Skip to navigation – Site map

Reading Between the Clauses”: an Exploration of Implicit Inter-clausal Relations in English

Fiona Rossette
p. 219-243

Abstract

Alors que la linguistique s’intéresse en priorité aux marqueurs de surface, ou au signifiant (en particulier en France dans le cadre de l’apprentissage de l’anglais), les ressorts de l’implicite retiennent moins l’attention. Cela est notamment le cas dans l’étude de l’enchaînement des propositions, où l’on traite principalement de la subordination, de la coordination, et des connecteurs adverbiaux (ex. however, therefore). Je propose ici d’identifier les relations qui ne sont pas ainsi marquées entre les propositions. Ces relations «implicites» ne posent aucun problème au lecteur, qui a l’habitude de lire entre les lignes ou, plutôt, entre les propositions. Je décrirai ici le modèle d’interprétation dit «interactif» élaboré par Hoey (2001) et qui met en avant le travail d’anticipation effectué par le co-énonciateur. Ensuite, je présenterai une étude de cas, basée sur la description des relations implicites dans deux textes argumentatifs. Dans ce genre, les relations de général-spécifique et de justification, ainsi que les schémas parallèles, sont prépondérants. Il en ressort également que le discours résiste très souvent à une analyse linéaire et segmentale. Toute tentative d’établir une liste exhaustive des relations, ou de coller à tout prix une étiquette à chaque enchaînement, s’avère artificielle et même contre-productive.

Top of page

Full text

  • 1 My translation.

La langue est le domaine des articulations, et le sens est avant tout découpage. Il s’ensuit que la tâche future de la sémiologie est beaucoup moins d’établir des lexiques d’objets que de retrouver les articulations que les hommes font subir au réel (…) (Language is made up of articulations, and meaning is primarily about division. Therefore, the task that lies ahead for semiology is not to draw up lexical lists but to determine the articulations which reality is subjected to.)1
R. Barthes, L’Aventure sémiologique, p. 53

1. Challenges regarding implicit relations

1.1. Interpretation as anticipation

1While description tends to concentrate on explicit rather than implicit clause relations, it can be argued that the role played by the latter is more essential to understanding the workings of discourse organisation. Links which are considered here to be explicit are those marked by some type of “connective”—a banner term grouping together subordinators, coordinators and adverbials, and a term which, it is worth pointing out, tends to be more widely used in the French meta-language compared to that of English. This does not mean that connectives do in fact always make the inter-clausal relation more explicit. Moreover, other means, which will be discussed below, exist for doing so. However, this working hypothesis reflects a generally accepted idea, if not within linguistic theory, at least within certain teaching practice. Indeed, while students, in my experience, are often encouraged to use connectives to make their writing clearer, this does not necessarily make for a more effective text, a point underlined by Hoey and Winter (1986: 127): “[I]t does not follow, of course, that if writers simply ‘flagged’ each sentence, all would be well; writers must consider the effects of inference in its own right”.

  • 2 These figures account for finite, non-embedded clauses, and therefore exclude relatives and subordi (...)

2In terms of their frequency, implicit relations by far exceed explicit links marked by a connective. In a corpus of written and oral English (editorials, essays, political speeches, televised debates), connectives are used in one third of interclausal relations, and the percentage is closer to one quarter if only written discourse is taken into account (Rossette 2003). Even in French, a language which often relies more on connectives than English (see Poncharal 2007), only one fifth of clauses in a corpus of press articles contains an explicit link of this type (Rossette to appear, a).2

  • 3 Rather than refer to a speaker and an addressee, I will adopt the terms writer and reader, as only (...)

3The question remains as to how readers deal with implicit relations.3 The interpretation task assumed by the reader has often been underlined, as well as the process by which a particular sequence paves the way for, or conditions, the discourse to follow. In this latter case, Anscombre and Ducrot (1997: 96) speak of “virtualités discursives”. The model described by Hoey and Winter (1986) presents text as an interaction between writer and speaker—as certified by the title of Hoey’s recent work Textual Interaction (2001). The authors liken written text to a virtual dialogue; they insist on the fact that the interpretation of a relation is just as much a question of anticipation as it is one of retrospection:

Some inferences are undoubtedly retrospective; faced with two sentences, the reader may consciously seek to find the connecting proposition that makes sense of their juxtaposition. But reading would be slow if this happened all the time. Indeed, if it did, one might question whether the reader properly understood what he or she was reading. For it seems that readers normally are able to anticipate the relations that are to come […] (Hoey & Winter 1986: 126) (my emphasis)

4According to such a model, the reader is constantly anticipating on a more or less conscious level where the text is heading. Such anticipation can be spelt out in the form of questions:

We might hypothesize that given previous information, readers have a number of weighted questions that they think might be answered in the next sentence(s). Each sentence is scanned as a possible answer to these questions and the question it most nearly fits is seen as the one it is answering […] (Hoey & Winter 1986: 126)

5The question technique can be applied to the introductory sentence of the essay which will be analysed here:

Geoffrey Blainey’s great phrase ‘the tyranny of distance’ when it was formulated nearly forty years ago, offered a powerful explanation of the problems of being Australian and of Australia’s relationship to the world.

6This sentence sets in motion a number of expectations for the reader, which can be formulated via the following questions: why did the phrase offer a “powerful explanation? What are the problems of being Australian? What are the problems of Australia’s relationship to the world? For the discourse to be viewed as coherent, the sentence which follows this first one needs to provide an answer to one of these questions. This is indeed the case, as the second sentence of the text evokes the role played by geography, which, it can be assumed, lies at the root of the “problems” referred to in the first sentence:

It pointed to geography, seen in terms of position and distance, as a determinant of what we call history, that is, of our daily lives as they are lived through events and conditions.

7Within the categories that will be highlighted here, these first two sentences realise a preview-detail relation, which is paralleled in the lexus, with the noun problems in the first sentence closely associated with geography in the second.

8A lot more can therefore be gathered from one single clause or sentence than one might originally think: “The point here is that the reader does half the work for the writer. The writer’s words activate knowledge in the mind of the reader which the reader brings into play in his or her interpretation of the text” (Hoey 2001: 120). To explain the acceptability of lexical combinations, Hoey (2001: 2002) develops a theory of “priming” which can also be applied here to underline the role of the preceding context in preparing the reader for what is to follow. Hoey and Winter (1986:126) distinguish between “default expectations” and “active interpretations”, depending on whether active effort is required on the part of the reader.

  • 4 Van Dijk uses such a principle to explain the inacceptability of a sequence such as We went to an e (...)
  • 5 Rumelhart, D.E. & A. Ortony (1977) ‘The representation of knowledge in memory’, in R.C. Anderson et (...)

9The authors note moreover that in the “scanning” process “we are bringing to bear our past experience of similar juxtapositions” (Hoey & Winter 1986: 122). While van Dijk (1985: 111) speaks, for instance, of the need for the reader to be able to identify stereotypical situations which conform to extralinguistic phenomena, noting that “the meaningfulness of discourse also depends on what we assume to be the normalcy of the facts, episode or situation described”,4 the reader also evaluates the sequence in terms of norms governing the order in which they appear in a specific discursive context. For example, Rumelhart and Ortony (1977), discussed in Hoey (2001),5 define prototypical scenarios, or “scripts” which make up narrative. For them, knowledge of the world is carefully organised and when one part is activated, the rest is brought to mind. It can be argued that the way knowledge is organised is dependent not only on culture but also on genre. Indeed, the reader activates his/her knowledge of a specific genre: expectations will not be the same in a narrative as they are, for instance, in argumentation. Before analysing specific examples of argumentation, we first need to ascertain what type of relations are to be identified.

