Approaching discourse markers in present-day english. A corpus-based study

! Approaching Discourse Markers in Present-day English. A Corpus-based Study Autor: David Varela Villaverde Titor: Ignacio M. Palacios Martínez Xullo 2018 Traballo Fin de Grao Facultade de Filoloxía Grao en Lingua e Literatura Inglesas Universidade de Santiago de Compostela ! Abstract This dissertation is a bid to clarify the status of discourse markers, DMs from now onwards, a tough task due to the large diversity and opposing approaches regarding this field of study. In the last 30 years, the research of discourse markers experienced an ever increasing interest; thus, under the same body of research we find various labels to refer to the members of this category, including discourse connectives, discourse operator, discourse markers, discourse particles, pragmatic expressions, pragmatic markers and pragmatic particles among others. In line with this, the main aim of this paper is the study of discourse markers in present-day written and spoken English with particular attention to one of the most frequent ‘well’ in written and spoken English. Most scholars do agree on the idea that discourse markers are words or expressions that relate and organise discourse. However, there is little agreement as to how they are to be defined or what their primary functions are. In order to shape a coherent definition and a proposal of their syntactic and semantic properties, I will rely on various approaches provided by different scholars, on previous theoretical research. All things considered, I define discourse markers as multifunctional linguistic items that occur within discourse, yet they are not part of the propositional content of the message and do not contribute to the meaning of the proposition. In addition, they are seen as free morphemes that occur in initial position, even though they can be also found in medial and final position, and they are used to mark the relationship within discourse. Moreover, they must follow the sequence S1- DM-S2, where S1 and S2 are understood as discourse segments; in some cases, the elision of the DM may have taken place, an issue to take into account in the analysis. List of Tables Table 1: Frequency of well as DM in spoken English (BNC) 35 Table 2: Types of well by genre (BNC) 37 Table 3: Frequency of well as DM in written English (BNC) 38 Table 4: Types of well in writtenness by genre (BNC) 40 Table 5: Frequency of well as DM in spoken and written English (BNC) 42 Table 6: Types of well in spoken and written contexts by genre (BNC) 42 Table 7: Frequency of well as DM in spoken English (COCA) 44 Table 8: Types of well in spokenness (COCA) 45 Table 9: Frequency of well as DM in written English (COCA) 47 Table 10: Types of well in writtenness by genre (COCA) 48 Table 11: Frequency of well as DM in spoken and written English (COCA) 49 Table 12: Types of well in spoken and written contexts by genre (COCA) 49 Table 13: Frequency of well as DM in spoken British (BNC) and American English (COCA) 51 Table 14: Types of well in spoken British and American English (BNC) vs (COCA) 52 Table 15: Frequency of well as DM in written British and American English (BNC) vs (COCA) 53 Table 16: Types of well in written British and American English (BNC) vs (COCA) 54 List of Figures Figure 1: Screenshot of the general search of well in BNC 26 Figure 2: First ten examples of a random list of well in BNC 26 Figure 3: Screenshot of the general search of well in COCA 27 Figure 4: First ten examples of a random list of well in COCA 28 Table of Contents Abstract Chapter 1. DMs: general overview 1 1.0. Introduction 1 2.0. Primary hardships of the field 2 2.1. Theoretical frameworks 2 2.2. Terminology 3 3.0. Conceptual background 6 3.1. Discourse markers: definition 6 3.2. Discourse markers: main features 8 3.3. Discourse markers: main functions 14 3.3.1. Set of functions 15 3.3.2. Textual and interpersonal functions 17 4.0. Discourse markers: taxonomy 18 5.0. Context of occurrence and importance of intonation in speech 20 Chapter 2. The DM well: analysis 23 6.0. Empirical study 23 6.1. Introductory remarks 23 6.2. Methodology 24 6.2.1. British National Corpus (BNC) 24 6.2.2. Corpus of Contemporary American English (COCA) 26 6.3.1. Well: introduction 28 6.3.2. Well: functions 30 6.3.3. Well: BNC evidence and analysis 34 6.3.3.1. Well: data scrutiny and analysis 34 6.3.3.2. Well: spokenness 34 6.3.3.2.1. Well: frequencies in spokenness 34 6.3.3.2.2. Well: pragmatic functions in spokenness (BNC) 36 6.3.3.3. Well: writtenness 38 3.0. Conceptual background 3.1. Discourse markers: definition It is necessary to mention, beforehand, the little agreement as to what is the finest and most accurate definition of the members of this category, hence, the following collection of definitions hold little resemblance to one another (Brinton, 1996: 30). Opposite to Fraser (2009: 298), I shall refer to discourse markers as linguistic items as it is, to my knowledge, a more inclusive and versatile category than lexical items. However, in spite of the fact that Fraser (2009: 298) refers to this set of items as lexical ones, he further maintains that discourse markers do not only consist of lexical expressions. In my view, Fraser’s approach regarding lexical items excludes “syntactic structures, prosodic features such as stress, pauses, and intonation, and non-verbal expressions such as a grunt or a shrug.” (Fraser, 2009: 298). In this respect, conversely, Schiffrin (1987: 41) approaches DMs as “discourse markers […] members of a functional class of verbal and (non-verbal) devices”, however, it is also to point out that she did not supply a non-verbal discourse marker. Once we reach the section 3.3, all these differences will become more clear since I will be dealing with the wide range of the functions of discourse markers, including here those which are considered to be primary. Therefore, in order to proceed to a definition, I shall display a list based on Brinton’s one (1996: 30). Such a list will be divided into five different approaches as to what the main functions of discourse markers are. Likewise, each and every of the approaches gathers together different pieces of evidence listed (A-O). Nevertheless, the major amount of scholars came to the same conclusion, the range (A-E) of functions are seen as primary, therefore, they forge a broad-based consensus on the definition. For this simple reason, these functions will be tackled at the very beginning in this following listing. !6 •Discourse markers express a relation or relevance between the preceding utterance or context and the current utterance or context in which they occur, that is to say, DMs are: A. Expressions used in order to indicate the relevant dependence of one discourse segment on another (Blakemore, 1987: 125). B. Expressions used to signal the relationship of an utterance to the immediate context (Redeker, 1990: 372) with the primary function of bringing to the listener’s attention a particular sort of linkage of the forthcoming utterance with the immediate discourse context (Redeker, 1991: 1168). C. Marking devices that demonstrate the speaker’s understanding of the contribution’s sequential relation or relevance to the information set as established by the immediately preceding contributions (Goldberg, 1980: 141). D. Devices that signal a sequential discourse relationship as well as how the speaker intends the message that follows to relate to the previous discourse (Fraser, 1988: 21; 1990: 387-392). E. Sequentially dependent elements which bracket units of talk (Schiffrin, 1987: 31), in other words, elements that work on a discourse level, not on smaller units of talk that conform discourse. •Discourse markers are primarily structural as the central function, that implies: F. Certain set of signals in the talker’s speech used to introduce level shifts within the conversation or to prepare listeners for the next turn in the logical argument (Keller, 1979: 220). G. Expressions that help the speaker divide the message into chunks of information and hence helping the listener decode the information units (Erman, 1986: 146). H. Vehicles for demarcation and concatenation which specifically express organisational relations both locally and formally (Even- Zohar, 1982: 179-170). •Discourse markers seen as response signals, in other words: !7 I. They constitute the range of conventionalised responses in English mediating between the covert thinking of participants and displayed verbal behaviour (Schourup, 1985: 3). J. They indicate, quite often in very complex ways, how the utterance that contains discourse markers is a response to, or a continuation of, some portion of the previous discourse (Levinson, 1983: 87-88). •Discourse markers as a means to achieve conversational continuity, which entails: K. They are conventionalised ways of plugging potential gaps, reducing those gaps to such an extent that these are not even perceived by the interlocutor (Edmondson, 1981: 154). L. They are used to maintain the continuity of discourse (Crystal & Davy, 1975: 88-91). M. They fill the silence and maintain the speaker’s right to speak while they organise the information they want to say (Brown, 1977: 109). •Discourse markers should be regarded as ‘essentially interactive’ (Stubbs, 1983: 70); that means,: N. Markers which implicitly anchor the act of communication to the speaker’s attitudes towards aspects of the on-going interaction (Östman, 1981: 5; 1982: 152). O. Vehicles for the establishment and maintenance of interpersonal relations between interlocutors (James, 1983: 193). 3.2. Discourse markers: main features Hereafter, I shall provide a numbered list, for ease of reference, that includes some of the main and common features or characteristics of discourse markers. The fact that there is not a consensus, in relation to such characteristics of discourse markers, led me to gather together the most remarkable ones suggested by different authors. !8 Taking Schiffrin’s 1987 work as a referential point in the light of the effect she caused on this field of study of discourse markers, I shall start this list of common features focusing on the ones supplied by this author. As Schiffrin suggests, the magnitude of such an approach does not only request knowledge of the shared characteristics of a vast and, most importantly, varied group of expressions, but also analysis across different languages in order to find out other linguistic resources as to what makes a certain expression a discourse marker. On top of it all, Schiffrin (1987: 328) makes a tentative suggestion as to what specific linguistic conditions allow a particular word or expressions to be used as a marker, that is, the delimitation of discourse markers as to what sorts of words or expressions are considered to be markers. Thus, according to Schiffrin (1987: 328): I. A discourse marker has to be syntactically detachable from a sentence; in other words, separable from the clause. Consider, for example, the following: (1) In other words, lack of financial resources can contribute to homelessness (BNCF87S_meeting) 1 II. It has to be commonly used in the initial position of an utterance although it can also be found medially and in final position. (2) The rest, of course, is history. (BNCCH1W_newsp_tabloid) III. It has to have a range of prosodic contours (e.g. tonic stress and followed by a pause, phonological reduction). Obviously, this feature can only be applied to speaking contexts. (3) They will have to pick us up cos our pass is not valued till eight o’ clock. (BNCD95S_meeting) The examples here included have all been extracted from the British National Corpus and they have been referenced 1 according to the codes provided by this corpus. S stands for spoken where W represents writing. !9 IV. It has to be able to operate at both local and global levels of discourse as well as on different planes of discourse. (4) The traditional forces for voting, namely party loyalty and social class (BNCW_ac- Polit_law-Edu) V. It either has to have no meaning, a vague meaning, or to be reflexive (of the language, of the speaker). (5) Client reports and things like that. (BNCFUKS_meeting) Zwicky (1985: 303-304), prior to Schiffrin, suggested that DMs tend to occur at the beginning of sentences to serve as a mechanism for continuing conversation. Moreover, he also argues that DMs are prosodically separated from their surrounding context by means of pauses and intonation breaks. However, Zwicky notes they are usually monomorphemic, yet morphologically complex and syntactically isolated; in other words, they are isolated from the sentence or context in which they occur. In addition, Schiffrin makes a generalisation, noting that any expression that operates within at least one discourse component has the particularity of being able to become a marker that operates within other discourse components, simply due to the integration among components. In order to support her own approach, the author provides the reader with a brief explanation of what she means by this. She says that those expressions by means of which speakers show their orientation toward a proposition, “e.g. an adverb such as surely, an interjection such as God, a polarity term such as yeh”, have the particularity of being able to become markers of other discourse components. Likewise, some other expressions such as those used by speakers either to organise action and exchange structures or to structure ideas and information states, also share this particularity of being able to become markers (Schiffrin, 1987: 329). !10 Jucker & Ziv (1998: 3) condense the conditions provided, previously, by Hölker (1991: 78-79) as to what a discourse marker entails. One aspect to bear in mind, as pointed out by Jucker & Ziv (1998: 3), is the fact that Hölker (1991: 78-79) approached these linguistic items as pragmatic markers, an event that reinforces the idea of such a fuzzy terminology tackled previously (see section 2.2). Hölker (1991: 78-79) applied the following features to discourse markers: VI. They do not affect the truth conditions of an utterance. (6) Well, I think that er, you know, just didn’t happen like that. (BNCD95S_meeting) VII.They do not add anything to the propositional content of an utterance, for instance, well in the previous example. VIII.They are related to the speech situation and not to the situation talked about. (7) Of course, we haven’t got there yet. (BNCW_fic_drama) IX. They have an emotive expressive function rather than a referential, denotative, or cognitive function. (8) Wow, it’s working. (BNCW_Newsp_other_report) Jucker & Ziv (1998: 3) go a step further, categorising the features VI and VII as semantic in nature, the VIII as pragmatic and, last but not least, the IX as functional. Jucker & Ziv (1998: 3) were responsible for the adaptation and abbreviation of Brinton’s (1996) list, a longer one including the basic features of DMs. However, it may lead to confusion due to no further explanation, in this case, I shall not stick to Jucker & Ziv (1998: 3) as I consider their proposal vague regarding some of the characteristics of discourse markers. For the purpose of offering a clarifying list that allows a better understanding of such features, I shall adopt the division illustrated by Jucker & Ziv, a reordered list which gathers together features that belong to the same level of linguistic description, going back to some particular features originally provided in Brinton’s whole list. !11 •Phonological and lexical features: X. They are short and often phonologically reduced. XI. They form a separate tone group with falling-rising or rising intonation. (see section 5.0) XII.They are marginal forms and hence difficult to place within a traditional word class. •Syntactic features: XIII.They are said to be restricted to sentence-initial position, or may always occur sentence initially. However, as Brinton (1996: 33) points out, basing herself on different studies of individual markers, they frequently appear sentence medially and finally as well. XIV.They occur outside the syntactic structure or they are only loosely attached to it and hence have no clear grammatical function. XV.They are optional rather than obligatory features. Schiffrin (1987: 55) claims that ‘the structure and the meaning of an