On the Use of the General Extenders in British English Discourses

(1)

On the Use of the General Extenders in British English Discourses

Om användningen av general extenders i brittisk engelska diskurser

Muntasir Essa

Faculty: Karlstad University Subject: English linguistics Level: D

Credits: 15

Supervisor: Andrea Schalley Examiner:Marie Tåqvist Term and year: Fall 2017

(2)

ii

Title: On the Use of the General Extenders in British English Discourses Titel på svenska: Om användningen av general extenders i brittisk engelska diskurser

Author: Muntasir Essa

Pages: 41

Abstract

General extenders are examined in a collection of spoken data drawn from the two sub-corpora Business and Leisure of the British National Corpus. This study investigates the use of the 110 instances of or something, and everything, and and stuff in native speakers‘ discourses. The speakers are males and females aged 15-59 years. This study focuses on the frequency and functions of general extenders, and explores whether there is a correlation between the use of GEs and speakers‘ ages as well as the gender of speakers. For the purposes of this study, both quantitative and qualitative methods are used. The analysis shows that the most frequent form is or something (representing 65.45% of the data), followed by and everything (20.90%) and and stuff (13.63%). This study reports that general extenders are multifunctional, and that they indicate ideational and interpersonal functions. However, there is variation in the use of general extenders. These forms are mainly used to implicate a category. The predominant use of these forms occurs in young speakers‘ discourses, and their frequency decreases as the speakers get older. In addition, this study points to the differences between female and male speakers in using general extenders, and that females use these expressions more frequently than males.

Keywords: general extenders; British English; ideational functions; interpersonal functions; speakers‘

ages and gender

Sammanfattning på svenska

General extenders undersöks i språkbrukares tal som härrör från två sub-corpora Business och Leisure från British National Corpus. Syftet med denna studie är att undersöka 110 instanser av or something, and everything, och and stuff hos brittiska mäns och kvinnors tal i åldersspannet 15-59 år. Denna studie fokuserar på general extenders frekvenser samt funktioner, och undersöker huruvida om deras användningen är korrelerade med språkbrukarnas ålder och kön. I denna studie används både kvantitativa och kvalitativa analysmetoder. Resultaten visar att den vanligaste formen är or something som motsvarar

(3)

iii

65.45% av datan, följt av and everything 20.90%, medan den lägsta användningen gäller and stuff 13.63%.

Denna studie visar att general extenders är multifunktionella det vill säga de indikerar ideella och interpersonella funktioner. Språkbrukare använder dock general extenders huvudsakligen för att implicera en kategori. Den övervägande användningen av general extenders sker i unga talares diskurser, deras frekvens minskas i äldre språkbrukares tal, samt kvinnor använder dem mer än män gör.

Nyckelord: general extenders, brittisk engelska, ideella och interpersonella funktioner, språkbrukares ålder och kön

(4)

iv Acknowledgement

I would like to thank my tutor Professor Andrea Schalley, Deputy Head, Department of Language, Literature and Intercultural Studies Faculty of Arts and Social Sciences, Karlstad University, for her guidance and support throughout this project.

I am indebted to the teaching staff at the English Institution, Uppsala University, for increasing my academic knowledge throughout the Master Program.

I wish to express my gratitude to Niklas Sundell, the Principle of Oxelskolan at Uppsala City, for providing his feedback for translating the abstract of this project into the Swedish language. However, any mistakes that remain are my own.

(5)

v

Contents

1. Introductions and aims………..………1

2. Background………5

2.1 General extenders: terminology, definition, contextualization……….5

2.2 The functions of general extenders………7

2.3 Speakers‘ ages and the use of general extenders……….……12

2.4 The use of general extenders and the gender of speakers………13

3. Methods………...…………13

3.1 Data………..13

3.2 Data analysis………18

4. Results………..19

4.1 The overall frequency of general extenders………..………...……20

4.2 The functions of general extenders………..………..……….………….21

4.2.1 General extenders and the ‗list construction‘………..………..22

4.2.2 Category implication……….………23

4.2.3 Approximation……….……….25

4.2.4 Intersubjectivity……….27

4.3 The use of general extenders and speakers‘ ages……...…………...……..………28

4.4 The use of general extenders and the gender of speakers……...……….29

5. Conclusion………...………31

References……….33

Appendix 1………35

Appendix 2………36

(6)

1

1. Introduction and aims

Many recent studies have focused on the use of general extenders (GE) by speakers of British English.

GEs are not only studied in a number of English varieties, for example, American English (Overstreet 1999), British English (Cheshire 2007), Australian English (Dines 1980), and Canadian English (Tagliamonte & Denis 2010), but some scholars also examine their use in English language learners' discourses, such as Swedish learners (Aijmer 2015), German learners (Terraschke 2007), and French learners (Dubois 1992; Secova 2012). GEs are expressions in the form of a noun phrase preceded by a conjunction, occurring in clause-final position. They are nonspecific in reference, and extend otherwise grammatically complete utterances (Overstreet 1999: 3). In a more recent study, Overstreet (2014) shows how GEs fill in a post-position, immediately after phrases, clauses and utterances, providing speakers with the opportunity to add modification and to influence the interpretation of what is being said (Overstreet 2014: 108). GEs can be divided into two types: those that begin with and, the adjunctive general extenders (AGEs), such as and stuff; and those beginning with or, called the disjunctive general extenders (DGEs) such as or something. GEs appear in spoken and written language, and they are most frequent and varied in informal spoken language (Overstreet & Yule 1997: 252; Overstreet 1999: 6; Biber, Johansson, Leech, Conrad, Finegan 1999: 116). The present study draws on data from the British National Corpus (BNC) and investigates the use of GEs or something, and everything, and and stuff in British English conversations.

