you don’t take the cough syrup, you’ll be coughing all night.
Coughing all night will make your Mum and Dad unable to sleep well. Not sleeping
well will make us unable to do our job well tomorrow. Not doing our job well will make
us lose our jobs. Losing our jobs will make us unable to buy you the computer games
you want…

To understand how the mother constructs the flow of information and the logic
of her argument, let us do a theme–rheme analysis of the above utterances. The
theme is the stable part, the anchor or the point of departure, and it is typically a
noun or a nominal group (usually the subject of the sentence, together with any
minor clause or phrase). It is also the given (or shared) information. The rheme is
the new information or the focus (usually the main clause) in a sentence or utter-
ance. Table 3.4 shows a theme–rheme analysis of the utterances.
We can see in the above hypothetical example that nominalization takes place to
‘pack’ or summarize the rheme (the main clause) of the previous sentence into the
54 3 Analysing Academic Texts

Table 3.4 Theme–rheme analysis of the mum’s utterances

Theme (given/shared information) Rheme (new information)

1 If you don’t take the cough syrup you’ll be coughing all night.
2 Coughing all night will make your Mum and Dad unable
to sleep well.
3 Not sleeping well will make us unable to do our job well
tomorrow.
4 Not doing our job well will make us lose our jobs.
5 Losing our jobs will make us unable to buy you the
computer games you want…

new theme (a nominal group) of the next sentence. And this process repeats itself to
move the argument forward step by step.
Imagine what you would feel if we interrupt this information flow or thematic
progression by reverting the theme–rheme sequencing (i.e. put new information in
the place of the theme and old information in the place of the rheme) as in the
following reconstructed utterances of the Mum above:
Mother (to Child): If you don’t take the cough syrup, you’ll be coughing all night. Your
Mum and Dad will be unable to sleep well if you’re coughing all night. We will be
unable to do our job well tomorrow if we are unable to sleep well. We will lose our jobs
if we are unable to do our job well. We will be unable to buy you the computer games
you want if we lose our jobs…

The above example helps us to understand the ways in which scientists or

academic writers present their information systematically and construct their argu-
ment logically. Halliday uses the following example from a science text to illustrate
how the presentation of logical reasoning hinges on mobilizing the linguistic
resources of nominalization and grammatical metaphor to ‘pack’ the rheme(s) in
previous sentence(s) into the theme(s) in new sentences:
If electrons were not absolutely indistinguishable, two hydrogen atoms would form a much
more weakly bound molecule than they actually do. The absolute indistinguishability of the
electrons in the two atoms gives rise to an ‘extra’ attractive force between them. (Layer
1990, pp. 61–62; cited in Halliday 2004, p. 125; italics added)

In the theme–rheme analysis of this example (Table 3.5), we see that what has
been presented in a clause in the theme of the first sentence (‘If electrons were not
absolutely indistinguishable’) is condensed into a nominal group and condensed as
a more compact theme in the next sentence (‘The absolute indistinguishability of
the electrons in the two atoms…’). This succinctly phrased or highly condensed
packet of information serves as a point of departure and anchor to which further
3.5 Thematic Progression and Logical Flow 55

Table 3.5 Theme–rheme analysis of a science text

Theme Rheme
1 If electrons were not absolutely two hydrogen atoms would form a much more
indistinguishable weakly bound molecule than they actually do.
2 The absolute indistinguishability of gives rise to an ‘extra’ attractive force between
the electrons in the two atoms them.

new information (‘gives rise to an ‘extra’ attractive force between them’) is

attached. Halliday argues that the linguistic resources of nominalization and
grammatical metaphor enable the academic or scientific writer to achieve system-
aticity and logicality—rationality—in their writing.
Learning how to mobilize these linguistic resources to achieve a systematic
information flow and logical argument in their writing is precisely that part of
invisible learning that confronts every school child if he/she is to participate suc-
cessfully in different school subject lessons. This task is often made more difficult
by the jumbled ways of presenting information in the school textbooks, especially
those written for EAL learners as the textbook writers tried to ‘make the language
simpler’ by turning text into a cluster of bullet points or scattering the information
among pictures and visuals. My colleague, Dr. Dennis Fung, who is a science
subject specialist, laments the lack of coherent texts in many of the science text-
books produced in Hong Kong. On one occasion, we were preparing for a teacher
seminar and were looking for a coherent text to present the process of energy
conversion in a closed circuit, and we looked in several science textbooks in Hong
Kong and could not find a good model text to illustrate the kind of thematic
progression and information flow discussed above. In the end, we worked together
to reconstruct a text as a model text. How we can make this invisible learning task
visible and help both teachers and students notice the operation and functioning of
these linguistic resources in academic texts will be discussed in Chap. 5.
Note 1:
A ‘morpheme’ is the smallest unit of meaning in a language; e.g., ‘love’ has one
morpheme, while ‘lovely’ is made up of two morphemes. A ‘phoneme’ is the
smallest unit of meaning-differentiating sound, e.g. /h/ in ‘hat’ and /s/ in ‘sat’, by
changing the sound from /h/ to /s/, the meaning of the word is changed.