argument can be preserved even without markers”. •Semantic features: XVI.They have little or no propositional meaning, or at least difficult to be difficult to specify lexically. Brinton (1996: 33) relied on Schiffrin’s meaning-minimalist view (see section 3.2. V) in order to propose this characteristic. •Functional features: XVII.They are multifunctional, operating on several linguistic levels simultaneously such as on the local (i.e., morphophonemic, syntactic, and semantic) and on the global (i.e., pragmatic), as well as on different planes within the pragmatic component. •Sociolinguistic and stylistic features: XVIII.They are a feature of oral rather than of written discourse and are associated with the informality of oral discourse and the grammatical ‘fragmentation’ caused by the lack of planning time characteristic of spoken contexts (Östman, 1982: 169). XIX.They appear with high frequency in oral discourse. !12 XX.They are stylistically stigmatised and negatively assessed owing to their high frequency and oral nature, especially in written or formal discourse. They are regarded as a cue of non-fluency and carelessness. This feature is closely related to (see section 3.2. XVIII; XIX), features characteristic of spokenness. XXI.They are gender specific and more typical of women’s speech. For instance, Östman (1981: 72; 75-76) finds you know to be more frequent in women’s than in men’s speech. However, Holmes (1986: 4) finds you know as frequent in women’s than in men’s. In my mind, the clarification that Jucker & Ziv (1998: 4) make concerning Brinton’s list is highly relevant, claiming that not all the features displayed in Briton’s are equally diagnostic. More precisely, as they propose, the three first levels (phonological/lexical, syntactic/textual and semantic) are those which provide the litmus tests as to what is considered to be as discourse marker or not. Likewise, Jucker & Ziv (1998: 4) suggest that those features mentioned on the functional and on the sociolinguistic/stylistic levels are primarily descriptive. Moreover, Jucker & Ziv (1998: 4) point out that sociolinguistic and stylistic distribution can only be applied once an item has been identified as a discourse marker. In addition, Jucker & Ziv (1998: 4) assert that occurrence either in oral or written contexts is not particularly helpful for the purpose of diagnosis and classification of DMs. Andersen & Fretheim (200: 40) also illustrate a list of characteristics of DMs taking as a reference the model introduced by Briton (1996: 32-35), a more diffuse one than the one reorganised and abbreviated by Jucker & Ziv (1998: 4). Conversely, Andersen & Fretheim (2000: 40) deal with the importance rendered to non-propositionality as an essential feature of discourse markers, yet, they claim that this feature should be considered as usual but not as essential. !13 As a conclusion, relying on the perspective adopted by Jucker & Ziv (1998: 4), the list previously provided above is a suggestive proposal of the range of properties discourse markers may display. Jucker & Ziv (1998:4) further argue that discourse markers may fulfil all of these characteristics or just a single one, yet, the higher the number of features is manifested by a certain discourse marker, the higher the possibility to consider such an item as a member of this class. 3.3. Discourse markers: main functions With regard to the features previously mentioned, we come to the conclusion that DMs are grammatically optional since they do not disrupt the grammatical correctness in both spoken and written contexts. However, where discourse markers are not present, the hearer has no lexical clue as to the relationship intended between the two segments (Fraser, 1999: 944). Nevertheless, Fraser also discusses that since, while, whereas and because, among others, are cases of DMs in which there are syntactic reasons why discourse markers cannot be deleted nor omitted. Furthermore, they are semantically empty, that is to say, as we previously assumed, DMs do not contribute to the proposition of the utterance or sentence in which they occur. However, they are not pragmatically optional or unnecessary, as they do contribute to a largely range of pragmatic functions (Brown, 1977: 107; Svartvik 1980; Erman, 1987: 2). In this respect, according to Brinton (1996: 35), a bid to omit such markers will not have an effect on grammar, that is, the discourse will be grammatically acceptable, conversely, this event will lead to an “‘unnatural’, ‘awkward’, ‘disjointed’, ‘impolite’, ‘unfriendly’, or ‘dogmatic’” communicative context. When discourse markers are missing within the communicative context, the factor of comprehension for the listener !14 might be compromised, and highly affected in the case of the speaker (Svartvik, 1985: 352), leading to a high likelihood of breakdown within the communicative context (Fraser, 1990: 390). Thus, according to Even-Zohar (1982: 180), DMs aid participants of a communicative context to navigate along a certain discourse, both encoder and decoder. Conversely, Ötsman (1982: 153) claims that DMs are to be solely and exclusively pragmatic in function, that is to say, their function is to hook the propositional content. The latter idea held by Ötsman (1982: 153) sheerly opposes one of the main features listed above (see section 3.2. XVII), as well as to the quintessential character of discourse markers previously admitted, multifunctionality (see section 2.1). 3.3.1. Set of functions Determining the functions of discourse markers is a complex and challenging task. As a result, in order to provide a set of functions of such linguistic items, it is necessary to focus on previous studies, preferably on those that adopt a general approach, setting aside those studies whose main centre of attraction is just a single discourse marker, as they may lead us astray, sticking out from our purpose. The reason why studies of individual markers are avoided is that these studies attribute a vast range of meanings to a single discourse marker and, as a consequence, some of these meanings might overlap. Therefore, I shall present a reorganised and abbreviated list, divided into textual and interpersonal functions, based on the proposal made by Brinton (1996: 35-40): •Textual functions, for instance, used: 1. To initiate and close discourse, including claiming the attention of the addressee. 2. To help the addressor in acquiring or relinquishing the floor. !15 !22 Chapter 2 The DM well: analysis 6.0. Empirical Study 6.1. Introductory remarks Section 6 aims to study in close detail a particular discourse marker: well in Present-day English in line with the description made in the first part of this dissertation. The afore-mentioned DM was chosen as a basis for this research due to its wide range of functions and its high frequency of appearance in everyday language use. Therefore, attention will be paid to its context of occurrence and its degree of spokenness or writtenness. I will also study up to what extent this discourse marker is more closely associated with a particular genre and its general frequency in British and American English on the basis of the analysis of comparable corpora samples. Regarding well, a total of 400 examples, 200 taken from BNC (British National Corpus) and 200 from COCA Corpus of Contemporary American English), were first selected in terms of their possible interpretation. The collected data were entered in a data-based (see Appendix A; B) and classified into three main categories, frame marker, mitigator and pause filler (Jucker, 1997). I will refer to each of them in detail below. The results were calculated and discussed accordingly in order to discover some linguistic regularities typical of the members of this category of linguistic items. !23 6.2. Methodology As mentioned above, this investigation is fully based on the data retrieved by means of online computer software, which was later on gathered together and stored by means of Microsoft Excel in order to be, subsequently, closely analysed. This means that this is a corpus-based study. The BNC (British National Corpus) and COCA (Corpus of Contemporary American English) were both crucial for collecting and categorising the range of real language examples. Such examples were accommodated according to their respective categories; function, level and label. It is to bear in mind that those samples in which well displays the functions of adverb, adjective, noun or verb were disregarded in the analysis as they are not considered to be DMs but false positives. Apart from raw frequencies, normalised frequencies were also provided. Since the number of tokens in each corpus and genre varies significantly from one to another, as we will see in the two upcoming sections, normalises frequencies per 1,000,000 words were calculated to make the corpora and all the samples comparable. 6.2.1. British National Corpus (BNC) BNC was, as explained on its website, “originally created by Oxford University Press in the 1980s - early 1990s, and it contains 100 million words of texts from a wide range of genres (e.g. spoken, fiction, magazines, newspapers, and academic)”. Furthermore, the original idea was to design a corpus which could be regarded as representative of the British English from the later part of the 20th century, both spoken and written. Although the accomplishment of the corpus took 3 years, from 1991 to 1994, and no new texts have been added after its completion, it was slightly revised prior to the release of the second edition of the BNC World (2001) and the latest edition, the BNC XML Edition which was released in 2007. !24 The BNC, according to its website, meets the following characteristics: 1) it is a monolingual corpus, in other words, although the corpus tackles only modern British English used in Britain, it also contains some non-British English as well as foreign language words. 2) It is synchronic, this means that it does not focus on the historical development of British English but on the late twentieth century. 3) It is general, that is, it contains many different styles and varieties, not being constrained to a certain subject field, genre or register. Furthermore, it includes examples of both spokenness and writtenness. 4) Samples of 45,000 words extracted from single-author texts (excerpts), and shorter than 45,000 words extracted from multi-author texts (magazines, newspapers, etc.) in full version. It is a very useful source of data for tracking down the contemporary use of a chosen linguistic element in British English and provides the chance of retrieving and analysing an amount of data otherwise inaccessible. In fact, sampling allows for a wider coverage of texts within the 100 million limit, and avoids over-representing idiosyncratic texts. BNC is as large as any other structured corpus of contemporary English and it contains real language examples collected within the time period of the late twentieth century. The corpus allows to retrieve a chosen number of randomly selected samples, a pivotal feature in the following investigation. Due to the vastness of the data available in this corpus (141,317 tokens of well), I decided to select 200 randomised tokens of this DM in particular, providing examples of the way it is represented in the corpus. Having gathered 200 real language examples from BNC, in total, was done automatically by means of the corpus utility. Figure 1 and 2 show how the data was accessed. !25 Figure 1: Screenshot of the general search of well in BNC Figure 2: First ten examples of a random list of well in BNC 6.2.2. Corpus of Contemporary American English (COCA) It is to bear in mind that both the BNC and the COCA are fully comparable corpora since they represent present-day British English and American English, respectively, and they have been designed following the same criteria regarding their organisation and structure. COCA is, as stated on its website, 5 to 6 times as large as the BNC. Furthermore, it supplies data for lower-frequency constructions that are not accessible from the BNC. Indeed, the American Corpus includes more than 560 million words of text, result from adding 20 million words each year from 1990 to 2017. What is more, it is equally divided among spoken, fiction, popular magazines, newspapers, and !26 academic texts. The corpus most recent addition of texts dates from January 2016, and it was accomplished in December 2017. COCA is certainly a very useful tool that allows us to trace the use of a certain linguistic element in contemporary American English. Furthermore, it helps you recover and analyse such an extent of data which would be otherwise inaccessible. The COCA can be regarded as the largest structured corpus of contemporary American English and it contains real language examples collected within the time period of 1980-2017. The corpus is a systematic collection of texts stored on computer, thus, provides a vast number of randomly selected samples, specifically (693,977 tokens of well). As a consequence of this and for the purposes of this study, I selected a second set of examples of 200 randomised tokens for the DM well, gathered together automatically by means of the corpus utility. Both corpora were accessed through BYU (Brigham Young University). 3 Figure 3 and 4 show how the data was accessed. Figure 3: Screenshot of the general search of well in COCA BNC-BYU: <https://corpus.byu.edu/bnc/>, last access 25/06/18. 3 COCA-BYU: <https://corpus.byu.edu/coca/ >, last access 25/06/18. !27 Figure 4: First ten examples of a random list of well in COCA 6.3.1. Well: introduction Despite the fact that well sometimes displays the functions of a noun, an adverb, an adjective or a verb, it is difficult to characterise its uses basing the analysis on any of these classes. Thus, this points out that the use of well is not only based on semantic or grammatical status, however, it goes a step further (Schiffrin, 1987). Therefore, there are many studies on the use of well that have been observed, one of them is investigated by Schiffrin (1987: 102). In her study she approached well from the perspective of a response marker, hence, well is described from the point of view of the interlocutor’s answer in a conversation. In addition, this is a research study in which she further argues that well “anchors its user in an interaction when the upcoming contribution is not fully consonant with prior coherence options […] it functions in the participation framework”. Likewise, De Klerk (2005: 1190) finds out and explains the tendency for well to start utterances in a natural way “since discourse markers typically act as a guide to addressees as to how to react to what is about to be said”. !28 In fact, the next important study on discourse markers is carried out by Biber et al. (1999:1086) and suggest that well is “a versatile discourse marker, but appears to have the general function of a ‘deliberation signal’ indicating the speaker’s need to give (brief) thought or consideration to the point at issue”. What is more, Biber et al. (1999:1096) also provide the frequency of well in English conversation, confirming that well is more frequent than, for instance, you know and I mean, either in British and American English. Fung (2007) claims that the DM well is of a great importance and that it has a great impact on someone’s own use of language. In addition, they suggested that as speakers improve their fluency in English, generally, we can expect them to use well both to take the turn and to signal a break in the utterance. However, Jucker (1997) approached the discourse marker well in the history of English, trying to condense each and every of the uses of well as a discourse marker, concentrating on the occurrence in written texts and on the diversity of forms and functions. He proposes that the distinct uses of well can be properly accommodated into four different categories depending on the use at issue: 1) as a frame marker which either introduces a new topic or prefaces direct reported speech; 2) as a qualifier marker that prefaces a reply which is only a partial answer to a question; 3) as a face-threat mitigator marker that prefaces a disagreement; 4) as a pause filler that bridges the interactional silence. (See examples for each of these below in my analysis). For the purpose of this investigation, I will rely on Jucker’s proposal as I consider it to be the most adequate and comprehensive, hence, it will be developed into close detail. Nevertheless, I will suggest a slight variation from Jucker’s proposal as is sheerly based on written texts, and it might mislead the purpose of this study, since he suggested some categories attending the uses of well either on-stage and off-stage, in literary works. !29 6.3.2. Well: functions As mentioned above, although there are many different existing categories of well in modern English, for the purpose of this dissertation, I will rely on Jucker’s (1997: 93) proposal, who claims that in contemporary English it can be classified under the following categories: (Jucker, 1997: 92). Jucker 1997 work suggests that on the propositional level, well can be used as an adverb or an adjective as in well-educated or everything is well. He further argues that under no circumstance can well be considered as a discourse marker in these cases as it lacks textual and interpersonal functions (see section 3.3) since it clearly functions as a lexical item typical of the adjectival or adverbial category. In this respect, I must say that I fully concur with Jucker, but I must add a slight objection; well can be also used as a noun as in “By this time all the wells had run dry” with the 4 meaning of pit or in contrast to woe, or as a verb as in “As she read the letter tears welled in her 5 eyes” with the meaning of the emergence of a liquid, hence, it is also at the propositional level, 6 7 thus, it does not meet the functions to be considered as a discourse marker either. In this respect, I Function Level Label Adverb, adjective Propositional — Discourse marker Textual Frame marker Discourse marker Interpersonal Face-threat mitigator Discourse marker Interpersonal Qualifier Discourse marker Interpersonal Pause filler Example taken from the Cambridge Dictionary. 