1.2. A typology of implicit relations?

10Establishing a typology of implicit relations requires a degree of conceptualisation and abstraction which is not necessarily compatible with the principles of segmentation or classification. An onomasiologic approach to clause relations still relies on the availability of words in the metalanguage to provide a label for the relations that are identified. Another problem arises out of the fact that implicit and explicit relations are generally presented in opposition, but I hope to demonstrate that they in fact prove complementary. I would argue that in each case, the analyst should not be looking for the same relations, or even the same types of relations. “Flagging” each clause with a connective is not feasible not only because their repeated use would overload the discourse and would prove more of a hindrance than an asset to comprehension, but also because connectives do not exist for all the types of sequences which structure discourse. That is, there exist implicit relations for which there is no equivalent connective, a problem already highlighted by Thompson and Zhou (2000: 137) in their study of relations linking sequences in which disjuncts are used: “the difficulty in pinning down the conjunctive function of the disjuncts, may arise because they are signalling a type of conjunction for which an alternative encoding by means of conjuncts is not available”. What types of relations therefore typically lie “between the clauses”? Martin (1992: 165) notes that “simply putting clauses next to each other suggests some logical connection between them” (my emphasis): the term logical points to processes of reasoning, such as causal relations, but does such a definition include, for example, time sequence relations?

11Few typologies have been made of clause relations, irrespective of their mode of realisation. Martin (1992) is, to my knowledge, one of the first to discuss both relations signalled by a connective and those that are signalled by other means, including inferences which can be made from the lexus (or the “experiential—i.e. extralinguistic—meaning of the clauses), in a chapter entitled “Conjunction and Continuity”. Among those who account for connectives, Halliday and Hasan (1976), Quirk et al (1985) and Biber et al (1999) all discuss additive, adversative, causal and temporal relations, even though these are sometimes organised differently: for example, Biber et al identify what they call “appositional linking”, which includes exemplification, reformulation or explication, all of which Halliday and Hasan consider as variants of the basic additive relation. These sequences belong not so much to an extralinguistic as to an intralinguistic or rhetorical level; that is, they are internal rather than external.

  • 6 While causal relations are often cited to illustrate the external/internal dichotomy, it can be arg (...)
  • 7 This is, of course, a largely simplified picture, and I do not dismiss the ambivalence of certain s (...)
  • 8 An extreme example of this is provided by the types of progressions identified between turns in con (...)

12The distinction between internal and external relations is well documented in the literature. External relations are “inherent in the phenomena that language is used to talk about”, whereas internal relations are “inherent in the communication process, in the forms of interaction between speaker and hearer” (Halliday and Hasan 1976: 241), or, as Martin (1992: 180) puts it, “[i]nternal relations […] structure semiosis; external ones code the structure of the world”. The distinction is not necessarily clear-cut, as Halliday and Hasan admit—for example, Verstraete (1998) provides an alternative classification—and accommodation needs to be made for a scale of relations, with some clearly more internal than others.6 It is not my aim here to review the different classifications which have been drawn up regarding explicit clause relations. However, it is worth pointing out the parallel that can be established between the external-internal spectrum and the syntactic level at which relations are realised. Matthiessen and Thompson (1988: 286) refer to “scalesensitive relations”; in later work, Matthiessen (2002) speaks of a “division of realizational labour” between external relations, which are typically local, for example in the form of subordination, and internal relations, which are situated at a more global level of the text. Martin (1992: 207) makes a similar observation, as do Thompson and Zhou (2000). A parallel can be drawn with van Dijk’s “pragmatic” uses of connectives, which “are often signalled by sentence-initial position in independent new sentences, whereas the semantic use of the connectives may also be interclausal” (van Dijk, 1979: 112), and with the opposition drawn by Deléchelle (1989: 630) between semantic and discursive relations; he describes the latter as “méta-énoncés” as they relate to the “internal workings” of the discourse. Deléchelle (1991: 127) also explains in this way why a different list of categories needs to be drawn up for connectives which appear in sentence-initial position as opposed to those that are situated within the sentence. Indeed, a glance at a list of subordinating conjunctions, which appear predominantly within the sentence, indicates the range of relations which can be realised at this level: those that introduce adverbial clauses can encompass relations which are more clearly external, such as temporal relations (e.g. when, before, after, since, as). In comparison, adverbial connectives, which mainly mark relations between sentences, realise relations which are internal rather than external (e.g. therefore, however, in addition) (Rossette 2003).7 If we apply the division of labour principle, it can be posited that implicit relations, which are more predominant between sentences (for example, two-thirds of relations devoid of a connective in my 2003 corpus of argumentation concern relations between sentences, compared to one third within the sentence), can be explained with reference to values situated at the internal, or rhetorical, end of the spectrum.8

1.3. Hoey’s categories

13Hoey (2001) adopts the distinction first made by Winter between “sequence” and “matching” relations. Sequence relations involve “putting propositions in some order of priority in time, space or logic” (Hoey 2001: 30), while statements linked by a matching relation “are brought together with a view to seeing what light they shed on each other” (Hoey 2001: 31). Sequence relations include time sequence, cause-consequence, means-purpose and premise-deduction; the main examples of matching relations are similarity, exemplification, preview-detail, and exception.

14Interestingly, this distinction is also based on the mode of realisation: connectives play a major role in the signalling of sequence relations, whereas parallelism (in the syntax) and repetition (in the lexus) come into play in the case of matching relations. Problems arise, however, when the two criteria—order and mode of signalling—do not necessarily correlate. For example, preview-detail and exemplification are considered matching relations due to the fact that they are generally signalled by repetition, but I would argue that they both rely on a specific ordering.

15The importance given to both parallelism and repetition as signalling devices prompts the question as to whether they provide sufficient reason to talk about explicit relations. The same question arises with other modes of signalling. The essay which will be analysed here contains a considerable number of examples of relations signalled either by adjuncts (optional adverbial elements) or by the process itself. Indeed, Martin (1992) draws attention to the fact that rhetorical relations can appear either between processes or within a process, particularly those which are attributive or identifying. In the extracts below, the adjunct in those terms and the finite verb mean both announce a deduction from a premise, just as a reason or justification is announced via the verb be based on.

We live in feelings as well as in conditions and events. Distance is also measured by the heart. In those terms, Europe, Britain, were close, not far off. (essav, 6-8)

The space they [Longreach and Sydney] exist in now belongs to mind, imagination and to the skills they demand to make them work, and this has meant a redefinition, a radical one, of what we do and who we are. (essay, 13)

There was only one place in Asia that we felt close to: Timor. (30) That closeness was based on events that went back half a century, but were still alive in our consciousness as real experience […] (essay, 30)

16In the essay, adjuncts and the finite verb assist in drawing attention to the nature of the relation in 11 cases (approximately 15% of the relations in the text). The newspaper article does not resort to signalling of this type. In all these examples, it can be argued that the reader is not required to “read between the clauses”, as the relation is in fact signalled in the language itself. However, it can be posited that each mode of signalling represents a different degree of explicitness. Of course, connectives should not be taken at “face value”; their use cannot be reduced to a pure signalling function (they are also cohesive, and can realise interpersonal meanings, which have come to be known in France as “meta-modal” values). They therefore require a certain amount of decoding, hence the considerable number of studies which focus on the meanings/effects of specific connectives (and these studies are not necessarily passed on to learners, who are given lists of “ready-touse” connectives). However, it can be argued that because they feature early in the clause, if not at the very beginning, connectives announce from the outset, or at least draw attention to, an inter-clausal relation. Adjuncts represent another degree: they do not necessarily stand out as much as connectives; they can be integrated into the clause and/or the sequence (e.g. through anaphora). Finally, relations signalled by the finite verb appear even more “buried” in the clause, to perhaps the same extent as parallelism and repetition.

17On top of sequence and matching relations, Hoey (2001: 122) identifies “culturally popular patterns of organisation”, which, akin to Rumelhart and Ortony’s “scripts”, assess a sequence of text more globally. Of note here is Hoey’s choice of the term pattern. The discussion here is centred on relations, which is, in essence, a neutral term (the term “connection” appears in most dictionary definitions). It can, however, be claimed that relation has come to be so closely associated with marked links that it would almost be preferable to use another term for those that are implicit. Moreover, a connective tends to project the two clauses as a unit, even isolating them somewhat from the surrounding context. If their meaning is not always explicit, connectives do work to “single out” couplets of clauses. Without such a cohesive agent, the discourse is more free-flowing, which is perhaps better expressed in terms such as pattern or progression. Having said all this, I will carry on using the term relation, for which pattern and progression will prove useful synonyms.

