Appendices - Masaryk University

wastecypriotInternet και Εφαρμογές Web

10 Νοε 2013 (πριν από 3 χρόνια και 7 μήνες)

82 εμφανίσεις



Department of English Language and Literature

se analysis: Interpreting Seven
Nation Army by the White Stripes

Diploma Thesis

Brno 2011


Writen by:

Mgr. Olga

Navratilova, Ph.D.

Bc. Markéta


ext, discourse, context,
, scripts


incoherence, coherence relations


I declare that I worked on my thesis on my own and that I used only the sources
mentioned in the bibliography.

I agree with storing of this work in the library of the Faculty of Education at the
Masaryk University Brno and making it accessible for study p


, 10
th January 2011

Markéta Ševčíková


I would like to thank to my supervisor, Mgr. Olga Dontcheva
Navratilova, Ph.D.,


kind guidance,

fair consideration, and

professional observations







1. Theoretical Background



1.1 Discourse a




Text, discourse, and c




The context of communicative s







1.5 Language and e



1.6 Models

of e




Frames, scripts, s







1.9 Incoherence and d



1.10 Coherence relations







General d



2.2 Description of a communicative s
ituation: Interpreting a song



2.3 Linguistic descript
ion of the song ly
ric Seven Nation Army with
the focus on the textual points of meaning and structural
indeterminacy in the l



3. Methodology



4. Analyses




Interpretation one






Interpretation two





Interpretation three







tion four





Interpretation five





Interpretation six





Interpretation seven





Interpretation eight





Interpretation nine















I n t r o d u c t i on

This works attempts a discourse analysis revolved around a comparison of texts that
can be categorized under a common heading of 'interpretations'. The nine texts
yzed bring reflections on a song lyric that shows linguistic features expected to
hinder making a coherent interpretation of it. Nevertheless, the lyric was not perceived
as incoherent by some its interpreters. At a first glance, some of the interpretation
made to this song bring distinct content structures read into the lyric and also a different
degree of coherence perceived with it. The present analysis of selected interpretations is
expected to enable answering two questions, or reaching two ends.

e first question addressed in this study is whether the degree of coherence perceived
with a text can be considered as influenced by the source of meaning structures read
into and/or used to interpret the text. For this part of the analysis, two broad cate
of a possible discourse source, semantic (or textually based) and pragmatic (or
contextually based), will be distinguished. Relating to this question, the experiential
structures, that is entities and actions, appointed as present in the content of
the song
lyric by its respective interpretations will be isolated in the texts and briefly discussed.

The second aim pursued is to identify coherence relations that can be perceived as
connecting distinct situational modes in the interpretations. Only th
e units that are seen
as expressing, or re
producing the content of the lyric interpreted are included for
analyses. The results of this analytical part are expected to suggest two things at least.
First, a conclusion will be drawn as for what discourse re
lations can be seen as the most
frequent, thus plausibly preferred, in connecting a newly presented situational mode
within an established discourse span. Second, it will be considered whether some
dependencies can be seen between the degree of coherence p
erceived with a text and
the coherence relations that are seen as attaching a newly presented situational mode
within a re
formulated content of this text.

Since the analyses are undertaken on a collection of ten text samples only, the
conclusions reach
ed are considered as very tentative, potentially deserving a more
profound testing to be carried out. The results could be seen as providing new insights



for the field of research that is concerned with coherence in text comprehension, and,
plausibly, also

text production.

The work is divided into four main sections. The
Theoretical B

section the
discusses theoretical issues regarded as basic for understanding the questions tackled in
this work along with the steps taken to answer these questions.

Thus, besides touching
upon the notions of e.g. text, discourse, context, coherence and experiential structures,
also the methods used for the classification of a communicative situation and of
coherence relations are introduced. The second section,
, is unsurprisingly
devoted to a rough description of the texts analyzed, including the description of the
communicative situation of interpretation, and the identification of linguistic features
that may disturb a coherent interpretation of the lyric.
As for the

two background standpoints will be revealed that served for the division of texts into
groups for making conclusions, followed by the explanation of the analytical steps
undertaken. The next section,
, brings inquiri
es into nine interpretations which
are, finally, assessed in Conclusions.




1.1 Di s c o u r s e a n a l ys i s

Perhaps the most expressive, even though not much informative, adjective to describe
discourse analysis is ´multid
isciplinary´. It appears an unmanageable task to assign all
the practical attempts subsumed under this theoretical label into one branch of science
only. Rather than an independent field of study, discourse analysis may be viewed as a
conjunction of intere
sts stemming from anthropology, ethnography, political science,
sociology, psychology, linguistics, and who knows what else. The variety of
approaches taken to discourse analysis reflect the diversity in stances and background
assumptions that this area of

research can incorporate
. The approach taken here
appears to follow cognitive and pragmatic strains in discourse analysis, supposedly
confirming such pragmatic assumptions as that “It is speakers/writers who have topics,
presuppositions, who assign infor
mation structure and who make reference. It is
hearers/readers who interpret and who draw inferences” (Brown and Yule 1983: IX

Pursuits and approaches in the field of discourse analysis are marked by a progressive
development. For its interdisciplina
ry character, the apprehension of even the most
basic concepts can differ from author to author in a relatively significant manner. The
notions of text and discourse are not exceptions. Even though some linguists perceive
and apply the two as interchangeab
le, this does not accord with the view taken here.
Therefore, so that undesirable controversies were prevented, it seems convenient to first
introduce these two fundamental terms as they will be understood in this work.

1.2 Te x t, d
i s c o u r s e, a n d
o nt e x t

To present the perspective on text and discourse assumed here, the description of the
standpoint expressed in Seidlhofer and Widdowson (1999) should serve well. As they
define it, and as this work holds, “text is the linguistic product of a discourse


For a detailed description of 6 (though not all) outstanding approaches cf. Schiffrin (1994).



cess…the trace of the pragmatic process of a communicative interaction”
206), providing evidence that a language is used and suggesting the wa
y(s) in that it is
used. As for the framework proposed by this work, this implies that a text is rather
erial, static and passive in nature, functioning as a collection of purely formal,
materially more
less stable linguistic cues, leaving the sender/receiver fully in
charge of the actual meanings (re

Regarding the concept of discourse,

idlhofer and Widdowson (ibid.) present it as:

the process of conceptual formulation whereby we draw on our
linguistic resources to make sense of reality. It is the continual
consultation and modification of the contexts which constitute our
world and ther
efore a continual renewal of our identity by interacting
with the third person world and second person others.

Slightly digressing from this conception, discourse will be further treated

here as the
product and, at the same time, the continuous process
of meaning reconstruction, that
can exploit chosen textual hints, and that necessarily involves the employment of
various contextual information, for the
purposes formulated by

itself. Therefore, a
discourse is supposed to be effected and/or coming into ex
istence on the necessarily
partial basis of a particular text, and can be conceptualized as a conflux of meanings
that stem to various degrees from the hints (that are thought of as) provided by the text,
with its isolated content and structure utterly dep
endent on the contextual configuration
valid for a respective processor at a particular point, or progression, in space and time.

