The Pragmatics of Quoting Now and Then 9783110427561, 9783110431759

In examining the phenomenon of quoting from multiple angles, The Pragmatics of Quoting Now and Then offers a fresh view

297 49 3MB

English Pages 436 Year 2015

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Acknowledgements
Table of Contents
Introducing Quoting as a Ubiquitous Meta-communicative Act
Part I: Quoting Now
1. Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News
2. Quoting in Online Message Boards: An Interpersonal Perspective
3. The Complexities of Thread-internal Quoting in English and German Online Discussion Fora
4. Quoting in Political Discourse: Professional Talk Meets Ordinary Postings
5. Quotation and Online Identity: The Voice of Tacitus in German Newspapers and Internet Discussions
6. “Ich kenne da so einen Jungen … kennen ist gut, wir waren halt mal zusammen weg.” On the Pragmatics and Metapragmatics of X ist gut in German
7. Only “nur”. Scare Quoted (Exclusive) Focus Particles at the Semantics/ Pragmatics Interface
8. Manufacturing Credibility: Academic Quoting Across Cultures
Part II: Quoting Then
9. Quotative Markers in A Corpus of English Dialogues 1560–1760
10. Histories of Talking about Talk: Quethen, Quoth, Quote
11. Quoting and Translating Latin in the Old English Homilies of the Vercelli Book
12. Quoting and Plagiarising – Concepts of Both Now and Then?
13. In-between Cognitively Isolated Quotes and References: Looking for Answers Lurking in Textual Margins
14. Quotations in Early Modern English Witness Depositions
15. Haunting Evidence: Quoting the Prisoner in 19th Century Old Bailey Trial Discourse. The Defences of Cooper (1842) and McNaughten (1843)
16. Quotations from 17th and 18th Century Medical Case Reports1
About the Authors
Index of Subjects
Recommend Papers

The Pragmatics of Quoting Now and Then
 9783110427561, 9783110431759

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Jenny Arendholz, Wolfram Bublitz, Monika Kirner-Ludwig (Eds.) The Pragmatics of Quoting Now and Then

Topics in English Linguistics

Editors Elizabeth Closs Traugott Bernd Kortmann

Volume 89

The Pragmatics of Quoting Now and Then Edited by Jenny Arendholz Wolfram Bublitz Monika Kirner-Ludwig

ISBN 978-3-11-043175-9 e-ISBN (PDF) 978-3-11-042756-1 e-ISBN (EPUB) 978-3-11-042762-2 ISSN 1434-3452 Library of Congress Cataloging-in-Publication Data A CIP catalog record for this book has been applied for at the Library of Congress. Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available on the Internet at http://dnb.dnb.de. © 2015 Walter de Gruyter GmbH, Berlin/Boston Cover image: Brian Stablyk/Photographer’s Choice RF/Getty Images Typesetting: PTP-Berlin, Protago-TEX-Production GmbH, Berlin Printing and binding: CPI books GmbH, Leck ♾ Printed on acid-free paper Printed in Germany www.degruyter.com

Acknowledgements This volume represents the work of a large number of people. First and foremost, we would like to thank the authors for producing such excellent contributions to the intriguing field of quoting. We have admired their patience, learned from their research and enjoyed the cooperation. We are also indebted to a reviewer, whose feedback enhanced the quality of the contributions. Our special thanks go to Elizabeth Traugott, who acted as supervising series editor. She gave us all the guidance and encouragement that we needed by reviewing papers and providing invaluable input for the overall conception and structure of the volume. We editors were surrounded by competent, trustworthy staff in Augsburg and Munich. We heartily thank Miriam Rothfeld and Alexander Schäferling, who helped us with the desk editing, and, in particular, Christine Bélanger, who did a marvellous job checking, correcting, stylistically improving and proofreading most contributions. Finally, we are grateful to Julie Miess, Birgit Sievert and Wolfgang Konwitschny of De Gruyter for their technical advice and professional support. Munich, Augsburg (Germany), Albany, NY (USA)

Table of Contents Acknowledgements | v Wolfram Bublitz Introducing Quoting as a Ubiquitous Meta-communicative Act | 1

Part I: Quoting Now Daniela Landert 1 Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News | 29 Jenny Arendholz 2 Quoting in Online Message Boards: An Interpersonal Perspective | 53 Birte Bös and Sonja Kleinke 3 The Complexities of Thread-internal Quoting in English and German Online Discussion Fora | 71 Anita Fetzer and Elisabeth Reber 4 Quoting in Political Discourse: Professional Talk Meets Ordinary Postings | 97 Andreas Musolff 5 Quotation and Online Identity: The Voice of Tacitus in German Newspapers and Internet Discussions | 125 Rita Finkbeiner 6 “Ich kenne da so einen Jungen … kennen ist gut, wir waren halt mal zusammen weg.” On the Pragmatics and Metapragmatics of X ist gut in German | 147 Jörg Meibauer 7 Only “nur”. Scare Quoted (Exclusive) Focus Particles at the Semantics/ Pragmatics Interface | 177 Klaus P. Schneider 8 Manufacturing Credibility: Academic Quoting Across Cultures | 209

viii   

   Table of Contents

Part II: Quoting Then Karin Aijmer 9 Quotative Markers in A Corpus of English Dialogues 1560–1760 | 231 Colette Moore 10 Histories of Talking about Talk: Quethen, Quoth, Quote | 255 Winfried Rudolf 11 Quoting and Translating Latin in the Old English Homilies of the Vercelli Book | 271 Monika Kirner-Ludwig and Iris Zimmermann 12 Quoting and Plagiarising – Concepts of Both Now and Then? | 291 Jenny Arendholz and Monika Kirner-Ludwig 13 In-between Cognitively Isolated Quotes and References: Looking for Answers Lurking in Textual Margins | 319 Ursula Lutzky 14 Quotations in Early Modern English Witness Depositions | 343 Alison Johnson 15 Haunting Evidence: Quoting the Prisoner in 19th Century Old Bailey Trial Discourse. The Defences of Cooper (1842) and McNaughten (1843) | 369 Bettina Lindner 16 Quotations from 17th and 18th Century Medical Case Reports | 401 About the Authors | 419 Index of Subjects | 425

Wolfram Bublitz

Introducing Quoting as a Ubiquitous Meta-communicative Act 1 Taking up Another’s Voice In much of what we say or write, we incorporate someone else’s (or, indeed, our own) prior voice. Our communicative creativity is as contingent on the originality of how we phrase our verbal output as it is on the ingenuity of how we transfer previous discourse actually authored by others – or simply “constructed” (Tannen 1986: 313) by ourselves – into current interaction. Manifestations of verbal relocation may be the simple act of self- or other-repetition (serving inter alia as pause fillers, hesitation indicators, and cohesive devices) or the much more complex act of quoting. And it is this truly intriguing phenomenon of quoting, i.e. of speaking or writing through someone else’s tongue or pen, to which the studies in this volume are dedicated. Together they focus on forms, meanings and functions of quoting across a broad spectrum of genres and media, ranging from Old English homilies, Middle English dialogues, Early Modern English witness depositions and Early Modern German medical case reports, to present day print and online news, online message boards and discussion fora as well as academic writing. Integrating what other people said or meant (and even how they said it) into one’s own discourse seems to be a universal feature of all natural languages and, what is more, one of the more prevalent joys of ordinary life. Small wonder that people who investigate speech reproduction love to quote Bakhtin on this aspect:¹ [I]n real life people talk most of all about what others talk about – they transmit, recall, weigh and pass judgement on other people’s words, opinions, assertions, information; people are upset by others’ words, or agree with them, contest them, refer to them and so forth. (Bakhtin 1981: 338)

While quotations may not be as widespread as other forms of recycled speech such as ad hoc self- and other-repetitions or pre-patterned sequences (proverbs, slogans, routine formulae), they are undeniably pervasive and productive. In fact, while people have been quoting people from time immemorial, highly frequent and excessively exact quoting (accomplished by ingenious software) has become a characteristic (and arguably even constitutive) feature of the more interactive 1 Among them are Rebecca Clift and Elizabeth Holt, who alerted me to this quote (Clift/Holt 2007: 1).

2   

   Wolfram Bublitz

forms of computer-mediated communication. It is scarcely possible to contribute to blogs, tweets or message boards without quoting, i.e. without using the integral ‘quote’ function. Depending on how this is achieved, the act of reproducing and representing another person’s (or one’s own) prior communicative output is referred to, inter alia, as citing and reciting, impersonating and enacting, echoing, copying, reporting and, of course, as quoting. Such a wealth of terms is a clear indication of the complexity of speech reproduction. At the same time, it displays a terminological and conceptual mishmash, which I will not try to disentangle. Instead, in this volume we will focus on quoting only and proffer a clear-cut definition which sets it apart from the alternatives just listed. It is the act of quoting spoken or written (or even imagined) text alone as well as its products (quotations or quotes), its indicators (quotatives) and its functions to which the present volume is dedicated.

2 Defining Quoting 2.1 Some Opening Examples The existence of a rich and incongruous inventory of alternative terms is a clear indicator of a wide-spread dissension among scholars on how to define the reproductive acts, forms and functions involved. Rather than indiscriminately following heterogeneous and frequently fuzzy understandings of quoting, the contributions to this volume operate on a common coherent understanding of quoting which reflects our usage- and media-based pragmatic perspective. According to this authoritative definition, quoting is defined as an act of taking up text and, in doing so, performing a shift of context, focus and perspective. Before elaborating on these features, some opening examples of quoting as an act and of quotation or quote as its product should be given. As they occur in all kinds of (verbal) communication, the examples are taken from spoken and written discourse as well as electronically mediated forms of communication. The first extract is an example of spoken discourse and exhibits both otherand self-quotation. A US-American high school senior talks to her friends in their school’s writing lab:² I was at the Lacrosse game last night and Fred Regnery comes up to me and he’s like so who’s the only guy in your class and I’m like Dan Gray and he’s like I feel so bad for him. He’s

2 Italics indicate quoted text.

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   3

like Mr. Harris comes into our class and tells us every day how he feels so bad for Dan Gray because he’s with three girls who won’t shut up … (private recording)

The second example, a paragraph from a typed letter, illustrates quoting conventions in written discourse: I bought a new computer. You thought that my antique was too much of an antique, but I didn’t complain because I knew its tricks. However, in the last couple of weeks, the old 1989 job developed new tricks. It suddenly decided that a system error occurred and it wouldn’t open. It just sat and looked at me with a blank look on its face. So, I bundled it into my carrier, carried it down to the bus stop (it is heavy), and took it in to the University’s help desk. I called ahead and asked what I needed to bring. “Do I need to bring the keyboard?” “No,” the voice on the phone said, “we have keyboards.” Well, they didn’t have any keyboards that would fit my old computer. (private letter)

The third example is a short extract from an email: Hi Jenifer, How about quoting as a topic? In the invitation they request that “The talk should be ‘general’ enough so that people who do not know your area of expertise can follow.” Do you think quoting is general enough? (private email)

The fourth exchange is taken from a personal weblog: JD: The saddest part is that there was no donut. But that’s OK! We’ll have a belated celebration. CONGRATULATIONS!!!!! Babs: Donut? … donut? You’re on a British blog ‘ere you know! (Hoffmann 2012: 186)

Finally, this is a very short exchange from a chat: monkeymorales_usa: CRISJAY R U AUTHOR crisjay_brentfrittz84: i enter the room just to read the exchanging messages monkeymorales_usa: SORRY BUT IS A LITTLE STRANGE U COME TO READ MESSAGES AND NOT TO CHAT crisjay_brentfrittz84: try to look for someone who are good chatters on line… monkeymorales_usa: WHAT DOES THAT MEAN CISJAY monkeymorales_usa: LOOKINHG FOR GOOD CHATTERS? monkeymorales_usa: I DONT UNDERSTAND WHAT U MEAN BY THAT STATEMENT (Yahoo Chat: Generation X:16, last accessed 22 Apr 2006, adapted courtesy of Jenny Arendholz)

4   

   Wolfram Bublitz

Quoting is a complex act that incorporates several sub-acts. Take the preceding chat exchange as an example. Besides introducing or framing the quote with “what does that mean Cisjay”, Monkey performs these sub-acts by quoting Crisjay’s reply as looking for good chatters?: – he relocates Crisjay’s utterance, i.e. shifts it into a new, topically different context; – (in doing so) he directs the recipients’ attention to this act of re-contextualization, i.e. puts it into a new focus, thus constituting a communicative disruption; – he views the quoted text from a new perspective, thus re-evaluating it by clearly indicating that he does not agree with Crisjay or does not understand what he means.

2.2 Definition Taking the meta-communicative acts of re-contextualizing, re-focusing and reflecting upon a prior communicative occurrence as constitutive for quoting, we can define quoting as follows: A quoter 1. takes up another person’s (or their own) source text (T1) and shifts it from its original, prior context (C1) to the present context (C2) as a target text (T2), and in doing so 2. draws the recipient’s attention to T2, thus disrupting ongoing discourse, 3. and puts T2 in a new (evaluating) perspective either explicitly (verbally) or implicitly (prosodically or kinesically).³ Accordingly, quoting is changing in three ways: change of context, change of focus and change of perspective. These three acts of creating new context, focus and perspective, which set a quote apart from repetition and other forms of textual reproduction, are pivotal in the transformation of a simple speaker/writer into a quoter. A brief account of these changing acts seems to be in order.

3 To this we need to add a medium-induced amendment: The quoter can employ software to shift T1 from C1 to C2 as T2, which renders quoting a semi-automatized process, or quoting is performed by such software even without active instigation by a human agent, especially in highly interactive forms of blogging and social networking (cf. Bublitz/Hoffmann 2011: 437–441) .

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   5

2.3 Changing Context Of course and trivially, any relocation of a text involves a shift from one location to another. However, what is at stake here is not the existence of two contexts per se, but the quoter’s and the recipient’s knowledge of both of them. To understand the quote (i.e. the target text), the recipient has to recognize the original context from which the quoted text (i.e. source text) derives its meaning. The quote has to be read through both its prior and its present context. The meaning of the target text is created against the backdrop of the meaning-in-context of the source text, as in the following example. The expression “und das ist gut so” (‘and that’s okay as it is’) on this leaflet announcing a diversity day in Augsburg, Germany (with a range of events dedicated to age, religion, ethnicity, sexual orientation, etc.) is a quote:

(https://www.google.de/search?q=und+das+ist+gut+so/espv=210/es_sm=122/source=lnms/ tbm=isci/sa=X/ei=NKbeUtzGI4XWswbhuIDoDQ/ved=0CAkQ_AUoAQ/biw=960/bih=489, last accessed 7 Nov 2013)

To understand the phrase as a quote, it is absolutely necessary for the recipient to know the original context. Before he was elected mayor of Berlin, Klaus Wowereit in a speech on a party convention in 2001, came out and said “Ich bin schwul und das ist auch gut so” (‘I am gay and that’s okay as it is’). Since then, other groups with minority status have repeatedly used this quote, e.g.:

(https://www.google.de/search?q=ich+bin+Christ+und+das+ist+gut+so/espv=210/es_ sm=122/source=lnms/tbm=isch/sa=X/ei=rajeUoSRGsHYswa06oC4Bg/ved=0CAcQ_AUoAQ/ biw=960/bih=489, last accessed 7 Nov 2013)

6   

   Wolfram Bublitz

The quote is equally part of a prior and of a present context; its prior contextual meaning resonates in the present use. From this angle, we can say that the new text (T2) is “parasitic upon the ‘old’ text” (Bublitz/Hoffmann 2011: 446). The precondition that, for a quote to be a quote, both prior and present context must be shared knowledge for the quoter and the recipient induces the question: for whom is an expression a quote? We have to distinguish between three basic scenarios. Firstly, the speaker quotes a text without overtly indicating that it is a quote (omitting conventional means like inverted commas, verba dicendi, etc.) and the addressee does not know and/or cannot place the text into its original context. In that case, the act of quoting has failed. It is then only a quote in the eye of the quoter but not in the eye of the beholding recipient, who understands the text’s propositional meaning but not its contextually loaded meaning. Secondly, the speaker uses quotatives to clearly indicate that the text is a quote but the addressee cannot place the text into its prior context. In this case, the act of quoting has not been fully performed either (cf. examples in Arendholz/ Kirner-Ludwig, this volume.) Thirdly, recipients treat an utterance as a quote even though there is no indication on the part of the alleged quoter that it is a quote. Here is an example (cf. also Bublitz/Hoffmann 2011: 436). In a speech on Labor Day in Milwaukee some years ago, President Obama briefly went off speech, uttering the phrase: “They talk about me like a dog”: And, Milwaukee, that’s what we’re going to do again. That’s been at the heart what we’ve been doing over these last 20 months: building our economy on a new foundation so that our middle class doesn’t just survive this crisis – I want it to thrive. I want it to be stronger than it was before. And over the last two years, that’s meant taking on some powerful interests – some powerful interests who had been dominating the agenda in Washington for a very long time. And they’re not always happy with me. They talk about me like a dog. (Applause.) That’s not in my prepared remarks, it’s just – but it’s true. (http://www.whitehouse.gov/the-press-office/2010/09/06/remarks-president-laborfestmilwaukee-wisconsin, last accessed 7 Nov 2013)

The line “They talk about me like a dog” was treated as a quote and extensively debated on the internet and by the press, even though we cannot be sure at all that it was meant as a quote. If you google the expression you get more than 130 million hits (7 Nov 2013). Some users, for example, considered the line to be an expression of “Obama’s true Muslim nature” (http://languagelog.ldc. upenn.edu/ nll/?p=2616 (last accessed 7 Nov 2013).) and his “Indonesian/Muslim … upbringing”:

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   7

Yes, this is probably his Indonesian/Muslim early upbringing speaking. Americans might say, “Treat me like a dog,” but in my 56 years I have never heard the expression, “They talk about me like a dog.” (http://michellemalkin.com/2010/09/06/dog-daze/, last accessed 7 Nov 2013)

In contrast, a number of journalists felt that Obama was actually quoting the Jimmy Hendrix song ‘Stone Free’, which features the expression in one of its verses: … people try to pull me down, they talk about me like a dog, talk about the clothes I wear … (http://languagelog.ldc.upenn.edu/nll/?p=2616, last accessed 7 Nov 2013)

Many speculated as to the expression’s origin: I’m 60 years old and born/raised in Georgia and trust me … Ive heard this phrase all my life. I was a country girl … so maybe it came from the language of “country folk” … inherited as much as our other “southernisms” from the Scots and Irish mishmash of language that evolved from our ancestry … throw in a little Black America … you have it! (http://languagelog.ldc.upenn.edu/nll/?p=2616, last accessed 7 Nov 2013)

Thus, speakers, writers and users can be unaware of using an utterance which was (actually or supposedly) used before by somebody else while their recipients may come up with interpretations in the light of all kinds of former contextualizations. One further aspect of ‘changing context’ as a feature of quoting needs mentioning. Quoting may or may not change a text. For our understanding of quoting, it is of little or no avail if the quote is verbatim or not. Exceptions are, of course, slogan-like expressions like ‘und das ist gut so’. But even there, minor lexical variation (‘und das ist gut so’ instead of ‘und das ist auch gut so’) is tolerated. We can thus support Clark/Gerrig (1990) and others, who argue against the requirement that for a direct quote to count as a quote it has to reproduce the actual words of the prior text.⁴ 4 Instead, they define quotation as a kind of “demonstration”, not in the sense of ‘pointing’ to the original text with, e.g., a demonstrative pronoun, but of exhibiting, showing or “depicting” (rather than describing) the original utterance event (cf. Clark/Gerrig 1990: 765f.). Following Goffman’s (1974: 74) distinction between serious and non-serious actions, they hold that demonstrations are ‘non-serious actions’. This accords with Sbisà/Turner (2013), who put quotation into the category of ‘aetiolated’, i.e. non-serious speech, “next to joking, fiction writing, playacting” which “are not … actions, but pretended actions” (2013: 7). Cf. also Pennesi (2007) and Tannen (1986) who characterise quotes as cases of “constructed dialogue” (1986: 313) because

8   

   Wolfram Bublitz

2.4 Changing Focus Any utterance that is obviously not authored by its present speaker but borrowed from someone else arouses the recipients’ attention and makes them change their focus. The quote causes the recipients to “attend to” the text (as a quotation) rather than “attend from it”, to borrow a distinction by Anton (1998: 199). When attending to the text, the text is no longer “an absent body … which disappears for the sake of the meaning … intended” (1998: 199) but is very present and in the recipients’ attentional focus. Quoting is thus a meta-communicative act, which creates “a disturbance in the flow of communication” (Hübler/Bublitz 2007: 7), a “communicative disruption” which “re-aligns the participants’ interpretive attention” (Bublitz/Hoffmann 2011: 434).

2.5 Changing Perspective The third and arguably most important feature of quoting is the ascription of a particular significance or value to the quoted (source) text by reflecting upon it, i.e. by putting it into a new perspective of assessment. The quoter creates a reflective surplus by conveying his or her perspective towards the quoted text. Take the “und das ist gut so” quote discussed above and the following example, a selfquotation from spoken discourse: Jemma is reporting to her friend Ivy an incident that had occurred earlier that day … “… so yeah when he told me that I just looked at him and I’m like … are you serious” (Sams 2007: 1)

At first sight, the old and the new text in both examples may appear to be similar in form, meaning and function, but, of course, they are different in at least these ways: – In the first example, das refers to two different referents: ‘Ich bin schwul’/‘I am gay’ versus “Augsburg ist vielfältig”, and so does you (“are you serious”) in the second example: from the present participants’ perspective, the person Jemma originally addressed is not the same as the person presently quoted and referred to (even though physically he is certainly one and the same person). – The respective contexts are different. quoters spontaneously construct a replica of what someone else actually said or wrote or, indeed, could have said or written.

Introducing Quoting as a Ubiquitous Meta-communicative Act   





   9

The quoted author and the quoter are, of course, different in the first example, and they are also different in the second example with self-quotation. The (current) quoting speaker and the (former) quoted speaker are only the same person physically, but not emotionally and psychologically. They play different roles and display different attitudes and stances, particularly so because the quoting speaker’s perception of the original speech situation and its participants has changed.⁵ The quoted speaker and the quoting speaker do, accordingly, look at their respective utterances in a different way.

The quote conveys a new stance; it carries a new message. In fact, this seems to be the essential motive for performing the act of quoting in the first place: not to recycle prior text to inform the recipient but to allow the quoter to express his or her stance towards the quoted text.

3 Types of Quoting Before turning to forms and functions of quoting, a brief account of the various types of quotes is called for. Most recent approaches have agreed on this set of major quotation types:⁶ – pure (or metalinguistic) ‘quotes’ (“She” is a pronoun), – direct quotes (She said: “I hate deadlines”), – indirect quotes (She said that she hated deadlines), – mixed quotes (She said that this handbook “tops all handbooks”), – echo quotes (John: “I hate deadlines” – Mary: “I hate deadlines – is that all you have to say?”), – scare quotes (She said that this “handbook” is neither comprehensive nor up to date), with which quoters distance themselves from someone else’s words, often in an ironical and ‘sneering’ way).⁷

5 Referring to Goffman, Golato (2002) relates that “the ‘I’ used in the quote does not necessarily have to correspond to the speaking ‘I’. The speaker might have changed or developed somehow in the time between the original event and its telling, possibly permitting him or her to thus look at the event itself from a different perspective” (Golato 2002: 60). 6 Cf., e.g., Cappelen/Lepore (1997), Recanati (2001), Brendel et al. (2007) and (2011) as well as the slightly different classification by Gutzmann/Stei (2011). 7 And there are others like “mendacity” or “claim quotes” (The author ‘committed plagiarism’) which also serve as a “distancing device …, to indicate that this is just a claim” by some other source (Geoffrey Pullum, http://languagelog.ldc.upenn.edu/nll/?p=1017, last accessed 7 Nov

10   

   Wolfram Bublitz

Of these, the so-called pure quote occupies a special status. It is a misnomer because the metalingual expression (here ‘she’) has not been mentioned before by someone else and neither involves a change of context, focus or perspective. It simply does not meet the requirements of our definition of a quote. Some linguists do, nonetheless, regard these metalinguistic forms as ‘quotes’ on the grounds that, for instance, ‘she’ in She is a pronoun is not used but merely mentioned. However, the use-mention distinction is not a valid criterion that can be employed to set quoted text apart from non-quoted text. Not all types of quotes display quoted expressions that are mentioned and not used. Typically, the usemention distinction is, for instance, applied to set direct quotes apart from indirect (and other types of) quotes (cf., e.g., Brendel et al. 2011: 4f., Saka 1998): while ‘I hate deadlines’ in the direct quote (cf. above) is mentioned but not used, ‘that she hated deadlines’ in the indirect quote is used but not mentioned.⁸ Only a minority of the aforementioned types of quotes are addressed in the contributions to the present volume. The most frequent type investigated is the direct quote, but there are also papers dealing with indirect quotes (e.g. Fetzer/ Reber, Johnson), mixed quotes (e.g. Musolff, Finkbeiner, Schneider, Rudolf), and scare quotes (Meibauer). Direct quotes are usually easy to detect because they display certain graphic or prosodic features, or are introduced (actually: announced) by some sort of quotative (cf. below).

4 Forms and Functions of Quoting Most of the means that are conventionally used in Present-Day English to indicate that the currently produced or about to be produced text is authored by someone else were unknown in the early stages of English. Moore accentuates the lack of “set punctuation conventions for the marking of quoted speech … in early English manuscripts”; quoting was not signaled through quotation marks but “through linguistic and scribal means: through the words of the texts (particularly the verbs of speaking) and through scribal marks on the manuscript page” (Moore, this volume). For quoting, the most frequently occurring verbum dicendi was the

2013); greengrocer’s quotes (“FRESH POTATOES!”), to emphasize and draw the passer-by’s attention to it; and emphatic quotes (Our potatoes are “local” produce!). 8 In mixed quotes and echo quotes, the quoted expression is both used and mentioned, as it is in scare quotes, where, however, it is not necessarily preceding, i.e. the scare-quoter is not always referring “to an original utterance” (Brendel et al. 2011: 5).

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   11

verb quethen (‘to say’),⁹ whose forms were used “to mark a piece of discourse as a de-dicto expression (i.e. a representation of the words as they were spoken)” (Aijmer, this volume).¹⁰ Depending on mode and medium, quoting rests on a variety of different formal devices. In spoken discourse we mostly find verbs of saying and prosodic signalling, in written discourse there are graphic as well as lexical and syntactic indicators, while electronically mediated communication also resorts to arrangement and layout. In the languages of the world, the range of such devices is very impressive.¹¹ Here is a moderately complete list of means that can be used in English to tag a verbal expression as a quote in spoken and written discourse: – kinesic and prosodic variation,¹² – quotation marks, which may be spoken (“quote … unquote”), gesturally enacted (doing air quotes) or written (inverted commas “…”, the quote-preceding colon), – verba dicendi et sentiendi, including recent ones like go, go like, be like (cf. Buchstaller/van Alphen 2012), and a plethora of corresponding quoting nouns of saying and meaning, – logophoric forms, i.e. pronouns whose antecedents refer to those persons whose speech, thoughts, feelings are reported (cf. Roncador 1988), – syntactic variation, including tense and reference shift, i.e. shift of the deictic centre (as in indirect quotes where “I hate deadlines” can become “She said that she hated deadlines”). In electronically mediated forms of communication, there are new ways of quoting that have been instigated by computer- and internet-based technology which range from copy-paste quoting (e.g. in weblogs) via semi-automatized quoting (e.g. in emails and online discussion fora) to fully automatized quoting (e.g., in social networking sites).¹³ Hand in hand with the development of new hard- and software, quoting has been getting increasingly automated. The various stages

9 Not to be confused with the verb to quote, as Moore points out in her paper on ‘Histories of talking about talk: quethen, quoth, quote’ in this volume. 10 For the ‘de re’ vs. ‘de dicto’ reading of a quotative verbum dicendi, Aijmer refers to Güldemann (2008: 133) and Moore (2011: 84). 11 For overviews and in-depth coverage see Klockow (1980), Roncador (1988), Güldemann/ Roncador (2002), Güldemann (2008) and Buchstaller/van Alphen (2012). 12 Cf. Günthner (1999), who notes that they are mainly used to animate the quoted author, and Sams (2007) and Jones/Schieffelin, who discuss the “vividly enacted voicings of the speech, thought, and demeanor of different personae” (2009: 78). 13 Cf. Bublitz/Hoffmann (2011: 437–441) and, for fully automated quoting in Facebook, Eisenlauer (2013).

12   

   Wolfram Bublitz

can be ranked according to their level of technical reproductivity, which goes in ascending order from quoting in weblogs to quoting in social networking sites, and to their degree of what one could call quoter intentionality or command of the act of quoting (cf. Eisenlauer 2013). Quoting as a meta-communicative act has become an indispensable discursive tool in some forms of computer-mediated communication, where quoting, supported by specifically designed software, has led to a previously unknown degree of formal congruence. While in traditional media it is not always easy to specify the boundaries of a quote, in particular where it ends, this is different in computer-mediated communication where quotes are precisely indicated by a variety of means (frames, indentation, etc.). A unique facet of the present edited volume is its focus on the functional evolution of quoting both from a chronological (“now and then”) and a mediumrelated, technological point of view. Traditionally, quoting has served a wealth of different functions; among them: – attesting the quoter’s knowledge (how well read he or she is), – establishing and increasing reliability, trustworthiness and credibility, i.e. the “credentializing function” (Bednarek/Bublitz 2006: 551), – establishing and increasing the quoter’s authority (by, e.g., referring to other authorities), – enhancing the stylistic value of a text, – and some others.¹⁴ The traditional stock of functions performed by quoting in speech and writing has grown considerably with the advent of internet-based forms of communication. This is not so say that in digital communication quoting is no longer or only rarely used to perform those long-standing functions. But communication in the age of the internet has fostered new and strengthened some old (rarely exercised) functions, among them: – creating narrative immediacy (e.g., quoters employ quotes “to propel their audience into the ongoing action of their verbal accounts”, Bublitz/Hoffmann 2011: 441), – simulating dialogicity (especially in narratives), – eliciting adjacency, as described by Herring who points out that quoting “creates the illusion of adjacency in that it incorporates and juxtaposes (portions of) two turns, an initiation and a response, within a single message. When portions of previous text are repeatedly quoted and responded to, the 14 Cf. Pennesi (2007: 2) for a listing of a number of (rather more marginal) functions, also Hübler (1989: 23).

Introducing Quoting as a Ubiquitous Meta-communicative Act   





   13

resulting message can have the appearance of an extended conversational exchange” (1999: 15), indicating affiliation or, indeed, “devaluation” of the quoted text; furthermore, “the evaluative surplus of quotes not only enables quoters to appraise the original utterance but also its author” (Bublitz/Hoffmann 2011: 443), generating serial knowledge (cf. Bublitz/Hoffmann 2011: 444f. and Hoffmann 2012: 193ff.) and general knowledge by, inter alia, sharing content.

In the present volume we find a sampling of these ‘old’ and ‘new’ functions, supplemented by one or two others not listed above (cf. section 6 below).

5 Previous Research Even though quoting is a common and ancient practice in every kind of communication¹⁵ and, given that it constitutes a longstanding area of linguistic and interdisciplinary inquiry, is by no means terra incognita in text-based sciences¹⁶, it remains a somewhat vague notion due to a number of vexing characteristics. To name the most significant ones: – Neither (the act of) quoting nor quotation (as its product) are per se distinct and stable phenomena, incontestably identifiable in any given text, but instead variant and fuzzy concepts that depend on an interpretation within a particular theory and field (apart from linguistics, rhetorical and literary science, philosophy, logic and psychology).¹⁷ Aspects of such elusiveness and constructivist conditionality are discussed in several articles of this volume (among them Arendholz/Kirner-Ludwig, Kirner-Ludwig/Zimmermann, Landert, Musolff).

15 Despite the fact that it is not generally cherished, neither in academia, if we go by Davidson’s dejecting verdict that a quotation is “a somewhat shady device” (1979: 79), nor in non-academic circles, e.g. internet-based communities whose stance Severinson Eklundh summed up in the following nutshell: “The results of a survey among users of eight discussion lists showed that this strategy [of quoting] was among the most negatively ranked of net behaviours, only exceeded by the posting of racist and sexist messages” (2010: 5). 16 There is no shortage in studies on quoting or quotation (our personal bibliography lists some 300 titles), neither in linguistics, cf. below, nor in literary and philosophical studies, cf. e.g. Mayer/Jacob (2010). 17 Cf. the overviews and articles in Brendel et al. (2007) and (2011), Mayer/Jacob (2010) and Cappelen/Lepore (2007).

14   







   Wolfram Bublitz

Looking back in history, it is obvious that not only forms and functions of quotations have changed over the last centuries (cf. the papers by Johnson, Lindner, Moore, Rudolf in this volume) but that the language users’ understanding of and attitude towards quoting and quotation have undergone a striking metamorphosis. This is also demonstrated by Kirner-Ludwig/Zimmermann (this volume) who look back in history and juxtapose the two concepts of ‘quoting’ and ‘plagiarizing’ in medieval texts. And today, as we all know and widely lament when adopting a pronounced moral stance, we are still far from general consent on how to distinguish clearly between these two concepts (or, for that matter, between quoting and paraphrasing). The notoriously intricate nexus between quoting and plagiarizing is complicated by the fact that different cultures frequently hold different views of quoting and plagiarizing (cf. Coulthard/Johnson 2007: 186ff.). Over the last two or three decades, new technology and media have bred alternate ways of quoting (e.g., by activating hyperlinks that automatically display ‘quoted material’) and new formats (among them quote frames); the contributions of Arendholz, Musolff, Fetzer/Reber, Bös/Kleinke in this volume bear witness to this enormous progress in technology and mediation. Outside internet-based communication, new forms and patterns of introducing and thus marking quoted speech are also permanently emerging; such new “quotatives” (Buchstaller/van Alphen 2012) are explored by Aijmer and Meibauer in this volume.

