Open Journal of Modern Linguistics
2013. Vol.3, No.4, 308-313
Published Online December 2013 in SciRes (http://www.scirp.org/journal/ojml) http://dx.doi.org/10.4236/ojml.2013.34039
Textual Metaphor from the Non-Finite Clausal Perspective
Guangdong Univer s it y of Foreign Studies, G ua n gz ho u , China
Received May 26th, 2013; revised J u ly 1st, 2013; accepted July 9th, 2013
Copyright © 2013 Qingshun He. This is an open access article distributed under the Creative Commons Attribu-
tion License, which permits unrestricted use, d is t ribution, and reproduction in any medi um, provided t he original
work is properly cited.
Grammatical metaphor in the Hallidayan sense only comprises ideational and interpersonal metaphors,
while Martin and others propose the inclusion of textual metaphor. Based on metafunctions of Systemic
Functional Linguistics, this paper analyzes the current discussions of textual metaphor, pointing out that
some textual metaphors by Martin and others are in essence representations of ideational and interper-
sonal metaphors in text, and some are not in accordance with the principles of grammatical metaphor.
Four types of textual metaphor with double functions are proposed from the perspective of non-finite
clause relators; they are (1) elaborative non-finite clauses, (2) extensive and enhancing non-finite clauses
without relators, (3) extensive and enhancing non-finite clauses with prepositions as relators, and (4) en-
hancing non-finite clauses with prepositionalized non-finite verbs.
Keywords: Grammatical Metaphor; Textual Metaphor; Double Functions; Non-Finite Clause; Relator
Grammatical metaphor proposed by Halliday (1985; 1994)
and Matthiessen (1999; 2004) includes only ideational meta-
phor and interpersonal metaphor. This propels people to think
of the inclusion of textual metaphor in grammatical metaphor.
Some systemisists around the world have been trying to carry
out research on textual metaphor from various perspectives.
However, Halliday himself has never accepted the existence of
textual metaphor, and only talked about the textual effect of
ideational metaphor (Halliday & Matthiessen, 2004: p. 642).
Based on the concept of grammatical metaphor in the Halli-
dayan sense, we will analyze the existing researches on textual
metaphor, point out their inadequacies, and then investigate
textual metaphor from the non-finite clausal perspective.
A Sketch of Grammatical Metaphor
According to SFL, any given semantic configuration can be
realized in a congruent form or various incongruent or meta-
phorical forms. Congruent forms are recognized as “the typical
ways of saying things” (Halliday, 1994: p. 343), and they are
“closer to the state of affairs in the external world” (Thompson,
1996: p. 164), while the metaphorical forms are glossed as “not
expressed through the most typical (and highly coded) form of
representation” (Halliday, 1978: p. 180) , Thus, Halliday pro-
posed the concept of grammatical metaphor, referring to “the
expression of a meaning through a lexicogrammatical form
which originally evolved to express a different kind of mean-
ing” (Thompson, 1996: p. 165) and “a rea1ignment between a
pair of strata: a remapping of the semantics on to the lexico-
grammar” (Halliday, 1998: p. 192).
Halliday classifies grammatical metaphor into two subcate-
gories: ideational metaphor and interpersonal metaphor. “The
ideational metaphor is metaphor of transitivity and the inter-
personal metaphor is metaphor of mood and modality” (Halli-
day, 1994: p. 343). Let’s take (1) as an example:
(1) a. They shredded the documents before they departed for
b. Their shredding of the documents preceded their departure
for the airport.
Here the clauses they shredded the documents and they de-
parted for the airport in (1a) are both rankshifted into nominal
groups their shredding of the documents and their departure for
the airport in (1b), and the conjunction before is verbalized into
preceded. In the process of nominalization and verbalization,
other lexicogrammatical items are also transcategorized corre-
spondently, i.e., nominal group they functioning as Thing into
adjective their functioning as Possessive Deictic, nominal
group the documents functioning as Thing into nominal group
the documents functioning as Qualifier, and prepositional
phrase for the airport functioning as Location into prepositional
phrase for the airport functioning as Qualifier. In other words,
ideational metaphors are realized through nominalization, ver-
balization and transcategorization.