In addition to identifying the forms of GEs and the positions in which GEs occur across registers, previous studies focused on the functions of GEs in speakers‘ discourses, and have shown that GEs are multifunctional expressions. However, up to this moment, there is no agreement on the core function of GEs. For example, some researchers suggest that GEs have ideational functions and that their core function is category implication (Dines 1980; Aijmer 2002). Consider example (1), which is drawn from the British National Corpus (BNC).

(1) Derek: that he's going to get some of then erm that might more of an incentive for him to

Wendy: Well he might for instance let us come over and and play with CorelDraw for an hour in one evening or something like that.¹ (BNC FM2 2137)²

1 Throughout the examples, I add italics in order to emphasize what part of phrases are extended by GEs.

2 BNC refers to the British National Corpus. FM2 is a text identifier code, which refers to the source text from which the current solution is taken. In the BNC, each sentence is consecutively numbered and 2137 refers to the sentence number from which the current solution comes (see Aston & Burnard 1997: 52).

(7)

2

Example (1) shows how Derek and Wendy talk about creating a company logo. In her utterance an hour in one evening or something like that, Wendy believes that the company logo needs one hour to be designed, but she is unsure about when this hour will be (on an evening or at another time of the day). In this example, the use of or something like that refers to a category of a time of the day when the logo creation might happen.

Additionally, Jefferson (1990) and Lerner (1994) hypothesize that the primary function of GEs is to complete three-part lists. Consider examples (2) and (3), in which GEs occur as a part of a three-part coordination.

(2) I mean I can't get my hands covered with flour and egg and stuff if he's going to cry and I've got to pick him up, can I? (BNC EBR 984)

(3) Wow! I will! I‘ll have it cut. I'll wander round this place and find a cheapo hairdresser and I'll get it done — Gothic, or spiky, or something. (BNC BMS 300)

In contrast, other studies reveal that GEs indicate interpersonal functions, for example, marking an assumption of shared knowledge between speakers, or indicating politeness (see table 1) (Overstreet &

Yule 1997; Overstreet 1999; Winter & Norrby 2000). Moreover, other studies have gone so far to claim that GEs have textual functions³ when, for example, speakers use GEs to signal the end of a turn (Dubois 1992: 180; Winter & Norrby 2000: 6; Cheshire 2007: 181), which leaves the floor to another speaker to begin his/her turn. Consider example (4), which is an extract of an interview with a lock maker, who talks about the advantage of science and new technique adopted in lock industry.

(4) Dick: I say the competition then, later on became financial. Them as who could make it a bit cheaper see, and all that. But <unclear> you said erm certain things could be <pause> we had, had a machine once and you'd got to pull four or five handles. (BNC G4N 314)

In example (4), and all that appears to function as a punctuation that marks the end of the speaker‘s turn. The speaker, then, moves on to talk about the lock manufacturing process.

GEs also enable speakers to construct their identities and maintain their social relationships with the other speakers on the basis of shared knowledge: addressees will fill in the missing information (Aijmer 2015: 213; Overstreet 1999: 73; Terrachke 2007: 143). Youssef (1993) examines the use of an ting ‗and thing‘ in medical settings (used by medical practitioners and patients) and shows that its use

3In the data investigated, the textual function exists, however, this study focuses on the ideational and interpersonal functions of GEs.

(8)

3

indicates solidarity and marks an information plane between speakers in Trinidad (Youssef 1993:

293).

Previous studies show that GEs serve a number of functions in discourses; however, there is no agreement on the primary function of GEs. The difference between, for example, Dines (1980) and Overstreet (1999) in assigning the main function of GEs highlights that research into the functions of GEs is rather inconclusive to date. In the present study, I will therefore investigate the functions of GEs, and in particular test Dines‘ (1980) hypothesis that the main use of GEs is to implicate a category. In this study, I hypothesize that the primary function of GEs is not restricted to indicate ideational functions such as category implication, because GEs can be used to indicate interpersonal functions such as approximation.

Another question in relation to GEs, in addition to what functions they serve, is who uses them mostly.

Many scholars have examined the frequency of GEs in young, middle, and old speakers‘ data and report that young speakers use GEs more frequently than older speakers do, suggesting that GEs are a feature of youth speech (Dubois 1992; Stubbe & Holmes 1995; Tagliamonte & Denis 2010).

Nevertheless, it is not clear whether the high frequencies observed in these previous studies can provide an accurate picture, in particular that the use of GEs is highly associated with young speakers.

The main difficulty may lie in the data from which such high frequencies are derived. GEs are collected and studied in, for example, interviews (Cheshire 2007: 160; Pichler & Levey 2011: 447;

Levey 2012: 266), face-to-face interactions and telephone conversation, news radio interviews, political debates, and academic discussions (Overstreet & Yule 1997: 252; Overstreet 1999: 6). The variation in the frequency of GEs may be a consequence of several factors, such as the level of formality of discourse, that some GE forms are frequently used in informal environments and talk among familiars, but rare in formal language (Overstreet 1997: 252; 1999: 8). In addition, age differences between speakers, what topics are being discussed between speakers, and what the social distance is between the speakers are correlated with the frequency of GEs (Cheshire 2007: 161).

Selecting discourses from a number of native speakers, from different age groups, from different topics, and from different contexts will contribute to our understanding of how speakers use GEs to express their thoughts in different situations. In British English, little attention has been paid to addressing speakers‘ ages and the use of GEs. For example, Cheshire (2007) examines the use of GEs in adolescents' discourses from three British towns, and Levey‘s (2012) study has only been concerned with how preadolescents aged 7-11 use GEs. This study explores the relationship between speakers' ages and the use of GEs in the following, by examining and comparing the frequency and functions of GEs, showing how elder speakers use GEs, compared to younger speakers‘ use of GEs. Based on the previous studies, I assume that young speakers use GEs differently and more frequently than older speakers.