Chapter Summary Points

• The SFL theoretical framework and the ‘Genre Egg’ as a metalanguage
for both content teachers and language teachers to talk about and analyse
the language demands of academic texts,
• Different conceptualizations of the language learning task: the bottom-up,
top-down and integrated approaches,
56 3 Analysing Academic Texts

• The Sydney School genre-based pedagogy, the teaching/learning cycle

(TLC) and analysis of school genres (text types),
• Construing technicality and abstraction through the use of nominalization
and grammatical metaphor, and
• Theme–rheme analysis, thematic progression and information flow.

End-of-Chapter Discussion Questions

1. What kind of learning goals do you want to set for your students ultimately?
Can you use the concepts of ‘information reader’, ‘rhetorical reader’ and
‘writerly reader’ to discuss how they are related to the content or language
focuses of the lesson?
2. How can we avoid showing students a list of language functions with a number
of sentence patterns under each? How can we teach functions and the sentence
patterns realizing these functions in a meaningful, contextualized way?
3. What would you do if you find that the curricular text that you are analysing
does not fit into the genre taxonomy and the predictable stages and phases in the
existing research literature?
4. By understanding the ‘linguistic engines’ of academic language (technicality
and abstraction), is it possible to develop a systematic way to help students
unpack and repack the abstract and technical sentences of academic texts?
5. What are the practical constraints on doing a guided analysis of academic
language in class, especially the possible impact on the logical flow and
coherence of content delivery?
6. If you were a language specialist, what would be the biggest challenge in
persuading content subject teachers to pay attention to the hidden linguistic
devices that may hinder students’ understanding of the content?

References

Cai, J. (2014). An integrated genre-based approach to scaffolding novice academic writers: genre
awareness, academic lexical phrases and student uptake. Doctoral dissertation, The University
of Hong Kong, Pokfulam, Hong Kong.
Collins, S. (2013). The hunger games book 2-catching fire. New York: Scholastic Press.
Coxhead, A. (2000). A new academic word list. TESOL Quarterly, 34(2), 213–238.
Dalton-Puffer, C. (2013). A construct of cognitive discourse functions for coneptualising
content-language integration in CLIL and multilingual education. European Journal of Applied
Linguistics, 1(2), 216–253.
Department of Education Queensland (1989). English language development across the
curriculum (ELDAC). Immigrant Education Services, Division of Special Services.
Derewianka, B. (2011). A new grammar companion for teachers (2nd ed.). NWS: Primary English
Teaching Association.
References 57