4 The OED records for example the following: "Most ethicists point out that personal matters have moral importance 5 not just because they give people power over each other, for well or for woe". <http://www.oed.com/view/Entry/ 226979?rskey=6Ay4y0&result=2&isAdvanced=false#eid>, last access 25/06/18. Example taken from the Cambridge Dictionary. 6 The Cambridge Dictionary records for example the following: “(of a liquid) to come to the surface or into view”. 7 <https://dictionary.cambridge.org/dictionary/english/well>, last access 25/06/18. !30 must make clear that I will not take into consideration those uses of well in which it does not strictly work as a discourse marker and they will thus be addressed as false positives. Interpersonal and textual functions are not mutually exclusive, which means they do not occur separately, both are present in each and every discourse marker. However, one of the functions is usually predominant over the other, that is, one of them gains importance at the expense of the other. Jucker (1997) discusses that this particular feature makes it easier for us to classify these linguistic items into the four different uses afore-mentioned. Nonetheless, depending on the approach taken, the outcome might result biased or deviated, that is why some of the data could be interpreted in different ways, hence, the final outcome should be understood as guidance notes and not as universal truths. The first of the uses is frame marker, which means that well is used whether to indicate a change in the topic or to introduce reported speech. Svartvik (1980) and Jucker (1997) contend that in both cases well is operating at the textual level as a means of text-structure device. (Examples provided in the analysis). The second of the uses of well is face-threat mitigator and, as Jucker (1997) explains, it shows some insufficiencies on the interpersonal level. According to Owen (1981, 1983), well is an indicator and a mitigator of some kind of confrontation between the addressor and the addressee. For example, this might be the case in which an assessment is followed by disagreement rather than agreement. (Examples provided in the analysis). As we can notice, these are situations in which the image of the participants is being threatened and needs to be saved, hence, this common practice forms the second of the uses. In this respect, I would like to add that the use of well as a face-threat mitigator is an indicator of politeness, that is, it is used in order to save the face either of the 8 addressor, of the addressee, or of both of them. In accordance to Jucker’s proposal, I strongly See Brown and Levinson (1987). 8 !31 contend that in spoken contexts the prevailing type of well is that of a pause filler. As a result, in line with what has been said above, considering well more likely to occur as a frame marker than as a mitigator, just because they belong to the textual and interpersonal levels, respectively, may lead to a biased assumption, hence, this is a practice to be avoided. 6.3.3.3. Well: writtenness As before, the analysis of the written component of the BNC will be tackled in two different sections. In the first one, I will display in Table 3 the outcome of the written data analysed and the general frequencies of well, while in the second one, the types of the DM in question. Besides, in order to maintain the findings of my analysis as trustful and objective as possible, I will also provide the normalised frequencies (per 1,000,000). 6.3.3.3.1. Well: frequencies in writtenness Table 3 below shows the frequencies of well in the written data of the BNC: Table 3: Frequency of well as DM in written English (BNC) Context Genre Samples Retrieved DMs | NF* False Positives Unclear •Written Fiction Prose 30 18 | 1.12 12 0 •Written Fiction Drama 1 1 | 21.69 0 0 •Written News Script 1 0 | - 1 0 •Written Newspaper 6 0 | - 6 0 •Written Pop 11 1 | 0.13 10 0 !38 NF* Normalised frequencies per 1,000,000 words. The data retrieved shows that only 25 out of 98 are the cases in which well works as a discourse marker. The first question that comes to my mind is: whether well is likely or not to happen in written contexts. The second one is whether there any written genre in which well is more likely to occur. The answer to the first question, based on the data retrieved, leads to a premise which claims that well is unlikely to happen in written contexts, an assumption that I truly expected from the very beginning. The answer to the second question, yet, requires a detailed examination of the data retrieved. During the course of my analysis, I came across a wide range of genres such as newspapers and academic, in which a large number of samples were retrieved, and in none of them well functions as a DM. Nevertheless, this is not what certainly draws my attention since they are false positives. What truly interests me are those cases of well in written contexts in which it works as a DM, as they constitute the subject matter of my dissertation. Until the analysis had been accomplished, I have got no clue about the fact that the DM well, in spoken contexts, is more frequent in (fiction) which stands first, followed by (miscellaneous) that stands second and (non-academic) that stands third. In my view, it is of interest that all those uses of well as a DM are mainly registered within the same genre, fiction. •Written Academic 12 0 | - 12 0 •Written Non- Academic 14 1 | 0.29 13 0 •Written Religion 1 0 | - 1 0 •Written Biography 4 0 | - 4 0 • Written Commerce 3 0 | - 3 0 • Written Miscellaneous 15 4 | 0.43 11 0 Total (Written) 98 25 | 23.66 73 0 !39 These questions that come to my mind, at this point, will be displayed in the next section, where, when possible, I will try to provide an answer to all of them. 6.3.3.3.2. Well: pragmatic functions in writtenness Having already answered the first set of questions, a new one emerged: what is the most prevailing pragmatic value of well? For this purpose, I will tackle the written component of the BNC in Table 4, displaying the classification of the uses of well by genre into the three different types aforementioned (see section 6.3.2). Likewise, I will also provide the normalised frequencies for each of the genres. Table 4: Types of well in writtenness by genre (BNC) DM Type Pause Filler| NF* Frame Marker| NF* Mitigator| NF* Total DMs •Written Fiction Prose 3 | 0.18 9 | 0.56 6 | 0.37 18 •Written Fiction Drama 0 | - 1 | 2.16 0 | - 1 •Written News Script 0 | - 0 | - 0 | - 0 •Written Newspaper 0 | - 0 | - 0 | - 0 •Written Pop 1 | 0.13 0 | - 0 | - 1 •Written Academic 0 | - 0 | - 0 | - 0 •Written Non- Academic 0 | - 1 | 0.29 0 | - 1 •Written Religion 0 | - 0 | - 0 | - 0 •Written Biography 0 | - 0 | - 0 | - 0 • Written Commerce 0 | - 0 | - 0 | - 0 !40 NF* Normalised frequencies per 1,000,000 words At this stage, we may claim that the prevailing use of well, in fiction, is that of a frame marker. However, it is also to be borne in mind its use as pause filler, since they both function on the textual level. Therefore, I contend that the main level in which well operates, in fiction, is on the textual one. Thus, the question is: is this assumption applicable to any other genre within writtenness? Indeed, based on the results provided by Table 4 above, I confirm that the prevailing use of well in all the genres analysed within the written component is that of frame marker, Due to the fact that the predominant use of well is, in written contexts, that of frame marker, I aim at examining whether they are used in order to indicate either a change in the topic of the conversation or to preface direct reported speech. Correspondingly, my data confirms that the most predominant use of well in written contexts as a frame marker, is to preface direct reported speech. 6.3.3.4. Well: comparing spokenness vs writtenness in BNC In this section, I briefly aim to illustrate the differences and similarities, regarding the uses of well, between spokenness and writtenness in British English. In Table 5 I show the outcome of spoken and written data submitted to analysis. • Written Miscellaneous 3 | 0.32 1 | 0.10 0 | - 4 Total (type) 7 | 0.63 12 | 3.11 6 | 0.37 25 !41 Table 5: Frequency of well as DM in spoken and written English (BNC) NF* Normalised frequencies per 1,000,000 words One of the main aims that this dissertation pursuits is to get to know the variation in the frequency of occurrence of the DM well, depending on the media of expression, that is, speech versus writing. The data obtained clearly indicated that the DM well is more likely to appear in spoken contexts. However, this does not mean that it is a DM that exclusively occurs in spoken contexts although here its frequency drops significantly. In spite of the fact that well is more likely to appear in spokenness than in writtenness, particularities were noticed by tackling each of the genres in close detail. Table 6 shows the classification of well by genre into the three different types afore-mentioned. Table 6: Types of well in spoken and written contexts by genre (BNC) NF* Normalised frequencies per 1,000,000 words Despite the fact that well is much more likely to occur in spoken contexts, my data indicates that it also appears in written ones, nonetheless, there are slight differences in its uses. In both media of expression, well takes place at the textual level. Thus, this is a feature that does not vary depending Contexts Samples Retrieved DMs | NF* False Positives Unclear • Spoken 102 79 | 7.87 22 1 • Written 98 25 | 0.28 73 0 TOTAL 200 104 | 8.15 95 1 DM Type Pause Filler | NF* Frame Marker | NF* Mitigator | NF* Total DMs | NF* Spoken 48 | 4.78 10 | 0.99 21 | 2.09 79 | 7.86 Written 7 | 0.08 12 | 0.13 6 | 0.06 25 | 0.27 Total (type) 55 | 4.86 22 | 1.12 27 | 2.15 104 | 8.13 !42 on the context of occurrence. Nonetheless, as previously suggested, the way in which well behaves depending on whether it occurs in spokenness or writtenness varies. The main use of the DM well is to bridge interactional silence, and everything that this entails, a characteristic extensible to all of the genres in spokenness and writtenness but fiction (writtenness), which deviates this use of well to a particular use, frame marker, that implies the addition of direct reported speech. 6.3.4. Well: COCA evidence and analysis 6.3.4.1. Well: data scrutiny and analysis Due to the wide variety of samples taken from COCA (200), this list of randomly chosen examples was scrutinised. Thus, once the scrutiny of the data was accomplished, the outcome is a list that consists of 98 examples in which well does work as a DM. As well as with the data taken from BNC, I will take into consideration the different genres provided by the corpus into which examples are classified. This classification is made into different genres in comparison to the ones provided in the BNC. COCA supplies in regard to spokenness, just spoken, and regarding writtenness, fiction, news, magazines and academic. There are then clear differences in this respect. 6.3.4.2. Well: spokenness Once again, the analysis of well in spoken contexts will be carried out within two different sections. The first section will be devoted to frequencies, while the second one to the different types of the DM at issue. To begin with, I will display in Table 7 the outcome of the spoken data analysed. Since there is no distinction among genres, regarding the spoken component, as in the BNC, I will only !43 display a table that shows the frequencies of well, in spoken contexts in general. As before, normalised frequencies (per 1,000,000) will be provided since the examples corresponding to the different media of expression and genre are different. Accordingly, at a second stage, I will tackle the pragmatic functions of the DM in question. 6.3.4.2.1. Well: Frequencies in spokenness Table 7 below shows the frequencies of well in the spoken data of the COCA: Table 7: Frequency of well as DM in spoken English (COCA) NF* Normalised frequencies per 1,000,000 words At this point, once the data have been presented, I will proceed to analyse and comment on everything that, in my view, is of interest to my analysis. To get started, I will deal with an example of a false positive: 12 (10) water in rural Belarus. " My mom had gotten some water out of the well, and she suspected that the water wasn't very good. So she would. (COCA2015NEWSAustin) Regarding false positives, it is to mention that, in this occasion, contrary to the data extracted from the BNC, I did find uses of well as a noun as in extract (10). Moving on, and leaving false positives aside, the data retrieved indicates that 75 out of 104 are the cases in which well works as a DM, that is, a percentage of around 72%. This is a fact that Context/ Samples Retrieved DMs | NF* False Positives Unclear Spoken 104 75 | 0.64 29 0 Total (Spoken) 104 75 | 0.64 29 0 The example here included has been extracted from the Corpus of Contemporary American English and it has been 12 referenced according to the codes provided by this corpus. !44 points out that well is highly likely to occur in speaking contexts as a DM. A premise expected from the very beginning. Since there is no spare genre to study as in the BNC, it is time to move on to the next section. 6.3.4.2.2. Well: pragmatic functions in spokenness Albeit well meets the function of DM, does it work on the same level? In order to answer this question, in Table 8 I aim at displaying the data regarding the types of this DM in the spoken component of the COCA. Nonetheless, it is to be borne in mind that one of the examples that I consider as a DM, hence it is counted, is unclear in terms of its classification. Thus, in order not to deviate my analysis, the column ‘unclear’ is added. Table 8: Types of well in spokenness (COCA) NF* Normalised Frequencies per 1,000,000 words. At this point the exhibition of data has been accomplished, consequently, I will proceed to analyse and comment on everything that, in my view, is of interest to my analysis. To get started, I will deal with the unclear example: 13 (11) Thank you. Thank you very much, folks. Thank you. Okey-doke. Well, thanks so much. Please be seated. Make yourselves at home. This (COCA1995SPOKInd_Limbaugh) DM Type Pause Filler| NF* Frame Marker| NF* Mitigator| NF* Unclear Total DMs Spoken 44 | 0.37 9 | 0.07 21 | 0.17 1 75 Total (type) 44 | 0.37 9 | 0.07 21 | 0.17 1 75 The example here included has all been extracted from the Corpus of Contemporary American English and it has 13 been referenced according to the codes provided by this corpus. !45 I consider extract (11) unclear because the turns of the interlocutors were not included in this excerpt, hence, I discard it. Additionally, I will also discard the 29 cases of well as a false positive, as for instance example (10) above. Getting into the analysis, my data points out that the primary type of well is that of a pause filler that stands first, mitigator would stand second and, last but not least, frame marker would occupy the third position. Thus, I contend that the main level in which well works is on the textual one since it may function not only as a pause filler but also as a frame marker. However, although the interpersonal level consists of just the mitigator function, this does not mean that well is more likely to occur as a frame marker rather than as a mitigator. 