(9)

4

To understand the role of the gender of speakers and their use of GEs, several studies have been carried out. However, previous research examining gender differences in the use of GEs report different results. For example, Norrby and Winter (2002) and Cheshire (2007) point out that there are no gender differences that influence the use of GEs (Norrby & Winter 2002: 4; Cheshire 2007: 163), while Levey (2011) provides a clear contrast in the frequency of GEs between males and females, and that females, especially aged 8-11, use GEs more often than males, of the same age group, do (Levey 2012: 270). Variations in the frequency of GEs according to the gender of speakers are also reported from other English varieties. For example, for New Zealand English, Stubbe and Holmes (1995) reveal that the frequency of GEs varies depending on the gender of speakers. They show that the most frequent use of GEs occurs in young middle-class females and young working-class males (Stubbe &

Holmes 1995: 77). In Canadian English, Tagliamonte and Denis (2010) reveal how speakers‘ gender has less effect than other variables such as speakers‘ ages. Instead, they show that the gender of speakers highlights, some certain, forms of GEs, and stuff is favored by males, while and thing is preferred by females (Tagliamonte & Denis 2010: 359). Thus, a third and final factor to be investigated in this study is the gender of speakers in order to uncover the extent to which the use of GEs has been affected. Based on the previous research that shows gender differences in the use of GEs, I assume that the use of GEs is influenced by the gender of speakers, and that females use GEs more often than males do.

This study thus examines the use of GEs in native speakers‘ discourses and attempts to answer these questions:

1. What are the functions of GEs in British speakers‘ discourses? What is the main function of GEs and stuff, and everything, and or something in native speakers‘ conversations?

2. How do older speakers and young speakers use GEs? What are the similarities and differences of frequency and functions of GEs between speakers‘ age groups?

3. How do male and female speakers use GEs, in terms of frequency and functions of GEs?

The rest of this paper is organized as follows. In section 2, I present the main theoretical issues related to GEs. First, I provide the terminology and definitions of GEs, and contextualization of them as used in previous studies. Next, the functions of GEs as uncovered in previous research will be demonstrated. Additionally, I address the relationship between speakers‘ ages and the use of GEs.

Further, the role of the gender of speakers and the use of GEs is also considered. Section 3 describes the corpus and methods used in data analysis. Because this study focuses on British English spoken language, the British National Corpus (BNC) is examined, and quantitative and qualitative approaches are applied in order to analyze the occurrences of GEs in speakers‘ discourses. Based on the findings of the quantitative and qualitative analyses, section 4 presents the most frequent forms of GEs, and

(10)

5

what ideational and interpersonal functions GEs perform, as well as their frequencies in British speakers‘ conversations. In addition, the similarities and differences in the use of GEs as used by young, middle-aged, and older speakers will be shown, and provide how older speakers use GEs differently from young speakers. This section also shows how GEs are used by males and females speakers, including the frequency and functions of GEs. Finally, section 5 summarizes the findings of this study and provides a brief conclusion regarding the use of GEs, followed by a discussion of the limitations of this study, and suggests directions for future research.

2. Background

As this study investigates the use of GEs, it is crucial to present an overview of previous studies related to GEs. In section 2.1, I introduce and discuss some terms that refer to GEs, how GEs are defined, and in which contexts GEs are analyzed. In section 2.2, I explore some core functions of GEs as found in previous studies. In section 2.3, I explain the relation between speakers‘ age and GEs.

2.1 General extenders: terminology, definition, contextualization

Recent studies describe GEs as being formed with a conjunction and/or plus an existential or universal quantifier and an optional form (like that) such as and all that, and everything, or something (Dines 1980: 18; Aijmer 2002: 213). However, the literature outlines terminology problems in relation to GEs, indicating that several terms are used for these expressions in the English language. GEs are differently labeled due to the fact that they have several functions in discourses. Although most of the recent studies use the term general extenders (Overstreet 1999), several studies still use other terms.

For example, the term set-marking tags was first coined by Dines (1980), who studied their use in middle-aged and working-class women‘s speech in Australian English. Dines (1980) defines GEs as expressions, which ―cue the addressee to interpret the preceding element as an illustrative example of some more general case" (Dines 1980: 22). Evidence is presented in example (5), which is an extract of a meeting between city council members.

(5) Okay, you were talking about, you you're talking about your target customers were sort of like abattoirs and hairdressers <-|-> and stuff like that?

Unknown speaker: <-|-> Yeah but you know the <-|-> mailers that we did, I mean <-|-> (BNC KGU 2008)

In example (5), the speaker uses and stuff like that with abattoirs and hairdressers to instruct his addressees to infer the member of a category that is related to craftsmen. Likewise, Aijmer (2002) examines the use of GEs in native speakers‘ spoken language. Aijmer (2002) draws on data from the London-Lund Corpus (LLC) and calls these expressions referent-final tags. She finds that these tags

(11)

6

occur after a referring expression in the preceding text, and their primary function is to instruct the hearer to interpret the category exemplified by the referring expression as subordinate to a more inclusive or general case (Aijmer 2002: 212). Consider example (6) in which Wendy and Clare talk about an amount of money that must be paid to careers services.

(6) Wendy: and they said well in that case could we make some money out of it? So I just gave them the statutory amount per tape

Clare: <-|-> Right.

Wendy: <-|-> erm <-|-> and they put it in their Christmas Fund or something. (BNC FM2 976)

In example (6), Wendy uses or something with Christmas Fund to indicate a category of a short-term savings account in which people deposit their money in order to be withdrawn in the near future. In addition, in her study of vague categories produced in Irish English, by drawing on discourses from an Irish talk radio show, O‘Keeffe (2004) includes all expressions used to identify these categories, and refers to GEs as vague category markers (VCM). She treats GEs as recognizable chunks of language indicating a shared knowledge between speakers (O‘Keeffe 2004: 7). However, this is not entirely accurate, because GEs are also found to be multifunctional within discourses (Overstreet 1999).

