Chapter Outline:

• Main tenets of Functionalism.

• Functional Sentence Perspective as a theory of

linguistic discourse analysis.

• The concept of communicative dynamism.

• Analysis of thematic and information structures.

8.1. Functionalism
Functionalism was one of the great linguistic paradigms of the 20th century, still in force, which grew as
an alternative to the abstract, formalized view of language presented by Transformational Grammar, an
approach to language that, according to functionalists, could not explain certain linguistic facts.
Functionalism relies, contrary to Formalism, on a pragmatic view of language as social interaction, and
therefore the approach focuses on the rules which govern verbal interaction. These rules are seen as a
form of co-operative activity, and their study helps the researcher describe the linguistic structures of
language in relation to the context in which they occur. Meaning, rather than form, is taken into account,
and both the extra-linguistic context and the purpose of communication play a crucial role in this
approach.
The origins of functionalism can be traced back to the Linguistic School of Prague, in the work
of scholars such as Vilèm Mathesius, Frantisek Danes or Jan Firbas. These scholars emphasized the
functional aspects of the organization of information. Danes (1974) developed the concept of
Communicative Dynamism and Firbas (1964, 1986) was the father of the theory of Functional Sentence
Perspective, which studies the distribution of information, organized in more or less dynamic elements
called theme and rheme.

8.1.1. Functional Sentence Perspective

Functional Sentence Perspective is a theory of linguistic analysis which refers to an analysis of utterances
or texts in terms of the information they contain. The role of each part of a given utterance is evaluated
for its semantic contribution to the whole. The notion of communicative dynamism is a key concept in this
theory, for it attempts to rate the different levels of contribution within a structure, in particular those
related to the concepts of theme and rheme. Firbas defines communicative dynamism as “the relative
extent to which a linguistic element contributes towards the further development of the communication”
(1992: 8). He also explains that:

It is an inherent quality of communication and manifests itself in constant development towards the
attainment of a communicative goal; in other words, towards the fulfillment of a communicative purpose.
Participating in this development, a linguistic element assumes some position in it and in accordance with this
position displays a degree of communicative dynamism. (1992: 7)

Thus, communicative dynamism assumes that when speakers say something, they have a communicative
purpose and that the elements of their language contribute to that purpose. It is important to note,
however, that even when functional sentence perspective relies on introspections regarding the goals of
communicative acts, any broader social or cognitive commitments are avoided.
Functional Sentence Perspective provides a functional explanation for word order: all other things
being equal, the order of words in a sentence corresponds to an increase in communicative dynamism.
There are four factors which determine communicative dynamism: 1) linear modification; 2) the
contextual factor; 3) the semantic factor and 4) prosodic prominence (in spoken language) (Chafe,
1994: 162).
 Linear modification 1 is a term used by Firbas to depict the relation between word order and
communicative dynamism, as illustrated in the following example:

A: Any news about your mom?

B: Yes, she has travelled to Brussels.

B’s response might have different purposes, but the most obvious seems to be “to state the destination of
B’s mom’s trip”. In B’s reply, then, the word she contributes the lowest degree of communicative
dynamism; has travelled contributes an intermediate degree, and to Brussels the highest degree, since it
serves the purpose of the response (destination).
The contextual factor has to do with the concepts of “retrievability and irretrievability from the
immediately relevant context” (Firbas 1992: 21), which have created the opposition context-dependent
versus context-independent.
The semantic factor deals with the so-called dynamic functions. The notions of theme and rheme are
crucial here. The theme is considered to be the part of a sentence which adds least to the advancing
process of communication, having the lowest degree of communicative dynamism. By contrast, the
rheme carries the highest degree of communicative dynamism. We shall deal with these two concepts in
more detail in section 8.1.1.1.
Prosodic prominence is a factor which can be studied only in spoken language, and it integrates
functional sentence perspective with intonation studies. It deals, for example, with the function that
certain tones or other prosodic features (such as pitch prominence) fulfill in discourse.

8.1.1.1. Thematic Structure: Theme vs. Rheme

The thematic structure of a clause contains two main elements: theme and rheme. Following Halliday
(1967: 212), we shall say that the theme of a clause is what speakers/writers use as the ‘point of
departure’. The rest of the message constitutes the other element of the thematic structure, namely, the
rheme.
Different elements can be chosen as the point of departure or initial constituent of a clause, as
illustrated in the following examples:

THEME RHEME
a) Peter doesn’t like that car
b) That car Peter doesn’t like
c) What Peter doesn’t like is that car
d) It’s that car Peter doesn’t like

The four clauses in the examples above have the same propositional content; however, the choice of
theme made by the speaker or writer in each case changes its meaning to a certain extent, because it
shows the angle from which the speaker projects his/her message. From the point of view of discourse
analysis, the possibility of making these choices is proof of the fact that a speaker producing one or other
choices will be making different assumptions about the stage of knowledge of his/her hearer. It is clearly
the case that all four clauses could not be the answer to the same question, and therefore each of them
assumes different presuppositions on the part of the interlocutor. For example, d seems to presuppose the
shared belief of both interlocutors that Peter does not like something, and c restricts what Peter does not
like among other things the hearer might consider ‘not likeable’.
The theme always contains an ideational element. This element is some entity which functions as
subject, object, complement or circumstantial adjunct. Halliday (1985: 54) refers to this entity as the
topical theme (due to the fact that it generally corresponds to the element identified as topic in topic-
comment analysis). The ideational function represents our experience of the world, namely, processes,
actions, events, processes of consciousness and relations, and that is why this type of theme has also been
called experiential theme.
Apart from the clausal/experiential themes, we may find what Downing and Locke (2006: 234) call
non-experiential themes, which can be divided into two main kinds: a) interpersonal themes and b)

