Skip to navigation – Site map
English Linguistics

And-Prefaced Utterances: From Speech to Text

Fiona Rossette
p. 105-135

Abstract

Cette étude a pour objectif de comparer les emplois du coordonnant and à l’oral en tant que « discourse marker » avec les emplois de and à l’écrit quand il est utilisé en début de phrase. Si de très nombreux travaux existent sur la coordination, il restait à établir un lien plus explicite entre langue orale et langue écrite. En effet, il semble que ce qui motive en grande partie les auteurs à passer outre la proscription traditionnelle ne pas commencer une phrase par and ou but ») est la volonté de rendre plus claire la macro-organisation du texte, ce qui rejoint la fonction des « discourse markers » à l’oral. Placée ainsi en tête de phrase, and facilite la compréhension du texte. Après avoir rappelé les fonctions décrites par Schriffin (1986 ; 2006) de and en tant que « discourse marker » dans un contexte de dialogue à l’oral, j’examine les occurrences de and en début de phrase dans un corpus écrit composé d’écrits scientifiques, d’articles de presse, d’essais et d’œuvres littéraires. Comme transition entre ces deux registres, je m’appuie sur l’étude des occurrences de and dans un discours de Steve Jobs, exemple de « monologue » à l’oral dans la mesure où un seul énonciateur s’adresse à un auditoire. Dans ces trois registres, and marque typiquement la différence, ou la dissociation, soit entre deux énoncés, soit entre des groupes d’énoncés, et négocie ainsi les transitions du discours. Il permet également la focalisation de l’information qu’il introduit.

Top of page

Full text

1. Introduction: Distinguishing speech and textual functions

1Descriptions of and above the intra-sentential (syntactic) level can be found within both the written mode, when the coordinator features in sentence-initial (S.I.) position, and the spoken, notably when it is conferred the status of discourse marker. This study is an attempt to explore the interface between the two, the hypothesis being that there is an overlap between speech and textual roles, and that writers flout prescription (cf. “Don’t start a sentence with and or but”) precisely so as to instantiate in the written mode meanings akin to those associated with and as a discourse marker in speech. Indeed, Schriffin (1987: 13) defines discourse markers as “sequentially dependent elements that bracket units of talk” (my emphasis). The structuring role is a common thread running through subsequent work on discourse markers: for example, Redeker (2006) defines a specific category which she calls “discourse operators” which display « discourse-structuring functions », signal notably “segment transitions” and “function as cues to direct the listener’s attention”. The structuring function is also the common thread of recent work on coordination at sentence level in text: Dorgeloh (2004) and Bell (2007) have highlighted the role and plays in structuring discourse.

2In comparison, other authors have preferred to identify an invariable value whatever the position (intrasentential or sentence-initial), and in addition, have not always considered text, as opposed to speech, for its own sake. For example, within relevance theory, Blakemore and Carston (2003: 588) affirm that their analysis of and can be “carri[ed] over satisfactorily to utterance-initial occurrences of and” (“detached conjuncts”), examples of which are taken from speech. Dichotomies opposing two different levels of clause linkage exist in the literature, such as van Dijk’s “semantic” versus “pragmatic” uses of connectives, the latter “signaled by sentence-initial position” (van Dijk 1979: 112; my emphasis) although examples given are also from speech. Similarly, in Halliday and Hasan’s 1976 study, which highlights the structural/cohesive divide by placing at a different level of the linguistic system relations within the sentence as opposed to those above the sentence, many of their examples of cohesive and are taken from reported speech. Of course, an overlap between speech and textual roles is generally taken as a given (for example, Bell’s 2007 study includes references to Schriffin’s work), just as speech serves as a point of reference to explain written usage (as noted by Lyons (1997: 638)) and quoted by Schriffin herself (1987: 6): “there is much in the structure of languages that can only be explained on the assumption that they have developed for communication in face-to-face interaction.” However, what I would like to do here is to consider speech and writing separately, precisely in order to grasp the parallels between the two. Indeed, at the crux of sentence-initial usage lies the way and negotiates ─ and in so doing, brings to light ─ the tension between continuity and discontinuity which is inherent in the dynamics of discourse. This finds an echo in Schriffin’s studies on the coordinator in speech: from the beginning of her work, she refers to the notions of continuity and differentiation. In what follows, Schriffin’s analyses will first be reviewed (section 2). Usage of and in an example of monologic speech (an extract of a Steve Jobs keynote) will then be studied (section 3), before turning to sentence-initial use (in text) (section 4).

2. Schriffin’s description of and as a discourse marker

  • 1 Schriffin is careful to underline that the coordinator displays, or brings into sharper focus, disc (...)

3As a discourse marker, and assumes two functions: it “coordinates idea units”, characteristic of linkage in the middle of a turn in dialogue (i.e. a discourse segment uttered by the same speaker), and it “continues a speaker’s action”, a function which is made particularly prominent at the beginning of a turn (Schriffin 1986; 1987). If the terms of “coordinating” and “continuing” are quite close, continuity appears, in Schriffin’s early metalanguage, to be more action-based (i.e. pragmatic, or interactional). The two functions seem to overlap in later work, when continuity is further defined in terms of the cumulative effect produced: the basic pragmatic meaning of and is “to mark the speaker’s communicative intention to continue a “cumulative set” (Schriffin 2006: 335). Moreover, and “contributes to the process of constructing discursive sequences whose smaller parts combine to form a larger structure” (Schriffin 2006: 15; my emphasis). In other words, the coordinator provides a window into the macro-organization of discourse, by highlighting the interplay between the local level (linkage between adjacent clauses to form a set or an idea unit/structure) and the global level (how these sets subsequently link up): “Idea structures […] are formed not only through relationships between individual ideas, however, but through relationships between sets of ideas” (Schriffin 1986).1

2.1. Association between adjacent utterances

4The example below, taken from Schriffin (1986), illustrates the use of and between adjacent utterances which are closely associated:

(1)

The speaker is talking about the old days:

(a)

What changed the whole way of living is the automobile.

(b)

You couldn’t go anywhere,

(c)

so you congregated together,

(d)

and y’got in one big truck, or something

(e)

and you went ─ went on a picnic,

(f)

and you had a good time.

(g)

Today, you could care less!

5The events from (c) to (f) “collectively describe the ‘way of living’ prior to automobiles […] and they form a collective contrast with the situation of today” (Schriffin 1986: 52). Linking by and at the local level (between adjacent clauses) is motivated by the fact that the clauses are similar in function, grouping “specific events together in service of a more general point” and hence constituting a set. The “general point” can be formulated via a question, if we draw a parallel with later descriptions made by Blakemore and Carston, for whom and requires “the conjuncts to be processed together”, linking a “conjoined proposition” which “provides a single answer to a single (implicit or explicit) question”.

6In (1), a contrast is set up with (g) due to the absence of any form of conjunction (i.e. zero), which, at a global level, signals if not the beginning of a new set, at least the end of the set to which the previous clause belongs. For the sake of comparison, I have removed and below:

(1a)

(a)

What changed the whole way of living is the automobile.

(b)

You couldn’t go anywhere,

(c)

so you congregated together,

(d)

y’got in one big truck, or something

(e)

you went ─ went on a picnic,

(f)

you had a good time.

(g)

Today, you could care less!

7Removal of the coordinator, and therefore of the contrast, results in a certain degree of confusion, with the onus now on the listener to seize on her own the rhetorical switch, which is no longer “signposted”.

2.2. Differentiation

8What matters is that a contrast is established between a textual norm which is set up in the first part of the text, and deviation from that norm (Schriffin 1986: 51), whether the coordinator coincides with the link at local level, as in (1), or at global level, as in (2) below. The sequence is about anti-Semite action: this time, a set of events is signaled via the repetition of zero, while And prefaces a conclusion to be drawn from the set. In this instance, the coordinator points to a higher, macro-relation precisely because it reflects, at the local level, a certain degree of dissociation, or differentiation:

(2)

(a)

They threw us in the fire,

(b)

they shot us,

(c)

they killed us,

(d)

they put us in the gas chambers,

(e)

and they couldn’t do it.

  • 2 « Être capable d’associer, c’est d’abord être capable de dissocier. Tout coordonnant se charge de l (...)

9Of course, it could be considered a contradiction that, depending on the example, and reflects either association or differentiation between adjacent clauses. Schriffin explains this in terms of a paradigmatic contrast with zero. Moreover, it can be argued that continuity and discontinuity represent two sides of the same coin. Lapaire (2005: 475) notes that linkage first requires dissociation, and that all coordinators are “differentiators”.2

10Text-chunking is motivated both functionally, grouping together clauses similar in function, and referentially/topically grouping those which share the same topic ─ note that in (1) and (2), the clauses belonging to the same set also share the same subject (you; they). Further, and can help to differentiate between topics, as in the sequence below, in which the speaker talks about sports he played when younger:

(3)

(What sports, particularly?)

(a)

Really football and baseball.

(b)

Cause two of ‘em play on little league teams

(c)

So I hadda learn to... understand the game,

(d)

or I was sitting on the bench like three days a week not knowing

(e)

what was goin’ on.

(f)

And with the football, they’re very big on football.

(g)

So I’ve been trying t’watch it on Sunday,

(h)

and trying t’understand it a little bit more.

11The sports football and baseball are introduced in (a); baseball is discussed from (b) to (e), while and prefaces a switch to football in (f), helping to “segment the explanation into the two prefigured topics” which makes it “differentiating” (Schriffin 1986: 55).

2.3. Imposing continuity when it is not obvious (including “re-entry” into a set)

12Turning to the more pragmatic, or interactional, “continuative” function of and, it is identified in contexts where it “impose(s) a meaning of continuation even when the prior interaction has not made such a meaning structurally obvious” (Schriffin 1986: 61). At the simplest level, it can negotiate the speaker’s “return to the floor” (Schriffin 2006) – i.e. at the beginning of a turn. A specific use features in Schriffin’s corpus of interviews when and appears at the beginning of a turn to “routinize a question that is potentially problematic because it is schematically and globally out of place” (Schriffin 2006: 42), and so “compensates for the problematic gap by glossing the question as an unproblematic continuation of the story” (Ibid. p.46). Or, in terms of macro-structure, encompassing one or several turns, and can counter the expectation that a particular set is finished, as illustrated below:

(4)

Jack:

(a)

He knew how t’get the ─ he knew the whole audience’d laugh.