Although Dines (1980), Aijmer (2002), and O‘Keeffe (2004) maintain that the primary function of GEs is a category implication, their definition fails to consider potential interpersonal functions GEs might indicate. Overstreet (1999) only finds six cases (of a total of 156 instances) in which GEs are used with lexicalized categories, due to the fact that GEs are used with, for example, actions or groups of entities where no established referring expression for the actions and groups is available (Overstreet 1999: 43). Other studies conducted on the use of GEs reveal that there is a clear decline over time in the frequency of category implication function. In their study of GEs on Australian English and Swedish, Winter and Norrby (2000) report that defining the function of GEs as set is problematic due to the reference, which GEs refer to, is ambiguous in adolescents‘ discourses (Winter & Norrby 2000:

4). GEs may refer to one or several elements or to the whole phrase(s) (Ibid). Evidence that GEs lose their categorizing function in English varieties such as Canadian English is also presented (Denis 2017: 178). In British English, Cheshire (2007) investigates the use of GEs in adolescents‘ data drawn from three English towns, namely Hull, Reading, and Milton Keynes, and reports that category implication is infrequent (Cheshire 2007: 178; Levey 2012: 275).

In order not to preempt a discussion of the functions of GEs, I will refrain from using function-based notions in the following to refer to these expressions under discussion (such as category implication, referent final tags, vague category markers) and instead use the form-based notion of general extenders (GEs) in the present study. For these, I will follow Overstreet‘s (1999) definition that GEs are expressions in the form of a noun phrase preceded by a conjunction, occurring in clause-final position,

(12)

7

and that they are nonspecific in reference, and extend otherwise grammatically complete utterances (Overstreet 1999: 3).

2.2 The functions of general extenders

In the literature, a number of functions of GEs are discussed (Dines 1980; Jefferson 1990; Overstreet 1999). They are ideational and interpersonal functions. Table 1 displays the functions of GEs found in previous studies. The ideational functions are category implication and three-partedness, while the interpersonal functions include approximation, speculation, quantity hedge, intersubjectivity, positive politeness, negative politeness, and intensifier.

Table 1. The primary functions of GEs

Type of function Example Explanation

Ideational: Category implication (Dines 1980)

I have moved pianos upstairs, in certain cases where it's a very open staircase and a straight one, but we won't normally do that cos er we need the equipment to do it, the proper sled and everything. (BNC KNC 1025)

The speaker uses and everything to refer to a category of equipment items needed to move, for example, house furniture.

Ideational: Three-part list (Jefferson 1990)

Bloody ages doing things like if, if mode so and so to work out whether it's a Monday or a Tuesday or something. (BNC H61 354)

The speaker uses or something, in a list completion, to refer to a list of the days of a week.

Interpersonal: Approximation (Overstreet 1999: 115)

Let's say I've known the guy for

<-|-> fifteen years or something like that <-|->. (BNC JA4 667)

The speaker uses or something as a hedge on the truth of his statement while estimating the period that he has known the guy.

Interpersonal: Speculation (Overstreet 1999: 114)

Maya: He—he always looks like he has really bad Bedhead.

Now that he's got more hair or less hair or something.

(Overstreet 1999: 114)

Maya does not know the reasons of her friend‘s bedhead.

She guesses that he's got more hair or less hair.

(13)

8 Interpersonal: Quantity hedge

If she's gonna go through labor and delivery and she's gonna have a baby and all this stuff you know I'd like to be able to be there. (Overstreet 1999: 127)

The speakers uses and all this stuff to indicate that there is more information around birth, but will not mention them.

Interpersonal: Intersubjectivity (Overstreet 1999: 68)

I really think as soon as she moves to Connecticut I won't know her anymore. I'm sure she's not gonna call me or you know write me or anything like that. (Overstreet 1999: 69)

Or anything like that appears to refer to a general knowledge related to ways of communications between friends.

Interpersonal: Positive politeness (Overstreet 1999: 98)

Y'know uh if I pay attention I can do pretty good on flat surfaces but as soon as the .hhh surface gets uneven an' stuff um. I—it—Yeah, it gets a little tricky. (Overstreet 1999: 101)

The speaker has recently had a brain surgery, and his use of and stuff appears to appeal for acknowledging his inability to hike on uneven trails, which can be tricky (Overstreet 1999:

101).

Interpersonal: Negative politeness (Overstreet 1999:

104)

Sara: I'd like to move out there 'n take over your apartment Roger: Right

Sara: an' all the stuff that's in it an' just buy it from you, or whatever, an‘ maybe like buy your truk or something. I dunno.

Sara‘s proposals may be interpreted as face-threatening.

However, her proposals are marked by the use of or whatever and or something, indicating that there may be other alternatives instead of buying all the stuff and the truck.

Interpersonal: Intensifier (Overstreet 1999: 93)

Donna: Do I have the cutest cat in the world or what?

The speaker uses or what to stress the idea that this cat is so cute.

In her study of lists in natural conversation, Jefferson (1990) shows that the lists are constructed by three items, as in example (7) (Jefferson 1990: 65). She proposes that GEs represent a third item in a category: they occur together with two other items. A different third item is either unavailable or will not adequately exhaust the array, which indicates that there are many more relevant nameables, as in

(14)

9

example (8) (Jefferson 1990:66). Like Jefferson (1990), Lerner (1994) also shows that a list construction, which is often formed with three items, creates a systematic resource for speakers by turn-taking and minimization; and that speakers use GEs to produce the third item (Lerner 1994: 22).