18The main organisation patterns identified by Hoey are: problem-solution, goal-achievement, question-answer, gap in knowledge, and claim-response. Hoey concentrates on narrative texts, although he notes that the last two patterns are common in academic writing. I have adapted his categories to describe relations in argumentation.

2. Examples of implicit relations in expository/persuasive prose

19The two texts examined here, “The People’s Judgment”, by the Australian writer David Malouf, published in a collection of essays, and “At last the focus is on talent”, a newspaper article presenting an opinion, published in The Guardian Weekly, share similar aims: to inform, to analyse, and to persuade. Initial observations can be made concerning the macro-structure of each text (see appendix for full texts). The article follows a question-answer format. Despite the fact that it relays an opinion, it conforms to a top-down construction, based on the model of the inverted pyramid generally adopted in the press. It announces from the outset the point being debated (are US voters ready for a knowledge-based presidency?— sentence 3). On the other hand, the structure of Malouf s essay may be described as bottom-up in that it is not until the end of the text (the second-last paragraph) that the core, topical issue is introduced (the 1999 Australian republic referendum); all the points mentioned beforehand (the “tyranny of distance”; the changing workplace; the link between Australia and Britain) build up to the writer’s opinion on the issue of the referendum. The text’s overall organisation can be likened to a claim-response pattern: the writer discusses the reasons generally put forward to explain why Australians voted no to the republic, only to challenge them and to suggest an alternative explanation. This pattern can also be identified on a local level, as the text presents a succession of different points which are introduced and then debated. Generally, the validity of each point is denied, which then sets in motion another claim-response as an alternate claim is then examined.

20In both texts, the number of explicit relations making use of a connective is inferior to that of implicit relations. The essay is made up of 49 sentences, containing 61 finite, non-embedded clauses, and boasts 10 explicit relations: 6 within the sentence (because, when, 2 instances of and, 2 instances of but) and 4 between sentences (but, too, 2 instances of and). The article is made up of 40 sentences, containing 46 finite, non-embedded clauses, of which 12 contain a connective: 6 connectives appear within the sentence (because, and, 2 instances of but and 2 instances of as), and 6 between sentences (also, for example, thus, and 3 instances of but). Unsurprisingly, in line with previous studies (e.g. Biber et al 1999), the coordinating conjunctions outnumber the other types of connectives. Moreover, among the latter, no specific connective/relation is favoured more than another.

21In what follows, I provide an exhaustive description of the introductory passages of each text; references to clause relations situated elsewhere are made when appropriate.

2.1. The first two paragraphs of the essay

(1) Geoffrey Blainey’s great phrase ‘the tyranny of distance’, when it was formulated nearly forty years ago, offered a powerful explanation of the problems of being Australian and of Australia’s relationship to the world.
(2) It pointed to geography, seen in terms of position and distance, as a determinant of what we call history, that is, of our daily lives as they are lived through events and conditions.
(3) The Australia Blainey was placing was nineteenth-century Australia, six weeks’ sailing distance from Europe, in the age before international cables had made possible the wonder of instant communication and, of course, by the time he formulated it the conditions it described had already changed.
(4) Air transport had reduced travel time to a single day; satellite images were about to make every event on the globe instantaneously visible.
(5) And it was never quite true, even before technology changed forever our notions of distance and the globe.
(6) We live infeelings as well as in conditions and events.
(7) Distance is also measured by the heart.
(8) In those terms, Europe, Britain, were close, notfar off.
(9) As for now and the century to come, the new communications systems mean that mere geography will never again determine our sense of where we stand.

  • 9 One definition of example provided by Merriam-Webster indicates “one (as an item or incident) that (...)

22(1) Preview-detail: As already noted in section 1.1 above, sentences 1 and 2 are linked by a preview-detail relation (cf. problems > geography/position/distance), that can also be identified between sentences 2 and 3, or at least the first clause of sentence 3, which picks up on the notions position and distance and specifies exactly how far Australia was from Europe in the nineteenth century in terms of travel. I consider preview-detail to realise a “zooming in” effect, with the initial clause framing, or providing a transition towards, the second clause. No connective exists to express such a progression. I make a distinction between this relation and exemplification, for which connectives exist (for example, for instance), and which is a type of preview-detail relation, with the specificity that it results from a choice between one of several possibilities.9 Exemplification can be identified between sentences 16 and 17 of the newspaper article:

(16) Encouragingly there are prospective political candidates in both parties who, whatever their individual flaws, have used their careers to acquire a body of knowledge about governance.
(17) As a matter of courtesy, let’s consider Gore first.
(18) Of all the American politicians I met during a long journalistic career, I believe that Gore and Richard Nixon knew most about all aspects of government and politics,
(article)

23Gore appears as one of several “prospective political candidates.” In contrast, the link between sentences 17 and 18 is a more general preview-detail, with sentence 17 introducing Gore and sentence 18 offering information about him. Sentence 17 could in fact be removed, but the content has been “stretched out” over two sentences according to key practice in discourse organisation: instead of getting to the point immediately, the text progresses in stages, introducing new information little by little—just as there is generally a progression from given to new information within the clause itself. Rather than be “thrown in at the deep end,” the reader needs to be “primed”, as Hoey would put it. Similarly, modern communication strategies have coined the adage “tell them what you’re going to tell them, then tell them, then tell them you’ve told them.” This is why I would argue that the preview-detail relation highlights one of the most powerful principles of text organisation. Further examples to be presented below demonstrate its prevalence in newspaper articles, in keeping with top-down organisation synonymous with the inverted pyramid model.

24(2) Claim-denial (explicit relation): The second clause in sentence 3 (and, of course, by the time he formulated it…) realises a change of tack, and introduces a denial, or at least a partial denial (= the geography/distance hypothesis would not prove valid for long), restricting the content of sentence 2, which, retrospectively, assumes the role of a claim. This restriction is introduced via the coordinator and in combination with the disjunct of course and I have therefore catalogued the relation as explicit. It can certainly be argued that the additive coordinator does not make the denial explicit, compared to the use of but or another adversative connective. It is worth noting, however, that the denial is not with respect to the content of the clause immediately to the left (i.e. the first clause of sentence 3), but with a clause situated higher up in the text. Elsewhere (Rossette, to appear, b), I have argued that the potential to establish non-linear relations is specific to coordinators, whereas subordinators and most adverbs mark relations between two consecutive clauses. The fact that this denial does not feature in a separate sentence is also significant: if this were the case, more weight would be given to the objection, which would appear more newsworthy, or unexpected, producing an effect that would not necessarily be consistent with the meaning expressed by of course.

25A further denial, again with respect to the content of sentence 2, and again introduced by and, is expressed in sentence 5 (And it was never quite true…). This time, the coordinator appears in sentence-initial position, which confers more argumentative weight, in keeping with the fact that the denial is not partial but now complete (cf. never).

26(3) Premise–reason: Reasons are provided for each of the denials identified above. It is in this type of relation that reader expectation is particularly born out: any sort of statement calls for justification, and for a persuasive text to be convincing, it needs to answer the question constantly at the back of the reader’s mind, and that is why? or on what grounds can such a statement be made? This point is underlined by Verstraete (1999: 7-8): “The need for argumentative support is particularly compelling in contexts of debate, where the speaker puts forward a claim in contrast with an alternative claim from the previous discourse.” A movement of justification appears between the second clause of sentence 3 and the sentence which immediately follows, which explains how the conditions “had already changed”. Similarly, sentence 6 provides a reason for the refutation expressed in sentence 5, suggesting that feelings are more important than physical distance. One could object that, in both instances, a preview-detail pattern is also validated, based on parallels that can be established within the lexus (conditions / air transport; satellite images, distance / feelings). The line between concepts such as justify, explain, specify, reformulate is a fine one. Indeed, preview-detail often combines with premise-reason, a point that will be developed below. For these examples, I would argue that the denial creates tension which calls forth a justification and that the premise-reason is therefore the most likely relation to be recognised.