The recounted technological developments (re hardware, software and Web 2.0 interaction) and changing communicative conventions (such as the emergence of new quotatives in youth language) have not only incited much interest in neighboring disciplines (and, with regard to the blurry distinction between quoting and plagiarizing, even in the general public). They have first and foremost alerted linguists, including the contributors to this volume, who have taken a renewed interest in investigating the forms, acts and functions of quoting. There is a lot of new uncharted territory to cover! The past decade has witnessed the publication of several edited volumes on quoting, which, however, adopt a different perspective from ours. Most notable are the two collections edited by Brendel/Meibauer/Steinbach (2007) and (2011), whose outline is primarily philosophical and semantic as opposed to our own pragmatic and media-driven approach. While they are mostly concerned with printed forms of communication from a syntactic point of view, we look at the whole range of spoken, printed and electronic forms of communication, both from a synchronic and a diachronic perspective. Furthermore, they investigate traditional, analog ways of signaling quotations in writing and speak-

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   15

ing, whereas our pragmatic approach is also concerned with new digital ways of signaling quotations in electronic media. Lastly, while they cover traditional functions of quoting like stylistic enhancement or authentification, the contributors of this volume additionally explore new medium-induced functions. Focusing on overt indicators of quoting, the recently published edited volume by Buchstaller/van Alphen (2012) investigates quotatives such as go and go like, be like, this is + speaker across various languages. We duly acknowledge the influence of previous scholarship on the subject of quoting, even though we rarely share the definitions of quoting advocated. Acclaimed edited volumes and monographs include Coulmas (1986), Roncador (1988), Lucy (1993) and, more recently, Güldemann/Roncador (2002), and Clift/Holt (2007).¹⁸ Together they regard quoting either as a synonym of direct speech or reported speech, or as a subcategory of reported speech or reflexive language.

6 Structure of the Volume and Preview of Articles The majority of papers collected here, actually 12 out of 16, originate from contributions to two international conferences.¹⁹ As we sought to proceed in a novel and arguably unusual way, both the original and the additional papers not only cohere in sharing the same concept of quoting (as presented in section 2.2 ff above), they also conform to the following twofold objective. To account for multimodal variance in form and meta-communicative bias in function, we align the now and then of quoting by adopting a synchronic or diachronic perspective, on the one hand, and by investigating media induced developments, on the other, i.e. exploring how and why speakers, writers or users quote in speech, in ‘old’ handwritten or printed, and in ‘new’ internet related media. This compilation pays a special tribute to the noticeable and linguistically rewarding difference between new digital and old analog forms and practices of quoting, thus forging a bridge between the forms and functions of quoting then and of quoting now.

18 Cf. also Bednarek/Bublitz (2006) and Bublitz/Hübler (2007) in which earlier research into forms of speech reproduction are reviewed. 19 The panel The Pragmatics of Quoting in Computer-mediated Communication organized at the 12th International Conference of Pragmatics (IPrA) held in Manchester, UK, in July 2011, and the 3rd International Conference on Quotation and Meaning (ICQM3) which the editors of this volume hosted in Augsburg, Germany, in April 2012, continuing the series of International Conferences on Quotation and Meaning I and II in Mainz 2007 and Berlin 2008, instigated by Brendel, Meibauer and Steinbach.

16   

   Wolfram Bublitz

To our knowledge, this edited volume thus presents the first concise linguistic investigation of quoting and quotation which encompasses synchronic and diachronic views, on the one hand, and electronically as well as non-electronically mediated forms of communication as data, on the other. Rather than focusing on literary, logical or philosophical facets of quotations, our goal is to uncover formal, structural and functional features by applying a decidedly linguistic approach (mostly with a pragmatic-interactional or semantic-cognitive basis). All papers offer a fresh perspective on various (mostly pragmatic) aspects of quoting now and then, whose coverage has been neglected so far. Adopting an interdisciplinary stance, we assemble (interpersonal) pragmatics, sociolinguistics, discourse analysis as well as historical and cultural studies in one volume, thereby providing a more comprehensive and integral understanding of the nature of quoting. The volume brings together contributions that deal with quoting in a broad variety of textual genres and media, taking their data from earlier and presentday stages of English (in most papers) or German. With the exception of Moore’s comparative study of (the historical emergence and development of) the quotative verbs quethen, quoth, quote, all articles are corpus-based studies which take their data from – homilies (late 10th century): Rudolf, – medieval manuscripts: Kirner-Ludwig/Zimmermann, Arendholz/Kirner-Ludwig, – witness depositions (16th to 18th century): Aijmer, Lutzky, – medical case reports (17th and 18th century): Lindner, – trial records (19th century): Johnson, – academic essays: Schneider, – newspaper headlines in print: Meibauer, – newspaper reports (and comments) in print: Musolff, – newspaper reports online: Landert, – TV interviews and their online comments: Fetzer/Reber, – online discussion fora/message boards: Bös/Kleinke, Musolff, Finkbeiner, Arendholz, Arendholz/Kirner-Ludwig. Of the inventory of quotation types acknowledged in linguistic research (cf. section 3 above), some papers focus on indirect quotes (e.g. Fetzer/Reber, Johnson), mixed quotes (e.g. Musolff, Finkbeiner, Schneider, Rudolf), and scare quotes (Meibauer). The majority of contributions, however, investigate direct quotes, which are easy to detect, even in medieval manuscripts (cf. Rudolf, Kirner-Ludwig/Zimmermann, Arendholz/Kirner-Ludwig) and post-medieval texts (cf. Aijmer, Lutzky, Lindner, Johnson), because they mostly display certain graphic (or prosodic) features or

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   17

are introduced (actually announced) by some sort of quotative. Of these quote indicators, verbs of saying, quotation marks, syntactic variation and hyperlink induced frames feature prominently throughout the collection. As mentioned above (cf. section 4), one overarching feature of all contributions is their functional orientation. The most prominent functions of quoting described in both diachronic and synchronic papers are these: – establishing and increasing credibility by alignment, support or distancing (Aijmer, Lutzky, Moore, Johnson, Fetzer/Reber, Landert, Schneider), – establishing and increasing the quoter’s authority (Kirner-Ludwig/Zimmermann, Musolff, Rudolf), – establishing and maintaining (group-)identity by, inter alia, face-work and personalization (Arendholz, Arendholz/Kirner-Ludwig, Bös/Kleinke, Schneider and Landert respectively), – foregrounding the quoter’s evaluation of the quotee or the source text (Landert), – sharing and negotiating content (Finkbeiner, Meibauer), – creating coherence (Bös/Kleinke). To reflect our dual synchronic and diachronic perspective on quoting, the 16 contributions to this volume are allocated to two sections: Quoting Now (eight articles) and Quoting Then (eight articles). As outlined above, these two groups of articles cohere in that they share not only the same concept of quoting but also the objective to explore the multimodal variance of its forms and the metacommunicative bias of its functions in relation to media induced developments from old analog to new digital quoting practices. The first, synchronic part, Quoting Now, features five papers (by Arendholz, Landert, Bös/Kleinke, Fetzer/Reber, and Musolff) which discuss various aspects of quoting in different forms of computer-mediated communication (including online newspapers), two papers on specific quoting formulae in German (by Finkbeiner, Meibauer), and a final article on intercultural academic quoting (Schneider). Obviously, the range of media-induced forms and mechanisms sets these papers apart from those in the second, historical section. Due to new technology, computer mediated texts can and regularly do combine oral with written and even (typo)graphical features, thus instigating a rich, previously impossible variety of quoting practices. In varying degrees, all papers in the synchronic part (covertly or overtly, as in Musolff’s paper) depict forms and functions of quoting that also occur in the study of quoting in old texts to follow. Noticeably, this applies to those papers which establish quoting today as a strategy for setting up social relations and evaluating (align with or challenge) a third party’s position (Arendholz, Bös/Kleinke, Fetzer/Reber, Musolff and Schneider).

18   

   Wolfram Bublitz

In more detail, the synchronic part assembles these contributions: In her opening article Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News, Daniela Landert examines how the functions that have generally been ascribed to direct quotes in traditional news media are currently undergoing significant modification in the news reports of online British newspapers. On the basis of a host of examples such as this, With Africa’s football championship set to open as planned on Sunday, other teams spoke of their fear that they, too, might become the target of attacks. “We have goose bumps … who knows what is going to happen to us,” said Amade Chababe, Mozambique assistant coach. (Landert, this volume)

Landert conclusively shows that two functions have substantially changed in online news: direct quotes are increasingly used to mark the quoted item as a reportable fact (thus corroborating the enduring process of ‘tabloidization’ in news media), on the one hand, and to enhance the quoter’s, i.e. the news actor’s own voice (which confirms the tendency towards personalization in the mass media), on the other. In Quoting in Online Message Boards: An Interpersonal Perspective, Jenny Arendholz deals with examples such as this one:

Adopting an interpersonal pragmatic point of view, she shows how quoting in message board communication is regularly used to establish, maintain and even redefine interpersonal relations between participants. To manipulate face, online quoters approve or disapprove of the quoted text, which verifies her thesis that the act of quoting is increasingly used as a tool for doing facework. For their contrastive investigation of The Complexities of Thread-internal Quoting in English and German Online Discussion Fora, Birte Bös and Sonja Kleinke have compiled a corpus of public message boards taken from Spiegel Online and BBC Have Your Say, of which this is an example: Significantly, in his welcome the President also said: “In a world where some treat life as something to be debased and discarded, we need your message that all human life is sacred and that each of us is willed.” (Bös/Kleinke, this volume)

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   19

While the German discussion forum is characterized by a higher degree of quoting and, more generally, interactivity than the English forum, both use quoting as a complex means of creating coherence and establishing social relations, though not in the same way. Anita Fetzer and Elisabeth Reber’s contribution Quoting in Political Discourse: Professional Talk Meets Ordinary Postings explores the forms and functions of quotations in mediated political TV interviews and their online comments within a pragmatic and interactional linguistic framework. The following interview question illustrates how the quoting of previous text (here by the quoter, William Hague, himself) can be strategically used (or even exploited) by the quoter to align himself with auspicious political positions (even with his own) in order to strengthen the current line of argumentation: no, no we say there are other changes to tax that we would like to make but that that depends on the state of the economy (Fetzer/Reber, this volume)

In perusing reports and comments in German newspapers and online discussion fora covering the 2009 bi-millennial commemoration of a Germanic revolt against Imperial Roman rule, Andreas Musolff in Quotation and Online Identity: The Voice of Tacitus in German Newspapers and Internet Discussions discovers that both groups of writers frequently quote from Tacitus (and other classical texts) to establish and increase their own authority or challenge the authority of others, as in this newspaper excerpt: [In den „Annalen“ des römischen Historikers Tacitus] war unter anderem von einem Mann namens Arminius die Rede, der „unbestritten der Befreier Germaniens“ gewesen sein soll. (‘Among other historical figures, Tacitus’s Annals mentioned a man called Arminius who was said to have been “without doubt Germany’s liberator”.’) (Musolff, this volume)

Using a meta-representational and relevance-theoretical framework, he convincingly demonstrates that in online discussion fora such reference to a classical authority is an appropriate means to establish and develop discursive identities in the respective community of practice. In her paper “Ich kenne da so einen Jungen … kennen ist gut, wir waren halt mal zusammen weg.” On the Pragmatics and Metapragmatics of ‘X ist gut’ in German, Rita Finkbeiner advances a pragmatic analysis of the German construction X ist gut (‘X is good’, with X being the quoted expression), using as data examples from online discussion fora such as: Aldiboz: scheint ein kleines problem vorzuliegen… ‘We seem to have a little problem here…’

20   

   Wolfram Bublitz

Gismaxx66: Hi, kleines Problem ist gut Der Server ist derzeit nicht zu erreichen. ‘Hi, little problem is good The server is not accessible at the moment’ (Finkbeiner, this volume)

The formulaic predicate ist gut indicates that the quoted item in the X-slot is not to be understood in its stereotypical meaning because it is incongruous with the original context. ‘X ist gut’ as a meta-communicative formula also has a clear evaluative function in either approving or disapproving the quoted author’s choice of words. In his paper Only „nur”. Scare Quoted (Exclusive) Focus Particles at the Semantics/Pragmatics Interface, Jörg Meibauer deals with newspaper headlines like Spears zahlt Mann „nur“ 750.000 Euro (‘Spears pays husband “only” 750,000 Euro.’) (Meibauer, this volume)

where the German focus particle nur (‘only’) elicits the interpretation that the sum of 750,000 Euros is not a lot of money for Spears but definitely is for the average citizen. The scare-quoting interpretation is triggered solely by the enclosing quotation marks. Meibauer explains that such an interpretation is the outcome of a specific type of implicature, as proposed in Levinson’s pragmatic ‘presumptive meanings’-model. In the final contribution to the first (synchronic) part of this volume, Manufacturing Credibility: Academic Quoting Across Cultures, Klaus P. Schneider compares the functions of quoting in academic texts that are practiced in different languages and cultures, academic disciplines, and communities of practice. Among other results he discovers that in academic writings quoting is often used as a positioning strategy and a means of establishing a positive professional identity, as in this example titled ‘Doing counter-attack’: “The polemical tone which leads him, for example, to describe our work imaginatively as a ‘rearguard action’ of a type ‘carefully designed […]’ is best ignored.” (Schneider, this volume)

The second, diachronic part, Quoting Then, presents eight papers that look into quoting practices in the early stages of the English language (Aijmer, Moore, Rudolf), ask how to identify quotes in general (Arendholz/Kirner-Ludwig) and how to delimit them from plagiarism (Kirner-Ludwig/Zimmermann), and discuss the forms and functions of quotes in various genres from the 16th to the 18th century (Lutzky, Johnson, and Lindner). As was already pointed out, some characteristics (formal and functional) of quoting have remained stable over the centuries, thus constituting an overarching feature shared by quoting in old and new

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   21

media. The aim of establishing and securing credibility by means of quoting, as discussed in Schneider’s paper (cf. above), for instance, is also a major facet of earlier language periods. Furthermore and possibly somewhat unexpectedly, old and new texts cohere in the broad variety of quoting practices employed, which is evident in all diachronic papers. They also demonstrate the rich complexities in formal indications of quotes common in old texts. In more detail, the diachronic part encompasses these contributions: In her corpus of witness depositions, Karin Aijmer in Quotative Markers in a Corpus of English Dialogues 1560–1760 did not come across a single case of unmarked quoting. Direct quotations are, on the contrary, regularly marked as such by a combination of several different devices. They range from the grammaticalized, i.e. semantically weakened verbum dicendi ‘say’ and deictic pointers (‘as followeth’, ‘namelye’, ‘to this effect’) to lexical and pragmatic markers (some of them no longer current in present-day English), which express the speaker’s attitudes towards the quote. Pragmatic markers typically co-occur with quotative verbs, but are, unlike the latter, internal to the quoted speech, as in this example with ‘well’: well [$sayde the quenes attourney$] well, can you make it come before vs nowe, if ye can we will (Aijmer, this volume; Aijmer uses dollar signs to mark off non-speech from direct speech quotations)

In Histories of Talking About Talk: Quethen, Quoth, Quote, Colette Moore disentangles the different meanings and functions of the two (phonetically and orthographically similar) verbs of saying ‘quoth’ and ‘quote’, which are easily perceived as being related and thus semantically matching words, even though they are etymologically unrelated. Moore traces the historical origins of the two verbs and their development from Early Modern to Present-Day English and demonstrates that they serve different pragmatic functions. ‘Quethen’ gradually loses its morpho-syntactic range (mainly occurring in inverted word order and in the first or third person) and turns into a grammaticalized quotative marker that is generally used as a fixed inquit formula, as in this example: To holy chirche quod he • for to here masse (Moore, this volume)

It did not, however, survive into modern English (except for the fixed form ‘quoth’) because of the rise of quotation marks (inverted commas) as direct speech markers which began in the eighteenth and was completed in the nineteenth century. At the same time, the verb ‘quote’ gained in frequency and importance in emergent text-based and literate societies.

22   

   Wolfram Bublitz

Many Old English homilies are based on Latin sources and display visible traces of these in the form of quotations or translations. As a case in point, Winfried Rudolf, in his paper Some Notes on the Latin Quotations and Their Old English Translations in the Homilies of the Vercelli Book, attends to the Latin quotations and translations in the Vercelli Homilies which belong to the oldest Anglo-Saxon preaching texts in England. The study of the bilingual passages composed in Latin and Old English sheds new light on the origins of vernacular preaching in England. The compositional process becomes clearer through the observation that incorporated Latin quotations may deviate notably in their Old English translation or are, indeed, not translated at all, as in this extract from a homily whose writer probably believed a translation would have been redundant: Ond þeah ure hwylc wið oðerne gegylte on worde oððe on worce, forbere he him þæt liðelice, þe læs him God þæt yrre witnige. Swa sylfa cwæð: Dim[it]te et dimit[t]etur uobis. (II:79–81) (And still whosoever becomes guilty against another in word or deed, let him gently refrain, lest God punish him with wrath for that. As he himself says: Dim[it]te et dimit[t]etur uobis. (Forgive and you will be forgiven.)) (Rudolf, this volume)

From both a diachronic and a crosslinguistic perspective, Monika Kirner-Ludwig and Iris Zimmermann, in Quoting and Plagiarising – Concepts of Both Now and Then?, probe into the question of whether there was a medieval consciousness of originality and plagiarism. With a focus on ‘quoting’, ‘citing’ and ‘referring’, they study Anglo-Latin and Middle English as well as Middle High German texts like this couplet from a political song (by Walther von der Vogelweide): Ahî, wie kristenlîche nû der bâbest lachet, swanne er sînen Walhen seit: ‘ich hânz alsô gemachet!’ (Hah, the pope laughs like a Christian whenever he tells his Romans: ‘I did it that way!’) (Kirner-Ludwig/Zimmermann, this volume)

Their close analysis conclusively suggests that medieval authors and writers were undoubtedly interested in legitimizing their work and placing themselves in a certain tradition, to which end they regularly paid their respect to distinguished authorities by referring to and quoting their work. The concepts of plagiarizing and quoting were, accordingly, not unfamiliar at all to medieval text producers. In their paper In-Between Cognitively Isolated Quotes and References: Looking for Answers Lurking in Textual Margins, Jenny Arendholz and Monika KirnerLudwig put present-day online message board entries and medieval manuscripts side by side as some of them share one essential feature: references and quotes that can be found both in the margins of message boards as parts of signatures

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   23

and in the margins of medieval manuscripts in the form of annotations. Here is an example of the former: If you strike me down, I shall become more powerful than you could possibly imagine. (Arendholz/Kirner-Ludwig, this volume)

As there are no overt, conventional indicators of a quote, the interpretation of such utterances rests solely on their positioning in the textual margins (as part of an entry’s signature or as a reflection of one of its expressions next to a medieval text) and the knowledge of the recipient. But, as the authors discover, even the presence of such indicators does not guarantee the recognition of a text as a quote should the interpreter lack the necessary knowledge. Spotting and interpreting an utterance as a quote is not at all self-evident in all cases and does, occasionally, require investigative ingenuity. Since inverted commas were not consistently used to mark the beginning and end of a quote in Early Modern English, Ursula Lutzky explores the linguistic means used instead in the preface position of direct speech quotations in Quotations in Early Modern English Witness Depositions. She also closely examines the role of pragmatic markers that appear in the preface position. The following example features the lexical items ‘what’, ‘I’ and ‘lame’, which occupy the preface position and mark the respective quotes’ beginnings: … presently there appeared to this Examinate the Blacke-Dogge, which appeared vnto her as before: which Black Dogge spake vnto this Examinate in English, saying;$] What wouldst thou haue me to do vnto yonder man? [$to whom this Examinate said,$] What canst thou do at him? [$and the Dogge answered againe,$] I can lame him: [$whereupon this Examinat answered, and said to the said Black Dogge,$] Lame him: [$and before the Pedler was gone fortie Roddes further, he fell downe Lame: (Lutzky, this volume; Lutzky uses square brackets and dollar signs to mark off non-speech from direct speech quotations)

In Haunting Evidence: Quoting the Prisoner in 19th Century Old Bailey Trial Discourse. The Defences of Cooper (1842) and McNaughten (1843), Alison Johnson’s choice of two discourses taken from the corpus of Old Bailey trials from Victorian London is particularly fortunate because of their wealth of direct and indirect quotations and the scribal decisions of how to represent them. The original questionings of witnesses are extensively quoted, requoted, and recontextualized in court. Quoting is used as a strategy (by the prosecution) to expose and support the guilt of the accused or (by the defence) to reveal his or her inability to commit a criminal offence. In the following example the indirect quote serves to support the defense council’s reasoning:

24   

   Wolfram Bublitz

Q. Try and recollect whether the stick Mott had was not broken over a paling? A. I do not know – … I heard the prisoner say he did not mean to shoot any one, but the first who came near him should have it (Johnson, this volume)

In the closing article Quotations from 17th and 18th Century Medical Case Reports, Bettina Lindner, switching to a different setting, investigates the use and development of quotations (their forms, functions and authors) in Early Modern German scientific medical literature, of which they are a characteristic and often defining feature. In the following example, the writing scientists directly quote Johann Bohn, an authority in his field, and mark the change in text and context by the fairly modern conventions of a colon, indentation and change of typeface: Bohnius de off. med. dupl. Part.I. Cap. 14. de Purgatione. hiervon ſaget, da dieſer das letʒtere Tempus matutinum ad ordinarium modum propinandi emetica rechnet; allwo er ſaget: Ordinarie Jejunis vomitoria exhibentur, illis tamen qui difficulter / cum ſingulari moleſtia hinc moventur, poſt paſtum paulo propinari valent, pro evacuatione facilitanda. (Lindner, this volume)

References Anton, Corey. 1998. “‘About talk’: The category of talk-reflexive words”. Semiotica 121, 193–212. Bakhtin, Michail M. 1981. The Dialogic Imagination. Austin: University of Texas Press. Bednarek, Monika and Wolfram Bublitz. 2006. “Reported speech: pragmatic aspects”, in: Keith Brown (ed.). Encyclopedia of Language and Linguistics. Vol. 10 (Art, ID 0369). 2nd ed. Oxford: Elsevier, 550–553. Bell, Allan. 1991. The Language of News Media. Oxford: Blackwell. Brendel, Elke, Jörg Meibauer and Markus Steinbach. 2011. “Exploring the meaning of quotation”, in: Brendel/Meibauer/Steinbach (eds), 1–33. Brendel, Elke, Jörg Meibauer and Markus Steinbach (eds). 2007. Zitat und Bedeutung. Linguistische Berichte, Sonderheft 15. Hamburg: Buske. Brendel, Elke, Jörg Meibauer and Markus Steinbach (eds). 2011. Understanding Quotation. Berlin: De Gruyter. Bublitz, Wolfram and Christian Hoffmann. 2011. “‘Three Men Using our Toilet all Day Without Flushing – This May Be One of the Worst Sentences I’ve Ever Read’: Quoting in CMC”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 433–447. Bublitz, Wolfram and Axel Hübler (eds). 2007. Metapragmatics in Use. Amsterdam: John Benjamins, Buchstaller, Isabelle and Ingrid van Alphen (eds). 2012. Quotatives. Cross-linguistic and Crossdisciplinary Perspectives. Amsterdam: John Benjamins. Cappelen, Herman and Ernie Lepore. 1997. “Varieties of quotation”. Mind 106, 429–450.

Introducing Quoting as a Ubiquitous Meta-communicative Act   

   25

Clark, Herbert H. and Richard J. Gerrig. 1990. “Quotations as demonstrations”. Language 66, 764–805. Clift, Rebecca and Elizabeth Holt (eds). 2007. Reporting Talk - Reported Speech in Interaction. Cambridge: Cambridge University Press. Clift, Rebecca and Elizabeth Holt. 2007. “Introduction”, in: Clift/Holt (eds), 1–15. Coulmas, Florian (ed.). 1986. Direct and Indirect Speech. Berlin: de Gruyter. Coulthard, Malcolm and Alison Johnson. 2007. An Introduction to Forensic Linguistics. Language in Evidence. New York: Routledge. Davidson, Donald. 1979. “Quotation”, in: Davidson, Donald. Inquiries Into Truth and Interpretation. Oxford: Oxford University Press, 79–92. [Originally published in Theory and Decision 11 (1979), 27–40.] Eisenlauer, Volker. 2013. A Critical Hypertext Analysis of Social Media. New York: Continuum. Eisenlauer, Volker and Christian Hoffmann. 2008. “The metapragmatics of remediated text design”. Information Design Journal 16/1, 1–18. Goffman, Erving. 1981. Forms of Talk. Philadelphia: University of Pennsylvania Press. Goffman, Erving. 1967. Interaction Ritual. New York: Pantheon. Golato, Andrea. 2002. “Self-quotation in German. Reporting on past decisions”, in: Güldemann and Roncador (eds), 49–70. Güldemann, Tom. 2008 Quotative Indexes in African Languages: A Synchronic and Diachronic Survey. Berlin: De Gruyter. Güldemann, Tom and Manfred von Roncador (eds). 2002. Reported Discourse. A Meeting Ground for Different Linguistic Domains. Amsterdam: John Benjamins. Günthner, Susanne. 1999. “Polyphony and the ‘layering of voices’ in reported dialogues: an analysis of the use of prosodic devices in everyday reported speech”. Journal of Pragmatics 31, 685–708. Gutzmann, Daniel and Erik Stei. 2011. “How quotation marks what people do with words”. Journal of Pragmatics 43, 2650–2663. Herring, Susan C. 1999. “Interactional coherence in CMC”. Journal of Computer-Mediated Communication 4/4. http://jcmc.indiana.edu/vol4/issue4/herring.html, last accessed 07 Feb 2014. Hoffmann, Christian. 2012. Cohesive Profiling. Meaning and Interaction in Personal Weblogs. Amsterdam: John Benjamins Holt, Elizabeth. 2000. “Reporting and reacting: concurrent responses to reported speech”. Research on Language and Social Interaction 33, 425–454. Hübler, Axel and Wolfram Bublitz. 2007. “Introducing metapragmatics in use”, in: Bublitz/ Hübler (eds), 1–26. Hübler, Axel. 1989. Citations in Academic Writing: A Text Linguistic Approach. Volume 21. Series C. Duisburg: LAUD (Linguistic Agency University of Duisburg). Jacob, Joachim and Mathias Mayer (eds). 2010. Im Namen des Anderen. Die Ethik des Zitierens. München: Wilhelm Fink. Johnson, Michael and Ernie Lepore. 2011. “Misrepresenting misrepresentation”, in: Brendel/ Meibauer/Steinbach (eds), 231–248. Jones, Graham M. and Bambi B. Schieffelin. 2009. “Enquoting voices, accomplishing talk: uses of be + like in Instant Messaging”. Language & Communication 29, 77–113. Klockow, Reinhard. 1980. Linguistik der Gänsefüßchen. Untersuchungen zum Gebrauch der Anführungszeichen im gegenwärtigen Deutsch. Frankfurt am Main: Haag und Herchen.

26   

   Wolfram Bublitz

Levinson, Stephen C. 2000. Presumptive Meanings. The Theory of Generalized Conversational Implicature. Cambridge, Massachusetts: MIT Press. Lucy, John A. (ed.). 1993. Reflexive Language. Reported Speech and Metapragmatics. Cambridge: Cambridge University Press. Mayes, Patricia D. 1990. “Quotation in English”. Studies in Language 14, 325–363. Moore, Colette. 2011. Quoting Speech in Early English. Cambridge: Cambridge University Press. Pennesi, Karen. 2007. “Be saying quotations as demonstrations of stance: a linguistic approach to environmental conflict”. Arizona Anthropologist 18, 1–27. http://anthropology.uwo.ca/ faculty/pennesi/AZ-Anthropologist.pdf, last accessed 7 Feb 2014. Recanati, François. 2001. “Open quotation”. Mind 110, 637–687. Roncador, Manfred von. 1988. Zwischen Direkter und Indirekter Rede. Nichtwörtliche Direkte Rede, Erlebte Rede, Logophorische Konstruktionen und Verwandtes. Tübingen: Niemeyer. Saka, Paul. 1998. “Quotation and the use-mention-distinction”. Mind 107, 113–135. Sams, Jessie. 2007. “Quoting the unspoken: an analysis of quotations in spoken discourse”. Colorado Research in Linguistics. June 2007, Vol. 20. Boulder: University of Colorado. http://www.colorado.edu/ling/CRIL/Volume20_Issue1/paper_SAMS.pdf, last accessed 7 Feb 2014. [again: 2010. Journal of Pragmatics 42, 3147–3160.] Sbisà, Marina and Ken Turner. 2013. “Introduction”, in: Marina Sbisá and Ken Tuner (eds). Pragmatics of Speech Actions. Handbooks of Pragmatics. Vol. 2. Berlin: De Gruyter Mouton, 1–21. Severinson Eklundh, Kerstin. 2010. “To quote or not to quote: setting the context for computermediated dialogues”. Language@Internet 7/5. http://www.languageatinternet.org/ articles/2010/2665/Severinson_Eklundh.pdf, last accessed 18 Mar 2013. Tannen, Deborah. 1986. “Introducing constructed dialogue in Greek and American conversational and literary narrative”, in: Coulmas (ed.), 311–332. Tulving, Endel. 1983. Elements of Episodic Memory. Oxford: Oxford University Press. Wade, Elizabeth and Herbert H. Clark. 1993. “Reproduction and demonstration in quotations”. Journal of Memory and Language 32, 805–819.

Part I: Quoting Now

Daniela Landert

1 Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News¹ Abstract: Reporting statements from news sources is an essential part of news reports. Such statements are often transformed into narrative reports and indirect quotes, but they can also take the form of direct speech quotes. Direct quotes have always been used in news writing with a variety of different functions, such as providing colour to a news report and distancing the journalist from what is said. However, in recent data from online news sites these functions show some differences compared to earlier data from print newspapers. The aim of this study is to investigate these shifts and to explore the extent to which they depend on the new technological setting online. Looking at data from the Times Online from 2010 and comparing it to data from the printed Times from 1985 and 2010, I identify two main shifts in the function of quotes. On the one hand, the integration of source material from which the quotes are taken reinforces one of the functions, namely presenting reportable facts. On the other hand, the function of expressing personal experience and emotion is much more prominent than in the print data from 1985. I argue that the first of these shifts is directly related to the different settings of print and online news, whereas the second shift has to be seen in the context of a more general trend towards more personalised news reporting. Keywords: news language, direct speech, pragmatic functions, Internet, emotions

1 Introduction Online news sites have introduced various innovations compared to print newspapers. They integrate text with video reports and image galleries, they update breaking news stories continuously without having to depend on regular print 1 I would like to thank the audience at the panel “The pragmatics of quoting in computermediated communication” at the 12th IPrA Conference 2011 in Manchester, Andreas H. Jucker, Elizabeth C. Traugott, an anonymous reviewer, as well as the editors of this volume for their valuable suggestions and comments on earlier versions of this paper. I would also like to thank Shane Walshe for his help with proofreading the article. It goes without saying that any remaining shortcomings are mine alone.

30   

   Daniela Landert

cycles, and they let readers comment on news articles and interact with each other. In many respects, though, they can be said to follow the model of print newspapers; in particular, they have adopted the linguistic conventions of news writing. Reporting statements from news actors and official sources is an essential part of any news report and, consequently, speech reporting has been a characteristic of news writing since the beginning of newspapers.² Therefore, one might assume that the use of speech representation more generally, and direct quotation in particular, does not show any interesting developments in online news. However, if we compare a typical instance of a direct quote from the printed Times from 1985 (Example 1) with a quote from the Times Online (Example 2) we can see that they do not necessarily fulfil the same functions.³ (1)

Mrs Thatcher said that she wanted a settlement, but the board had to manage and had to close uneconomic pits. “It has to make the decision: that must be clear from the outset.” (tp-uk-850125)

(2)

With Africa’s football championship set to open as planned on Sunday, other teams spoke of their fear that they, too, might become the target of attacks. “We have goose bumps … who knows what is going to happen to us,” said Amade Chababe, Mozambique assistant coach. (to-wn-100110, ellipsis in original)

The main function of the quote in Example (1) is to present Thatcher’s statement as a reportable fact. This is very typical of quotes in the printed Times from 1985. The sources of most quotes in these data come from official actors, such as politicians and spokespersons of interest groups, who present positions and interpretations of current affairs (see also Landert 2014: 194). Statements that represent personal experience and emotion are hardly ever found in the Times of 1985. However, this is an important function of quotes in the Times Online, as illustrated in Example (2). The quote appears in a report of a terror attack on the Togolese football team at the 2010 Africa Cup of Nations. In this case, the quote is not newsworthy for announcing a political decision or an official stance, but rather 2 The term “news actor” is used in this paper to refer to individuals who appear in news reports, for instance as sources of quotes. Bell (1991: 191) uses the term to refer to individuals “whose […] utterances have news value”. See also Gans (1979: 8), Jucker (1996). 3 The examples quoted in this paper identify the article from which they come in the format SOURCE-SECTION-DATE. Source is either “tp” for a printed issue of the Times or “to” for an online article. The section is “wn” for world news articles and “uk” for UK news. And the date of the issue from which the article is collected is given in the format YYMMDD. For more information, see section 4.

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   31

for presenting the personal perspective of someone who was affected by a news event. The quote in Example (2) also differs from quotes in the Times from 1985 in another respect, namely that further evidence for its factual status can easily be found online. Not only does the same quote appear in various online newspapers, but there are also videos in which one can watch Chababe being interviewed.⁴ There are various factors that play a role in the differences between the functions of quotes in the printed Times from 1985 and the Times Online from 2010. On the one hand, technological innovations in the online setting can facilitate changes. On the other hand, news media are affected by more general diachronic changes, such as tendencies towards popularisation and increased immediacy in news reporting. Disentangling these two aspects is not trivial, since innovations in online news that originated from technological factors can influence print news (see, for instance, Bucher/Schumacher 2008, Cooke 2005). In order to understand the shifts in the functions of direct quotes, it is important to combine a comparison of different types of data (diachronic and print versus online) with a detailed investigation of how these functions came into effect. The aim of this study is to investigate recent shifts in the functions of direct quotes that can be observed on the Times Online. I will compare data from the 2010 Times Online with data from the printed Times of the same year, as well as with data from the 1985 printed Times. The first shift I identify consists of a new way in which quotes can function as reportable facts, which relies on additional evidence for the statement that is made available online. The second shift concerns an increase in quotes that present personal experience and emotions. I will argue that this second development is not directly related to technological factors, unlike the first development. Instead, it should be seen in the context of personalisation tendencies that affect news media more generally. Before starting my analysis, I will provide a brief overview of previous accounts concerning the functions of quotes in print newspapers in section 2. Following this, I will show that there are systematic relations between these functions and other aspects of quotes. More specifically, I will argue that certain characteristics of the context, content and language of quotes serve as triggers for specific functional interpretations. The model introduced in section 3 will not only facilitate the analysis of quote functions, it will also help identify factors that lead to changes in these functions over time, and differentiate between technological factors and more general tendencies in news media. After introducing my data in section 4, I will proceed with the analysis in two steps. In section 5, I will investigate how the online setting can strengthen one of the functions of direct quotes

4 http://www.youtube.com/watch?v=O6w99EFF8a0, last accessed 29 October 2014.