(7) Roger: He was given three bottles a champagne, three horses, and three addresses of uh patronizing women y‘know. (Jefferson 1990: 66)

(8) Ernie: I said no I know his name is something else. Teddy or Tom or something. (Jefferson 1990: 66)

Overstreet (1999) provides an interesting account of how GEs are employed by speakers in discourses, by discussing their ideational and interpersonal functions. She examines GEs, including their forms and functions, by analyzing their use in the informal settings of telephone conversations and face-to- face interactions. Overstreet (1999) shows that, with regard to previous studies, category implication and complete three-part lists are good examples of the ideational functions of GEs due to the language is used to encode the cultural experience, and the speakers encode their own experiences as members of the culture (Halliday 1994: 27). She shows how speakers use GEs with exemplars to indicate category implication, while the addressees can infer additional or alternate members of a category. In regards to their interpersonal functions, she explains how GEs are used to mark an attitude toward the message expressed (Halliday 1994: 27), and speakers use GEs to mark, for example, an assumption of shared knowledge, function as positive and negative politeness strategies, as intensifiers, and as hedges on the Gricean (1975) maxims of quality and quantity (Overstreet 1999). The interpersonal functions of GEs will be discussed in detail.

When GEs are collocated with, for example, you know, and you see, GEs are used to mark an assumption of shared knowledge between participants (Aijmer 2002: 228), and speakers tend to use GEs to communicate the following message to the addressee ―because we share the same knowledge, experience, and conceptual schemes, I do not need to be explicit; you will be able to supply whatever unstated understandings are required to make sense of the utterance‖ (Overstreet 1999: 68; also Winter

& Norrby 2000: 4). Support for this perspective is provided in example (9) in which two workers, Williams and Fremantle, discuss some information at their work place.

(9) S Williams: Steve works in the engineering office and has taken over some of the er you know purchasing function as well, like enquiries and stuff.

T Fremantle: Yeah. (BNC JP2 8)

In example (9), Williams informs Fremantle about the recent functions that Steve ‗co-worker‘ is assigned to: purchasing function as well, like enquiries and stuff. Williams does not mention what

(15)

10

exactly Steve does. Instead, Williams relies on Fremantle‘s ability to guess. That is, Williams assumes, by using a combination of you know and and stuff, that Fremantle already knows about what these job tasks roughly are. When and stuff signals the end of Williams‘ utterances, Fremantle confirms that he has a similar knowledge with yeah (Overstreet 1999: 76; Cheshire 2007: 179). The use of and stuff also indicates solidarity between speakers, especially when it co-occurs with you know, thereby underscoring an assumed closeness as the speakers William and Fremantle belong to the same group, the group of employees of their company, and have certain things in common (Overstreet & Yule 1997: 254; Overstreet 1999: 104; 2005: 1852; 2014: 121).

Disjunctive GEs can function as a hedge in relation to the truth or accuracy of a statement, indicating that the content of the message is inaccurate (Overstreet 1999: 112; Aijmer 2015: 228). In example (10), a male speaker talks about his previous job on a farm. There, he used to go with the keeper to breed the hens.

(10) And then the old keeper used to come round and see that we were all right and they'd be there till the following day and carry on for about three or four weeks you see. And then as they hatched you see, you used to take them in-- into the incubator then. Because we had finished

with the hens then you see, and they used to sell the hens for about two shillings or something like that you know. (BNC HER 12)

In example (10), because the speaker does not remember exactly what the price of hens was, he uses or something with two shillings. In this example, or something functions as a hedge on the truth of the speaker‘s statement. Using or something suggests alternative possibilities that the price of each hen could be more or less than two shillings. Therefore, the speaker avoids risking a violation of the maxim of quality.

AGEs can function as a hedge of quantity to indicate that ‗there is more‘ than might be expected (Overstreet 1999: 127). Consider example (11), which shows how Mike talks about the introduction of technology in relation to individuals‘ skill levels.

(11) Mike: It produced a small number of <-|-> highly skilled Unknown speaker: <-|-> <unclear> <-|->

Mike: people didn't it? The tool makers and the people with their machinery and all the rest of it. (BNC H47 374)

In example (11), the speaker uses and all the rest of it to signal that there are many more advantages of technology than the production of the tool makers and the people with their machinery.

(16)

11

In her study of GEs in American English and German, Overstreet (2005) notices that many differences are found regarding, for example, the position, frequency, and functions of GEs. GEs typically occur in a clause-final position in English, while German GEs can occur in a clause-internal position, immediately preceding past participles; and GEs are used more frequently in English than German (Overstreet 2005: 1849).

Overstreet‘s (2005) analysis also reveals that GEs have, in both languages, similar functions, such as making an assumption of shared knowledge, solidarity, evaluation, accuracy, politeness. However, one of the functions of GEs in English that does not exist in German is that GEs function as intensifiers to solicit agreement with an evaluation being made in the English language, as in example (12) (Overstreet 2005: 1858).

(12) Cathy: Is that the best picture or what?

Jean: Shh! It‘s absolutely priceless! (Overstreet 2005: 1858)

In example (12), not only Cathy uses or what to express how this picture is magnificent, but also to solicit agreement from Jean about her evaluation. Jean responds to the invited agreement with Cathy‘s assessment Shh! It’s absolutely priceless.

Another important study in this field is Romero-Trillo (2015) who presents a prosodic analysis of GEs in the London-Lund Corpus. Romero-Trillo (2015) refers to AGEs as endocentric general extenders (ENGEs) and DGEs as exocentric general extenders (EXGEs). He examines the occurrence of GEs according to their intonation contour. ENGEs are either stressed in medial or final position, while EXGEs appear to be stressed in final position (Romero-Trillo 2015: 57, 60). He reveals that ENGEs indicate shared knowledge between speakers, whereas EXGEs orient the addressee to infer alternative options (Romero-Trillo 2015: 61).

Aijmer (2015) examines the use of GEs in Swedish learners‘ spoken language compared to their use by native speakers, by drawing on data from the Louvain International Database of Spoken English Interlanguage Corpus (LINDSEI-SW) and the Louvain Corpus of Native English Conversation (LOCNEC). Aijmer (2015) finds that Swedish learners use GEs, as native speakers do, to indicate, for example, intersubjectivity, and when GEs function as hedges, they are collocated with expressions that indicate uncertainty such as maybe, I don’t know, I think (Aijmer 2015: 226-229).