Interpersonal function is realized in the system of modality
and the system of mood, in both of which interpersonal meta-
phor can occur. Halliday (1994: p. 89) distinguishes four types
of modality: probability, usuality, obligation and inclination.
Each type of modality is realized in four semantic domains:
subjective, objective, implicit and explicit. Metaphor of modal-
ity occurs when “the speaker’s opinion regarding the probabil-
ity that his observation is valid is coded not as a modal element
within the clause, which would be its congruent realization, but
as a separate, projecting clause in a hypotactic clause complex”
(Halliday, 1994: p. 354). See (2) - (4):
(2) a. Mary will know.
b. I think that Mary knows.
(3) a. Fred usually sits quite quiet.
Q. S. HE
Open Access 309
b. It’s usual for Fred to sit quite quiet.
(4) a. John is supposed to go.
b. It’s expected that John goes.
Examples (2), (3) and (4) are probability, usuality and obli-
gation types of modality respectively. Modalities in (2a), (3a)
and (4a) are realized by modal elements within the clauses, and
those in (2b), (3b) and (4b) by projecting clauses in hypotactic
clause complexes. Therefore, (2b), (3b) and (4b) are interper-
The speech roles in exchange are giving and demanding, and
the commodity exchanged are goods-and-services or informa-
tion. “These two variables, when taken together, define the four
primary speech functions of offer, command, statement and
question” (Halliday, 1994: p. 69). The four speech functions are
expressed by mood, and the speaker can select different mood
types, i.e., declarative, interrogative and imperative. The meta-
phor of mood occurs when the correspondence between speech
functions and mood types shifts. For example:
(5) a. I would do my homework first if I were you.
b. Do your homework first!
The declarative type of mood in (5a) metaphorically realizes
the speech function of command, the congruent realization of
which is (5b).
Existing Researches on Textual Metaphor
Halliday has never mentioned the concept of textual meta-
phor. Therefore, “it is debatable whether the label ‘textual
metaphor’ is really justified” (Thompson, 1996: p. 176). Sys-
temicists like Martin (1992, 1993), Thompson (1996), Hu
(1996) and Liu (2002, 2003) believe that “textual metaphor”
should be included in the study of grammatical metaphor since
the three metafunctions of SFL are intertwined. Martin is the
first systemicist proposing textual metaphor. In Martin’s think-
ing, certain discourse elements organize text rather than field,
which include meta-message relation, text reference, negotiat-
ing text and internal conjunction, all of which are text-orga-
nizing pro-forms (Martin, 1992: pp. 416-417). For example:
(6) a. I think governments are necessary at different levels
for a number of reasons.
b. “Curtsey while you’re thinking what to say. It saves time.”
Alice wondered a little at this, but she was too much in awe of
the Queen to disbelieve it .
c. Next, he inserted the key into the lock.
d. Next, he was incapable of inserting the key into the lock.
e. Let me begin by pointing out that the Federal Government
fixes up problems that occur in the community.
In (6a), the meta-message a number of reasons is the nomi-
nalization of the logico-semantic relation of reason, functioning
to organize the text, hence textual metaphor. In (6b), the second
pronoun it is a discourse anaphora, referring back to a fact,
concept or expression rather than a specific thing, hence also
textual metaphor. In (6c), next is part of the discourse content,
functioning as external reality. In (6d), next represents the
speaker’s choice of conversation roles and rhetoric devices, his
attitude and judgment, functioning as internal conjunction. This
kind of linking words realizing internal conjunction is textual
metaphor. In (6e), the negotiation structure Let me begin in
dialogue is used in monologue as internal interpretation, hence
Subsequently, in his work coauthored with Halliday, Martin
(1993: pp. 241-243) further groups textual metaphor into meta-
phorical Themes and metaphorical News, as in Examples (7)
(7) a. Between 1937 and 1945 the value of industrial produc-
tion almost doubled.
b. This increase was faster than otherwise would have oc-
(8) a. Probably that pudding never will be cooked.
b. I don’t think that pudding ever will be cooked.