  • 10 Importantly, the adverb also does not elucidate the relation with the previous clause. Its scope re (...)

27(4) Implication (+ re-elaboration): Similar overlapping involving a causal relation occurs between sentences 6 and 7 (We live in feelings as well as in conditions and events. Distance is also measured by the heart.) I first identified premise-deduction, with a paraphrase possible using therefore or that is why (e.g. We live in feelings as well as in conditions and events. That is why distance is also measured by the heart). However, the opposite movement, in the form of premise-reason, which would allow a paraphrase using because, cannot be ruled out altogether (e.g. We live in feelings as well as in conditions and events because distance is also measured by the heart). Instead of persisting in the attempt to identify a specific type of causal relation working in a specific direction, I would suggest that it is sufficient to recognise a general relation of implication—with the meaning of one clause implied by that of the other.10

28It may also be appropriate in certain cases to talk about “reformulation”, a relation described and sub-categorised by Martin (1992), or “re-elaboration”, a notion developed by Cotte (e.g. 1992) to describe various phenomena, including the use of the auxiliary in English in certain verbal forms: in the theoretical genesis of the utterance, a primitive, discrete form is re-elaborated into a two-part analytical form in order to produce new, expressive nuances. Both sentences 6 and 7 deal with abstractions rather than distinct, concrete, extralinguistic events, and it can be argued that there is little difference between the notions heart and feelings. In fact, the second sentence could almost be considered redundant, as it simply provides a reformulation of the first. However, it would be hasty to claim that repetition on the surface of the discourse reproduces the exact same meaning. Even when repeated word for word, a meaning is modified, if simply by reinforcement. I would argue that the preponderance of repetition in discourse is often underestimated, as it goes against the grain of segmental, constituent analysis. Discourse is constantly repeating itself, and meanings are forever being re-elaborated, a characteristic of construction proving equally as important as the stage-by-stage progression engendering preview-detail.

  • 11 We can go further and postulate that for this natural order to be upturned, and deduction preferred (...)

29(5) Premise-deduction: A less ambivalent premise-deduction can be identified between sentences 6/7 and sentence 8. As already noted in section 1.3, the deduction may be considered explicit to a certain degree, as it can be traced to the adjunct “in those terms.” Premise-deduction is in fact rare in the two sample texts. In the essay, I have only identified 3 unequivocal examples. The other two are situated between the two clauses of sentence 12, linked via a connective, even if it is simply the additive coordinator, which could well combine here with so or therefore (This is the new form of geography we live with and [so] if it collapses the distance between hemispheres it also collapses the distance between, say, Longreach and Sydney) and between sentences 45 and 46 (An Australian republic can only be argued for convincingly at the level of feeling on what we feel towards the place and for one another. [On these grounds,] when it comes at last, some time in the next century, it had better be a true republic, one that is founded not on the loyalty of its citizens to their head of state, but on their loyalty to one another: on bonds, which already exist and which we already recognise, of reciprocal concern and care and affection.) The extremely low number of premise-deduction contrasts with the preponderance of premise-reason. Halliday and Hasan (1976: 257-258) had already noted that marked justification is more common than marked consequence. I would suggest that, whatever the mode of realisation, premise-reason reflects a more innate organisation pattern; that is, it is more natural to make a statement and follow it up with a justification, than to carry through a deduction process in keeping with the tradition of the syllogism. Put more simply, writers, at least in expository/persuasive prose, spend their time justifying themselves.11 This challenges the view that linear relations in the form of time sequence or cause-consequence are characteristic of asyndetic clauses—a point underlined for example by Quirk et al (1985: 1472) concerning clauses in which only the preterit or the present appear: “Asyndetic relation of this kind […] raises the expectation that the second utterance follow[s] the first as an iconic representation of being sequential in time or consequential in reasoning.” It is certainly necessary to distinguish between clause links within the sentence (which I believe are more conspicuous, and for which the term asyndeton is used) and links between sentences; moreover, we can expect to encounter different progressions depending on the genre, with the claim made by Quirk et al proving more pertinent in the case of narrative text.

30(6) Matching: I have adopted Hoey and Winter’s notion of matching relations in a more restricted sense, to refer to a progression between clauses which fulfil the same rhetorical purpose in a text. A matching relation can therefore be identified between the two clauses in sentence 4 (Air transport had reduced travel time to a single day; satellite images were about to make every event on the globe instantaneously visible), which both provide justification for the denial expressed just before. On a more macro level, a matching relation can be established between the two sentences expressing denials: sentence 5 appears fully integrated into the passage as it enters into relations with two distinct clauses: matching with respect to the second clause in sentence 3—which would justify the use of the additive coordinator—and claim-denial with respect to sentence 2.

31Just as exemplification is a specific type of preview-detail pattern, matching relations include the sub-category of contrast. In the newspaper article, two examples of former presidents are given one after the other (matching). At the same time, the aim is clearly to draw a distinction between the two presidents: Reagan’s beliefs contrasts with Clinton’s knowledge, and there is an opposition between luckily in the first sentence and bad luck in the second:

(11) Reagan did value beliefs over facts, but luckily his strategic beliefs meshed almost perfectly with a moment of opportunity to remake cold war geopolitics.
(12) Bill Clinton’s noble effort to reform healthcare was knowledge-based, but its political collapse gave learning a bad name with half of the electorate, just as the country was seized with one of the religious manias that sweep America every century.

32Similarly, the last two sentences of the article present a contrast between war and knowledge:

(39) We’ve given war a chance.
(40) Now, when it seems possible that one or both parties could allow it, let’s give knowledge a chance.

33(7) Chronology: A progression on the temporal axis can be identified between sentences 8 and 9 (In those terms, Europe, Britain, were close, not far off. As for now and the century to come, the new communications systems mean that mere geography will never again determine our sense of where we stand), with use of the preterit tense in sentence 8 (were close, not far off) appearing in contrast with the adverbial in sentence 9 (As for now and the century to come). This is the only example of progression on the temporal axis to appear in either of the texts, and no doubt we would find more in narration than in argumentation. Sentence 9 also serves as a transition towards the theme of the following paragraph, hence preview-detail, or preview-example (new communications systems > the net). And this pattern overlaps with premise-reason: in sentence 10 the issue of language is addressed, which provides a reason why geography will never again be influential.

2.2. The first four paragraphs of the newspaper article

34I do not have the space to present as detailed an analysis of the article, and will concentrate on the patterns it contains which are specific to the press. The article contains a less complex structure in that paragraphs are short and many are organised around a preview-detail pattern, with the top-down organisation located in the macro-structure echoed on a local level, governing the progression from clause to clause. Another example of where macro and local structure meet is in the use of question-answer, which can be identified between consecutive clauses. The second sentence in each of the following extracts provides at least a partial answer to the question which immediately precedes it:

What are the signs that America’s disenchantment with knowledge and elected officials who possess it may be ending? Encouragingly there are prospective political candidates in both parties who, whatever their individual flaws, have used their careers to acquire a body of knowledge about governance, (article, 15-16)

Will he [Gore] run in 2008 if there’s an opening? Remember what his father, the late senator Albert Gore, said when Gore went on the Democratic ticket in 2002: “He was raised for it.” (article, 21-22)

35I will now describe the relations present in the first four paragraphs of the article, reproduced below:

(1) Al Gore’s recent appearance on Capitol Hill marked his dominance in Washington’s debate about the science of global warming. (2) It also raised a supremely important question about the state of political science in the US. (3) Given the pain inflicted on the world by George Bush’s faith-based presidency, are US voters readyfor a knowledge-based presidency?
(4) It is a question worth exploring in advance of the 2008 election, because Gore and several other prospective candidates represent a reappearance on the US campaign scene ofan endangered breed - elected officials who have used their time in office to become expert in government.