(b)

So he must’ve had something to him.

Freda:

Even the teachers, huh?

Jack:

(c)

Even this teacher, this one that─ she laughed.

(d)

She couldn’t help it!

(e)

And I’m playin’ (hums melody)

(f)

and I’m playin’ y’know!

  • 3 Similarly, in her discussion of discourse markers as “attentional cues”, Redeker (2006) includes an (...)

13For Schriffin, (a) ends the “storytelling” set, while (b) reiterates the point of the story, and Freda’s contribution together with (c) and (d) provide further evaluation. What happens in (e) however is that Jack decides to re-enter the story-telling set, by referring to another event: “And marks this re-entry, thus imposing a continuative meaning on a discourse whose structure actually warranted otherwise” (Schriffin 1986: 61). And enables the speaker to reach back up into the discourse and pick up on a previous thread, an example of discursive dynamics which go beyond linearity, unlike the more linear chunking present in examples (1) to (3).3

14The potential displayed by and to negotiate either association or differentiation, and hence bring into sharper focus the macro structure of the discourse, is a dialectic which captures the essence of what it does in sentence-initial position in writing. It will now be used in the analysis of an example of monologic speech, which will prove a useful stepping stone towards the study of (monologic) text.

3. And in an example of monologic speech

  • 4 The transcript is mine, based on a recording of the presentation, as no official transcript exists (...)

15The following sequence is a transcript of part of Steve Jobs’ keynote address during which he presented the iPad 2 (March 2, 2011). It provides an example of monologic speech (as opposed to dialogue, and so devoid of turn-taking), characteristic of public addresses, where the code is oral and, while obviously prepared in advance, does not consist of reading from a script. Indeed, Jobs represents an interesting case for, if he was famous for rehearsing in minute detail his product presentations, he spoke without notes and fostered informality and spontaneity. Several minutes into the presentation, after having first talked about the speed and the cameras in the new iPad, he turns to the issues of its size and weight. To facilitate the discussion below, the sequence has been broken into four sections, based on changes in local discourse topics, although there is some movement backwards and forwards between them: 4

(5)


















[…] / now, having built in all of this stuff, one of the most startling things about the iPad 2 is it is dramatically thinner / not a little bit thinner, a third thinner / 33% thinner / that’s what it looks like / so if you look at the numbers, when you look at the numbers, gone from 13.4 mm down to 8.8 mm thick, it’s dramatic /

and for those of you that have iPhone 4s, the new iPad2 is actually thinner than your iPhone 4 / so we’re incredibly happy with this / and when you get your hands on one, it feels totally different / and all these other tablets are coming out, most of them even thicker than the original iPad, nothing even approaching this /

in addition to thinner, it’s lighter as well, going from 1.5 pounds down to 1.3 / and you might not think that’s a lot / but when you get down to 1.5 pounds, a tenth of a pound is a lot / and uh it feels quite a bit lighter /

and it’s got an all-new design / it’s just beautiful / so this is what it looks like / it’s really thin /

and it comes in two colours, black and white / we’re going to be shipping white from day 1 / and to give you some scale, this is what it looks like / again, you can just pick this thing up / it almost floats / […]

16What is striking in this excerpt is the quasi systematic use of and over spans of speech. In writing, such a high frequency is extremely rare and considered marked. Several examples come to mind, and can be explained due to their “orality”: the Bible, initially written to be read out loud and transmitted orally (Alter 1999), features “biblical parataxis”, and certain writers, such as Hemingway, are known for their polysyndetic, “oral” style. In addition, scripts of speeches (e.g. political speeches) ─ a corpus of monologic writing-cum-speech worthy of study for its hybridity ─ exploit and as a connective, no doubt to create the impression of spontaneity and orality, for example:

(6)





This is historically a marginal seat and it’s a seat that has moved around very much according to the quality of its representation and we have to work and fight very hard in order to retain the seat of Ballarat at the next election and I am delighted Charles has been chosen as the Liberal Party candidate and the standard bearer to replace Michael Ronaldson (J. Howard, 10/10/2001; source: www.pm.gov/au)

17To return to Jobs’ address, the first relevant question to ask is whether and helps to make the macro-structure of the discourse clearer.

3.1 Differentiation within a set (including “re-entry”)

18The first utterance of the first section contains the claim “it is dramatically thinner” and initiates a set, with the subsequent units prefaced by zero providing justification for this claim. So introduces the utterance “so if you look at the numbers…” which provides detailed figures supporting the previous claim (“33% thinner”). Jobs has therefore made his point. However, he does not leave it at that, and the first instance of and in the excerpt (“and for those of you that have iPhone 4s…”) introduces a comparison which serves as a further argument. The second instance of so (“so we’re incredibly happy with this”) foregrounds a move towards closure (cf. “Speakers often repeat the point which their story was intended to establish as a way of ending its conversational relevance” (Schriffin 1986: 61)), with and marking re-entry (“and when you get your hands on one…”). The final utterance of the section, prefaced by and (“and all these other tablets are coming out”) broadens the perspective by comparing the iPad with competitors’ tablets, providing closure which is this time definite. Just like in (2) above, this last instance of and is locally differentiating, appearing between consecutive utterances which do not share the same rhetorical function, and therefore providing a window into a higher, macro-relation. In fact, the three instances of and in this first section are all differentiating in that they inhibit the assimilation between adjacent utterances (p and ≠ q), which serve different rhetorical purposes.

19The first two examples of and play out a one-step-forward, one-step-back movement, and can be paraphrased by “but that’s not all, I have something else to add.” This function is common to other instances of the coordinator in Jobs’ speech. The second section displays a similar format to the first:

in addition to thinner, it’s lighter as well, going from 1.5 pounds down to 1.3 / and you might not think that’s a lot / but when you get down to 1.5 pounds, a tenth of a pound is a lot / and uh it feels quite a bit lighter /

20The first utterance contains the claim that the iPad is lighter, for which sufficient justification is provided with the figures presented in the non-finite clause “going from 1.5 pounds down to 1.3”. The coordinator (“and you might think…”) then marks re-entry, inducing a “that’s not all” colouring, while in the second instance (“and uh it feels quite a bit lighter”), it again introduces an utterance which reiterates the main point and hence serves as a conclusion. Similarly, the final instance of the coordinator in the extract also marks re-entry, an example I will discuss in more length below:

and it comes in two colours, black and white / we’re going to be shipping white from day 1 / and to give you some scale, this is what it looks like / again, you can just pick this thing up / it almost floats / […]

21This example of monologic speech reflects the ambivalence of the coordinator: it is associative (e.g. above: “and it comes in two colours”, where it introduces a new property of the iPad, and adds to the list at the macro-level), and at the same time dissociative, “separating out” the elements. It proves valuable in the discourse of Jobs, who wants to constantly surprise his audience, and avoid the impression of serving up “more of the same” of what has just gone before. It can be considered an ingredient of Jobs’ up-beat delivery style, contributing to the hype of his address. However, would comprehension be jeopardized if and were removed? As an example, here is the first section, reproduced without the coordinator:

(5a)







/ now, having built in all of this stuff, one of the most startling things about the iPad 2 is it is dramatically thinner / not a little bit thinner, a third thinner / 33% thinner / that’s what it looks like / so if you look at the numbers, when you look at the numbers, gone from 13.4 mm down to 8.8 mm thick, it’s dramatic / for those of you that have iPhone 4s, the new iPad2 is actually thinner than your iPhone 4/ so we’re incredibly happy with this / when you get your hands on one, it feels totally different / all these other tablets are coming out, most of them even thicker than the original iPad, nothing even approaching this /

  • 5 At the same time, removal of the coordination can produce the effect of arguments being thrown like (...)

22If the resulting sequence in (5a) does not engender comprehension problems, a certain “lack of purpose” can be felt ─ a similar lack of purpose highlights the difference between (1) and (1a) above. The listener is inclined to think “so what?”, and needs to work through seemingly disparate elements whose function(s) can only be grasped retrospectively. In comparison, the original produces the feeling that the text is “going somewhere”. The same effect is produced by the repetition of and in the political speech quoted in (6).5

  • 6 In contrast, in a corpus of televised interviews (dialogic speech, i.e. featuring turn taking betwe (...)

23And acts like a signpost, negotiating us through what therefore comes across as a more motivated rhetorical structure ─ and therefore a more motivated rhetoric. It prevents listeners from taking time out ─ keeping them, as it were, on their toes. A rhetorical/textual tension is created. And pushes the discourse forward, “launching” each utterance – indeed, Lapaire and Rotgé (1998: 306) observe the role of and to “launch” discourse, describing it as a “discursive tool favouring textual progression” (my translation). In relation to a written example, Sekali (2010: 245) talks about and making “the text advance in an explicit manner” (my emphasis). Biber et al. (1999) describe sentence or turn-initial and as an “initiator”, rather than a connector of equal elements, therefore challenging traditional theories on coordination (linkage of items of “equal” status) compared to subordination (linkage of items of “unequal” status). Interestingly, Jobs pronounces and quite emphatically, with a full vowel (/æ/) and it is only reduced to /ə/ in one instance: at the beginning of the second section “and when you get your hands on one…”. Emphatic pronunciation suggests an emphatic reading, compatible with an exclamatory value, glossed by “can you believe it?” 6 The coordinator therefore contributes to the persuasiveness of Jobs’ arguments, and it is in its differentiating capacity that this is accomplished.

3.2 Differentiation between sets

  • 7 Pausing cannot be the unique criterion on which to base a discourse boundary: Jobs makes numerous “ (...)