What this conception of discourse is, too, meant to imply is its assumed meta

and character. Under this view, any i
ndividual’s discourse becomes a ʽthingʼ impossible
to capture in the present conditions, thus a ʽthingʼ impossible to re
construct and
analyze in its full.

Having the role of context in discourse shown, it seems relevant to touch upon the
question of wha
t to imagine under the contextual information. Although conceptions of
context generally share the feature of referring outside a text, the apprehension of this
term may, again, harbor very different ideas depending on the author. For instance, the


Except when used in quotations.




mentioned above delimit context to “an internal conceptual construct”
(Seidlhofer and Widdowson 1999: 210), hence attributing it to the realm of cognition
solely. This insight may, however, appear as excluding the share of perception on the
development of

discourse, or in other words, it may look like completely ignorant of
the features of physical environment that, too, mark the discourse construction. To
illustrate, just remember such an ever
present environmental factor as temperature is.
Its adverse le
vel may cause attentional difficulties which can result in narrowing the
range of semiotic cues perceived and consequently utilized in discourse building.

Upon consulting the work of an earlier discourse analyst, one can encounter a rather
contrary persp
ective of context. Although Van Dijk acknowledges his awareness of the
existence of its mental reflection

or context model, as he calls it, he seems to employ
the term to refer to “a communicative event or situation” (1997: 193), that is, to the
s of) elements of realities that are extrinsic to a human processor and/or her/his

Seeking to consolidate both perspectives, the concept of context is widened here and
can henceforth be used as a cover term for any elements of any reality tha
t are
perceived as relative in a current discourse by its processor, and/or are discursively
influential without her/his conscious perception, and/or are discursively related to a
current discourse.

1.3 Th e c on t e x t of c o mmu ni c a t i v e s i t u a t i o n

It has man
y times been proposed, and is still agreed on, that the context of
communicative situation, including its conception lived by the processor, influences the
choice of language means used in interaction. To set one influential example,



authors with similar perspective to language

as listed by Van Dijk 1991:
Argyle, Furnham &
Graham 1981; Cicourel 1987; Cook 1990; Dascal &

Weizman 1987; Duranti & Goodwin 1992; Forgas
; Givón 1989; Gumperz 1989; Gumperz & Hymes 1972; Hymes 1972; Watson & Seiler 1992.

Nevertheless, in his recent writings (e.g. 2006), a con
ception similar to the one by Seidelhofer and
Widdowson is presented.



Halliday´s observation

can be cited that we can better explain “the actual sentences and
words” when “we can pay attention to situations of language use” (1978: 28).

It seems relevant to mention here, that the discourse analysis undertaken in this work
examines ten text samp
les that, however, do not instantiate one register only. To be
more specific, while one of the texts can be thought of as belonging to the genre of
song lyrics, and, therefore, as embodying a component of a message that would be
comprehensively classi
fied as multimodal, the remaining nine texts can be
assigned to the register of written interpretations and/or summarizations. The latter
texts were produced with at least one common purpose

to provide an interpretation
and/or summarization of the former
, a song lyric. Thus, referring back to the above
paragraph, also (the conception of) the communicative situation lived by the
interpreters of the song discussed in this work influenced the development of their
discourse of interpretations.

It is, presum
ably, not possible to reconstruct the perception of situation that the
interpreters were experiencing when they came across the song lyric and/or when they
wrote their interpretational contributions; even less possible it appears to do so in a
account such as this thesis is. The interpretations discussed are retrieved
from, a webpage that by a mere title incites to contribute with a
comment on the lyrics of miscellaneous songs. While making their contributions, the
(more than likely) perceived distinct textual parts as subjectively salient,
conceived of its content under distinct, untraceable pragmatic configurations, and it can
also be expected that the understanding of semantics of the lyric decoded varied from

interpreter to another depending on their social and cultural background, their
idiolect, etc. What appears possible, though, is to comment on some aspects of such a
discourse situation, or, in linguistic terms, some features of the text’s register can be

described. This attempt will be performed in the

part. What follows is the
theoretical introduction of the model used for this end.



1.4 Re gi s t e r

Textual descriptions with regard to selected aspects of a surrounding communicative
situation ha
ve become conventionally facilitated with the help of outlines hiding behind
the explanatory linguistic construct of register (Biber and Conrad 2009: 6). Turning to
one entrenched sociolinguistic paradigm of register, a communicative situation
a text can be differentiated in terms of its three outstanding parameters:

Types of linguistic situation differ from one another, broadly speaking,
in three respects: first, what is actually taking place [field]; secondly,
who is taking part [tenor]; and
thirdly, what part the language is
playing [mode]. These three variables, taken together, determine the
range within which meanings are selected and the forms which are
used for their expression.

(Halliday 1978: 31)

Such a formulation of situational

dimensions is perhaps rather nebulous but, on the
other hand, it can endow an analyst with a freedom in selection of such situational
features for discussion that are more relevant to her/his thematically motivated work
than the others. Actually, Halliday
´s framework has become a frequent working tool of
analysts who deal with the genre distinction; it has become a framework commonly
used in aiding genre
based studies and defining the varying categories of registers.

Nevertheless, the division of a commun
icative situation need not be done along the
dimensions referred to above only. As Halliday himself points out, there is a number of
schemata that are comparable with his descriptive model (cf. Beedham, 2005: 142
Biber and Conrad, 2009). For instance,

Biber and Conrad (ibid.: 40) propose seven
situational variables that, nevertheless, all except one, communicative purpose, appear
as issuing from the Halliday´s conception. Although this seven
parameter descriptive
direction may be seen as more comprehe
nsive, it can also, and rather adversely,
sidetrack the analyst’s attention and, subsequently, induce the negligence to those
situational aspects that are not explicitly articulated in this framework, but that can be
more significant for an interpretation
of a particular text than the ones appointed as
generally major.



In this work, Halliday´s model will serve to draw near some aspects of the
communicative situation that could be sheltered under the general heading of “song

1.5 La n gu a
ge a n d Ex p e r i e n c e

In the previous parts, text has been established as one of the formal inputs/outputs to
generate/be generated by discourse, or the processes of meaning construction.
Discourse was, in turn, conceptualized as a matrix of realities, experi
enced via making
sense, thus, establishing relations. For an individual, a reality can be viewed as
composed of all that s/he can recognize, believe and suppose to know. To compose a
picture of reality, one can only exploit the information that has come to

her/him through
own physical, emotional, dream, or any other world experience. As Blahnik puts it, to
explain any reality on a different base than experiential would be “not only superfluous
but inaccurate, misleading, and illusory” (1997: 23

The p
revious claim can, no doubt, be extended to a language analysis too, since also the
meaning of an utterance is more than likely to be grasped as a means of and/ or a part of
some world experience. Even discourse, or meaning reconstruction, presents itself

What can be seen as following from this observation in particular is the possibility to
furnish the analysis of the meaning (i.e. semantico
pragmatic) level of a language with
the units that are delimited experientially. Such an approach to language an
alysis is
actually exploited in many fields of soft science, including, for example, cognitive
linguistics, and will be employed for the analyses included in this work too.