32   

   Daniela Landert

by emphasising the role of quotes as reportable facts. In section 6, I will then examine the increase in quotes that present personal experience and emotions.

2 Functions of Quotes in Print Newspapers In the past couple of decades, various approaches to the functions of quoting have been developed. On the one hand, general theoretical accounts have discussed, for instance, the functional differences between various forms of quotation, such as direct, indirect and mixed quotation. Examples are Clark/Gerrig (1990), Cappelen/Lepore (1997), and various contributions in the recent volume edited by Brendel/Meibauer/Steinbach (2011). On the other hand, the functions of quoting have been investigated in relation to specific contexts, such as informal conversations (e.g. Mayes 1990, Tannen 1986), courtroom settings in the past (e.g. Collins 2001, Włodarczyk 2007) and computer-mediated communication (Bublitz/Hoffmann 2011). These latter approaches show that the communicative context in which quoting takes place can have an influence on its functions. For instance, several of the functions identified by Bublitz/Hoffmann (2011: 441–446) in their data from (interactive) computer-mediated communication are closely related to the setting, e.g. when semi-automatised quoting is used for “bridging disrupted adjacency” between a post and its reply. In this section, I will focus in particular on approaches that deal with quoting in the context of news reporting (where quotations are usually called “quotes”) and I will introduce the functions that have been suggested for this type of data. Sections 5 and 6 will later investigate two of these functions in more detail, identifying the shifts they undergo in the online setting. A general account of the functions of quotes in news media is provided by Bell (1991: 207–209). While this is not the most recent account, it provides an excellent characterisation of quotes in up-market print newspapers before the widespread use of the Internet. Bell names three main functions. The first of these concerns their ability to present “incontrovertible fact[s]” (1991: 207). Journalists are required to report reliable information and, if necessary, they need to be able to provide proof for their statements. By quoting, the responsibility for the content of the statement is shifted to the source of the quote. Journalists do not need to verify the information in statements; they only need to verify that the statement was made. Given that many of the statements quoted in news media come from press conferences or interviews, where there is usually the possibility to record such statements, this verification is relatively easily done. Thus, quotes are “particularly incontrovertible fact[s]” for journalists, since it is hard to contest that the

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   33

quoted statement was made if the journalist is in possession of a recording (1991: 207–208). Example (3) illustrates one such use of a quote as an incontrovertible fact. It is taken from an article in the printed Times from 1985, in which Mr Shultz, the US Secretary of State, is quoted using the term “equitable agreements”. (3)

Mr Shultz by contrast made no reference to space, and talked only in vague terms of “equitable agreements”, raising the thorny problem of verification. (tp-wn-850108)

It is characteristic in such usage that the quoted source is a public figure with authority. The same statement made by an unknown individual without some official function would have had a low likelihood of being quoted. In contrast, statements made by influential individuals like the Secretary of State are quite likely to become newsworthy facts. The second function mentioned by Bell (1991: 208) is to “distance and disown”. It is comparable to the function Clark/Gerrig call “dissociation of responsibility” (1990: 792). Journalists are required to report in an unbiased and objective manner. Opinions and evaluations are therefore often attributed to sources and the more extreme a statement, the more likely it is to be represented as a direct quote. (4)

But there are also powerful American figures in Geneva who, although not present at the negotiating table, exert influence behind the scenes. They include Mr Richard Perle, Assistant Defence Secretary, who has profound mistrust of the Russians, describing them as a “nation of liars”. (tp-wn850108)

In Example (4), the offensive expression “nation of liars” is presented as a direct quote in order to distance the newspaper from the content of the statement. The direct quote signals a neutral stance on the part of the newspaper while still reporting the statement. The third of Bell’s functions consists in enriching the news article with colourful expressions (1991: 208–209). Such expressions would be inappropriate outside of quotes due to their style. In the same way as newspapers are expected to report content objectively, they are normally expected to use formal language as well. Usually, therefore, colloquial and non-standard language, colourful metaphors and verbal puns are restricted to quotes. In the following example, Arthur Scargill, the president of the National Union of Mineworkers, is metaphorically described as the Godfather.

34   

(5)

   Daniela Landert

But the appeals fell on deaf ears and Mr John Gummer, the chairman of the Conservative Party, returned to the offensive, urging striking pitmen to “refuse to be intimidated by Scargill the Godfather”. (tp-uk-850109)

In this case, colouring the report with an expression relating to a popular film (Francis Ford Coppola’s epic crime movie The Godfather from 1972) is the most salient function of the quote, but other functions are also present. Using a direct quote distances the newspaper from the content of the quote, which associates Scargill with a fictional mafia boss. Moreover, the quoted statement was made by a central actor in the reported news event and, therefore, the quote constitutes a reportable fact. Thus, all three of Bell’s functions are present to a certain extent. There is one function of quotes in news reports which is not discussed by Bell (1991), but which has received attention from other scholars: reporting emotions. Example (6) illustrates this function. The quote appears in a report about the 2010 earthquake in Haiti and it presents the emotions of the sister of a British citizen who was killed. The quote does not add any newsworthy facts to the report  – instead, it repeats information that was already provided, but formulates it in a more personal way. (6)

At least one British citizen is among the missing. Ann Barnes, 59, a United Nations worker originally from Leighon-Sea, Essex, has been unaccounted for since Tuesday. Ms Barnes’s sister, Irene Marquet, said she felt sure her sister was among the casualties. “There’s been absolutely no trace. One wants to remain hopeful but it gets more and more difficult as time goes on,” said Mrs Marquet, a pensioner living in southern France. (tp-wn100116)

Burger (2005: 97, 101) lists integrating the emotions of the source into a news report as one of the central functions of direct quotes. Stenvall (2008: 1572) points out that direct quotes play a particularly important role in the expression of emotion in hard news reports. She explains that the objective style of these texts requires the absence of the text producers’ emotions and concludes that “[s]trictly speaking, the only way to report other people’s individual feelings ‘objectively’ is to resort to direct quotes, that is, to let the person in question describe how s/he feels.” (Stenvall 2008: 1572, emphasis in the original). Several researchers furthermore make a link between direct quotes reporting emotion and personal experience, on the one hand, and the emotional involvement of readers, on the other (Bös 2010a: 239–240; Luginbühl 2012: 258–259; Santulli 2012: 278; Schäfer 2005: 226; Tuomarla 2000: 70–72). Quotes that represent the personal experience and emotions of those directly affected by the news offer readers a more immedi-

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   35

ate experience of news events.⁵ They provide a personal touch to otherwise distanced reporting and can be used to dramatise news reports. Such dramatisation of news reports with quotes containing emotion or personal experience has been described for French (Schäfer 2005: 226; Tuomarla 2000: 70–72), Italian (Santulli 2012: 278), German (Schäfer 2005: 226) and American (Luginbühl 2012: 258–259) newspapers. Bös (2010b: 77) analysed the occurrence of personal quotes in British newspapers from a historical perspective and found an increase between 1700 and 2000. She further found that down-market newspapers took the lead in this development and that it is only towards the end of the 20th century that personal quotes began to occur in up-market newspapers like the Times (2010b: 84–85; see also Schneider 2002: 157). This might also explain why Bell (1991) does not include the reporting of emotion and personal experience in his discussion of the functions of direct quotes in news reports.

3 The Function of Quotes in Relation to Their Context, Content and Language In order to analyse shifts in the functions of quotes it is important that these functions be clearly identifiable. As illustrated above, several functions of quotes are often present simultaneously, which is one of the reasons why it is sometimes difficult to pinpoint the exact function of a specific quote. Another reason is the fact that functions generally are more difficult to identify than formal characteristics. They depend on the interpretation of textual effects, which is to some extent subjective. This poses a problem for empirical studies that aim at comparing the function of quotes in different sets of data. One way to approach this problem is to analyse factors that facilitate or even trigger a specific functional reading. As I am going to argue in this section, such factors can be identified for all functions and they can be assigned to three different levels: the context of the quote, its content and its language (see Table 1).

5 Creating immediacy is not restricted to quotes reporting emotions. Bublitz/Hoffmann (2011: 441–442), for instance, point out that the use of direct speech, indirect speech and free indirect speech allow authors to shift between different levels of narrative immediacy. In the case of quotes containing personal experiences and emotions, this effect is particularly strong, however, since the content of the quote invites readers to become emotionally involved.

36   

   Daniela Landert

Table 1: Factors inviting specific functional interpretations of quotes

Context

Content

Language

Reportable authoritative source; facts evidence available / traceable Distance and disown

not objective, opinionated, potentially offensive

Personal touch

personal experience and emotion

Colour

(first person reference)

violating stylistic requirements of journalistic reporting: nonstandard, colloquial, colourful metaphors

The context of a quote is related to the function of quotes as reportable facts. As mentioned above, in order for a quote to be a newsworthy reportable fact, its source needs to have a certain authority. Generally speaking, the more authoritative a source, the more likely it is that a quote will be perceived as a reportable fact. The status of the source is thus an important indicator for the first of Bell’s three functions of quotes. There are other aspects of the context which can also relate to this function. For example, in order for a quote to be perceived as a fact, it needs to be highly credible that the quote is faithful. Generally, direct quotes evoke faithfulness claims that include the speech act value of the quote, its propositional content, and its wording (see Jucker 2006: 109). However, Short/Semino/ Wynne (2002) point out that contextual factors can weaken or even cancel these faithfulness claims.⁶ For instance, if the anterior discourse in which the quote originally occurred is not accessible, there is less expectation on behalf of the addressee that the quote is faithful. This is the case in spoken conversation and storytelling, for which it has been shown that quotation is often inaccurate (see Mayes 1990, Tannen 1986). In contrast, quotes that function as reportable facts need to evoke strong faithfulness claims. Readers must believe it credible that the journalist had access to the anterior discourse and in most cases journalists are expected to be in possession of a record documenting this anterior discourse, either in the form

6 See also Clark/Gerrig (1990: 795–800), who provide a detailed discussion of the various problems of the “verbatim assumption”, i.e. the view that direct speech reproduces previous statements verbatim.

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   37

of notes or, even better, as a sound or video recording. As I will further argue in section 5, the presentation of such records along with the article in which a quote appears is one of the factors that emphasise the factual function of quotes. In other words, whenever evidence of the anterior discourse is made available, the factual function is reinforced. The content of a quote plays an important role in the distancing function of quotes. Whenever news articles include content that obviously violates the journalistic norms of objectivity, there is a high likelihood that it is not only explicitly attributed to a source, but that it is also presented as a direct quote. In general terms, quotes which contain such openly non-objective statements invite a reading that disassociates the newspaper from the reported position. Quotes that are likely to be read in this way include, for instance, those which express strong opinions or which are potentially offensive. Quotes that provide a personal touch by expressing emotion and personal experience can also be included here. Like opinions and offensive statements, emotion and personal experience violate the journalistic norms of objectivity by privileging the view of an individual over general and abstract processes, and by providing an affective rather than a factual account. In this respect, quotes that report emotion and personal experience can be seen as a special case of the more general distancing function. One difference between these quotes and other distancing ones concerns their linguistic realisation. Given that emotion and personal experience are generally expressed in the first person, linguistic features like first person pronouns can help identify such quotes. This does not mean that first person pronouns are necessarily reliable indicators for quotes expressing emotion and personal experience. They can also occur in contexts in which the source states an opinion, takes over responsibility, expresses commitment or gives advice. Moreover, first person plural pronouns have a variety of different possible referents, including vague and generic uses that do not refer to the speaker in particular (see Kitagawa/Lehrer 1990, Wales 1996: 58–68). I will discuss the various functions of quotes containing first person singular pronouns in more detail in section 6. Finally, the language of quotes is clearly responsible for their function of enriching news reports with colourful expressions. As discussed in section 2.1 above, this includes all expressions that would violate the stylistic requirements of formal news reporting if they appeared outside of quotes. The fact that different functions of quotes are triggered by different factors not only helps identify functions; it can also explain why some functional shifts are more closely related to technological innovations than others. As I will further argue in section 5, presenting quotes as reportable facts has taken on new facets in the online setting. Given that this function is triggered by contextual factors, it

38   

   Daniela Landert

is not surprising that it is affected more directly by the technological setting than other functions.

4 Data Most of the data on which this analysis is based were collected for a larger research project on personalisation in online news (Landert 2014). In the present study, I only focus on a subset of these, published on Times’ website, which is most suitable for a diachronic comparison with earlier data from the printed Times, available via the Times Digital Archive. The online data were collected in January 2010 from the news site Times Online (www.timesonline.co.uk).⁷ They consist of 28 news articles selected on the 14 even days between 4 January and 30 January. On each collection day, the top-listed articles from the sections “world news” and “UK news” were selected, resulting in a corpus of articles that can be described as typical front page news, i.e. news reports on current events with very high news value. The data from the printed Times from 1985 were selected to match the online data as closely as possible. They were taken from the 18 issues of the Times published between 7 and 26 January 1985 and include for each day the most prominently placed front page article on a world news topic and the most prominently placed article on a UK news topic. A larger number of articles was collected to compensate for the shorter average length of these earlier articles, compared to the online data. In order to be able to distinguish between purely diachronic changes in print newspapers, on the one hand, and changes related to the difference between print and online setting, on the other, the two sets of data were complemented with data from the printed version of the Times from January 2010. The articles for this data set come from the issues published on the days on which the online data were collected and, again, include two articles from each issue. Due to the fact that the print version from 2010 does not always contain a UK news topic and a world news topic on its front page, some of the articles come from subsequent pages. In one case, no suitable article with an international focus was found in the first section of the newspaper, so that the world news article was taken from page 29. Table 2 gives an overview of the number of articles and number of words in each part of the corpus.

7 In the meantime, the online site of the Times has been redesigned. It was re-launched in the summer of 2010 and the old domain www.timesonline.co.uk was replaced by two separate domains, www.thetimes.co.uk and www.thesundaytimes.co.uk.

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   39

Table 2: Overview of collected data in number of articles and number of words

Articles

Words

Times 1985 Times 2010 Times Online 2010

36 28 28

23,188 19,004 22,443

Total

92

64,635

All articles were copied or typed into an XML file and annotated for quotes and first person singular pronouns.

5 Quotes as Reportable Facts in Online News As mentioned before, the first shift that I investigated concerns the function of quotes as reportable facts, which is closely related to technological innovations on online news sites. In contrast to print newspapers, online news sites can not only quote source material, but they can also embed it into their articles. Videos and transcripts of press statements can be added to news articles as additional material and hyperlinks can be used to direct readers to external websites where source material is officially released. In this section, I will show with the help of some examples how the availability of such material affects the function of quotes as reportable facts. One of the most common topic areas for which source material is provided is politics. Important press conferences and interviews are often recorded and integrated as videos, and central documents are sometimes provided in addition. In January 2010, for example, the Times Online reported on US president Obama’s first State of the Union address. The textual report was accompanied by a video, integrated between the headline and the by-line. The video consisted of a brief extract of the speech, put together from various segments. The video extract covered two of the quotes occurring in the textual report, reproduced in Examples (7) and (8). In the second case, only the end of the quote, the emphasised part, was included in the video. (7)

No detailed strategy was expected for saving the healthcare Bill that dominated Mr Obama’s first year in office, and none was offered. Mr Obama restricted himself to a plea “to come together and finish the job for the American people”. (to-wn-100128)

40   

(8)

   Daniela Landert

“We have finished a difficult year,” he said. “We have come through a difficult decade. But a new year has come. A new decade stretches before us. We don’t quit. I don’t quit. Let’s seize this moment to start anew.” (to-wn100128, my emphasis)

The video showing Obama making these statements provides evidence that the quotes in the textual report are accurate and it reinforces their status as facts. Readers can see for themselves that Obama gave a speech and that he made the quoted statements. In the case of this article, additional material is provided through a link named “Full text: Obama’s State of the Union Address”. This leads to a separate page on the Times Online website which contains the transcript of the entire speech. The subtitle of the page indicates that this transcript was issued by the White House and a quick web search leads to the White House website, which not only provides the full transcript but also a video of the entire speech that can be downloaded or viewed online.⁸ The online setting thus allows journalists to make their source material available to readers, and it allows readers to trace statements back to their original context and check their accuracy. Irrespective of whether or not readers actually do this, the fact that they are able to do so already influences the effect of quotes. While all direct quotes claim to accurately represent a statement from an anterior discourse, the availability or reproduction of this anterior discourse reinforces the claim (see also Short/Semino/Wynne 2002: 349–350). In other words, the mere possibility of verifying a quote emphasises its factual function. The emphasis is particularly strong in the case of material located on official external websites and for videos, which provide the information through visual and auditory channels, allowing readers to see and hear for themselves. Videos and transcripts of public speeches are not the only way in which source material is presented with news reports. Of particular interest are cases in which documents are centrally involved in causing news events, so that reports of the events are likely to quote these documents. A news story which illustrates this point concerns a conflict within Britain’s Labour Party in January 2010. Two MPs, Geoff Hoon and Patricia Hewitt, wrote to all Labour MPs asking for a secret ballot about the degree of support for Prime Minister Gordon Brown as a candidate for the general election. Like most British online news sites, the Times Online reported prominently on the message and its consequences. The report quoted several paragraphs from the letter and the beginning of this quote is reproduced in Example (9). 8 http://www.whitehouse.gov/photos-and-video/video/2010-state-union-address, last accessed 8 October 2012.

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

(9)

   41

In their letter, Mr Hoon and Mrs Hewitt said the party was “deeply divided” over the leadership. “The continued speculation and uncertainty is allowing our opponents to portray us as dispirited and disunited,” they said. (to-uk-100106)

The quote can be compared to the text of the entire letter, which is made available as a PDF file via a link in the left-hand sidebar. Unlike the White House press release discussed above, this letter cannot be traced back to an official release by an independent website. Moreover, the PDF file looks like a screenshot from an e-mail without an official letterhead. In this respect, the letter is rather weak evidence for the factual status of the quotes in the article. It provides additional context, but otherwise the document is no more credible than the quote itself. However, there is one factor that still contributes to the perception of the letter as a credible record, namely the (implied) knowledge that other reproductions of the letter are available online. Indeed, news articles about the story on the online sites of the Guardian, BBC News and the Sun all reproduce the full letter, either in a sidebar, on a separate page or integrated into the news report. With the help of search engines, the respective articles are quickly identified. Knowing that the content of this crucial document can easily be compared across different platforms increases the credibility of the letter’s reproduction and, by extension, of quotes from it in the news reports. The importance of cross-platform references can also be observed in other quotes from the same news article. Several of the statements attributed to the two main news actors, Hoon and Hewitt, are said to come from interviews with the BBC, rather than with the Times’ own journalists. This is particularly noteworthy since the article is accompanied by a SkyNews video in which very similar statements are made. Example (10) reproduces the quote from the news report and Example (11) provides a transcript for a section of the video embedded in the article. (10)

Mr Hoon told the BBC he had not discussed the move with anyone other than Mrs Hewitt. “We both came back from the Christmas vacation, having independently had these kinds of conversations, both independently deciding that it was time that we issued this letter. […]” (to-uk-100106)

(11)

Patricia Hewitt and I quite separately over the Christmas vacation came to the view that it was necessary to address this matter. (transcript from the video embedded in to-uk-100106)

42   

   Daniela Landert

There are several possibilities as to why the article may be quoting Hoon from an interview with the BBC rather than from an interview with an associated news channel reporter.⁹ A likely explanation is that at the time of writing the article, only the interview with the BBC was available to the journalist. However, given the ease of accessing online content, it is not difficult for a reader to locate the video in which Hoon makes the statement quoted in Example (10). It is embedded in an article about the same news story published on BBC News.¹⁰ These examples show that the Internet affects the context and function of quotes by facilitating access to the anterior text. It has become quite common for news articles to embed and link to source material that provides further evidence for quoted statements and thereby emphasises their status as reportable facts. To some extent, this is not only true for online news but also for print newspapers. Print newspapers cannot contain embedded videos and linking does not provide access to external material as directly as on online sites. Still, readers of print newspapers are frequently directed to online sites on which additional material can be found. The article about Obama’s State of the Union address in the printed version of the Times ends with such an online reference, reproduced in Example (12). (12)

The State of the Union speech in full. Video, pictures and analysis at www.timesonline.co.uk (The Times, 28 January 2010, p. 39)

Even though access to the speech is less direct in this case, pointing readers towards the full transcript of the speech affects the status of the quotes in this article in a very similar way as the link to the transcript affects the quotes of the online article. While clearly tied to technological developments and media changes from print to online, the effects of this change in the function of quotes are thus not restricted to online publications.

9 The Times and SkyNews are both part of Rupert Murdoch’s News Corporation (see www.newscorp.com, last accessed on 8 October 2012). SkyNews videos are often used to illustrate news articles on Times Online. 10 http://news.bbc.co.uk/2/hi/uk_news/politics/8443769.stm, last accessed 16 January 2013.

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   43

6 Quotes Presenting Personal Experience and Emotion Another shift in the function of quotes can be observed in recent online news data, namely an increase in the presentation of personal experience and emotion. As explained in section 3, this is a special case of the distancing function of quotes, whereby newspapers dissociate themselves from the views expressed in the quote. For the purpose of this study, I investigated quotes with first person reference in the data from the 2010 Times Online and compared the results to the data from the 1985 printed Times. In addition, the comparison with data from the 2010 printed Times will show whether the observed change is restricted to the online setting or whether it also occurs in the print version. Before presenting the results for quotes containing first person reference, it is interesting to look at some more general results for the frequency of quotes in these three data sets. As can be seen from Tables 3 and 4, quotes occur more frequently and take up a considerably larger proportion of the texts overall in the more recent data compared to the data from 1985.¹¹ Moreover, quotes are slightly longer in the data from 2010 (see Table 5), even though this difference is not statistically significant. The voices of sources thus generally have a higher presence in the news articles from 2010 than in those from 1985, and this is true for both the print and the online version of the Times. Previous research has shown that quotes increased in the Times between 1833 and 1988 (Jucker/Berger 2014), and in British (and American) newspapers, more generally, between the 1960s and the 1990s (Mair/Hundt 1997: 75–76). The data in Tables 3 and 4 indicate that this trend might still be ongoing. Table 3: Number of words in quotes in percent of total number of words

Times 1985

Times 2010

Times Online 2010

Words total

23,188

19,004

22,443

Words in quotes

 3,220

 4,466

 5,179

In %

13.9 %

23.5 %

23.1 %

11 For both Tables 3 and 4, the difference between the printed Times from 1985 and the Times Online from 2010 is statistically significant at p < 0.001. The difference between the printed issues from 1985 and 2010 is statistically significant at p < 0.001. The difference between the printed version and the online version from 2010 is not statistically significant.

44   

   Daniela Landert

Table 4: Frequency of quotes Times 1985

Times 2010

Times Online 2010

Words total

23,188

19,004

22,443

Number of quotes

   213

   255

   305

Frequency per 1,000 words

   9.2

  13.4

  13.6

Times 1985

Times 2010

Times Online 2010

Words in quotes

3,220

4,466

5,179

Number of quotes

  213

  255

  305

Average length

 15.1

 17.5

 17.0

Table 5: Average length of quotes

In order to investigate the presence of personal viewpoints, all forms of the first person singular pronoun were tagged, including the possessive determiner my. Table 6 shows the percentage of quotes containing any first person singular pronoun in each of the three datasets. Only 7 % of all quotes have first person singular pronouns in the data from the printed Times from 1985. In contrast, 15 % of all quotes contain first person singular pronouns in the Times Online from 2010. This higher frequency is not due to the online setting, as a comparison with the printed Times from 2010 shows. Here, the percentage of quotes with first person singular pronouns is even higher.¹² It is possible that this difference between the printed and the online version of the 2010 Times is related to the slightly shorter length of quotes in the online version (see Table 5), which in turn could be a consequence of different typographic conventions of marking quotes in print and online. If longer quotes are split up into two quotes, which often happens when they span more than one paragraph, this could lead to a lower percentage of quotes containing first person singular pronouns. Quotes with first person singular pronouns often contain more than one pronoun. In addition, such quotes tend to occur in clusters, especially in the more recent data. 11 of the 28 articles in the online data do not contain any quotes with first person singular pronouns. In contrast, the article on the Labour Party conflict discussed above contains four quotes with a total of six first person sin12 The difference between the printed Times from 1985 and the printed Times from 2010 is significant at p < 0.001. The differences between the other pairs of publications are not statistically significant.

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   45

Table 6: Percentage of quotes containing any first person singular pronouns

Times 1985

Times 2010

Times Online 2010

  213

   255

   305

Containing 1  pers. sg. pronouns

   15

    51

    46

in %

7.0 %

20.0 %

15.1 %

Number of quotes st

gular pronouns. These are all quotes in which politicians specify their role in and attitude towards the call for a secret ballot (see Example 13). One article in the printed version of the Times from 2010 even contains 11 quotes with a total of 17 first person pronouns. This is an article on an earthquake in Haiti and several victims and witnesses report their own experience (see Example 14). (13)

“I have received the letter and I support their call.” (to-uk-100106)

(14)

“It’s tearing at my leg. I can’t wait long.” (tp-wn-100120)

This suggests that the likelihood of articles containing quotes with first person singular pronouns varies depending on topic and that differences in the use of first person singular pronouns in quotes need to be interpreted with care, especially when the analysis is based on a relatively small number of complete articles. Nevertheless, the earlier data show a lower proportion of articles containing quotes with first person singular pronouns than the more recent data. Only 10 of the 36 articles from the Times from 1985 contain quotes with first person singular pronouns, compared to more than 60 % for the data from 2010 (see Table 7).¹³ Table 7: Percentage of articles containing quotes with first person singular pronouns

Times 1985

Times 2010

Times Online 2010

    36

    28

    28

Articles with 1  pers. sg. quotes

    10

    19

    17

in %

27.8 %

67.9 %

60.7 %

Number of articles st

13 This difference is not statistically significant, though.

46   

   Daniela Landert

Frequency is not the only difference between the earlier and the more recent data. A detailed analysis of the function of all quotes with first person singular pronouns shows that some uses of quotes are restricted to the data from 2010. I will now discuss the functions of first person quotes, starting with those also found in the earlier data and then proceeding to those that appear to be innovations. The most common function for all sets of data consists in expressing the opinion of the source of the quote. Often this is realised with verbs expressing mental processes like think and believe or with verbs of communication like say. (15)

Mr Peter Rogers, head of Manufacturers’ Hanover Trust’s foreign exchange department in New York said: “The British Government has got a breathing space rather than a permanent turnaround for the pound. I can’t believe, the Government would allow it, but the market here believes parity can happen.” (tp-uk-850116)

(16)

“I don’t think American elections should be bankrolled by America’s most powerful interests, and worse, by foreign entities,” he [= US president Obama] said. (to-wn-100128)

(17)

“I’ve always said that because of all the uncertainty around we should be very cautious,” the Chancellor said. (to-un-100126)

In addition, quotes sometimes report past, ongoing or future actions carried out by the source in an official capacity. Again, such quotes can be found both in the 1985 printed Times (Example 18) and in the 2010 Times Online (Example 19). (18)

Yesterday Dr Mesfin Dmise, the national co-ordinator [of the World Health Organization], said: “I asked the Ministry of Health to let me know the results of their inspection of the camps last week.” They promised to let him have them today, he said, and “if the tests are positive I will inform the WHO in Geneva. If the result is negative I might well carry out independent laboratory tests”. (tp-wn-850124)

(19)

“I have sought an urgent meeting with the First Minister to discuss the implications for the Office of the First Minister and Deputy First Minister.” (to-uk-100108)

First person quotes expressing emotions occur in two main ways. On the one hand, emotions are used as a rhetorical resource in argumentation. Politicians and other official actors are often quoted in this way and this use can again be

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   47

found in all data sets, for instance in Example (20) from the 1985 printed Times. In this case, the emotion does not express an emotional state of the source, but rather an attitude towards the quoted statement. Even though the quote contains a verb expressing an emotion (hate), this is very close to expressing an opinion. (20)

“I hate to disagree with Mr Gromyko before the talks start, but that clearly was not our position,” Mr Weinberger said. (tp-wn-850114)

On the other hand, emotions in quotes can serve to actually represent the emotional state of news actors. Often this co-occurs with the description of personal experience by those directly affected by a news event, for instance in the role of victims or witnesses. (21)

Outside court Mr Hussain’s son, Awais, 22, said he was “extremely grateful” to the court for releasing his father. He added: “I just hope other families have people like my dad.” Awais said of the attack on his family: “It was quite terrifying. I don’t think our family will get over it.” (to-uk-100120)

(22)

“It has been the hardest thing I have ever experienced and will ever experience in my whole life, no matter what happens to me. There will be nothing that will compare to the pain and heartbreak of watching my beautiful daughter leave this world.” (tp-uk-100126)

This use was observed in the online and the print data from 2010, but not in the earlier print data from 1985. In contrast to quotes that report actions carried out in an official capacity (Examples 18 and 19 above), these quotes provide a much more personal perspective on news. They describe the experience of individuals who became victims or witnesses of events, and when they do quote people who are represented in their official role, like the priest in Example (23), then the experience clearly surpasses the normal duties of this person. (23)

“I’ve been up and down this country and every door is closed,” the priest said. “We are running out of food; our children are throwing up water.” “They keep saying that the government is having a meeting and that soon someone will come. But no one comes and I no longer know what I should do.” (to-wn-100124)

These uses of quotes tend to occur in articles reporting on crimes and catastrophes. Therefore, one might perhaps think that their absence from the 1985 print data could be a consequence of a stronger focus on political topics in the front

48   

   Daniela Landert

page articles of this earlier data set. However, there are strong indications that the differences in topic do not provide a sufficient explanation. For instance, the earlier data contain two articles about a Cholera outbreak in an Ethiopian refugee camp, which would have provided the opportunity to integrate quotes expressing personal experience and emotion. Example (24) reproduces a brief extract (without any quotes) from one of the articles. (24)

Yesterday I visited Harbo, a new camp in a part of Wollo region which has only recently been affected by the more severe effects of a famine that is slowly spreading south. […]. In Harbo, where an average of 50 people still arrive daily, they show all the signs of comparatively recent arrivals. There is a resignation in the faces, a despair in the eyes and a listless quality to their movements. They sit and peer aimlessly out of tents or stand and watch the world without curiosity, a people who have given up. (tp-wn850123)

From the article, it becomes clear that the journalist was present at the scene and that he interacted with individuals who could have made statements similar to those in Example (23). Indeed, towards the end of the article, relief and charity workers are quoted. In contrast to the priest in Example (23), however, they do not describe their personal experiences in the crisis, but present an analysis of the political situation. (25)

“Colonel Mengistu came to power because of the failure of the last government to manage a famine. Now here he is with one which is five times as severe.” (tp-wn-850123)

This example suggests that the absence of quotes expressing emotion and personal experience in the printed Times from 1985 cannot be attributed to differences in topic and access to sources alone. The article about the cholera outbreak in an Ethiopian refugee camp would have provided many opportunities for integrating a more personal perspective on the events by quoting relief workers or individuals who were directly affected. Instead, the parts that allow readers the most direct and emotional access to the events occur outside of quotes, in the voice of the journalist (see Example 24). A more likely explanation for the absence of this function in the earlier data lies in a recent trend towards more emotional, sensational and personalised reporting in up-market newspapers. It has often been shown that up-market newspapers are taking over features that are typically associated with down-market newspapers (Bös 2010b: 84–85, Conboy 2010: 130–135, Schneider 2002). This

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   49

tendency, sometimes referred to as “tabloidisation” (see, for instance, Conboy 2006, Holly 2008, Sparks/Tulloch 2000), has affected the layout, the language and the content of up-market newspapers like the Times. Compared to the issues from 1985, the 2010 Times contains more images, more “soft news” (e.g. human interest stories), and a stronger focus on private actors in the news (see Landert 2014 for results based on the Times Online). The fact that news articles – in both print and online versions – also contain more quotes in general, more quotes with first person singular pronouns, and more quotes expressing emotion and personal experience fits very well into this overall trend.

7 Conclusion Over the last few decades, newspapers have undergone quite far-reaching changes, not least due to the development of new media. It is taken for granted today that high-profile newspapers offer content online, either as a freely accessible service financed through advertisement or as a service restricted to their paying subscribers. While such news sites clearly continue many of the conventions of their printed counterparts, it is also reasonable to assume that the new setting invites and facilitates change – not only in how news is produced and consumed, but also in the form and structure of texts and the functions of textual elements. One of the challenges of studying such changes lies in the difficulty of differentiating between general diachronic developments in news media, on the one hand, and the consequences of new technologies, on the other. In light of the many interdependencies of media changes and technological changes, a strict separation between these two aspects is not possible. Nevertheless, there are changes that can be shown to be more closely related to technological factors than others. In this study, I have investigated changes in the functions of quotes in articles from one online news site, compared to their usage in articles in the newspaper’s printed counterpart 25 years earlier. I have determined two main shifts. On the one hand, quotes expressing personal experience and emotion are characteristic of the more recent data. This function is closely related to the content of quotes, i.e. the kind of statements with which sources are quoted. The fact that there is no obvious relation between the content of quotes and technological factors is one indication that this change is not directly related to the online setting. Further evidence is provided by the finding that quotes with such personal content can also be found in more recent articles of the print version. Moreover, this development fits into the more general tendency of up-market newspapers to adopt

50   

   Daniela Landert

features typical of down-market ones. All this suggests that the online setting is of secondary importance for the increase of quotes expressing personal experience and emotion. On the other hand, I have argued that the online setting plays an important role for the function of quotes to present reportable facts. By integrating the source material from which quotes are taken and by allowing readers to compare such source material across different platforms, the factual status of quotes is reinforced. This reinforcement is directly related to the technological setting of online news sites, which allows the integration of videos and which facilitates hyperlinking. While references to online resources can also be found in print newspapers, linking in this case is less immediate and, moreover, is still dependent on the online setting. The potential for a seamless integration of written articles with videos, sound files and material on external websites is therefore one of the fundamental differences between print and online news.