The literature reveals that GEs indicate ideational and interpersonal functions. However, it remains unclear whether GEs are primarily used to indicate, for example, a category implication, or

(17)

12

intersubjectivity. It would be thus of interest to investigate the functions of GEs in British English discourses.

2.3 Speakers’ ages and the use of general extenders

Recently, the relationship between speakers‘ ages and the use of GEs has been extensively studied in a number of speech communities. Earlier research, focusing on the use of GEs in light of speakers‘ ages, indicates that GEs are socially conditioned, and speakers‘ ages are a salient feature that influences their use in discourses. For example, Tagliamonte and Denis (2010) examine the frequency of GEs in the Toronto English Archive that consists of 1.2 million words representing spoken language of contemporary Canadian English, and based on comparable data taken from previous studies, show that, overall, GEs occur more frequently in adolescents‘ data than in older speakers‘ discourses (Tagliamonte & Denis 2010: 349). The overall frequencies of GEs provided in previous studies lead Tagliamonte and Denis (2010) to the consistent observation that GEs are a characteristic of youth speech (see also Winter & Norrby 2000: 4, 8). Denis (2017) finds that GEs co-occurring with other discourse markers are frequently used by young speakers, and by older speakers in Canadian English (Denis 2017: 176). In their study of pragmatic expressions, Stubbe and Holmes (1995) investigate the use of GEs in New Zealand English and find that middle-class young speakers use GEs twice as often as middle-class middle-aged speakers do (Stubbe & Holmes 1995: 72). This is similar to what Dubois (1992) finds in data drawn from two Montréal French corpora. Dubois (1992) reports a robust age- variation in the frequency of GEs, showing that speakers use fewer GEs as they grow older (Dubois 1992: 185). Likewise, Secova (2012) reveals the existence of age differences in the frequency of GEs, and young speakers use GEs more frequently than older speakers do in European French, and adolescents tend to use GEs to function as, for example, hedges, intensifiers, and mark a shared knowledge between speakers (Secova 2012: 24). In the northeast of England and across three age cohorts young (17-23), middle-aged (27-48), and older speakers (60-81), Pichler and Levey (2011) investigate the use of GEs in synchronic data and notice an age-graded effect that GEs occur less frequently as the speakers get older (Pichler & Levey 2011: 454). In addition, their data show that GEs are multifunctional, and that speakers frequently use GEs to signal interpersonal and textual functions, whereas category implication functions are in decrease (Pichler & Levey 2011: 458). Similarly, Cheshire (2007) reveals that adolescents aged 14-15 years old mainly use GEs to indicate interpersonal functions such as intersubjectivity and politeness in their discourses (Cheshire 2007:

179-181). A second aim of this paper is to investigate the relationship between speakers‘ ages and the use of GEs in order to uncover whether younger speakers use GEs more frequently and differently from those older speakers, by examining and comparing the frequency and functions of GEs in speakers‘ discourses, that is, how older speakers use GEs compared to the use of GEs by younger generations.

(18)

13

2.4 The use of general extenders and gender of speakers

The relationship between the gender of speakers and the use of GEs has been examined from two perspectives. The first perspective addresses the frequency of GEs, and the second is associated with the functions of GEs in order to explore the similarities and differences in the use of GEs by male and female speakers (Norrby & Winter 2002; Chechire 2007; Tagliamonte & Denis 2010; Levey 2012).

However, earlier research focusing on the role of the gender of speakers and the use of GEs indicates that either very small gender differences emerge (Norrby & Winter 2002 on their study of GEs in Australian English and Swedish; Tagliamonte & Denis 2010), or gender differences do not exist at all (Cheshire 2007). In contrast, more recent research reports different results with regard to gender differences. For example, Levey (2012) finds that females use GEs more frequently than males do.

Murphy (2010) also investigates the use of GEs in the Female Adult Corpus (FAC) and the Male Adult Corpus (MAC) and reports that males use GEs more frequently than females do (Murphy 2010:

107). For FAC, she posits that speakers aged 70-80 use GEs twice as often as speakers aged 20 do, and three times as often as speakers aged 40 (Murphy 2010: 93). For MAC, Murphy (2010) reports that speakers aged 20 use GEs more often than speakers aged 40 and speakers aged 70-80 do (Murphy 2010: 106). Pichler and Levey (2011) provide a robust gender variation in the use of GEs. They show that males in young, middle-aged, and older speakers groups use GEs more often than females do (Pichler & Levey 2011: 454). Nevertheless, there is still inconclusive evidence as to whether female speakers use GEs differently from male speakers or not. In this study, I assume that male speakers use GEs differently from female speakers in regard to the frequency and functions of GEs.

3. Methods

The present study originates from a need to investigate the functions of GEs in British English conversations, and to explain how GEs are employed by different age cohorts and speakers of different gender. This section introduces the corpus, and describes the procedures used in the data collection and analysis. In section 3.1, I will first detail information regarding the British National Corpus (BNC) that this study is based on, and outline the process by which the data were collected. Section 3.2 describes the methods that were used in data analysis.

3.1 Data

This study investigates the three most frequent forms of GEs found in previous studies (Biber, Johansson, Leech, Conrad, Finegan 1999: 116; Overstreet 1999: 7). These GEs are or something (like that), and everything (like that), and stuff (like that), in order to detect their use in native speakers‘

discourses. Previous research investigates the use of GEs in, for example, interviews (Cheshire 2007:

160; Pichler & Levey 2011: 447; Levey 2012: 266), face-to-face interactions and telephone conversation (Overstreet 1999: 6). This study draws data from the BNC. The BNC represents the most

(19)

14

significant electronic corpus of British English and was designed to characterize British English in its various social and generic uses. The BNC consists of 100 million words, of which written texts comprise 90% and transcribed spoken materials contain 10% (Aston & Burnard 1997: 28-29). Due to the fact that GEs are characteristic of spoken language, this study examines the use of GEs in two spoken sub-corpora: Business and Leisure. Business and Leisure have been selected as they tend to consist of informal discourses, in which GEs frequently appear. Regarding their type of discourses, Business and Leisure include conversations drawn from, for example, company and trade union talks or interviews, business meetings, sales demonstrations, sports commentaries, broadcast chat shows, phone instructions as well as club meetings (Aston & Burnard 1997: 32, 33), conversation that overall represents a good source for studying the use of GEs. Together, Business and Leisure comprise 735,195 words, as displayed in Table 2.