24And also articulates differentiation between certain sub-sections, or sets. These can be summed up via the following questions: (1) what is the iPad’s width? (cf. “it’s thinner”); (2) what is its weight? (cf. “it’s lighter”); (3) what does it look like? (cf. “an all-new design”); (4) what colours does it come in? (cf. “black and white”). Transitions between these sets all feature an additive connective: there are two examples of and (and it’s got an all-new design; and it comes in two colours, black and white) and one example with in addition introducing an adverbial phrase, in combination with post-predicate as well (in addition to thicker, it’s lighter as well). Unlike the adverbial, the coordinator is preceded by a noticeable pause (of one or more seconds in length).7 A one-second pause precedes and in the extract below:

[…] / it feels quite a bit lighter / [pause]

and it’s got an all-new design / it’s just beautiful /

25On its own, and does not necessarily create a dissociating effect, but it has the potential to do so if accompanied by a pause. This is the case in the other context where the coordinator coincides with a discourse boundary, and is preceded by a pause of two seconds:

[…] / so this is what it looks like / it’s really thin / [pause]

/ and it comes in two colours, black and white/ we’re going to be shipping white from day 1/ and to give you some scale, this is what it looks like / again, you can just pick this thing up, it almost floats / (….)

26If notionally, the question of colour belongs to the topic “what the iPad looks like”, the pause creates a dissociation which leads us to identify a discourse boundary. Moreover, the question of colour is taken up in the subsequent clause, in which immediate shipping of the white version is announced. This is met with great applause by the audience: clearly, the question of colour is a key argument, and Jobs wants to labour the point. The pause is akin to a “dramatic pause”, conferring informational prominence, which, it can be posited, is heightened by and. The < pause + and > variable is not discussed by Schriffin, no doubt because pausing in the context of dialogic speech is instead interpreted as a cue to the listener that it is his/her turn to speak. The informational prominence produced here is specific to monologic discourse and will be an essential factor motivating sentence-initial usage in text. It ties in with the emphatic, “can-you-believe-it” potential of and noted above. A propos, the following example of S.I. and taken from the script of a political speech, while it does not operate between sets, is worthy of note:

  • 8 The National Teacher and Principal Recruitment Act. Speech given March 7 2001 (source: http://clint (...)

(7)



The teacher shortage in the United States is projected to reach a staggering 2.2 million teachers in the next ten years. And, these shortages have already begun for communities across my state as well as throughout the country. (political speech, H. Rodham Clinton)8

27The coordinator is followed by a comma, suggesting a pause after it (and making it quite difficult to reduce the vowel), which can also be explained due to the wish to confer informational prominence.

3.3 Imposing continuity

28Let us turn to the instance of and which follows Job’s announcement that white iPads will be shipped from day one ─ the last example of the coordinator in the extract. It is preceded by the longest pause of the extract, lasting three seconds. However, unlike the examples discussed above, the combination < pause + and > does not articulate a transition between sets, and does not contribute to informational prominence. Rather, the pause results from the audience’s applause in response to the news of the shipping of the colour white. As for and, this example was analysed in section 3.1 as marking re-entry into an otherwise closed set. If this analysis remains valid, it can be taken a step further. Here, and provides a bridge on the interactional level, between the audience and the speaker: it materialises the transition back to the speaker, after he has been interrupted (meditated though this may be) by the applause of the audience, the only form of audience participation in this (verbally) monologic discourse. In this, the coordinator resembles Schriffin’s “turn-taking” and, to mark the speaker’s return to the floor. Moreover, the content of a previous utterance is repeated (so this is what is looks like > this is what it looks like), as if the interruption then necessitates the main point being brought back into focus.

29Finally, this content is preceded by a fronted adverbial (And, to give you some scale, this is what it looks like), which, in thematic position sets up a new framework for the repeated content and creates a certain hiatus, or barrier, on the syntactic level: the heart of the clause structure (subject + verb) is delayed. It can be argued that and softens the hiatus. Compare with:

(5b)

/ and it comes in two colours, black and white / we’re going to be shipping white from day 1 / to give you some scale, this is what it looks like

30Likewise, the first instance of and in the extract is followed by a fronted adverbial:

/ so if you look at the numbers, when you look at the numbers, gone from 13.4 mm down to 8.8 mm thick, it’s dramatic / and for those of you that have iPhone 4s, the new iPad2 is actually thinner than your iPhone 4/

31to be compared with:

(5c)

[…] it’s dramatic / for those of you that have iPhone 4s, the new iPad2 is actually thinner than your iPhone 4 /

32These cases exemplify continuity imposed in the face of discontinuity, this time on the syntactic level, with and providing a syntactic bridge to soften the transition between adjacent utterances. In the corpus of writing about to be examined, the majority of instances of S.I. and appear in conjunction with a fronted adverbial ─ a common denominator worth underlining, and it is interesting to note that examples of the combination exist in speech.

33To summarise the way and functions in this extract: it is more characteristic of differentiation than association. It assumes a differentiating role within a set, to bring into sharper focus the different functions of adjacent utterances, including re-entry into a set, and suggesting a “that’s not all” meaning. It also signals boundaries between sets, by prefacing either the last utterance of a set, or the first of a new set (often in combination with a pause). However, its associative potential is demonstrated through the way it imposes continuity, providing a bridge which is otherwise lacking (e.g. due to a long pause or to a fronted adverbial, for example). Finally, what appears specific to monologic speech (as opposed to dialogue) is and’s potential to confer informational prominence ─ an essential factor in sentence-initial usage, due to the very status of the sentence as opposed to that of the clause.

4. Sentence-initial And

4.1 Genre variation

34Before turning to a discussion of the different types of examples of S.I. and observed in the written corpus, a few words need to be said about the latter. It represents different genres and hence highlights variation in depending on text type. Biber et al. (1999) underline the less frequent status of coordination in S.I. position in text compared to turn-initial position in speech. Within the written genres, they also note, together with Dorgeloh (2004), that the lowest frequency is to be found in academic prose. This is confirmed in the present corpus (see table in appendix 2), containing examples of press articles (from dailies and from the weekly The Economist), essays, fiction, and academic prose. In fact, three academic articles contain no instances of S.I. coordination (and/but). Why is this so?

  • 9 Stigmatization may also result from the fact that, when compared with subordination, coordination i (...)

35S.I. coordination is stigmatized in academic prose, and represents a problematic area of usage, particularly in academic writing by non-native speakers (Bolton, Nelson & Hung ; Lee 2005). Dorgeloh (2004) documents how the frequency of S.I. and dropped off from early Modern English (1500-1710) to the present day, mirroring the move from narrative to analytical text structuring, the latter exemplifying the modern mode of scientific/academic enquiry. In turn, “sentence-initial and became associated with the older, more narrative, and hence less professional style and thus became increasingly stigmatized” (Dorgeloh 2004: 1770).9 Moreover, there has been a trend over the last forty years towards increased S.I. usage which is part of the more general “shift of English towards colloquiality” (Ibid.: 1762), hence for example the higher number of S.I. and she finds in modern-day tabloids compared to broadsheets, and the observation by Biber et al. (1999: 84) that initial coordinators are often based on a model of “more spontaneous discourse.”

  • 10 A corpus of fiction (the first 30 pages of 12 works) was included at a later stage of this study. E (...)

36Bell (2007) compares different disciplines within academic prose and notes that S.I. coordination is more frequent in the humanities than elsewhere (e.g. medicine, chemistry). He views it as a mark of speaker involvement, not so much in terms of colloquiality but in terms of the attention paid to making the text’s rhetorical organization clear to the reader, to produce a more reader-friendly text. In the present study, the academic prose does not provide examples of different fields of academia (all texts deal with language-related topics), but it does highlight an interesting contrast between academic articles and academic textbooks, which share a priori the same writers (academics) but not the same target audience (peers versus students/a wider public). In fact, the highest ratios of S.I. coordinators in the corpus are found in the essays and the academic textbooks – perhaps the trace of a heightened level of “reader-friendliness”.10

4.2 On the specific status of the sentence: initial examples

  • 11 « Chaque phrase est en principe l’enjeu d’un différend entre des genres de discours, quel que soit (...)

37The tension between association and differentiation also lies at the heart of the ambivalent status of the sentence within discourse, a point highlighted for example by Lyotard (1983: 200) when he describes the sentence as both forming a separate entity but also as constituting a part of a text.11 Starting a new sentence creates a new discursive beginning, which pushes the text forward. To the question “what makes a sentence different from a clause?”, the answer must be discourse-oriented, a position summed up by Fabricius-Hansen and Ramm in their introduction to a 2008 study on coordination and subordination:

What ‘is’ a sentence? As is well known, this question cannot be answered in a theory-independent and at the same time precise manner. From a syntactic point of view the sentence is the domain of syntactic theory: a unit that is governed by syntactic principles (constituent structure, dependency, government, binding, movement, etc.). From a discourse perspective a sentence might be a unit representing a single illocutionary act on the part of the speaker/reader and intended to be understood as such by the hearer/reader. Thus, it is a unit the reader/hearer is expected to process ‘in a single step’, assigning to it a communicative purpose and a content to be integrated into her/his representation of the current discourse, i.e. what corresponds to an utterance in spoken language. (my emphasis)

  • 12 A less marked status can be posited for but: instances of S.I. but outnumber S.I. and whatever the (...)

38In other words, the sentence boasts a discursive potential that the clause does not share: it corresponds to a higher discursive unit. The principle of “processing within a single step” echoes one of the main values of and as a discourse marker in speech (cf. grouping “specific events together in service of a more general point”), and also the general role of the coordinator for Blakemore and Carston (2005) (and requires “the conjuncts to be processed together”). This suggests that the most natural position for and in writing is intra-sentential rather than sentence-initial. Indeed, the lower frequency of S.I. and compared to intra-sentential and (Biber et al 1999; Hoarau-Gournay 1997) confirmed in the written corpus under examination here (see appendix) points to the marked status of S.I. and.12

39What therefore motivates prefacing a sentence with and? Let us first examine some particularly “marked” examples. In (8), borrowed from Sekali (2010) ─ an instance of direct speech ─ and introduces a nominal phrase, a segment lacking a subject and a finite verb, which would not normally appear in a separate sentence:

(8)

I see books, Harry, don’t you? I see hundreds of books. And not just any books, but first editions, even signed first editions. (Auster, The Brooklyn Follies, 207)

40And “promotes” the segment it introduces to sentence status. For Sekali, it acts as a counter to the punctuation which signals the end of the previous utterance, picking up on it in order to insert it into a “macro-utterance” ─ (i.e. “to be processed in a single step” according to Fabricius-Hansen and Ramm). The coordinator appears here within reported speech; similar interactive dynamics are present in the excerpt below, a humorous opening to the introduction of Baron’s academic textbook From Alphabet to Email:

(9)

“You didn’t write,” she chides.