6.3.4.3. Well: writtenness In order to analyse the written component of the COCA, it will be tackled, as before, in two different sections. Firstly, I will display in Table 9 the outcome of the written data analysed and the general frequencies of well and, secondly, the types of the DM in question. Furthermore, I will also provide the normalised frequencies (per 1,000,000). 6.3.4.3.1. Well: frequencies in writtenness Table 9 below shows the frequencies of well in the written data of the COCA: !46 Table 9: Frequency of well as DM in written English (COCA) NF* Normalised frequencies per 1,000,000 words My data shows that only 23 out of 96 are the cases in which well works as a discourse marker, that is, a percentage of about 20%. Therefore, this event indicates that well is unlikely to occur in written contexts, a presumption expected from the very beginning. However, there is a genre in which it seems to be slightly used, fiction. Although well does not work as a DM in a large number of cases, those in which it does will be analysed. Once again, I did not expect well to be mainly used in written contexts, in fiction. Nevertheless, the fact is that fiction stands first, followed by magazines that stands second and news that stand third and academic that stands fourth. 6.3.4.3.2. Well: pragmatic functions in writtenness Again the question is: what is the most predominant type of well? For this purpose, I will tackle the written component of the COCA in Table 10, showing the classification of the uses of well by genre. Likewise, I will also provide the normalised frequencies for each of the genres. Context Genre Samples Retrieved DMs | NF* False Positives Unclear •Written News 17 3 | 0.02 14 0 •Written Fiction 35 14 | 0.12 20 1 •Written Academic 16 1 | 0.00 15 0 •Written Magazines 28 5 | 0.04 22 1 Total (Written) 96 23 | 0.18 71 2 !47 6.4.4. Well: pragmatic functions in written (BNC) vs (COCA) What is the most prevailing type of well? In this section, the written component of the BNC and the COCA will be tackled. Table 16 shows the classification of the uses of well by type in written British and American English. In addition, I will also supply the normalised frequencies for each of the types in both varieties. Table 16: Types of well in written British and American English (BNC) vs (COCA) NF* Normalised frequencies per 1,000,000 words At this stage, I cannot state which is the major type of well in written contexts since the difference is minimal. However, it seems that as a frame marker it stands first, pause filler as second and mitigator as third. As regards the function of well as frame marker, it was not possible to determine whether it is used to introduce a new topic in conversation or to preface direct reported speech in American English. However, as previously mentioned, my data shows that in written British English well is mainly used as a device to preface direct reported speech. Nonetheless, my data shows that in both British and American English, the most prevailing level in which well works is on the textual one, formed by its use as pause filler and frame marker. DM Type Pause Filler| NF* Frame Marker| NF* Mitigator| NF* Total DMs •British 7 | 0.08 12 | 0.13 6 | 0.06 25 •American 9 | 0.01 9 | 0.01 5 | 0.01 23 Total (type) 16 | 0.09 21 | 0.14 11 | 0.07 48 !54 6.5. Discussion After having analysed the data from the BNC and the COCA, the question that comes to my mind is: do these categories shape an absolute categorisation of the DM well? Can we talk about subtypes within these three categories depending on the purpose for which they are used? As I explained before, the structure of my dissertation, regarding the classification of well, is based on Jucker’s (1997) proposal. His proposal, as well as mine, does not deny the existence of any other sort of approach such as well as a response marker, as an initiation marker, turn initiator or as a pre-closing device. However, what I mean to do, following these three types suggested, is to propose the different ways, or subtypes, in which, in my view, well functions. To get started, I will deal with the categories that I consider to work on the textual level, pause filler and frame marker, but they both have different functions that I will comment on. In those cases of well used as a frame marker, we may find it; either to introduce a new topic in conversation or to preface direct reported speech: (12) originally. (SP:PS03S) Well they don't look like there's, the boots (SP:PS03T) Well they're damn clean! Now, nip all that up and eat it. (BNCKBBS_Conv) 14 (13) her friend died in the (unclear)2. (SP:KBEPSUNK) (unclear) (SP:KBEPSUNK) Listen me Blanche she said, well I knew the truth she said (unclear) was a friend to me. (SP:KBEPSUNK) Yeah. (BNCKBES_Conv) In line with this, I have nothing else to add apart from what Jucker (1997: 93) suggested. In extract (12), well marks the introduction of a new topic brought into the conversation. As we can observe in this example, well has little to do with the previous utterance, hence, it is used in order to move onto a new topic in conversation. The speaker may need to change or shift the present topic for various reasons, but this may lead to incoherency, for this reason, well is introduced in order to indicate a SP:[…] stands for a particular speaker in a particular conversation. 14 !55 switch in the topic and to keep the coherency in the conversation, in other words, to avoid the switch to be abrupt. In my opinion, this way of using well may overlap, to some extent, with the function of mitigator, as I will explain later on. Witness the following: (14) the door open with it. (SP:PS535) Yeah. (SP:PS532) My feet ache. (SP:PS532) Well I can't cook cos I've got (SP:PS535) Well my feet ache as well. (BNCKCAS_Conv) To my mind, this event makes the classification of well as a DM difficult, especially taking into account that these functions work on different levels. However, despite sharing the label frame marker, in extract (13) we find a different use. As Jucker (1997: 93) claims, it introduces direct reported speech separating the reported speech from the immediately preceding reporting clause “she said”. Furthermore, he argues that this practice implies a deictic reorientation (Jucker, 1997: 93). In my view, this happens due to the action of reporting a speech, which entails a reformulation of the original utterance, not due to the presence of well. It is to be borne in mind that well in these cases is likely to occur preceded by reporting verbs such as “said” or “suggested”. In those cases in which well is used as a pause filler, it implies a use that suggests bridging interactional silence: (15) they're always mucky! (pause) I think they're horrible they are! (SP:PS03S) Well they look mucky in the shops! And that's (SP:PS03T) Well no they do. (BNCKBBS_Conv) Based on Jucker (1997: 95), well as a pause filler is used in order to bridge interactional silence, further arguing that it signals a speaker’s claim to the floor. I do believe, as well, that the main purpose of using well as a pause filler is claiming the floor, however, its particular purpose may differ from one situation to another. In extract (15) the speaker is just claiming the floor because there is something else to be added to the conversation. Therefore, well signals the switch of turns between speakers and who is the one who holds the floor at a specific time. Analysing the data retrieved from both corpora, I came to the conclusion that, in these cases, well is likely to appear, !56 for instance, followed by I think, I suppose or I believe. Whether these collocations are present or not, well, in my view, is used with the aim to add a personal opinion, as in (15) and (16). (16) found she couldn't get anything better. That's probably nearer the truth (SP:PS4BU) Well I think, I've got ta say that I think it might be ha—. (BNCJN7S_meeting)! However, something that called my attention was that there are also instances where we find well followed by, for instance, I think, I suppose and I believe, in which it functions as a mitigator, therefore, this will later on explained. Although well is used in order to claim the floor, there are also instances in which it does not work, in other words, it does not fulfil its initial function: (17) next year. (SP:PS27U) Mhm. (SP:PS27T) Have we decided for next year? (SP:PS27R) Well (SP:PS27S) Well I think we are round to any other business I suppose. (SP:G59PSUNK). (BNCG59S_meeting) In extract (17), we can clearly see that the use of well in this situation does not work as it is intended to, since the speaker is not allowed to say anything else, on the contrary, the interaction is interrupted by another speaker. (18) that subsidy should continue? (SP:PS5VG) Well, whether I think it should or (SP:PS5VF) Well, but I ask you that one. (SP:PS5VG) Whether I think it should or. (BNCKRKS_brdcast_discussn) Extract (18) can be considered as a good example of how well may be used in order to claim the floor and interfere in a conversation while somebody else is speaking. The speaker (SP:PS5VG) uses well as a way to claim the floor and interact in the conversation, however, the exchange is interrupted by the other speaker, (SP:PS5VF). The latter of the speakers uses well in order to interrupt the person who is speaking. This event is, to my mind, an illustrative example since the speaker (SP:PS5VG) cannot finish the utterance at a time, and finishes it once (SP:PS5VF) has intervened. !57 These two are not, in my opinion, the two single ways of using well as a pause filler, as in my estimation, another way of claiming the floor while using well is as follows. (19) just because of promises either, you know, not generally. (SP:PS03W) No. Well no, I mean, a-- (pause) tha-- even tha-- that one the other day. (BNCKBDS_Conv) In extract (19) what the speaker is trying to do is to claim the floor but, actually, what I believe is that well is used in order to keep the floor in situations of hesitation. The speaker is, then, looking for something to say or searching the right words. To my knowledge, the speaker is trying to bridge the silence as Jucker proposes, but the reason why well is used is that the conversation needs to be reorientated. This occurs since we are tackling spoken contexts, a context in which participants do not have time to think about what they are going to say. This event prompts the appearance of well, in these cases, accompanied by many other DMs such as I mean, as we can see in the example, actually or yeah, filling up the gaps within the conversation. Furthermore, it is also probable to find it followed or proceeded by several repetitions, which point out the lack of time, hesitation, and the reorientation of the utterance. Likewise, it is highly probable to find well surrounded by pauses, as the one in the example above. On the other hand, I will deal with the category of mitigator that, to my mind, works on the interpersonal level (see section 3.3.2). On this level, well can be used as a way to save someone own’s face or someone else’s face, with the purpose of being polite, as in the following extracts: (20) which they are funded and (SP:PS5VF) Do you think that subsidy should continue? (SP:PS5VG) Well, whether I think it should or (SP:PS5VF) Well, but I ask you that In extract (20) I believe that well is used with the purpose of mitigating, of not compromising someone own’s face. The answer to a yes/no-question is avoided by breaking Griece’s maxims, above mentioned. The speaker in this extract tries to avoid to answer by using well and replying with something that is not to be expected from such a question. !58 (21) firm and I think that you have a feather one haven't you? (SP:PS1F2) Well I always have a feather one and I, I have (SP:PS1F1) Yeah (SP:PS1F2) two. (BNCKCSS_Conv) In extract (21), yet, well is used with the purpose of mitigating, but in this case, the speaker (SP:PS1F2) tries not to compromise someone else’s face, the face of the speaker (SP:PS1F1), indeed. The speaker (SP:PS1F2) not wanting to attack the image of the others uses well and asserts that not only one, but two feathers. Furthermore, the absence of well, in this case, would lead to impoliteness by the speaker (SP:PS1F2). This event, made me reflect upon the absence of well in those cases in which, in the event of being missing, it would lead to impoliteness: (22) are! (SP:PS03S) Well they look mucky in the shops! And that's (SP:PS03T) Well no they don't, not originally. (SP:PS03S) Well they don't look like. (BNCKBBS_Conv) This extract illustrates that in the event that well had been missing, it would have led to an impolite behaviour. I consider this assumption relevant since it is likely to happen in those cases in which one of the speakers is opposing any other speaker in the conversation, for instance, invitations that result in rejections. In other words, the absence of well in such cases would lead to changes in the politeness level of the utterance or even of the conversation. As a conclusion, I believe that the two ways of using well as a mitigator are bounded to the event of adding strain or taking it away from the conversation. Thus, they are supposed to happen since they mitigate any kind of bad feeling, misunderstanding, disagreements, refusals, rejections and clashes within a conversation, keeping the image of all the participants safe and avoiding the tone of the conversation to be elevated. To put an end to my analysis, I will show one of the examples that I previously mentioned, in which I notice, a certain degree of overlap between the two categories, mitigator and pause filler: 15 The example here included (23) has been extracted from the Corpus of Contemporary American English and it has 15 been referenced according to the codes provided by this corpus. !59 (23) TAVIS-SMILEY-1HOS# Those persons are Africans, though. We should point that out. GRETA-VAN-SUSTEREN# Well, yes, but the thing is that, also, I should say -. (COCA2012SPOKABC_ThisWeek) In extract (23), I consider well to be used as a mitigator, however, it was difficult to determine whether it was pause filler or mitigator. Likewise, it could also be interpreted as pause filler, yet, as in many other cases, everything I could do was to resort to the context in order to conclude its use. Thus, in a wide number of occasions, I had to rely on the context in order to classify the uses of such a versatile DM, well. !60 7.0. Summary and concluding remarks The main purpose of this dissertation was to study the nature and function of discourse markers in present-day written and spoken English, with particular attention to one of the most frequent in both contexts, ‘well’. This means to present a proposal that may clarify the complex status of these items. DMs have been investigated in this dissertation by considering the work of some linguists who over the last few years have shed some light on this topic. Thus, the first chapter reviewed the main theories and research studies conducted to clarify the status of these words. First of all, I relied on the arguments provided by Jucker (1998), Schourup (1999), and Schiffrin (1987) that let me make a choice as regards terminology, ‘discourse marker’. Taking Brinton’s 1996 work as a starting-point, in order to shape a definition of DMs, I displayed a general list divided into five different approaches as to what the main functions