According to Martin (1993), the clausal Rheme the value of
industrial production almost doubled in (7a) is nominalized as
phrasal/group Theme This increase in (7b), forming meta-
phorical Theme, hence textual metaphor. The adverbial inter-
personal Theme Probably in (8a) is transcategorized as clausal
interpersonal Theme I don’t think in (8b), hence textual Meta-
phor. If such clausal News as Australia’s scientific and re-
search capabilities were inadequate in (9a) is transferred into a
nominal group inadequacies in Australia’s scientific and re-
search capabilities in (9b), metaphorical News occurs. For
(9) a. The war had also revealed that Australia’s scientific
and research capabilities were inadequate.
b. The war had also revealed inadequacies in Australia’s
scientific and research capabilities.
Thompson (1996: p. 176) includes textual metaphor in the
category of grammatical metaphor. According to Thompson,
two types of thematic structure, i.e., thematic equatives and
predicated Themes, are textual metaphors because both need
double transitivity analyses. Thompson (1996) does not men-
tion information metaphor, but Liu (2003) argues that in cleft
sentences or pseudo cleft sentences where the new information
is placed at the beginning of a sentence, or in sentences where
phonological or graphological means suc h as contra stive stresses,
italics or uppercase letters etc. are used to highlight relevant
information, metaphorical News occurs. He also points out that
marked Themes function to foreground emphases, highlights or
cohesion within texts, hence metaphorical Themes.
Apart from these, Lassen (2003a: p. 43; 2003b: p. 283) dis-
cusses the textual effects of grammatical metaphor, proposing
five types of textual metaphor based on the structural and non
structural organizations of text, i.e., compound nouns, passive
voice, reference, non-finite clause and ellipsis, and classifies
textual metaphor into syntagmatic and paradigmatic metaphors.
See Lassen’s examples in (10) - (14).
(10) a. Straw walker rear shaft.
b. (the) rear shaft (which is) (on the) walker (which moves
(11) a. The grain is moved to the front of the top sieve.
b. (A mechanism) moves the grain to the front of the top
(12) Connect (the) pipes to (the) c ylin der.
(13) a. Operate the valve, checking for continuous flow.
b. Operate the valve and check for continuous flow.
c. Operate the valve while (you) check for continuous flow.
(14) Cleaning shoe drive belt (is) slipping.
According to Lassen (2003), Target straw in (10b) changes
into Classifier in (10a), and Actor and Circumstance walker
into Thing; active Process in (11b) changes into passive Process
in (11a), Actor a mechanism into Agent, and Aim the grain into
Medium. Therefore, (10a) and (11a) are syntagmatic textual
metaphors. The omission of the definite article the in (12) and
that of the is in (14) are the result of systematic choices. (13a)
has two possible interpretations, i.e., (13b) and (13c). Therefore,
Q. S. HE
(12), (13a) and (14) are paradigma t i c t e x t ual metaphors.
Inadequacies of the Research on
According to SFL, “language evolved, in the human species,
in two complementary functions: construing experience, and
enacting social processes” (Halliday & Matthiessen, 1999: p.
xi). The ideational metafunction is a way of knowing the world
and construing experiences, and is realized by the system of
transitivity. In ideational metafunction, grammatical metaphor
is realized as rankshift through nominalizing Process into Actor
or Circumstance. Interpersonal metafunction is realized by the
system of Mood and that of Modality. In interpersonal meta-
function, grammatical metaphor is realized through transcate-
gories among categories of Mood and those among categories
of Modality. “The text base provides the resources that enable
the speaker to produce contextualized discourse and to guide
the listener in interpreting it” (Halliday and Matthiessen, 1999:
p. 12), and so “it is a second-order mode of meaning” (ibid, p.
398). “‘Metaphor’ in general is intrinsically a ‘second-order’
phenomenon in language” (Taverniers, 2006), that is, a gram-
matical structure is substituted for by another grammatical
structure, which leads to the occurrence of an extra meaning.
For example, (15a) is a clause of mental process, which is
rankshifted as what you want and is identified with this, and is
realigned as a relational clause of the identity type (15b), form-
ing a thematic equative, hence ideational metaphor .
(15) a. You want this.
You want this
Senser Process Phenomenon
b. What you want is this.
What you want is this
Identifier Process Identified
Similar analysis also functions in predicated Themes. For
(16) a. We have learned how to use it.
We have learned how to use it
Carrier Process Attribute
b. It is we who have learned how to use it.
It is we who have learned how t o us e it
According to Halliday and Matthiessen (1999: p. 242), gram-
matical metaphor is the result of rankshift or transcategorization.