1.6 Mo d e l s o f e x pe r i e n c e

For answering the main questions of this work, the c
oncepts that are rooted in
psychological, sociological, and computational approaches to language will be



employed in the analytical part (Brown and Yule
1983: 236, 241). Mental
models, frames, schemata, scripts, scenarios, scenes, domains, mental

spaces: all these
conceptual categories have proved as working tools in aiding, for example, the
capturing and exploration of the background knowledge that is used in the meaning
reconstruction (Brown and Yule 1983: 236
255). Besides, framing is seen as a

semiotic principle, realized by different semiotic resources in different semiotic modes
(Van Leeuwen 2005: 14). Despite, unsurprisingly, the treatment and understanding of
the above terms differ with authors, there is a conceptual core that the ab
ove mentioned
categories are expected to share: “they all designate a coherent organization of human
experience” (Kövecses 2006: 64). Hence, not only background knowledge but also the
contentual structures built in discourse can be analyzed down into these

determined units.

Expectedly, a distinct (and likely even the same) formal input will generate
heterogeneous mental models. Their structure would show varying complexity and their
content distinct composition, contingent upon the differen
ces in the complexity and
composition of both the formal input and the cognitive and sensory properties of a
particular processor. It can seem interesting to note that the mental models established
by a processor after interpreting a text can be thought of

as highly complex and, at the
same time, relatively complete units of meaning that

in case they are stored in a
processor’s long
term memory

may be further retrieved in their full, partial, and/or
more abstract form, and modified in still new and new
discoursal/cognitive operations.
It is also plausible to compare discourse to one existing frame (system) that undergoes
constant re
construction, shifting and re
fillment by manifold sub
frames, sub
systems, and structures. Out of the two, it is the

latter possibility that corresponds with
the understanding of discourse assumed here.

As it
has been indicated before, some of the above mentioned terms were used in order
that further analyses could be endeavored. In the analytical part, discrete catego
ries are
established in order that some meaning structures that can be seen as coming off the
interpretations could be classified
. In the next sub
sections, the reader will be presented
both with the sketches of the original conceptions and with their modi
fications utilized
in this work.



1.7 Fr a me s, s c r i p t s, s c e n a r i o s

The frame theory was originally developed by Minsky with a purpose slightly different
than to facilitate language analysis. Described as indefinite data
structures that are
stored in an indi
vidual’s memory and recalled by her/his cognition when the need
arises, frames were designed to explain the way in which knowledge is operated by
human brains. These structures can be expected to “represent stereotyped situations”
and their components can
be subdivided into obligatory and optional (to provide
Minsky´s example, ´wall´ is to be considered as obligatory element and ´decorations´
an optional element of a ´room´ frame). “The basic structure of a frame contains labeled
” which can be filled
by other frames, therefore, a number of frames may be
recalled at once, on the stimulus of one textual hint (Brown and Yule 1983: 238
This permits to think of a discourse as of an endless process of the recall, (re
transformation and storag
e of experiential hierarchies, or systems.

Later, the frame theory served Schank as grounds for the development of another
conceptual category, that of scripts. In contrast with the conventional use of frames for
rather inactive, stationary (conceptual)
entities, scripts have been associated with
knowledge structures that yield actions, activities and “event sequences” (ibid.: 241
244). Although further discussions of this concept could definitely be incorporated in
this works, such an attempt will be exc
luded for the sake of its irrelevance to the topic
of this work.

The last terminological construct that will be made use of here, scenario, is to

describe the ´extended domain of reference´ which is used in interpreting
written texts, ´since one can thi
nk of knowledge of settings and situations
as constituting the interpretative scenario behind a text´.

(ibid.: 245)

The motivation for the choice of experiential structures for the present analytical
attempts is, besides pre
conceptions already expressed
, based on the observation that
people (or discourses) in every
day situations freely handle experiential structures
which can be seen as representing people, things, actions, events, and situations. The



modification of the above outlined concepts can thus

reveal what particular people,
things, actions, and situations could be perceived as coming from the song lyric

As for this work, the conceptions of frame, script and scenario introduced above will be
slightly modified for reaching the aims in
tended. Hence, the first term sketched, frame,
will be used here to shelter basically any entity identified in a particular interpretation,
and seen as employed by the particular interpreter to express the meanings read into the
song lyric. A frame is thus

recoverable after any lexeme or text part that designates an
animate or inanimate entity, i.e. a person, a thing, a collection, etc.

In the real world, people and things can remain quiet and stable at one place, but they
can be involved in some actions t
oo. To create a conceptual category for the isolation of
various kind of states, actions, happenings, and events, that frames may (and
conceivably must) be parts and/or conductors of, the term script will be used further.
Therefore, a script is recoverable

after any lexeme or text part that designates a state, an
action, and an event. These two categories will be used for the identification of
particular meaning structures read into a text, which will serve for answering the first
question posed in the intr

Last, people, things and actions are naturally bound to some place and time which
actually co
defines the idea of the surrounding reality, and, as for discourse, helps to
locate and order scripts and frames involved in it.

To illustrate, peop
le commonly make future plans, which involves creating hypothetical
situations, relating them to past events, and, finally, attempting the plans construed. All
these operations are definable as employing various experiential structures that adhere
to varyi
ng modes of reality, or, simple terms, situations. Analogously, the last
terminological construct presented, scenario, will be used to set boundaries for the
assortment of experiential structures under its respective mode of reality, or situation.
As scrip
ts can be seen as structures that, at least potentially, involve frames, in the same
way scenarios can be seen as involving scripts (and thus also frames).



1.8 Coh e r e n c e

In the
Text, Discourse, and Context

part, there has been established a covert divisi
line separating (linguistic) reality into two spaces: the material and the ideal. For its
nature, text has become to represent an integral component of the material, physical
world, while the existence of discourse has been appointed as projecting in th
e ideal,
mental space of a processor. The terminological construct, that will be turned to now,
may be seen as interfacing the two.

Although generally recognized, coherence appears as the most elusive of the notions
developed in linguistics, conceivably
involving manifold phenomena, and, therefore,
achieving various dimensions. Even more puzzling might seem the fact that this aspect
of discourse progression can be extended to the explanation of extra
linguistic affairs.
The lack of a shared definition may

even induce a picture in that coherence works in a
distinct manner for each of us, which would, theoretically, exclude the possibility to
ever agree on common measures of (not only) coherence. What can be seen as, at least
implicitly, present in several w
orks that try to handle this topic are some faculties that a
coherent text, or rather its interpretation shows, for example, unity, continuity,
comprehensibility, smooth flow in reading, connectedness, and good overall
organization. At this point, the desc
ription claiming that coherence shows as “the
dependent interpretative perception of the semantic unity and purposefulness of
discourse” (Dontcheva
Navratilova 2009: 100) is adopted

For a brief and synoptic
comparison of other perspectives on cohe
rence cf. Kehler (2004).