of discourse markers could be. As I pointed out above (see section 3.1), the range (A-E) of this list is regarded as the primary set of functions by a wide amount of scholars. Section 3.2, taking Schiffrin’s 1987 and Juckers’ 1998 works as a referential point in the light of the effect they caused on this field of study of discourse markers, presented the main features of DMs. Section 3.3 was devoted to the discussion of the main functions based mainly on the arguments supplied by Brinton (1996) and Fraser (1999). The latter gained importance in section 4.0 where the taxonomy of discourse markers was dealt with. Finally, section 5.0, based on what Schiffrin pointed out in her 1987 work, was devoted to the contexts of occurrence broadly speaking, and to the importance of intonation in speech. After this brief review of the theoretical background, I analysed in close detail the ‘well’ as a DM in spoken and written British and American English. For this purpose, I selected two main corpora, the BNC (British National Corpus) and COCA (Corpus of Contemporary American English) (see section 6.2). !61 Their different registers in both spoken and written language seemed highly suitable for this in-depth analysis given that the main objective was to observe up to what extent well is used in present-day British and American English, according to the context of occurrence. After a general review of the DM well, all the potential occurrences of this selected DM in the BNC and COCA were examined and contrasted (see section 6.3.3; 6.3.4). The second step was to separate those hits that could not be considered as actual cases of DM. Therefore, they were addressed to as false positives, separated from those that represent true cases of this DM, hence, not taken into consideration in my analysis. The classification of the uses of well was carried out relying on Jucker (1997); however, discrepancies were found and commented in section 6.3.2. One of the most immediate conclusions drawn from the analysis of well in BNC and COCA is that it is far more frequent in spokenness rather than in writtenness. A reasonable explanation could be that DMs are, to some extent, more characteristic of spoken language and related to a certain degree of informal registers. Another issue that my data shows and, thus, undeniable is that well as a DM works on the textual level and that applied to both media of expression, speech and writing. For further observations see Discussion section, 6.5, where I discuss the different uses of well in particular instances. Nevertheless, the data used for this dissertation, although useful to obtain a glimpse of the behaviour of the DM well in spoken and written British and American English may be insufficient to make solid statements about their use in present-day English. It would be very interesting to carry out further research by expanding the sample selected and by using other types of corpora that may contain meta-information about the speakers. Attention could also be paid to how well is used in ELF instances and whether they are used in the same way a native speaker does. !62 !63 1980 Well in Conversation. In: Greenbaum, Sidney, Leech, Geoffrey & Svartvik, Jan (eds.), Studies in English Linguistics for Randolph Quirk. London: Longman, 167–177. Svartvik, Jan & Stenström, Anna-Brita. 1985 Words, Words, Words: The Rest is Silence. In: Bäckman, Sven & Kjellmer, Göran (eds.), Papers on Language and Literature Presented to Alvar Ellegård and Erik Erykman. Göteborg: Acta Universitatis Gothoburgensis, 342-353. Zwicky, Arnold. 1985 Clitics and Particles. Language 61: 283-305. !70 Appendix A: well in BrE !71 Nº Genre Function Level Label Remarks 1 KBB S_conv they're always mucky! (pause) I think they're horrible they are! (SP:PS03S) Well they look mucky in the shops! And that's (SP:PS03T) Well no they do DM Textual Pause Filler Claiming the floor. 2 KBB S_conv are! (SP:PS03S) Well they look mucky in the shops! And that's (SP:PS03T) Well no they don't, not originally. (SP:PS03S) Well they don't look like DM Interpersonal Mitigator It is not polite just saying no. 3 KBB S_conv ! And that's (SP:PS03T) Well no they don't, not originally. (SP:PS03S) Well they don't look like there's, the boots (SP:PS03T) Well they're damn DM Textual Pause Filler Claiming the floor. 4 KBB S_conv originally. (SP:PS03S) Well they don't look like there's, the boots (SP:PS03T) Well they're damn clean! Now, nip all that up and eat it. DM Textual Frame Marker New Topic 5 KBB S_conv that! (SP:PS03T) It's not. French do it all the time! (SP:PS03S) Well we're not French. (SP:PS03T) Ah! You want to eat more bread you DM Interpersonal Mitigator Politeness 6 KBC S_conv 's (pause) fine now. No problems. He's great! He's sleeping well. (SP:PS1A9) Yeah. Elaine said (SP:PS1AD) Altogether (SP:PS1A9) he had a nightmare last night False Positive Propositional - 7 KBD S_conv just because of promises either, you know, not generally. (SP:PS03W) No. Well no, I mean, a-- (pause) tha-- even tha-- that one the other day DM Textual Pause Filler Restructuring speech 8 KBD S_conv Yeah but they were we-- like they shot Catholics (unclear) (SP:PS03W) Yeah. (SP:PS03Y) as well! (SP:PS03W) (laughing) That's right. Yeah I know. (SP:PS03Y) (laugh) (SP:PS03W) Yeah False Positive Propositional - 9 KBE S_conv her friend died in the (unclear)2. (SP:KBEPSUNK) (unclear) (SP:KBEPSUNK) Listen me Blanche she said, well I knew the truth she said (unclear) was a friend to me. (SP:KBEPSUNK) Yeah, DM Textual Frame Marker Reported Speech. Precedes she said. She goes to I. 10 KBG S_conv (SP:PS051) (unclear) (SP:PS052) Funny that in n it? (SP:PS051) (laughing) Andrew? (SP:PS052) Well well (laugh) sort out what's like worried him. (pause) I never rememb-- what did DM Textual Pause Filler Even 2 well’s + pause (laugh) 11 KBK S_conv certain point but they were always making a loss in the past and (unclear) (SP:PS05X) Well their, their accounting was cock-eyed in commercial terms Lynda. It, it really DM Interpersonal Mitigator Face saving 12 KBK S_conv think it's a bit dubious what they do with the profits after. (SP:PS05X) Well. Oh that, that's a different matter. That is a (pause) surely DM Interpersonal Mitigator Saving someone own’s face 13 KBL S_conv (SP:PS06A) And she can't afford anything for freebies definitely! (SP:PS06G) Freebies? (SP:PS06A) Well I mean any sort of food or anything. (SP:PS06G) She's said I'm DM Textual Pause Filler Restructuring speech 14 KBL S_conv Typical man! (SP:PS06G) (unclear) (SP:PS06A) Expecting her to cook for you food! (SP:PS06G) Well if I'm there and I'm a guest (pause) if she cooked for her DM Textual Pause Filler Claiming the floor. 15 KBW S_conv on a Sunday (SP:PS08P) And (SP:PS087) afternoon. (SP:PS08P) you've been before? (SP:PS087) Well we've never been this place before, but we've been on a similar DM Interpersonal Mitigator Someone’s own face 16 KBW S_conv been on a similar do. (SP:PS08P) And what do you do then? (SP:PS087) Well, we have (SP:PS08P) Yeah. (SP:PS087) all sorts of things organised (SP:PS08P) Yeah, DM Textual Pause Filler Restructuring speech. Well + pause (comma). 17 KC0 S_conv that phrase used but I don't know what it's for. (SP:PS096) Oh well it's, it's where (pause) they say (pause) say a sequence of underground DM Textual Pause Filler Restructuring speech 18 KC4 S_conv Beefeaters. (SP:PS1EB) Did he? Oh yeah. (SP:PS1E4) The Tower of London. (SP:PS1EB) Well I knew, when I was in, when we were at Donnington, living DM Textual Pause Filler Claiming the floor. 19 KCA S_conv always thought for a, a woman who wears (pause) really loads of make-up (SP:PS0DP) Well (clears-throat) (SP:PS0DN) and then they sleep with a fella right, honestly (SP:PS0DP) And they DM Textual Pause Filler Claiming the floor + pause (by means of clearing the 20 KCA S_conv fella right, honestly (SP:PS0DP) And they wake up and it's terrible. (SP:PS0DN) Well I mean that fella must (SP:PS0DP) And they (SP:PS0DN) look around and think (pause) tt DM Textual Pause Filler I mean, reformulates. 21 KCA S_conv 's like a land-- but Barry (-----) is a landscape (SP:PS0DL) (unclear) (SP:PS0DP) gardener as well isn't he? (SP:PS0DL) (unclear) a coalman, a landscape gardener you name it False Positive Propositional - 22 KCF S_conv (SP:PS1ES) Well (pause) I've tried it occasionally. (pause) Might be Conservative (unclear)2. (SP:PS1ET) Well (unclear) usually comes. (pause) A-- (pause) at er (pause) (unclear)2. (SP:PS1EN) Who? (pause) DM Textual Pause Filler Claiming the floor. 23 KCF S_conv (pause) And I didn't canvas for him, I just marked up. (SP:PS1EP) Well this (SP:PS1EN) Oh well this was only for the local! (SP:PS1EP) Who did you DM Textual Pause Filler Claiming the floor. 24 KCF S_conv n't canvas for him, I just marked up. (SP:PS1EP) Well this (SP:PS1EN) Oh well this was only for the local! (SP:PS1EP) Who did you canvas for? (SP:PS1EN) DM Textual Pause Filler Restructuring speech 25 KCH S_conv the butter? Thank you. I must admit I'm very hungry. (SP:PS1BT) Well so you should be. Working at the garden. (pause) Like you've had DM Textual Pause Filler Claiming the floor. 26 KCN S_conv Which is the one nearest Cockney's? (SP:PS0FF) Electric. (SP:PS0FG) Electric. (pause) Well the people going in and out there as well. (SP:PS0FF) They're paying their DM Textual Frame Marker New Topic 27 KCN S_conv Electric. (SP:PS0FG) Electric. (pause) Well the people going in and out there as well. (SP:PS0FF) They're paying their bills, they're (pause) but there's nothing False Positive Propositional 28 KCS S_conv (SP:PS1F3) You do get a lot of fibre now (SP:PS1F2) yes, the, yeah well, well it's like a spongy now (SP:PS1F3) Yeah yeah (SP:PS1F1) Yeah (SP:PS1F2) We DM Textual Pause Filler Restructuring speech 29 KCS S_conv do get a lot of fibre now (SP:PS1F2) yes, the, yeah well, well it's like a spongy now (SP:PS1F3) Yeah yeah (SP:PS1F1) Yeah (SP:PS1F2) We have two DM Textual Pause Filler Restructuring speech 30 KCS S_conv firm and I think that you have a feather one haven't you? (SP:PS1F2) Well I always have a feather one and I, I have (SP:PS1F1) Yeah (SP:PS1F2) two DM Interpersonal Mitigator Not just 1, but 2. Saving someone else’s face 31 KCS S_conv me head (SP:PS1F1) Yeah I think feathers are more yielding aren't they? (SP:PS1F3) Well I've er, I've er fibre I've a fibre underneath and a DM Textual Pause Filler Restructuring speech. Doubting (er, er) 32 KCT S_conv ta work all day Saturday you see? (SP:PS0GA) Yeah you said that. (SP:PS0FP) Well that's something to do with they're working this Saturday and then they break DM Interpersonal Mitigator Sb telling you that you said sth so you do not want to 33 KCT S_conv Oh I see, yeah. (SP:PS0FP) Something like that, so (pause) I said well that's a (pause) good job I'm going this Saturday in n it really DM Textual Frame Marker Linked to I said 34 KCW S_conv We'll do that picture! Straight after registration we're doing it? (SP:PS0H8) Well I'm not. I've got a, I've got (SP:PS12C) I am. DM Interpersonal Mitigator Negation 35 KD0 S_conv that, do they have those in Scotland? (SP:PS0HM) (unclear) (SP:PS0HN) (unclear) may as well (SP:PS0HM) Word of warning, I'll give you a word of warning, I False Positive Propositional 36 KD1 S_conv , erm what other ones do you learn? (SP:PS0JB) I don't know (SP:PS0JA) Well you're the ones that go to playschool I don't (SP:PS0JB) the one that DM Interpersonal Mitigator You are the ones going there, you should know. Face 37 KD5 S_conv I haven't done anything have I Colin? (SP:PS0K0) Yeah, eat my dinner well. (SP:PS0JX) (unclear)2. (SP:PS0K0) (unclear)2. (SP:PS0JX) Cos (unclear) do it. (SP:PS0K0) (unclear)2. (SP:PS0JY) She Unclear 1 !72 38 KD8 S_conv do that then? (SP:PS0LK) The full turn in (pause) count. (SP:PS0LT) Suppose, well (pause) supposing we count ins and outs and the (SP:PS0LK) Yeah. (SP:PS0LT) sample. DM Textual Pause Filler Restructuring speech. Well + pause 39 KDB S_conv and you're just adding bits (SP:PS0KY) No. (SP:KDBPSUNK) to it or something. (SP:PS0KY) Well no I did it last week and one of the references I couldn't remember DM Interpersonal Mitigator +Negation 40 KDJ S_conv minds up (SP:PS0N4) That's right (SP:PS0N3) er and I sort of, I thought well you know, I mean, I sort of went of that erm and I DM Textual Pause Filler +DMs 41 KDJ S_conv , right, and, and if you actually knew what you were doing and well aware of what you're doing you would do it a hundred times better and False Positive Propositional 42 KDN S_conv were a new one. (SP:PS1HP) What are all those different colours for? (SP:PS1HN) Well it's (pause) blue. (SP:PS1HP) Mm. (singing) Doo be doo doo. (pause) You DM Textual Pause Filler Restructuring speech. Well + pause 43 KE0 S_conv lot of this then? (SP:PS0SX) Wednesday. Wednesday, it must have Thursday, well (pause) he sat in the store as they called him (unclear)2. (laugh) (SP:PS0SY) Mm. (SP:PS0SX) DM Textual Pause Filler Restructuring speech. Well + pause 44 KE6 S_conv to bits, try to move everything in here that she wouldn't chew (SP:PS0X9) Well you couldn't (SP:PS0X8) you couldn't do it could you? (SP:PS0X9) No, DM Interpersonal Mitigator +Negation 45 KP0 S_conv get bust and erm (SP:KP0PSUNK) He'll sit on them. (laugh) (SP:KP0PSUNK) (unclear) (SP:KP0PSUNK) Well no but if I do bust' em (SP:KP0PSUNK) Whether or not they are easily DM Interpersonal Mitigator +Negation 46 KP9 S_conv the door open with it. (SP:PS535) Yeah. (SP:PS532) My feet ache. (SP:PS532) Well I can't cook cos I've got (SP:PS535) Well my feet ache as well DM Textual Frame Marker New Topic. It could also be seen as a mitigator as the person is 47 KP9 S_conv My feet ache. (SP:PS532) Well I can't cook cos I've got (SP:PS535) Well my feet ache as well. I, my legs as well. (SP:PS532) Twenty DM Interpersonal Mitigator I’m sorry to let you know that mine are also in 48 KP9 S_conv Well I can't cook cos I've got (SP:PS535) Well my feet ache as well. I, my legs as well. (SP:PS532) Twenty past twelve we finished last False Positive Propositional 49 KP9 S_conv 've got (SP:PS535) Well my feet ache as well. I, my legs as well. (SP:PS532) Twenty past twelve we finished last night. (SP:PS535) That's what I False Positive Propositional 50 KPG S_conv (SP:PS6R3) So what's the difference between black (unclear) and (pause) Malcolm X? (SP:PS555) Well actually I like Margaret Thatcher cos she's a (SP:PS55A) Well there's more violence DM Textual Pause Filler Restructuring speech. 51 KPG S_conv Malcolm X? (SP:PS555) Well actually I like Margaret Thatcher cos she's a (SP:PS55A) Well there's more violence in it. (SP:PS555) white dude! (SP:PS55A) And th--, DM Textual Pause Filler Interfere in somebody else’s speech 52 KST S_conv yes? (SP:PS6RG) Yeah. (SP:PS6TH) Sh-- She told me er (SP:PS6RG) She might as well be a (unclear)2. (SP:PS6TH) Don't know. (SP:PS6RG) Why? (SP:PS6TH) She doesn't False Positive Propositional 53 KST S_conv you not got (unclear)? Oh I don't understand that! Oh! (SP:PS6TH) Well she says (unclear) (SP:PS6RG) Oh! (SP:PS6TH) now (unclear) (SP:PS6RG) Oh! Oh dear! DM Interpersonal Mitigator Face saving. She is saying that that is what some other 54 GYU S_intervi ew_oral_ history was the m-- the president of the area council, myself and the secretary. Well we er (unclear) duly arrived down at quarter to ten as (unclear) twenty six of DM Textual Frame Marker New Topic 55 GYU S_intervi ew_oral_ history meet you until eleven o'clock. An-- an-- (unclear) of course and naturally I says well why, why is he getting (unclear)? Is he in the building? No DM Textual Frame Marker Reported Speech 56 H5G S_intervi ew_oral_ history (cough) (SP:PS22C) You remember Cliff Quay being developed? (SP:PS22D) I remember it yes. Well before I went to Hemel Hempstead, my father was dredging the first part of DM Textual Pause Filler Restructuring speech 57 HDM S_intervi ew_oral_ history Mm. (SP:PS22P) he had a way with him and he was, got on very well with Mr (-----) and er he told us afterwards that erm the whole trouble was False Positive Propositional 58 J8G S_intervi ew_oral_ history Harlow that sold (SP:PS1N4) Oh yes at one time erm after the New Town got well started, after The Stow was built got to the, goes up to The False Positive Propositional 59 J8G S_intervi ew_oral_ history built got to the, goes up to The Stow and be-- beyond er, well it's just, going on into erm, er, towards the swimming bath DM Textual Pause Filler Restructuring speech 60 K6M S_intervi ew_oral_ history e-- in given that there weren't many women at that time er in (SP:K6MPS000) Well (SP:PS5M1) a a manual working environment if you like. (SP:K6MPS000) At (-----) aye and DM Textual Pause Filler Claiming the floor. 