Mismatches, approximations, mistakes and misunderstandings: the common group of evils likely to impede on communication; without counting blurred meanings, vagueness, or ambiguity. Communication, particularly oral, in theory would require several repetitions and tests in advance to reach a real level of correctness – or, at least, a harmony between what enunciators say and mean. However, situations where a locutor can erase the first formulation of his or her thought and substitute it with a more developed version are rare. This type of vertical re-elaboration (relative to the paradigmatic axis, we could say) is specific to writers, or speakers of different backgrounds (journalists, politicians, teachers, to cite but a few), who prepare speech before its actual enunciation. They benefit from a kind of preparation “laboratory”. Correlatively, in these precise cases, there is a delay between the time taken to develop the speech and the actual moment of enunciation. The situation of communication is therefore artificial and relatively unidirectional because there is no interaction, so to speak, with an interlocutor: only a “receiver” is targeted. But an ordinary locutor, in concrete situations of everyday life, generally does not have this possibility to prepare in advance his or her words, particularly when expressing him or herself orally. What we could describe as the “constraints of direct speech” is particularly relevant as we rarely speak alone: the interaction with an interlocutor does not allow for vertical re-elaboration, and only horizontal “readjustments”, that is to say only variations, revisions and adaptations are possible. These readjustments are necessary due to the frequent difficulties of putting our thoughts into words in a satisfactory way. The readjustments in question seem especially necessary as interlocutors rarely share the same mental representations. And this goes for words themselves, which are far from always being understood or receiving the same connotations.

Failure in communication is, for that matter, a favored theme in the world of certain writers. Pinter’s theatre (Night School¹, or Betrayal²) in particular approaches the subject. It is also the case in Kundera’s novels: in The Unbearable Lightness of Being³, for example, the author illustrates this irrevocable “incommunicability” through his “dictionary of misunderstood words”. Here, he approaches the question of the difference that separates individuals in their perception of words and things. More precisely, he presents these gaps as a set of dictionary entries, with completely distinct definitions according to the characters. To resume the message of the work, Kundera adopts the following wording, touching on the misunderstandings that he ties together in his characters’ dialogues: “although they had a clear understanding of the logical meaning of the words they exchanged, they failed to hear the semantic susurrus of the river flowing through them.”⁴ In the novel, this divide in perceptions creates a wedge between the characters, a void where misunderstanding makes a nest for itself. Such is, precisely, the risk that we run when we look to interact with others, even more so when we know little about our interlocutor, his or her perceptions and lexical universe. Indeed, sharing the same language and culture is not enough to rule out misunderstandings.

It seems, nonetheless, that we are not condemned to accepting (or at least, not completely) these problems of misunderstanding, and that languages themselves do offer locutors ways of trying to fill these voids and overcome communication failures. And so, if understanding is often put into peril, it is not necessarily a failure either. Communication dead ends would not, then, be a fatality. It is at least possible to reduce the effects of this on-going difficulty to communicate accurately. By observing languages, we realize that they offer the possibility to break away momentarily from discursive linearity⁵ and make way for reflexivity, to come back to our own words to amend and sharpen them, to overturn them, and make changes to the wording or enunciative perspective. All these reworkings, following an already uttered segment, are what we call discourse readjustments.