Robin’s innocent retort: “I never learned how.”

In this imagined sequel to the familiar saga, the film “Robin and Marian” starkly captures the great linguistic divide between medieval and modern times in European-based cultures. Marian presupposes a twentieth-century view of the writer world (“Drop a line to let me know how you’re getting on”). Robin, a product of his times, makes no apology for being unable to write. And apologize he shouldn’t, for literacy in the Middle Ages was hardly widespread. Your average warrior or nobleman had no more use for reading or writing than for eating with silverware or regular bathing.

41Sentence-initial and is followed by a marked word order (he shouldn’t apologize > apologise he shouldn’t). Just as it would in (8), removal of the coordinator, regardless of the word order, creates a less natural, more marked sequence in (9a), whereas substituting it for intra-sentential and, by grouping the two sentences into one in (9b), appears more acceptable:

(9a)



[…] Robin, a product of his times, makes no apology for being unable to write. Apologize he shouldn’t/ He shouldn’t apologize, for literacy in the Middle Ages was hardly widespread. Your average warrior or nobleman had no more use for reading or writing than for eating with silverware or regular bathing.

(9b)

Robin, a product of his times, makes no apology for being unable to write, and apologize he shouldn’t, for literacy in the Middle Ages was hardly widespread.

  • 13 In this example, the desire to create informational focus justifies the marked word order. Emphasis (...)

42Rather than viewing S.I. and in (8) and (9) as the equivalent of other S.I. adverbial connectives (e.g. in addition, as well, etc.), it can be analysed as a re-elaboration (cf. Cotte 1996) of intra-sentential and. But why promote the and-prefaced segment to separate-sentence status? In order to confer informational prominence (the full stop is iconic of the pause which can occur in speech) and therefore increase its discursive potential. The segment is brought into sharper focus, providing a window into the rhetorical organisation of the text: in (9), and introduces the main point of the paragraph ─ that is, that literacy was not widespread in Medieval times.13 This specific usage of and is identified by Bell (2007: 189), in cases where “the elliptical privileges of intra-sentential coordination and are carried over to inter-sentential connections, hence increasing the importance of an otherwise equal last item of a list”.

43More generally, separate-sentence status, whatever the mode of conjunction, is synonymous with informational prominence, a point underlined by Matthiessen (2002: 272), who identifies a cline of “informational prominence and density” to distinguish intra-sentential from sentence linkage. Indeed, examples (8) and (9) are not typical in that most and-prefaced sentences cannot be “tagged on” to the previous sentence, and therefore cannot be analysed as the re-elaboration of an intra-sentential and. However, the separate-sentence status already confers a certain degree of informational prominence which, it can be argued, is heightened by the discrepancy created by the use of a typically intra-sentential mode of conjunction in S.I. position. Thus posited, S.I. and works to re-elaborate this time the informational focus ─ and the discursive potential ─ already inherent in the separate-sentence status. In so doing, it functions like a sign-post, in the same way it functions as a discourse marker in speech. A less marked example appears in (10), taken from the first chapter of another academic textbook, Eckert & McConnell-Ginet’s Language and Gender (Cambridge Press, 2003). Here, the authors discuss the way gender is constructed within society.

(10)











Everywhere we look, we see images of the perfect couple. (For a still compelling discussion of the construction of male and female in advertising along these lines, see Goffman, 1976.) They are heterosexual. He is taller, bigger, darker than her. They appear in poses in which he looks straight ahead, confident and direct; she looks down or off into the distance, often dreamily. Standing or sitting, she is lower than him, maybe leaning on him, maybe tucked under his arm, maybe looking up to him. And from the time they are very young, most kids have learned to desire that perfectly matched partner of the other sex. Girls develop a desire to look up at a boyfriend. A girl begins to see herself leaning against his shoulder, him having to lean down to kiss her, or to whisper in her ear. She learns to be scared so she can have him protect her; she learns to cry so he can dry her tears. […] (Eckert & McConnell Ginet, Language and Gender, 2003, p. 28)

44The opening sentence of the paragraph refers to “images of the perfect couple”, examples of which are then provided (e.g. heterosexual… he is taller… she looks down…). The sentence introduced by And actually marks a departure from this list of examples, to move on to the fact that such images create in children desires which are described in the subsequent sentences. If removal of the coordinator in (10a) does not make for as marked a sequence as in (9a), it does affect the reader’s processing of the segment.

(10a)










Everywhere we look, we see images of the perfect couple. (For a still compelling discussion of the construction of male and female in advertising along these lines, see Goffman, 1976.) They are heterosexual. He is taller, bigger, darker than her. They appear in poses in which he looks straight ahead, confident and direct; she looks down or off into the distance, often dreamily. Standing or sitting, she is lower than him, maybe leaning on him, maybe tucked under his arm, maybe looking up to him. From the time they are very young, most kids have learned to desire that perfectly matched partner of the other sex. Girls develop a desire to look up at a boyfriend. A girl begins to see herself leaning against his shoulder, him having to lean down to kiss her, or to whisper in her ear.

45In (10a), the change in orientation is not clear, at least not immediately. While the shift is mirrored by the change of topic after a sequence of topic continuity (they > he > they > she > she> most kids), we are first inclined to assign to the sentence the same rhetorical role as the preceding ones. A new paragraph break would clarify the shift. Rather than guaranteeing a tighter flow with the immediately adjacent sentence (which is the case in (8) and (9)), S.I. and actually has a separating effect, and we can identify the same potential for differentiation as observed in speech.

46For Bell, all instances of S.I. and serve to “simplify identification and processing”. He differentiates three uses: listing, signaling continuation/ developing arguments by topic development, and signaling a shift in authorial perspective. The latter two are motivated by the “need to separate out stages in argumentation” (Bell 2007: 193). I will now build on this categorization, to both account for examples not just in academic prose but in the other text types represented in the corpus, and also to place S.I. usage within the wider context of text and speech functions. As in the example of monologic speech, S.I. and proves to be more differentiating than associative, first at the level of discourse transitions, and secondly at the local level, between adjacent sentences.

4.3. Differentiating in discourse transitions

4.3.1. Ending a set

  • 14 Each example will be identified as belonging to a particular text type: academic article (aa); acad (...)

47Halliday and Hasan (1976) pinpoint a use of cohesive and which links “a series of points all contributing to one general argument” and creates a retrospective effect. Lapaire and Rotgé (1998: 307) speak about the dialectic of opening/closure in association with and, and underline its use to close a segment (p. 475). Dorgeloh (2004: 1776) speaks of “continuity created by the speaker” which “foreshadows a functional ‘turn’”, and for Bell (2007) the most common use of S.I. and in academic prose is “signaling the last item on a list”. Signaling the last sentence in a set is one of several common uses in the present corpus. This is the only instance in which and proves associative ─ that is, at a local level, to link adjacent sentences sharing a similar role ─ although final status equates with a certain alterity, and it can be argued that the main function is to wind the set up, foreground a transition and differentiate the set from the next. In the extract below, taken from an academic article,14 and introduces the last in a series of three questions (placed in italics, as in the other examples in this section), just before the sentence which concludes the paragraph:

(11)








[…] The UK’s language needs are similar to those of many other countries, but the recent history of language teaching in this country has been relatively encouraging. But as well as encouraging, it has been challenging because we have suddenly found ourselves in a new situation where doors that were previously closed are now open and we are invited to display our wares. This raises urgent questions: Which of our wares are relevant? What educational needs are relevant to us? And do we stand to gain from a closer relation to education? These are questions that I shall try to answer below. (aa, Hudson, 108)

48Similarly, in (12), and fronts the last of three direct quotations from Tony Blair:

(12)








Describing householders’ fear of opening the door to him as a Labour canvasser in the early 1980s in east London, he (Mr Blair) said focusing on crime was not “clever politics – just instinct.”

It was time to “recapture crime and law and order from rightwing politics”, he told an audience of community leaders in Somers Town, a deprived London ward near King’s Cross.

And he described the criminal justice system as “the public service most unfit for purpose” when his government took over in 1997. (The Guardian, 19/7/2004)

49And works retrospectively to unite not only the last two but the last three sentences – much like the way it is used within the sentence to unite elements within the same constituent structure. It acts here as a textual signpost, encouraging the reader to group the sentences together and prepare for the rhetorical change to come. The identical rhetorical status of the sentences in italics is sometimes mirrored by topic continuity (e.g. “he” ─ Blair in (12)), or reinforced, as noted by Bell, by parallel structure. In (13), and marks off a set of four sentences which each introduce a different character and provide a vignette of the “maraud” announced in the first sentence. Parallelism is created by repeated use of parentheses apposing additional information about each character.

(13)









They descended on Soho, swarming through the bars in a Friday night, tequila drinking, office workers’ maraud. Little Sharif Mumtas (features assistant) got separated from the others and was helped home by a kind man whom she married nine months later. Jeanie Geoffrey (assistant fashion editor) was bought a bottle of champagne by a man who declared she was ‘a goddess.’ Gabbi Henderson (health and beauty) had her bag stolen. And Ally Benn (recently appointed editor) clambered on a table in one of the livelier pubs in Wardour Street and danced like a mad thing until she fell off and sustained multiple fractures to her right foot. In other words, a great night. (M. Keyes, Sushi for Beginners, 10)

50And introduces the last of the four vignettes, which is the longest and perhaps the most amusing (observing the principle of end-focus, and “saving the best until last”). The sentence “In other words, a great night” concludes not without humour/irony (heightened by its concision and separate paragraph status), like a punch-line.