61 F7C S_meetin g door (SP:PS1LL) The blind people respond really (SP:PS1LK) wanted to come back, yeah (SP:PS1LL) well to somewhere they've been (SP:PS1LJ) Yeah (SP:PS1LK) That's right (SP:PS1LL) and if they DM Textual Pause Filler Claiming the floor. 62 G59 S_meetin g next year. (SP:PS27U) Mhm. (SP:PS27T) Have we decided for next year? (SP:PS27R) Well (SP:PS27S) Well I think we are round to any other business I suppose. (SP:G59PSUNK) DM Textual Pause Filler Claiming the floor. 63 G59 S_meetin g . (SP:PS27U) Mhm. (SP:PS27T) Have we decided for next year? (SP:PS27R) Well (SP:PS27S) Well I think we are round to any other business I suppose. (SP:G59PSUNK) (laugh) (SP:PS27S) DM Textual Pause Filler Claiming the floor. 64 HYF S_meetin g se-- extend that er, later in the season. So, it's going well. (SP:HYFPSUNK) Erm, I seem to remember that when we met earlier in the False Positive Propositional 65 HYF S_meetin g right and I remember it correctly, whether that's still the view? (SP:PS303) Well, I think that I bore people endlessly by saying that erm, the first DM Interpersonal Mitigator Saving someone own’s face 66 J3T S_meetin g mean, otherwise one assumes simply that people don't know. (SP:PS3NV) Mm, well th-- the answer to the first question is that, no we don't have DM Textual Pause Filler Restructuring speech. Doubting (Mm) 67 JJG S_meetin g plus an Annual General Meeting of it's full council. (SP:PS45S) Miss (-----) (SP:JJGPSUNK) Well, I mean, I think, er from, I mean certainly I do DM Textual Pause Filler Restructuring speech 68 JN7 S_meetin g found she couldn't get anything better. That's probably nearer the truth (SP:PS4BU) Well I think, I've got ta say that I think it might be ha-- DM Textual Pause Filler Claiming the floor. 69 JNK S_meetin g 'm very much looking forward to being a member of N C V O as well as being a consumer of its excellent services, and while I've got the False Positive Propositional 70 JS9 S_meetin g rather hope they do, it'll be a little while yet, we could well see a change in the security of tenure issue, which might make it easier False Positive Propositional Nº Genre Function Level Label Remarks 2 !73 71 KLV S_meetin g not back till after Easter anyway. (SP:PS3SK) So when's that happening? (SP:PS3SF) Well I'll leave that (pause) problem for Mary to sort out. But it, DM Interpersonal Mitigator Saving someone own’s face 72 KLV S_meetin g what do we do? Now (SP:PS3SH) Sorry what reports are they? (pause) (SP:PS3SF) Well Phil knows what they are. (SP:PS3SL) Well we've (pause) we're gon na DM Interpersonal Mitigator Saving someone own’s face 73 KLV S_meetin g what reports are they? (pause) (SP:PS3SF) Well Phil knows what they are. (SP:PS3SL) Well we've (pause) we're gon na spend it now anyway. No I mean DM Textual Pause Filler Restructuring speech. Well + pause 74 FM5 S_classro om you could say what it looked like and what was happening and then say, Well I'm not quite sure what the convention is for writing them (SP:PS1SC) Right. DM Textual Frame Marker Reported Speech 75 FM5 S_classro om quite sure what the convention is for writing them (SP:PS1SC) Right. (SP:PS1SB) down. Well that's fine. I mean the convention is (pause) each of these ex-- I DM Interpersonal Mitigator Saving someone own’s face 76 FMJ S_classro om and they said I want to work out this angle. (SP:PS1T2) Mhm. (SP:PS1T1) Well there's you right angle that's still the hypotenuse which one would be the DM Textual Pause Filler Claiming the floor. 77 G3W S_classro om you come across the one that (SP:PS1VN) Erm (pause) instrument that I know. (SP:PS1VM) Well that, that you might know. (pause) It's not actually a percussion one DM Textual Pause Filler Restructuring speech. 78 JJS S_classro om Coming together. They're coming together. That's also in six, oh well then so the two boys raced down the harbour (SP:PS47L) With the (SP:PS47K) Where the DM Textual Pause Filler Restructuring speech 79 JSU S_classro om (cough)2. (SP:PS4RA) Lucy. Any other reasons why it's a shadow? (SP:PS4RF) (pause) Well it's shy isn't it, so it's got to try to keep DM Textual Frame Marker New Topic 80 F8R S_lect_h umanitie s_arts democracies erm aren't the same as we are, they're (unclear) (SP:F8RPSUNK) Yes well it's (unclear) (SP:F8RPSUNK) (unclear) (SP:F8RPSUNK) And there, there are shades of democracy, DM Textual Pause Filler Restructuring speech 81 G5V S_consul t twenty four hours yes. (SP:PS2C3) Oh aye, (unclear) yes (unclear) know them very well. (SP:PS2C1) if he has any queries he can give me a ring but (SP:PS2C3) False Positive Propositional 82 G5V S_consul t me a ring but (SP:PS2C3) I mean (unclear) (SP:PS2C1) Yeah. (SP:PS2C3) know me very well. (SP:PS2C1) Yeah. (SP:PS2C3) (laugh) (unclear) (SP:PS2C1) So (unclear) (SP:PS2C2) Yeah. (SP:PS2C1) You False Positive Propositional 83 KRT S_brdcas t_news mile tailback in both directions. You are also advised to avoid that area as well. On the M twenty five its very slow anti-clockwise, junctions twenty to nineteen False Positive Propositional 84 FLR S_brdcas t_discuss n (SP:FLRPS000) Has anyone, I mean has anyone had that experience? Yes? (SP:FLRPS005) Well I got a video (pause) and, like, that's all we got a DM Textual Pause Filler Claiming the floor. 85 FLR S_brdcas t_discuss n are actually (pause) given any instruction on how to use them? Yes? (SP:FLRPS006) Well (pause) yesterday we actually went to (pause) to a school and (pause) you show them DM Textual Pause Filler Restructuring speech. Well + pause 86 HMP S_brdcas t_discuss n at Lloyds who stake everything they have in order to rake in enormous profits. Well they had a particularly bad time many of them lost absolutely everything and now they DM Interpersonal Mitigator Saving someone else’s face 87 KJS S_brdcas t_discuss n Russians indeed feel oppressed in the Baltic and, and probably in other Republics as well. I'm not sure that withholding diplomatic recognition is the best way to approach False Positive Propositional 88 KM2 S_brdcas t_discuss n Yes that's right. (SP:PS30G) Which animal charities are you interested in? (SP:PS30E) Well mine's the National Canine Defence League. (SP:PS30G) I see now that's one DM Textual Pause Filler Claiming the floor. 89 KM2 S_brdcas t_discuss n D S A but the National Canine Defence League what makes it different? (SP:PS30E) Well we have a non-destructive policy. But I've been a member for many years DM Textual Pause Filler Claiming the floor. 90 KM2 S_brdcas t_discuss n as they haven't got anyone in this area. We've been going since well September eighty three and I agree with what you say. When we first started False Positive Propositional 91 KRK S_brdcas t_discuss n which they are funded and (SP:PS5VF) Do you think that subsidy should continue? (SP:PS5VG) Well, whether I think it should or (SP:PS5VF) Well, but I ask you that DM Interpersonal Mitigator Saving someone own’s face 92 KRK S_brdcas t_discuss n that subsidy should continue? (SP:PS5VG) Well, whether I think it should or (SP:PS5VF) Well, but I ask you that one. (SP:PS5VG) Whether I think it should or DM Textual Pause Filler Claiming the floor + interference in someone else’s 93 KRK S_brdcas t_discuss n mean for these people who are listening and and involved with these schools? (SP:PS5VG) Well that means that (pause) I personally believe that the problems which many small village schools DM Textual Pause Filler Restructuring speech 94 KS7 S_brdcas t_discuss n a prize. And not only that, there are four runners up prizes as well to be won, and these are a copy of the Time Form's' False Positive Propositional 95 KS7 S_brdcas t_discuss n on the flat in England in 1990, and lot's more as well. Well that's going free to the first person to ring us on three double one DM Textual Pause Filler Claiming the floor. 96 HYU S_speech _unscript ed lot of time on this (pause) because it's covered on various other courses as well. Things like appraisal skills er level one level two, various others, other False Positive Propositional 97 HYU S_speech _unscript ed run. There's your job (pause) the first thing I should do is say well what are the key tasks, what are the areas that would mean results? DM Textual Frame Marker Reported Speech 98 K6Y S_speech _unscript ed I'm pretty sure it said two hundred pounds a week. (SP:PS5MU) Right. Well the actual State er the statutory sick pay is fifty six pounds ten pence per DM Interpersonal Mitigator Saving someone else’s face 99 K6Y S_speech _unscript ed , so how do you know you're right and I'm wrong? (SP:PS5MU) Well that is the, that is the erm the government guidelines that are actually set DM Textual Pause Filler Restructuring speech 100 JT5 S_unclas sified from there. (SP:JT5PSUNK) Right, okay. (SP:JT5PSUNK) Alright? (SP:JT5PSUNK) Right. (pause) Well Ian we've, we've finished the completion of the document erm the planning DM Textual Pause Filler Restructuring speech 101 JT5 S_unclas sified to er increase me pension and (unclear) was very good. (SP:JT5PSUNK) Right. Okay well as I said previously what I'm gon na do is go away and then DM Textual Pause Filler Restructuring speech 102 JTE S_unclas sified there, hammer some more tenterhooks in somewhere else and put it on there as well. Erm even tapestries like the Devonshire hunting tapestry which you may have seen at False Positive Propositional 103 A0L W_fict_p rose your case, why not? You're stunning,'' I suppose. Well, you're not having my photo. I know what I'm giving you DM Interpersonal Mitigator Explaining himself. Saving his face. 104 A7J W_fict_p rose Sister Cooney took it and shook it warmly.' They'll look after you well there,' she smiled.' If they don't, let us know False Positive Propositional Nº Genre Function Level Label Remarks 3 !74 105 CKC W_fict_p rose accidental or deliberate? If we could hoist the Delos to the surface we might well find out.'' We might well indeed. Still, first things first False Positive Propositional 106 CKC W_fict_p rose the Delos to the surface we might well find out.'' We might well indeed. Still, first things first.' Hawkins looked briefly at the signals False Positive Propositional 107 AT7 W_fict_p rose is Conway. But what the hell is she doin' here?'' Well, you'd better ask her, Ma, hadn't you? As she DM Interpersonal Mitigator Saving someone own’s face. 108 BP0 W_fict_p rose marriage, I should have thought, is a false step you must have been well warned against. With such as you as his wife, how can a man False Positive Propositional 109 CJT W_fict_p rose , Stephen Holly's voice carried a curious brusqueness. What did he want? Well, you don't tell two old people that a year and a bit too DM Interpersonal Mitigator Politeness 110 CJT W_fict_p rose that. We worked very hard to get him out of the Soviet Union -- well, you know all that, Mr Carpenter will have told you, and he DM Textual Pause Filler Restructuring speech 111 FB9 W_fict_p rose 's the Ketterings' dog. It seems to have come home.'' Well, we can't be expected to look after it. Not on top of DM Interpersonal Mitigator Politeness 112 FEE W_fict_p rose you got to say to explain that place, I should like to know? Well, never mind. Egalitarianism's never been my strong suit, but I overcame DM Textual Pause Filler Restructuring speech 113 FR0 W_fict_p rose ' This is it, then,' she said.' We might as well face it.'' No way,' Bernice shouted bitterly.' I False Positive Propositional 114 FS3 W_fict_p rose you were the father of Jane Davenant's son.' Will laughed.' Well, well,' he said,' people say that, do they? DM Textual Frame Marker Reported Speech. Linked to “he said”. 115 FS3 W_fict_p rose the father of Jane Davenant's son.' Will laughed.' Well, well,' he said,' people say that, do they? What will DM Textual Frame Marker Reported Speech. Linked to “he said”. 116 G0E W_fict_p rose .' Both of you! We've more to worry about.'' Well, just keep an eye on it,' said Duvall, without taking his DM Textual Frame Marker Reported Speech. Linked to said Duvall 117 G1L W_fict_p rose said left and I said right.'' Which was it?'' Well, it was right, actually,' said Fenella.' We consulted the DM Textual Frame Marker Reported Speech. Linked to said Fenella 118 G3E W_fict_p rose 's right, mate. This member of your staff anyway. I know damn well what you're up to and I don't like it. So stow it False Positive Propositional 119 GV7 W_fict_p rose I think you are probably mad. But I will stay.'' Very well,' said the stranger.' Now remember your promise. You've been False Positive Propositional 120 GVT W_fict_p rose , when I was about fifty, and goin' through you know what, well this Sunday I was feelin' there wasn't much left to go on livin DM Textual Frame Marker New Topic 121 H86 W_fict_p rose snapped. The member stiffened, then edged away in a fading mum-ble:' Well, I suppose things must be rather trying for you, what with...' DM Textual Pause Filler Claiming the floor. 122 H8B W_fict_p rose know what I mean. Gave them a chance to test themselves. Death as well as life. Not so different, a soldier's job, from the clergy False Positive Propositional 123 HA5 W_fict_p rose .'' Why should I?' She assumed the air of sang-froid so well known to her acquaintances back in England.' I'm not so green that False Positive Propositional 124 HA7 W_fict_p rose 's hush.' Have you, now?' he said softly.' Well, well. It didn't take you long, did it? I'd DM Interpersonal Mitigator Taking strain away. Diminish the importance of 125 HA7 W_fict_p rose .' Have you, now?' he said softly.' Well, well. It didn't take you long, did it? I'd forgotten your DM Interpersonal Mitigator Taking strain away. Diminish the importance of 126 HGE W_fict_p rose and well cared for, he noted professionally. She was also a young person well accustomed to having her own way -- there was no doubt of that.' False Positive Propositional 127 HGJ W_fict_p rose She'd put her finger on it. Sometimes it takes people years.' Well,' said Ken,' don't expect me to look after Apricot while DM Textual Frame Marker Reported Speech. Linked to said Ken 128 HGM W_fict_p rose 've got to work pretty intensely with the team, and we need you as well. Don't worry about Jason; IMP have promised there won't be a False Positive Propositional 129 HGV W_fict_p rose she was still attired in the old cotton gown, now stained with blood as well as dirt. She sat, leaning her cheek on her hand, gazing down False Positive Propositional 130 HHA W_fict_p rose your mother and your brothers.' And as she remained stubbornly silent,' Well, at least you're going to take a look -- and see just what DM Textual Frame Marker New Topic 131 HWE W_fict_p rose .' She now poked her finger into Peggy's arm, saying,' Well, there's a cure for that, you know. Oh! Here they DM Textual Frame Marker Reported Speech. Preceded by “saying”. 