The term “readjustment” (or its french equivalent) is occasionally employed by Gilbert [GIL 89, p. 40]⁶, or Lapaire and Rotgé [LAP 98, p. 319]⁷. It is also mentioned by Ranger [RAN 12, p. 39]. In the true sense of the term, as in the metalinguistic sense⁸ used in this study, readjustment is a means of adaptation: it is a question of making an element conform with a value taken as a target. The term readjustment of course resonates with the term adjustment, which is at the center of the Theory of Enunciative Operations, and we will specify why it is useful to distinguish the two concepts. The two terms have, in any case, the common factor that they presuppose a lack of harmony, or leeway, or play (meaning lack of juncture, leaving a gap or hole, as Culioli points out [CUL 99a, p. 98]) between two elements. With regard to readjustment, this gap is manifested between an actual formulation and a targeted formulation. Bridging this gap is part of a dynamic where various markers or constructions intervene, which this study aims to examine. We will focus on oral as well as written communication. The latter is not characterized by the constraints linked to spontaneity that we previously mentioned with regard to oral communication. Yet, we will see that readjustments are also frequent here, and not only in cases where writing retranscribes or mimics orality: this frequency can first of all be explained by the fact that certain forms of writing remain quite spontaneous and are not worked on to the point of being able to evoke referents in one single description perfectly. Moreover, even in cases of perfected writing, referents often need to be characterized by successive reworkings so as to be evoked faithfully, in keeping with their complexity. Sometimes, in writing, readjustment also becomes a stylistic element, and seeks to reproduce and expose the stream of thought underlying the utterance. But the latter is far from being direct and linear in the majority of cases.

In terms of structure, in Part 1 this study will aim to define more accurately the notion of readjustment, understood here at the discursive level as reworkings on the basis of a first formulation. We will set up this notion in contrast to adjustment, employed (regarding a work on notions) in Culioli’s Theory of Enunciative Operations (henceforth TEO)⁹. We will then move onto a wider framework, mixing enunciation and pragmatics, given that our subject will be made up of discursive sequences. Still in Part 1, we will examine the reasons underlying the use of such readjustment processes in speech, as well as the modalities that characterize them. The sections that follow will aim to clarify the concrete manifestations of discourse readjustments. They will group together linguistic phenomena according to the types of speech acts that they allow. They will also go on to analyze in detail the introducers linked to these phenomena. Part 2 will examine the case of reformulation, whether paraphrastic or non-paraphrastic. We will move back and forth between the questions around accuracy in nomination and self-correction. Part 3 will focus on the phenomena of re-examination in the form of recentering, upgrading and downgrading processes, to show the extent to which the point of view itself can be reworked. The enunciative stance is, for that matter, sometimes modified to such a point that a change in enunciative perspective ensues, as is the case in distancing processes, which we will explore in Part 4. We will then leave these phenomena that reflect the existence of play in language to analyze, throughout Part 5, those that bear witness to this play on language. The use of segments that we will call “inserts”, that is to say, metalinguistic expressions, parenthenticals or even added-in structures with dialogical characteristics, will highlight these possibilities at the heart of language and revealed in speech. The last part of this work will concentrate on phenomena that interest us on a microdiscursive as well as macrostructural level, as we will examine readjustments that are characteristic of oral speech, and that allow speech to be (re)structured depending on the co-enunciators.

This work does not claim to be exhaustive. Nevertheless, it seeks to give a global view of these phenomena, which we can group under the label “discourse readjustments”. Whilst the starting point is onomasiological (based on the very notion of readjustment), the essential bulk of the work will then be more semasiological (meaning, that it will concentrate on the linguistic forms and discursive devices associated with that notion, which will then shed light on the notion itself), so that the reader may come to an idea of the diversity of the processes concerned.

Relatively few works have adopted such an approach. We can, however, identify two categories of connected works. On the one hand, we can find studies of the notion of adjustment (that we will contrast with the notion of readjustment) but these are still limited in number. Beyond the first references to adjustment found in Culioli (in a very specific form, since it deals with intersubjective adjustment), one major publication has approached the question explicitly: L’ajustement dans la TOE d’Antoine Culioli (Adjustment in Antoine Culioli’s TEO), a collective work whose articles were brought together and presented in detail by Filippi-Deswelle [FIL 12]¹⁰. We will give particular attention to an article by Ranger [RAN 12], as it paves the way for a connection between adjustment and readjustment phenomena. We may also note that the concluding article in this work – an article written by Filippi-Deswelle – also brings the two notions of adjustment and readjustment together. The author specifies that “when the enunciator makes him or herself not only producer but also interpreter, and attracts attention either to his or her own words or another’s […] this will enable “readjustments” where necessary”. The linguist also adds that “the study of such ‘readjustments’ has yet to be done”. In parallel to this work, we may also mention an article by Celle [CEL 09], which approaches adjustment from an “inter-enunciative” perspective, or furthermore an article by Deléchelle [DEL 11], which focuses on discourse readjustments. The work by Lebaud and Paulin [LEB 16] examines adjustment phenomena in relation to questions of variation.