References Bell, Allan. 1991. The Language of News Media. Oxford: Blackwell. Bös, Birte. 2010a. “Dialogic quotation patterns in historical news reports”, in: Nicholas Brownlees, Gabriella Del Lungo and John Denton (eds). The Language of Public and Private Communication in a Historical Perspective. Newcastle upon Tyne: Cambridge Scholars Publishing, 228–244. Bös, Birte. 2010b. “People’s Voices. Eine diachrone Betrachtung persönlicher Zitate in der britischen Presse“, in: Martin Luginbühl and Stefan Hauser (eds). MedienTextKultur. Linguistische Beiträge zur kontrastiven Medienanalyse. Beiträge zur Fremdsprachenvermittlung, Sonderheft 16. Landau: Verlag Empirische Pädagogik, 67–93. Brendel, Elke, Jörg Meibauer and Markus Steinbach (eds). 2011. Understanding Quotation. Berlin: de Gruyter. Bublitz, Wolfram and Christian Hoffmann. 2011. “‘Three Men Using our Toilet all Day Without Flushing – This May Be One of the Worst Sentences I’ve Ever Read’: Quoting in CMC”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 433–447. Bucher, Hans-Jürgen and Peter Schumacher. 2008. “Konstante Innovationen: Vom OnlineJournalismus zum konvergenten Journalismus. Wie neue Medien und alte Paradoxien die Kommunikation verändern“, in: Bernhard Pörksen, Wiebke Loosen, Armin Scholl (eds). Paradoxien des Journalismus. Theorie – Empirie – Praxis. Wiesbaden: VS Verlag für Sozialwissenschaften, 477–501. Burger, Harald. 2005. Mediensprache. Eine Einführung in Sprache und Kommunikationsformen der Massenmedien. Vol. 3. Berlin: de Gruyter. Cappelen, Herman and Ernie Lepore. 1997. “Varieties of quotation”. Mind 106, 429–450. Clark, Herbert H. and Richard J. Gerrig. 1990. “Quotations as demonstrations”. Language 66/4, 764–805.

Reportable Facts and a Personal Touch: The Functions of Direct Quotes in Online News   

   51

Collins, Daniel E. 2001. Reanimated Voices. Speech Reporting in a Historical-Pragmatic Perspective. Amsterdam: John Benjamins. Conboy, Martin. 2006. Tabloid Britain: Constructing a Community Through Language. London: Routledge. Conboy, Martin. 2010. The Language of Newspapers. Socio-Historical Perspectives. London: Continuum. Cooke, Lynne. 2005. “A visual convergence of print, television, and the Internet: charting 40 years of design change in news presentation”. New Media & Society 7/1, 22–46. Gans, Herbert J. 1979. Deciding What’s News. A Study of CBS Evening News, NBC Nightly News, Newsweek, and Time. New York: Pantheon Books. Grösslinger, Christian, Gudrun Held and Hartmut Stöckl (eds). Pressetextsorten jenseits der ‘News’. Sprache im Kontext 38. Frankfurt am Main: Peter Lang Holly, Werner. 2008. “Tabloidisation of political communication in the public sphere“, in: Ruth Wodak and Veronika Koller (eds). Handbook of Communication in the Public Sphere. New York: de Gruyter, 317–344. Jucker, Andreas H. 1996. “News actor labelling in British newspapers”. Text – Interdisciplinary Journal for the Study of Discourse 16/3, 373–390. Jucker, Andreas H. 2006. “‘but ‘tis believed that …’: speech and thought presentation in Early English newspapers”, in: Nicholas Brownlees (ed.). News Discourse in Early Modern Britain. Selected Papers of CHINED 2004. Bern: Peter Lang, 105–125. Jucker, Andreas H., and Manuel Berger. 2014. The Development of Discourse Presentation in The Times, 1833–1988”. Media History 20/1, 67–87. Kitagawa, Chisato and Adrienne Lehrer. 1990. “Impersonal uses of personal pronouns”. Journal of Pragmatics 14/5, 739–759. Landert, Daniela. 2014. Personalisation in Mass Media Communication: British Online News between Public and Private. Amsterdam: John Benjamins. Luginbühl, Martin. 2012. “‘I felt the glass windows of my parents’ apartment shaking’. Ein intermedialer und interkultureller Vergleich zur Inszenierung von Nähe in Zeitungsberichten und Fernseh-Korrespondentenberichten”, in: Grösslinger/Held/Stöckl (eds), 249–264. Mair, Christian and Marianne Hundt. 1997. “The corpus-based approach to language change in progress”, in: Hans Sauer and Uwe Böker (eds). Anglistentag 1996. Proceedings. Tübingen: Niemeyer, 71–82. Mayes, Patricia. 1990. “Quotation in spoken English”, Studies in Language 14/2, 325–363. Santulli, Francesca. 2012. “Zitierte Rede. Authentizitätsbeweis oder einfach Infotainment?”, in: Grösslinger/Held/Stöckl (eds), 265–279. Schäfer, Patrick. 2005. “Porträts in der Regionalpresse. Ein deutsch-französischer Vergleich”, in: Hartmut E. H. Lenk and Andrew Chesterman (eds). Pressetextsorten im Vergleich – Contrasting Text Types in the Press. Germanistische Linguistik Monographien 17. Hildesheim: Georg Olms, 223–242. Schneider, Kristina. 2002. The Development of Popular Journalism in England from 1700 to the Present. Corpus Compilation and Selective Stylistic Analysis. Doctoral dissertation. Universität Rostock. Short, Mick, Elena Semino and Martin Wynne. 2002. Revisiting the Notion of Faithfulness in Discourse Presentation Using a Corpus Approach. Language and Literature 11/4, 325–355. Sparks, Colin, and John Tulloch (eds). 2000. Tabloid Tales. Global Debates over Media Standards. Lanham: Rowman & Littlefield.

52   

   Daniela Landert

Stenvall, Maija. 2008. “On emotions and the journalistic ideals of factuality and objectivity – tools for analysis”. Journal of Pragmatics 40/9, 1569–1586. Tannen, Deborah. 1986. “Introducing constructed dialogue in Greek and American conversational and literary narrative”, in: Florian Coulmas (ed.). Direct and Indirect Speech. Berlin: de Gruyter, 311–332. Tuomarla, Ulla. 2000. La citation mode d’emploi: Sur le fonctionnement discursif du discours rapporte direct. Helsinki: Academia Scientiarum Fennica. Wales, Katie. 1996. Personal Pronouns in Present-day English. Cambridge: Cambridge University Press Włodarczyk, Matylda. 2007. Pragmatic Aspects of Reported Speech. The Case of Early Modern English Courtroom Discourse. Frankfurt am Main: Peter Lang.

Jenny Arendholz

2 Quoting in Online Message Boards: An Interpersonal Perspective Abstract: When linking the act of quoting to computer-mediated communication, on the one hand, and interpersonal pragmatics on the other, remarkable insights can be gained. Based on previous findings in both fields of research (Arendholz 2013, Herring 1999, Locher/Watts 2005), this paper focuses on quoting in online message boards in order to see how message board participants use quotes to refer to and – more importantly – to comment on other contributions and their authors. In this process, they strengthen or weaken, or in a word, redefine interpersonal relations between online interlocutors. By approving or denigrating the quoted message, quoters dispose of an effective tool to manipulate the face (Goffman 1967) of the person behind the quote, which is why quoting gains momentum as a key to doing facework. The nature of quoting to express interpersonal bonds is explored on the basis of a corpus of message board entries and signatures. Using a formal classification, the frequency of different types of quotes in these two message board templates can be determined. These frequencies are also analyzed on a functional plane with a view to their topical orientation and to their role in (re)negotiating face. Keywords: quotes, quoting, interpersonal relations, facework, message boards, ventriloquizing quotes

1 Introduction When engaging in conversation – be it oral, written or computer-mediated – we regularly put quotes to good use. For a long time now, pragmatics has recognized quoting as an important metacommunicative act which induces a wide array of different purposes in discourse research (cf. Cappelen/Lepore 2007, Jones/Schieffelin 2007, Bublitz/Hübler 2007). To mention only three of the most obvious functions, we quote to generate cohesive ties between otherwise sometimes unrelated elements of texts. In this way, we foster the establishment of coherence in texts and of common ground between interactants. Besides these structural and cognitive functions, we can also detect interpersonal benefits in the use of quotes. It is this particular function of quotes and, for that matter, of the more complex

54   

   Jenny Arendholz

act of quoting that will take center stage in this paper.¹ Located at the interface of interpersonal pragmatics and computer-mediated communication (CMC), this study focuses on online message boards, where quoting someone is – at least from a technological point of view – highly facilitated by simply clicking on a quote button (see Arendholz/Kirner-Ludwig, this volume). To find out what users achieve on an interpersonal plain, terminology and ideas are borrowed from Goffman (1967). Accordingly, it will be interesting to see how facework is negotiated inter alia with the help of quoting. For this reason, it is vital to work out what kinds of quotes can be identified in online message boards, which templates lend themselves for these quotes and who is being quoted. With this in mind, seven threads² randomly chosen from the British message board system The Student Room (TSR, URL 1) were put to a quantitative and qualitative test. This paper is thus structured as follows: after a short definition of both quotes and the act of quoting, a general, technology-driven introduction of quotes in message boards is given in section 2.1. Section 2.2 is concerned with face negotiation cycles and outlines interpersonal processes based on quotes in entries and signatures. Following a description of the corpus in 3.1, the quantitative and qualitative interpersonal results of the study are presented in sections 3.2 and 3.3 respectively. Last but not least, quotes are traced back to their original authors in section 3.4, in which the special case of ventriloquizing quotes is introduced as well. Finally, the conclusion in section 4 will sum up the most important insights and findings of the present study.

2 Quoting and Face Negotiation in Online Message Boards 2.1 Definitions and Technological Prerequisites At the outset of this study, some general definitional remarks are called for to introduce quotes, quoting and (their relations to) message boards. Based on

1 I would like to thank Christian Hoffmann for his collaboration in preparing and presenting an earlier version of this paper at the 12th International Conference of Pragmatics (IPrA) in Manchester, 2011. 2 The seven threads form part of a larger TSR message board corpus (consisting of 50 threads), which was originally compiled in 2010 for Arendholz (2013). It should go without saying that potential changes in the contents and/or appearance of threads – especially over time – can be expected and not avoided.

Quoting in Online Message Boards: An Interpersonal Perspective   

   55

Bublitz/Hoffmann’s definition, quoting is here viewed as “the act of taking up a source text (T1) of an author (A1) and shifting it from its context (C1) to another context (C2) as a target text (T2)” (2011: 434; cf. also Bublitz, this volume). As a result, a quote is generated. More often than not, it is formally marked as such in one way or another, classically by means of quotation marks and verba dicendi et sentiendi. In the case of message board entries, the process of marking (part of) a source text T1 as a component of a target text T2 is carried out by the message board system, viz. triggered by users’ simply clicking on a quote button (see also Arendholz/Kirner-Ludwig, this volume). Above that, users also creatively resort to a change of color or font, use italics, bold print or insert tildes and the like to set apart quotes from non-quotes. Although the quote is without a doubt the central element of the more complex act of quoting, it is not the only relevant element. After all, the quoter, who can be A1 or A2, also reflects on T1 either explicitly (verbally) or implicitly (prosodically or kinesically), draws the recipient’s attention to the text by attending to it, [and] indicates both T1 and C1 as common knowledge (or at least as knowledge mutually shared by quoter and recipient) (Bublitz/ Hoffmann 2011: 434).

It is important to note that the explicit or implicit reflection on T1 extends a mere quote to the act of quoting, viz. the fact that the quoter comments and thus complements and evaluates the quote. For this reason, the combination of quote and subsequent reflections by the quoter, i.e. the entire act of quoting, will be of interest for the upcoming examination. Entries, however, are not the only place to find quotes. Theoretically, every template which can be filled by creative users can thus feature quotes. The question of whether the quoter is given the chance to comment on these quotes, thus completing what has been defined as the act of quoting before, is, however, a different matter. We might therefore also have a look at a user’s profile information and message board signatures. The former does not form part of a current message board contribution as such but can be inspected on an adjacent webpage, mostly linked to the post by means of clicking on the nickname (cf. Arendholz 2013). The latter, on the other hand, is comparable to a business card underneath each post and visually separated from the actual post. Both profile information and signatures offer unlimited room for users’ personal content, possibly including quotes and corresponding evaluative comments by the quoter. Figures 1a-c give examples of these three different environments of quotes.

56   

   Jenny Arendholz

Figure 1a: Quote in an entry (thread 3, post 12)

= original context (profile)

= enlargement of quoted passage

Figure 1b: Quote as part of profile information (thread 3)

= original context (signature)

= enlargement of quoted passage Figure 1c: Quote as part of a signature (thread 3, post 4)

Quoting in Online Message Boards: An Interpersonal Perspective   

   57

Figure 1a displays the outcome of clicking the quote button: as part of a target text T2, the source text T1, including the bracketed mention of its originator A1, is automatically framed and put into quotation marks by the system. The act of quoting is completed by a comment drafted by A2 underneath the box. In 1b, the quote is indicated by the verbum dicendi “wrote” and the reference to its original author A1 (Sango). Although the system allows for evaluative comments by the quoter A2 (Tha_Black_Shinobi) in the profile section – after all, it is his very own profile section in which he can write and comment on whatever he pleases – any kind of evaluative response to this particular quote is lacking. Although the mere fact that the quote was reproduced in the profile might be seen as an implicit commentary, the act of quoting still remains incomplete. Exactly the same holds true for 1c, in which the quote as part of the signature is (at least originally) formally marked in a redundant fashion, i.e. by a change of color to green, quotation marks, italics and the mention of the author A1. The following interpersonal analysis is, however, limited to quotes found in entries and in signatures. After all, it could be argued that only these quotes have an immediate impact on the interpersonal relations between message board interlocutors that can be perceived at once. This is only the case with entries and signatures, which are integral parts of a message board thread, and not, however, with profile information.

2.2 Cycles of Face Negotiation Based on Quotes in Entries and Signatures Now that message board quotes have been located within threads and sketched from a formal point of view, their interpersonal potential can be illustrated – beginning with the quotes in entries. The starting point for the upcoming discussion is content provided by an A1 which is later on quoted and thus turned into a quote via A2, the quoter. This quoted content can be interpreted as lines emitted by an A1, i.e. as “a pattern of verbal and nonverbal acts by which he expresses his view of the situation and through this his evaluation of the participants, especially himself” (Goffman 1967: 5). In deciding to quote a T1 by an A1, A2s do not only establish the cohesion necessary for the argumentative structure of their own contributions. More importantly they also pick up A1’s lines and react to them: firstly by repeating them in the form of a quote, and secondly, by commenting on and evaluating them in their own contributions underneath

58   

   Jenny Arendholz

the quote-box.³ Accordingly, an A1 is seen in a certain light by their interlocutors (A2, 3, n) and evaluated inter alia on the basis of their (behavior in) T1. Again borrowing from Goffman’s terminology, the end-product of this evaluative process is face, “the positive social value a person effectively claims for himself by the line others assume he has taken during a particular contact [and] an image of self […]” (1967: 5). Note, however, that although a certain face is desired by A1, Goffman argues that it can only be ascribed to them by their fellow-users A2, 3, n based on their (non-)verbal behavior, their lines. Figure 2 graphically displays this linebased cycle of face negotiation found in entries.

Figure 2: Complete face negotiation cycle in entries (to be read from right to left)

As can be seen in Figure 2, A1’s message board (MB) contribution (I’m superfrankie!), which is read as a line by interlocutors (and future quoters), undergoes recontextualization when quoted (“I’m superfrankie!”) and evaluation when commented on (Good for you!! ᄕ) by A2, 3, n. In this way, both the act of quoting and consequently the face negotiation cycle are complete. In the end, face is given to A1 by the quoter(s) on the basis of A1’s lines, which can either be judged by A2, 3, n as appropriate and reasonable (when the lines are ratified) or not (when the lines are rejected). This cycle is then repeated in the course of interaction with new lines, i.e. new verbal or non-verbal behavior, ready to be evaluated. A different picture presents itself when looking at quotes and interpersonal processes in signatures. Reconsider the example presented in Figure 1c. Here the author A1 is not a message board user but instead the American poet Oliver

3 This mirrors the fact that we do not only communicate on a purely ideational level, but also always on a relational one (cf. Watzlawick et al. 1967, Halliday 1978). In pragmatics, this basic dichotomy has been labeled transactional vs. interactional function of language (Brown/Yule 1983) or message vs. metamessage (Tannen 1992).

Quoting in Online Message Boards: An Interpersonal Perspective   

   59

Wendell Holmes, Sr., who is quoted by an A2, Ghost Grey, the creator of the signature under investigation. Similar to the analysis of the entry above, we could make an – arguably moot – point about A1’s, i.e. Holmes’ lines and resultant face attributions by A2. Far more interesting insights can be gained, however, when looking at the quoter A2’s lines and face, who is in fact an active participant in the message board exchange. After all, using a quote by a famous author in one’s signature can be interpreted as a line strategically placed to be given a certain face by another message board user A3, 4, n. Whether targeting A1 or A2 in search of lines and processes of face attributions, evaluative comments are lacking in both cases: despite being a possible option from a technological point of view, neither is Holmes’ quoted T1 commented on by Ghost Grey in his signature (see Figure 3 below for a counter-example), nor are Ghost Grey and his signature evaluated by fellow-users in surrounding entries. For this reason, we can establish that the act of quoting remains incomplete in both these cases. The same is true for the face negotiation cycle. Although lines are available, face attribution is only laid out potentially, which is why users might still pick up on one line or another perceived here and there when composing their entries. For the remainder of this paper, (quotes in) signatures are not taken into consideration due to their lack of actual evaluative components vital for the attribution of face.

3 Results of the Study 3.1 The Corpus Material As already mentioned, the seven threads under investigation represent a random sample taken from a larger message board corpus (Arendholz 2013). This corpus consists of altogether 50 threads downloaded from the British message board system The Student Room and was originally compiled in 2010. While the first column of Table 1 indicates the position of each thread in the corpus, the thread title is followed by information regarding the number of posts and contributors in each thread, as well as the date of the first and the last post added to the thread. In the upcoming analysis, 437 posts authored by 293 users and amounting to approximately 66,000 words (minus the content automatically inserted by the system), were taken into account.

60   

   Jenny Arendholz

Table 1: Details of threads under investigation

#

thread title

No. posts

No. date of first users & last post

3

Why Are So Many People On Here Homeschooled?

81

49

12/08/2009 – 13/08/2009

4

booze illegal?

67

37

13/08/2009 – 14/08/2009

8

‘Life is too short’

57

31

10/08/2009 – 13/08/2009

14 How to stop strange people speaking to me

54

41

13/08/2009 – 14/08/2009

15 To drink or not to drink….?

67

54

12/08/2009 – 01/09/2009

17 Burton’s rejected me for a credit card, I think :S

23

12

12/08/2009 – 13/08/2009

28 Women Make Crappy Musicians

88

69

21/07/2009 – 31/08/2009

3.2 A Quantitative Analysis of Message Board Quotes I will now provide an overview of the various types of quotes that were encountered during the quantitative part of the analysis, again distinguishing between quotes in entries and in signatures (see Table 2). Table 2: Quantitative findings for quotes in message board entries and signatures

position

marked as quote

originator (A1)

No.

%

entries

Yes (quote button)

MB author

197

60.3

Other

MB author

 19

5.8

Other

non-MB author

 19

5.8

No

MB author

  5

1.5

No

non-MB author

  8

2.5

Yes

MB author

 20

6.1

Yes

non-MB author

 59

18.0



327

100

signatures

As can be seen in Table 2, the investigation of the seven message board threads yielded 327 instances of quotes, 79 of which occur in signatures and 248, the

Quoting in Online Message Boards: An Interpersonal Perspective   

   61

majority, which appear in entries. Similarly to the example shown in Figure 1c, almost two thirds of quotes in signatures stem from A1s that are not and never have been (cf. 3.4) message board participants. The remaining third can be traced back to other message board users who, at one point or another, authored something worth quoting – at least in the eyes of some A2s. The signature found in thread 17 (post 1) bears witness to this phenomenon (see Figure 3).

Figure 3: Quote by a MB author in a signature (thread 17, post 1)

The quote in the signature displayed in Figure 3 is formally marked by simple quotation marks and a change of font size as well as by a mention of the source A1, DJkG.1, obviously a fellow-user of the quoter A2. Another way to indicate the fact that we are faced with a statement by a fellow-user is the use of a quote box in a signature, which is not unlike those used in entries (see Figure 4).

Figure 4: Quote box inserted in a signature (thread 3, post 29)

While all the quotes in signatures were formally marked as such one way or another, the same cannot be said of the quotes in the entries, which will be of interest in the following. Most of them, 235 to be precise, were either marked automatically when clicking on the quote button or by creative keyboard tricks produced by resourceful users (i.e. tildes, changes of fonts and/or colors etc.). But we also find 13 cases of unmarked quotes, which can only be detected through a close reading of the contributions.

62   

   Jenny Arendholz

221 of the quotes in entries were originally drafted by message board users, while the remaining 27 were written by external, non-MB authors. Table 2 also shows that the prototypical message board quote (with 197 of all 327 quotes, i.e. 60.3 %) was formally the product of the quote button and was originally authored by a TSR-user. In the following qualitative, interpersonal analysis, this more prominent type of message board quotes will take center stage.

3.3 A Qualitative Analysis of Interpersonal Functions of Quoting In the second step of this analysis, only those 197 quotes produced by the quote button in the entries are investigated from an interpersonal point of view. The omission of signature quotes leading to a reduction of the sample is unavoidable, considering the fact that the act of quoting is only complete in entries. After all, at this point the relevant question is how the (complete) act of quoting – and not the mere quotes – gains momentum as a key to doing facework, i.e. how quoting is used to strengthen or weaken, or in a word, redefine interpersonal relations between online interlocutors. As it turns out, by approving or denigrating the quoted message, quoters dispose of an effective tool to manipulate the face of the person behind the quote. Based on the remaining 197 quotes in the entries, the following classification scheme was devised (see Figure 5).

Figure 5: Classification scheme of interpersonal alignments in comments

As demonstrated in Figure 5, quoting A2s have several possibilities to turn their subsequent comments into means of evaluation which are either directed at the proposition of the quote or at its originator, A1. These two possible ways, respec-

Quoting in Online Message Boards: An Interpersonal Perspective   

   63

tively labeled ad res and ad hominem in Figure 5, both lead to face attributions for A1s executed by A2s. On the one hand, ad hominem evaluations are always direct in my data and they exhibit either a positive or a negative alignment in the form of direct appreciation (see top right) or direct flaming⁴ (see top left). On the other hand, ad res evaluations can feature three different alignments: positive and thus face-enhancing, negative and thus face-threatening, and also neutral. This latter alignment goes to show that not all content-related, i.e. ad res comments, automatically yield face-changing effects. When they do, however, as in the other two cases, face-threatening disapproval and face-enhancing approval for the proposition at hand can again be brought forth directly or indirectly. As indicated by the circular arrangement of the chart and the dashed arrows, all ad res evaluations – minus, of course, the neutral ones – necessarily lead to ad hominem evaluations, so that in the end all evaluative comments with a positive or negative alignment affect A1’s face. After all, tearing down someone’s argument is bound to have some impact on the feelings and on the image of the person behind it. The term evaluative drift, which I used at the bottom of the chart, hints to the fact that we deal with scalar and not polar phenomena. This also becomes obvious in the following five examples in Figures 6–10.

Figure 6: Example of a negative ad hominem evaluation, flaming (thread 3, post 34)

The example in Figure 6 displays a comment underneath a quote box, which can be counted as a direct ad hominem attack on Funkymonkey21’s face. The lexeme fool and the sarcastic undertone of the comment can only be taken as attack on A1’s feelings and a threat to their face, not only from A1’s point of view but also from that of other users. In this case, the act of quoting functions as an act of flaming. 4 In trying to define the term flaming, a lot of different propositions have been put forward in the last two decades. Flaming as used in this paper is best defined broadly as “hostile and aggressive interaction” (Thurlow et al. 2004: 70) or with regard to a single utterance as “a highly negative message that functions like a metaphorical flamethrower that the sender uses to roast the receiver verbally” (O’Sullivan/Flanagin 2003: 68f.).

64   

   Jenny Arendholz

Figure 7 illustrates another ad hominem reaction to a comment, this time, however, featuring the opposite, viz. positive alignment:

Figure 7: Example of a positive ad hominem evaluation, appreciation (thread 28, post 47)

Yawn-er’s quoter shows their consent and thus their appreciation of A1’s taste and also of A1 themself in the most unambiguous way. Helped by the honorific sir, the user gives A1 a positive face value (i.e. that of having good taste) – at least when ruling out an ironic reading of this comment. Again this unambiguously positive face attribution is perceivable for its recipient as well as for all their fellow-users. The next screenshot (see Figure 8) showcases an example of an ad res comment which carries indirect disapproval of Kiwiguy’s opinion and, in the end, of Kiwiguy himself, who thereby experiences a threat to his/her face as a competent discussion partner.

Figure 8: Example of an ad res evaluation, indirect disapproval (thread 4, post 46)

Although Figure 9 illustrates yet another case of ad res evaluation, the alignment is opposite this time. A1, Flying Cookie, experiences direct and unambiguous approval from the quoter A2, which results in the strengthening of A1’s face. Last but not least, an example of a neutral ad res evaluation is given in Figure 10. In this act of quoting, the accompanying comment underneath the quote box is neither positively nor negatively marked. Instead, it should be counted as a neutral attempt to advance the conversation in a coherent way without entailing any perceivable changes in cpj1987’s face.

Quoting in Online Message Boards: An Interpersonal Perspective   

   65

Figure 9: Example of an ad res evaluation, direct approval (thread 3, post 60)

Figure 10: Example of a neutral ad res evaluation (thread 3, post 6)

The distribution of the seven possible types of evaluative comments found in my corpus is represented in Table 3: Table 3: Qualitative findings for quotes produced by the quote button in entries

type

subtype

directness

No.

ad hominem

flaming

direct

 17

8.6

appreciation

direct

  2

1.0

disapproval

direct

 20

10.2

indirect

 44

22.3

neutral



 68

34.5

approval

direct

 31

15.7

indirect

 15

7.6



197

100

ad res

%

66   

   Jenny Arendholz

As mentioned earlier, not every comment within an act of quoting has implications for the quoted person’s face. In fact, 34.5 % of the 197 entries under investigation are neutral, probably used for the sake of topical cohesion. The second most common type is represented by ad res evaluations that convey indirect disapproval (22.3 %). This might be expected from message boards, in which topics are discussed controversially despite the fact that flaming is usually forbidden by the netiquette. This also explains the comparably high number of ad res comments designed to put across direct approval (15.7 %), a face-enhancing and desirable strategy, which also promotes discussion. On the whole, ad res comments are comparatively face-saving and thus definitely preferred (90.3 %) – something that cannot be said about their ad hominem counterparts. It is, in fact, interesting to note that when users decide for direct ad hominem evaluations, which do not take topic related “detours” before hitting their target, face concerns for their interlocutors seem to vanish: while we find 17 cases (8.6 %) of flaming, only 2 instances (1.0 %) of appreciation could be detected.

3.4 Ventriloquizing Quotes As has been established before (cf. Table 2), we can draw a distinction between message board internal A1s (i.e. MB-authors) and message board external A1s (i.e. non-MB-authors) whose content is quoted by A2s, be it in entries, signatures or other sections. What has so far been neglected is the following question: who are these external authors that constitute no less than 26.3 % of all A1s? Looking again at the corpus material, we find famous people, among them authors, poets, musicians, politicians, actors and the like, who are readily quoted by message board users. While these external A1s are rather rare in entries – after all only 27 out of 248 cases (10.9 %) were found – the figures differ considerably when looking at signatures where there are 59 out of 79 examples of external A1s (74.7 %). These diverging percentages can be explained when reconsidering the prevalent communicative purpose of (quotes in) entries, on the one hand, and (quotes in) signatures on the other: the creation of coherence within the communicative encounters of internal message board users vs. the users’ self-presentation with the help of external authorities. There is, however, another subtype of external A1s besides those listed above. Sometimes originators of content turned into quotes are neither fellow message board users nor well-known people, but purely imaginary authors. Quoting fictitious content by fictitious authors creates what can be called ventriloquizing quotes (Tannen 2003, cf. Sams 2007). See, for example, Figures 11a and b (my own underlining, JA).

Quoting in Online Message Boards: An Interpersonal Perspective   

   67

Figure 11a: Ventriloquizing quotes (thread 4, post 41) ((blank line inserted))

Figure 11b: Ventriloquizing quote (thread 17, post 8)

Judging by the expressions which function very similarly to verba dicendi, viz. “was told” and “the most education kids get is” (both Figure 11a), the quotative like (Figure 11a) and the quotation marks (Figure 11b), both figures present passages that look like ‘real’ quotes from a formal point of view (see underlining). However, close inspection reveals them to be pseudo-quotes since the actual source is an imagined rather than a real person. Based on a shared cultural background, users of these ventriloquizing quotes put words into the mouth of a cardboard figure adopting, for example, the generic role of a parent, a teacher, or a friend, including the impersonal “they” or “people”, who might as well have spoken these words. What is also striking about these quotes is the absence of an authentic T1, as it is in fact A2 (not A1) who is responsible for the phrasing of T1. We could thus say that A2 is quoting a maybe remembered or maybe purely imagined type of T1, but not a real token thereof.⁵ The use of ventriloquizing quotes reflects the need to disassociate oneself from the statement put forth by such a fictitious person (cf. echo theory of irony, Wilson/Sperber 1992: 60). Since everyone involved can be expected to be familiar with real-life representatives of these fictitious people – especially in a relatively tight-knit community of practice such as TSR – it is not necessary to spell out an actual, real source.

5 I would like to thank Wolfram Bublitz for his valuable comments concerning ventriloquizing quotes.

68   

   Jenny Arendholz

4 Conclusion As this paper has tried to show, it is useful and thus strongly advisable to distinguish between a quote as text shifted to a new context of use, on the one hand, and the act of quoting on the other. The latter is defined by the quoter’s explicit or implicit (verbal or nonverbal) expression of attitude which is added to the quote and which is directed either toward the proposition (ad res) of the quoted text or its designated author (ad hominem). While quotes occur in entries, signatures and profile information, the act of quoting is only completed in entries. This does not, however, mean that quotes in signatures and profile information can never serve as a basis for a complete act of quoting. The difference lies, however, in the fact that someone else’s expression of attitude toward it can only be placed in another message board post and not directly underneath the quote itself. The latter option is technologically reserved for entries alone. One way or the other, quotes can be seen as lines, which are used by interlocutors as an indicator and a springboard for the attribution of face. In the case of entries, face is given to the quoted person A1 by means of comments, while quotes in signatures and profile information – obviously lacking such comments by others – reveal more about the quoting A2 and not so much about the quoted A1. This study has shown that almost two thirds of all quotes in the corpus were generated using the quote button feature. While 90.3 % of these quotes were found to be topic-related, only 9.6 % were explicitly author-related. Also, more than a third of the quote-button quotings found in the MB entries have neutral attitudinal orientations, most likely serving the purposes of mending disrupted adjacencies and synchronizing communicative moves. Two other types of quoting frequently found in the corpus were indirect disapproval and direct approval. This shows that MB users are keen to either mitigate potential facethreats or enhance each other’s face when giving ad res comments. It also became obvious that when commenting on users, not on their propositions, concerns about possible face threats hardly played a role (cf. 17 instances of flaming). Last but not least, I outlined two types of original authors: other MB users and external (famous) authors. The latter also includes the subtype of imaginary authors, which are usually produced when MB authors wish to disassociate themselves from a stereotypical claim or judgment.