132 J10 W_fict_p rose ' he said.' He's fixed it.' Owen nodded.' Well,' said Mahmoud,' it was always on the cards. He's DM Textual Frame Marker Reported Speech. Linked to said Mahmoud 133 FU4 W_fict_d rama . LADY DAVERS: Jackey, sit down and don't touch the creature. Well, child, how dost find thyself. Thou'rt got into a fool's paradise DM Textual Frame Marker New Topic 134 K1N W_news _script . He says he'd like Prince William and Prince Harry to enrol later as well. Then it was onto Oxford's more established University and a chance to meet False Positive Propositional 135 A3P W_news p_brdsht _nat_mis c better spending a weekend in Antwerp, about 15 miles to the east, less well preserved architecturally but with more comprehensive museums and galleries, and much more the feel False Positive Propositional 136 K4Y W_news p_other_ arts Northern Gains the Northern League magazine with which we have some slight connection.' Well produced and excellent value for money, it has the professional look you would expect False Positive Propositional 137 K5A W_news p_other_ social , jun, who took 77, were alongside Montgomerie, while Brian Marchbank did well to also join the qualifiers. He recovered from an opening 80 to beat par False Positive Propositional 138 BM4 W_news p_other_ report football and almost went further ahead when Whiston headed just over. Churt were competing well and also had their chances. After the interval, with the wind and slope False Positive Propositional 139 CBM W_news p_other_ report Day (Chester 6f, Gd-Sft). Previously progress from halfway, ran on well to lead last strides, bt JOHANNA THYME (rec 22lb) hd, with False Positive Propositional Nº Genre Function Level Label Remarks 4 !75 140 K97 W_news p_other_ report waited a long time. The saying goes that everything comes to he who waits well they've waited longer than most' Newcastle manager Kevin Keegan after his side's False Positive Propositional 141 A0V W_pop_l ore and I like to maintain my contact with the live action on court, as well as the monitored information in the press room. It pays to keep in touch False Positive Propositional 142 A7P W_pop_l ore bright future as less resilient designers wring their hands or close shop.' As well as opening more boutiques in England, France and Germany, I'm working on False Positive Propositional 143 ABD W_pop_l ore failure to hold a direct election for the presidency last year. Mr Gorbachev might well have won such a vote, and added to his formal powers the authority he False Positive Propositional 144 CKW W_pop_l ore # February's off-peak sales of Contemporary Art in New York produced respectable results as well as evidence that the price-gap between first-rate works and the rest may be widening. False Positive Propositional 145 C9E W_pop_l ore the 1990 Irish Amateur before turning pro, that he completely revamped his swing as well as changing his attitude to the game.' The difference is vast,' False Positive Propositional 146 C9K W_pop_l ore and saxophonist has lost none of his initial zeal for the Chapman Stick.' Well, I'm still using the standard setup: five strings on the treble, DM Textual Pause Filler Reported Speech 147 CAU W_pop_l ore years to age fifty, then annually to age 65. From age 65 they well be carried out every 6 months. # d) #, Renewal lung function False Positive Propositional 148 CB8 W_pop_l ore advice. # BODY TALK Babies communicate with body movements and facial expressions. As well as smiles and cries, look out for: #' I'm pleased to False Positive Propositional 149 EBW W_pop_l ore for the nearby Prado, which is urgently in need of more space, as well as the kind of modern facilities which have' put the Villahermosa among the ranks False Positive Propositional 150 ECX W_pop_l ore made fairly flat. The Chevron is certainly an interesting aeroplane. The compromises work well and whilst I am not an experienced glider pilot Derek Piggott believes that the type False Positive Propositional 151 G2S W_pop_l ore . The rest of the tent is then laid and pegged out. A clever well thought out design feature is the use of the bike pump as the upright internal False Positive Propositional 152 B2P W_ac_po lit_law_e du legacies was that the list for legacies was closed: there were set wordings, well established by long tradition. If they were not employed, that did not mean False Positive Propositional 153 CS6 W_ac_po lit_law_e du and Chinamen whom he considered were defiling the race, particularly in seaports, as well as the alleged Jewish menace. To Leese immigration and race-mixing was a jewish plot False Positive Propositional 154 EDD W_ac_po lit_law_e du . We may have mixed feelings about scientific triumphs which have included nuclear weapons as well as electric light but we can not deny the comprehensive debt we owe to the False Positive Propositional 155 EW7 W_ac_po lit_law_e du will be selected, and examined for its facets, concepts and possible sub-units as well as for the skills that are necessary for its full comprehension and others that can False Positive Propositional 156 FSS W_ac_po lit_law_e du user does not. Nevertheless in many factual situations, a plaintiff will succeed equally well either under the rule or in nuisance. Despite the original relationship of the rule False Positive Propositional 157 G1F W_ac_po lit_law_e du balance of different lessons and their place in the curriculum as a whole, as well as a very clear concern with progression, continuity, the acquisition of underlying skills False Positive Propositional 158 J79 W_ac_po lit_law_e du appropriate to such a transaction (19(JP)), requires execution by the transferees as well as by the transferor, because it contains a declaration that the survivor can ( False Positive Propositional 159 K8U W_ac_po lit_law_e du it had in the thirties. Codetermination received enormous support. In 1950 and 1951 well over 90 per cent of metal workers and miners voted in favour of strike action False Positive Propositional 160 E9V W_ac_hu manities _arts had a special place in the king's affections, and he had acquitted himself well in Aquitaine. In April 1372 he was appointed Lieutenant in Aquitaine and assembled a False Positive Propositional 161 GV0 W_ac_na t_science section. In the right type of country projected profiles may illustrate the relief remarkably well: in Fig. 9.17 it is obvious that there is an area of high ground False Positive Propositional 162 GV0 W_ac_na t_science of certain features of a map, which some people can probably detect just as well by scanning the map but others may not. The same is true for any False Positive Propositional 163 H9F W_ac_so c_scienc e of development has been broadened again to take account of social and cultural, as well as economic, features; and more attention has been given to development as a False Positive Propositional 164 BNB W_non_ ac_huma nities_art s help to clear the way for a successful invasion. Charles III of Spain was well aware that his country, unlike France, could not support a long, full-scale False Positive Propositional 165 CCN W_non_ ac_huma nities_art s play. She may serve a cause greater than herself. This could just as well be art as social work or joining a religious movement. She is uninterested in False Positive Propositional 166 CCS W_non_ ac_huma nities_art s Like almost all commando colonels, Colonel A.C. Newman was a man with technical as well as military ability: a civil engineer of cheerful but quiet character, seldom seen False Positive Propositional 167 CD8 W_non_ ac_huma nities_art s enough to allow quite large vessels to come very close to the castle-stacks, certainly well within range of their cannon and mangonels.' Booms,' Ramsay announced. False Positive Propositional 168 CDV W_non_ ac_huma nities_art s do not reach for another three chapters. Now this unnatural form of presentation works well for surprise and suspense. It is a shock to have the battle of Helm False Positive Propositional 169 CK2 W_non_ ac_nat_s cience was the sort of environmental planning that takes into consideration the needs of wildlife as well as those of humans. Indeed, even in 1991 there is little real evidence False Positive Propositional Nº Genre Function Level Label Remarks 5 !76 170 CEG W_non_ ac_nat_s cience difference is not usually of much consequence because in both cases the practical strengths are well below what they ought to be. The causes of weakness in adhesion are rather False Positive Propositional 171 HLH W_non_ ac_polit_ law_edu as the month progressed it became clear that the scandal involved opposition party members as well as members of the LDP. A senior SDPJ legislator, Ryoichi Yasutsune, agreed False Positive Propositional 172 B24 W_non_ ac_soc_s cience ' So I said,' Just coming out of my refreshments.'' Well,' he said' it's now six thirty-two. I'm reporting you DM Textual Frame Marker Reported Speech. Linked to “he said”. 173 CGD W_non_ ac_soc_s cience old flat. And then arrangements had to be made to transport his cats as well as himself to the new flat. By discussion, planning and reviewing the achievements False Positive Propositional 174 HJ1 W_non_ ac_soc_s cience and another. These comparisons and evaluations are also made by children. This is well recognized by parents who are invariably concerned that their children make friends with the' False Positive Propositional 175 HR0 W_non_ ac_soc_s cience revenue loss resulting from smuggling by reducing duty levels to lower its profitability, as well as by strengthening the preventative service. By 1790 the yield from wine duties had False Positive Propositional 176 CPN W_non_ ac_tech_ engin to what Advanced Micro says is cleanroom microcode, and other copyrighted programs, as well as the microcode that is acknowledged to come from Intel. Intel also notes that False Positive Propositional 177 CSN W_non_ ac_tech_ engin .' There is no negotiation with SunSoft,' Cunningham says. He is well aware however, that' if USL were to hold the keys for an industry False Positive Propositional 178 C8L W_religi on the building have been drawn in. There will usually be those who are quite well disposed towards the church but who are drifting. Some may be converted and disillusioned False Positive Propositional 179 CL2 W_biogr aphy was crazy about football. He idolised Ifor. He had a bad back as well, you know, just like Rich. He injured it in the Normandy landings False Positive Propositional 180 B3H W_biogr aphy given clear directions from where the picture should be painted. The cloud cover as well as the atmospheric conditions are precisely defined. He suggested the time of day and False Positive Propositional 181 BN3 W_biogr aphy of the heat would then be dissipated. A question of: you may as well have it, for' it's on the table'. No! Anything False Positive Propositional 182 G39 W_biogr aphy by the constant treadmill of work on the land. During Hannah's childhood and well beyond, Dalesfolk lived in a closed world where to travel more than a dozen False Positive Propositional 183 CBV W_com merce table below shows HM Treasury forecasts for growth in components of aggregate demand, as well as total gross domestic product and manufacturing output, made at the time of the False Positive Propositional 184 H0C W_com merce rare, although poor visibility caused through low cloud and rain is common (the well known Scotch Mist). However, we ran out of the fog somewhere near False Positive Propositional 185 J6S W_com merce Act 1986, the board of directors will, if insolvency is imminent, be well advised to put the interests of creditors (including those of employees) before those False Positive Propositional 186 A7F W_misc in the Telford area. These hotels are offering competitive corporate business rates, as well as superior accommodation and leisure facilities which the Valley Hotel does not have. But False Positive Propositional 187 APC W_misc rainforest that have been definitely aged, in terms of their lifespan. TOM: Well, is there a time when Big Tree stops being a plant and starts being DM Textual Pause Filler Claiming the floor. 188 B1M W_misc , and the remedy lies partly in the hands of each one of us as well as in the hands of governments and large industries throughout the world. You'll False Positive Propositional 189 B31 W_misc noted regularly at other localities on the chalk and east of Hastings. There were well over 100 birds summering in 1976. The birds arrive on the cliffs in January False Positive Propositional 190 CDK W_misc out something else if it is not successful, build on the things that go well? # 5. # Improvements Do you look at different ways of doing things and False Positive Propositional 191 CE4 W_misc been well cleaned. Once you have covered the mirror with the glass you may well find that several pieces of fluff have been trapped between the two, and are False Positive Propositional 192 EEL W_misc Convocation has met five times, receiving regular reports on the progress of CUCGA as well as dealing with Convocation affairs. Convocation was again represented on GRADUATION DAYS 11 and False Positive Propositional 193 EER W_misc moved her and her mother to a separate pen. Other breeds of goats may well be different but my Goldens are very kind to each other and have a strong False Positive Propositional 194 F9J W_misc something' is more correct than' to recommend that someone do something'. Well, well. Hasn't Mr Brittlebrain ever heard of the subjunctive? I suggest DM Textual Pause Filler Restructuring speech. Followed by well 195 F9J W_misc is more correct than' to recommend that someone do something'. Well, well. Hasn't Mr Brittlebrain ever heard of the subjunctive? I suggest that he DM Textual Pause Filler Restructuring speech. Preceded by well 196 FSN W_misc about what others will understand when making off the cuff remarks to those you know well. Leave out obscure remarks, jargon, and private jokes, otherwise you'll False Positive Propositional 197 GW9 W_misc know why, so I asked him how much he knew about me.' Well, we know a lot,' he said.' But there's a DM Textual Frame Marker. Reported Speech 198 HC4 W_misc that it will now have considerably enhanced status and authority and will become even more well respected.' Simon N Davis FHCIMA Catering Audits #' A note to say False Positive Propositional 199 HC4 W_misc the new Hospitality. Interesting, good use of colour, easy to read. Well done!' Melvyn Green OBE FHCIMA Consultant to the hotel, food service and False Positive Propositional 200 K9D W_misc vests like the one shown here, from the Decca range in Belgium, as well as into running shirts, thermal tops, leggings and other garments aimed at professional False Positive Propositional Nº Genre Function Level Label Remarks 6 Appendix B: well in AmE !77 Nº Genr e Function Level Label Remarks 1 2017 SPOK NBC: T o d a y Show A B C Baby Driver with Ansel Elgort and Jon Hamm, that's doing really, really well so -- HODAKOTB# Okay. All right. KATHIE-LEE-GIFFORD# How are you doing? HODA-KOTB# False Positive Propositional 2 2017 SPOK NBC: T o d a y Show A B C I liked seeing how they make it. I thought it was cool. SHEINELLE-JONES# Well, it was really cool having you guys in. All the best. I DM Textual Pause Filler Restructuri ng speech 3 2017 SPOK NBC: T o d a y Show A B C fans at home. Guess what? You can win a cruise at home as well. Thanks to Windstar Cruises, we're giving away one more Mediterranean cruise to False Positive Propositional 4 2017 FIC Bk:Gra vePredi ctions A B C parade, That gives us cheer! A toast, my dear! # All well and good - but are we on the soup, or the dessert? Who False Positive Propositional 5 2017 FIC Bk:Ebb Tide A B C " We all have work to do, ain't so? " # " Well, Sallie's excused from redding up the kitchen this mornin', " Mamm DM Interpersonal Mitigator Saving someone else’s face 6 2017 FIC New England Review A B C sounds like she is giving him advice, good advice, advice he would do well to take. # Strike two, fouled off. Long pause, flaps the False Positive Propositional 7 2017 NEW S Baltimo re Sun A B C . # On Saturday, many singers waited hours, forming a line that stretched well over a block outside. They filed into a large area holding a few hundred False Positive Propositional 8 2016 SPOK PBS: PBS Newsho ur A B C sign it or you don't, that there's no compromise? KELLYANNE-CONWAY: Well, he also said today -- I was sitting right there, Judy -- he DM Interpersonal Mitigator Saving someone own’s face 9 2016 SPOK Fox: Sunday Mornin g Futures A B C BARTIROMO# Well, thank you so much. It is sunny in New York as well. Let me ask you about the implications of a leave vote. If the False Positive Propositional 10 2016 SPOK ABC: The View A B C So if you have a needy husband, you can only have one child. Well then some people need three kids. Well you also fight about chores, who DM Textual Pause Filler Restructuri ng speech. 