1The term linear modification was first used by Dwight Bolinger (1952: 1125) to explain a somewhat different
phenomenon from that explained by Firbas under the same term.
textual themes. Within interpersonal themes we find continuative themes, manifested as pragmatic
markers of attention, response, request, surprise, hesitation, etc. (such as Oh, Well, Please, Hey!),
Adjuncts of stance (such as apparently, surely, certainly) and vocatives and appellatives (such as Dad!
Mr. Wilson! Ladies and gentlemen). Textual themes include connective adjuncts/discourse markers
(such as anyway, however, first, finally, etc.) which connect a clause to the previous part of the text by
indicating relations of consequence, addition or concession, among others.
Table 1 presents these main kinds of theme more schematically, with illustrative examples:

Table 1: Experiential and non-experiential themes

TYPES OF THEMES
A) EXPERIENTIAL Examples
1. Subject Tom is coming for dinner.

2. Object “No!” I said.

3. Circumstantial adjuncts This morning at the cafeteria we

had a lot of fun. (Time and place
adjuncts)

4. Complement A terrible fiasco it was. (subject

comp.)
General Director he was
appointed. (object comp.)

5. Verb Coming up is the latest news.

(thematized verb)

B) NON-EXPERIENTIAL Examples
1. INTERPERSONAL
a) Continuative pragmatic Well, see you tomorrow.
markers
b) Adjuncts of stance Frankly, I don’t understand your
point.

c) Vocatives and appellatives Peter, call your Mom, please.

2. TEXTUAL (connective adjuncts/ She is sick. Consequently, she

discourse markers) won’t go to work today.

8.1.1.1.1. Multiple Themes

As Table 1 shows, the three macro-functions of language (experiential, interpersonal and textual) can be
represented within the Theme. Two or the three types of theme may be concurrent in the same utterance,
and in such cases we may talk of multiple themes. When the three of them appear together, the typical
sequence of these elements is textual/ interpersonal/experiential, although the interpersonal theme may be
marked (see 8.1.1.1.4) in certain circumstances, and thus it may appear before the textual theme. But the
experiential theme always has to be the final one. Here are some examples of multiple themes:
 do you
Well then you little brat what
want?
connective
continuative vocative object
adjunct
Interpersonal Textual Interpersonal Experiential
theme theme theme theme
THEME RHEME

rejected my
Unfortunately though she
offer
adjunct of
connective
stance subject
adjunct
(evaluative)
Interpersonal Experiential
Textual theme
theme theme
THEME RHEME

8.1.1.1.2. Detached Themes - Absolute thenes

\ Dislocatimos
Some themes are detached from the main clause in one way or another. Downing & Locke (2006) note
that a subtype of these does not function as a constituent of the clauses that follow them. They are
normally detached lexical noun phrases which stand outside the clause, and are called Absolute Themes.
Absolute themes are common in the spoken registers of many European languages, such as English or
Spanish. Consider this example:

The financial crisis, we are all aware that some measures have to be taken.
Absolute Theme

Here the noun phrase the financial crisis is completely detached and thus has no grammatical relations
with the second part of the message. This is to say that the financial crisis is neither the subject nor the
object, nor any other constituent of the clause that follows (we are all aware that some measures have to
be taken). However, there is still a connection between both parts: the absolute theme provides a
pragmatic framework which allows the hearer to infer the relationship between them. This is one of the
many instances in linguistic communication where grammatical competence is not enough, and the
interlocutors have to resort to the wider pragmatic/discursive/communicative competence in order to sort
out meanings.
Dislocations are another subtype of detached themes. They are different from Absolute Themes “in
that the ‘dislocated’ element is a constituent of the clause, frequently subject […], and is repeated by a co-
referential pronoun […] in its normal position within the clause” (Downing & Locke, 2006: 232). In this
type of theme, then, the connection is encoded grammatically. When the dislocated element is placed at
the beginning of the clause, it is known as left-dislocation; when it is placed at the end of the clause, it is
called right-dislocation. Consider these examples:

a) Left-dislocations:

That scream, where did it come from?

Left-dislocated Theme

That house, that is the one I’ve always dreamed of!

Left-dislocated theme

Here, the normal, non-dislocated forms would be, respectively: Where did that scream come from? and
That is the house I’ve always dreamed of!

b) Right-dislocations:

Is it yours, that jacket?

Right-dislocated theme
It’s fantastic, this book.
Right-dislocated theme

The normal non-dislocated forms for these examples would be, respectively, Is that jacket yours? and
This book is fantastic. Not being initial, this kind of theme is more problematic to analyze, but as
Downing & Locke suggest, the cognitive motivation for such a structure could be that of making explicit
a referent (that jacket and this book in our examples) which was accessible to the speaker in the context,
but perhaps not so obvious to the hearer, or not in the speaker’s mind at the moment he uttered the first
part of the clause (2006: 233).
Two or more detached themes may co-occur, in which case the relationship between them must be
pragmatically relevant, as in the following example:

Your friend, the car outside her house, they’ve stolen it.
Absolute theme + left-dislocated theme

The co-occurrence of detached themes is more likely to be found in spoken English than in written
English. Downing & Locke explain that the function of double or multiple detachments “is to ‘anchor’ the
final referent to other related referents which are presumed to be accessible to the hearer” (2006: 234).