Quoting in Online Message Boards: An Interpersonal Perspective   

   69

References Arendholz, Jenny. 2013. (In)Appropriate Online Behavior. A Pragmatic Analysis of Message Board Relations. Amsterdam: John Benjamins. Brown, Gillian and George Yule. 1983. Discourse Analysis. Cambridge: Cambridge University Press. Bublitz, Wolfram and Christian Hoffmann. 2011. “‘Three Men Using our Toilet all Day Without Flushing – This May Be One of the Worst Sentences I’ve Ever Read’: Quoting in CMC”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 433–447. Bublitz, Wolfram and Axel Hübler (eds). 2007. Metapragmatics in Use. Amsterdam: John Benjamins. Cappelen, Herman and Ernie Lepore. 2007. Language Turned On Itself. The Semantics and Pragmatics of Metalinguistic Discourse. New York: Oxford University Press. Goffman, Erving. 1967. Interaction Ritual. New York: Pantheon. Halliday, Michael A. K. 1978. Language as Social Semiotic: The Social Interpretation of Language and Meaning. Baltimore: University Park Press. Herring, Susan. 1999. “Interactional cohesion in CMC”. Journal of Computer-Mediated Communication 4/4. n.p. Jones, Graham and Bambi Schieffelin. 2007. “Enquoting voices, accomplishing talk: uses of be + like in Instant Messaging”. Language and Communication 29, 77–113. Locher, Miriam and Richard J. Watts. 2005. “Politeness theory and relational work”. Journal of Politeness Research 1/1, 9–33. O’Sullivan, Patrick B. and Andrew Flanagin. 2003. “Reconceptualizing ‘flaming’ and other problematic messages”. New Media and Society 5/1, 69–94. Sams, Jessie. 2007. “Quoting the unspoken: an analysis of quotations in spoken discourse”. Colorado Research in Linguistics 20, 1–16. Tannen, Deborah. 2003. “Power maneuvers or connection maneuvers? Ventriloquizing in family interaction”, in: Deborah Tannen and James E. Alatis (eds). Linguistics, Language, and the Real World: Discourse and Beyond. Washington, DC: Georgetown University Press, 50–62. Tannen, Deborah. 1992. That’s not What I Meant! How Conversational Style Makes or Breaks Your Relations with Others. London: Virago Press. Thurlow, Crispin, Laura Lengel and Alice Tomic. 2004. Computer-Mediated Communication: Social Interaction and the Internet. Los Angeles et al.: Sage. Watzlawick, Paul, Janet B. Bavelas and Don D. Jackson. 1967. Pragmatics of Human Communication: A Study of Interactional Patterns, Pathologies, and Paradoxes. New York: Norton. Wilson, Deirdre and Dan Sperber. 1992. “On verbal irony”. Lingua 87, 53–76. URL 1: http://www.thestudentroom.co.uk/, last accessed 1 May 2013

Birte Bös and Sonja Kleinke

3 The Complexities of Thread-internal Quoting in English and German Online Discussion Fora Abstract: This study is part of a large-scale research project on English and German forum discussions. It provides a contrastive perspective on the characteristics of quoting in such online communities, which are exemplified by two extensive sample threads from the public message boards BBC Have Your Say (HYS) and Spiegel Online (SPON). The German SPON-thread proves clearly more favourable of quoting and is generally more interactive than the English HYS-thread, as demonstrated by the higher number and structural complexity of quotation patterns. In both fora, quoting fulfils important coherence-related and relational functions, and can thus be considered as a constitutive characteristic of online discussions. This study expands the functional characterisation of quoting by highlighting the trigger/mental access function of quotes and by elaborating on interpersonal and intergroup-related aspects. That quoting is a multidimensional process with forum-specific realisations is demonstrated in two qualitative case studies on the allocation of targets and the processes of perspectivising interpersonal and intergroup criticism via metonymic elaboration. Keywords: thread-internal quoting, adjacency, metonymic elaboration, evaluation, identity construction, group constellations

1 Introduction The systematic application of quoting as an interactive practice is one of the most prominent features of online discussion fora, which has, however, not received much in-depth treatment in research on Computer-Mediated Discourse (CMD) so far.¹ Our study investigates thread-internal quotations and quoting in English and German discussion fora, and aims at a detailed comparison of their formal and functional characteristics. In distinguishing quotation and quoting, we acknowledge that it is not the quote as such (in Volosinov’s terms, the secondary discourse) which fulfils certain functions, but its contextual embeddedness in the primary discourse, i.e. the actual posting (cf. Fairclough 1988: 125f). Through1 But see, e.g. Severinson Eklundh/Macdonald 1994, Herring 1999, Barcellini et al. 2005, Severinson Eklundh 2010, Bublitz/Hoffmann 2011.

72   

   Birte Bös and Sonja Kleinke

out this paper, we thus use the term quotation – the combination of quote and comment (cf. also Bublitz/Hoffmann 2011, Bublitz, this volume) – to refer to the actual linguistic material, when the focus is on its structural characteristics. The term quoting indicates a functional perspective, referring to the actual process of transferring linguistic material as an activity performed by the user (cf. Arendholz/Kirner-Ludwig, this volume). Accordingly, we define thread-internal quoting as the de- and re-contextualisation of material from previous postings in subsequent postings in the thread of an online discussion by a quoter, i.e. the poster actively and intentionally transferring the source material. As typical of CMD, this process is heavily determined by the technical facilities provided by the respective online platforms. For our close-up study of quoting practices, we selected two sample threads from a corpus of 40,000 postings from discussion fora accompanying non-tabloid mass media. Our English data were drawn from the BBC platform Have Your Say (HYS) and the German material was taken from Spiegel Online (SPON). Both threads are comparable with regard to their topics and number of postings. HYSparticipants discuss the question Should the US give the Pope such a presidential welcome? in 881 postings. In SPON, the 754 postings of the thread deal with the question Der Papst in den USA – Retter in der Not? (‘The Pope in the US – the saviour?’) Both fora are frequented by a large number of (usually anonymous) users and can be considered as public spaces (cf. Eysenbach/Till 2001: 1104). The data were retrieved electronically. Both authors tagged the corpus individually, focussing on the structure and targets of quotations, relational functions of quoting, and instances of metonymic elaboration. Cases in which tags diverged were discussed and then categorised jointly. While both quantitative and qualitative features were compared regarding the formal characterisation of quotations in the two threads, we opted for a focus on qualitative results in our discussion of the functional properties of quoting in order to explicate the complexities of quoting phenomena in our material. The specific formal characteristics of thread-internal quotations in our English and German data are described in section 2. Section 3 gives an overview of the major functions of quoting in the two discussion fora investigated. Finally, section 4 presents the targets of contextualising comments and a case study of metonymic elaboration. Our data provide evidence of the systematic interplay of relational, coherence-related and perspectivising functions of quoting processes in forum discussions.

Thread-Internal Quoting in Online Discussion Fora   

   73

2 Major Formal Characteristics: Frequency, Position and Structural Complexity of Thread-Internal Quotations Our data show that the range of quotations traditionally found in written texts can also be observed in the threads of forum discussions. The data provide evidence of canonical direct quotations, free direct quotations (i.e. quotations which are not marked by quotation marks and/or reporting verbs – illustrated in Table 1 below by an example from SPON including an English passage) and short, incorporated quotations, which highlight specific elements of the original posting; and they also feature indirect quotations and discourse summaries (see Table 1).² Table 1: Formal realisations of quotations in the corpus

Form of quotation

Examples

Direct (canonical)

Significantly, in his welcome the President also said: “In a world where some treat life as something to be debased and discarded, we need your message that all human life is sacred and that each of us is willed.” (HYS#417)

Free direct

Viele Katholiken ignorieren z.B. die Lehren des Vatikan zur Empfängnisverhütung komplett³: In 1968 the Pope reaffirmed the traditional view with the encyclical Humanae Vitae. Many lay people simply decided that this was a teaching they did not need to follow in order to be Catholic. … (SPON#593)

Incorporated

The Pope said before arriving that he was “deeply ashamed” and that the scandal had caused “great suffering” for the church and for “me personally.” (HYS#19)

Indirect

Especially as Bush has openly said that god has been guiding him on the Iraq war. (HYS#155)

Discourse summary

Several here write of respect for The Pope. Why? (HYS#29)

2 For more information on the formal classification of quotes see Leech/Short’s (1981) comprehensive classification system, and also Short (1988), Clark/Gerrig (1990), Short et al. (1997), Kurz (2000) and Burger (2005) (summarised in Bös 2010). 3 Transl.: Many Catholics completely ignore the doctrines of the Vatican on issues such as contraception.

74   

   Birte Bös and Sonja Kleinke

Here, we focus on direct thread-internal quotations, which are typically processed through the semi-automatised quoting devices offered by the fora.⁴ Both HYS and SPON allow participants to include original posts via mouse click on the Quote (HYS) or Zitieren (SPON) buttons. In both fora, the posting to be cited is first presented in a comment box, where it can be edited, e.g. by selecting and thus highlighting specific elements of the source posting (see Figure 1). The formal marking of the quote in a subsequent posting is generated by the system, and consists of a box and the label Posted by [name] (HYS)/Zitat von [name] (SPON). In SPON, a mouse click on the box links the reader back to the original posting, a technological feature which obviously proves important with respect to the creation of coherence in that it allows the receiver to reactivate the original context of the quoted material in the ongoing discussion.⁵ Additionally, the application of manual copy/paste-techniques is, of course, possible. Such quotations are usually indicated by classic quotation marks. Despite these technical similarities, the frequency and complexity of thread-internal quotations varies in the two fora analysed.

Figure 1: Technical set-up on BBC Messageboard

4 For the technical details see Bublitz/Hoffmann (2011). 5 To what extent this technical facility is actually exploited by users, and whether knowing the original context of the quote (cf. Bublitz/Hoffmann 2011) is actually necessary and common practice in forum discussions is not altogether clear for threads as complex as the ones we have studied – see also Marcoccia (2004) and section 3.

Thread-Internal Quoting in Online Discussion Fora   

   75

Figures 2 and 3 provide an overview of the distribution of quotations across the two threads. In the figures, each posting is represented by one box. The boxes of those postings which contain thread-internal quotations are marked with regard to position of the quote within the posting and differentiated from ‘unmarked’ postings without any quotes. As a comparison of the two figures shows, the SPON thread is clearly much more favourable of quoting than the HYS thread. In fact, 94 % of the postings in SPON (i.e. 710 out of 754) contain thread-internal quotations. In contrast, this is only the case in approximately 18 % of the HYS postings (157 out of 881). Furthermore, quoting patterns are much more complex in SPON, where multiple-quote messages,⁶ i.e. postings including more than one threadinternal quote, are quite common (358, i.e. 47.5 % of all postings). In HYS, the technical default option of single initial quotes is preferred (127 cases). There are only 25 postings featuring multiple quotes and in five messages, the quote occurs in mid-message position.

Figure 2: Thread-internal quotations in SPON

6 Terminology based on Barcellini et al. (2005: 308f), cf. also ‘shell quoting’ (Bublitz/Hoffmann 2011).

76   

   Birte Bös and Sonja Kleinke

Figure 3: Thread-internal quotations in HYS

Thread-internal quotations are not only much more frequent in SPON, they are usually also much longer than in HYS, with the longest SPON-quotation extending to 313 words. In HYS, it is quite common to repeat the topic question of the thread in posting-initial position. Where source material from a previous posting is used, it is usually presented at the beginning of the posting, often rigorously edited by the poster and cut down to the relevant passages, thus even deleting the names of the users quoted (which are originally included in the automatic processing of source material, cf. ex. (1))⁷. In the following, examples can be identified by the name of the thread and the number of the posting in brackets. The quoted source material from previous 7 For ethical reasons, all examples discussed in the text are given in their fully anonymised version rather than as screenshots, leaving the actual text, punctuation, spelling, etc. otherwise unchanged for reasons of authenticity.

Thread-Internal Quoting in Online Discussion Fora   

   77

postings is marked in bold type, the posters’ comments are marked in italics such as in ex. (2). In the case of long comments, analytically relevant passages are indicated in bold italics. (1)

Forum: BBC Should the US give the Pope such a presidential welcome? #37 Author: {{HYS_45}} “had caused “great suffering” for the church” What about the children? Date: 2008-04-16 13:09:00

(2)

“had caused “great suffering” for the church” What about the children? (HYS_45 in #37)

The posting-initial quotations in SPON often reproduce the whole source posting, as provided by the automatic system, or at least substantial parts thereof. However, the multiple-quote messages in SPON display more pronounced editing techniques. Here, the quoters intentionally select specific elements of previous postings and insert them in their own posting. This is illustrated by ex. (3), where, in posting #180, SPON-participant 48 (SPON_48) reproduces two elements from a previous posting in reverse order to the original and comments on them. The coherence of the original posting is thus destroyed, yet, the quoter re-contextualises the source text by adding asterisks to highlight a typo by SPON_6, who had complained about the lack of orthographical competence of SPON-participants. (3)

Zitat von **{{SPON_6}}** Wieso muss die Anti-Papst-Häme in diesem Strang immer mit so mangelhafter Beherrschung der deutschen Sprache einhergehen? Hm, sehr gute Frage! Zitat von **{{SPON_6}}** Sicher **merh** als Sie, aber so ist das Leben. Halten Sie es doch einfach so wie die Mehrzahl der hier vertretenen Foristen und halten sich in dem Wissen um die eigenen Unvollkommenheit mit Hinweisen zur Beherrschung der deutschen Sprache zurück - trägt sowieso nichts zum Thema bei. (SPON_48 in #180) [Quote by **{{SPON_6}}** Why does the anti-Pope malice in this thread always have to go hand in hand with such insufficient mastery of the German language? Hm, very good question! Quote by **{{SPON_6}}** Certainly **moer** than you, but such is life. Just do as the majority of the posters in this forum do and, in the awareness of your own imperfections, hold yourself back with comments regarding the mastery of the German lan-

78   

   Birte Bös and Sonja Kleinke

guage – this doesn’t contribute anything to the topic anyway. (SPON_48 in #180)] In sum, the findings regarding the formal characteristics of quotations further attest to the general tendency for greater brevity and structural simplicity in HYS than in SPON, which other studies on the corpus have also indicated (cf. e.g. Kleinke 2010, Kleinke/Bös 2015).

3 Major Functional Characteristics of Thread-Internal Quoting Our investigations show that participants in both fora apply quoting practices for a similar range of functions capturing textual and interpersonal uses⁸, yet with diverging prominence. We conceptualise this functional variety as a continuum, with the two major functions, coherence and relational (i.e. interpersonal and intergroup) uses, at its extreme ends (cf. Figure 4). Clearly, this idealised scale is not completely balanced, and our data indicate that, typically, quoting patterns display a complex functional profile. Primarily coherence-related



Primarily relational

Figure 4: Functional continuum of quoting in discussion fora

3.1 Coherence Due to the asynchronous, polylogal nature of forum discussions, one of the most vital functions of thread-internal quoting is to provide context and create adjacency.⁹ This coherence-establishing function is the one that has received most attention in the literature, although the term coherence itself often refers to quite different concepts, which are not always clearly delineated. Barcellini et al. (2005), Herring (1999), Severinson Eklundh/Macdonald (1994) and Severinson Eklundh (2010) postulate two levels of coherence in online discussion fora, which are outlined below and to which we will add a third layer. 8 See e.g. Halliday (2003). 9 See e.g. Severinson Eklundh/Macdonald (1994), Herring (1999), Severinson Eklundh (2010), Bublitz/Hoffmann (2011).

Thread-Internal Quoting in Online Discussion Fora   

   79

(i) Text-deixis – a ‘reminder for the reader’ As a macro-level context-preserving mechanism (Barcellini et al. 2005: 302, Herring 1999, and Severinson Eklundh/Macdonald 1994: 200 on e-mail interaction), the quotation merely denotes or points at those parts of the topic in previous discourse which the quoter wants to refer to in his/her comment, in a text-deictic function¹⁰ (cf. ex. (4), (5)). In both threads, the discussions are very long and involve many participants. Often sub-discussions between alternating groups of participants unfold simultaneously.¹¹ Therefore, coherent material is often spread over long passages. Text-deictic functions, where users almost literally point at the portion of the discussion they want to comment on, seem to supersede the cohesive function usually related to the verbatim repetition of lexical material in quotations (cf. Bublitz/Hoffmann 2011: 446). However, at the micro-level of a quotation which links up quote and comment in a single posting, the cohesive and coherent functions of grammatical constructions and lexical repetition often go hand in hand. They may furthermore be enriched by relational functions, as repetitions of linguistic material often prove to be instances of mimicking, which lends the discussion a clearly non-convivial flavour. (4)

So one elderly German in a badly fitting white frock says sorry in English and that makes everything OK does it? {{HYS_13}}  – No: the offenders will be judged, Judged that is, when time comes. {{HYS_66}}  – It also helps that the church is compensating many of the victims with money. (HYS_151 in #156)

(5)

Zitat von **{{SPON_1}}** Pädophilenskandal, Glaubwürdigkeitsverlust, Finanzdebakel - die katholische Kirche in den USA steckt in der Krise. >>> Was kann der Papst bewirken? Warten wir ab, was er am Ground Zero sagen wird. Dann wissen wir`s. (SPON_5 in #5) [Quote by **{{SPON_1}}**Pedophile scandal, loss of credibility, financial debacle – the Catholic church in the US is in a crisis >>> What can the Pope do about it? Let’s wait what he says at Ground Zero. Then we’ll know. (SPON_5 in #5)]

10 In thread-internal quotations, the text-deictic function is realised by a ‘fully-fledged’ portion of a text referring to itself – we thank Wolfram Bublitz for this comment on an earlier version of the paper. 11 See also Bublitz (2008, 2012) on multilinearity and multilaterality, as well as Marcoccia (2004) on online polylogues (Marcoccia’s term) in public forum discussions.

80   

   Birte Bös and Sonja Kleinke

(ii) Creating adjacency, dialoguicity and polyloguicity In forum discussions, quotations systematically bridge disrupted adjacency at the macro-level of the entire thread “by juxtaposing portions of two turns – an initiation and a response – within one message” (Herring 1999: 8, cf. Barcellini et al. 2005: 302), and they thus often go hand in hand with dialoguicity, which is a typical feature of many postings. Dialoguicity is defined here as involving two or more participants (cf. Roulet 1984: 42, Schwenter 2000: 260).¹² Users in Severinson Eklundh/Macdonald’s study see dialoguicity as one of the most important functions of quotations, as it allows people “to reply directly, as in a conversation, without having to explain what point [they are] referring to” (1994: 199). In addition, quoting contributes to the overall polylogal nature of the discussions at two levels: as observed by Herring (1999), Marcoccia (2004) and Barcellini et al. (2005), quoting enables many users to respond directly (and sometimes almost simultaneously) to a previous posting, thus turning the dialogue between two participants into a polylogue between many participants. In addition, by quoting material from different previous postings in one multi-quote message, as in ex. (6) and (9) (see also Bublitz/Hoffmann 2011), users can enter into a direct exchange with several users in one single message, thus achieving a density of the polylogal interaction that is typical of discussion fora and cannot usually be reached within one single turn in natural face-to-face group discussions.¹³ Thus, quoting is not only induced by multi-user interactions, but, especially in the German thread, helps to create polyloguicity and adjacency, both on the macro-level of the whole thread as well as the micro-level of individual postings, as a constitutive element of forum discussions. In our threads, participants use a variety of linguistic means to create adjacency, including (rhetorical) question/answer sequences (ex. (7)) or question/ counter question sequences (ex. (8)), interjections, cohesive devices such as repetitions, elliptical constructions (ex. (9)), sentences starting with coordinating conjunctions, etc. (6)

“Why do you not criticize Islam or Judaism with the same venom you reserve for us.” {{HYS_275}} “I am sure if it was any other non Chris-

12 However, we embrace the distinctions, pointed out by Roulet (1984) and Schwenter (2000), between dialogal discourse/dialoguicity (as defined above) and dialogical discourse/dialogicity, which relates to the expression of opposing viewpoints. In our data, the latter often co-occurs with the former. 13 See also Herring (1999) on combining response moves referring to several postings by the same poster and the greater intensity of CMD interaction in general and Günthner (1998) on similar phenomena in reported speech in oral interaction – though in far less complex communicative events involving fewer participants.

Thread-Internal Quoting in Online Discussion Fora   

   81

tian church leader the BBC would not publish.” {{HYS_33}} Oh do stop playing the victim card. All faiths and faith leaders get the same disdain on Have Your Say. You only need look at any thread that has remotely anything to do with Islam or the Archbishop to see evidence of such. Then again evidence has never been the religionists’ strong point. (HYS_479 in #514) (7)

“Does the Pope have a moral obligation to promote Catholic doctrine to wider society?” No, definitely not. I have absolutely no problem with Catholic people and communities believing what they like, and practising their religion. But I really resent the Vatican trying to influence our politicians’ votes, whenever an issue of personal conscience comes up in Parliament. The law of the land is for everyone, whether we’re Christian or not. (HYS_10¹⁴ in #1)

(8)

“I don’t understand why anyone is remotely interested in this gentleman’s visit. He wears funny clothes and recites a lot of mumbo jumbo in rituals that are favoured by those simple folk who enjoy the myths and superstitions that make up this particular flavour of this particular religion. What does he actually do?” Are we talking about the pope or Gordon Brown’s visit? (HYS_389 in #415)

(9)

Zitat von **{{SPON_42}}** Da “WIR” ja Papst sind, ist PBd.16. meiner Meinung nach im Auftrag der CDU nach Amerika geflogen….oder ? CSU ! (SPON_35 in #57) [Quote by **{{SPON_42}}**As “WE” are Pope, PBd.16., in my opinion, has flown to America on behalf of the CDU…hasn’t he? CSU !¹⁵ (SPON_35 in #57)]

Although dialoguicity is a prominent side-effect of adjacency, the dialogal flavour may vanish if the comment is written in a letter format (containing such structural devices as salutations), if it is too long, or clearly monologally structured (including enumerations, conclusions, etc). Examples where quoting creates adjacency, but lacks dialoguicity, are far more common in SPON (18.3 instances per 100 postings) than in HYS (1 instance per 100 postings). Sometimes, the structure and length of the quotation itself seems to reduce dialoguicity, which may

14 Deviating post and participant numbers due to deletion of insulting posts by the BBC as the host of the forum. 15 CDU refers to the Christian Democratic Union, CSU to their more conservative Bavaria-based sister party, the Christian Social Union.

82   

   Birte Bös and Sonja Kleinke

also lead to interesting hybrid patterns, with a monologally structured quotation at the beginning followed by a short comment creating dialoguicity (as well as dialogicity in the sense of opposing viewpoints, which may have an impact on relational functions of quoting, cf. section 3.2, cf. fn. 12) (ex. (10)). (10)

“The Catholic church is much maligned and now it has become abundantly clear that they were right all along in rejecting the Copernican heliocentric system and subjecting Galileo and his heresies to the inquisition, then their influence in the world should be greater. In the same way I look to them to taking leadership in rejecting the obviously false cult that has built up around the man made climate change myth. Climate change (if it is happening at all) is God’s work and it is sacrilege and arrogant for man to promote the falsehood that he is more powerful than the creator… {{HYS_71}}” This is a joke, right? (HYS_230 in #239)

(iii) Trigger function – mental access and perspectivisation In addition to the two layers of coherence ((i) and (ii)) commonly discussed in the CMD literature, we postulate a third layer of coherence termed trigger function. As our data show, posters also use quotes as a mental access point for the conceptual perspectivisation of their own views and their alignment to the views of others (see section 4.2 for a more detailed discussion). In the two threads, quoted material not only relates back to a previous posting, but frequently initiates or triggers processes of metonymic elaboration by providing a conceptual (and contextual) anchorage point from which the quoter accesses the situation referred to in the comment (Kleinke 2012a). This is one of the prototypical characteristics of cognitive metonymies, which “not only make target meanings accessible but also available, e.g. as new topics, for further elaboration in the ensuing discourse” (Panther/Thornburg 2004: 91). Metonymic elaboration is understood here as rooted in the mechanisms of cognitive metonymy – “a cognitive process in which one conceptual entity, the ‘vehicle’ (also often called the ‘source’) provides mental access to another conceptual entity, the ‘target’ within the same cognitive model” (Radden/Kövecses 1999:  21). The linguistic material of the vehicle resides in the quote (realised as a morpheme, word, phrase, sentence) (Panther/Thornburg 2004: 96). In their quotes, users highlight a usually carefully selected aspect of a situation expressed in a previous posting acting as a vehicle that triggers and perspectivises the comment following the quote in a quotation. By the same token, the poster incorporating a quote into his/her own message provides inferential pathways for the reader (Panther/Thornburg 2007: 248f, Barcelona 2003: 97, 2011: 25).

Thread-Internal Quoting in Online Discussion Fora   

   83

Stirling (1996: 69) describes the special type of metonymic inferencing that is relevant for our data as indirect anaphora, i.e. “instances of reference where there is no explicit, identical antecedent for the referring nominal expression, yet nevertheless it is felt that its interpretation depends in some systematic way upon the linguistic context” (cf. ex. (11) from the HYS-thread, and Barcelona 2003, 2011 on more complex cases).¹⁶ The metonymic link in quotations such as exemplified in ex. (11) is in fact established between two utterances, a quote from a previous message posted by user HYS_314 and a comment directly attached to it by HYS_151, who instantiates the actual conceptual link. It is important to point out here that HYS_151 does not simply resolve a metonymic implicature via inferential processing, thus arriving at an interpretation that HYS_314 may have intended, but that s/he productively¹⁷ and textually elaborates on an entity mentioned in the quote s/he selected from the previous stream of discourse. (11)

Bush is a man who killed thousands of people in iraq in the name of terrorism, what the hell is the pope doing with a murderer. Regards {{HYS_314}} I thought God loves everyone, including sinners. (HYS_151 in #574)

In ex. (11), HYS_151 relates to a trigger expression in the quote (the pope) that was neither consciously nor unconsciously intended by user HYS_314 to be used as a metonymic source by another participant. The trigger expression the pope merely serves as an access point from which HYS_151 enters the two subdomains god and sinner of the more complex target domain the church in a part for part metonymy. As many instances of such metonymic links in the two threads show, the productive aspect of metonymic elaboration does not only concern the range of source and target variations in a text by one author as described by Brdar-Szabó/Brdar (2011), but may be extended by an additional interactional and supra-individual level. We thus consider the trigger/mental access function to be an important extension of the two established layers of coherence, as it helps to illuminate cases in which the linguistic material in the quote serves as the con16 The phenomenon of metonymic inferencing has not escaped researchers striving for an account of coherence outside the framework of a cognitive-functional approach – cf. for example Erkü/Gundel (1987), Cornish (1999), Schwarz-Friesel (2007), Schwarz-Friesel/Consten (2011), who discuss similar inferential processes such as indirect anaphora, conceptual/bridging inference, or (inferential) bridging. 17 Cf. also Schwarz-Friesel (2007), Cornish 2010 and Schwarz-Friesel/Consten (2011), who include the productive perspective in their work on indirect anaphora and coherence without, however, discussing the role of cognitive metonymy in this process.

84   

   Birte Bös and Sonja Kleinke

ceptual anchorage for an elaboration on the intended target.¹⁸ How the process of metonymic elaboration can be exploited to highlight a specific perspective on an entity¹⁹ will be briefly discussed and exemplified in section 4.2. Our analysis indicates that quotes with mere text-deictic functions are rather rare in both threads (1.1 instances per 100 HYS-postings and 5.3 cases per 100 SPON-postings). In contrast, the adjacency function with a dialogal flavour is particularly prominent (13.9 instances per 100 HYS-postings and 63.3 cases per 100 SPON-postings)²⁰, and the trigger function also proves quite frequent (8.5 instances per 100 HYS-postings and 17.6 cases per 100 SPON-postings). In addition to creating text-deictic coherence, dialoguicity and mental access to forthcoming material, there is often a transition to and overlap with other functions such as hostility, solidarity and group construction, which are captured in the second major group of functions discussed in section 3.2 below.

3.2 Relational Functions of Quoting In our data, quotes are systematically linked to interpersonal comments which target individual posters, and intergroup comments aimed at members of a certain camp (e.g. pro- and contra-church camps in the forum). Quoting thus strongly contributes to the negotiation of interpersonal and intergroup relations, and features particularly prominently in interpersonally tense passages of forum discussions where users verbally attack each other, triggering metapragmatic comments as illustrated in ex. (12) (cf. also Schütte 2000). As pointed out by Bublitz/Hoffmann (2011: 444), quotative judgements are an important means of relational work and for the generation and negotiation of identities in online discussion fora. (12)

“Ah yes, the lack of condoms force these ‘millions’ into a lifestyle of promiscuity and prevent them from working the land to grow food. Amazing how necessary condoms are for human survival.

18 Brendel/Meibauer/Steinbach (2011: 21) discuss a related phenomenon from a context-related perspective, stressing the receptive side of the process, in which the context of the quotation provides access to the interpretation of the context of the quoted passage (but cf. Bublitz/Hoffmann 2011 on the same mechanism in opposite direction). 19 Cf. Barcelona (2011: 12–13) drawing on Langacker’s (1999: 207ff) notions of reference point and construal. 20 The relative differences between HYS and SPON result from the far higher frequency of quotations in SPON. The actual percentages are almost similar here: 72.6 % of the HYS-quotations and 67.2 % of the SPON-quotations show adjacency + dialoguicity.

Thread-Internal Quoting in Online Discussion Fora   

   85

{{HYS_88}}” Very ignorant and flippant comment {{HYS_88}}. Would you be one of those charming neo conservatives we’ve been hearing so much about in the UK. (HYS_82 in #271) Figure 5 provides an overview of the major relational functions of quoting, which we understand as overlapping and often implying each other. Starting out from a major differentiation of consensual and contentious uses, we can ascribe evaluative and identity-constructing functions to quoting (Hübler/Bublitz 2007: 18, Bublitz/Hoffmann 2011: 443). The left-hand column of Figure 5 shows particular characteristics of consensual usage, whereas the right-hand column relates to specific aspects of contentious usage. For a more detailed explanation of the categories listed, see below.

CONSENSUAL

CONTENTIOUS

EVALUATION Positive evaluation

Negative evaluation – Patronising – Mimicking

Elaboration (Lecturing) Agreement

Elaboration (Lecturing) Disagreement Defence

IDENTITY CONSTRUCTION Self-oriented Other-oriented: Individual

– Other-oriented: Individual

Fuzzy

Fuzzy

Group

Group

Ingroup Outgroup: Type II Type III Type IV

– Outgroup: Type II Type III Type IV

Figure 5: Relational functions of thread-internal quoting in online discussion fora

The upper half of Figure 5 lists possible realisations of evaluation in the comments about the quotations. In consensual quoting, quoters express their agree-

86   

   Birte Bös and Sonja Kleinke

ment with the poster of the source posting more or less explicitly (e.g. You are absolutely right, HYS_428); they provide positive evaluations regarding the actions or personal characteristics of certain targets (represented in the lower part of Figure 5), and elaborate on selected aspects raised in the quote. Elaboration, which (especially in the German thread) may even extend into lecturing with long sequences displaying e.g. enumerations and a sophisticated argumentative structure, is also found in contentious quoting. Here, the quoter might express disagreement with the poster quoted or defend his/her position or that of another poster (typically an ingroup member, see below). Negative evaluations are provided not just explicitly, but also via patronising and mimicking strategies (cf. ex. (13) below). The lower half of the figure shows the relevant targets involved in the construction and negotiation of identity via quoting. As quotation practices in the two threads show particular differences with regard to these targets, some basic explanations are provided here, before the category target receives a more detailed treatment in section 4.1. Obviously, quoting is mostly other-oriented, relating to an individual or a group target, or, in fuzzy cases, to unclear or combined targets. Our data clearly indicate that in the two threads investigated (and probably also in other political discussion fora), quoting does not only serve important interpersonal functions, but is also highly relevant in the construction and negotiation of related group identities. Regarding potential group constellations, we postulate four major types based on our study on the use of rudeness tokens in the two discussion fora (Kleinke/ Bös 2015). This classification extends Kienpointner’s (1997) twofold distinction of intergroup rudeness, which focussed on hierarchical group constellations: Type I refers to members of a majority group directing at minority groups; Type II, vice versa, refers to members of a minority group directing at majority groups. Whereas Type I does not prove relevant for our data, Type II relates to comments with a venting function which are directed against majority targets outside the forum (i.e. targets equipped with a certain institutional power in the sense of Tajfel 1981 such as the Pope, Catholic Church, etc.). To these types we have added two categories which capture non-hierarchical group constellations: Type III covers the construction of ingroups and outgroups within the forum; Type IV refers to the construction of outgroups outside the forum. The most prominent realisation of a Type III constellation in our data are the pro- and contra-church camps, which are constructed and negotiated with much vigour throughout the threads. As previously indicated, the pragmatic classification suggested in Figure 5 above has to cope with the fuzziness of categories and the multifunctionality of quoting – see ex. (13) and (14) for illustration:

Thread-Internal Quoting in Online Discussion Fora   

   87

(13)

Heck, its an election year. Gotta line up the voters. With GW and the Pope (along with conservative Christians) sharing like views on homosexuality, abortion, etc. I am not surprised to see Washington roll out the red carpet. {{HYS_46}} Hmmm. Yes, there are similarities, but most ‘conservative christians’ are evangelicals who think that the Catholic Church is the harlot from the Book of Revelation. They may appear united in purpose but there are fundamental ideological divisions! (HYS_68 in #107)

(14)

I am astounded by so much hate filled verbiage by the liberal elites. And I am amazed how little understanding many have of history and the Catholic church. I however suspect that this is religious intolerance coming from the followers of the profit Darwin. {{HYS_329}} I am astounded that a church like the catholic curch (for that matter many others) is still revered around the world, even though it has commited so many crimes against humanities in the name of god…… You can believe what ever you want but do not indoctriante me with your believes. As an athisist I can reassure you that i am 1000X more tolerant than an religiouse fanatic. (HYS_370 in #392)

Ex. (13) demonstrates the fuzzy boundaries between consensual and contentious uses in the case of ‘yes-but’ strategies, where the quoters mitigate their agreement to the point of transition into disagreement, so that the comment oscillates between [consensual] and [contentious]. In contrast, ex. (14) is clearly [contentious], as are roughly 82 % of all quoting cases in HYS and 92 % in SPON, reflecting the general tendency that agreement is rarely acknowledged in these discussions. It also illustrates the multifunctionality of quoting in both threads: the comment provides a [negative evaluation], and more precisely it is [mimicking] the quotation. It is targeted not just at the [individual poster], but contributes to the group construction in the thread by attacking the religiouse fanatic[s] of the pro-church camp, i.e. [outgroup, Type II], and providing a selfaffiliated group label (athisist) for the quoter. This example provides a first illustration of how interpersonal and intergroup functions are tied to different targets and often overlap. The differences regarding the manifestation of simple and complex targets in HYS and SPON, and how participants handle them in relation to interpersonal and intergroup functions are discussed in more detail in the next section.

88   

   Birte Bös and Sonja Kleinke

4 Illustrating the Complexities 4.1 Targets of Contentious Quoting and Their Interpersonal and Intergroup Effects Regarding the targets of contentious quoting in the two groups, two main patterns can be distinguished. In roughly half of the quotations in HYS (54.8 %) and two thirds of all cases in SPON (65.5 %), a critical comment is directly related to the actual quote and directed at a single target. Such simple targets may include the individual user who has produced the posting from which the quote was taken, as in ex. (15), or one or several members of a majority outgroup (Type II targets, in ex. (16) former US-President Bush representing the US-government). Comments may also relate to a member of the outgroup inside the thread (Type III targets) – in our case members of the pro- or anti-church groups as in ex. (17). (15)

Zitat von **{{SPON_2}}** Was hat der Pabst uns zu sagen? Nichts. Wenn Sie den Titel noch nicht mal schreiben können, glaube ich das gern (SPON_6 in #5) (‚Quote … What can the Pobe tell us? Nothing. If you can’t even spell the title correctly, I can well believe that.’)