Table 2. The frequency of words in two sub-corpora of the BNC

Word counts

Business 519,622

Leisure 215,573

Total 735,195

Table 2 shows that Business (519,622) is larger than Leisure in terms of absolute word counts (215,573). The difference in terms of absolute word counts may influence the frequency of GEs, that is, GEs may on absolute counts appear more frequently in Business than in Leisure. Murphy (2010) mentions several advantages of a corpus-based approach to investigate age and language, stating that

While questionnaires gather information on people‘s perception of their linguistic patterns, a more reliable corpus-based approach focuses on samples of real language used by real people in natural[sic] settings and provides us with a context within which to interpret our findings, which is lacking if we rely solely on questionnaires. (Murphy 2010: 16).

Additionally, corpus-based tools allow researchers to examine the data quantitatively and qualitatively, thereby determining, for example, the most frequent forms used and their functions in the corpus. As demographic information is available for the BNC spoken language, speakers‘ ages and their use of GEs can be correlated easily. In this study, data from four age cohorts (15-24), (25-34), (35-44), and (45-59) are collected. However, data from speakers aged (0-14), (60+) and (unknown speakers‘ age) are not considered in this study. Table 3 shows the frequency of words investigated in Business and Leisure according to speakers‘ ages.

(20)

15

Table 3. Distribution of word counts across speakers‘ age-groups in Business and Leisure

Speakers‘ ages No. of words investigated

15-24 42,916

25-34 194,302

35-44 121,973

45-59 376,004

Total 735,195

As can be seen in Table 3, there is a variation in the size of samples found in the BNC. For example, the number of words produced by young speakers (42,916) is fewer than the amount of words produced by older speakers (376,004). The difference in the frequency of words is caused by the exigencies of sampling and availability (Aston & Burnard 1997: 28).

In spoken texts in the BNC, discourses are stratified by the gender of speakers (Aston & Burnard 1997: 31). In Business and Leisure, conversations are defined according to female, male, and unknown speaker gender. However, as this study aims to give an account of how female and male speakers use GEs, discourses comprising unknown speaker gender are disregarded. Table 4 provides the distribution of words produced by male and female speakers in Business and Leisure.

Table 4. Distribution of word counts according to gender of speakers in Business and Leisure

Gender of speakers No. of words investigated

Female 170,914

Male 564,072

Total 734,986

Table 4 presents a variation in the frequency of tokens used by male and female speakers. As can be seen, words produced by male speakers (564,072) are close to three times as many as those of female speakers (170,914).

In Business and Leisure, I searched for or something, and everything, and and stuff. In the process of data collection, all the occurrences of or something, and everything, and and stuff were compiled.

Next, these examples were manually examined to omit instances that did not match the characteristics of GEs. As GEs are typically formed by a noun phrase preceded by a conjunction (and, or), occurring in clause-final position, and as they are nonspecific in reference and extend otherwise grammatically

(21)

16

complete utterances (Overstreet 1999: 3), instances that did not match the definition were omitted from the analysis of this study. This applied to examples (13) and (14).

(13) I will try and get something like that brought up from North Wales (BNC FM2 1035) (14) I mean it's so easy to put stuff like that in isn't it? (BNC H61 1674)

In examples (13) and (14), something like that and stuff like that did not refer to preceding elements.

I found an instance of and stuff that was further specified by a relative clause, as in example (15), and two instances in which stuff was used as a verb, as in example (16), and eleven instances of and stuff that were used with a verb phrase, as in example (17). Those instances were omitted from the analysis of this study.

(15) Alright, just <pause> make some notes <pause> and stuff that you've got already. (BNC KGT 818)

(16) Erm there's nine hundred statements coming back in today but the units team that you all kindly lent me, we're all gonna sort through those and stuff them in envelopes and get them out the door. (BNC KLV 998)

(17) There's the call it's coming in I knew it would work in fact we'll talk straight away now just come straight through to me O nine O four six four one six four one and there's another one there so just pick up the phone and hear as we sit and talk and chew the fat and stuff like that hello? (BNC HV0 1215)

For and everything, in twelve instances everything was used as the subject of the clauses. These were thus removed from the analysis. Consider example (18).

(18) Can I, can I just say chairman, finally for point of [pause] clarification in case there's any doubt about it [pause] I'm not looking to turn it into the, into an absolutely like efficient business you know and everything succumbs to that. (BNC F7A 445)

In example (18), everything occurred in a coordinated clause introduced by and, and used as the subject of the clause everything succumbs to that. Moreover, the additive adverb else (Biber et al.1999: 563) occurred eleven times with and everything, as in example (19), and in nine instances and everything referred back to a verb phrase, as in example (20). These were also omitted from the analysis of this study.

(19) Well Bill stood up and was a bit negative really to the whole idea of parish constables, farm watch and everything else, it was quite disappointing. (BNC K6W 908)

(22)

17

(20) And that's it and the contracts had been signed and everything and it was <pause> (BNC J92 497)

Similarly, the adverb else was found three times with or something, as in example (21), nine instances of or something occurred in a coordinated structure by which or was used to indicate alternativeness (Biber et al. 1999: 79), and something was modified by an adjective, as in example (22). These instances were omitted from the analysis of this study. In addition, in thirty-five instances or something was coordinated with a verb phrase, as in example (23), in four instances or something was specified by a relative clause, as in example (24), and in ten instances or something was used with prepositional phrases, as in example (25). All of these cases were omitted from the analysis of this study as a consequence of not matching the definition of GEs.