8.1.1.1.3. Thematic clauses

When two or more clauses are joined together in a complex clause, the clause that is placed first is said to
be thematic with respect to the whole complex clause. This applies for cases of coordination as well as for
cases of subordination. Examine these examples:

1) Coordination
Tommy hit his sister and she burst into tears.
THEME RHEME

2) Subordination
When I saw her I realized she had been crying.
THEME RHEME

Clauses related by coordination are said to be paratactically 2 related. Clauses related by subordination
are said to be hypotactically 3 related. Paratactically related clauses are typically placed in the
chronological order in which the events described occur. In example 1, the event described in the rheme
clause (Tommy’s sister bursting into tears) is understood to have occurred after the event described in the
theme clause (Tommy hitting his sister). In hypotactically related clauses, however, the speaker/writer
does not necessarily have to maintain chronological order, thus in example 2 both the main and the
subordinate clauses could be used as the starting point of the message (or theme). Therefore, I realized
she had been crying when I saw her is equally possible, the only difference being the intention, on the
part of the speaker/writer, to place what he considers to be new and important information at the end, and
as part of the rheme in each case.
The thematic organization of the different clauses in a text is of great importance to the discourse
analyst, because it reveals the method of textual development: “…by analyzing the thematic structure of a
text clause by clause, we can gain an insight into its texture and understand how the writer made clear to
us the nature of his underlying concerns” (Halliday, 1985: 67).

8.1.1.1.4. Theme, subject and topic

It is important to note that theme is a different category from syntactic Subject and from Topic, even
though “these three tend to coincide in one wording” (Downing & Locke, 1992: 222). Whereas theme is
the starting point of the message, subject is a syntactic element of clause structure (the other elements
being Predicator, Complements, Objects and Adjunct) and Topic refers to what the text is about (it may
refer to the whole or only to a given part of the text). The following examples illustrate the fact that the
three elements may coincide (a) or, contrariwise, may not (b):

2A paratactic relationship is that holding between clauses of equal status. It is, thus, a relationship of equivalence.
3 A hypotactic relationship is that holding between clauses of unequal status. It is, thus, a relationship of non-
equivalence.
a) The new president has been strongly criticized for his foreign policy.
Subject
Theme
Topic

b) In Spain, the people criticized the new president for his foreign policy.
Theme Subject Topic

As can be seen in b, the theme does not coincide with the syntactic subject. Instead, the theme is,
syntactically, an adjunct (circumstance) which is realized by a prepositional phrase (In Spain). The topic
does not concur with the subject either: when turning a into its active counterpart in b, the agent (the
people) that was not named in a appears as the subject of the clause and the topic is now the direct object
(the new president). However, clauses may have more than one type of topic. Those topics which are
introduced into the discourse for the first time are called new topics. New topics may become old or
known topics as the discourse proceeds, or they may be abandoned completely. In b, for example, the
people could be taken as an old topic and the new president as the new topic. We may also speak of
global and local topics (Downing & Locke, 2006: 225). The former refer to those that organize a whole
piece of language; the latter, to the topics of utterances and sentences, and these are the only ones that
have a direct grammatical realization. In between global and local topics it can be said that paragraphs or
sections in writing, as well as episodes in talk, each have their own topic. All levels of topic help to build
up the coherence of discourse as a whole.

8.1.1.1.5. Marked and Unmarked Themes

Depending on the purposes of communication, certain types of information may be foregrounded or

thematized. When the theme does not coincide with the expected first constituent of each mood structure, 4
we speak of a marked theme. If, on the contrary, the theme co-exists with such a constituent, it is an
unmarked theme. For example, the expected first constituent of a declarative clause is the subject, so if
in such a clause the subject appears as the first constituent, the theme will coincide with the subject (as is
the case in example a in 8.1.1.1.4.), and we shall say that the theme is unmarked. Contrariwise, in
example b the adjunct In Spain has been thematized or foregrounded, and thus we say that the theme is
marked. Here are two more examples:

1)

Sally will never pass that exam.

UNMARKED THEME RHEME

Never will Sally pass that exam.

MARKED THEME RHEME

2)

He popped the question.

UNMARKED THEME RHEME

The question he popped.

MARKED THEME RHEME

In example 1, the adjunct never is fronted and thus thematized; in example 2, it is the direct object (the
question) that is thematized, both fronted elements thereby turning into marked themes. If we go back to

4The different mood structures with their corresponding (normal or expected) first constituents are: a) Declarative
subject (e.g. Tommy fell off his bike); b) Polar interrogative Finite + subject (e.g. Do you believe in ghosts?);
c) Wh- interrogative Wh- element (e.g. Who told you such a thing?); and d) Imperative Predicator or let +
Subject (e.g. Do it now!/ Let’s do it now!). For a more detailed description of this topic see Downing and Locke
(2006:224-25).
Table 1 (in 8.1.1.1.) we shall see that in all the examples, except for the first one (Tom is coming for
dinner), the themes are marked.