(16)

Was President Bush right to give him such a special welcome? ---Was the Pope right to accept such a special welcome from President Bush? (HYS_186 in #192)

(17)

Yes, the US administration is right to welcome the spiritual leader of about a billion people including many Americans. If the atheists are upset about it, perhaps it’s about time they elected their own leader. {{HYS_322}} … (Atheists don’t need leaders  – we’re intelligent enough to figure out right from wrong without being told. There’s an inverse relationship between degree of religiousness and level of education.) (HYS_498 in #534)

Particularly SPON has a high number of simple target realisations where the user quoted is also the personal addressee²¹ of a contentious comment, which certainly contributes to the overall non-convivial tone of the discussion (57.7 instances in 100 SPON-postings vs. 9.7 instances in 100 HYS-postings). Equally interesting

21 See also Kleinke (2010), Kleinke/Bös (2015), Arendholz (this volume).

Thread-Internal Quoting in Online Discussion Fora   

   89

from a contrastive perspective is the second pattern: quotations in which participants direct their criticism at combined targets and go beyond the quote of the individual poster to include group targets, thus moving away topically from the content of the original quote. Such extended targets may include majority Type II targets (ex. (18)) as well as thread-internal outgroup Type III targets, or even a combination of both (ex. (19)). (18)

“The British are always the first to make anti-catholic comments on HYS but are probaly to scared to do the same about Islam (or any other religon). {{HYS_1}}” The British have no problem making antiIslamic comments in every day life [Target = Individual], [contentious], [disagreement], its just that on forums such as HYS the BBC censors us heavily, hoping to maintain their beloved multicultural fiction. [Target = Outgroup type ii], [contentious], [Negative evaluation] (HYS_58 in #51)

(19)

So one elderly German in a badly fitting white frock says sorry in English and that makes everything OK does it? {{HYS_13}} – No [Target = Individual], [contentious], [disagreement]: the offenders [Target = Outgroup type ii], [contentious],[ Negative evaluation] will be judged, Judged that is, when time comes. Now there’s a working system of moral economy that the postmodern atheists on here [Target = Outgroup type iii], [contentious], [Negative evaluation] have to do without. (HYS_66 in #59)

In ex. (18), the comment starts with a contentious disagreement directed at an individual poster (HYS_1). HYS_58 achieves a mitigating effect through the downtoning phrase it’s just that combined with a secondary intergroup focus when s/ he resorts to assumed common ground by extending his/her criticism beyond this disagreement to a negative evaluation of the censoring practice and the beloved multicultural fiction of the BBC (majority outgroup Type II). Quotations with a mitigating secondary intergroup focus realised by an extended target in addition to the personal criticism against an individual poster constitute a relatively high percentage of quotes in the HYS group, but are relatively rare in the German thread. In ex. (19), the target of the critical comment is even more complex. User HYS_66 also starts out with a personal criticism directed at the individual poster HYS_13, with whom s/he blatantly disagrees. But in contrast to ex. (18), the mitigating addition of a secondary Type II target (the offenders) is followed by further criticism with a rather fuzzy target. HYS_66 directs his/her negative evaluation

90   

   Birte Bös and Sonja Kleinke

unmistakably at the thread-internal anti-church outgroup (the postmodern atheists on here – a Type III target). Clearly, using disparaging denominations of the Catholic Church and its official representatives, especially the Pope, is an ingroup marker of the anti-church group in this discussion. Thus, by his/her flippant remark, HYS_13 constructs him-/herself virtually as a member of the anti-church camp and therefore, he/she turns indirectly into an individual, personal target of HYS_66’s criticism of the postmodern atheists on here. Type II targets used in addition to individual targets often have a clearly mitigating effect, which is often reinforced by additional linguistic material such as downtoners or hedges. By contrast, the combination of a critical comment directed at an individual poster or posting with a criticism against a thread-internal outgroup (i.e. Type III target) often appears to enforce rather than weaken the impact of a potential interpersonal criticism. As illustrated by ex. (19), such a reinforced reading is further supported by the absence of additional downtoners in the criticism against the offenders and the specific perspectivisation of the object of criticism achieved by metonymic elaboration, which will be discussed in more detail in section 4.2. A similar reinforcing effect seems to be linked to instances of extended criticism against hierarchically equal outgroup targets (Type IV) as in ex. (20), which are rare in the two threads studied here. The fact that the forum is run by the BBC and that the moderators posed the thread-initial question “Is the US right to give the Pope such a welcome?” gives the combination of targets in ex. (20) another additional twist: (20)

“Is the US right to give the Pope such a welcome?” Why are you asking this question BBC? [Target = Outgroup type ii/Individual], [contentious], [Negative evaluation] I get the impression this is just an excuse to invite all the anti US brigade to say what they like with your permission. [Target = Outgroup type ii/Individual], [contentious], [Negative evaluation] The US can invite who they like to their country, it’s their business and I don’t really understand why this warrants a debate. I bet you wouldn’t dare ask is the US right to invite Osama Bin Laden? Oh no, we can’t have our beloved Islamists [Target = Outgroup type iv], [contentious], [Negative evaluation] insulted now can we..? [Target  = Outgroup type ii], [contentious], [Negative evaluation] (HYS_639 in #688)

On the one hand, the BBC as a majority organisation clearly constitutes a typical Type II outgroup target. On the other hand, the thread-initial question not only nominates the general topic of the discussion, it is also often quoted postinginitially in the same way as contributions of individual participants. Thus, in a

Thread-Internal Quoting in Online Discussion Fora   

   91

way, the BBC simultaneously assumes the role of an individual poster and the first target of the comment oscillates between an individual and a Type II-reading. In his/her comment, HYS_639 first replies with a contentious counter-question, open to an interpretation as rhetorical and challenging, as it is followed by a lengthy self-reply giving detailed reasons for why this question should not be asked in the first place (The US can invite who they like to their country, it’s their business and I don’t really understand why this warrants a debate. I bet you wouldn’t dare ask is the US right to invite Osama Bin Laden?). Another potentially rhetorical question is added at the end of the posting, extending the fuzzy individual/Type II target: Oh no, we can’t have our beloved Islamists insulted now can we..?. Argumentatively, this potentially rhetorical question²² further backs the negative evaluation of the reporting practice of the BBC (who would not insult Islamists). At the same time, the religious Type IV outgroup of Islamists is sarcastically characterised as a beloved group, thus reinforcing the criticism against the BBC, who allegedly supports it.

4.2 Illustrating the Complexities II: Metonymic Elaboration and Target Perspectivisation Both groups use metonymic elaboration in order to tie their comments to a quote selected from a previous posting. This technique allows users not only to create coherence, but also to perspectivise their elaboration on a selected entity (cf. Barcelona 2011: 12f). Through metonymic perspectivisation, quoters can highlight a certain aspect of an entity mentioned in the quote, which acts as the source domain.²³ Ex. (19), repeated here as (21), aptly illustrates these processes. (21)

So one elderly German in a badly fitting white frock says sorry in English and that makes everything OK does it? {{HYS_13}} – No [Target = Individual], [contentious], [disagreement]: the offenders [Target = Outgroup type ii], [contentious], [ Negative evaluation] will be judged, Judged that is, when time comes. Now there’s a working system of moral economy that the postmodern atheists on here [Target =

22 For a more detailed account of the use of rhetorical questions in online fora see also Kleinke (2012b). 23 In our data, this highlighting function exceeds predicative uses of metonymy as discussed in Ungerer/Schmid (2006) and involves a broader range of structural realisations, cf. ex. (10) and (20).

92   

   Birte Bös and Sonja Kleinke

Outgroup type iii], [contentious], [Negative evaluation] have to do without. (HYS_66 in #59) In the quote, user HYS_13 implements the complex Noun Phrase one elderly German in a badly fitting white frock, which potentially acts as a cognitive access point or source for a straightforward cognitive referential part for whole, more specifically, salient characteristic for person/garment for person metonymy: the complex metonymic expression elderly German in a badly fitting white frock represents salient aspects of the person denoted and – in the context of this discussion – provides access to the intended referent the Pope. User HYS_66 has obviously mastered the intended metonymic inferencing and can now for his/her part base his/her own metonymic elaboration on the merely inferred referent. In a second layer of the complex process of metonymic elaboration, the conceptual entity the Pope provides cognitive access to the (Catholic) Church in a conventionalised part for whole (more specifically: salient member for institution) metonymy. In step three, user HYS_66 implants another conceptual metonymy, now based on the inferred cognitive entity the (Catholic) Church. In the context of this discussion, which was topically overshadowed by the childabuse scandal in the US, this whole for part metonymy (again, more specifically, institution for salient member of an institution) does not merely provide access to the offenders as the conceptual entity that HYS_66 wants to discuss. In addition to the cognitive access function, HYS_66 exploits this whole for part metonymy (the catholic church for offenders) in order to highlight the limited scope of his/her criticism against a Type II target (here probably inserted for the mitigating effect). This also helps to perspectivise his/her access to the complex cognitive domain the (Catholic) Church in that it activates an access route – offenders and will be judged – into a juridical frame the Church can resort to in dealing with the scandal. A similar, though less complex, process of perspectivisation can be observed in HYS_13’s initial cognitive referential metonymy, which offers the outstanding garment of the Pope and his age as a conceptual access route into the cognitive domain the (Catholic) Church, exploiting its ridiculing potential. A comparison of metonymic elaborations in both threads revealed that the SPON-participants – due to their sheer quantity of quotations – use metonymic elaboration in more diverse patterns and more frequently than HYS-users. The two threads also have different preferences in their choice and granularity of conceptual metonymies. Further contrasts emerge at interpersonal and intergroup levels: almost as in scapegoat repair (Cheepen/Monaghan 1990: 81ff), HYSparticipants use metonymic elaborations more often than SPON-participants to

Thread-Internal Quoting in Online Discussion Fora   

   93

mitigate a personal criticism against an individual poster by adding an additional criticism to a common outgroup target (most frequently Type II), as exemplified in ex. (18). In addition, due to the overall greater complexity of quotations, the German thread shows a broader range of coherence-creating devices beyond conceptual metonymy (such as the direct repetition of lexical material, lending their postings a clearly provocative flavour as indicated in section 3).

5 Conclusions Quoting is an important constitutive practice in the two online communities examined in this study. Yet, the detailed analysis of the data revealed some interesting quantitative and qualitative contrasts. SPON-participants quote in higher quantities and apply more complex structural quotation patterns than HYS-users. Both groups systematically use quoting to establish macro-level deictic coherence, and macro- and post-internal micro-level adjacency, which often creates a dialogal flavour and fosters the emergence of a truly polylogal interaction. With processes of metonymic elaboration, the paper introduces a third layer of coherence at the micro-level of the quotation itself, focusing on the prospective aspect of how participants tie their comments conceptually to the linguistic material of the quote. Concerning relational functions, both threads mainly use quotations for negative evaluations. However, a detailed qualitative investigation of the complexities of target allocation revealed that SPON-users tend to exploit the contentious potential of quoting more systematically at an interpersonal level inside the thread. In HYS, mere interpersonal criticism is less frequent. Instead, interpersonal comments are often combined with negative evaluations of commonground thread-external majority targets, a pattern which can generate mitigating effects. Additionally, processes of metonymic elaboration help users to conceptually attach such second-order targets to the quoted text and to perspectivise their criticism. It is these complexities that render quoting in CMD a fascinating object of research, which invites a more systematic treatment of questions regarding the correlation of topic and (non-) consensuality in quoting, the cross-cultural nature of differences in quoting practices, and the complexities of cognitive mechanisms underlying the speaker-related productive side of coherence.

94   

   Birte Bös and Sonja Kleinke

References Barcellini, Flore, Françoise Detienne, Jean-Marie Burkhardt and Warren Sack. 2005. “A study of online discussions in an open-source community: reconstructing thematic coherence and argumentation from quotation practices”, in: Peter van den Besselar, Giorgio De Michelis, Jenny Preece and Carla Simone (eds). Communities and Technologies 2005. Dortmund: Springer, 301–320. Barcelona, Antonio. 2003. “The case for a metonymic basis of pragmatic inferencing: evidence from jokes and funny anecdotes”, in: Klaus-Uwe Panther and Linda Thornburg (eds). Metonymy and Pragmatic Inferencing. Amsterdam: John Benjamins, 81–102. Barcelona, Antonio. 2011. “Reviewing the properties and prototype structure of metonymy”, in: Réka Benczes, Antonio Barcelona and Francisco José Ruiz de Mendoza Ibáñez (eds). Defining Metonymy in Cognitive Linguistics. Towards a Consensus View. Amsterdam: John Benjamins, 7–57. Bös, Birte. 2010. “People’s Voices – Eine diachrone Betrachtung persönlicher Zitate in der britischen Presse”, in: Martin Luginbühl and Stefan Hauser (eds). MedienTextKultur. Linguistische Beiträge zur kontrastiven Medienanalyse. Beiträge zur Fremdsprachenvermittlung, Sonderheft 16/2010. Landau: Verlag Empirische Pädagogik, 67–93. Brdar-Szabó, Rita and Mario Brdar. 2011. “What do metonymic chains reveal about the nature of metonymy?”, in: Réka Benczes, Antonio Barcelona and Francisco José Ruiz de Mendoza Ibáñez (eds). Defining Metonymy in Cognitive Linguistics. Towards a Consensus View. Amsterdam: John Benjamins, 217–248. Brendel, Elke, Jörg Meibauer and Markus Steinbach. 2011. “Exploring the meaning of quotation”, in: Elke Brendel, Jörg Meibauer and Markus Steinbach (eds). Understanding Quotation. Berlin: de Gruyter, 1–33. Bublitz, Wolfram. 2008. “‘Sailing the islands or watching from the dock’: the treacherous simplicity of a metaphor. How we handle ‘new (electronic) hypertext’ versus ‘old (printed) text’”, in: Andrea Gerbig and Oliver Mason (eds). Language and Computers, Language, People, Numbers: Corpus Linguistics and Society. Amsterdam: Rodopi, 249–273. Bublitz, Wolfram. 2012. “From speaker and hearer to chatter, blogger, and user. The changing metacommunicative lexicon in computer-mediated communication”, in: Ulrich Busse and Axel Hübler (eds). Investigations into the Meta-Communicative Lexicon of English. A Contribution to Historical Pragmatics. Amsterdam: John Benjamins, 151–176. Bublitz, Wolfram and Christian Hoffmann. 2011. “‘Three Men Using our Toilet all Day Without Flushing – This May Be One of the Worst Sentences I’ve Ever Read’: Quoting in CMC”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 433–447. Burger, Harald. 2005. Mediensprache. Vol. 3. Berlin: de Gruyter. Cheepen, Christine and James Monaghan. 1990. Spoken English. A Practical Guide. New York: Columbia University Press. Clark, Herbert H. and Richard J. Gerrig. 1990. “Quotations as demonstrations”. Language 66, 764–805. Cornish, Francis. 1999. Anaphora, Discourse, and Understanding. Oxford: Clarendon. Cornish, Francis. 2010. “Anaphora: text-based or discourse-dependent? Functionalist vs. formalist accounts”. Functions of Language 17/2, 207–241.

Thread-Internal Quoting in Online Discussion Fora   

   95

Erkü, Feride and Jeanette Gundel. 1987. “The pragmatics of indirect anaphors”, in: Jef Verschueren and Marcella Bertuccelli Papi (eds). The Pragmatic Perspective. Amsterdam: John Benjamins, 533–546. Eysenbach, Gunther and James E. Till. 2001. “Ethical issues in qualitative research on Internet communities”. BMJ 323, 1103–1105. Fairclough, Norman. 1988. “Discourse representation in media discourse”, in: Sociolinguistics: Newsletter of the Research Committee on Sociolinguistics 17, 125–139. Günthner, Susanne. 1998. “Polyphony and the ‘layering of voices’ in reported dialogues. An analysis of the use of prosodic devices in everyday reported speech”. InLiSt Interaction and Linguistic Structures 3. http://kops.ub.uni-konstanz.de/bitstream/handle/ urn:nbn:de:bsz:352-opus-4731/473_1.pdf?sequence=1, last accessed 27 Nov 2013. Halliday, Michael A. K. 2003. “On the ‘architecture’ of human language”, in: Jonathan J. Webster (ed.). On Language and Linguistics. Vol. 3 in the Collected Works of M.A.K. Halliday. London: Continuum, 1–29. Herring, Susan. 1999. “Interactional coherence in CMC”. Journal of Computer-Mediated Communication 4/4. http://jcmc.indiana.edu/vol4/ issue4/herring.html, last accessed 2 July 2006. Hübler, Axel and Wolfram Bublitz. 2007. “Introducing metapragmatics in use”, in: Wolfram Bublitz and Axel Hübler (eds). Metapragmatics in Use. Amsterdam: John Benjamins, 1–26. Kienpointner, Manfred. 1997. “Varieties of rudeness”. Functions of Language 4, 251–287. Kleinke, Sonja. 2010. “Interactive aspects of Computer-mediated Communication – ‘disagreement’ in an English and a German public news group”, in: Sanna-Kaisa Tanskanen, Marja-Liisa Helasvuo, Marjut Johansson and Mia Raitaniemi (eds). Discourses in Interaction. Amsterdam: John Benjamins, 195–222. Kleinke, Sonja. 2012a. “Metonymic inferencing and metonymic elaboration in quotations: creating coherence in a public internet discussion forum”, in: Sonja Kleinke, Zoltán Kövecses, Andreas Musolff and Veronika Szelid (eds). Cognition and Culture. The Role of Metaphor and Metonymy. Budapest: Eötvös University Press, 87–98. Kleinke, Sonja. 2012b. “Responses to rhetorical questions in English and German public news groups in the Internet”. Functions of Language 19/2, 174–200. Kleinke, Sonja and Birte Bös. (2015). “Intergroup rudeness and the metapragmatics of its negotiation in online discussion fora”, in Miriam Locher, Brook Bolander, Nicole Höhn (eds), Relational Work in CMC, special issue, Pragmatics 25: 1, 47–71. Kurz, Josef. 2000. “Die Redewiedergabe”, in: Josef Kurz, Daniel Müller, Joachim Pötschke and Horst Pöttker (eds). Stilistik für Journalisten. Wiesbaden: Westdeutscher Verlag, 173–214. Langacker, Ronald. 1999. Grammar and Conceptualization. Berlin: de Gruyter. Leech, Geoffrey N. and Mick H. Short. 1981. Style in Fiction. A Linguistic Introduction to English Fictional Prose. Harlow: Pearson Education Limited. Marcoccia, Michel. 2004. “Online polylogues: conversation structure and participation framework in Internet newsgroups”. Journal of Pragmatics 36, 115–145. Panther, Klaus-Uwe and Linda Thornburg. 2004. “The role of conceptual metonymy in meaning construction”. metaphorik.de 06, 91–116. Panther, Klaus-Uwe and Linda Thornburg. 2007. “Metonymy”, in: Dirk Geeraerts and Hubert Cuyckens (eds). Oxford Handbook of Cognitive Linguistics. Oxford: Oxford University Press, 236–263.

96   

   Birte Bös and Sonja Kleinke

Radden, Günter and Zoltán Kövecses. 1999. “Towards a theory of metonymy”, in: Klaus-Uwe Panther and Günter Radden (eds). Metonymy in Language and Thought. Amsterdam: John Benjamins, 17–61. Roulet, Eddy. 1984. “Speech acts, discourse structure, and pragmatic connectives.” Journal of Pragmatics 8, 31–47. Schütte, Wilfried. 2000. “Sprache und Kommunikationsformen in Newsgroups und Mailinglisten”, in: Werner Kallmeyer (ed.). Sprache und neue Medien. Berlin: de Gruyter, 142–178. Schwarz-Friesel, Monika. 2007. “Indirect anaphora in text: a cognitive account”, in: Monika Schwarz-Friesel, Manfred Consten and Mareile Knees (eds). Anaphors in Text. Amsterdam: John Benjamins, 3–20. Schwarz-Friesel, Monika and Manfred Consten. 2011. “Reference and anaphora”, in: Wolfram Bublitz and Neil R. Norrick (eds). Foundations of Pragmatics. Handbooks of Pragmatics, vol. 1. Berlin: de Gruyter, 347–372. Schwenter, Scott A. 2000. “Viewpoints and polysemy: Linking adversative and causal meanings of discourse markers”, in: Elizabeth Couper-Kuhlen and Bernd Kortmann (eds). Cause – Condition – Concession – Contrast: Cognitive and Discourse Perspectives. Berlin: Mouton de Gruyter, 257–281. Severinson Eklundh, Kerstin and Clare Macdonald. 1994. “The use of quoting to preserve context in electronic mail dialogues”. IEEE Transactions on Professional Communication 37/4, 197–202. Severinson Eklundh, Kerstin. 2010. “To quote or not to quote: setting the context for computermediated dialogues”, Language@Internet, 7/5. http://www.languageatinternet.org/ articles/2010/2665, last accessed 12 Sep 2011. Short, Mick H. 1988. “Speech presentation, the novel and the press”, in: Willie Van Peer (ed.). The Taming of the Text. London: Routledge, 61–81. Short, Mick H., Martin Wynne and Elena Semino. 1997. “Using a corpus to test a model of speech and thought presentation”. Poetics 25, 17–43. Stirling, Lesley. 1996. “Metonymy and anaphora”. Belgian Journal of Linguistics 10, 69–88. Tajfel, Henri. 1981. Human Groups and Social Categories. Cambridge: Cambridge University Press. Ungerer, Friedrich and Hans-Jörg Schmid. 2006. An Introduction to Cognitive Linguistics, 2nd ed, Harlow: Pearson.

Anita Fetzer and Elisabeth Reber

4 Quoting in Political Discourse: Professional Talk Meets Ordinary Postings Abstract: This paper examines the different forms and functions of quotations in political discourse, comparing their instantiations in a corpus of political TV interviews and online comments on political interviews which are reproduced on the same website. Its methodological framework is an integrated one, feeding on pragmatics and interactional linguistics. Quoting is seen as a context-dependent, multi-layered communicative practice which can be realised in various linguistic forms, and those forms can have multiple functions. As regards their form, the quotations in the interviews comprise (1) the quoter, (2) the ascribed source of the quoted stretch of discourse, (3) a verb of communication or possible functional synonym which is assigned the status of quotative, and (4) the quoted. The quotations in the online comments show a more telescoped format, generally lacking a verb of communication. As regards their function, they are deployed to express alignment or disalignment with political positions and policies, but at the same time are used to enact particularised participation frameworks. Keywords: quotation, dis/-alignment, argumentation, political discourse, participation framework, perlocutionary effect

1 Introduction¹ This paper analyses the practice of quoting in the context of political discourse in the traditional media (TV interviews and panel interviews) and in the new media (online discussion forums), giving particular consideration to the question of how quotations are used strategically to construct, deconstruct and reconstruct ideological coherence of political agents and the party they represent in the media and across the media. To capture the inherent dynamics of mediated and medialised political discourse, the practice of quoting is examined from a pragmatic perspective, accounting for implicatures and perlocutionary effects, and from a formal perspective, describing the specific linguistic formatting of the quotations 1 This analysis was presented at the “3rd International Conference on Quotation and Meaning – Quoting Now and Then” (19–21 March 2012). Thanks are due to the participants of the conference and to the editors and reviewers for their valuable input.

98   

   Anita Fetzer and Elisabeth Reber

in the two data sets. Further attention is given to sequential aspects, considering in particular the negotiation of the validity of quotations, which comprises the quoter and source and their participant roles. The contrastive approach allows us to provide a fine-grained micro analysis of the context-specific distribution of the linguistic marking of self- and other-quotation, and demonstrates the communicative function and enactment of discursive frames across the two domains (interviews and online discussion forums).

1.1 Political Discourse The interviews and online discussion forums under investigation can be attributed to the wider field of political discourse. Political discourse is a multifaceted notion, which comprises discourse about politics on the one hand, and discourse by politicians on the other. However, not every instance of discourse produced by one or more politicians may count as political discourse, just as not every discourse about politics may count as political discourse either. What are the necessary and sufficient conditions for a stretch of discourse to count as political discourse? Political discourse in our westernised mass democracies is public discourse, it is institutional discourse, and it is – to a large extent – professional discourse. This is because the traditional concept of politics is anchored to the public domain of society. Since political discourse is also mediated and medialised discourse (Fairclough 2006), it interfaces with micro domains, entering the private spheres of life through the media, and in medialised mass democracies, this is the only way in which most people ever encounter politics (cf. Fairclough 1995, Fetzer/Lauerbach 2007, Fetzer/Weizman/Reber 2012). Political discourse in the media is a complex phenomenon: it is institutional discourse, it is media discourse, and it is mediated and medialised discourse. As institutional discourse, it differs from everyday conversation in that it is subject to institutional constraints. As media discourse, it is different from other types of institutional discourse by being, above all, public discourse addressed to a mass media audience. As mediated and medialised political discourse, it is the outcome of the encounter between two different institutional discourses: (1) those of politics and the media, feeding on the inherent constraints of mediated discourse, i.e. the communication through a medium and thus the uncoupling of space and time, and the movement of meaning from one text, discourse or event to another, with the constant transformation of meanings, and (2) of medialised discourse, i.e. the professionalisation of politics and management of the mediation of political ‘messages’ by spin doctors or political branding, for instance (Fairclough 2006).

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   99

Political discourse uses various discourse genres, such as statements and speeches in the contexts of election campaigns, summit meetings, business meetings or party conferences; interviews in the context of TV or printed media; and multi-party discourse in the context of panel interviews and parliamentary debates. Political discourse may also employ reports, analyses, commentaries, editorials or letters to the editor, to name but the most prominent ones. All of these discourse genres are employed strategically to ‘talk politics’ and to talk about politics. Moreover, these discourse genres do not occur in a void but rather are embedded in journalistic news discourse and may be repeated as sound bites in later programming. Ordinary people may also participate in mediated political discourse in audience participation programmes, such as panel interviews, standing in for the interviewer and asking questions, or by members of the home audience, calling in or sending emails. Most recently, the evolution of the internet has brought about new forms of communication and opened up new arenas for political discourse, e.g. social networks (Kushin/Kitchener 2009), online discussion forums, twitter or blogs (cf. Atifi/Marcoccia 2012 for an overview of the different types of computer-mediated political discourse). As with all new forms of discourse, this technical development has led to the formation of context- and culture-specific conventions of language use in the different kinds of mediated and medialised political discourse, as is the case with quotations, for instance (cf. Buchstaller et al. 2010, Johansson 2007). Against this backdrop, the present paper provides a contrastive analysis of the forms and functions of quotations in two types of political discourse: pre-election interviews and online discussion forums.

1.2 Data Our data consist of a corpus of pre-election interviews broadcasted by the BBC and ITV (1997–2003) and of news blogs/sites which contain written transcripts of TV interviews and discussion forums, which feature the users’ responses towards the interviews (2008/2010). The televised data comprises 20 interviews with the leaders of the major political parties. These data were recorded and transcribed according to interactional principles, adhering as closely as possible to the speakers’ wording of their conversational contributions and to the turn-taking mechanism (cf. Fetzer/ Bull 2006, Fetzer/Johansson 2010). This corpus was manually tagged with regard to the relevant linguistic structures and then hand-counted in order to capture subtle aspects of analysis. As the focus of the investigation lies on the verbal forms and functions of quotations in their local linguistic context, the transcription presented here adheres to orthographical standards to facilitate readability.

100   

   Anita Fetzer and Elisabeth Reber

The collection of computer mediated discourse (CMD) consists of a total of 101 comments, posted in two online discussion forums; one featured by Lynn Sweet’s blog on the web site of Chicago Sun-Times, which offers the transcript of the Obama Interview with Bill O’Reilly (2008) broadcasted on Fox News and the subsequent comment section (87 comments),² and the other by a page on the news site msnbc.com, where the transcript of an NBC News Obama Interview (2010) with Brian Williams was made available in the form of a video clip and transcript (14 comments).³ In spite of the different members’ labels used for the online discussion forums, viz. “comments” and “discuss this post”, both constitute similar communities of practice (Wenger 1998) in that they show a “sustained pursuit of a shared enterprise” (ibid, 45), namely the expression of alignment and disalignment with the conversational contributions in the interview and in other comments.⁴ Furthermore, the communicative situation may be described as non-synchronous, one-to-many, public and – at least to some extent – anonymous. While users have to give their email addresses when registering for these discussion forums, a nickname suffices for identification on this web interface. In this respect, the comments examined are more anonymous than letters to the editor, where the full name and address needs to be supplied.

1.3 Methodology The methodological frame of reference adopted in this research is informed by compositionality, drawing on the pragmatic premises of intentionality of communicative action and conversational implicature, and on the interactional-linguistic premises of language as a socially situated form, the unit of investigation of speech activity, and linguistic variation, i.e. variations and alterations are not random or arbitrary but communicatively functional and meaningful (Gumperz 1992). The starting-point of the study was the observation that participants deploy quotations in the online comments in order to refer to the interview, which is 2 http://blogs.suntimes.com/sweet/2008/09/foxs_bill_oreilly_obama_interv.html, last accessed 18 April 2012. 3 http://www.nbcnews.com/id/38907780/ns/nbcnightlynews/t/full-transcript-obama-interview-nbc-news/#.UTRpvFf09Vs, last accessed 18 April 2012. 4 Online discussion forums may not only be hosted by news blogs/sites but also by other potentially non-political websites, e.g. YouTube (Jones/Schieffelin 2009). What they have in common is that they enable the user to respond to specific web content on a web site and other users’ posts on the same site.

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   101

reproduced on the same website. This means that the original wording of what is presented as a quotation can be retrieved directly from the shared discursive context of the comment. This contrasts with political interviews and even more so with everyday interaction, where the source of quotation may not be retrievable for all conversational parties in the same way. Nevertheless, the quotable (that is, the chunks of speech to be quoted in the comments) and the quotation are enacted in different discursive frames in the websites examined (Goffman 1986, cf. Fetzer 2006 for a model of frames of reference in political interviews). This is reflected in Goffman’s seminal work on “Frame Analysis” (1986), especially his notion of framing and reframing, as well as in Burger’s (2005) analysis of the media-specific use of language in media interviews, and by Heritage’s research on the media-specific use of back-channels in interviews (Heritage 1985). Fetzer’s investigation of the obligatory references to the media frame in the opening and closing sections of an interview as well as the strategic use of references to the media frame in the topical sections also explores how the quotable and quotation are enacted in different discursive frames (Fetzer 2006). They have all shown that the interviews between interviewer (IR) and interviewee (IE) are not actually directed towards the direct communication partner, i.e. IR and IE, but rather towards the media audience. The multi-layeredness of media communication is captured best by a multiframe discourse scenario, as is illustrated in Figure 1 below, and exemplified with the following exchange between the IR Jonathan Dimbleby (JD) and the IE Tony Blair (TB), which provides the local context of excerpt 1a analysed and discussed in section 2.1. The first-frame interaction comprises the face-to-face interaction between JD and TB (1), which is represented by the TV broadcast (1a); this faceto-face interaction between the first-frame participants may be further supplemented by an optional internet video ((1b)) and a transcript ((1c)) available on the TV station’s website. In excerpt 1, the IE asks the IR if he is expected to elaborate on a discourse topic which they have been discussing, targeting the first-frame interaction. The IR agrees with the IE and then quotes Peter Mandelson, an ally of the IE, inviting the IE to comment on the quotation. The use of the quotation as well as parts of the IE’s response display a dual orientation: towards the firstframe participants (IE and IR), and towards the media audience. This is reflected in the generalized form of address ‘why not just judge us on what we do’, which is recycled and recontextualised several times:

102   

   Anita Fetzer and Elisabeth Reber

Excerpt 1⁵ 1TB Do you want me to stay [inc]? JD More or less, yes. On- on- on spin, you know, Peter Mandelson said the other day, there’s been too much spin, there ought to be more vision. Do you accept the charge that there’s been too much spinning? TB Well, I- I sometimes think we’re more, sort of, spun against than spinning, in the sense that people have this huge thing about it all. But in the end, why not just judge us on what we do? I mean, judge us on what we do. I’m perfectly happy to be judged on what we do, because I will defend the record that we have. And, you know, you- you talk a lot about presentation and everything. Look, l- l- let’s t- t- try and- er- get- get to the heart of this. In today’s age, it’s a twenty four hour media age. And, you know, news comes in and out and it develops very, very quickly …. (Ask Tony Blair 5 May 2001) To adapt and extend the frame of reference to CMC as illustrated below by Figure 1, the first-frame interaction (1) comprises the studio setting and first-frame participants, i.e. IR and IE of the TV broadcast interview (1a), or alternatively, the internet video (1b) and/or the transcript provided by the website (1c). The second-frame interaction consists of the first frame interacting with the media audience, i.e. audience watching the televised interview or the computer-mediated interview as a video or reading it as a transcript (2). The third-frame interaction comprises the users as audience, commenting on the interview between the firstframe participants on a website provided by the TV channel, or alternatively in some face-to-face interaction (3). The fourth-frame interaction is composed of audience members responding to the third-frame commentators commenting on the interview (4), and the fifth-frame interaction includes those audience members referring to the fourth-frame interaction (5). This multi-frame discourse illustrates the multi-layeredness of CMD which goes beyond the dual frame of reference of traditional media communication:

5 Relevant cues for the analysis are printed in italicised small capitals, and the quoted material is underlined.

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

(1) (1a) ((1b) ((1c) (2) (3) (4) (5) (6)

   103

1st frame: IR + IE TV broadcast internet video) Transcript) 2nd frame: first frame + audience 3rd frame: commentators + first frame 4th frame: commentators + commentators (+ first frame) 5th frame: commentators and readers of comments etc. …

Figure 1: multi-frame discourse scenario

Our paper examines the forms and functions of quoting as context-sensitive instantiations, which are deployed strategically to achieve particular goals in a complex multi-frame discourse scenario. It is structured as follows: the next section examines practices of quoting in discourse and applies the results obtained to an investigation of mediated political discourse, and the final section summarises and concludes the paper.