“The grammatical metaphor thus shifts both the rank and the
class” (Halliday, 1998: p. 40). However, metaphorical Themes
and metaphorical News proposed by Martin (1993) are not the
result of rankshift or transcategorization in the system of
Theme or in the system of information; they are descriptions of
ideational metaphor from different perspectives. For example,
according to Martin (1993), the clausal Rheme the value of
industrial production almost doubled in (7a) is nominalized
into phrasal/group Theme This increase in (7b), resulting in the
occurrence of ideational metaphor in (7b). From the perspective
of textual metafunction, This increase in (7b) is the unmarked
Theme. In the system of information, it acts as the Given in-
formation, and in the system of cohesion, it functions as refer-
ence, hence no grammatical metaphor occurring in textual
metafunction. The adverbial interpersonal Theme Probably in
(8a) is transferred into the clausal Theme I don’t think in (8b)
from implicit objective category to explicit subjective category,
hence metaphor of Modality occurring. However, adverbial
interpersonal Theme and clausal interpersonal Theme are both
unmarked Themes, so the clausal interpersonal Theme is not
the necessary condition to produce textual metaphor. Similarly,
for example, the nominalization from clausal New information
Australia’s scientific and research capabilities were inadequate
in (9a) into phrasal/group New information inadequacies in
Australia’s scientific and research capabilities in (9b) is also
ideational metaphor, rather than textual metaphor.
Although Huang (2009) also discusses textual metaphor, he
does not agree that thematic equatives and predicated Themes
are textual metaphors. According to Huang (2002), metaphoric
Themes or thematic metaphors are actually ideational meta-
phors appearing in text.
In addition, changing the thematic and informational struc-
tures has the effect of rhetorical or textual emphases. However,
marked Thematic structures or marked informational structures
are not grammatical metaphors resulting from rankshift or
transcategorization. “Grammatical metaphor can be considered
as a marked form” (Goatly, 1996). However, “when we say
markedness is relevant to metaphor, we do not mean that mark-
edness is itself grammatical metaphor” (Fan, 2001: p. 27). In
fact, incongruence and markedness are not the same thing, and
in specific contexts, metaphoric form may also be unmarked
(Ravelli, 2003). For example, in the system of polarity, positive
form is unmarked, and negative form, marked; and in the sys-
tem of tense, present and past tenses are unmarked, and future
tense, marked (Halliday and James, 2005). However, we can
not say that negative form and future tense are grammatical
Ravelli (2003) does not accept metaphor of cohesion pro-
posed by Martin. According to Ravelli (2003), if the items used
to organize texts are grammatical metaphor, abstract nouns
such as fact are also grammatical metaphors. Abstract nouns
can be called “‘meta-comments’ rather than grammatical
metaphor” (Derewianka, 2003). Interestingly, Martin (1997)
also clearly distinguishes abstract Things from grammatical
metaphor, and puts that abstract nouns like fact are not at all
Since grammatical metaphor is looked at “as variation in the
expression of a given meaning” (Halliday, 1994: p. 342) rather
than as variation in the function of a given expression of inter-
nal cohesion and external cohesion, then textual metaphor pro-
posed by Martin (1992) is to be reconsidered. In fact, the
metaphors of cohesion proposed by Martin (1992) should be
understood as various grammatical functions of a given expres-
sion. This does not meet the conditions for creating grammati-
Q. S. HE
Open Access 311
cal metaphors, hence not textual metaphor.
The change of Target straw in (10b) into Classifier in (10a)
takes place at the semantic level, while the change of gram-
matical met aphor ta ke s plac e at the le xico -gramma tic al le vel. In
this sense, the compound noun in (10a) is not the metaphoric
form of (10b). In addition, a nominal group itself has not its
own structures of Theme, information or cohesion. The passive
voice is a marked syntactic structure, and the changing of active
voice to passive voice is not a process of rankshift. The change
from Actor in active voice to Agent in passive voice and the
change from Goal to Medium both take place at the semantic
level, with no rankshift or transcategorization at the lexico-
grammatical level, hence no grammatical metaphor occurring.