1.9 I n c o h e r e n c e a n d Di s t u r b e d Co he r e n c e

It seems clear that coherence need not be evoked in every discourse, or, simply said,
not every text we encounter necessarily has to make sense to us. Texts that appear in
situations of t
his character, that is situations that are conceived of as nonsensical, or



difficult to understand, have become to be labeled as incoherent
. In 'lighter' cases, that
can be encountered in numerous communicative situations “when the extent to which
the tex
t is only partly understood is no longer tolerated by the hearer” (Bublitz and
Lenk 1999: 155), the texts appearing would be classified as bringing disturbed
coherence. Bublitz and Lenk (1999) suppose, too, that “even though there
italics] de
viations, we permanently assume that our fellow
speakers [and writers] are
creating coherent discourse” (157), which they refer to as default principle of
coherence. If this assumption holds, then coherence can also be seen as a kind of
discourse norm agai
nst which the mental models constructed are tested, helping
discourse to overcome breaks in the continuity of information flow/exchange. This
would imply that a discourse progression is directed so that it always finally arrives at a
personally coherent in
terpretation of any incoming information

paradoxically, even
with the utterances, or communicative situations, that succumb to a temporal state of
incoherence, resulting in apparent moments of break in comprehension.

Since this work is primarily intend
ed as a quest into interpretations of a text that, in my
view, shows language features of disturbed coherence, an illustrative outline of (some
of) its sources as identified by the authors quoted above follows:

Unclear reference
does not have to come wit
h deictic expressions only.
Virtually any lexeme whose referent does not appear as obvious falls into
this category

(ibid.: 165

Topic drifts
can be, by virtue of their occurrence on either global or local
, distinguished

into two types. Insta
nces of the former type may
emerge when “the global topic is temporarily neglected or even
permanently abandoned”, while the latter type disturbances can be observed
provided/on condition that coherence respected on the global level hampers
“checking on lo
cal coherence”
(ibid.: 167


The observation of this phenomenon leads me to note that coherence can be seen not only as perceptual
but also as attributive quality.


At the

level, micro
structures (i.e. ph
rases, clauses, sentences, etc.) and relations in language are
studied, while at the
, or macro, discourse level structures (e.g. topics, speech acts) and relations
are of concern.



Topic change

is, as the previous source, discernible only on recognition of
global and local (though this time not topic but) coherence, and has been
classified into two categories: “abrupt topic change and announced topic

change…[A]brupt topic changes are always locally incoherent but coherent
on a global level provided the new topic is a noticeable contribution either
to a superordinate topic or to a general conversational goal on the
interpersonal level of communication”

(Bublitz and Lenk 1999: 168

Another common cause of disturbed coherence may be discovered when
explaining discourse in terms of experiential structures, and it is
(ibid.: 169

And, finally, the last plausible conception explaining

for breakdowns in
coherence has been appointed as
register break,
that is, unsignaled (and
unrecognized) switch in register (ibid.: 170

1.1 0 Co h e r e n c e r e l a t i o ns

The apparent difficulties in defining coherence, however, prevent neither academics
me from researching on this topic: a growing number of studies have been conducted
mainly on the types of coherence and, what is relevant for the analyses attempted in this
work, on the categories of coherence relations. As Taboada (2006: 2) points out
coherence relations may sometimes be interchangeable with discourse, or rh
relations. In this work
, the terms coherence relations and discourse relations are used
interchangeably, referring to the relations that hold between distinct situation mo
, or

that are yielded by content
retelling parts of an interpretation in discourse.
These relations can be seen as elements connecting meaningful fragments of textual
experience in coherent discourse. In the words of the same author:

nce relations

relations that hold together different parts of the

are partly responsible for the perceived coherence of a



text. More specifically, the recognition of coherence relations by the
hearer or reader enables them to assign coherence t
o a text.

(Taboada 2006
: 1)

Although they need not be formally marked, there have been identified certain
expressions, commonly termed as discourse
markers, which “guide

the text receiver in
the recognition of those relations” (
: 1). Apart from sema
ntically empty lexemes
such as connectives (
and, but, anyway, nevertheless
), or interjections (
oh, damn!),
some fixed expressions (
I mean)
have been identified to perform this function. Despite
the guidance that discourse markers are assumed to provid
e, it is still important not to
forget that there is no one
one mapping between discourse markers and discourse
relations: a marker need not signal one possible relation in discourse only (Knott and
Sanders 1998: 9). What makes the whole issue even more

omplicated is that various
clues show as influential in the identification of connections between interpreted text
parts. Next to discourse markers, e.g. tense, mood, and verb meaning can serve as
grounds to establish relations within and across text spa
ns too (Taboada 2006: 2).

Embarking on varying background assumptions and aims, several theories have been
proposed to describe, analyze, and classify discourse relations (cf. Longacre 1983,
Hobbs 1985, Grosz and Sidner 1986, Martin 1992). One outstanding

framework, The
Rhetorical Structure Theory by Thompson and Mann
, “identifies hierarchic structure
in the text” (1988: 243), and puts forth a definite number of possible discourse

though, as its authors acknowledge, also other relations can be

added on the
list. Relying on the reader’s/analyst’s recognition of plausible writer’s intentions, this
descriptive paradigm enables to draw a possible rhetorical structure of a text (spans)
with any length. The typology of relations proposed is claimed t
o be founded on a
parameter method of classification: the four fields that serve to define a relation
are its assumed Constraints on the Nucleus, Constraints on the Satellite, Constraints on
the combination of Nucleus and Satellite, and its Effect.

The approach to the treatment of coherence relations that is made use of in this work
was developed by Sanders, Spooren

and Noordman (1992, 1993) and it can be


For the comparison of RST to other approaches see Taboada and Mann




understood as advocating the claim that coherence relations “model cognitive
constructs” (Knott

and Sanders, 1998: 6). The authors argue that coherence relations
are synthesized concepts that can be analyzed down into smaller, cognitively based
pieces. This framework proposes four cognitive dimensions, namely Basic Operation,
Source of Coherence, Po
larity, and Order of Segments, expected to be shared by all
coherence relations. Therefore, as Knott and Sanders handily summarize, any
discourse relation recognized is deemed to be definable via the values of its:


BASIC OPERATION: every relation is deem
ed to have either a
CAUSAL or an ADDITIVE component. CAUSAL relations are
those where a ´relevant´ causal connection exists between the spans;
all other relations are ADDITIVE.


SOURCE OF COHERENCE: It is semantic if the spans are related
in terms of their
propositional content and pragmatic if they are
related because of their illocutionary force.


POLARITY: a relation is POSITIVE if its basic operation links the
content of the two spans as they stand, and NEGATIVE if it links the
content of one of the spans

to the negation of the content of the other
span. Negative polarity relations typically involve either a violation
of expectation, where the expectation derives from a causal basic
relation; or a contrast, where the basic relation is additive.