51When the set contains at least three sentences, Bell (2007: 188) notes that signaling finality is “more clearly seen”. However, I will argue that it is a necessary condition for recognizing such a value. Otherwise, in the case of only two sentences, the coordinator takes on a “furthermore” colouring, as will be demonstrated in section 4.4.

4.3.2. Prefacing a new set

52Conversely, And can preface the first sentence of a new set, to introduce for instance a new discourse topic:

(14)










From infancy, male and female children are interpreted differently, and interacted with differently. Experimental evidence suggests that adults’ perceptions of babies are affected by their beliefs about the babies’ sex. (...) Such judgments then enter into the way people interact with infants and small children. People handle infants more gently when they believe them to be female, more playfully when they believe them to be male. And they talk to them differently. Parents use more diminutives (kitty, doggie) when speaking to girls than to boys (Gleason et al. 1994), they use more inner state words (happy, sad) when speaking to girls (Ely et al. 1995). They use more direct prohibitives (don’t do that!) and more emphatic prohibitives (no! no! no!) to boys than to girls (Bellinger and Gleason 1982). […] (ab, Eckert & McConnell Ginet, 17)

53The transition is materialized via the paragraph break. When Biber et al. (1999) note the tendency for S.I. coordinators to occur at paragraph boundaries and the “marked effect” this creates, they no doubt have in mind examples such as (14), in which the coordinator serves to negotiate transitions by signaling differentiation between sets. Both paragraphs here deal with the way babies are treated differently depending on their sex, but while the scope of the first is more general (e.g. interpretation and interaction), the second relates directly to language. This is made explicit in the topic sentence (“talk”), which begins with and.

54Transitions do not necessarily coincide with a paragraph boundary ─ for example, (10) quoted earlier (“And from the time they are very young, most kids have learned to desire that perfectly matched partner of the other sex”), or this passage from Bradford’s introduction to stylistics:

(15)






The relation between classical philosophy/rhetoric and Saussurean linguistics is far more complicated than my brief comparison might suggest, but it is certain that Saussure makes explicit elements of the divisive issue of whether rhetoric is a potentially dangerous practice. And this leads us to a second problem: the relationship between language and literature. Plato has much to say about literature – which at the time consisted of poetry in its dramatic or narrative forms. […] (ab, Bradford, 8)

55The transition is made explicit by the clause “this leads us to…” As underlined in the discussion of monologic speech, the coordinator does not negotiate the transition alone; rather, it combines with other elements in context (e.g. a pause, a paragraph break, lexis indicating transition) to bring it into sharper focus: again, the coordinator is the result of re-elaboration. Instead of “stringing out” the transition as in (15), a more concise wording could be envisaged ─ e.g. “A second problem is the relationship between language and literature”, however, the reader would not have as much time/space to seize where the text is heading. Emphatic transitions make for a more reader-friendly text; the re-elaboration realised by the coordinator provides an extra layer of emphasis. Dorgeloh (2004: 1775-1776) identifies the way S.I. and helps to negotiate transitions between topics, and the following sequence (of academic prose) is borrowed from her discussion:

(16)








In other words, what is crucially lacking ─ just as it was lacking for Anderson and indeed for Erikson ─ is material from which inferences might be made, with some assurance of representativeness, about the patterns of social action that are of interest within particular collectivities. As Clubb has observed, the data from which historians work only rarely allow access to the subjective orientations of actors en masse, and inferences made in this respect from actual behavior tend always to be question begging. And Marshall, it should be said, like Anderson, sees the difficulty clearly enough. He acknowledges […] (example borrowed from Dorgeloh 2004)

56The sentence prefaced by and contains the textual anaphor “the difficulty”, referring back to the topic of the previous sentences (“what is crucially lacking…”). Again, a more concise wording would be possible (“Marshall, like Anderson, acknowledges…”). This example highlights the ambivalence of textual transitions, which are built on both similarity (e.g. the textual anaphor) and on difference: the sentence includes interpersonal elements (“should”/ “clearly enough”), which reflect a heightened degree of speaker involvement, on top of that already present, if we follow Bell’s reasoning, in the attention paid to make the text’s rhetorical organization clear.

57The textual anaphor in (16) provides an example of repetition, a common feature of transitions involving and ─ or rather, repetition of some part of the previous discourse realised in a different form, or with some variation. Bell (2007: 189) identifies “argumentative chains” to refer to the way and articulates the interplay between given and new information and “the way that arguments are derived from and built on prior arguments”. Similarly, Schriffin (2006: 32) notes that and specifically prefaces questions which “build upon” the most recent list-item (e.g. “And what does your son do?” follows a mention at the very end of the previous turn of the (other) speaker’s son.) Bells’ “chains” are most frequent in the humanities journals. Below, and introduces a nominalization (prices remain fixed > fixity of prices), which becomes the object of an appraisal by the writer (“a certain plausibility”). The coordinator introduces differentiation in what might otherwise be viewed as too repetitive:

(17)



A small country in the Ricardian model, then, cannot lose from fragmentation so long as prices of final goods remain fixed. And fixity of prices has a certain plausibility if the rest-of-world is integrated, as noted above. (North American Economics and Finance) (example borrowed from Bell 2007)

58And underscores the way transitions – and discourse more generally – move forward by building on what has gone before.

4.3.3 Prefacing a transition in mode

59And combines with repetition and an appraisal by the writer in the following press article, about a man who has decided to run across the U.S. in less than seventy days:

(18)







He was feeling breezy after a relatively clam 38-mile run but, somewhat surprisingly, confessed to not liking running “that much really”. Instead, he puts doing the 3,000 mile run down to the “Everest factor”. “I’m doing it because it is a huge challenge,” he explains.

And quite a challenge it is: although the scenery is stunning, desert temperatures give way to mountains and, where it is not safe to run on roads, competitors must make their way along dirt or gravel tracks. (The Guardian, 19/7/2004)

  • 15 Remember that repetition contributes to Jobs’ return to the floor after audience applause (“we’re g (...)

60The transition builds on previous content, modified by an appraisal (it is a huge challenge > quite a challenge it is), and marks the move from direct speech to an evaluation by the writer. Indeed, after the “intrusion” of another voice, S.I. and can here be likened to discourse marker and when it negotiates the speaker’s return to the floor.15 Halliday and Hasan identify a “transitional” and, in moves between narrative and direct speech in fiction. Examples abound in fiction and in the press, with and prefacing the switch either from speech to narrative, as in (19), or from narrative to direct speech, as in (20):

(19)





“What do you do?”
“I’m a longshoreman.”
“No, I mean really.”
“I mean really too.”
And he would have showed her his palms to prove it if he hadn’t been afraid she could tell the difference between calluses and blisters. (R. Yates, Revolutionary Road, p. 23)

(20)




Sex and the City star Sarah Jessica Parker watched the grand finale of her HBO series Sunday night, just like any other fan. And she was thrilled with what she saw.

“I’m extremely happy,” the actress, who played sex columnist Carrie Bradshaw for six seasons, said Monday. (USA Today, 25/2/2004)

61In this last example, and introduces a short sentence which could have been integrated into the previous one: while placing it in a separate sentence could simply signal finality, in the press it frequently precedes direct speech. It therefore not only helps us to anticipate (and therefore process) the transition. At the same time, it increases informational prominence and underlines the cause-consequence relation (cf. reaction: “she was thrilled”).

62Unlike examples presented in the previous two sections, the transition is not so much between sets of arguments which can be attributed to the writer, but between different voices, or even between different modes, again in fiction (cf. Bell talks about a “shift in authorial perspective”). In (21), and coincides with a shift from narrative to description:

(21)




I thought he would be more glad if I came upon him with his breakfast, in that unexpected manner, so I went forward softly and touched him on the shoulder. He instantly jumped up, and it was not the same man, but another man! And yet this man was dressed in coarse gray, too, and had a great iron on his leg, and was lame, and hoarse, and cold […] (C. Dickens, Great Expectations, 15)

4.4. Differentiating between adjacent sentences

4.4.1. Informational prominence

63In addition to the role it plays in transitions, S.I. and also assists the reader’s processing of the text by accentuating at a more linear level the difference between adjacent sentences. The first type involves the afore-mentioned contrast in informational prominence: and prefaces the second and last of two arguments, which receives more focus than the first. In (22), it links two sentences providing justification for the premise announced at the beginning of the paragraph: that is, that Ms Short’s new appointment does not appear to be a demotion, particularly in her eyes. In (23), and links two reasons why it is difficult for immigrants who are settled in Britain to return to their homeland (cf. But there is no going back now”).

(22)





The move (to make Clare Short minister for International Development) may have been intended as a demotion, but someone forgot to tell Ms Short. She kept her cabinet rank and, more importantly, a close relationship with Gordon Brown, the chancellor of the exchequer. And her willingness to speak unvarnished truths won her admirers in the development world, where polite euphemisms abound. (The Economist, 17/7/2004) (end of paragraph)

(23)





But there is no going back now. Given enough time, you can get a liking for anything – kebabs and rain, egg and two and the 38 bus. And a fashion for retiring to Jamaica – where an English pension would make possible a life of prosperous ease – has been diminished somewhat by the fact that a few have been murdered for their money by gangs, and many have been shunned as foreigners. Home is what rusts on. Home is what sticks. (essay, Rundle, 193)

64Compared to ternary or list-type units, binary structures are more contrastive — a point confirmed by other linguistic dichotomies such as topic/predicate or theme/focus. It is in this context that S.I. and most closely resembles the adverbials furthermore or moreover ─ for which it offers a “lighter” equivalent. In the extract below, the difference between verbal/sign languages and gestures is discussed. And links two justifications of the premise made in the first sentence (cf. “the different development trajectory”). Just as the and-prefaced sentence is far longer in (23), that of (24) gives rise to extra commentary (“This is striking in light of the fact that….”) and therefore carries more discursive weight.