The ellipsis of the definite article in (12) can make the state-
ment vague and difficult to interpret, but ellipsis itself is not
According to Lassen (2003), (13a) is textual metaphor be-
cause a non-finite clause may have multiple possible interpreta-
tions. However, multiple interpretations can only show that
such a structure may be ambiguous in meaning, which is not a
process of generating grammatical metaphor. The ellipsis of the
verb is in (14) will not cause difficulty in interpretation, but this
kind of ellipsis is syntactically unacceptable. Ellipsis is a device
of textual cohesion discussed by Halliday and Hasan (1976),
which is a linguistic device to avoid repetition. However, Ellip-
sis itself will not cause the occurrence of grammatical meta-
Textual Metaphor: A Non-finite
The types of textual metaphor mentioned above from various
perspectives are actually representations of ideational and in-
terpersonal metaphors in text, some of which are not even
grammatical metaphors. “As a consequence, there is no need to
include ‘textual metaphor’ as a separate term” (Yang, 2003: p.
Grammatical metaphors are the result of rankshift or tran-
scategorizaton, and the process of rankshifting and transcatego-
rizing is at the same time a semogenic process which is referred
to as semantic blend, semantic fusion or semantic junction
(Halliday and Matthiessen, 1999: pp. 243-259) and “has a feed-
back effect into the semantics” (Ravelli, 1988: p. 137). “All
such junctional elements involve grammatical metaphor” (Hal-
liday and Matthiessen, 1999: p. 243). Semantic junction “en-
ables the key defining motif of grammatical metaphor to be
captured” (Ravelli, 2003) and is “a notion that is at the root of
grammatical metaphor” (Lassen, 2003: p. 34). For example,
although a nominalized verb construes a Thing, it still realizes a
Process. Thus, nominalization is characterized with double
functions, i.e., Process and Thing. Therefore, “double function
is the identifying principle of grammatical metaphor” (Fan,
Non-finite clauses and finite clauses usually constitute clause
complexes. When discussing relations between clauses, Halli-
day proposes two dimensions: interdependency and logico-
semantic relation. The former consists of two relations, i.e.
parataxis and hypotaxis, and the latter, expansion and projec-
tion. In expansion, the secondary clause expands the primary
clause by elaborating, extending or enhancing. In projection,
the secondary clause is projected through the primary clause,
which instates it as a locution or an idea. The logico-semantic
relation between clauses in clause complexes is mainly deter-
mined by conjunctions (Yang, 2003: p. 78). Halliday and Mat-
thiessen (1999: p. 177) recognize a category of element, the
relator, which is construed by conjunctions or conjunction
groups. The relation between a finite clause and a non-finite
clause in a clause complex tends to be extension and enhance-
ment rather than elaboration, because “the relators themselves
embody meanings of extension or enhancement” (Yang, 2003:
p. 79). When there is not a conjunction functioning relator, the
logico-semantic relation is usually construed by the non-finite
element. Thus, the non-finite element has double functions,
relator and process. Under this circumstance, the logico-se-
mantic relation is usually of elaboration or enhancement and of
additive and adversative extension. In other words, the realiza-
tion of extension requires a conjunction or a preposition, which
further indicates that extension is dependent upon relators. Con-
junctions operate outside the rank scale, so they have no other
functions except realizing relators. Non-finite verbs realizing
relators will not change the logico-semantic relation between
clauses, nor will rankshift which is required by ideational
metaphor occur. This kind of double function only occurs in
textual organization, hence textual metaphor. Here is an exam-
(17) a. I worked for a local firm at that time, selling office
b. Instead of finding the perpetrators, they criminally
charged the Earth First!
c. We used to go away at the weekend, taking all our gear
d. You won’t get away without the work being completed.
e. While pondering which way to go, I completely lost my
(17a) is an elaborating clause complex, non-finite element
selling in which realizes both relator and process, which can be
considered as the first type of textual metaphor from a non-
finite clause perspective. (17b) is an extending clause complex
of the variation type, conjunctive preposition phrase instead of
can not be omitted and the non-finite element finding has no
double functions, hence no textual metaphor. (17c) construes
the additive extension without conjunction or conjunctive pre-
position, and the non-finite element taking realizes both the
relator and the process. (17d) and (17e) construe enhancement,
and they will change into (17f) and (17g) with the conjunctions
omitted, and the non-finite element realizes both relator and
process. Therefore, (17f) and (17g) are the textual metaphors of
(17d) and (17e) respectively. These extending and enhancing
non-finite clauses without explicit relators are the second type
of textual metaphor from a non-finite clause perspective.