8.1.1.1.6. Thematization/Staging

As has been suggested so far, sentence word order is of capital importance for the organization of the
information. The same applies for the order in which sentences are put into texts, because this order will
influence the hearer’s or reader’s interpretation of the whole discourse in question. The speaker/writer
always has to choose a beginning point: what s/he puts first will influence the interpretation of the text
which follows it. This process, which has to do with the linear organization of sentences and texts, has
been called thematization. A more inclusive and more general term than thematization is staging 5, a
term that, according to Grimes, refers to the fact that “Every clause, sentence, paragraph, episode, and
discourse is organised around a particular element that is taken as its point of departure” (1975: 323).
It is interesting to note how the linear organization can be manipulated to bring certain items or events
into greater prominence than others by means of the process of thematization or staging. The title of a
newspaper article, or the title of a book can be considered a powerful thematization device used by the
author. Thematization creates certain expectations in the readers or hearers in that the thematized
elements provide a starting point which constrains their interpretation of the discourse that follows.
Brown & Yule illustrate this by presenting the results of an exercise they conducted using the following
text:

A Prisoner Plans His Escape

Rocky slowly got up from the mat, planning his escape. He hesitated a moment and thought. Things were not
going well. What bothered him most was being held, especially since the charge against him had been weak.
He considered his present situation. The lock that held him was strong, but he thought he could break it
(1983: 139).

Brown & Yule used this text to conduct an exercise in which they asked their subjects several questions,
and they found that “there was a general interpretation that Rocky was alone, that he had been arrested by
the police and that he disliked being in prison” (1983: 139). Another group of subjects were given the
same text to read but with a different title, namely, A Wrestler in a Tight Corner. This group gave very
different answers to the questions: they said Rocky was a wrestler who was being held in some kind of
wrestling ‘hold’ and was planning to get out of it. They also thought that Rocky was not alone and that
that he had had nothing to do with the police. This experiment shows how, by providing different themes
or starting points in the thematized elements of the two different titles, the authors constrained the ways in
which the texts were interpreted by the subjects.

8.1.1.2. Information Structure: Given vs. New

In his search for correspondences between linguistic elements and their functions, Halliday found that
the tone group, apart from being a phonological constituent, “functions as the realization of something
else, namely, a quantum or unit of information in the discourse” (1985: 274). Halliday explains that
spoken discourse takes the form of a sequence of information units, and he uses the term information to
mean “a process of interaction between what is already known or predictable and what is new or
unpredictable” (1985: 274-75). The information unit, thus, is a structure made up of two functions: the
New and the Given. From a structural point of view, it can be said that all information units have an
obligatory new element and an optional given element. The latter is concerned with information which is
presented by the speaker as ‘recoverable’ (either from the linguistic co-text, from what has been said
before, or from the situational or cultural context). The former (the new element) concerns whatever
information the speaker presents as not recoverable by the hearer.
We say that the Given is optional from the structural point of view because, by its own nature, this
element is referential or ‘phoric’ (i.e. it refers to something already present in the verbal or non-verbal
context), and reference is often achieved through ellipsis. Ellipsis is a grammatical form in which certain
features are not realized in the structure.
The Given typically precedes the New, and the New is always marked by tonic prominence. The
element which has this prominence is said to be carrying information focus. Consider this example, in
which the syllable in capitals represents the intonation nucleus of the tone unit:

5 Although some authors, like Brown & Yule (1983), make no distinction between these terms (thematization and
staging).
A: Where have you guys been?
B: Well, I’ve been to the GROcery store, and Tim to the LIbrary.
GIVEN NEW GIVEN NEW

In A’s utterance, you guys is a deictic expression that refers to B (I) and to Tim, so by the time of B’s
reply, I and Tim are a Given, and the New elements are found towards the end of each coordinate clause.
As noted above, on many occasions the Given is ellipted, and therefore the clause structure consists
only of the New element, as in the following example:
A: What are you writing?
B: (I am writing) An essay for my English class.
(Ellipted Given) NEW

Here the first part of B’s response has been ellipted, since it would involve a repetition of part of A’s
clause, and therefore would be redundant and unnecessary. Thus, the whole of B’s response can be said
to be New.

8.1.1.2.1. Marked and Unmarked Focus

Regarding the information focus in normal, unemphatic discourse, the unmarked distribution starts with
the Given and progresses towards the New. Downing & Locke (2006: 241) explain that this is often
called the principle of end-focus. The focus normally marks where the New element ends (because it
typically falls on the last lexical item in the clause) but it is not always clear where it begins, or where the
boundary between Given and New would be. The distinction of such a boundary is highly dependent on
other elements of the text or context, which are not always available to the analyst. If, for example, we
take an utterance out of context, we will be able to tell that it culminates with the New, but will not be
able to tell whether there is a Given element first. Suppose you overhear the following statement:

All the people were running for their LIVES.

NEW

Here we know that their lives is New, because the prominence falls on that element, but we would not be
able to tell whether the New extends also to were running and all the people. The whole statement would
be New if it were the answer to the following question: What happened immediately after the 9/11 attacks
in New York? However, if the questions changed as shown in Table 1, the New (in bold type) would also
vary:

Table 1
What were all the people doing immediately after (All the people) were running for their lives.
the 9/11 attacks in New York?
What were all the people running for immediately (All the people were running for) their lives.
after the 9/11 attacks in New York?
Who were running for their lives immediately… All the people (were running for their lives…).