(24)










The second issue has to do with the finding that the development trajectory followed by gestures is different not only from that followed by words, but also from that observed for signs. Thus, there was little, if any, growth over time in the size of gestural repertoire. And although Marco occasionally combined two representational gestures (unlike the monolingual children), two-gesture combinations never became productive in the way that two-sign combinations did. This is striking in light of the fact that gestures and signs are produced in the same modality, and it suggests that what distinguishes gestures from words is not simply modality of production. Words and signs belong to structured linguistic systems, while gestures do not, and this is what differentiates the development of gestures from that of words and signs. (aa, Capirci et al., 35-6)

65Here is a last example, from an academic article by Hudson, entitled “Why education needs linguistics (and vice versa)”, in which the and-prefaced sentence is again lengthier and carries more weight:

(25)







[…] Like all other countries, we need to learn foreign languages, though (like most other English-speaking countries) we dislike these at school level and postpone enthusiasm for adult language courses (Kelly & Jones 2003). In short, the context of language teaching is much the same in the UK as in other countries. And of course, as in all other countries, language supports education in a very specific way as the medium of instruction, the medium of testing, the medium of exercise and the medium of a great deal of thought. (aa, Hudson, p. 108) (end of paragraph)

4.4.2. Highlighting a build-up in the argumentation process

66And also confers informational focus in another context: to emphasise an essential, intermediate step in a reasoning process. This is the case for the coordinator analysed as inherently intra-sentential in section 4.2:

(9)



Robin, a product of his times, makes no apology for being unable to write. And apologize he shouldn’t, for literacy in the Middle Ages was hardly widespread. Your average warrior or nobleman had no more use for reading or writing than for eating with silverware or regular bathing.

67The content of the and-prefaced sentence constitutes the cornerstone of the argumentation, hence its separate sentence status. It provides both an appraisal of the content of the previous sentence, and gives rise to a move towards justification in that which follows. Unlike examples (22) to (25), a paraphrase with furthermore would be difficult: indeed comes more readily to mind. This type of usage is observed in any type of argumentation: that is, academic prose, or the essays, excluding fiction and the press. In (26), Brazil discusses the difficulty of systematising intonation patterns in order to present them in a book. The paragraph begins with of course, which foregrounds the concession (cf. but later in the paragraph). The two sentences linked by and constitute an embedded deduction which is conceded:

(26)







It is fair to say, of course, that the only research procedure available is to make tentative phonetic observations and try to associate them with generalisable meaning categories. And what applies to the discovery of the meaning system applies equally to its presentation in such as book as this. We have no alternative but to refer to the variables in pseudo-phonetic terms; but we have to stress that such a characterisation is no more than an approximation and a convenience. It is necessary always to preserve a certain distance from the phonetic fact. […] (ab, Brazil, 4)

68And can be glossed by indeed or so, marks a process of deduction, and leads on to further justification (“We have no alternative…”) before (intra-sentential) but announces the counter-argument. Again, and introduces the crux of the matter (the presentation of the book in hand), with maximum pertinence made all the more explicit by the emphatic wh- turn of phrase. A gloss via French “or” also comes to mind (for example, when it prefaces the intermediate stage of a syllogism). Equivalent dynamics underscore (27), taken from an essay entitled “In Distrust of Movements”, which contains an example of and repeated in consecutive sentences.

(27)









[…] Sharing a cultural heritage is more than “negotiating an identity” (1). It is perhaps time for those of us who have lost religious, linguistic, or cultural ties with our ancestors to admit to that and let go (2). Finally, and I think this goes to the heart of the matter, we should recognize that truth is not just a point of view (3). There are facts that are not made up but real (4). And to pretend there is no difference between fact and fiction, or that all writing is fiction, is to paralyze our capacity to distinguish truth from falsehood (5). And that is the worst betrayal of Primo Levi and all those who suffered in the past (6). For Levi’s fear was not that future generations would fail to share his pain, but that they would fail to recognize the truth (7). (essay, Buruma, 31) (end of paragraph)

69Sentence (3) introduces a claim (“we should recognise that truth is not just a point of view”), and the following three sentences linked by and make up an embedded, 3-step justification of this claim. They do not share the same rhetorical status: the coordinator emphasises this fact, and in each instance can be paraphrased by “so”. Removal of and would not compromise meaning, but the effect of the two-fold build-up would be lost. Again, and ‘separates out’ the content of the conjuncts in order to confer argumentative weight.

70In 4.4.1, and links sentences sharing a similar rhetorical purpose, but each worthy of specific attention, particularly the second, and hence guarantees that they remain distinct. Here, it calls attention to the separate rhetorical status of the sentences, which build on from one another, generally making up embedded deductions. Both cases reflect the potential of the coordinator to produce a more deliberate, motivated rhetoric ─ akin to its use in Jobs’ speech, and a strategy to ensure that nothing is lost on the reader.

4.4.3. Re-entering a set/picking up on an earlier discursive thread

71Finally, and dissociates the sentence it introduces from the previous one(s) by signalling re-entry into an earlier discourse segment. In fiction, for example, to return to description after a short prolepse (in the middle paragraph – “The time was to come…”):

(28)










The wine was red wine, and had stained the ground of the narrow street in the suburb of Saint Antoine, in Paris, where it was spilled. […] Those who had been greedy with the staves of the cask, had acquired a tigerish smear about the mouth; and one tall joker so besmirched, his head more out of a long squalid bag of a nightcap than in it, scrawled upon a wall with his finger dipped in muddy wine-lees – BLOOD.

The time was to come, when that wine too would be spilled on the street-stones, and when the stain of it would be red upon many there.

And now that the cloud settled on Saint Antoine, which a momentary gleam had driven from his sacred countenance, the darkness of it was heavy […] (C. Dickens, A Tale of Two Cities, 37-38)

72or to mark re-entry into the main line of argument after a side-comment in this extract of an academic article, in which the film adaptations of Jane Austen’s novels are discussed:

(29)








Then, too, the idealizing culture of Hollywood demands the wan perfection of a Gwyneth Paltrow in Douglas Mc Grath’s Emma or the manly vigor of a Colin Firth in Andrew Davies’s Pride and Prejudice. Neither may fit our imaginings of Jane Austen. And if cinema is idealistic, it is also famously realistic. Roger Michell makes his mud-stained, weather-beaten protagonists innocent of make-up in Persuasion, as if their imperfection guaranteed their “real” existence. […] But it is a sleight of hand to suggest that the authenticity of realism guarantees authenticity to Jane Austen. Thus, both the idealism and the realism of cinema make fidelity to her texts impossible. (aa, Harris, 46-47)

73Without the coordinator, we are tempted to analyse the sentence as a justification of what precedes (e.g. “Neither may fit our imaginings of Jane Austen. If cinema is idealistic, it is also famously realistic”). In contrast, and dissociates the two sentences and relegates the first to the status of side comment. Finally, and appears in the second paragraph of the beginning of the following essay. The first paragraph is reproduced below to illustrate how the same topic “Geoffrey Blainey’s great phrase ‘the tyranny of distance’” carries through both, appearing as the grammatical subject in the first sentence, and pronominalised by “it” (all italicised below) in all subsequent sentences, with the exception of that immediately preceding and:

(30)














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 (1). 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 (2).

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 (3). (Air transport had reduced travel time to a single day; satellite images were about to make every event on the globe instantaneously visible (4).)
And it was never quite true, even before technology changed forever our notions of distance and the globe. We live in feelings as well as in conditions and events. Distance is also measured by the heart. […] (essay, Malouf, 11)

74The sentence between round brackets equates with a departure from the main line of argument: it provides two justifications for the claim made in the previous clause (“by the time he formulated it the conditions it described had already changed”). Interestingly, it is the only sentence up to this point not to carry over the pronoun it, which reappears as the grammatical subject of the sentence fronted by and. The coordinator invites the reader to reach up higher into the text to retrieve the antecedent of it. At the same time, the sentence picks up from the line of argument left open after sentence (3), and the relation between the two can be glossed by ‘moreover’.

4.5. Imposing continuity (and + fronted item)

75A relatively common denominator in the written corpus is the combination < and + fronted item >, where an item precedes the grammatical subject. Among the examples quoted, and combines with an adverbial (e.g. 10: And from the time they are very young, most kids have learned …; e.g. 28: And now that the cloud settled on Saint Antoine, which a momentary gleam had driven from his sacred countenance, the darkness of it was heavy), a finite adverbial clause (e.g. 24: And although Marco occasionally combined two representational gestures… two-gesture combinations never became productive…; e.g. 29: And if cinema is idealistic, it is also famously realistic), an adverb (e.g. 21: And yet this man was dressed in coarse gray); disjunct + adverbial (e.g. 25: And of course, as in all other countries, language supports education). Fronted adverbials push the text along, setting up a new “discursive frame” often materialising a discourse boundary, and so it is natural to find them associated with and, whose role in transitions has been demonstrated. However, it can be argued that in this case, and is not dissociative ─ the dissociation is already created by the fronting of the other item ─ rather, it smoothes over the transition, guaranteeing continuity in the face of discontinuity. Bell (2007) observes that and is preferred over additive adverbials when accompanied by an enumerator (firstly, secondly) or a disjunct (e.g. critically) ─ no doubt, because and smoothes over the transitions realised by these fronted items.

76In academic articles 50% of the instances of S.I. and or but occur in combination with a fronted item ─ the highest rate among the different text types. In comparison, the figure is 25% in the essays, with academic textbooks and press articles situated in between these two extremes. More than elsewhere, a main motivation for S.I. coordination in academic articles is the cohesion they guarantee within textual transitions.