(21) You see I mean it it did occur to me that er erm on on the three occasions that we've done this now, it always turns out to be er me involved in about just you know either playing the piano or something else in just about everything. (BNC FXR 750)

(22) It's not really a very clear definition is it? From a work point of view, degree of excellence, it's, it's too vague. So unfortunately, we're not able to manage quality if we use the dictionary definition of degree of excellence. We need to have something better, or something different.

(BNC H47 199)

(23) If you decide you want a couple of days off to take your grandson shopping or something like that, er or your husband even with his cheque book, then you just have to tell us. (BNC JAE 564)

(24) I'm just thinking maybe of a central diary or something that someone would keep, someone who you know you could relay them with phones. (BNC F7C 239)

Another criterion which led to the exclusion of some instances of or something was ambiguity. I found fifteen instances in which or something indicated unclear meaning and ambiguous reference, as in example (26).

(26) Up and talked to the guy <unclear> sitting in front of them and he said Yes we've got it but we haven't got it on those yet. So they have the C D ROM, they just haven't managed to <unclear> yet. <pause> Or something. <unclear>. (BNC FM2 600)

In example (26), the use of or something referred back to incomplete utterance they just haven't managed to <unclear> yet.

(23)

18

3.2 Data analysis

To answer the questions raised in this study, following Cheshire (2007), a combination of quantitative and qualitative methods was applied to explore the use of GEs and to track their use by different age- groups. Johnson (2007) presents an overview of mixed methods and states:

During the data analysis stage, quantitative data can facilitate the assessment of generalizability of the qualitative data and shed new light on qualitative findings. Alternatively, during the data analysis stage, qualitative data can play an important role by interpreting, clarifying, describing, and validating quantitative results, as well as through grounding and modifying. (Johnson 2007: 115)

To uncover the primary functions of GEs occurring in native speakers‘ data, the qualitative method identified whether GEs were used, for example, to complete three-part list, indicate intersubjectivity, category implication, or function as hedges, while the quantitative method provided the statistical analysis of the functions of GEs indicated by native speakers.

To test Jefferson‘s (1990) hypothesis that GEs occur in response to three-part lists, each item (noun phrase) used with GEs was counted, highlighted, and then coded. Consider examples (27) and (28), which show the coding system for items occurring with GEs.

(27) It's like interleave factors <-|-> and stuff <-|-> (BNC H61 1094) One item (28) And the chiller and the waterways and everything should all have been done all at <-|-> the

same time <-|->. (BNC JTC 282) Two items

In example (27), the speaker uses and stuff with the one item interleave factors, whereas example (28) shows that and everything is used with two the items the chiller and the waterways.

In the process of data analysis, the category of intersubjectivity was interpreted broadly by considering collocations co-occurring with GEs. For example, the use of the discourse marker you know has been taken into account due to the fact that you know is used to indicate intersubjectivity between speakers, and can be considered as a frame, allowing analysts to interpret the use of GEs on the basis of shared knowledge (Overstreet 1999: 74). Further, the addressees‘ responses were also considered, for example, the use of backchannels such as mm, umhm, and yeah indicating shared knowledge between speakers.

This study adopts the Gricean (1975) maxims of quantity and quality to analyze speakers‘ use of GEs.

For the maxim of quantity, Grice (1975) explains that successful communication between participants requires an adequate amount of information provided in talk, while the maxim of quality concerns a

(24)

19

high degree of truth and certainty (Grice 1975: 45-46). The use of and stuff, and and everything represents a good example of GEs operating on the maxim of quantity (see Overstreet 1999). As speakers have a preference for minimization in regards to the amount of information exchanged, AGEs such as and stuff and and everything are used to indicate that there is more information, even though it is not provided in the discourse (Dubois 1992: 182).

For the category of approximation, all expressions that denote an amount of money or time expressions, modal verbs that indicate uncertainty such as may, might and expressions such as I think occurring with or something were considered to show that speakers used or something as a hedge when their utterances were inaccurate. Consider examples (29) and (30).

(29) Gill: That is just stupid isn't it? It'll be about three pounds or something.

Nancy: Exactly.

Gill: Or it'll be about twenty quid or <-|-> something. (BNC KC7 1434)

(30) And I came home about I don't ken, maybe eight o'clock in the morning or something [unclear] a beautiful morning. (BNC HEA 238)

In example (29), Gill does not know how many pounds to pay. Therefore, he gives two estimations:

the first is three pounds, and the second is twenty quid, followed by or something. Likewise, in example (30), the speaker is unable to remember when he came back home. Because the speaker might have come earlier or later than eight o’clock in the morning, he uses or something to avoid violating the truth of his utterance. In addition, the use of maybe refers to probability (Biber et al.

1999: 868) and emphasizes that the speaker is uncertain about the exact time he mentions.

In order to detect whether speakers‘ ages and the gender of speakers have an essential role in the frequency and functions of GEs, I examined data available from male and female speakers aged (15- 24), (25-34), (35-44), and (45-59). The quantitative and qualitative methods required each occurrence and function of GEs to be counted, and then coded for speakers‘ age cohort and gender. Having established the quantitative method, following Biber, Conrad, and Reppen (1998), the frequency of GEs was counted, divided by the size of the corpus, and then multiplied by 100,000 words, to adjust raw frequency counts from texts of different lengths so that the texts can be compared accurately (Biber et al. 1998: 263). This process is called normalization.

4. Results

This section presents the results obtained from the quantitative and qualitative analyses. In section 4.1, I provide the overall frequency of GEs occurring in the data collected from two sub-corpora in the