GMENTS: this distinction only applies to CAUSAL
relations; they are deemed to have BASIC order if
the antecedent is
on the left, and NON
BASIC order if it is on the right.

(1998: 7)




2. 1 Ge n e r a l d e s c r i p t i on

As was imparted in the
retical background
, it is the situation

including the


decides on how a discourse operates with words

be it in the
stage of their processing or production. Similar meanings can be inserted into dissimilar
words and similar words
can convey diverse meanings. Nevertheless, it stays a fact that
after recognizing words, or a language utterance, almost every person can provide a
report on what s/he read/heard

be it verbally in his discourse for her/himself only, or
be it for public t
hrough a
speech, writing, and even other semiotic mode. The analytical
efforts in this work are centered around nine texts

that were produced to provide such a
report. In particular, the analyses are pursued based on ten interpretations and/or
ns of the song

Seven Nation Army

performed by the White Stripes.

Despite the
lyric of this song is not directly inquired into in the analytical part, it was
exploited to address one of the issues of this work
. Since

one of the main


this work

is concerned with the source of experiential structures read into a text, the
lyric served for comparing the experiential structures that were read into it by its
different interpreters. Only after it became possible to set the source of these structures
and use the results for making appropriate conclusion. For this reason, the text is
included in the appendices. As for the texts analyzed, every interpretation is inserted
before its respective analysis.

Although varying versions of the song lyric are av
ailable on the web, all the above
mentioned texts were retrieved from one web page
. By its name, composition, and the
authors´ note, this website invites its visitors to express a comment on the meaning, that
is the content, of an array of song lyrics pre
sented there. Such an invitation presupposes
a visitor’s own initiative to contribute, as well as her/his state of being acknowledged
with the lyrics s/he is commenting on to a certain extent at least.





What can be seen as contributing
to the genuineness
of the comments inserted is the
virtual arrangement of this Webpage, is that it offers

a relatively high degree of
anonymity and, therefore, freedom for expression to its contributors, avoiding the
adjustment of content according to societally based restri
ctions, or conventions
(including the norm of coherence).

One thing may be seen as a potential modifier of the messages articulated there, and
this is the fact that only persons that are subscribed and identifiable according to the
nickname can

make their

contributions there. Nevertheless, as this virtual identity is
valid only in the space of the website only, even this feature of the website need not
pose any great limitation for expression.

Of course, the discussion concerning the constraints on expre
ssion that can be created
by the actual context of contributors to the webpage cannot be included here. But this is
a matter that, expectedly, can be included in no analytical work. Hence, it is assumed
that the commentaries

to be found on the webpage

de the meanings that are very
close to the ones that the interpreters actually discoursed. The contributors are supposed
to have expressed taking the voice of their own, rather than operating within the mode
of their socially confirmed persona on the webpa

When providing their comments, the interpreters, I believe, included the linguistic
expression of meaning structures they arrived at after decoding the same language
input. This input did not have to get to the interpreters in the same form exactly,
is simply because the interpreters might have try to introduce the structures they
reconstructed either on the basis of audio input (i. e. listening to the song) or on the
basis of audio
visual input (i. e. listening to the song while watching its vi
deo clip).
Nevertheless, both these variants of presentation introduce the linguistic aspect of the
message in approximately the same format. What is typical about the text interpreted is
that it can be assigned neither the status of a spoken text, nor the

status of a written text:
though transmitted in sound, we cannot treat it as speech for its distinct prosodic
character and it’s pre
formulated content and structures make it close to writing.



2. 2 De s c r i pt i on

of a c o mmun i c a t i ve s i t u a t i on: I
n t e r p r e t i n
g a

s o n g

At this point, some general observations concerning some features that might have been
influential upon the interpreters will be included. It is assumed that the communicative
situation described must have preceded giving the interpretations of
its text since it is
very unlikely that other people than its listeners woul
d deliberately decide
to express
their comments on the song lyric. This part has been included so that a reader
her/himself could consider what aspects of a communicative situation

might have
modified the meanings read into the lyric. As a basis for this task, Halliday´s schema
introduced in the Theoretical background will be used.


The participants in this communicative

situation are not involved in a direct

act, or exchange of information. Thus, it can be claimed that there
exists no direct relationship between the participants. There are two types of
participants engaged in this communicative event. On one side, there stands a
performer of a song, who is a
t the same time its author in this case. This person
represents an active force in moulding the articulation of the verbal message
transmitted, which, moreover, can be seen as a part of more complex, pre
multimodal message. On the other side of
this communicative act, one from the range
of its receivers, most likely a listener, would be found decoding the message sent.

It seems relevant to mention the fact too that the
participants in this communicative

event perhaps did not share one spatio
mporal setting, since, most likely, the listeners
processed a recording of the song. Next, the interpreters probably do not know the
singer personally, that is they could use only such contextually relevant information in
interpreting that came to them thr
ough media. Interpreters might have been influenced
by their personal attitudes towards the singer and/or music that he (with his co
produces, and even by the assumed ideology he puts forth.




What is going on in this communicative situatio
n can be imparted in one phrase:
listening to a song, or, in a more sophisticated way, a perception of musical production.
Such an activity is basically an indirect social (human
human) interaction belonging to
the sphere of entertainment. This interaction

is indirect in that one side of the
communicative event is passive, purely receiving and interpreting, not actively
influencing the form and content of the message received.

Assumedly, the interpreters did not hear the song once, which means that they
have more opportunities for making an interpretation and, thus, for the (re
establishment of its mental content model. As the type of communication described is
indirect, with the producer not physically present in the environment where the
etation takes place, it is not even possible for the interpreters to ask for any
clarification of inconsistencies in the message. This implies an interpreter’s
involvement in a highly independent, individualistic mode of interpretation. Since a
song embodi
es a multimodal form of a message, also other aspects than verbal could
affect the content structure assigned to the interpretations.


The song lyric interpreted is a text written in advance, possibly with some underlying
conception, to be transmitt
ed repeatedly, and primarily intended to be received through
the sound channel. Though, this cannot exclude the possibility that its interpreters got to
the text

for any sake

in a purely written form, too. Besides, the text can be viewed
as forming a c
omponent in a multimodal message that is commonly transmitted in two
modes of delivery: the first mode is purely audial

the song alone, with the form of its
verbal component defined by the vocal mode of singing, and acted upon by
instrumental components.

The second mode involves audio
visual input

the song
accompanied by an intentional visual support in the form of a video. (It is assumed here
that the interpretations are not based on a live performance of the song. This mode of
delivery can influence t
he verbal message proposed by a singer to different degrees, but
usually based on an assumption that a concert participants are acknowledged with



words of the song in its recorded version.) Both variants offer a slightly different formal
potential for the
reconstruction of meaning, as the “story” presented in a music video
may add to the content structure of it interpretation.