On the other hand, works on some of the phenomena gathered in this study do exist but are addressed in isolation by other authors, without being bound to notions of adjustment or “readjustment”. Studies on reformulation or certain uses of discourse markers particularly come to mind. Major works on reformulation focus on French (sometimes in contrast with Italian), and here we can cite Rossari [ROS 97] or Roulet [ROU 87]. This study will highlight the existing common points (in respect to their differences, evidently), at the level of discourse organization, between the phenomena of reformulation and other processes using discourse markers. Our questions will join other texts exploring these issues (on the English language, this time) by researchers such as Aijmer [AIJ 02] or Schiffrin [SCH 87]. The interpersonal dimension of the markers in question has often been pointed out, but we will try here – in line with this dimension – to show what their role is in the field of discourse readjustments.

This work is part of a perspective that is complementary to the previously cited studies. It presents the phenomena in their diversity and aims to systematize them. With this study aiming to link together the enunciative¹¹ and pragmatic frameworks, we will set ourselves the goal of specifically examining the markers or constructions upon which the phenomena are based. Their effects, and particularly in relation to intersubjectivity, will be systematically linked to the values underlying their usage. The enunciative dimension of this approach can – at a first glance – seem paradoxical because the study is focused primarily on linking devices and markers connecting utterances, as well as discourse relationships, knowing that these objects are prototypical of discourse analysis and pragmatics. This being said, the paradox is removed when we consider that the enunciative approach corresponds to a method of analysis and not the subject of this study. The goal is to bring out the fundamental values of the markers or fixed phrases¹² that will be examined, that is to say, extracting the stable elements which characterize them, beyond the plasticity presented by their contextual interpretation. The field of adjustments (and readjustments), as highlighted by Mélis [MEL 12], is the ultimate connection point between enunciation and pragmatics, to the extent where he places the phenomena of regulation (and therefore interpretation) at the heart of the analysis. This is, in short, the issue of co-enunciative relationships that are fundamental to the theories of enunciation but also close to the questions dealt with in pragmatic frameworks, which will be the focus of this study. Let us remember that, according to Anscombre and Ducrot [ANS 76], pragmatics which deals with the use of sentences, and semantics, which discusses their meaning, both focus on what, in the meaning of a sentence, is linked to its use in discourse. The theories of enunciation, which reflect on syntax and semantics, establish links with pragmatics when they study linguistic markers whose value consists of directing the very interpretation of the utterance¹³. This study will precisely explore linguistic forms “whose meaning is pragmatic rather than descriptive”, to use a phrase by Récanati [REC 81, p. 29]. This being so, the analysis will be rooted each time in an enunciative perspective, considering that the markers used are the traces of mental operations.