(17) f. You won’t get away the work not being completed.
g. Pondering which way to go, I completely lost my bear-
There are also such textual metaphors in non-finite -ed
clauses. For example, (18b) is a textual metaphor of (18a):
(18) a. After Interviewed, Disxon made a statement.
b. Interviewed, Disxon made a statement.
Relators are usually realized by conjunctions or conjunctive
groups linking two items at the same linguistic level, such as
words, phrases/groups and clauses. However, when a finite
clause and a non-finite clause are linked to form a clause com-
plex, prepositions can also function as relators. For example:
(19) a. When you become a member, you will receive a mem-
bership card and a badge.
Q. S. HE
b. When becoming a member, you will receive a membership
card and a badge.
c. On becoming a member, you will receive a membership
card and a badge.
In (19c), the preposition on realizes relator, functioning the
same as the conjunction when. The former is of double func-
tions and the latter not. For example, we can say on your be-
coming, but not when your becoming, and when you become in
(19a) can not be changed into on you become. Accordingly, in
(17b), although the conjunction group instead of functioning
substitutive extension can not be omitted, non-finite element
finding is not of double function. However, preposition phrase
instead of realizing relator is of double functions. In this sense,
(17d) is also textual metaphor with preposition phrase realizing
relator. These extending and enhancing non-finite clauses with
preposition phrases realizing relators are the third type of tex-
tual metaphor from a non-finite clause pers pective.
The theory of cline in SFL has important implications to the
research of textual metaphor from the non-finite clause per-
spective. According to SFL, there is not a distinct boundary
between categories, and the members of two categories form a
gradually graded continuum (Yang, 2007). For example, non-
finite clauses and preposition phrases form a cline.
(20) a. Looking at the picture, I couldn’t help missing my
middle school days.
b. Weather permitting, they will go on an outing to the beach
c. Considering the distance, he arrived very quickly.
d. During the war, people suffered food shortages.
e. At the news, he co uld hardly say a word.
Verbs tend to be grammaticalized into prepositions, which
can be verified in both Chinese and English. For example,
words like yan (沿), jiang (将) and pin (凭) in Chinese and
considering, regarding, and concerning in English are all the
result of prepositionalization. Verbs require participants more
than prepositions do. Looking in (20a) and permitting in (20b)
are the strongest in action and require logical subjects most
among these example sentences. During in (20d) and at in (20e)
are the weakest in action and have the least requirement of
logical subjects. Considering in (20c) lies in between non-finite
verbs and prepositions and relatively weak in action; it is an
internal mental activity and has no external relations with the
subject of the clause. In this sense, considering has double
functions in the process of prepositionalization of verbs, hence
textual metaphor. The enhancing non-finite clauses with pre-
positionalized non-finite verbs are the fourth type of textual
metaphor from the non-finite clause perspective. Preposition at
in (20e) is a core member of the category of prepositions and it
has no double functions. Preposition during in (20d) evolved
from ancient French verb durer, it was loaned to English in the
middle ages, with the present participle form duren. And now,
this word has completely prepositionalized and should not be
treated as textual metaphor.
The occurrence of ideational metaphor and interpersonal
metaphor is a semogenic rankshift or transcategorization. This
is marked by qualitative changes in language form with double
functions. As a second-order meaning mode, textual metafunc-
tion can be organized in a conventional or an unconventional
way. The conventional organization is unmarked, and the un-
conventional organization is marked. The marked structure
functions emphasize and cohere rhetorically or pragmatically,
but do not meet the conditions to create grammatical metaphor.
The occurrence of ideational and interpersonal metaphors in
textual circumstances provides these metaphors with textual
effect, but it does not qualitatively change the organization of
text. Based on the double functional identification principle,
this paper proposed four types of textual metaphor from the
non-finite clause perspective, i.e., elaborative non-finite clauses,
extensive and enhancing non-finite clauses without relators,
extensive and enhancing non-finite clauses with prepositions as
relators and enhancing non-finite clauses with prepositionalized
Derewianka, B. (2003). Grammatical metaphor in the transition to ado-
lescence. In: A.-M. Simon-Vandenbergen, M. Taverniers, & L. J.