11 2015 SPOK NPR: F r e s h Air A B C BENEDICT-CUMBERBAT#... But if I was a girl of a certain age. (LAUGHTER) BENEDICT-CUMBERBAT# Well, OK. All right - being really sort of hyper objective about it... DM Textual Pause Filler Restructuri ng speech. Well + comma 12 2015 SPOK CNN: Situatio n Room A B C response, saying, this is the first question, we need this question. Well, Hillary Clinton saying that we should not respond that way and also, of DM Textual Pause Filler Restructuri ng speech. Well + comma 13 2015 FIC Bk:Pinn acleEve nt A B C in tradecraft, talent spotting, and agent handling, everything he had done so well in his own country. His advice had helped them penetrate the U.S. Navy, False Positive Propositional 14 2015 FIC FilmHis tory A B C get there, and he rolled into bed in one of the Bowery lodging houses well content with his first day. He had seen eight Florede pictures in twelve hours False Positive Propositional 15 2015 FIC Souther nRev A B C ? " Naoki nodded. " It's a dam n shame. " " Well, " I said. " We tried. We did. And I'm DM Textual Frame Marker Reported Speech 16 2015 MAG Natural Hist A B C a traveling naturalist, once he got back home safely he would say, " Well, that's enough of that. " But next, at age thirty-one, DM Textual Frame Marker Reported Speech 17 2015 MAG People A B C a Showtime series, Bloom's musical dramedy isn't afraid to embrace the, well, crazy. " We've all been under the spell of love, " Unclear 18 2015 MAG SatEven Post A B C such as the insurance exchanges, and establish new ground rules for many activities as well as for the key players in the system -- insurers, hospitals, physicians, False Positive Propositional 19 2015 NEW S Austin A B C water in rural Belarus. " My mom had gotten some water out of the well, and she suspected that the water wasn't very good. So she would False Positive Propositional 20 2015 ACA D Emergi ngInfect ious A B C control measures for pandemic influenza are antiviral medications and vaccines (9), as well as nonpharmaceutical interventions, such as social distancing measures, school closures, and hygienic False Positive Propositional 21 2014 SPOK NBC: T o d a y Show A B C . NATALIE-MORALES# I have to say I was very impressed. And I-- it bodes well down the road for the Olympics. I mean, they're going to get False Positive Propositional 22 2014 SPOK PBS: PBS Newsho ur A B C her reference to the fact that there's a problem with the methodology... DR-GILBERT- WELCH# Well... JUDY-WOODRUFF#... and number two, the fact that there are many other studies DM Textual Pause Filler Claiming the Floor 23 2014 SPOK PBS: F r e s h Air A B C if anyone was schizophrenic, I'd say Keith was. But I got on well with Keith every time I saw him. And I was a good friend to False Positive Propositional 24 2014 FIC Bk:Nig htingale Before A B C 'd been trying to pull a fast one on her by impersonating a decorator. Well, I probably could if I wanted to. I couldn't tell a finial DM Textual Pause Filler Restructuri ng speech. Well + comma 25 2013 SPOK NBC: T o d a y Show A B C . GESINE-BULLOCK-PRA# We're going to put it on the sheet pan as-- as-- as well as you can, because it can run over a little bit so-- WILLIE-GEIST# But False Positive Propositional 26 2013 SPOK Fox: The Five A B C , it's hard to believe. PERINO# I found something hard to believe as well, believe it or not. GUTFELD# Unicorns. PERINO# Today, the chief of False Positive Propositional 1 !78 27 2013 SPOK NPR: Talk Of The Nation A B C gives us a sense of identity. Now what do we mean by that? Well, the University of California Santa Barbara psychologist Stan Klein has distinguished a number of DM Textual Pause Filler Restructuri ng speech. Well + comma 28 2013 FIC Bk:All GodsCh ildren A B C of her uncle's or not, this man was certainly not someone they knew well. # " We have a guest, Liesl, " Franz continued, nodding False Positive Propositional 29 2013 FIC Fantasy SciFi A B C " All right, then. You won't hear from me again. Be well, Alice. " He tilted his head to one side, as if studying False Positive Propositional 30 2013 FIC Fantasy SciFi A B C was young enough to resent being ordered around, yet didn't protest because -- well, because, speaking of eyes, Andy's were gray, something unknown in DM Textual Pause Filler Restructuri ng speech. Well + comma 31 2012 SPOK CNN_C ooper A B C about the election after all whatever some of these polls show enthusiasm faltering? BILLBURTON-SENIO: Well, I don't know if they are enthusiastic about the election. If you DM Interpersonal Mitigator Saving someone own’s face 32 2012 SPOK ABC_T hisWee k A B C TAVIS-SMILEY-1HOS# Those persons are Africans, though. We should point that out. GRETA-VAN-SUSTEREN# Well, yes, but the thing is that, also, I should say - DM Interpersonal Mitigator Politeness 33 2012 FIC Bk:Tec hnologi stsNove l A B C in anticipation of a tragic drowning. # Despite knowing the water and boats pretty well, Marcus had not grown up indulging in such impractical pursuits as rowing for pleasure False Positive Propositional 34 2012 FIC Bk:Won dersInvi sible A B C asked, # " What were you? " # Even he knew Durham that well. " An angel. " # " What's that? " # " False Positive Propositional 35 2012 NEW S Denver A B C search found cash and a bag of multicolored balloons like party favors inside the stick-shift well. The driver took the fall, and the police let Angel off. She False Positive Propositional 36 2012 ACA D JAdolA dultLite racy A B C were used for the content review of items. The list of statements, as well as the conceptual definitions for each of the four categories, was presented to 56 False Positive Propositional 37 2011 SPOK THE CHAR LIE ROSE SHOW A B C been has he been damaged, do you think, the prime minister? JOHN-BURNS: Well, I think in the end in that American political cliche, it's the DM Interpersonal Mitigator Saving someone own’s face 38 2011 SPOK PBS_N ewsHou r A B C positions... JUDY-WOODRUFF: They're saying it's a combination of attrition and layoffs. BERT-ELY: Well, and also I suspect that some of those employees will be -- go with DM Textual Pause Filler Restructuri ng speech 39 2011 SPOK ABC_G MA A B C (Off-camera) We talked about that earlier this week, and you wrote about it so well. Okay, I want I want to say something, you know, some False Positive Propositional 40 2011 SPOK CNN Newsro om A B C be some people who need to stay away from this whole gluten-free phenomenon? TEH: Well, yes. There is - if you don't have a problem tolerating gluten DM Interpersonal Mitigator Saving someone else’s face 41 2011 FIC Bk:Und erAttac k A B C over his lower lip. " Nice of you to ask. They're going well. We've been able to reach out to more key figures in the Underworld False Positive Propositional 42 2011 MAG GoodH ousekee ping A B C . 4. Return pasta mixture to saucepot and immediately add milk mixture. Stir well to evenly coat. Divide evenly among prepared dishes or pour into prepared dish. False Positive Propositional 43 2011 NEW S AssocPr ess A B C a modern politician, Kennedy's famous " ask not " call to service might well be derided as a socialist pitch for more government. His idyllic clamoring for a False Positive Propositional 44 2011 NEW S NYTim es A B C club. " They can be too scene-y for me, but they're very well done.' # Suffocating door policies are nothing novel, but Mr. Becker romanticizes False Positive Propositional 45 2010 SPOK NPR_S cience A B C American populations who were here and then bringing in subsequent African populations. FLATOW: Well, we had a huge number of Italians and Irish and Jews, all these DM Textual Pause Filler Claiming the Floor 46 2010 SPOK Fox_Su steren A B C malfeasance and the incompetence of the federal government in dealing with this crisis, as well. VAN-SUSTEREN: All right, let me play sort of devil's advocate with False Positive Propositional 47 2010 SPOK Fox_Su steren A B C until elections until they get their next message. KLEIN: That's right. Well, look, Democrats -- Democrats don't have much time left, so they DM Textual Pause Filler Restructuri ng speech. 48 2010 FIC Bk:Mill ionBles sings A B C , she added silently, not wanting to hurt her mother's feelings. " Well, what can I do to help? " Sandra pressed a kiss against her DM Textual Frame Marker Reported Speech 49 2010 FIC Bk:De monsPa rchment A B C . " I shall have to see your apartments. And to do that, well, it will be difficult. I must raise my fee and charge one shilling DM Textual Pause Filler Restructuri ng speech. 50 2010 ACA D Writer A B C was about human drama, and that is universal. I shot back, " Well, hell, we don't sell any papers in Afghanistan either, but every DM Textual Frame Marker Reported Speech 51 2009 SPOK NBC_T oday A B C be drinking, either. That -- a person might not be drinking, as well, and that would help one with one's memory. KOTB: What are False Positive Propositional 52 2009 SPOK NPR_S cience A B C completely safe. It's still kind of airless. Dr. GARVIN: Right. Well, we really have a multistrategy, Joe, and the first thing is to DM Textual Pause Filler Restructuri ng speech 53 2009 SPOK PBS_N ewshou r A B C details to others. What do you make of this right now? MARK-SHIELDS: Well, I think, unless were going to go through on stem cell research what DM Interpersonal Mitigator Saving someone own’s face 54 2009 SPOK CBS_E arly A B C JOHN-MCENROE: Im talking -- CHRIS-WRAGGE: Have you heard of him -- JOHNMCENROE: Well, how could -- how could -- how can you not say Federer. Im DM Textual Pause Filler Claiming the Floor Nº Genr e Function Level Label Remarks 2 !79 55 2009 SPOK NPR_T alkNati on A B C : Alcoholism as - I don't remember. (Soundbite-of-laugh CONAN: Well, I'm sure we could look it up. KATRINA: I'm really DM Textual Pause Filler Claiming the Floor 56 2009 SPOK CBS_E arly A B C so. As we head into tomorrow the first major snowstorm in the Cascades. Well keep an eye on it. So watch it. Tough traveling tomorrow there. DM Textual Pause Filler Restructuri ng speech 57 2009 FIC Analog A B C upscale the neighborhood, the more housing color varied. The houses got larger as well. The talking guidebook explained all of this, also mentioning the perks of the False Positive Propositional 58 2009 MAG Americ anSpect ator A B C small world or what? I gave a speech after the reception and it went well. I forgot to tell you I also posed for photos. That is standard False Positive Propositional 59 2009 MAG Entertai nment A B C actress, you must have wanted to be famous one day, right? Fox Well, yeah, because that's how you measure success in Hollywood, really. DM Textual Pause Filler Restructuri ng speech 60 2009 MAG FieldStr eam A B C looking for tracks and rubs to determine which fields are most productive generally, as well as which specific places within each field the bucks prefer. " Water is a False Positive Propositional 61 2009 NEW S Housto n A B C It was a big confidence factor for us. # " We didn't play well in the College Station tournament, so that really affected them. I told them False Positive Propositional 62 2008 SPOK Fox_O Reilly A B C site yet. O'REILLY: You know, it looks like a movie marquee INGRAHAM Well, Bill, here's... O'REILLY: It just looks like, you know DM Textual Frame Marker New Topic 63 2008 SPOK Fox_Hu me A B C difficult for Hillary Clinton to win in South Carolina and perhaps other Southern states as well. MORT-KONDRACKE: You remember there were all these polls early on, where there False Positive Propositional 64 2008 SPOK Fox_El ection A B C tragedy that we all shared. What do you say to them? MEEHAN: Well, I realize that, but as much as they are offended, Heather, DM Textual Pause Filler Restructuri ng speech 65 2008 SPOK PBS_N ewshou r A B C a process for approving monuments. Chief Justice Roberts, for example, said, Well, we have a Statue of Liberty. Do we have to have a Statue DM Textual Frame Marker Reported Speech 66 2008 FIC Bk:Da mien A B C get a fashionable form of the truth of the goings-on. " # " Very well, " Noah conceded. " But let me know if you change your mind False Positive Propositional 67 2008 MAG USATo day A B C the state-controlled Russian media is engaged in a propaganda war against the just-mentioned trio as well as the Baltic countries, Europe, and the U.S. # What do nonfree nations False Positive Propositional 68 2008 ACA D Anthrop olQ A B C educated women interviewed placed a high value on educating their own children, girls as well as boys, and many expressed the sentiment that girls could do as well as False Positive Propositional 69 2007 SPOK NPR_T ellMore A B C That had to have been devastating. How did you cope? Councilman EVANS: Well, it was - the whole process was obviously a very difficult process. My DM Textual Pause Filler Restructuri ng speech 70 2007 SPOK NBC_T oday A B C , right? Ms-BAUER: Yeah. And tattoos are a good idea, as well. My kids love when they get tattoos. Hard to wash off, but False Positive Propositional 71 2007 SPOK NBC_T oday A B C The luge was your favorite moment? VIEIRA: (Unintelligible). CURRY: Well, you know,... (unintelligible). LAUER: You've never -- DM Textual Pause Filler Restructuri ng speech 72 2007 FIC Bk:Brea kingFre e A B C He paused and looked them each in the eye. " I hope you... well, learn all you can. It's the first time this program has been DM Interpersonal Mitigator Politeness 73 2007 ACA D DrugIss ues A B C have been directed at the formulation and implementation of an integrated, coordinated, and well balanced national drug policy, and one that is evidence based. It is clear False Positive Propositional 74 2006 SPOK CNN_Z ahn A B C , so we feel good about this. ZAHN: Thank you. COHEN: Well let's take a look. First of all, you are not yet 50 DM Textual Pause Filler Claiming the Floor 75 2006 SPOK NPR_M orning A B C take strong action if the public supports what the administration is doing? Sen-LEAHY: Well I think that what we have done is actually start doing oversight. The Republican DM Interpersonal Mitigator Saving someone own’s face 76 2006 SPOK CBS_M orning A B C aged 79 to 102. Unidentified Man 3: Can you see the monument fairly well? Unidentified Man 4: Yeah. Man 3: Well, over to your False Positive Propositional 77 2006 SPOK CBS_M orning A B C see the monument fairly well? Unidentified Man 4: Yeah. Man 3: Well, over to your left is the Pacific. Man 4: OK. GEIST: DM Textual Pause Filler Claiming the Floor 78 2006 SPOK PBS_Ta vis A B C . Because over the last few days, as your name has become much more well known because of what you? ve accomplished over the last two days, leading False Positive Propositional 79 2006 SPOK MSNB C_Matt A B C got in to the inner sanctum to get briefed himself this morning? ALLEN: Well Tony was very astute about sticking to the facts that he knew, he has DM Textual Pause Filler Restructuri ng speech 80 2006 FIC Triquart erly A B C is open to the passengers? " # " The rest of the train? Well, I suppose it is. Did you, um, need something? A DM Interpersonal Mitigator Saving someone own’s face 81 2006 NEW S Atlanta A B C at the finish line with a competitor, one cousin pulling a knife. # Well, the summer of' 80 dragged on and attendance began to lag, and DM Textual Frame Marker New Topic 82 2006 ACA D SchoolP sych A B C early education settings, school psychologists may consult with teachers with limited formal training as well as specialists and allied health professionals who have in-depth knowledge of physical, occupational, False Positive Propositional 83 2005 SPOK CNN_L iveFro m A B C the background or have a son or daughter serving in the military? TAYLOR: Well, let me first say I have a tremendous amount of respect and admiration for DM Textual Pause Filler Restructuri ng speech 84 2005 SPOK NPR_S cience A B C , the aluminum industry on a global basis could agree to reduce their emissions, well, that's something that should be sort of acceptable under the treaty as well DM Textual Pause Filler Restructuri ng speech 85 2005 SPOK NPR_S cience A B C well, that's something that should be sort of acceptable under the treaty as well. If we could get the automobile industry to do the same thing, which False Positive Propositional 86 2005 MAG Atlantic A B C Wall Street Republicans). His stands against stem-cell research and gay marriage, as well as abortion, make him appealing to social conservatives. His state adjoins New Hampshire False Positive Propositional 87 2005 NEW S USATo day A B C he's offering a little less to a public that is rarely sated. The well isn't dry, but the pump needs more priming. In solitude lies solace False Positive Propositional 88 2005 NEW S Denver A B C comparing them to my predecessor's appointments. My predecessor appointed his political supporters as well. The person they touted as an example of a " good " appointment as False Positive Propositional 89 2005 ACA D PublicI nterest A B C , dressed, and out the door in time to get to the day-care providers well before the parents are due at their jobs. At 5:00 p.m. the parents leave False Positive Propositional Nº Genr e Function Level Label Remarks 3