5. Conclusion

77This study has attempted to explore the link between speech and text usage as regards and. Originally, I was uniquely interested in sentence-initial and, in order to describe its functions, particularly to help non-native speakers in their writing (notably in an academic context), where difficulties do arise. Native speakers (or rather, writers) use S.I. and in many contexts (c.f. the variety of text types referred to here), in a way that is not gratuitous (c.f. the problems encountered if and is removed), and in a way that is probably quite intuitive. Indeed, they flout the “Don’t-start-a-sentence-with-and-or-but” rule because, in S.I. position, and serves a specific purpose: to help the reader navigate through the macro-structure/rhetoric, to assist in its processing, and make for a more reader-friendly text. In light of the structuring role of S.I. and, the parallel with discourse markers in speech became obvious, and with it the importance of (re)turning to speech ─ not only dialogue, which has received considerable attention, but monologic speech, a (new?) mode, of particular interest, due to its hybrid nature and therefore the theoretical junction it provides between speech and text. While space only allowed for the analysis of one (short) sample of monologic speech (which in itself proved complex), it can still be considered indicative.

78It is significant that if, in dialogue, discourse-marker and can be both associative and differentiating, and is predominantly differentiating in monologic speech and in S.I. position in text. As a discourse marker, it typically marks association between adjacent utterances, or differentiation between adjacent utterances and therefore between macro-utterances, including when it imposes continuity in the face of discontinuity, for example to negotiate the speaker’s return to the floor, to re-enter a higher set. In monologic speech, and signals differentiation between adjacent utterances, including re-entry into a set or the speaker’s return to the floor, and differentiation between sets. In writing, S.I. and differentiates between macro-utterances (in transitions), and between adjacent utterances (particularly in conferring informational prominence to the sentence it introduces). In all modes, use of and reflects a more motivated rhetoric, a text which is “definitely going somewhere” and is more persuasive. Finally, and highlights the tension between connection and disconnection, and between separation and unity, which characterizes the ambivalence of the status of the sentence as a unit within discourse.

Top of page

Bibliography

Alter, R., 1999. L’art du récit biblique. Brussels : Lessius.

Bell, D., 2007. ‘Sentence-initial And and But in Academic Writing’. Pragmatics 17:2, 183-201.

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

Blakemore, D. & Carston, R., 2005. ‘The Pragmatics of Sentential Coordination with and’. Lingua 115, 569-589.

Bolton, K., Nelson, G. & Hung, J., 2002. ‘A Corpus-Based Study of Connectors in Student Writing’. International Journal of Corpus Linguistics 7 (2), 165-182.

Clachar, A., 2003. ‘Paratactic Conjunctions in Creole Speakers’ and ESL Learners’ Academic Writing’. World Englishes 22 (3), 271-289.

Cotte, P., 1996. L’Explication grammaticale de textes anglais. Paris : PUF.

Dorgeloh, H., 2004. ‘Conjunction in Sentence and Discourse: Sentence-Initial and and Discourse Structure’. Journal of Pragmatics 36, 1761-1779.

Fabricius-Hansen, C. & Ramm, W., 2008. ‘Subordination’ versus ‘Coordination’ in Sentence and Text: A cross-linguistic perspective. Amsterdam: John Benjamins.

Gournay-Hoarau, L., 1997. Linguistique contrastive et traduction: étude contrastive de la coordination en français et en anglais. Paris : Ophrys.

Gournay, L., 2006. Approche énonciative des catégories de marqueurs. Document de synthèse en vue de l’HDR. Université Paris 7-Denis Diderot.

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

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

Hasselgård, H., 2004. ‘The Role of Multiple Themes in English’. Pragmatics and Beyond 120, 65-87.

Hoey, M., 2001. Textual Interaction: An Introduction to Written Discourse Analysis. London: Routledge.

Kitis, E., 2000. ‘Connectives and Frame Theory: the Case of Hypotextual Antinomial and’. Pragmatics and Cognition 8 (2), 357-409.

Lakoff, G., Peters, S., 1966. ‘Phrasal Conjunction and Symmetric Predicates’. The Computation Laboratory of Harvard University Mathematical Linguistics and Automatic Translation, Report No. NSF-17. 

Lakoff, R., 1971. ‘If’s, And’s and But’s about Conjunction’. In: Fillmore, C.J. & Langendoen, D.T. (Eds.) Studies in Linguistic Semantics. New York: Holt, Rinehart & Wilson, 115-149.

Lapaire, J.-R., 2005.  ‘Coordination et Cognition’. Études Anglaises, 58, 473-494.

Lapaire, J.-R & Rotgé, W., (1998) Linguistique et grammaire de l’anglais. Toulouse : Presses universitaires du Mirail.

Lee, S.-W., 2005., Information Distribution in Native and Non-Native University Student Writing. Ph.D. dissertation. University of Melbourne.

Lyons, J., 1977. Semantics. 2 vols. Cambridge: CUP.

Lyotard, J.-F., 1983. Le différend. Paris : Éditions de Minuit.

Martin, J., 2001. ‘Cohesion and Texture’. In: Schriffin, D., Tannen, D., Hamilton, H. (Eds.) The Handbook of Discourse Analysis. London: Blackwell, 35-53.

Matthiessen, C. & Thompson, S., 1988. ‘The Structure of Discourse and ‘Subordination’’. In: Haiman, J. & Thompson, S. (Eds.) Clause Combining in Grammar and Discourse. Amsterdam: John Benjamins.

Matthiessen, C., 2002. ‘Combining Clauses into Clause Complexes: A Multi-Faceted View’. In; Noonan, L., Thompson S. & Bybee. J. (Eds.) Complex Sentences in Grammar and Discourse: Essays in honor of Sandra Thompson. Amsterdam: John Benjamins, , 235-319.

Morgenstern, A. & Sekali, M., 2009. ‘What Can Child Language Tell Us about Prepositions ? A Contrastive Corpus-Based Study of Cognitive and Social-Pragmatic Factors’. In: Zlatev, J., Andrén, M., Johansson, M. & Falck, C. (Eds.) Studies in Language and Cognition. Newcastle upon Tyne: Cambridge Scholars Publishing, 261-275.

Ong, W., 1982. Orality and Literacy: The Technologizing of the Word. London: Methuen.

Redeker, G., 2006. ‘Discourse Markers as Attentional Cues at Discourse Transitions’. In Fischer, K. (ed.) Approaches to Discourse Particles. Amsterdam: Elsevier, 315-338.

Schriffin, D., 1986. Functions of and in Discourse. Journal of Pragmatics 10, 41-66.

Schriffin, D., 1987. Discourse Markers. Cambridge: CUP.

Schriffin, D., 2006. ‘Discourse Marker Research and Theory : Revisiting ‘and’’. In: Fischer, K. (Ed.) Approaches to Discourse Particles. Amsterdam: Elsevier, 315-338.

Sekali, M., 2010. ‘Coordination et dynamique discursive: étude comparative des coordonnants anglais AND, OR, BUT et FOR’. In Florea, Papahagi, Pop, & Curea (Eds), Directions actuelles en linguistique du texte, Casa Cărţii de Ştiinţă, Cluj-Napoca.

Thompson, G., Zhou, J., 2000. ‘Evaluation and Organization in Text: The Structuring Role of Evaluative Disjuncts’. In: Thompson, G., Hunston. S. (Eds.) Evaluation in Text: Authorial Stance and the Construction of Discourse. Oxford: OUP, 121-141.

Van Dijk, T., 1979. ‘Pragmatic Connectives’. Journal of Pragmatics 3 (2), 447-456.

Top of page

Annex

Appendix A: Corpus

Academic articles

Allen, C., 2003. “Deflexion and the Development of the Genitive in English.” English Language and Linguistics 7, 1-28.

Bauschatz, P., 2003. “Rhyme and the Structure of English Consonants.” English Language and Linguistics 7, 29-56.

Boyd, J., 2001. “Virtual Orality: How Ebay Controls Auctions without an Auctioneer’s Voice.” American Speech. 76 (3), 286-300.

Capirci, O., Iverson, J.M., Montanari, & S., Volterra, V., 2002. “Gestural, Signed and Spoken Language Development: The Role of Linguistic Input.” Bilingualism: Language and Cognition 5 (1), 25-37.

Carr, P., 2004. “The Recent History of Linguistic Relativity.” The European English Messenger. 13 (2),17-22.

Fergus, J., 2003. “Two Mansfield Parks: Purist and Postmodern.” In: Macdonald, G., Macdonald, A.F. (Eds.) Jane Austen on Screen. Cambridge, CUP, 69-89.

Harris, J., 2003. “Such a Transformation: Translation, Imitation, and Intertextuality in Jane Austen on Screen.” In: Macdonald, G., Macdonald, A.F. (Eds.) Jane Austen on Screen. Cambridge: CUP, 44-68.

Holloway King, T. & Darrymple, M., 2004. “Determiner Agreement and Noun Conjunction.” Journal of Linguistics 40, 69-104.

Hudson, R., 2004. “Why Education Needs Linguistics (and vice-versa).” Journal of Linguistics 40, 105-130.

Montrul, S., 2002. “Incomplete Acquisition and Attrition of Spanish Tense/Aspect Distinctions in Adult Bilinguals.” Bilingualism: Language and Cognition. 5 (1), 39-68.

Nagy, N., 2001. “Live Free or Die’ as a Linguistic Principle.” American Speech. 76 (1): 30-41.

Waksler, R., 2001. “A New all in Conversation.” American Speech 76 (2), 128-138.

Chapters from academic books

Baron, N.S., 2000. Alphabet to Email: How Written English Evolved and Where it’s Heading. London: Routledge. Chapter 1 “Robin Hood’s retort”,1-25.

Bradford, R., 1997. Stylistics. London: Routledge. Chapter 1 “Rhetoric”; chapter 2 “Textualism 1: Poetry”, 3-37.

Brazil, D., 1997. The Communicative Value of Intonation in English. Cambridge: CUP. Chapter 1 “A Preliminary Sketch of the Tone Unit”, 1-20.

Corbett, E. P. J., (1965) 1990. Classical Rhetoric for the Modern Student. Oxford: OUP. “Introduction”, 3-31.

Crystal, D., 1997. English as a Global Language. Cambridge: CUP. Chapter 1 “Why a Global Language?”, 1-28.