Where were all the people running for their lives? (All the people were running for their lives…) in
New York.

When were all the people running for their lives in (All the people were running for their lives…)
New York? immediately after the 9/11 attacks.

The principle of end-focus allows us to say that the unmarked option for the focus is to fall on the
last lexical item of the clause. The focus will be marked, therefore, when it does not fall on the last
lexical item. The third example in Table 1 (All the people) illustrates an instance of marked focus, placed
at the beginning of the clause. A focus is marked when the speaker wants to contrast or correct something
which has been said or implied in the previous discourse or in the situational context. It can also be
marked for emotive purposes. Consider example 1, which has an unmarked focus, in contrast with
examples 2, 3 and 4, which contain marked foci for contrastive purposes:

1. Sarah took her car to the gaRAGE. (unmarked focus)

2. Sarah took her CAR to the garage. (not her motorbike)
3. Sarah TOOK her car to the garage. (not “brought it from there”)
 4. SArah took her car to the garage. (not Susan)

Table 2 displays examples of foci marked for emotive purposes.

Table 2
UNMARKED FOCUS MARKED FOCUS
It sounds ODD. It DOES sound odd.
She is SHY. She IS shy.
You will see me in the FUture. You WILL see me in the future.

8.1.1.2.1.1. But how do we identify the focus?

It was explained in 8.1.1.2. that speakers divide their messages into segments of information, namely,
information units. These are features of the spoken language, for they are not realized by any given
grammatical unit but by a phonological unit called tone unit. Tone units always contain one syllable
which is more prominent. This syllable contains the intonation nucleus of the unit and constitutes the
focus of information. Prominence can be given by means of pitch movement, increased duration and tonic
stress. Pitch movement corresponds to the different tones of intonation, which may be falling, rising or
level.
The information focus represents the peak or highest point of the unit, and its correct placement is of
utmost importance in English, due to the fact that it constitutes the main strategy used by speakers for
communicating contrast and emphasis in the spoken language.
It must be emphasized that there is not a one-to-one correspondence between the tone unit and any
grammatical unit. Downing & Locke (2006) pinpoint the fact that speakers may lengthen or shorten their
tone units according to their communicative needs, and that variations in the length of tone units also
depend on other factors, such as speed of utterance, the syntactic structures and/or the lexical items
chosen, familiarity with the content of the message and the consequent relative need to plan ahead,
acoustic conditions, self-confidence, etc.
Considering all these factors, it is not surprising that the identification of the information focus is not
always as easy and clear-cut a task as one would wish it to be. Several authors have pointed this out, and
have therefore proposed new alternatives to the analysis of the phenomenon of information structure (see
8.3.).

8.1.1.2.1.2. Cleft Constructions: it-cleft and wh-cleft

Speakers often re-organize the content of a single clause into two related parts or units in order to place
the focus on a new element that always follows a form of the verb be. This is called clefting, and it can be
done by using two different constructions: the it-cleft and the wh-cleft. Thus, a simple clause like “I want
a hamburger” may be re-organized as:

a) It’s a HAMBURGER (that) I want. (It-cleft)

New Given

b) What I want is a HAMBURGER. (Wh-cleft)

Given New

The focus of both clefts in a) and b) is hamburger, which is thus marked with tonic stress. In a) the focus
is near the front of the first unit; in b) it is at the end of the second unit. The verb want contains Given or
presupposed information and for this reason it has a lesser stress.
The main discourse function of cleft constructions is to mark contrastive focus, but they may also be
used to highlight expressions of time or place (e.g. It was in December that I met her), to signal the
beginning of an episode in discourse (e.g. It is with pleasure that I introduce you to Dr. Stephens tonight),
to signal a shift to a new episode (e.g. It was only after some years that I realized I had to change); or to
suggest exclusiveness (e.g. What I want is care and attention = (Care and attention is the only thing I
want = I don’t want money or anything else))
 8.2. Information Structure and Thematic Structure: Given + New, and Theme
+ Rheme
Information and thematic structures are closely related from the semantic point of view. Under normal
conditions, the speaker/writer will choose the Theme from within what is Given and will locate the New
within the Rheme. However, neither Given and Theme nor New and Rheme are the same. As Halliday
puts it, “The Theme is what I, the speaker, choose to take as my point of departure. The Given is what you, the
listener, already know about or have accessible to you. Theme + Rheme is speaker-oriented, while Given + New is
listener-oriented” (1985: 278). Thus, the speaker can use thematic and information structure to produce a
wide variety of rhetorical effects; s/he can play with the two systems by using different strategies which
will bring about different results in the interpretation of the message on the part of the hearer. Example 1
shows a prototypical case, where the Given information is found within the Theme and the New within
the Rheme. Example 2, on the contrary, shows an example in which B plays with the two systems in
order to produce a contrastive effect, and in some way contradict or correct his/her interlocutor’s
statement.

1) A: What does Mary think of John?

B: She HATES him.
Theme Rheme
Given New

2) A: Bill likes tennis

B: GOLF is what he really likes.
Theme Rheme
New Given

In these examples it is clear that the New has not been mentioned before and therefore did not appear
in the previous structure of discourse. However, it is important to note that it is not the structure of
discourse which determines whether information is treated as New or Given. On the contrary, the factor
determining this choice is the speaker’s moment-to-moment assessment of the relationship between what
he wants to say and his/her hearer’s informational requirements. For instance, and as Brown & Yule state,
“it is not the case that if a speaker has just mentioned a referent he must necessarily repeat it low in pitch,
treating it as a given” (1983: 168). Consider example 3:

3) (A mother is upset when she sees her youngest son bothering her older son and eventually fighting
with him.)