2 Practices of Quoting in Mediated Political Discourse This section begins with a discussion of quotation in discourse, and applies the results obtained to an examination of the forms and functions of quotation in the political interviews (2.1) and in the online comments (2.2), and ends with a contrastive analysis of the two (2.3). Quotations can be represented overtly, e.g. by quotation marks and/or the use of a verbum dicendi, or non-overtly, i.e. they may not be marked at all as quotations.⁶ Lexico-grammatical devices which introduce what follows as a quotation have been labeled quotatives (e.g. Buchstaller 2013, Buchstaller/van Alphen 2012). According to Mathis and Yule (1994), “the omission of a quotative [i.e. socalled ‘zero-quotatives’]” can be functional in everyday conversation in that it “may be used to achieve a dramatic effect which cannot as easily be achieved

6 In spoken discourse, prosodic cues may also support the identification of reported portions of speech and the staging of narrative figures in reported dialogues (Klewitz/Couper-Kuhlen 1999). Visual gestures, such as so-called air quotes, can further be assigned the status of a pragmatic marker or “contextualisation cue” (Gumperz 1992), showing how current speech is intended by the speaker and how the speaker intends the addressee to interpret it.

104   

   Anita Fetzer and Elisabeth Reber

with the presence of a full quotative form” or even “to reflect the urgency of the interaction being reported” (ibid, 67). Along these lines, Günthner (2005) provides empirical evidence that speakers in everyday conversation may “[remodel] the past text according to the situative communicative intention and [imprint their] perspective onto the reconstructed event” (ibid, 293). This suggests that (especially direct) quotations should be seen as situated products of remodelling and imprinting, serving particular communicative goals in non-mediated discourse (cf. e.g. the contributions to Buchstaller/van Alphen 2012). Furthermore, all types of quotation require the contextualisation and recontextualisation of the quoted, to use Linell’s terminology (1998), or pragmatic enrichment, to employ relevance-theoretic terminology (Sperber/Wilson 1986). In a relevance-theoretic frame of reference, quotation is differentiated along the lines of the fundamental distinction between mention and use: the former does not allow the use of indexicals while the latter allows their use. Quotations generally serve to import context into situated discourse in that they make unbounded context importation bounded by spelling out the source of the quoted material and their contextual embeddedness in regard to its temporal, local and discursive coordinates (cf. Fetzer 2011). The importation of particular contextual information, which is required for the interpretation of the stretch of discourse to be quoted, invokes particular contextual frames, in accordance with which the quoted is interpreted. At the same time, the quoted is assigned the status of quote-worthiness, to adapt a term from Bell (1991), or of quotability, to adopt a term from Clayman (1995), and considered relevant to the ongoing discourse, thus contributing to the negotiation and joint construction of discourse coherence and to the negotiation and construal of discursive common ground (Fetzer 2007). The functions of quotations in CMD are generally similar to those in nonmediated discourse. For instance, Severinson Eklundh (2010) describes “quoting as a way to provide a context or important information relevant to the current message” (ibid, 18). Context importation of this kind can be used strategically as a means of stance-taking in online discussion forums (cf. also Bös/Kleinke this volume, Severinson Eklundh 2010). Bublitz/Hoffmann (2011) observe that direct quotations may serve to simulate dialogicity in CMD: “If multiple direct speeches are presented in adjacent positions, they simulate the turn-taking structure of real-life dialogues” (ibid, 442). The following analysis of the forms and functions of quotations in political discourse (excluding prosodic and visual aspects) is informed by the form-based requirements for quotation, i.e. quoter, source, quoted, and indexical references to the contextual coordinates of the quoted, the communicative functions of quotation examined above as well as the indexical nature of quotation, and adapts them accordingly. In the two types of mediated political discourse under

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   105

investigation, the stretch of discourse which is constructed as quoted speech can generally be retrieved from prior discourse and may be accounted for by the participants in communication, making explicit the source of the quotation and its temporal and local embeddedness.

2.1 Quotations in Political Interviews This section presents the results of a study on the forms and functions of quotation in televised political interviews, paying particular attention to the participation framework with respect to self- and other-quotation. Self- and other-quotations import some prior conversational contribution into a new (or different) discourse by placing it in a different context, thereby recontextualising form, function and content. Inherently connected with the importation of the quoted and its ascribed source is its evaluation, i.e. the quotation is assigned the status of quote-worthiness or quotability, which may imply a possible re-evaluation, such as the re-affirmation of its relevance but also its deconstruction. In discourse in general, and in political discourse in particular, quotations are used strategically to import context into the ongoing discourse and to express the quoter’s attitude towards the ascribed source of the quotation and/or towards its content. However, it is not only the content of the quotation which is of relevance, but also the attitude of the present speaker in her/his role as quoter towards the ascribed source, content, illocutionary force and context. In media discourse, quotations allow the quoter to align and disalign with particularised sets of the mass audience because the quoted is not only directed at the first-frame participants, the face-to-face addressee(s), but also, if not primarily, at the mediated audience, or second-frame participants, as has been illustrated in the discussion of Figure 1 above. Because of these contextual constraints and requirements, the ascribed source of the quotation is generally made explicit, e.g. proper name, affiliation, party programme, or a more indeterminate entity, and communicated to the mass audience. This does not only serve as a means of identification; rather, by making explicit the identity of the source, the present speaker assigns it the status of quote-worthiness, and by the strategic selection of a quotative which makes explicit the illocutionary force of the quoted, the quoter implicates her/ his evaluation of the stretch of discourse to be quoted, as is the case with, e.g., the more neutral verb say and the more specific ones make the point or admit. Moreover, in the political interviews at hand, the quoters generally make their intended perlocutionary effects explicit and assign them the status of an object of

106   

   Anita Fetzer and Elisabeth Reber

talk for both the first-frame participants and the second-frame participants. This may, of course, be taken up by third-, fourth-, and further-frame interactants. In the mediated political interviews, both IR and IE use quotations to challenge their communication partner’s argumentation, if not their credibility. There is, however, quite a difference between the use of self-quotation and the use of other-quotation. The former is employed to strengthen the argument of self or of the political party self represents, as is the case in Excerpts 2, 3, 4, 5, 6 and 7 below. In the interviews, self-quotation tends to be used in particularised contexts, i.e. only when self intends to reassert her/his credibility. The infrequent use of selfquotation in the British media interview data is in line with the modesty principle (Leech 1983). What is more, the function of the self-quotation is not influenced by its formatting as direct or indirect quotation in the data at hand. Other-quotation needs to be differentiated with regard to the status of other: other as an ally, as is the case with Excerpt 1 above and its extracted quotation in 1a below, or other as a political opponent. Both forms of other-quotation tend to be employed to confront the politician with what s/he said at some prior stage in the ongoing interview, or what s/he said or is supposed to have said in some prior interview or speech. The strategic employment of other-quotation, especially recycling what other said/wrote before, is less constrained than the use of self-quotation. It is generally used to challenge a politician’s argumentation, signifying a lack of coherence and/or credibility on her/his side. Excerpt 1a contains an other-quotation formatted as a direct quotation. The IR uses the quotation to import context into the ongoing interview, entextualising⁷ an ally of the IE and assigning the ally (Peter Mandelson, a former spin doctor of the Labour government) the status of an object of talk. The quotative, which represents the illocutionary force of the ascribed source, is neutral (‘say’), and the original local and temporal embeddedness is left underspecified (‘the other day’). However, the quoter’s attitude towards the content (‘too much spinning’) is made explicit and the intended perlocutionary effect (‘charge’) is assigned the status of an object of talk in the meta-comment ‘do you accept the charge that there’s been too much spinning’; through the use of the perfective aspect, the IR extends the quoted’s domain of validity to include present time:

7 The use of ‘entextualisation’ in this contribution differs from that promoted by Park/Bucholtz (2009) who define entextualisation primarily in terms of institutional control and ideology. It shares their stance of approaching entextualisation in terms of “conditions inherent in the transposition of discourse from one context into another” (ibid, 489), while accommodating both local and global contexts, and their dialectical connectedness.

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   107

Excerpt 1a⁸ IR Peter Mandelson said the other day, there’s been too much spin, there ought to be more vision. Do you accept the charge that there’s been too much spinning? (Ask Tony Blair 5 May 2001) Other-quotations are not only anchored to particularised allies or political opponents, they can also be attributed to indeterminate sources. Excerpt 2 formatted as a direct quotation is such a case with the ascribed source ‘people’ and the neutral quotative ‘say’. The quoted material displays features of spoken language (‘well look here’). As above, the other-quotation is used strategically to underline the quoter’s argument, which is a constitutive part of an argumentative chain, illustrating the politician’s point. However, the intended perlocutionary effect of the quotation is not made fully explicit and it is not assigned the status of an object of talk, as was the case with the IR’s use of other-quotation: Excerpt 2 IE Well I want to change it, as Anne Widdecombe has set out, I I want to change this because it has a counter-productive effect. At the moment if a lorry driver discovers that people have got on to his lorry and then that, says that to the police - he gets penalised, he gets fined. Erm one firm has got fined many tens of thousands pounds as a result, so that needs changing, erm that is not an effective rule, we need to reform it so that there is a penalty for knowingly abusing the system, but that people who say well look here, there’s an abuse of system going on, I’ll tell you about it, they should not be penalised. (Ask William Hague 30 May 2001) In mediated political interviews, unlike the quoter’s interactional role, the formatting of other-quotation as direct or indirect quotation does not have an impact on their function: IRs generally spell out the intended perlocutionary effects of the quotation and assign them the status of an object of talk, while IEs use otherquotations to support their argument. Self-quotations tend to be used by politicians only and other-quotations are used by both IRs and IEs. While an IR may entextualise quotable material from particularised allies or particularised politi-

8 The focus of the analysis is on the verbal form and function of quotation. To facilitate readability, the transcription mode employed here adheres to orthographic conventions. Relevant cues for the analysis are printed in italicised small capitals, the ascribed source and quotative are printed in small capitals, and the quoted material is underlined. In the data analysis, the use of personal pronouns is in accordance with natural gender.

108   

   Anita Fetzer and Elisabeth Reber

cal opponents in order to unbalance the politician and challenge their credibility and/or argumentative coherence, the IE generally entextualises quotable material from allies or political opponents to support their argument. Both IRs and IEs may also use other-quotation with an indeterminate source in order to illustrate and underline their argument; frequently the use of such quotations is a constitutive part of an argumentative chain. Excerpts 3 and 4 are instances of self-quotation introduced by collective selfreference with a neutral quotative (‘we say’) in the context of clarifying the position taken by the Conservatives in (3) and in illustrating the Conservatives’ position in (4). Excerpt 5 is an instance of individual-anchored self-quotation with neutral ‘say’ used in a parasitic manner exploiting the ‘I-say-you-say’-language game. In all of the excerpts the quoter imports context into an ongoing argumentative sequence and assigns the entextualised context the status of an object of talk which he uses to underline his argument: Excerpt 3 IE no, no we say there are other changes to tax that we would like to make but that that depends on the state of the economy (Ask William Hague 30 May 2001) Excerpt 3 contains a direct quotation in a negatively loaded context, which is used strategically to refute the argument of the IE and at the same time support the politician’s argument. This is reflected in the doubling of the negative operator ‘no no’ in the initial position of the turn, which refutes the IR’s previous contribution (“After that you in the manifesto, Michael Portillo in interviews say there will be further cuts in the second part of that parliament”) carrying the loaded presupposition that the Conservatives intend to cut taxes further. The refutation is not mitigated by any attenuation devices and indicates that the IE does not intend to discuss the matter any further. Excerpt 4 IE Well let’s look at some of those things, we say we’ll save one billion pounds from welfare fraud. The Government have admitted that there are seven billion pounds in welfare fraud going on in this country, and they haven’t done anything about it, and we’re saying what you can do about it, you can give benefit inspectors the same powers as tax inspectors so we would be able to tackle a bit of the fraud in the welfare system. (Ask William Hague 30 May 2001)

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   109

Excerpt 4 contains an orchestrated interplay of direct and indirect quotations in an argumentative sequence. This contributes to the dialogicity of the argumentation as in the ‘I-say-you-say’ language game, underlining the IE’s argument. While the collective self-quotation formatted as a direct quotation is part of an affirmative context (‘we’ll save one billion pounds’), the other-quotation is an instance of an indirect quotation and is part of a negatively loaded context framed by the highly negatively loaded quotative (‘have admitted’), and the negative meta-comment (‘they haven’t done anything’). The negative other-evaluation is juxtaposed with a more positive self-quotation framed by a neutral quotative and the presentation of the solution to the problem under discussion (‘we’re saying what you can do about it’), which is elaborated on. Excerpt 5 IE Well that is why I say you’ve got to respect the fact that people want to migrate economically. They want to to do better for themselves and their families. But no-one believes that you can have unrestricted of movement of populations around the world, obviously we can’t do that. And that’s why we have to have rules. And all I’m saying is the rules have to be firm and fairly enforced. (Ask William Hague 30 May 2001) Excerpt 5 contains another self-quotation used parasitically in the ‘I-say-you-say’ language game, which provides backing for the politician’s argument (‘well that is why I say you’ve got to respect the fact that …’). The next two excerpts, 6 and 7, are instances of negotiated quotations, that is, the content and force of the quoted are not accepted by the communication partner and the quoter needs to account for the ascribed source and context. Excerpt 6 IR1 You see the the last week you said that you would face the no-vote in any referendum, because of what you’ve just described, your allegations about how it would be, almost certain defeat. Now you say there’s a fighting chance IE1 no no, I I what I said was that they would have, they would have a strong chance of being able to win. IR2 No no you did more than that forgive me IE2 No I did not say, did not say almost certainly IR3 It was on a programme that you did at this time last Wednesday night you said, you said well almost certainly, the government will almost certainly win the referendum. Do you withdraw the ch, the

110   

   Anita Fetzer and Elisabeth Reber

the the view that you expressed then that they would almost certainly, the precise words you used, win the referendum. (Ask William Hague 30 May 2001) In Excerpt 6, the multilayered status and embeddedness of the direct and indirect quotations surface in the argumentative sequence IR1 – IR3. The IR challenges the argumentation of his communication partner with the indirect quotation (‘you would face the no-vote in any referendum’) specified by its temporal (‘last week’) coordinates, contrasting it with a statement just (‘now’) made public, formatted as direct quotation (‘there’s a fighting chance’) with the neutral quotative ‘say’. The IE refutes the IR’s direct quotation ‘there’s a fighting chance’ in IE1 (‘no no I I what I said was that [corrected quotation]’), which is refuted by IR2 (making explicit the generalised communicative action (‘do’) supplemented by an explicit reference to the politician’s face wants (‘forgive me’). IE2 refutes this directly (‘no I did not say did not say ‘almost certainly’), spelling out the non-said. To substantiate his challenge, IR3 spells out the location of the quoted (‘on a programme’) and temporal embeddedness (‘this time last Wednesday night’). The degree of explicitness as regards illocutionary force and content is very high, and the status of the quoted, which has been controversial at some stage, is made explicit by ‘the precise words you used’. The negotiation of the validity of the quotation about the Labour government winning the referendum gets more and more controversial and is only resolved when the IR invites a member of the audience to take on his role (‘Let’s bring the audience in this. Right down in the front here’). Excerpt 7 is another instance of a negotiated quotation, in which the contextual coordinates are also made explicit to underline the validity of the quoted material: Excerpt 7 IR1 But you’re also the only party leader who says, as you said to meIE1 Indeed I did. IR2 not so long ago, erm, when I asked you whether users of cannabis were criminals, you said, I don’t regard them as criminals. And you say – I’m right, aren’t I? – you don’t regard them as criminals. IE2 I- I- that’s what I said to you, in a- in another studio, in an equivalent programme some time ago, that is my personal view. It is not the position of the Liberal Democrats, let me be quite clear about this (Question Time Special – challenge the leader 17/05/2001)

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   111

Excerpt 7 is quite interesting as it shows that it is not only the content or pragmatic force of the quoted which is of relevance to importing context into an ongoing discourse, but also the footing of the ascribed source, that is either animator, principal or author (Goffman 1981). It also displays multi-layered and multi-frame quoting with a complex procedure of ascribing the source of the quoted, that is the face-to-face IE, as a constitutive member of the first-frame interaction (‘you’re also the only party leader who says’). Moreover, the illocutionary force of the quoted material is made explicit (‘asked’), as is its status of the yes/no-question (‘whether’). The multi-layered status of the interaction is further explicated by providing the IE’s answer, formatted as mixed quotation (‘you say – you don’t regard them as criminals’). The linguistic realization of the quotation is also remarkable, switching between past tense (e.g., ‘as you said to me’; ‘indeed I did’; ‘you said’, ‘I said’) and past time (e.g., ‘not so long ago’; ‘in another studio’) and present tense (e.g., ‘you’re also the only party leader who says’; ‘you say’; ‘that is my personal view’; ‘it is not the position of the Liberal Democrats’) and present time (e.g., ‘I’m right, aren’t I?’), displaying all of features necessary to count as the (conversational) historic present (cf. Montgomery 2007, Schiffrin 1981). As in Excerpt 6, the participants format their contributions with a high degree of explicitness, and both agree upon content, force and temporal and local embeddedness of the quotations. What is controversial, however, is the footing of the ascribed source in Excerpt 7. The IE claims that he made the statement in his role as author of the quoted (‘personal view’), in Goffman’s terms (1981), but not as principal, i.e. the leader of the Liberal Democrats, on whose behalf he did not make that statement (‘it is not the position of the Liberal Democrats’). As above, the IR uses the other-quotation to challenge the IE’s argumentative coherence and credibility. Quotations in political interviews can be used to boost the force of an argument, thus supporting the interlocutor’s argumentation, and they can be used non-supportively, challenging the coherence and credibility of the other (cf. Fetzer, Weizman and Reber 2012). Quotations in the traditional media tend to be introduced by a neutral quotative, mainly ‘say’, and their ascribed source and temporal and local embeddedness are generally made explicit. While the source needs to be more determinate, time and place can be left underspecified.

2.2 Quotations in Online Comments Contrary to the quotations in the political interviews examined above, the quotations found in the news discussion forums are only rarely prefaced by an explicit reference to their ascribed source and a verb of communication. Excerpt 8 is one

112   

   Anita Fetzer and Elisabeth Reber

of the few examples which contain the ascribed source of the quoted, a verb of communication, and the quoted.⁹ Excerpt 8 a. transcript (1) President Obama: ((5 lines omitted)) A year ago today, we were still losing jobs, (2) we’re now gaining them. The economy was still contracting, it’s not expanding. It’s (3) not happening as fast as people would like. But it’s moving in the right direction. (4) And the thing we can‘t do is to try to go back to the same policies that created this (5) mess in the first place. b. comment (1) Transparency At Its Best (2) Obama said, “A year ago today, we were still losing jobs, we’re now gaining them. (3) The economy was still contracting, it’s not expanding. It’s not happening as fast as (4) people would like. But it’s moving in the right direction. And the thing we can’t do (5) is to try to go back to the same policies that created this mess in the first place.” ((comment continues)) In this comment, the ascribed source of the quoted is named (‘Obama’) and followed by a general verb of communication (tensed form of ‘say’, simple past) and the direct quotation, marked by double quotation marks. Instead of a verbum dicendi (verb of communication) users may use a nomen dicendi (that is, a noun, such as summary, announcement) to refer to a forthcoming quotation.¹⁰ Consider Excerpt 9:

9 The portions in italics are quotables. The direct quotations in the comments are underlined, the ascribed source and verb/noun of communication are printed in small capitals. 10 Unlike Halliday’s notion of ‘verbiage’ (Halliday/Matthiesen 32004, 294), which refers to what is said, our term ‘nomen dicendi’ refers to the action performed by an utterance.

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   113

Excerpt 9 a. transcript (1) MR. O’REILLY: Who is the enemy? (2) SEN. OBAMA: Al Qaeda, the Taliban, a whole host of networks that are bent on (3) attacking America, who have a distorted ideology, who have perverted the faith of (4) Islam. So we have to go after them. b. comment (1) M. L. Russell | September 5, 2008 4:18 AM | Reply (2) The Senator’s synopsis of the “enemy” as being those “…who have perverted the (3) faith of Islam” is of concern. ((comment continues)) In this comment, the ascribed source of the quoted is the ‘Senator’s’, with the noun of communication being ‘synopsis’. Two portions of the utterance are marked as quotations of the Senator’s contributions in the interview: ‘enemy’ and ‘…who have perverted the faith of Islam’. However, as becomes visible in the transcript, ‘enemy’ was produced as part of the interviewer’s question which elicited the Senator’s contribution. The commentator thus provides what may be seen as a shorthand presentation of the contents of the interview. More frequently, there is no introductory verb of communication and no reference to the ascribed source of the quoted. Consider Excerpt 10: Excerpt 10 a. transcript (1) MR. O’REILLY: So you’re going to pull out and let the Islamic fundamentalists (2) take them over? (3) SEN. OBAMA: No, no, no, no. What we say is, look, we’re going to provide them (4) with additional military support targeted at terrorists, and we’re going to help (5) build their democracy and provide the kinds of funding –

114   

   Anita Fetzer and Elisabeth Reber

b. comment (1) nelly | September 5, 2008 7:01 AM | Reply (2) The bottom line: (3) Obama: (4) I’ll fight Iran by diplomatic means. I would never take a military option off the (5) table, but unfortunately haven’t any plains yet. (6) What a CHANGE!!!! (7) In Afghanistan “we’re going to provide them with additional military support (8) targeted at terrorists, and we’re going to help build their democracy and (9) provide the kinds of funding –“ (10) The same things that we’re doing now! (11) What a CHANGE!!!! (12) And by all means I’ll never admit that I was wrong about the surge in Iraq. (13) That’s also sound familiar doesn’t it? Not a CHANGE at all !!!! While the quotable (lines 3–5 of the transcript) is presented as a collective selfquotation by Obama, the comment seems to limit the source of the quoted to Obama alone (line 3 of the comment). Moreover, the status of the direct quotation in lines 7–9 can only be inferred from the context since there is no explicit verb of communication which links Obama to the quoted portions in the comment: Cues that contextualise lines 7–9 as a quotation are 1) the use of the quotation marks, and 2) their contextual placement in the discussion forum below the transcript.¹¹ An example from the fourth-frame interaction (commentators + commentators (+ first frame)) shows a division of labour between a full-fledged marking (including a verb of communication and a reference to the ascribed source) on the one hand and the use of quotation marks on the other. Excerpt 11, which shows a sequence of comments, exemplifies this:

11 Note that what is presented as an allegedly direct quotation by Obama in lines (4)-(5) of the comment constitutes reported speech which seems to be fictitious, i.e. it was very likely made up by the user in order to summarise Obama’s position from her/his point of view.

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   115

Excerpt 11 a. transcript (1) MR. O’REILLY: Who is the enemy? (2) SEN. OBAMA: Al Qaeda, the Taliban, a whole host of networks that are bent on (3) attacking America, who have a distorted ideology, who have perverted the faith of Islam. (4) So we have to go after them. b. comment (1) M. L. Russell | September 5, 2008 4:18 AM | Reply (2) The Senator’s synopsis of the “enemy” as being those “…who have perverted (3) the faith of Islam” is of concern. The true faith of Islam that follows the Koran (4) will always be perverted - and political correctness will not protect America (5) against a religion of hate. ((comment continues)) c.

comment (1) F. Atalay | September 5, 2008 6:28 AM | Reply ((7 lines omitted)) (9) As for the poster who calls Islam a religion of hate; thank you for making my point (10) regarding Republicans and their preconceived notions…

d. comment (1) Connie | September 7, 2008 3:32 PM | Reply (2) True Islam is NOT a “religion of hate” and anyone who thinks so needs to go read the (3) Koran and speak to an American Muslim about his practice. ((comment continues)) Comment 11c shows the full-fledged structure: it contains the source of the quoted ‘poster’ (note that the reference to ‘poster’ rather than to the user’s name may be treated as a more impersonal choice which creates more social distance), a verb of communication (‘calls’) and the quoted (‘religion of hate’), which is, however, not marked off as such through quotation marks. This contrasts with comment 11d, where the quoted is again taken up in a follow-up comment, this time marked exclusively by quotation marks. The question why comments c and d show this variation may be explained by interactional reasons (comment 11c is posted in

116   

   Anita Fetzer and Elisabeth Reber

close temporal proximity (about 2 hours later) as opposed to comment 11d (about 2 days later) or by the complex-transitive verb complementation pattern in which the quotation is embedded in 11c.¹² What is crucial here, however, is that the quoted portions are indexed in both comments by conventional devices of the written medium, which have the same function as quotatives. Because of this function, quotations which are only indexed by quotation marks in online discussion forums can hardly be classified as cases of zero-quotatives. But why is the use of quotation marks sufficient here? The writers of these posts rely on the assumption that the material quoted from the first- and fourthframe contributions represents a shared discursive context which may be ’reactivated’ through a minimal set of contextualisation cues. Moreover, the formatting of the quotations suggests that these comments (and the quotes used there) are designed for recipients who treat the interview and the computer-mediated discourse which evolves from it as a multi-framed and shared discourse. Turning to the discourse functions of quotations in online discussion forums, we will now discuss the role of direct quotations in users’ stance-taking towards the interview and as a resource for enacting frames. To begin with the former, the major function of the online discussions forums we examined is the evaluation of the style and contents of the given interview. Similar to the interviews analysed, quoting (both direct and indirect) in the online comments has the communicative function of stance-taking and positioning. Users import context (Fetzer 2011) from the first-frame into the fourth-frame interaction (thus constructing a third frame) by quoting. They thereby construe their object of stance (Du Bois 2007) visibly to their audience, that is, the other participants in the fourth frame and the readers of the comments. When presenting a quotation in their comments, users create an object of stance with which they may align or disalign, positioning themselves in relation to that object (cf. Du Bois 2007). Consider Excerpt 11’: Excerpt 11’ a. transcript (1) MR. O’REILLY: Who is the enemy? (2) SEN. OBAMA: Al Qaeda, the Taliban, a whole host of networks that are bent on (3) attacking America, who have a distorted ideology, who have perverted the faith of Islam. (4) So we have to go after them.

12 The latter point was raised by an anonymous reviewer.

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   117

b. comment (1) M. L. Russell | September 5, 2008 4:18 AM | Reply (2) The Senator’s synopsis of the “enemy” as being those “…who have perverted (3) the faith of Islam” is of concern. The true faith of Islam that follows the Koran (4) will always be perverted - and political correctness will not protect America (5) against a religion of hate. ((comment continues)) In the transcript (the first-frame interaction), the IE (Obama) makes negative attributions about Al Qaeda and the Taliban (‘a whole host of networks that are bent on attacking America, who have a distorted ideology, who have perverted the faith of Islam’): he constructs them as objects of stance from which he disaligns. In the subsequent discussion, the commentator imports chunks of these attributions (marked as direct quotations through double quotation marks) into her/his comment. The shared discursive context of the first-frame interview, to which all comments make reference, is thus bounded (Fetzer 2011). What is more, the quoted attributions are in turn embedded in an evaluative move: presenting them in subject position (‘The Senator’s synopsis of the “enemy” as being those “…who have perverted the faith of Islam”’), the user continues with an evaluative expression in a copulative construction (‘is of concern’). The imported objects of talk are transformed from first-frame attributions to third-/fourth-frame objects of evaluation. Moreover, the user displays disalignment with his/her object of stance: in this way they disalign themselves from the first-frame attributions. The data show that these kinds of processes involving the context-importation and transformation of an object of evaluation can also be observed between fourth-frame participants. The evaluative label/attribution ‘religion of hate’ coined there is imported by subsequent comments and becomes an object of stance in its own right. Compare comments 11’c and d: Excerpt 11’ c. comment (1) F. Atalay | September 5, 2008 6:28 AM | Reply ((7 lines omitted)) (9) As for the poster who calls Islam a religion of hate; thank you for making my point (10) regarding Republicans and their preconceived notions…

118   

   Anita Fetzer and Elisabeth Reber

d. comment (1) Connie | September 7, 2008 3:32 PM | Reply (2) True Islam is NOT a “religion of hate” and anyone who thinks so needs to go read the (3) Koran and speak to an American Muslim about his practice. ((comment continuer)) The analysis has illustrated that evaluative attributions such as e.g. ‘enemy’ and ‘religion of hate’ may be imported from one interactional frame into another and be rendered objects of stance themselves. In this sense, these quotations may be described as multilayered constructions: a chunk of discourse is imported from one interactional frame (here the first-frame interview transcript) into another one (the fourth-frame interaction between commentators) through practices of direct quoting. What may constitute an evaluative expression for stance-taking and positioning in the first frame becomes an object of stance in the fourth frame and thus an object for alignment and disalignment. Nevertheless, users may not always make their stance and positioning so explicit: Excerpt 10’ is an example of a comment where the author’s stance is expressed in an ironic fashion: Excerpt 10’ b. comment (1) nelly | September 5, 2008 7:01 AM | Reply (2) The bottom line: (3) Obama: (4) I’ll fight Iran by diplomatic means. I would never take a military option off the table, but (5) unfortunately haven’t any plains yet. (6) What a CHANGE!!!! (7) In Afghanistan “we’re going to provide them with additional military support targeted at (8) terrorists, and we’re going to help build their democracy and provide the kinds of funding –“ (9) The same things that we’re doing now! (10) What a CHANGE!!!! (11) And by all means I’ll never admit that I was wrong about the surge in Iraq. (12) That’s also sound familiar doesn’t it? Not a CHANGE at all !!!!

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   119

The policy quoted (lines 7–8) was proposed by Obama in an interview which took place during his 2008 election campaign. The quoted is first assessed as ‘The same things that we’re doing now!’ (line 9), that is, as the same policy pursued under the Republican government at the time when the interview was conducted. It is then re-assessed as ‘CHANGE’ (line 10), i.e. the opposite of what was stated in the previous line. By alluding to the slogan “change” from Obama’s 2008 campaign and stating the opposite of what is meant, line 10 comes off as ironic. Furthermore, quotations do not only have an important function in stancetaking but are also used as a resource for enacting discursive frames and thus particular participation frameworks. The online comments exclusively contain other-quotations with the ascribed source of the quotation framed in third-person constructions. Straightforward examples are for instance: Excerpt 8’ comment Obama said, “A year ago […]” Excerpt 11’ comment c As for the poster who calls Islam a religion of hate; […] Excerpt 8’ is exemplary for cases where the user quotes what was said in the firstframe interaction without addressing the IR or IE, i.e. the sources of the quoted, in the first frame. Excerpt 11c’ shows an other-quotation taken from another comment, thus serving as a resource to address both the producer of the quotation and the other participants in the fourth frame. To use Goffman’s (1986) terminology, while the original producers of the quotation in Excerpt 8’ may be treated as ratified but not as one particular addressee singled out of the indeterminate mass audience, the producer of the poster mentioned in Excerpt 11c’ is a ratified participant in the frame. From a different angle (Levinson 1988), we may consider Obama (and the Democrats) or O’Reilly (and his followers) as indirect targets of the interaction between the users in the online discussion forum. However, this kind of framing is not always made explicit. For instance, Excerpt 11d’ shows that a lack of reference to the ascribed source of the quoted may be exploited for communicative purposes:

120   

   Anita Fetzer and Elisabeth Reber

Excerpt 11’ comment d (1) True Islam is NOT a “religion of hate” and anyone who thinks so needs to go read the (2) Koran and speak to an American Muslim about his practice. Since the quotation lacks information about its source and is not introduced by a verb of communication, it is left indeterminate with regard to its status as an other- or self-quotation in deflecting responsibility and distancing the self from the quotation. To sum up, this formal analysis has demonstrated that quotations without source-marking and a verbum dicendi are most frequent in the online discussion forums. This does not mean that they are not recognisable for what they are but that they are contextualised with cues specific to the written medium, quotation marks. It is therefore argued that quotation marks represent quotatives in their own right, i.e. they constitute formal devices to introduce a stretch of speech as a quotation. Similarly to other work (Bös/Kleinke in this volume, Severinson Eklundh 2010), direct quotations in the online discussion forums we examined function as stance objects for construing alignment/disalignment with the ascribed source of the quoted, which may, however, not always be named explicitly. Furthermore, direct quotations are used as a resource to enact specific participation frameworks, constructing the other users in the discussion forum and/or the protagonists of the interview under discussion as addressees.

2.3 Contrastive Analysis The form-based analysis of quotations shows a context-sensitive employment in both the pre-election interviews and computer-mediated discussion forums. In the pre-election interviews, the quoted is generally prefaced by an explicit reference to the ascribed source, for instance the interviewee or some ally or opponent as is the case with the ascribed source Peter Mandelson in excerpt 1, the collective self-references in (3) and (4), the self-reference in (5), the generic other-reference ‘people’ in (2), and the collective other-reference ‘government’ in (4). While the quotative is realised by the neutral verb of communication ‘say’ in the self-quotations in excerpts 3, 4 and 5 and in the specific and generic other-quotations in (1) and (2), the collective other-quotation is prefaced by the evaluative quotative ‘admit’. This is different to the practice of quoting in the negotiated quotations analysed in excerpts 6 and 7, where the quotative is made more specific in the

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   121

process of negotiating the validity of the quote, starting with neutral ‘say’ and then using the more particularised forms ‘ask’ and ‘withdraw’. As for the additional information supplemented on the local and temporal embeddedness of the quotation, only the proper-name based other-quotation is supported by the underspecified temporal reference ‘the other day’ in excerpt 1. The negotiated quotations analysed in excerpts 6 and 7 contain a number of references to their local embeddedness, e.g. ‘equivalent programme’, ‘another studio’, and temporal embeddedness ‘last week’, ‘now’, and ‘this time last Wednesday night’. This is necessary to secure the validity of the quoted material and reflects positively on the credibility of the quoter. In the computer-mediated discussion forums such formats are rare. Here quotations are often exclusively indexed by quotation marks, as is the case in Excerpt 10, for instance. This provides evidence that the formatting of quotations is reflective on the discourse and contexts in which they are produced: in the interviews, the explicit anchoring of quotations contributes to the first-frame participants’ construal of discursive common ground amongst themselves and their mediated audience. The sources of the quotations are entextualised and may thus be attributed to the collective discursive common ground without further negotiation. In contrast, the formatting of the quotations in the online discussion forums shows that the users assign the interview under discussion the status of shared discursive common ground. As regards their functions, quotations are used to import context into the local situated interaction in both the interviews and online discussion forums. Users deploy these quotations as objects for alignment and disalignment, reproducing what was said in a (past) discursive frame in the current frame. In the interviews, quotations allow the quoter (i.e. the interviewer, usually a journalist, interviewee, i.e. the politician, or some other political ally or opponent) to align or disalign with particularised sets of the mass audience, e.g., specific groups of the electorate. The quotations in the comments are deployed to construct the quoters as supporters of particular political ideologies. This demonstrates that such displays are particularly functional in constructing participants’ political identities in these political discourse contexts although the display of alignment/ disalignment through quotations may be typical of argumentative contexts in general. Furthermore, the analysis shows that the speakers in both interviews and online discussion forums use other-quotations to challenge the addressee’s/target’s argumentation, if not their credibility. Self-quotations were only found in interviews where they support self’s argument by reasserting their credibility. This may be because the participants in the interviews are public figures who are accountable for what they say. By using self-quotation, they can demonstrate

122   

   Anita Fetzer and Elisabeth Reber

that they are responsible communicators who know what they have said before. Participants in online discussion forums, by contrast, belong to an anonymous mass audience who are not generally challenged to account for their sincerity and credibility.