(5) Gore stands in a long line of congressional figures who came to Washington intent on mastering one or more essential issues. (6) His range was unusually broad — the environment, energy, communications technology and nuclear control.
(7) As the Bush presidency winds down, one can feel the electorate holding its breath, as if to reassure itself it can will the centre into holding. (8) In the press, the opposition and his own party, Bush’s critics are restraining themselves in the belief that admitting how dangerous this man really is will diminish the chances of safely running down the clock until Inauguration Day 2009. (article)

36Applying Hoey and Winter’s technique, the first sentence triggers questions such as why did Gore’s appearance mark his dominance? or What did Gore have to say? But in fact the second sentence introduces another consequence that can be drawn from Gore’s appearance on Capitol Hill. If we had to formulate a question to elucidate the progression between the sentences it would be What else can be drawn from his appearance/his active role in the debate? However, I am not convinced that this is a question that naturally springs to mind. Moreover, the two sentences enter into a matching relation, which is quite rare in the early stage of a text, where preview-detail dominates. I would therefore argue that this relation does not meet the expectations of the reader, and that an “active interpretation” in Hoey and Winter’s framework is needed. Moreover, I would posit that this is why the connective “also” has been used. In fact, the first sentence presents a known, accepted fact which the reader recognises (= Gore knows a lot about global warming) and which serves to introduce the new idea that the writer would like to get across.

37The additive relation, as it were, “delays” the preview-detail pattern, which appears immediately after, with “a supremely important question” in sentence 2 announcing the interrogative in sentence 3. Preview-detail also links the second clause of sentence 4 with sentence 5, with expert foregrounding the notion of mastering one or more essential issues. In turn, essential issues provides a prelude to the identification of specific examples relating to Gore in sentence 6 (the environment, energy, communications technology and nuclear control).

38A minor relation which has not yet been illustrated is statement-reaction, linking the question formulated in sentence 3 with an evaluation by the writer in sentence 4 (a question worth exploring). This leads onto premise-reason within the same sentence, this time made explicit by the connective because.

39Sentence 7 marks a change of tack. After the example of Gore developed in the first three paragraphs, the focus is now on counter-examples, beginning with Bush. In the absence of any of the relations already discussed, I observe here a change of topic, both facilitated and emphasised by the paragraph break. (Similar changes in topic can be observed with respect to sentences 9, 15, 23, 31, 34 and 37 in the article, and sentences 15, 34, 41, 44 and 49 in the essay). A parallel can be drawn with the changes of topic at paragraph breaks in news stories, which conform more rigidly to the inverted pyramid model. In this case, White (1998: 202) talks about macro-relations between the lead and each paragraph of the article.

40Finally, sentences 7 and 8 offer another example of a matching relation: both provide proof of certain apprehensions concerning the Bush presidency, with the notion of holding [one’s] breath, which applies to the general public in sentence 7, finding a parallel in restraining themselves, applying to officials and the press in 8.

3. When the discourse withstands linear analysis

  • 12 This is, for example, a feature of asyndeton in Marguerite Duras, in which a clause is repeated wit (...)

41Much is written about the linear construction of language and the question is often raised as to how meanings or thought patterns, which are dynamic and multi-dimensional, are “tamed” or adapted to fit such a format. The dynamic nature of discourse is recognised for example with regards to the interpretation process, which involves anticipation and retrospection and is paralleled by certain linguistic forms (e.g. anaphora, cataphora). However, I would suggest that it is not sufficiently accounted for in the organisation of the content itself. At the syntactic level, constituent analysis promotes the notions of linearity and segmentation, but it would be wrong to transfer this model to the rhetorical organisation of text. Discourse is not always linear, and clauses do not represent discrete entities of meaning. Extreme examples of circular or loop constructions are characteristic of spontaneous, oral discourse (Rossette 2003), and also that of children (e.g. Morgenstein & Sekali 2004). Moreover, repetition in the form of what I have labelled re-elaboration can produce interesting effects in literature.12

42In the crafted, revised writing under study here, two types of phenomena bear out the dynamic nature of discourse, which therefore withstands a linear, segmental analysis. It has already been noted that what “holds a text together” is not necessarily relations between consecutive clauses but between one clause and another situated higher up in the text. Martin (1992: 166) observes that “[i]t is also important to note that clause complex analysis is somewhat biased towards local relations between pairs of adjacent clauses and that certain texts, especially abstract written ones, may have a global rhetorical organisation requiring a different analytical perspective.” Studies of subordinators and adverbial connectives have trained us to look for relations between adjoining clauses, however clause links devoid of a connective, as well as those that are articulated via a coordinator, allow for relations between non-adjoining clauses. Secondly, clauses can enter into several relations simultaneously, either with different clauses, or with the same clause in the case of overlapping relations. All phenomena are illustrated in the following paragraph taken from the article:

(23) What is striking about this presidential cycle is that for the first time in a long time, a number of plausible candidates in both parties seem, at a minimum, informed enough for the top job.
(24) Hillary Clinton fits that description.
(25) She is a meticulous student of government who has used the Senate as a graduate school. (26) One sees a similar seriousness of purpose in John Edwards and Barack Obama.

43The first sentence (23) enters into a preview-detail relation via exemplification with both sentences 24 and 26, which each provide examples of plausible candidates·. (1) Hillary Clinton; (2) John Edwards and Barack Obama. A matching relation can also be identified between these two sentences, despite the fact that they are non-adjoining. In between, sentences 24 and 25 are linked via another preview-detail pattern, with sentence 25 providing information on Hillary Clinton. Moreover, it can be argued that preview-detail combines here with premise-reason, in that sentence 25 provides justification for the statement made in 24 (Clinton fits that description).

44The fact that clauses and even sentences do not always represent discrete entities of meaning is illustrated by the culminating rhetorical effect that can sometimes be observed between adjacent clauses. This is the case for sentences 35 and 36 below. The paragraph opens with a concession signalled by but.

(34) The vagaries of American politics and the unpredicted pressures of the presidency chasten one’s optimism.
(35) But it is worth remembering that our most naturally gifted president, Abraham Lincoln, emerged after the horrid term of James Buchanan set the stage for the civil war.
(36) I am reliably informed that no less a presidential scholar than the late Arthur Schlesinger had tipped Bush as a challenger for Buchanan in the running for worst president ever.
(article)

45The line of reasoning could stop at sentence 35, however the content of 36, which places Bush on a par with Buchanan, adds considerable argumentative weight. The content of both sentences are to be taken together. At the same time, there is a progression between the two, and so I would not talk about a matching relation here (the two are not equivalent, and sentence 35 cannot be suppressed). The culminating effect is similar to that marked by the connective moreover.

  • 13 As exemplified in the following sentence analysed by Martin (1992: 265): Government and police sour (...)

46The main problem faced in the identification of clausal relations pertains to those that overlap within the two same clauses. As in the progression between sentences 24 and 25 of the article just mentioned, preview-detail + premise-reason is the most common combination. It can be posited that overlapping relations are more frequent in written discourse, which features a higher rate of lexical density compared to conversation (Halliday 1985), as well as cases of “grammatical metaphor”, in which processes are packed into non-finite elements.13 Moreover, exemplification can be “packed” into a nominal group in subject position, which is the case in sentence 25 (Hillary Clinton fits that description), while justification is localised in the predicate. The rare cases of premise-reason to be encountered in the newspaper article all appear in combination with preview-detail. Here is another example, situated just after the introductory paragraphs examined above. Sentence 9 introduces Ronald Reagan, who becomes the topic of 10. At the same time, an explanation/justification is provided to understand why Reagan heralds Bush: both can be described as “conviction politicians”:

(9) American politics has been building up to Bush since Ronald Reagan’s election in 1980.
(10) He was part of a generation of forceful “conviction politicians” (Margaret Thatcher in Britain, Mikhail Gorbachev in the former Soviet empire) that gave US voters an exaggerated reverence for leaders with strong beliefs,
(article)

47The essay, which boasts more complex macro and local structures, contains a great number of progressions which cannot be reduced to one unique relation. On top of the combination with preview-detail, premise-reason can also be associated with a matching relation. Once again, we run into problems when abstract notions are being discussed. While sentences 41 and 42 enter into a matching relation (indigenous people and non-English speaking Australians are given as examples of groups whose heritage is respected), the link between 42 and 43 combines contrast (non English/ English) and a deduction, accepting a paraphrase via then or therefore (= how odd then/therefore that…):

(41) We admire indigenous people for belonging deeply to time and drawing strength from it.
(42) We encourage non-English-speaking Australians to hang on to their language, accepting that in doing so they will also hang on to what is inextricably one with language, the culture it embodies: not just in song and story but in patterns of thought that are inherent in syntax and idiom.