Mother: Please, don’t fight with your BROther any more. Your BROther,
Theme Rheme Theme
Given New New

sweetheart, is one of the most important people in your life.

Rheme
Given

Your BROther, will always love and protect you, even after dad and I die.
Theme Rheme
New Given

This example shows how, if the speaker so judges it, she can treat information that has already been
mentioned in the discourse as New by giving it phonological prominence. In this case, the mother wants
to stress the fact that it is Tommy’s older brother (and nobody else) who will always care for and love
him, especially when she and their father eventually die. Thus she gives the highest prominence to the
word brother in all the clauses of her discourse in order to signal, in each case, a new aspect of the
importance of brotherhood as it affects Tommy. Even when the information in the Rheme portion of her
clauses might be considered partially new, it is more important for the mother to single out Tommy’s
brother as a key person in his life, and therefore treat him as the New information in each case.

8.3. Some Considerations Related to Halliday’s Information Structure Analysis

Brown & Yule (1983) note that Halliday makes the simplifying assumption that the sole function of pitch
prominence is to mark the focus of new information within the tone group. Indeed, if we examine pitch
prominence carefully, we shall see that it may have several other discourse/pragmatic functions, such as
marking the beginning of a speaker’s turn or the beginning of a new topic. Brown & Yule very
graphically explain that phonological prominence has “a general watch this! function” (1983: 164), that
is, it is used by speakers to mark any kind of information that requires being paid attention to, but by no
means does it only and exclusively mark the information focus.
In addition, some scholars have found that it is very difficult to find such perfect tone groups as the
ones Halliday describes (constituted around a tonic syllable). It is, on the contrary, common to find tightly
rhythmically bound structures with several peaks of prominence. Experiments such as those made by
Brown, Currie & Kenworthy (1980) have shown that the phonetic cues which were traditionally thought
to mark the tonic syllable rarely cumulate on one word in spontaneous speech. Contrariwise, they tend to
be distributed separately or paired over words which introduce new information. Therefore, many
scholars who do research on conversational speech do not believe that the information unit should contain
only one focus and that it should then be realized with only one tonic.
Furthermore, some authors such as Ward & Birner (2001) have noted that the term focus means
different things to different people, and that a two-way division of information into Given and New is
inadequate. Prince (1992), for instance, classifies information by means of a pair of cross-cutting
dichotomies: on the one hand information may be discourse-old or discourse-new, and, on the other hand,
it can be either hearer-old or hearer-new. Such a distinction sheds light on the fact that what is new to the
discourse need not be new to the hearer. Consider this example:

Yesterday the sun was shining, so I invited a neighbor to go to the pool.

Here, the sun represents information that is discourse-new but hearer-old, while a neighbor refers to
information that is both discourse-new and hearer-new.
In very general terms, however, Halliday’s approach is accepted by most information structure
analysts: it can be said that the information that is felt or judged to be New is going to be prominent,
while the Given information will be produced without prominence. However, it seems reasonable to
suggest that information structure is not only realized by the phonological system, but also by the
syntactic system (e.g. by word order and thematic organization) and the textual system (e.g. the
organization of the different paragraphs in a given text).
All in all, the analysis of discourse is not an easy task, and the approach studied in this chapter, as well
as all other approaches, is but one more attempt to describe and understand the different ways human
beings organize their linguistic messages. No approach is perfect or all-embracing, but each and all of
them contribute with different and useful tools for linguistic analysis. It is up to analysts to choose a given
approach or certain elements from different approaches in order to best suit the needs of their research.

8.4. Sample analysis of data

8.4.1. Thematic structure
From the information given in 8.1.1.1., we can infer that the analysis of the thematic structure of a text
helps us draw conclusions as to the way the writer or speaker has organized the information s/he wants to
convey. Likewise, thematic analysis can help us identify the topic area as well as the discourse type of
the text in question. It can be used to establish the relative coherence of a text and to show how
paragraphs can be arranged across sentence boundaries by means of patterns of theme and rheme
development. Consider the thematic structure of the following letter in a financial advice column of a
very well-known American magazine:

After 12 years of marriage, I have just begun divorce proceedings. We have three children ranging from 1 to
10, and I’ve been a stay-at-home mom for the entire marriage. My physician husband has a substantial
income, and our expenses are high because of our large mortgage. I know that we will have to sell our
beautiful house. My attorney says that, taking into account child support and alimony, I should be able to stay
home until the baby is in kindergarten or first grade. My question is this: How, after never having taken care
of myself, do I plan for my future? I don’t want to end up as my mom did after my father died –living on a
$500 Social Security check.
(The O Magazine, January 2005, p. 30)