Fung, D., & Yip, V. (2014). The effects of the medium of instruction in certificate-level physics on
achievement and motivation to learn. Journal of Research in Science Teaching, 51(10),
1219–1245.
Gee, J. P. (2003). What video games have to teach us about learning and literacy?. New York:
Palgrave Macmillan.
Goodman, K. S. (2005). What’s whole in whole language (20th ed.). RDR Books.
Halliday, M. A. K. (1994). An introduction to functional grammar. London: Arnold.
Halliday, M. A. K. (2004). The collected works of M.A.K. Halliday. In J. J. Webster Series (Ed.),
The language of science (Vol. 5). London: Continuum.
Halliday, M. A. K., & Martin, J. R. (1993). Writing science: Literacy and discursive power.
London: The Falmer Press.
Hasan, R. (1992). Rationality in everyday talk: From process to system. In J. Svartvik (Ed.),
Directions in corpus linguistics: Proceedings of Nobel Symposium 82, Stockholm, 4–8 August
1991 (pp. 257–307). Berlin: de Gmyter.
Hirvela, A. (2004). Connecting reading and writing in second language writing instruction. Ann
Arbor: The University of Michigan Press.
Hyon, S. (1996). Genre in three traditions: Implications for ESL. TESOL Quarterly, 30(4),
693–722.
Kong, S. (2009). Content-based instruction: What can we learn from content-trained teachers’ and
language-trained teachers’ pedagogies? The Canadian Modern Language Review, 66(2),
233–267.
Kong, S., & Hoare, P. (2008). Late immersion in Hong Kong: Still stressed but making progress?
In T. W. Fortune & D. J. Tedick (Eds.), Pathways to multilingualism: Evolving perspectives on
immersion education (pp. 242–263). Clevedon, UK: Multilingual Matters.
Kress, G., & van Leeuwen, T. (2006). Reading images the grammar of visual design (2nd ed.).
London: Routledge.
Lemke, J. L. (1990). Talking science: Language, learning, and values. Norwood, NJ: Ablex
Publishing Corporation.
Lin, A. M. Y. (2010). How to teach academic science language. Keynote speech delivered at the
Symposium on Language and Literacy in Science Learning, 24 June 2010. Hong Kong
Education Bureau (Curriculum Development Institute—Science Education Section), Hong
Kong.
Lin, A. M. Y., & He, P. C. (2014). Academic language functions and examples from the Sarasas
corpus. Unpublished manuscript. Faculty of Education, University of Hong Kong.
Littlewood, W. (1981). Communicative language teaching. Cambridge, UK: Cambridge
University Press.
Lyster, R., & Ranta, L. (1997). Corrective feedback and learner uptake: Negotiation of form in
communicative classrooms. Studies in Second Language Acquisition, 19(1), 37–66.
Martin, J. R. (2010). Modelling and mentoring: The Yin and Yang of teaching and learning from
home through school. [PowerPoint slides]. Seminar presented at the Faculty of Education, the
University of Hong Kong, Pokfulam, Hong Kong.
Martin, J. R., & Rose, D. (2007). Working with discourse: meaning beyond the clause. London:
Continuum.
Martin, J. R., & Matthiessen, C. M. I. M. (2014). Modelling and mentoring: Teaching and learning
from home to school. In A. Mahboob & L. Barratt (Eds.), Englishes in multilingual contexts
(pp. 137–163). Dordrecht: Springer.
Martin, J. R., & Rose, D. (2008). Genre relations: Mapping culture. London: Equinox.
Martin, J. R., & Rose, D. (2012). Genres and texts: Living in the real world. Indonesian Journal of
SFL, 1(1), 1–21.
Mercuri, S. (2010). Teaching English versus teaching in English: Understanding the complexity of
academic English [PowerPoint slides]. Seminar presented at the Faculty of Education, the
University of Hong Kong, Pokfulam, Hong Kong.
Mohan, B. (1986). Language and content. Reading, UK: Addison-Wesley.
58 3 Analysing Academic Texts

Rose, D. (2010). Reading to learn—Book 1—Preparing for reading and writing. Sydney: Reading
to Learn.
Rose, D. (2012a). Integrating SFL theory with literacy teaching. In Z. Yan, J. Webster, & F. Yan
(Eds.), Developing systemic functional linguistics: Theory and application. London: Equinox.
Rose, D. (2012b). Designing curriculum genres. [PowerPoint slides]. Seminar presented at the
Faculty of Education, The University of Hong Kong, Hong Kong.
Rose, D. (2013). The potential of detailed reading for second language literacy [PowerPoint
slides]. Seminar presented at the Faculty of Education, the University of Hong Kong,
Pokfulam, Hong Kong.
Rose, D. (2015). Building a pedagogical metalanguage II: Knowledge genres. In J. R. Martin
(Ed.), Applicable linguistics and academic discourse (pp. 29–58). Shanghai: Shanghai Jiao
Tong University Press.
Rose, D., & Martin, J. R. (2012). Learning to write, reading to learn: Genre, knowledge and
pedagogy in the Sydney School. Sheffield, UK: Equinox.
Rothery, J. (1994/2008). Exploring literacy in school English (write it right resources for literacy
and learning). Sydney: Metropolitan East Disadvantaged Schools Program. Republished 2008,
Sydney: Adult Migrant Education Service.
Swales, J. M. (1990). Genre analysis: English in academic and research settings. New York:
Cambridge University Press.
Tang, G. (1992). The effect of graphic representation of knowledge structures on ESL reading
comprehension. Cambridge: Cambridge University Press.
Zimmerman, F. (1989). English for science. New Jersey: Prentice-Hall.