(43) How odd that when it comes to those among us who are of British origin we feel they will only be fully Australian when they have cut themselves off from what we see in others as the nourishment ofa complete life.

48There is also the issue of causal relations which I have proposed to label “implication” and which can combine with re-elaboration. While the following passage contains verbs which elucidate a causal relation, it is often difficult to pinpoint its “direction” (i.e. which clause coincides with the “cause” and which is the “consequence”):

(19) The new economy is based on softer skills.
(20) The predominance now of service industries such as teaching and tourism means that we have had to redefine what we think of as real work and this is a psychological change as well as a change in ‘conditions’.

(21) At this level, the level of feeling, it involves real pain, a strong sense, especially in the bush, that older Australian values, like the older skills, are no longer wanted and, for that reason, no longer respected; a sense of fracture, of alienation.
(22) And this has been intensified by the belief that those who manage our lives are driven by theories that take no account of how people actually live; that they have no ear, behind what sometimes appears as truculent opinion, for the pain, the anger, the frustration and
foiled pride of individual lives.
(23) But people, in our system, always have the last word.

(24) Policies that take no account of feeling inevitably fail.

49Once again, the passage deals with abstract notions (cf. values, feeling, belief). The first two sentences of the paragraph combine preview-detail (softer skills > service industries such as teaching and tourism) with justification, which can be traced to the predicate (mean). However, I consider that the use of the verbs involve and intensify expounds a general relation of implication (between sentences 20 and 21, and 21 and 22 respectively), which can also be identified between the last two sentences of the paragraph—for which it is both unproductive and impossible to determine whether it is because people have the last word that policies ignoring feelings are bound to fail, or the reverse.

4. Conclusion

50Originally, this study was to include a graph giving the frequency in the two texts of each category of relation. To arrive at the results presented here, I did examine every clause relation in each of the two texts. However, the non-linear nature of discourse, together with the preponderance of relations that overlap and which do not fit neatly into one unique category, quickly made it apparent that to attempt to account for relations in quantitative terms would not be extremely enlightening and could even end up contrived. As it is, “dissecting” even a sample of relations as I have done here can but present an artificial vision of the interpretation process. Discourse flows, and because it flows, it does not favour segmental, linear analysis.

51Therefore, is it appropriate to say that we “read between the clauses”? I would argue that the answer is yes, but on an unconscious level: reading between the clauses is so engrained in the interpretation process that we do it without thinking. Following Hoey and Winter’s model, reading (and, for that matter, “listening”) entails two steps: anticipating where the text is heading, and then having such expectations confirmed or not. In argumentation, we expect to find a high frequency of preview-detail, matching relations, and claim-reason. The latter is more common than premise-deduction.

52I have addressed in passing the issue of what makes for an explicit or an implicit relation. Even though the working hypothesis here has equated explicit relations with connectives, these do not necessarily make a clause-relation crystal clear, and other types of signalling exist, namely within the lexus itself. The use of a connective should be seen as a marked option (in both senses of the term), in that the question should not be “why hasn’t a connective been used?” but rather “why has a connective been used?” I have suggested that one reason why certain connectives are used is that they are “summoned” by progressions which run counter to reader expectation.

53Persuasive discourse has been the object of study here, and further work will involve an examination of fiction, specifically the way certain writers overturn linearity using circular constructions and re-elaboration. While the non-linear nature of narrative is recognised (e.g. in the form of prolepsis or flash-backs) that of the local progression between clauses warrants particular attention.

Top of page

Bibliography

Anscombre, J.-C. & O. Ducrot (1997). L’argumentation dans la langue, Sprimont: Pierre Mardaga.

Biber, D., S. Johansson, G. Leech, S. Conrad & E. Finegan (1999). The Longman Grammar of Spoken and Written English, Essex: Pearson Longman.

Brazil, D. & M. Coulthard (1992). “Exchange structure”, in M. Coulthard (ed.) Advances in Spoken Discourse Analysis, London: Routledge.

Chafe, W. L. (1979). “The Flow of Thought and the Flow of Language”, in T. Givon (ed.) Syntax and Semantics, Volume 12: Discourse and Syntax, London, New York: Academic Press, pp. 159-181.

Cotte, P. (1992). “Réflexions sur la linéarité”, in L’ordre des mots, Travaux LXXVI du CIEREC, Université de Saint-Etienne, pp. 53-73.

Cotte, P. (ed.) (1999). Langage et linéarité, Lille: Presses universitaires du Septentrion.

Delechelle, G. (1989). L’expression de la cause en anglais contemporain, étude de quelques connecteurs et Opérations, Thèse d’état, Université Paris III-Sorbonne nouvelle.

Delechelle, G. (1991). “Les connecteurs transphrastiques”, in C. Guimier et P. Larcher (eds.), Les états de l’adverbe, Travaux linguistiques du CERLICO, pp. 115-127.

Halliday, M.A.K. (1985b, 1989). Spoken and Written Language, Oxford: OUP.

Halliday, M.A.K. & R. Hasan (1976). Cohesion in English, Harlow, Essex: Longman.

Hoey, M. (2001). Textual Interaction: an Introduction to Written Discourse Analysis, London: Routledge.

Hoey, M. (2002). “Language as Choice: What is Chosen”, Plenary Paper, International Systemic Functional Linguistics Conference, July, Liverpool.

Hoey, M. & E. Winter (1986). “Clause relations and the writer’s communicative task”, in B. Couture (ed.), Functional Approaches to Writing: Research Perspectives, London: Pinter.

Martin, J. (1992). English Text: System and Structure, Amsterdam: John Benjamins.

Matthiessen, C. & S. Thompson (1988). “The structure of discourse and ‘subordination’”, in J.Haiman & S. Thompson (eds.) Clause Combining in Grammar and Discourse, Amsterdam: J. Benjamins.

Matthiessen, C. (2002). “Combining clauses into clause complexes: a multifaceted view” in M. Noonan, S. A. Thompson & J. Bybee (eds.), Complex Sentences in Grammar and Discourse: Essays in Honor of Sandra Thompson, John Benjamins: Amsterdam.

Morgenstern, A. & M. Sekali (2004). “De la parataxe aux premiers connecteurs: les prémices de l’argumentation chez le jeune enfant”, in Les actes des Journées sur l’argumentation, L’ENS de Lyon, ENS Edition.

Poncharal, B. (2007). “Cohérence discursive en anglais et en français: fonction des connecteurs dans la traduction”, in A. Celle, S. Gresset & R. Huart (eds.) Les connecteurs, jalons du discours, Bern: Peter Lang, pp. 117-136.

Quirk, R., S. Greenbaum, G. Leech & J. Svartvik (1985). A Comprehensive Grammar ofthe English Language, London, New York: Longman.

Rossette, F. (2003). Parataxe et connecteurs: quelques observations sur l’enchaînement des propositions en anglais contemporain, Thèse de doctorat, Paris IV Sorbonne.

Rossette, F. (to appear, a). “News versus Commentary: a Comparison of Theme, Discourse Structure and Forms of Subjectivity between the English and French Press.”

Rossette, F. (to appear, b). “More Ands and Buts about Conjunction: a Study of Sentence-initial Coordination with Respect to Genre.”