Several interesting conclusions can be made from the thematic analysis of this letter. Firstly, if it were
presented in isolation (without a title, introduction or context), by merely looking at the themes (which
are in italics), the reader/analyst could easily identify this text as coming from an advice column, and thus
its genre or type of discourse could be pinpointed. In the very first clause of this letter we see an instance
of thematization, where a circumstancial adjunct of time (After 12 years of marriage) is fronted, probably
in order to give the proper frame to the situation the writer is going to present: having been married for 12
years, this woman is now divorcing and needs advice as to what to do with her house and finances,
because throughout those twelve years she was financially protected by her husband and thus did not
develop skills to protect herself in that respect.
The first themes (After 12 years of marriage, We, My physician husband, our expenses, My attorney)
clearly show that the writer is talking about herself, narrating some aspect of her personal experience. The
last themes (My question, How, I) point to the fact that she is asking a question which is centered on her
particular problem (described at the beginning of the letter), and consequently needs some advice.
Secondly, if we look at the themes of this text, we will observe that most of them are related to the writer
or to some aspect of her life, which gives this letter the characteristics of a narrative of personal
experience. In this type of narratives there is often a tendency to thematize the teller or narrator. Thirdly
and consequently, we may say that most of the themes in this letter have an interactional nature.
McCarthy & Carter explain that “a theme is interactional if it contains words or phrases which
specifically refer to the sender or receiver(s)” (1994: 71). The fact that most of the themes are
interactional also tells us something about the features of this text: we can infer that the writer’s intention
is to interact with someone who will read the letter and somehow try to help her. The way in which the
interactant is going to help can be inferred from the last themes, which show that the writer is asking for
some type of advice. The two types of theme found in this letter lead us to reach the conclusion that we
are facing a case of ‘mixed register’ (Fairclough, 1989), i.e., it contains the characteristics of a narrative
of personal experience but at the same time it communicates with the recipient of the letter on a different
plane, asking for advice, which also makes it a multifunctional and multivalent discourse.

8.4.2. Information Structure

The following fragment has been taken from Nafá (2005) and is an accurate phonological transcription of
a part of a lecture given at the European Parliament. We shall examine how the speaker treats the New
and Given information by assigning prominence to certain words and not to others. The bold type
indicates the focus in each information unit.

DO4 // [p] ´how ´important // [p] is the ´New Delhi ´Conference? // [p] Now, // [o] the
´Commissioner is ´writing ´down: // [hk] [p] “´very // [r] ´important”. // [p] But of ´course, //
[p+] the ´answer // [p] to the ´question // [r] I ´then asked: // [p] if it is ´very important, // [r] ´why
is she ´here? // (2005: 395).

As can be observed, the speaker wants to give rhetorical importance to the adjectival phrase very
important, and therefore, in all cases, the focus is placed either on the word important or on the word
very. In the first clause (How important is the New Delhi Conference?) we may clearly speak of a
marked focus because it is placed at the beginning and not at the end: the word important is stressed and
treated as New, while the New Delhi Conference is treated as Given information. In the successive
clauses, very important and then very are treated as New, even when both words were mentioned before,
due to the fact that the speaker uses her rhetorical skills in order to give a final ironic effect. She uses a
final rhetorical question in relation to the Commissioner’s comment on the New Delhi’s Conference
being ‘very important’, thus signalling the Commissioner’s inconsistency in being present at the
parliamentary discussion instead of at the ‘very important’ Delhi Conference. The analysis of this
example sheds light on the fact that speakers may play with information structure (i.e. the organization of
the Given and the New in discourse) in order to produce certain effects, such as irony or indirect
criticism.
1. Functionalism relies, contrary to Formalism, on a pragmatic view of language as social interaction, and therefore the
approach focuses on the rules which govern verbal interaction. Both the extra-linguistic context and the purpose of
communication play a crucial role in this approach.
2. The origins of Functionalism can be traced back to the Linguistic School of Prague.
3. Danes (1974) developed the concept of Communicative Dynamism (the relative extent to which a linguistic element
contributes towards the further development of the communication) and Firbas (1964, 1986) developed the theory of
Functional Sentence Perspective, which studies the distribution of information, organized in more or less dynamic
elements called Theme and Rheme.
4. The Theme of a clause is what speakers/writers use as the ‘point of departure’. The rest of the message constitutes the
Rheme.
5. Considering the three macro-functions of language, there are two main types of themes: experiential and non-
experiential. Experiential themes contain an ideational element and they represent our experience of the world
(experiential macro-function). Non-experiential themes contain textual and/or interpersonal elements (textual and
interpersonal macro-functions).
6. When non-experiential themes co-occur with the obligatory topical/experiential theme, we speak of multiple themes.
7. Depending on the purposes of communication, certain types of information may be foregrounded or thematized. When
the theme does not coincide with the expected first constituent of each mood structure, we speak of a marked theme.
If, on the contrary, the theme coincides with such a constituent, it is an unmarked theme.
8. Spoken discourse takes the form of a sequence of information units. The information unit is a structure made up of
two functions: the New and the Given.
9. The New is always marked by tonic prominence. The element which has this prominence is said to be carrying
information focus. The focus is unmarked when it falls on the last lexical item of the clause. The focus is marked,
therefore, when it does not fall on the last lexical item.
10. Information structure is closely related to thematic structure from the semantic point of view. A speaker can play
with the organization of the thematic and the information structures in order to produce a wide variety of rhetorical
effects, bringing about different results in the interpretation of the message on the part of the hearer.
11. Some scholars have found that it is very difficult to find such perfect tone groups as the ones Halliday describes
(constituted around a tonic syllable). Furthermore, some authors, such as Ward & Birner (2001) have noted that the
term focus means different things to different people, and that a two-way division of information into Given and New
is inadequate. However, and in very general terms, Halliday’s approach is accepted by most information structure
analysts: it can be said that the information that is felt or judged to be new is going to be prominent, while the given

12/15
12. information will be produced without prominence.
SELF EVALUATION QUESTIONS

Choose the answer that best suits the information given in Chapter 8.