With regard to terminology, we will henceforth use the term utterance rather than sentence. We will also use the term enunciator. We consider, actually – and this is really the basis of our approach – that an utterance consists of a set of markers, or viewpoints originating from the producer of the utterance, whom we should consider as an enunciator (and not only a locutor). The term enunciator does not only refer to a person in a situation where words are exchanged, but to an enunciative origin, who locates the utterances and thus constructs a system of referential values associated with the locutor, with the time and place of enunciation. Time, aspects, and pronoun references are calculated in relation to this enunciative origin. But the enunciator also marks a subjective origin, constructing the utterance and functioning as a source of intersubjective relationships. In correlation, the term used for interlocutor will be co-enunciator. An enunciator produces, indeed, utterances with the co-enunciator(s)’s interpretation in mind, depending on the perspectives that he or she has of them. In TEO, for that matter, the expression “semiotic loop” is used to describe this phenomenon. Firstly, as Mélis reminds us [MEL 12, p. 65], it is a matter of “producing content which is meant to be interpreted (by someone else), rather than content which is just to be expressed (the manifestation of latent content which would exist prior to speaking)”. But the phenomenon does not stop there. Culioli [CUL 03, p. 144] specifies – hence describing it as a loop – that “if I produce text, I must produce a text as it is recognized by another subject as having been produced in order to be recognized as interpretable”. It is for this reason that the recipient of the utterance is described as the “co-enunciator”. Indeed, he or she is not only the recipient of transmitted content, but well and truly takes part in constructing the utterance. Correlatively, the co-enunciator is an abstract representation of the interlocutor. This is taken into account from the moment the utterance is formed. Such a denomination seems all the more significant when the objective is to study readjustment phenomena. Nonetheless, this terminology does not presuppose that we will stay strictly in the field of enunciation: the enunciation/discourse/pragmatics trio will, in fact, be used to strive to describe, in the most comprehensive way possible, these readjustment phenomena. Moreover, we will use the terms locutor and interlocutor when we seek to interpret specific examples, bringing in real participants within any given dialogue. We prefer to reserve the terms enunciator and co-enunciator for the phases focusing on the conceptualization of the processes being studied, respecting the fact that we are dealing with theoretical constructs and that these terms do not designate people in flesh and bone.

In addition, this book is based on the analysis of authentic examples of contemporary English, whether British, American, or occasionally Canadian English. Corpora used are mainly the British National Corpus (BNC)¹⁴, the Corpus of Contemporary American English (COCA)¹⁵, the Freiburg Lancaster-Oslo-Bergen Corpus (FLOB)¹⁶, the London Lund Corpus (LLC)¹⁷, as well as extracts from novels or short stories. We will not question the fact that certain differences may exist in terms of distribution of the markers that we will be studying, depending on the variety of English used. We will, for that matter, mention the cases where significant differences can arise. The objective is not, however, to perform a sociolinguistic analysis but to identify the values underlying the use of the forms in question, in order to better understand their contribution to the English language, as well as their input in terms of enunciative and communicational strategies. Our analyses will be part of a mainly synchronic perimeter, even if we will examine etymological data from time to time in order to shed light on the value of the markers being studied.

Mutual intelligibility, which is supposedly the objective of any form of communication, is not always obvious; it can sometimes seem laborious and will generally have to be negotiated. As enunciators, we do this on a daily basis. In fact, linguist Culioli has left us with a meaningful maxim, stating that “understanding is but a specific instance of misunderstanding”.¹

Mutual comprehension, far from being immediate, always has to be conquered, by the very tools that language gives us. More precisely, this mutual intelligibility is the result of adjustments (mainly notional) and readjustments (on the discursive level), intended to refine comments made and create a shared space between the co-enunciators’ representations.

But what does the notion of readjustment cover? How can it be defined, particularly in relation to the notion of adjustment used in TEO? What linguistic elements does it bring into play? How useful is it in the field of discourse analysis, and communication, which is its end goal?

The first mention of the term “adjustment” as a linguistic metaterm² is given to us by Culioli, who links it closely to the term “intersubjective” [CUL 90, p. 43; CUL 99b, p. 10, 18]. It has yet to be determined whether this association with intersubjectivity is necessary, and if it also concerns readjustments on a discursive scale. Are we not able to imagine situations where the enunciator aims, above all, for a better formulation (and not directly a better understanding) in a perspective of harmonization with his/her own thoughts?

First of all, we will define and articulate these notions of mutual intelligibility, adjustment, readjustment and intersubjectivity, in order to better characterize the motivations as well as the modalities of discourse readjustments. This will then enable us to focus on their concrete manifestations in linguistic terms.