2. 3 Li n gu i s t i c d e s c r i p t i o n o f t he s o n g l yr i c Se ve n Na t i o n
Ar my wi t h t h e f o c u s o n t h e t e x t u a l po i n t s o f me a n i n g a n d
s t r u c t u r a l i
n d e t e r mi n a c y i n t h e l yr i c

An underlying

assumption of this work is that the song lyric, that served as semiotic
grounds for origination of the interpretations examined further, shows linguistic
features that can be considered as complicating its decoder’s

arrival at a coherent
interpretation. Therefore, the following description of the lyric is focused on
those of its

aspects that may cause incoherences, or bring disturbances to coherence in its
procession. For the limited range of this work, do not consid
er the survey comprised
within the next paragraphs as comprehensive. Only the most salient features at the
lexical, grammatical, and pragmatic levels have been included
in the following

From the linguistic point of view, the original text can

seem to require the listener, or
, to show a relatively

high degree of independence, creativity, involvement, and,
to perform many other

pragmatically describable operations, so that links
for achieving a coherent content interpretation
could be established. Evidence
supporting this claim can be found at multiple levels of language representation.

As it emerges after reading through the lyric interpretations, the textual features that
can be expected to hinder its coherent interpreting
need not have such an effect.
Conversely, the semantically indeterminate points in the lyric may offer space for the
establishment of more plausible and relatively diverse interpretations. Indeed, the
extent of textual ambiguity and inexplicitly encoded r
elations that seems coming with
the lyric actually offers more possibilities for inserting interpreter
relations than it appears to be with texts where relations on various levels are formally



hinted in a more direct, overt way. The outline o
f the (most salient) textual features that
can be considered as providing space for a varying degrees of ambiguity and more
directional reference follows.


There have been identified three (potential, of course,) sources of ambiguity in the lyric
hat can be seen as emergent from the employment

of certain lexical expressions.
text brings:


first, semantically indeterminate expressions and lexemes with an indeterminate
reference (cf. subsection pragmatics below),

I, me, myself; them, their
; me, it, you, that, seven nation army, what you
want to hear


second, lexemes with generic/general meaning or reference

nation, army, everyone, take time, a story to tell


and, third, lexical units with abstract meaning or reference

mind, f
orget, know, want, home, forever more

As for the first category, the disambiguation of these textual items is conventionally
assumed to be fostered by the pragmatic trace of their some
world referents, the
establishment of links being effected via exopho
ric and endophoric reference
expressions. This, however, does not appear to be the case with this text, with its partly
like character providing the most handy explanation. Even such a basic
requirement on a successful comprehension as identifying t
he referent of the speaker, or
is made complicated here, since it cannot be set with certainty whether the speaker of
the lyric can be identified with its singer/author, or with some other, possibly even
invented, person.



Regarding the second potential

source of ambiguity, lexemes with general reference,
every interpreter may have a different idea of (some
world) referents that are hidden
behind these expressions, and, in turn, influence the relations coined to the text. In
addition, the

ambiguity of le
xical units with abstract reference rests in that the
composition of the concepts they shelter is highly idiosyncratic, thus likely to have
different dimensions with each of us, as they, compared to concrete concepts, have a
purely cognitive basis.

des, the lack of contextual information can also contribute to the indeterminate
meaning of phrases as, for example
fight off, rip off, hold back, the stains coming from
my blood,

etc. These can be apprehended either literally (i. e. accepting their semant
or figuratively, which is proposed here as a more likely variant, resulting in the
pragmatic search for analogies and resemblances, which can even widen the scope of
plausible referents in result. Besides, there are also items that allow to be interpr
eted in
terms of figurative language only, e.g.
the hounds of hell, the stains…tell me


Regarding grammar, the structural patterns and irregularities that the lyric yields
approximate the structural layout of speech, which brings omissions, uncl
ear structural
boundaries, transitions, etc., and thereby allows for a greater number of possible
combinations of meaning relations read into it. Considering the fact that the lyric is
transmitted as neither spoken nor written variety of language (it is su
ng), the prosody
that accompanies the message cannot be exploited as a reliable clue for disambiguation.
Another salient grammatical feature that can considerably confuse an interpreter of this
text is the inconsistent use of tenses throughout.



atic level

One essential task that the receiver will face when striving for a coherent interpretation
is to assign referents to the deictic expressions that are to be found in the text. One can
encounter these instances of personal deixis in the text
: I
(me, my, myself), them (their),
it, you
. There is also an instance of a discourse deictic expression:
. It
is in the full

competence of a receiver to disambiguate these lexemes and attribute them the co
textually missing connections, or referents. What

makes the whole task even more
complicated is that at some

points it can seem not clear at all whether the deictic
expression in use is pointing to the same entity each time it is employed. Also the
temporal settings of the events mentioned in the
lyric remains unclear and,
moreover, it stays a question whether the events reconstructed upon the text are to be
assigned to the real
world or hypothetical situations.

With respect to rhetorical relations, the lack of their overt marking, or, in other w
the lack of marked linking across discourse
level information units, invokes further
decisions that must be made so that the text was disambiguated at this level too, and so
that this
level units could be connected in a coherent manner. The only con
juncts, or
means that can be perceived as explicitly establishing discourse relations are
both of them, however, conventionally classified as semantically

The frequent change of the topics in consequent clauses can be p
erceived as another
aspect that disturb the process of interpretation (cf. Bublitz and Lenk), widening the
range of meaning relations that are possible to be generated in discourse too. The
listener/reader is left to explain the switches in the content of
consequent information
units on her/his own, or, to put it differently, to furnish these unmarked relations (that
are, though, necessary for arriving at coherence) between propositions via pragmatic

One general conclusion that can be made now
is that, facing all the discourse
complicating features that has been outlined above, and assuming that one reader’s



main aim is to arrive at a coherent interpretation, it seems evident that, in case s/he
succeeded in accomplishing this aim, s/he (I dare s
ay) could not manage based on the
attributable information only.




The Theoretical Background and Material parts have, hopefully, contributed the
information necessary to comprehend both the questions tackl
ed in this thesis and the
material serving for answering these. Nevertheless, two important background
assumptions have not been introduced yet. The opening lines of this section repair this
drawback and, in the rest of this part, the framework used for an
alyses is

As mentioned above, two interconnected background standpoints remain to be
revealed. They are related to the issues of coherence perceived as coming with a text
and of the certainty expressed about the meanings read into this text. T
hese standpoints
are taken as valid for those interpretations which can be typified as containing an
expression of a stance towards the interpreted text and an actual interpretation of it. All
the instances of interpretations chosen for analyses in this wo
rk belong to this class.
Hence, for such cases of interpretations it is assumed that:


A certainty about the meaning of a text cannot be reached and expressed unless
the perception of coherence is present in discourse. To put it as an example, if I
feel th
at a song does not make sense to me, I will hardly believe and claim that
“I know” what it is about, excluding the possibility to provide a deliberate
report of its content.


The textual parts conveying a writer’s stance
to a text being interpreted
can be

as suggesting (the degree of)
her/his certainty about the meaning
reconstructed upon a text. In other words, I will say “I know what it is about”
only when I am certain about the meaning of a text, while, conversely, when I
am not certain about
the meaning of the text I am going to report on, I will
rather say “I think it is about”, “It might be about”, etc.