¥
1) Functionalism relies on…
a) a formal view of language.
b) a pragmatic view of language.
c) an anthropological view of language.

2) Communicative dynamism has to do with the extent to which a linguistic element contributes towards
…
a) the development of communication.
b) the grammaticality of the clause.
c) the accuracy and truth of a proposition.

¥
3) A tenet within Functional Sentence Perspective is that…
a) themes and rhemes are always marked options.
b) the rheme carries the lowest degree of communicative dynamism.
c) the relative order of constituents in a clause serves different functions in discourse.

4) What is the type of theme in this clause?

Honestly, I don’t understand what you say.

a) Experiential.
b) Non-experiential.
\
c) Detached.
:
5) Choose the appropriate combination of themes corresponding to the following clause containing
multiple themes:

O.K., then I will do what you want.

1 +2+3

a) Non-experiential (Textual) + experiential (object) + Non-experiential (interpersonal).

b) Non-experiential (interpersonal) + Non-experiential (Textual) + experiential (subject).
c) Experiential (subject complement) + Non-experiential (interpersonal) + Experiential (object).
6) What sub-type of detached theme is your sister in the following clause?

Does she like music, your sister?

a) Left dislocation.
b) Absolute theme.

É
c) Right dislocation.

7) The theme, subject and topic of a clause…

a) always coincide in the same wording.
b) do not necessarily have to coincide.
c) never coincide.

8) The themes in i, ii, and iii are, respectively…

a) marked, marked, unmarked.
b) unmarked, marked, unmarked.
c) unmarked, unmarked, marked.

i) What I want is to know your secret.

ii) Under no circumstances will I tell you my secret.
iii) Can you tell me your secret?
✓
9) The information unit is made up of…
a) four elements: theme, rheme, given and new.
b) two elements: theme and given.

¥
c) two elements: given and new.

10) The Given is optional from the structural point of view because…
a) the speaker knows what s/he is talking about.
b) it is marked by tonic prominence.
c) it is referential or phoric.

11) The focus of a clause…

a) is always at the end.
b) always coincides with the rheme.
c) does not necessarily coincide with any of its grammatical units.

:
12) The foci (marked in capitals) in i, ii and iii are, respectively:
a) marked, marked, unmarked.
b) unmarked, unmarked, marked.
c) marked, unmarked, marked.
i) I DO like your new car. It’s beautiful.
ii) I like your NEW car, not the old one.
iii) She came here to ask for a FAvor.

13) In the clause:

She wants to BORROW your car, not just see it.

a) both Theme and New coincide in the word borrow.

÷
b) the Theme does not coincide with the New element.
c) the Rheme coincides with the Given element.

14) Pitch prominence…

a) fulfills the sole function of marking the focus of new information within the tone group.
b) fulfills several discourse/pragmatic functions.
c) fulfills the sole function of marking a speaker’s turn.

15) Information structure…

a) is only realized by the phonological system.
b) is realized by the syntactic and semantic systems.
c) is realized by the phonological, syntactic and textual systems.

A) ANALYSIS: IDENTIFY the themes in each of the following examples and say whether they are
marked or unmarked. If marked, specify which clause constituent has been thematized in each case:

a) Out she came in a rush. Market

b) Susan called to say good-bye. vnnartaed
c) Are you for or against McCain? vnnarked
d) Never say never again. Marted
e) Right you are!
Market

B) ANALYSIS: IDENTIFY the type of theme: Experiential/non-experiential/detached?

ex
a) Hey dude, give me my pen-drive.
b) I love him. However, I don’t want to talk to him.
c) She will never forgive you.
d) =
That kid, is he your child? aiíin
e) The Rolling Stones. There will be a concert tomorrow night.
ditached
C) ANALYSIS: The following is a fragment of a Joint Debate on Pharmaceutical Products at the European
Parliament, which is part of the anotated corpus found in Nafá’s (2005) study. ANALYZE the thematic
structure of the fragment, as well as its organization structure (Given-New) and explain how both
structures (Theme-Rheme & Given-New) are used to cause certain rhetorical effects. Pay attention to
tonic prominence (see notation conventions below) in the text.

1 /// [HK] [p+] President, ´I have ´admiration // [p] for the

2 ´accomplishment // [p+] of ´scientific research // [p] in developing
3 ´medicines // [p] which have ´proved // [r] of ´great ´benefit. // [hk]
4 [p+] Yet ´also, // [r] I have a ´healthy ´scepticism // [r] ´both of our
5 ´pharmaceutical industry // [o] and ´our… // [p] of our
6 ´exaggerated ´confidence // [p] in ´some of its ´products, // [o]
7 ´many of which ´cause // [hk] [r] a ´great deal more ´harm // [r] than
8 the ´illegal // [hk] [r] ´recreational drugs // [lk] [p] which ´attract //
9 [p] the ´bulk of ´public ´attention. // [p+] And for ´that reason, //
10 [hk] [p] it would be ´wrong // [p] to place ´more ´unnecessary