Thompson, G. & J. Zhou (2000). “Evaluation and organisation in text: the structuring role of evaluative disjuncts”, in G. Thompson & S. Hunston (eds.), Evaluation in Text: Authorial Stance and the Construction of Discourse, Oxford: OUP.

Van Dijk, T.A. (1985). Handbook of discourse analysis, London: Academic Press.

Verstraete, J.C. (1998). “A semiotic model for the description of levels in conjunction: external, internal-modal and internal-speech functional”, Functions of Language, 5:2.

Verstraete, J.C. (1999). “The distinction between epistemic and speech act conjunction”, Belgian Essays in Linguistics and Literature, pp. 119-130.

White, P. (1998). Telling Media Tales: the News Story as Rhetoric, Ph. D. Dissertation, University of Sydney.

White, P. (2006). “Evaluative semantics and ideological positioning in journalistic discourse – a new framework for analysis”, in I. Lassen (ed.) Mediating Ideology in Text and Image: Ten Critical Studies, Amsterdam: John Benjamins, pp. 37-69.

Top of page

Annex

Two examples of expository/persuasive prose

Essay: “The People’s Judgment”, David Malouf (in Craven P. (ed) The Best Australian Essays 2000. Black Inc. Melbourne)

(1) Geoffrey Blainey’s great phrase ‘the tyranny of distance’, when it was formulated nearly forty years ago, offered a powerful explanation of the problems of being Australian and of Australia’s relationship to the world. (2) It pointed to geography, seen in terms of position and distance, as a determinant of what we call history, that is, of our daily lives as they are lived through events and conditions.

(3) The Australia Blainey was placing was nineteenth-century Australia, six weeks’ sailing distance from Europe, in the age before international cables had made possible the wonder of instant communication and, of course, by the time he formulated it the conditions it described had already changed. (4) Air transport had reduced travel time to a single day; satellite images were about to make every event on the globe instantaneously visible. (5) And it was never quite true, even before technology changed forever our notions of distance and the globe. (6) We live in feelings as well as in conditions and events. (7) Distance is also measured by the heart. (8) In those terms, Europe, Britain, were close, not far off. (9) As for now and the century to come, the new communications systems mean that mere geography will never again determine our sense of where we stand.

(10) In space as the net defines it—language space—El Paso Texas, Aberdeen and Longreach are equidistant from what can only be an imaginary centre: some forms of international business can be conducted as easily from Longreach as from Los Angeles. (11) Once again the fact that our language is English has made us powerful well beyond our size. (12) This is the new form of geography we live with and if it collapses the distance between hemispheres it also collapses the distance between, say, Longreach and Sydney. (13) The space they exist in now belongs to mind, imagination and to the skills they demand to make them work, and this has meant a redefinition, a radical one, of what we do and who we are. (14) No wonder that some of us need time to catch up.

(15) Australians, and especially Australian men, have traditionally defined themselves by the sort of work they do. (16) This is where their pride in themselves, their sense of worth and honour resides. (17) That work had until recently been on the land. (18) It depended on muscle, on hard physical labour and endurance.

(19) The new economy is based on softer skills. (20) The predominance now of service industries such as teaching and tourism means that we have had to redefine what we think of as real work and this is a psychological change as well as a change in ‘conditions’. (21) At this level, the level of feeling, it involves real pain, a strong sense, especially in the bush, that older Australian values, like the older skills, are no longer wanted and, for that reason, no longer respected; a sense of fracture, of alienation. (22) And this has been intensified by the belief that those who manage our lives are driven by theories that take no account of how people actually live; that they have no ear, behind what sometimes appears as truculent opinion, for the pain, the anger, the frustration and foiled pride of individual lives. (23) But people, in our system, always have the last word. (24) Policies that take no account of feeling inevitably fail.

(25) Take the attempt to convince Australians that they are really part of Asia. (26) It failed because it was based on a ‘fact’ of geography that had no life in what people actually felt. (27) Asia, of course, is a loose concept, but even when it was translated into something particular - India or Thailand or Vietnam people still could not feel the tie, even those who had been to Phuket or Bali, were interested in zen or yoga, or had neighbours from one of these countries that they had been to school with; neighbourliness is about sharing things here. (28) Neither did it touch us that we had strong trading links with these places. (29) There was only one place in Asia that we felt close to: Timor. (30) That closeness was based on events that went back half a century, but were still alive in our consciousness as real experience, and on a debt of gratitude and responsibility that we also feel, and for similar reasons, to the people of Papua New Guinea, a bond of feeling based on shared suffering and sacrifice in which any difference of culture or skin colour is cancelled out by our common humanity. (31) Bonds of this sort are not easily forgotten and not honourably shrugged off. (32) They are the only ones that really move us. (33) In this case, of course, there was a gap between what most Australians felt and what politicians thought was practically good for us, but in the end it was feelings that won out.

(34) Then there is ‘our link with Britain’. (35) It upsets many among us that after 150 years of de-facto independence this link should still be so strong. (36) A bond of emotion, of spirit, that for the vast majority of those who feel it has no hint of colonialism and in no way compromises their sense of themselves as wholly Australian, it has to do with family, identity in that sense, personal identity rather than their identity as Australians; it is no coincidence that so many Australians have a passionate interest in family trees and understandable that they might be curious about where their ancestors live, for centuries in some cases, before they found themselves here. (37) It is a link of language too, and of culture in the sense of shared associations and understanding, of shared objects of affection, and a history of which we are a branch a growth quite separate and of itself, but drawing its strength from an ancient root. (38) We will know that this link has been broken when we hear an old hymn such as ‘To be a Pilgrim’, a folk song such as ‘Waly Waly’ or a ballad such as ‘Comin’ through the Rye’ and are no longer moved, or when we no longer laugh spontaneously at old jokes from ‘The Goon Show’ or ‘Fawlty Towers’. (39) The fact is that the part of ourselves in which we live most deeply, most fully, goes further back than one or two generations and takes in more than we ourselves have known. (40) To have no roots in time is to have no roots in place either.

(41) We admire indigenous people for belonging deeply to time and drawing strength from it. (42) We encourage non-English-speaking Australians to hang on to their language, accepting that in doing so they will also hang on to what is inextricably one with language, the culture it embodies: not just in song and story but in patterns of thought that are inherent in syntax and idiom. (43) How odd that when it comes to those among us who are of British origin we feel they will only be fully Australian when they have cut themselves off from what we see in others as the nourishment of a complete life.

(44) This, I suspect, had more influence on the recent referendum than we care to recognise; not because Australians are still colonial or have a weak sense of national identity or have not yet come of age, but because the case for the republic was put in terms that people had no strong feeling for, or which ran counter to what they actually felt. (45) An Australian republic can only be argued for convincingly at the level of feeling - on what we feel towards the place and for one another. (46) When it comes at last, some time in the next century, it had better be a true republic, one that is founded not on the loyalty of its citizens to their head of state, but on their loyalty to one another: on bonds, which already exist and which we already recognise, of reciprocal concern and care and affection. (47) A republic based on loyalty only to a head of state is a monarchy in disguise, even when the monarch is elected and temporary. (48) An elected monarch, as in too many republics one might name, can very easily become an autocrat.

(49) Perhaps after a century of theories and ideas and ideologies, some of them murderous, we might try listening at last to what people have to say; paying attention to what they have to tell us; accepting, too, and without resentment, that in being human they are imperfect, and that theories, even the most beautiful and idealistic, are for angels of the imagination, not real men and women. (50) We might grant people the dignity of a life determined not by cold principle, but by what they will recognise as true to what they are.

Newspaper article: At last the focus is on talent, Howell Raines (Guardian Weekly, April 13-19 2007)

(1) A1 Gore’s recent appearance on Capitol Hill marked his dominance in Washington’s debate about the science of global warming. (2) It also raised a supremely important question about the state of political science in the US. (3) Given the pain inflicted on the world by George Bush’s faith-based presidency, are US voters ready for a knowledge-based presidency?