Based on these standpoints, it is further supposed that the interpreter’s degree of
certainty about the meanings read into a text can be e
quated to the degree of coherence
perceived with this text. Therefore, the expression of an interpreter’s stance towards the



content that s/he is attributing to the song lyric

can and will be taken as communicating
also the degree of coherence perceived b
y the interpreter as coming with this lyric.

Embarking on this view and, thus, trying to reflect the
degree of an interpreter’s
perception of coherence, interpretations will be classified into the following groups



interpretations suggesting t
he highest degree of coherence
perceived with the lyric





interpretations suggesting the highest degree of coherence
perceived with the lyric

With regard to the units used for analyses, three types of them are employed, each

a different type of experiential structures that the interpretations can recall
in discourse. The first two categories of units are provided by the modification of
'frame' and 'script' conceptions and will be used to report on the particular entities and
actions that can be read into the song lyric (for the definition

of the concepts as used in
this work cf. section 1.7). These will be consequently compared in terms of their
assumed source. For this end, there are differentiated two possible sources of me
structures: the semantic (or textual) source and the pragmatic (or contextual) source.
This distinction can be seen as following the lines proposed by Sanders, Spooren and
Noordman (1992) and Van Dijk (1977).

In more specific terms, every meaning st
ructure coming off a particular interpretation
that can be described as a frame (i.e. an animate/inanimate entity) or a script (i.e. an
action, state, event) will be pointed out and, next, its source will be set. In cases where a
particular structure appea
rs as possible to be either directly read into the song lyric, or
shows as based on the propositions of the song lyric, this structure will be classified as
issuing from the semantic source. In cases where the structure does not show any such
connection wi
th the song lyric, it is classified as a contextually based structure featuring
the pragmatic source.


In the interpretations expressed by structures such as
I know it´s about, I´m sure its about, it´s about,



Having done this, and with the coherence grouping established above, the grounds will
have been prepared for answering the first main questions of this w
ork that asks
whether some dependencies can be traced between the degree of coherence perceived
with a text and the source of meaning structures read into and/or used to interpret this

The discussion that relates to the second main issue of this wo
rk, coherence relations

will, too, be a discussion exploiting the units of meaning that have been determined on
experiential basis. In the parts of analyses that touch this issue, the third type of
experiential structures, the modified conception of ´scen
ario´, will be used. Drawing on
its original definition (cf. Theoretical Background), scenario is understood here as
designating a content situation behind, that, in a rather abstract way, interconnects
related frames and scripts within one distinguished s
ituational mode. A particular
scenario will be isolated, thus an individual discourse
level structure established for an
analytical model, anytime a switch between situations is perceived in the span of an
interpretation that re
tells the content of the ly
ric. To put it differently, the conception
of a scenario is used here to divide the report of the lyric content into discourse level
units, between which coherence relations are established. As it is outside the scope of
the interest

of this work to use sc
enarios for anything else but setting boundaries to
mental model situations, any characterizations of the scenarios isolated will be
withdrawn. The employment of scenarios for the delimitation of discourse units should
prevent confusion of various levels o
f discourse production, and grant that only the
relations holding at one discourse plane will be studied.

As it was mentioned in the previous paragraph, only the relations seen as holding
between the parts of interpretations that convey the reformulation
of the lyric content
will be discussed. This is expected to draw near the relations that hold together one
discourse span, a span that project an interpreter's mental model discoursed upon, and
attributed to, the song lyric. Therefore, the textual parts us
ed to directly express the
stance of an interpreter, and to provide a support/explanation of the content structures
presented are left out of considerations.


or discourse relations (cf. 1.10)



For each interpretation, an analytical model was drawn. For

making the work more
transparent, th
e models are placed in the text under a corresponding analysis
. These
models contain the parts of interpretations that were identified as re
telling the content
of the song lyric and the parts functioning as the discourse markers used by an
interpreter to
connect these content expressing parts. The former are written in boxes,
with each of the boxes supposed to contain one scenario. The latter can be found
highlighted in italics next to the arrows that are used to indicate the presence of a
relation. The la
st type of an item that can be encounter in the models is the
classification of a relation identified according to the taxonomy delivered by Sanders,
Spooren and Noordman (1992, 1993), written in Capital letters.

In the analytical models, two
level (re)p
resentation of scenarios is established. The
scenarios connected in horizontal lines are the ones that, on one hand, can be viewed as
independent, distinct discourse situation modes, but that, on the other hand, only when
taken together form a discourse st
ructure standing on the same level in discourse as the
scenarios drawn in vertical line (order).

Next, in the models, each newly arriving information unit, or a scenario, is regarded as
being linked to all the units that preceed it understood as a whole.
Thus, for example, a
relation seen between

4 and

5 is not perceived as linking these two
scenarios only, but is intended as a formalization of the relation established between the
(given, processed) complex information read into

1,2,3, and 4 together, and
the information newly arriving in

5. This approach may suggest what
relational steps, or coherence relations, can be seen as incorporation new information
into one established content model in discourse
. For the discou
rse markers offered by
interpreters were reflected during the classification of relations, they are printed in the
analytical models too.

In the analytical part the interpretations are ordered sequentially. Each analysis consists
of three sub

with each sub
section focused on a selected language feature:
frames, scripts, and coherence relations. As for the parts of the analyses that deal with


Of course, at times, the appointed type of relation can seem
as not the only one possible to be inserted.



frames and scripts, they serve to introduce the instances of this particular type of
experiential stru
cture that have been identified in a respective interpretation as
employed for retelling the content of the lyric. Their number, brief characteristics, and
the identification of their source will be proposed. The third
part of each analysis

with cohe
rence relations. In particular, the coherence relations are discussed that were
identified as connecting a newly recognized scenario to an established discourse span.
Comparable to the analytical models drawn, the analyses are based on those textual
of interpretations only that are seen as conveying the (re
formulated) content of
the lyric.

The analytical part starts with texts where the degree of coherence coming with the
lyric is supposed to be the greatest, and finishes with the texts in that the
degree of
coherence with the lyric is, expectedly, the lowest. In between, the group of
interpretations label
ed “Middle coherence” comes. Since

the interpretations discussed
are not very lengthy, every interpretation is inserted before its respective analy
ses, with
its parts that are not of concern for analysis printed in grey font.

of the
analyses are summarized in three graphs and two tables that can be found in




4. 1 I n t e r p r e t a t i on o n e

Ce r t ai n
gr o u p

you just talks about when

there's a whole bunch of people giving you a hard
time making rumors and getting all up in your face and you just want to explode and go
all crazy and suddenly make everything go back to normal

it's pretty simple but a
gain i
can't mess with other people's sad and complicated meanings


In this interpretation, all the frames identified are brought about by the expressions with
general reference, thus pointing to experiential entities that can be reconstructed in
discourse, but that at the same time can be expected to have no concrete real
referent. Although there can be found five linguistic expressions that yield frame
structures in this interpretation, one of them, “you”, is employed twice. Each time this
structure is used, it plays the same (syntactic) role of a general subject and recalls the
same entity,
thus can be considered as

a dialogical form addressing any participant in
the web
site discussion. The frame
your face

is relating to this

for it ca
n be
understood as designating its part. One more frame in the interpretation,
a whole bunch
of people
, refers to an

animate entity, which can be categorized as a collection of

The two remaining frames,


prompt the re
nstruction of
general experiential entities of inanimate nature. The former frame can be seen as
providing access to a smaller number of insertable, more specific entities than the latter
frame. In spite of the fact that no main character has been reconstr

as appearing in
the lyric, the generic
frame fulfils the function of interconnecting the scripts and
scenarios read into there.