Donnelly, C., 1994. Linguistics for Writers. New York: State University of New York Press. Chapter 1 “Language and Linguistics”, 1-18.

Eckert, P., & McConnell-Ginet, S., 2003. Language and Gender. Cambridge: CUP. Chapter 1 “Constructing Gender”, 9-32.

Hoey, M., 2001. Textual Interaction: An Introduction to Written Discourse Analysis. London: Routledge. Chapter 1 “What to Expect and what Not to Expect”, 1-10.

Yule, G., 1985. The Study of Language. Cambridge: CUP. Chapter 1 “The Origins of Language”; chapter 2 “The Development of Writing”; chapter 3 “The Properties of Language”, 1-29.

Essays

Aciman, A., 2000. “The Last Time I Saw Paris.” In: Atwan, R. (Ed.) The Best American Essays 2000. New York: Houghton Mifflin, 1-12.

Berry, W., 2000. “In Distrust of Movements.” In: Atwan, R. (Ed.) The Best American Essays 2000. New York: Houghton Mifflin, 13-19.

Blainey, G., 2000. “Globalisation: Unpacking the Suitcase.” In: Craven, P. (Ed.) The Best Australian Essays 2000. Melbourne: Black Inc, 34-42.

Buruma, I., 2000. “The Joys and Perils of Victimhood.” In: Atwan, R. (Ed.) The Best American Essays 2000. New York: Houghton Mifflin, pp.20-31.

Gass, W. H., 2000. “In Defense of the Book.” In: Atwan, R. (Ed.) The Best American Essays 2000. New York: Houghton Mifflin, 52- 63.

Gordon, M., 2000. “Rome: The Visible City.” In: Atwan, R. (Ed.) The Best American Essays 2000. New York: Houghton Mifflin, 64-79.

James, C., 2001. “The All of Orwell.” In: James, J. Reliable Essays. London: Picador, 3-24.

Malouf, D., 2000. “The People’s Judgment.” In: Craven, P. (Ed) The Best Australian Essays 2000. Melbourne: Black Inc, 11- 14.

Rundle, G., 2000. “A Town Called Hackney Nation.” In: Craven, P. (Ed) The Best Australian Essays 2000. Melbourne: Black Inc, 187-196.

Waterford, J., 2000. “Capital Letters.” In: Craven, P. (Ed) The Best Australian Essays 2000. Melbourne: Black Inc, 136-145.

Press articles taken from the following:

The London Times

The Guardian

The International Herald Tribune

USA Today

The Sydney Morning Herald

The Australian

The Economist

Corpus 2 (Fiction):

Atwood, M., (1996), 1997. Alias Grace. London: Virago.

Atwood, M., (2000), 2001. The Blind Assassin. New York: Anchor.

Austen, J., (1814), 1994. Mansfield Park. London: Penguin.

Bryson, B., 2000. Down Under. London: Black Swan.

Dickens, C., (1861), 1992. Great Expectations. Ware, Hertfordshire: Wordsworth.

Dickens, C., (1859), 1994. A Tale of Two Cities. London: Penguin.

Du Maurier, D., (1936), 2003. Jamaica Inn. London: Virago.

Hemingway, E., (1939), 1993. The First Forty-Nine Stories. London: Arrow.

Keye, M., (2000), 2001. Sushi for Beginners. London: Penguin.

Winton, T., 2008. Breath. London: Picador.

Winton, T., 2004. The Turning. London: Picador.

Yates, R., (1961), 2009. Revolutionary Road. London: Vintage.

Appendix B: Table: Instances of and and but in different text types

The table presents the frequency of and and but in each text type, both in sentence-initial (S.I.) and intra-sentential (I.S.) positions. Intra-sentential coordinators accounted for here link finite, non-embedded clauses, which could potentially stand in a separate sentence. The statistics exclude coordinators featuring within quoted direct speech. The length of texts is not equivalent and so the figures can only be interpreted comparatively within each category of text type. For example, the frequency of coordinators is not necessarily lower in the newspapers than elsewhere. Rather, the newspaper articles are short texts and contain a lower number of words than the texts representing the other genres.

Top of page

Notes

1 Schriffin is careful to underline that the coordinator displays, or brings into sharper focus, discourse structure, but does not create it: “and has a structural role only because speaker and hearer are able to interpret textual material well beyond that of and itself, including the overall pattern of textual connection, and the content and structure of ideas within the surrounding discourse. This dependency between and and its text means that and does not in itself create the idea structures which comprise a text; rather, and displays those idea structures” (Schriffin 1986: 55). This concords with the notion of context-driven meaning defended by Gournay (2006: 142-3) about coordination in general: “(les coordonnants) expriment des relations construites dans le discours”─ “coordinators express relations constructed within discourse” (my translation).

2 « Être capable d’associer, c’est d’abord être capable de dissocier. Tout coordonnant se charge de le rappeler en agissant comme un démarcateur ou un différenciateur » – Association requires dissociation. Coordinators remind us of this fact, and function as a boundary marker or differentiator (my translation; Lapaire 2005 : 475).

3 Similarly, in her discussion of discourse markers as “attentional cues”, Redeker (2006) includes and in her examples of “pop-markers”, which signal “the return from a parenthetical segment” – as opposed to “push-markers”, which signal the beginning of a parenthetical segment. She also notes that “the ‘return function’ can be “further expressed” with stressed pronominal reference, indicating that a previously introduced referent has not been active in the immediate context and must be retrieved from the discourse record”.

4 The transcript is mine, based on a recording of the presentation, as no official transcript exists (on the Apple website or in any other media). The slashes (/) indicate the boundaries between units recognised as clause complexes in the Systemic Functional tradition (e.g. Matthiessen 2002) or segments which, while not syntactically independent, have been dissociated from the left-hand context by a relatively long initial pause (e.g. “Not a little bit thinner, a third thinner”). These units will be referred to as “utterances”. Space prevents me here from rendering paralinguistic components which contribute to meaning and which deserve a study of their own, such as body language and the interplay with the visual mode (e.g. when Jobs turns and points to his Powerpoint presentation), or interaction with the audience (e.g. when he pauses while the audience applauds).

5 At the same time, removal of the coordination can produce the effect of arguments being thrown like ammunition at the listener, particularly directly before the personal pronoun you in “you might not think that’s a lot but…” The combination and + you of the original makes for a less aggressive tone.

6 In contrast, in a corpus of televised interviews (dialogic speech, i.e. featuring turn taking between interviewer and interviewee). I have noted that and is generally pronounced with a reduced vowel. The handful of examples of full-vowel pronunciation induce a more emphatic “can-you-believe it?” reading.

7 Pausing cannot be the unique criterion on which to base a discourse boundary: Jobs makes numerous “dramatic pauses” (of two or three seconds in length), and these sometimes accompany a shift from the linguistic to the visual medium (when he turns to present a new slide) or result from (or encourage) applause by the audience.

8 The National Teacher and Principal Recruitment Act. Speech given March 7 2001 (source: http://clinton.senate.gov – July 2003)

9 Stigmatization may also result from the fact that, when compared with subordination, coordination is considered “simplistic” because it corresponds to the more primitive “oral” form. This is compounded in the case of the additive coordinator which, as noted by Sekali (2010: 236), some regard as “empty of meaning”. Hence the preference in analytical prose for linking by the more explicit means provided by adverbial connectives (e.g. however, furthermore, in addition).

10 A corpus of fiction (the first 30 pages of 12 works) was included at a later stage of this study. Excluding direct speech, it contains 29 examples of S.I. and (a relatively low number, on a par with the frequency of S.I. and in the essays, which contain fewer words). Two novels (Austen’s Mansfield Park, and Atwood’s Blind Assassin) contain no S.I. and outside usage in direct speech. Stylistic variation aside (e.g. Hemingway), the highest frequency of S.I. and is not to be found in fiction.

11 « Chaque phrase est en principe l’enjeu d’un différend entre des genres de discours, quel que soit son régime. Ce différend procède de la question : Comment l’enchaîner ? qui accompagne une phrase. Et cette question procède du néant qui « sépare » cette phrase de la suivante » ─ Each sentence plays out the difference between discourse genres, whatever the type, and provides an answer to the problem of linkage, bridging the divide created by the nothingness which “separates” the sentence from the next (my translation), Le différend, J.M. Lyotard, 1983, Éditions de Minuit, p. 200.

12 A less marked status can be posited for but: instances of S.I. but outnumber S.I. and whatever the context (Biber et al 1999; Gournay-Hoarau 1997; Bell 2007), and in some cases, outnumber intra-sentential but. Unlike the notion of addition, it can be argued that the adversative relation lends itself to S.I. position, with the values of opposition or contrast enhanced by the sentence break.

13 In this example, the desire to create informational focus justifies the marked word order. Emphasis is also created due to the repetition of the lexical verb apologize in such a short interval, which contributes to the forced character of 2a and 2b (compare with: Robin, a product of his times, makes no apology for being unable to write. He shouldn’t have to, for literacy in the Middle Ages was hardly widespread.)

14 Each example will be identified as belonging to a particular text type: academic article (aa); academic book (ab); essay. For press articles and fiction, the title of the publication/work is given directly.

15 Remember that repetition contributes to Jobs’ return to the floor after audience applause (“we’re going to be shipping white from day 1 / and to give you some scale, this is what it looks like”).

Top of page

List of illustrations

URL http://anglophonia.revues.org/docannexe/image/234/img-1.png
File image/png, 27k
Top of page

References

Bibliographical reference

Fiona Rossette, « And-Prefaced Utterances: From Speech to Text », Anglophonia/Sigma, 17 (34) | 2013, 105-135.

Electronic reference

Fiona Rossette, « And-Prefaced Utterances: From Speech to Text », Anglophonia/Sigma [Online], 17 (34) | 2013, Online since 10 December 2013, connection on 21 August 2017. URL : http://anglophonia.revues.org/234 ; DOI : 10.4000/anglophonia.234

Top of page

About the author

Fiona Rossette

Université Paris X (EA n°370- CREA)

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