(4) It is a question worth exploring in advance of the 2008 election, because Gore and several other prospective candidates represent a reappearance on the US campaign scene of an endangered breed – elected officials who have used their time in office to become expert in government.

(5) Gore stands in a long line of congressional figures who came to Washington intent on mastering one or more essential issues. (6) His range was unusually broad—the environment, energy, communications technology and nuclear control.

(7) As the Bush presidency winds down, one can feel the electorate holding its breath, as if to reassure itself it can will the centre into holding. (8) In the press, the opposition and his own party, Bush’s critics are restraining themselves in the belief that admitting how dangerous this man really is will diminish the chances of safely running down the clock until Inauguration Day 2009.

(9) American politics has been building up to Bush since Ronald Reagan’s election in 1980. (10) He was part of a generation of forceful “conviction politicians” (Margaret Thatcher in Britain, Mikhail Gorbachev in the former Soviet empire) that gave US voters an exaggerated reverence for leaders with strong beliefs. (11) Reagan did value beliefs over facts, but luckily his strategic beliefs meshed almost perfectly with a moment of opportunity to remake cold war geopolitics.

(12) Bill Clinton’s noble effort to reform healthcare was knowledge-based, but its political collapse gave learning a bad name with half of the electorate, just as the country was seized with one of the religious manias that sweep America every century. (13) Thus was the stage set for the long night of Bush and his rigid belief in petroleum and prayer. (14) US foreign policy became wars-for-oil and its fiscal and environmental policy has been to use the blunt terms of this White House—about preserving at all costs the wealth of Texans and Saudis.

(15) What are the signs that America’s disenchantment with knowledge and elected officials who possess it may be ending? (16) Encouragingly there are prospective political candidates in both parties who, whatever their individual flaws, have used their careers to acquire a body of knowledge about governance.

(17) As a matter of courtesy, let’s consider Gore first. (18) Of all the American politicians I met during a long journalistic career, I believe that Gore and Richard Nixon knew most about all aspects of government and politics. (19) (Knowledge, alas, can be morally neutral.) (20) My bottom line on Gore is that he would be a competent president who fully grasped the choices before him. (21) Will he run in 2008 if there’s an opening? (22) Remember what his father, the late senator Albert Gore, said when Gore went on the Democratic ticket in 2002: “He was raised for it.”

(23) What is striking about this presidential cycle is that for the first tinte in a long time, a number of plausible candidates in both parties seem, at a minimum, informed enough for the top job. (24) Hillary Clinton fits that description. (25) She is a meticulous student of government who has used the Senate as a graduate school. (26) One sees a similar seriousness of purpose in John Edwards and Barack Obama.

(27) On the Republican side, John McCain and Chuck Hagel have used their Senate terms to educate themselves about national security issues. (28) Neither of them would have stood by while his vice-president and defence secretary ran amok in Iraq. (29) Despite his personal idiosyncracies, the former New York mayor Rudy Giuliani is a seasoned executive, having run and to some extent reformed New York’s City Hall. (30) Although ideologically shifty, Mitt Romney crafted a successful financial career that did not depend entirely on his father’s money, patronage and connections.

(31) I’m not saying I would vote for all these candidates, or that they are uniformly talented or even equal in their intelligence and preparation. (32) But they do reach a threshold of competence and preparation that Bush did not, despite his family and educational credentials. (33) His late and simplistic religious conversion, for example, has been a much stronger force in his presidency than his degrees from Yale and Harvard.

(34) The vagaries of American politics and the unpredicted pressures of the presidency chasten one’s optimism. (35) But it is worth remembering that our most naturally gifted president, Abraham Lincoln, emerged after the horrid term of James Buchanan set the stage for the civil war. (36) I am reliably informed that no less a presidential scholar than the late Arthur Schlesinger had tipped Bush as a challenger for Buchanan in the running for worst president ever.

(37) None of us can say with certainty that Bush is simply a dullard. (38) But we cannot deny what is self-evident from the past six years. (39) We’ve given war a chance. (40) Now, when it seems possible that one or both parties could allow it, let’s give knowledge a chance.

Top of page

Notes

1 My translation.

2 These figures account for finite, non-embedded clauses, and therefore exclude relatives and subordinate clauses which constitute the Subject or Object of the main clause, for which there is a syntactic link but no rhetorical link. The same segmentation has been applied here: the “inter-clausal relations” studied include both links between sentences, and links between finite, non-embedded clauses within the sentence.

3 Rather than refer to a speaker and an addressee, I will adopt the terms writer and reader, as only written texts are under investigation here.

4 Van Dijk uses such a principle to explain the inacceptability of a sequence such as We went to an expensive restaurant. John ordered a big Chevrolet.

5 Rumelhart, D.E. & A. Ortony (1977) ‘The representation of knowledge in memory’, in R.C. Anderson et al. (eds) Schooling and the Acquisition of Knowledge. NY: Halsted Press.

6 While causal relations are often cited to illustrate the external/internal dichotomy, it can be argued that establishing any causal sequence is a matter of interpretation. For example, appraisal theory (e.g. White 2006) lists clausal relations—be they of any type—among interpersonal markers, which reflect a high degree of speaker involvement.

7 This is, of course, a largely simplified picture, and I do not dismiss the ambivalence of certain subordinating conjunctions, particularly those which express clausal relations. As for coordinating conjunctions, these are the only category of connective to feature in both positions. While their traditional role is limited to within-sentence relations, the number of sentence-initial coordinators can sometimes outnumber those situated within the sentence (Rossette, to appear, b).

8 An extreme example of this is provided by the types of progressions identified between turns in conversation, within what has come to be known as exchange structure theory (see for example Brazil & Coulthard 1992).

9 One definition of example provided by Merriam-Webster indicates “one (as an item or incident) that is representative of all of a group or type”.

10 Importantly, the adverb also does not elucidate the relation with the previous clause. Its scope relates to the notion of distance and a parallel can be established between distance measured by travel time, mentioned in sentence 3, and, as stated here, distance “measured by the heart.”

11 We can go further and postulate that for this natural order to be upturned, and deduction preferred over justification, some type of explicitation proves particularly useful to counter the expectations held by the reader.

12 This is, for example, a feature of asyndeton in Marguerite Duras, in which a clause is repeated with only a slight surface variation, e.g. Je descends toujours du car quand on arrive sur le bac, la nuit aussi, parce que toujours j’ai peur, j’ai peur que les câbles cèdent, que nous soyons emportés vers la mer. (Duras, L’Amant, Editions de Minuit, p. 18) The effect is not as emphatic in the published English translation, in which the repetition of the finite verb has been suppressed: I always get off the bus when we reach the ferry, even at night, because I’m always afraid, afraid the cables might break and we might be swept out to sea. (Bray, Pantheon Books, p. 11)

13 As exemplified in the following sentence analysed by Martin (1992: 265): Government and police sources agreed that the force’s problem is lack of morale through a lack of discipline resulting from the absence of a strong figurehead, which can be “unpacked” into three messages: (1) the police force has problems; (2) because they aren’t being disciplined; (c) because the force doesn’t have a strong commissioner.

Top of page

References

Bibliographical reference

Fiona Rossette, « Reading Between the Clauses”: an Exploration of Implicit Inter-clausal Relations in English », Anglophonia/Sigma, 12 (24) | 2008, 219-243.

Electronic reference

Fiona Rossette, « Reading Between the Clauses”: an Exploration of Implicit Inter-clausal Relations in English », Anglophonia/Sigma [Online], 12 (24) | 2008, Online since 13 December 2016, connection on 27 June 2017. URL : http://anglophonia.revues.org/1009 ; DOI : 10.4000/anglophonia.1009

Top of page

About the author

Fiona Rossette

Université Paris 10

By this author

Top of page

Copyright

Licence Creative Commons
Anglophonia – French Journal of English Linguistics est mis à disposition selon les termes de la licence Creative Commons Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International.

Top of page
  • Logo Presses universitaires du Midi
  • Revues.org