With one exception, none of the five frames listed above is considered as directly, or
semantically, relating to the
structures in the song lyric. The exception mentioned is
represented by
a whole bunch of people
frame, which can be seen in the song lyric.
Thus, the conclusion at this point is
that 80 per cent of

frames appointed to retell the



content of the lyric have b
een supplied based
on context, springing from the p


Emphasizing the existence of a state of events, the first
script serves to
introduce a reader into a certain situation, or rather to establish an certain experiential
aming within that the next three frames are connected.
Giving (you) a hard time

is an
action script that brings about the main action perceived as taking place within this state
of events. It recalls the idea of a series of events that are perceived as out
standing, or
defining for the current state of events. This scripts is further adjusted by

getting all up in (your face)

scripts. These two can be seen as
subordinate scripts bringing about two distinct actions that can actually be tre
ated as
causing the existence of the previous, superordinate
giving (you) a hard time

With the next two scripts,
want to

, it is possible to say too, that one is
meant to modify the other. More specifically, the latter script, that evokes a o
ne shot
physical action carried out by an animate entity, develops the former script that
represents a cognitive activity of one and the same animate frame of reference

The last two scripts,
go all crazy

make (everything) go back to normal
, actu
relate to the
want to

script, too. They bring about desired, hypothetical actions that are
springing from this mental activity. Although they evoke the idea, or experiential
model, of partly physical actions, they do not take place anywhere but in the

mind of
the (general) agent.

None of the scripts shows direct resemblance to the structures employed in the lyric
and, therefore, all these can be classified as derived on the pragmatic basis. Though

from a different point of view, the
g all up in (your face)
script can be
seen as,
pointing to the “don't want to hear about it” structure of the interpreted lyric,

hence, can be attributed the s
emantic source. The remaining five scripts

perceived as based on the p
ragmatic source.
Expressed in percentage, 83 per cent of the



structures used to re
tell the
content of the lyric is of the p
ragmatic source, while 17 per
cent of the stru
ctures shows its origin in the s
emantic source.

Coherence relations

Four insta
nces of the type of coherence relations relevant for this study can be isolated

in this interpretation. They all share the PRAGMATIC source for the establishment of
coherence and are POSITIVE in their polarity. Apart from the rest of the relations that
w as CAUSAL, the first relation employed is classified as

ADDITIVE in its basic

value. As it is evident from the respective analytical model
basic operation

was employed only for the case where a scenario that stand
s on a lower
l of discourse

is seen as a
djusting the previous,

level scenario. All the



scenarios that stand on the same level in discourse are linked by the CAUSAL type of
relation. From this observation it follows that, since this interpreter employed a
se marker “
” to connect the scenarios textually, s/he did not use this marker
to signal the same type of discourse relation each time. This can be seen as signifying
that even within the boundaries of one utterance by one individual

one discourse
r need not signal one type of discourse relation only, thus supporting the claim
that a discourse marker alone cannot be considered as a reliable cue for the recovery of
particular coherence relation working. As for the order of elements, a cognitive
tive that can be set the CAUSAL type of relation, all its instances identified show
the BASIC order arrangement.

Clearly, there can be seen a preference for
one type of relation in linking a newly
arriving situation mode with this interpreter. 75 per cent

of this type of discourse
operation is perceived as effected through The CAUSAL
BASIC type of discourse relation effects 75 per cent of the linking. Two values that are
shared by all the relations found are the PRAGMATIC SOURCE OF COHER



4. 2 I n t e r p r e t a t i on t wo

Ce r t ai n
gr o u p

It's about drugs. Pure and simple. Cocaine in particular.

He's ignoring everyone, what
they say, how they try and warn him because he's fallen in love with it. He's addict


ble to the majority of interpretations inquired into, a

frame is employed by
this interpreter to introduce the presence of, and to refer to,

the character that is
perceived as main

the story discoursed upon the song lyric. This
achieves a number of

textual representations that are executed via two
lexical items, "he" and "him". One more frame structure that yields

picture of an
animate entity can be isolated.

This frame

can be recovered after two




both "everyone" and "they" signify the same collection of


n the context of this interpretation,

are considered as recalling one and the same
experiential entity.

The third frame isolated,
what they say
, is

remarkable for its inherent complexity.
Although, out of the context of this interpretation,
the lexicogrammatical structure

employed to textually represent this frame


be further decomposed into
discourse segments, the

it yields within t
he boundaries of the interpretation
a unified, relatively complete mental representation.

this frame is abstract

in nature,
it provides
a flexible mental structure that can take



specific shapes and,
therefore, shelter a number of


The last frame detected is invited by
deictic "it" and expected to refer to the

introduced at the very beginning of this

The total number of entities identified in this interpretation is four. For both


are not recoverable following the semantics of the lyric




from the p
ragmatic source. On the other hand everyone and what they say is
seen as based on the propositions

of the lyric and th
us take advantage of the lyric
. Expressed in percentage, 50 per cent of the frames recognized is based on
the semantic source and 50 per cent is marked by the pragmatic sources of this





number of

scripts used to furnish the reconstruction of the lyric cont
ent in this
interpretation is five
. With one of the scripts discovered,
(he)'s fallen in love with

it is
difficult to distinguish whether the structure serves to introduce a recent event or
state. If understood as a temporary state of events
an observ
ation can be made that
like scripts represented
in this interpretation
the main

character. On the
other hand,
the scripts
recognized as

like structures are pertinent to
the other animate frame
occurring in this interpret

With the state
like scripts

(he)´s ignoring

(he)'s addicted
, it becomes nicely visible
how the mere use of tenses can signal a time interval, or, in this case, the durati
on of a
state in discourse. When
the present progressive


used to

pt a script, the state it

will be considered as temporary, whereas the present simple


signal a


As for the actions identified with

and warn
, they are both seen as
introducting acts

on the main ch
aracter, who, hence,

the role of

affected agent

. Both the sc
ripts are rather general and potentially involve

insertion of more specific
experiential structures so that they
could achieve


outline in a partic
ular discourse of a particular reader.

Since none of the experiential structures discussed above can be directly read into the
song lyric, the conclusion
made at this point is that 100 per cent
of the scripts employed
in this interpretation

the pragmatic source.

Coherence relations

The total number of coherence relations used to attach a new scenario into an