Ravelli (Eds.), Grammatical metaphor: Views from systemic func-
tional linguistics (pp. 185-219). Philadelphia, PA/Amsterdam: John
Fan, W.-F. (2001). Theoretical study of grammatical metaphor. Bei-
jing: Foreign Language Teaching a nd Re sea rch Press.
Fan, W.-F. (2007). A comprehensive pattern of grammatical metaphor.
Foreign Language Edu cation , 4, 12-15.
Goatly, A. (1996). Green grammar and grammatical metaphor. Journal
of Pragmatics, 25, 537-560.
Halliday, M. A. K. (1978). Language as social semiotic: The social
interpretation of language a n d me a ning. London: Edwar d Arnold.
Halliday, M. A. K. (1985). An introduction to functional grammar.
London: Edward Arnold.
Halliday, M. A. K. (1994). An introduction to functional grammar (2nd
ed.). London: Edward Arnold.
Halliday, M. A. K., & Hasan, R. (1976). Cohesion in English. London:
Halliday, M. A. K., & James, Z. L. (2005). A Quantitative study of
polarity and primary tense in the English finite clause. In M. A. K.
Halliday, & J. J. Webster (Eds.), Computational and quantitative
studies (pp. 93-12 9). London: Conti nuum.
Halliday, M. A. K., & Matthiessen, C. (1999). Construing experience
through meaning: A language-based approach to cognition. London/
New York: Cassell.
Halliday, M. A. K., & Matthiessen, C. (2004). An introduction to func-
tional grammar (3rd ed.). London: Edward Arnold.
Hu, Z.-L. (1996). Grammatical metaphor. Foreign Language Teaching
and Research, 4, 1-7.
Huang, G.-W. (2002). Cleft sentences as grammatical metaphors. In
G.-W. Huang, & Z.-Y. Wang (Eds.), Discourse and language func-
tions (pp. 32-41). Beijing: Foreign Language Teaching and Research
Huang, G.-W. (2009). The analysis of grammatical metaphor in transla-
tion studies. Chinese Translators Journal, 1, 1-5.
Lassen, I. (2003). Accessibility and acceptability in technical manuals:
A survey of style and grammatical metaphor. Philadelphia, PA/Am-
sterdam: John Benjamins,.
Liu, C.-Y. (2003). The stylistic value of grammatical metaphor. Mod-
ern Foreign Languages, 2, 120-127.
Liu, C.-Y. (2005). The reverse direction of rankshift between ideational
metaphor and interpersonal metaphor. Foreign Language Teaching
and Research, 5, 289-293.
Martin, J. R. (1992). English text: System and structure. Philadelphia,
PA/Amsterdam: John Benjamins.
Martin, J. R. (1993). Life as a noun: Arresting the universe in science
and humanities. In M. A. K. Halliday, & J. R. Martin (Eds.), Writing
science, literary and discourse power (pp. 241-293). London: Palmer
Martin, J. R. (1997). Analyzing genre: Functional parameters. In F.
Q. S. HE
Open Access 313
Christie, & J. Martin (Eds.), Genres and institutions (pp. 3-39). Lon-
Ravelli, L. J. (1988). Grammatical metaphor: An initial analysis. In: E.
H. Steiner, & R. Veltman (Eds.), Pragmatics, discourse and text:
Some systemically-oriented approaches (pp. 133-147). London: Pin-
Ravelli, L. J. (2003). Renewal of connection: Integrating theory and
practice in an understanding of grammatical metaphor. In A.-M.
Simon-Vandenbergen, M. Taverniers, & L. J. Ravelli (Eds.), Gram-
matical metaphor: Views from systemic functional linguistics (pp.
37-64). Philadelphia, PA/Amsterdam: John Be njamins.
Taverniers, M. (2006). Grammatical metaphor and lexical metaphor:
Different perspective on semantic variation. Neophilologus, 2, 321-
Thompson, G. (1996). Introducing functional grammar. London: Ed-
Yang, B.-J. (2003). A study of non-finite clauses in English: A systemic
functional approach. Beijing: Foreign Language Teaching and Re-
Yang, B.-J. (2007). Realization and meaning of cline in non-finite
clauses. Foreign Languag e Research, 3, 50-54.