3 Conclusions This analysis of quoting practices has shown that quoting is not a context-independent endeavour but a context-sensitive practice which adapts to the contextual constraints and requirements of the respective genre and medium. Quotations play an important role in stance-taking, and are used strategically in both televised political interviews and online discussion forums to align or disalign with party-political or governmental policies and affiliations. The formatting of the quotations is reflective on the participants’ identities as public figures or members of an anonymous mass audience. By speaking about politics and the policies put forward in the interviews under discussion, users do not only take part in the political discourse but are also part of the process of political decisionmaking. Discourse processes, which have been primarily situated in the private sphere of society, are thus exposed publically in online discussion forums. Fetzer/ Weizman (2006) observe: Contrary to the rights and obligations of a ratified coparticipant in an ordinary face-to-face interaction, the media-discourse audience is not in a position to take part in the process of negotiating meanings in a direct manner by negotiating the status of a communicative contribution there and then. Instead, the members of the audience are confronted with the product of the media interaction with which they can agree or disagree. (ibid, 144–145)

It is open to conjecture whether electronic devices such as discussion forums or twitter will bring about a change, allowing the media-discourse audience to participate more actively in the political decision-making processes of mediated, public and professional politics.

References Atifi, Hassan and Michel Marcoccia. 2012. “Follow-ups in online political discussions”, in: Fetzer, Weizman and Reber (eds), 22–34. Bell, Allan. 1991. The Language of News Media. Oxford: Blackwell.

Quoting in Political Discourse: Professional Talk Meets Ordinary Postings   

   123

Bös, Birte and Sonja Kleinke. “The complexities of thread-internal quoting in English and German online discussion fora”. This volume. Bublitz, Wolfram and Christian Hoffmann. 2011. “‘Three Men Using our Toilet all Day Without Flushing – This May Be One of the Worst Sentences I’ve Ever Read’: Quoting in CMC”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 433–447. Buchstaller, Isabelle. 2013. Quotatives: New Trends and Sociolinguistic Implications. Oxford: Wiley-Blackwell. Buchstaller, Isabelle, John Rickford, Elizabeth Traugott, Tom Wasow and Arnold Zwicky. 2010. “The sociolinguistics of a short-lived innovation: tracing the development of quotative all across spoken and internet newsgroup data”. Language Variation and Change 22, 191–219. Buchstaller, Isabelle and Ingrid van Alphen (eds). 2012. Quotatives: Cross-linguistic and Crossdisciplinary Perspectives. Amsterdam: John Benjamins. Burger, Harald. 2005. Mediensprache: Eine Einführung in Sprache und Kommunikationsformen der Massenmedien. 3rd ed. Berlin: de Gruyter. Clayman, Steven. 1995. “Defining moments, presidential debates, and the dynamics of quotability”. Journal of Communication 45, 118–146. Du Bois, John W. 2007. “The stance triangle”, in: Robert Englebretson (ed.). Stancetaking in Discourse: Subjectivity, Evaluation, Interaction. Amsterdam: John Benjamins, 139–182. Fairclough, Norman. 1995. Media Discourse. London: Arnold. Fairclough, Norman. 2006. Language and Globalisation. Abingdon: Routledge. Fetzer, Anita. 2006. 2“Minister, we will see how the public judges you. Media references in political interviews”. Journal of Pragmatics 38, 180–195. Fetzer, Anita. 2007. “Reformulation and common grounds”, in: Anita Fetzer and Kerstin Fischer (eds). Lexical Markers of Common Grounds. London: Elsevier, 157–179. Fetzer, Anita. 2011. “‘Here is the difference, here is the passion, here is the chance to be part of a great change’: strategic context importation in political discourse”, in: Anita Fetzer and Etsuko Oishi (eds). Context and Contexts: Parts Meet Whole?. Amsterdam: John Benjamins, 115–146. Fetzer, Anita and Peter Bull. 2006. “Who are we and who are you? The strategic use of forms of address in political interviews”. Text and Talk 26, 1–36. Fetzer, Anita and Marjut Johansson. 2010. “Cognitive verbs in context: a contrastive analysis of English and French argumentative discourse”. International Journal of Corpus Linguistics 15, 240–266. Fetzer, Anita and Gerda Lauerbach (eds). 2007. Political Discourse in the Media: Cross-cultural Perspectives. Amsterdam: John Benjamins. Fetzer, Anita and Elda Weizman. 2006. “Political discourse as mediated and public discourse”. Journal of Pragmatics 38, 143–153. Fetzer, Anita, Elda Weizman and Elisabeth Reber (eds). 2012. Proceedings of the ESF Strategic Workshop on Follow-ups Across Discourse Domains: A Cross-cultural Exploration of Their Forms and Functions, Würzburg (Germany), 31 May - 2 June 2012. Würzburg: Universität Würzburg. http://opus.bibliothek.uni-wuerzburg.de/volltexte/2012/7165/, last accessed 28 Feb 2013. Goffman, Erving. 1981. Forms of Talk. Oxford: Blackwell. Goffman, Erving. 1986. Frame Analysis. Boston: Northeastern University Press.

124   

   Anita Fetzer and Elisabeth Reber

Gumperz, John J. 1992. “Contextualisation and understanding”, in: Alessandro Duranti and Charles Goodwin (eds). Rethinking Context: Language as an Interactive Phenomenon. Cambridge: Cambridge University Press, 229–252. Günthner, Susanne. 2005. “Narrative reconstructions of past experiences. Adjustments and modifications in the process of recontextualising a past experience”, in: Uta Quasthoff and Tabea Becker (eds). Narrative Interaction. Amsterdam: John Benjamins, 285–301. Halliday, Michael A. K. and Christian Matthiessen. 2004. An Introduction to Functional Grammar. 3rd ed. London: Arnold. Heritage, John. 1985. “Analysing news interviews: aspects of the production of talk for an overhearing audience”, in: Teun A. van Dijk (ed.). Handbook of Discourse Analysis 3: Discourse and Dialogue. London: Academic Press, 95–117. Johansson, Marjut. 2007. “Represented discourse in answers. A cross-linguistic perspective on French and British political interviews”, in: Fetzer and Lauerbach (eds). 139–162. Jones, Graham M. and Bambi B. Schieffelin. 2009. “Talking text and talking back: ‘My BFF Jill’ from Boob Tube to YouTube”. Journal of Computer-Mediated Communication 14, 1050–1079. Klewitz, Gabriele and Elizabeth Couper-Kuhlen. 1999. “Quote-unquote. The role of prosody in the contextualisation of reported speech sequences”. Pragmatics 9, 459–485. Kushin, Matthew and Kelin Kitchener. 2009. “Getting political on social network sites: exploring online political discourse on Facebook”. First Monday 14. http://firstmonday.org/article/view/2645/2350, last accessed 28 Feb 2013. Leech, Geoffrey. 1983. Principles of Pragmatics. London: Longman. Levinson, Stephen C. 1988. “Putting linguistics on a proper footing: explorations in Goffman’s concepts of participation”, in: Paul Drew, and Antony Wooton (eds). Erving Goffman. Exploring the Interaction Order. Cambridge: Cambridge University Press, 161–227. Linell, Per. 1998. Approaching Dialogue. Amsterdam: John Benjamins. Mathis, Terrie and George Yule. 1994. “Zero quotatives”. Discourse Processes 18, 63–76. Montgomery, Martin. 2007. The Discourse of Broadcast News: A Linguistic Approach. Abingdon: Routledge. Park, Joseph Sung-Yul and Mary Bucholtz. 2009. “Public transcripts: entextualisation and linguistic representation in institutional contexts”. Text and Talk 5, 485–502. Schiffrin, Deborah. 1981. “Tense variation in narrative”. Language 57, 45–62. Severinson Eklundh, Kerstin. 2010. “To quote or not to quote: setting the context for computermediated dialogues”. Language@Internet 7. http://www.languageatinternet.org/ articles/2010/2665/Severinson_Eklundh.pdf, last accessed 28 Feb 2013. Sperber, Dan and Deirdre Wilson. 1986. Relevance: Communication and Cognition. Oxford: Blackwell. Wenger, Etienne. 1998. Communities of Practice: Learning, Meaning, and Identity. Cambridge: Cambridge University Press.

Andreas Musolff

5 Quotation and Online Identity: The Voice of Tacitus in German Newspapers and Internet Discussions Abstract: This paper* discusses quotation data from German newspaper articles and online discussion fora relating to the 2009 bi-millennial commemoration of a Germanic revolt under Arminius against Imperial Roman rule. In these press articles and online discussions, quotations from the writings of classical authors such as Tacitus play a key role in establishing speaker authority regarding historical details and interpretative perspectives. Some discussants adopt the voice of Tacitus for themselves; others try to undermine these assumed voices/identities, e.g. by quoting dissonant passages from Tacitean and other classical texts. The debates between quoters and critics are characterized by a high degree of implicit or explicit meta-communicative commenting. In using a metarepresentation-/ relevance-theoretical framework, this paper distinguishes quotations in press articles from those in the online discussions by highlighting the specific function of the latter in establishing and maintaining discursive identities in an emergent community of practice. Keywords: internet fora, metarepresentation, quotation, relevance, stance, voice

1 Introduction 2009 was a good year for popular history in Germany, as it marked the 2000th anniversary of an event that had been treated for centuries as the birth of a distinctive Germanic/German national identity that was supposedly defined by its longing for freedom from foreign rule.¹ In 9 AD, three Roman legions under the command of the governor Publius Quinctilius Varus were defeated by a coalition of Germanic tribes under the leadership of a Cheruscan warlord whose historical name was only recorded in its Latin form Arminius (later Germanised to Hermann) after the so-called “Battle of the Teutoburg Forest” which took place in North-

* This paper has benefited from critical comments and suggestions by David Lilley and three anonymous reviewers. The remaining errors are the author’s sole property. 1 See e.g. Der Spiegel 2008; Bendikowski 2009.

126   

   Andreas Musolff

West Germany, somewhere between the Weser, Lippe and Ems rivers.² No firsthand battle accounts have survived from ancient times but there are enough references to Varus’s defeat in the works of Roman and Greek historians to leave no doubt about its historicity,³ despite the fact that until the 1980s no archaeological traces of a large-scale battle had been found on German soil. The battle was commemorated publicly in 1875 with the erection of a monument (Hermannsdenkmal) near the town of Detmold in North-Rhine Westphalia, which used to be regarded as the best guess for the likely location of the Teutoburg Forest.⁴ This situation changed in the 1980s with the discovery of a large ancient battlefield near the village of Kalkriese in Lower Saxony (Schlüter 1992, Clunn 1999, Derks 2009). Besides Roman coins, excavations in Kalkriese have also produced evidence of weaponry, human and animal remains, and makeshift ramparts; the latter have been made publicly accessible in a museum park whose administration claims that their site is the location of the battle between Varus and Arminius.⁵ The exact identification of the main Teutoburg battlefield is, however, still a matter of debate among archaeologists (Berke 2009: 137–138, Wolters 2008: 161– 173), and the Detmold tourist agency, whilst avoiding assertions that the monument itself marks the precise site of the battlefield, continues to present the modern-day Teutoburger Wald region as the place of the battle.⁶ In the anniversary celebrations of 2009, the Hermann monument was presented as a ‘symbol for Europe’s integration, cultural diversity, international understanding and peace’ (Zeichen für ein zusammenwachsendes Europa, für kulturelle Vielfalt und […] für Völkerverständigung und Frieden; Lippe Tourismus & Marketing AG 2009: 1). Such a presentation stands in marked contrast with the monument’s previous uses. During the Second German Empire (1871–1918), for instance, anniversary celebrations at the monument suggested a continuity of German resistance against alien

2 According to Tacitus’s Annals, the Roman army was defeated in the region of the teutoburgiensis saltus (Tacitus 2005: 344–347). The location of this ‘forest’ was assumed to be situated in the wider Westphalian region; since the late 17th century, the mountain ridge of the “Osning” near Detmold has been popularly identified as the most likely site of the battlefield (Berke 2009: 133–135). 3 The main classical sources besides Tacitus are Cassius Dio, Velleius Paterculus, Frontinus, Florus, Seneca, Plinius the Older, Strabo, Suetonius, Orosius. For their rediscovery, especially during the Renaissance, see Ottomeyer 2009; Krebs 2011: 56–104. 4 See Berke 2009; for the history of the monument, see Wolters 2008: 188–192; Mellies 2009a, b. 5 See Varusschlacht im Osnabrücker Land. Museum und Park Kalkriese, http://www.kalkriesevarusschlacht.de/varusschlacht-information/selbstverstaendnis/info-selbstverstaendnis.html, last accessed 5 Oct 2012. 6 See Das Hermannsdenkmal – Geschichte erleben! Ed. Landesverband Lippe, Hermannsdenkmal-Stiftung, http://www.hermannsdenkmal.de, last accessed 5 Oct 2012.

Quotation and Online Identity   

   127

(“Western”) domination (Dörner 1995, Mellies 2009a), and during the 1920–30s, the National Socialists used it as a rallying point for chauvinistic and racist campaigns (Halle 2009; Mellies 2009b, Niebuhr/Ruppert 1998, 2007). Given the ideologically charged pedigree of the Teutoburg Battle memory in popular history, it is no surprise that in the run-up to and during the anniversary year, German internet discussion fora and blogs were filled with debates about its commemoration and its political and cultural significance for present-day Germany. Some mainstream press media, such as the daily newspaper Die Welt and the weekly magazine Der Spiegel, created discussion groups on their websites. One of these, the Die Welt-online discussion forum, provides the data for this study. Our sample includes 18 articles and the online discussion commentaries for seven of these articles, amounting to a total of 64,844 words. The Die Welt data are part of a larger corpus which was assembled for the purpose of collecting and analysing metalinguistic commenting in online discussion fora.⁷ The focus here will be on (purported) quotations from authoritative historical sources in Die Welt articles and online fora. In particular, we will study quotations from Tacitus’s (AD 56–117) writings, i.e. the Annals and Germania that are used to boost a speaker’s credibility as regards to her⁸ argumentative stance in terms of objectevaluation, self-positioning and alignment with others (Du Bois 2007), and thus to construct her online identity.

2 Quotation and Metarepresentation Quotations can be viewed as cases of metarepresentation insofar as they constitute “utterances about attributed utterances” (Wilson 2000: 413). They metarepresent a (public) representation, i.e. the quoted utterance, as part of another (public) representation, i.e. the speaker’s utterance. All varieties of metarepresentation, “public, mental and abstract, can be analyzed in terms of a representation by resemblance”; direct quotations typically serve to increase “the salience of formal or linguistic properties” (i.e., metalinguistic resemblances), indirect quotations typically increase “the salience of semantic or logical properties”

7 The whole corpus comprises 71 articles from eight press media (Die Welt, Westfalenblatt, Der Spiegel, Neue Westfälische, Süddeutsche Zeitung, Die Zeit, Der Freitag, Das Parlament), plus internet fora texts, and amounts to 158,091 words, see Musolff 2010, 2012. 8 When referring to communication partners, I follow the convention, established by Sperber/ Wilson (1995), of designating the speaker as she and the hearer as he. Only in those cases where a real-world speaker or a constructed discussion participant explicitly identifies himself as male, grammatical gender attribution will follow suit.

128   

   Andreas Musolff

(i.e., interpretive resemblances), and mixed quotations exploit “both metalinguistic and interpretive resemblances” (Wilson 2000: 425–426). Of these three main subtypes of representational resemblance, it is the second, i.e. interpretive resemblance, that is of greater importance here because it concerns the speaker’s attitude towards the content of the metarepresented utterance, in particular regarding its credibility and its possible use as a back-up for her argumentative stance in a given utterance. This does not mean, however, that metalinguistic resemblance is irrelevant – as we shall see, the question of formal linguistic accuracy of Tacitus-quotations can become important for the interpretive dimension, e.g., when the linguistic accuracy of quotations is explicitly contested as part of a wider argument. In the rhetorical tradition, the utilisation of quotations for argumentative purposes has been prominently theorised as a sub-type of the “argument from authority” which “uses the acts or opinions of a person or a group of persons as a means of proof in support of a thesis” (Perelman/Olbrechts-Tyceca 1971: 305). Whilst the argument from authority has been attacked for being a form of pseudoargumentation in some strands of anti-rhetorical rationalism, its prominence in non-formal argumentation from ancient times to the present day should not be underestimated (Perelman/Olbrechts-Tyceca 1971: 177, 306–307). The quotations attributed to Tacitus in our data thus invoke his authority as one of the most famous classical Roman historians to corroborate statements about the historical context of the Teutoburg Battle and its significance for the development of Germany. The use of quotations by press journalists in particular imitates that of scholarly historical research, albeit at a reduced level of accuracy and text-critical reflection. The articles in Die Welt treat Tacitus as the main historical authority on the Teutoburg Battle and on ancient Germany. Of the eighteen articles, eight make use of extended quotations from Tacitus’s writings; the only other Roman historian mentioned (twice and briefly so) is Velleius Paterculus (c. 19 BC – c. 31 AD).⁹ The following examples are representative of the way in which Tacitus’s writings are quoted in Die Welt: ¹⁰

9 This situation is similar across other German press media that are represented in the wider corpus; the only classical historians mentioned besides Tacitus are Suetonius and Cassius Dio but they are not nearly as often referred to as Tacitus (i.e. each with just one quotation). 10 Italicizations and English translations of all following examples are by AM. For the usage of the terms “Germany” and “Germans” in present-day English historical terminology when referring to ancient “Germanic” territories and people, see current Tacitus-translations (e.g. Tacitus 1999 and 2005) and Wells 2001 and Krebs 2011.

Quotation and Online Identity   

   129

(1)

Den wahren Beweggrund der Römer, hartnäckig am Krieg festzuhalten, entlarvt der römische Geschichtsschreiber Tacitus: „Krieg war zu dieser Zeit nur noch gegen die Germanen zu führen, mehr um die Schande zu tilgen, die mit dem Verlust des Heeres unter Quinctilius Varus verbunden war, als aus dem Bestreben, das Reich zu erweitern, oder wegen der Aussicht auf entsprechenden Gewinn.“ (Husemann 2009) (‘The Roman historian Tacitus reveals the true reason for the Romans’ stubborn insistence on continuing the war: “There remained only one war to be fought: that against the Germans, more in order to obliterate the shame associated with the loss of Quinctilius Varus’s army than for the purpose of enlarging the empire or gaining profit”.’).

(2)

[In den „Annalen“ des römischen Historikers Tacitus] war unter anderem von einem Mann namens Arminius die Rede, der „unbestritten der Befreier Germaniens“ gewesen sein soll. (Seewald 2009a) (‘Among other historical figures, Tacitus’s Annals mentioned a man called Arminius who was said to have been “without doubt Germany’s liberator”.’)

(3)

Tacitus‘ Bonmot vom „Befreier Germaniens“ erscheint auf einmal als ironisch-melancholische Wendung, dass die Germanen mit ihrem Sieg auch frei von allen Segnungen der Zivilisation geblieben sind. (Seewald 2009b) (‘Tacitus’s famous quote about “Germany’s liberator” suddenly appears in an ironical-melancholic perspective, for it implies that victory for the Germans meant that they also remained untouched by all blessings of civilisation.’)

In all these examples, the quoted text is formally indicated by quotation marks. In (1), the passage consists of a whole (translated) sentence from the Annals (book I, 5),¹¹ which is rendered in direct quotation and introduced by the stanceexplicit verbum dicendi entlarven (‘to reveal’). In (2), the Tacitean phrase unbestritten der Befreier Germaniens is part of an indirect quotation attributed to Tacitus via the sollen-construction. In (3), part of the same phrase is quoted to qualify the noun phrase Tacitus’ Bonmot, which is then commented on from a present-day perspective. We can characterise the communicative purpose of these utterances as an invitation for the reader to in the first place accept the quoted passage as a reliable (translated) quotation of Tacitus’s original text (‘formal and interpretive resemblance’, in terms of Sperber’s theory of metarepresentation), and as 11 The passage is identical to the text of the German translation by M. Fuhrmann in Tacitus 2000: 11.

130   

   Andreas Musolff

information worthy of being processed at a reasonable effort (‘communicative relevance’, in terms of general relevance theory). In addition, we find that the speakers (i.e., the journalists Husemann and Seewald) either adopt Tacitus’s stance or at least do not contradict it. In (1), the verb entlarvt (‘reveals’) explicitly endorses the quotation from the Annals. In (2), the formulations war die Rede von (‘mentions’) and sollen (‘being said to’) provide a non-committal perspective on Tacitus’s formulation. Even in (3), where Seewald gives a new, ironic reading of Tacitus’s “Bonmot vom ‘Befreier Germaniens’” by adding a further meaning aspect to the notion of “liberation” (i.e., ‘liberation from civilisation’), he does not contradict the original statement but represents it as having been additionally validated in a way that Tacitus could not have foreseen. In order to be able to fully understand these quotations, readers of these articles have to (de- and then re-)construct the metarepresentational complex at several levels: (A) the original proposition from a (translated) text by Tacitus (e.g., ‘There remained only one war to be fought….’, ‘Arminius was the liberator of Germany’); (B) its integration into the metarepresentation by the Welt journalist (e.g., ‘Tacitus reveals / mentions / uttered a Bonmot’); (C) the contextually inferable stance of the journalist towards the original proposition (e.g. as being reliable information and/or at least of general interest) (D) with specific regard to quotation (3): further conclusions that interpret the Tacitean quotation in a special (ironical) perspective, i.e., that Germany was ‘liberated’ in more senses than those intended by Tacitus. Of these four representation levels, the first two are necessary for readers to understand the respective utterances as instances of quotation; the third one is necessary to understand the speaker’s attitudinal and epistemic stance and evaluation. The fourth is a special case pertaining only to example (3); it involves the representation of further assumptions about the speaker’s stance regarding ancient Germany’s civilisatory status which are brought to bear on the understanding of the quotation. Generally, the Die Welt journalists’ attitude towards Tacitus’s relevant writings, the Annals and Germania, is one of qualified acceptance: they take both books seriously as being the most widely known sources on the historical context of the Teutoburg Battle and ancient Germany. However, the journalists do not take Tacitus’s statements at face value but instead try to assess them from their own historical context as politically biased in representing ancient Germanic warrior society as a counter-image to imperial Rome. Husemann and Seewald also pay particular attention to the (mis-)use of the Germania

Quotation and Online Identity   

   131

for nationalist propaganda in Germany from the 16th to the 20th century (Husemann 2009; Seewald 2009a, c). On the other hand, the vast majority of participants in the discussion fora attached to the Die Welt articles enthusiastically accept the Roman historian’s texts as providing a truthful depiction of Germanic life in the first century AD. Most of the 26 quotations from and allusions to Tacitus in the fora endorse his key utterances wholeheartedly or presuppose such an endorsement and derive sweeping conclusions from them to support the bloggers’ own argumentative stance. The following example of a posting, by a blogger who calls himself “Heinrich” in his comment on Seewald’s article ‘The Varus battle is no anniversary to celebrate’ (“Die Varusschlacht taugt uns nicht als Gedenktag” = Seewald 2009c), is representative in this respect: (4)

Heinrich sagt: Dem Autor Seewald ein Dankeschön für diesen ausgewogenen Artikel. […] Arminius, Ritter Roms und Befreier Germaniens – lt. Tacitus – gilt [….] nicht zu unrecht als Anker für das spätere Deutschland [….] Aus römischer Sicht ein Verräter – er hat den römischen Fahneneid und sein Bürgerrecht gebrochen  – hat er den Appetit Rom [sic] auf Germanien nachhaltig verdorben und ist der 1. deutsche Held. (‘Heinrich says: Thank you, Mr Seewald, for this well-balanced article! Arminius, being a Roman knight and the liberator of ancient Germany – according to Tacitus – is rightfully regarded as the cornerstone of what later became Germany. Whilst he was a traitor when seen from the Roman perspective – after all, he broke his oaths of Roman allegiance and citizenship – he thoroughly spoilt the Roman appetite for conquering Germany and thus is the first German hero.’)

Heinrich identifies with the Germanic side of the ancient conflict and links the battle to the later course of German history, presumably up to the present day. He concedes that the Roman viewpoint may have been different, not without reason, but clearly shows where his loyalties lie by praising Arminius as the “first German hero”. Tacitus is invoked as the source for the epithet “liberator”; the Roman historian is thus treated as a kind of witness in an (imaginary) dispute over Arminius’s hero-status. Heinrich’s posting is the 29th of altogether 64 comments on Seewald’s article, the overwhelming majority (N = 40) of which are critical of the journalist for allegedly not being patriotic enough and for submitting to the rules of ‘political correctness’. Heinrich’s posting occurs after a number of attacks on Seewald have already been posted (i.e., contributions 1–6, 9–12, 14–17, 19–28) and can therefore be seen as an attempt to set the record straight by praising the journalist’s ‘well-balanced’ consideration of both sides whilst still

132   

   Andreas Musolff

showing a modicum of patriotic pride. The speaker’s online name, Heinrich, is conceivably a real-life name but it also fits in a particular strand within the online forum: it is a traditional German first name, famous in history as that of many German kings and princes as well as legendary characters (see e.g. Grimm’s fairy tale, Der Froschkönig und der eiserne Heinrich): it provides a patriotic resonance and compares to many other German(ic)-sounding identities used in the forum, such as Germania, Hermann, Leopold von Ranke, duorp, Der vorletzte Germane, and Ostgote. The suggestive name, the evaluating stance on the accuracy of Tacitus’s and Seewald’s accounts of ancient Germany and the explicit conclusion as regard to Arminius’s hero status in German history show that Heinrich’s contribution to the online debate is not just an independent statement but also a contribution to a discursive series of utterances that shows a strong tendency to emphatic stance-taking and identity-construction which is characteristic for the blogosphere (Myers 2010: 95–114). Online fora can be viewed as dynamic “communities of practice” (Wenger 1999, Holmes/Meyerhoff 1999) that are established ad hoc by the “commentary” function of the forum management and which include all contributions that have been left online by the forum editors (i.e. have not been deleted for breach of forum “netiquette”) and are thus freely accessible to internet users. The members of such a community of practice have to establish and defend their online identities vis-à-vis other participants. This is a situation not dissimilar from face-work in social interaction, as introduced into social psychology by Goffman (1972) and developed into “politeness theory” in linguistic pragmatics (Brown/Levinson 1987, Watts 2003, Watts/Ide/Ehlich 2005).¹² Goffman’s classic definition describes face as the “positive social value a person effectively claims for himself by the line others assume he has taken during a particular contact” (Goffman 1972: 320); it is constructed for the purpose of social interaction, and hence neither a ‘naturally’ given nor a wholly privately assumed identity. The ‘constructedness’ of such an online identity is, arguably, even more evident to the participants of online discussions than in real-life face-to-face interactions. Usually, the competent adult participants of internet fora are (or should be) aware of the fact that the online identities of other participants are not necessarily identical with the identities of their real-life holders. We may speak metaphorically of an internet identity as an (online) voice (in analogy to

12 Goffman’s view on face is a social, interactional one which according to Watts (2003) and Werkhofer (2005) was disregarded by Brown/Levinson (1987). The merits of this critique cannot be discussed here, but the application of the face-theoretical approach to internet discussion fora, as attempted here, assumes an interactional perspective rather than a mainly speakercentred one.

Quotation and Online Identity   

   133

identity construction in real-life conversations as conceived by Tannen 2007), rather than as a face, to highlight its partial, discursive and dynamic character. The assumed awareness of the constructed character of online identities does not mean, however, that forum participants treat other members of their community of practice as disembodied, abstract entities. A number of studies (e.g., Angouri/ Tseliga 2010; Hardaker 2010; Yus 2011: 271–279) have shown that emotional and interactive stance-taking is typical of computer-mediated communication (CMC) and that the use of impoliteness is facilitated by the relative anonymity of internet identities. Within this competitive voice-establishment and voice-maintenance system, references to authority play a key role insofar as a speaker’s voice has greater weight when she can refer to high-prestige voices that share her stance. Quoting from “Tacitus” as a (metonymic) reference to a famous classical source thus counts as an argumentative move to boost the quoter’s stance and her voice. In Heinrich’s usage (example (4)), the Roman historian’s acknowledgement of Arminius as “Germany’s liberator” may even count as double as it were, because “Tacitus” is viewed as an authority on the (Roman) enemy’s side who, against his own political loyalties, had enough objectivity to recognise the achievements of the “first German hero”.

3 Identity Construction Through Quotation The above-quoted references to Tacitus fit the standard argument from authority: a passage from a source text is quoted (in translation) and referenced to Tacitus as the original author. This is achieved either through a matrix clause that identifies him as the original speaker of the quoted passage, by attribution (“Tacitus’s Bonmot…”), or via the “according to” construction (in German lt. [= laut]). These quoted references to speaker and source texts are more or less precise, e.g., the Die Welt journalists call Tacitus a “Roman historian” (example 1) and refer to a specific source text, such as the Annals (example 2). In examples (3) and (4), the readers are only vaguely reminded of the provenance of the “liberator” epithet; thus, we may assume that the quoters rely on sufficient shared knowledge about Tacitus among their readers to fill in the gaps and accept the respective quotations as bonafide, broadly correct translations based on reliable editions. The forum debate data show, however, that these background assumptions do not always remain uncontested. One striking case of questioning the standard sources is the claim by one blogger, “wjbeckermann”, that he alone uses Tacitus’s “own words” differently from those in the established editions and translations:

134   

(5)

   Andreas Musolff

wjbeckermann sagt: Arminius war auch ein Freiheitskämpfer. Tacitus zitiert ihn wie folgt: wenn ihr die Knechtschaft wollt, dann folgt dem Segestes, wenn ihr die Freiheit wollt, dann folgt mir. Segestes […] ist übrigens der Typ, der dem Arminius die Frau gestohlen hat, seine schwangere Tochter Thusnelda. […] Arminius kämpfte für seine Liebe, für sein Land und seine Kultur. […] Nachzulesen bei Tacitus, aber nur in meiner Übersetzung “De Bello Arminico-Arminsieg” bei verlag-neuwiese.de. (‘wjbeckermann says: Arminius was also a freedom fighter. Tacitus quotes him as follows: “if you want to be slaves, follow Segestes, if you want your freedom, follow me”. Segestes was the guy who stole Arminius’ wife, his own, pregnant daughter, Thusnelda. Arminius fought for his love, his land and his culture. You can find it all in Tacitus but only in my translation “De Bello Arminico-Armin‘s victory”, available from verlag-neuwiese.de’)

The speaker, who we may presume to be, exceptionally, the real-life person Wolfgang Josef Beckermann,¹³ uses a matrix-clause-plus-direct-quotation construction to invoke Tacitus as an authority for Arminius’s status as a heroic freedom fighter but which can only be found in his own translation of an alleged pep-talk which the Cheruscan warlord is supposed to have given to his followers and in which he rallies them not only against the Romans but also against traitor-Germans like his father-in-law, Segestes. As in examples (1)–(4), Tacitus’s account is assumed to be truthful, but wjbeckermann goes beyond the previous Tacitus-quotations in two respects: a) he privileges his own translation and b) he treats Tacitus’s text as the source for a further, secondary quotation, i.e. that of Arminius’s own words. He thus insinuates that his translation not only gives the only correct version of Tacitus’s Latin text but also that of Arminius’s speech (which would have had to be delivered in some kind of early north-west Germanic dialect). Such a suggestion is, of course, a product of pure fantasy insofar as Tacitus, who wrote his books nearly a century after the Battle of the Teutoburg Forest, is unlikely to have had access to (translated!) records of Germanic warlords’ speeches. The inclusion of such speeches and also of dialogues was a wellknown literary technique among Roman historians, as was the heroic depiction of barbarian enemy leaders which served to highlight the achievements of their Roman vanquishers (Günnewig 2009). However, such text-critical considerations evidently play no role in wjbeckermann’s statement which constructs his online

13 See the publisher’s website “Verlag Neuwiese” referred to in the example which identifies “Dr.rer.nat. Wolfgang Josef Beckermann, Dipl.Biologe” as its legal representative (http://verlagneuwiese.de, last accessed 5 Oct 2012).

Quotation and Online Identity   

   135

identity as the true voice (qua ‘correct’ translation) of Tacitus who, in turn, is assumed to have given voice to Arminius’s own words. A possibly even stronger identification between quoter and quotee can be found among the comments on another Welt article, ‘Not everything was bad in Varus’s Germany’ (“In Varus’ Germanien war nicht alles schlecht”, Kellerhoff 2009): (6)

So schrieb Tacitus: Die Gebiete Germaniens bieten einen traurigen Anblick. Sie sind voller schauriger Wälder, grässlicher Sümpfe, die Rinder unansehnlich, die Pferde weder schnell noch schön. Diese Barbaren leben in Gehöften und trinken Met… (‘Thus wrote Tacitus: The Germanic lands are a dreary sight. They are full of horrible forests and dreadful swamps. The cattle are unimpressive, the horses neither fast nor beautiful. These barbarians live in homesteads and drink mead…’)

At first sight, this contribution seems to be a simple case of matrix clause plus direct quotation; in fact it is a collection of snippets from the standard German translation of the Germania (cf. Tacitus 2000: 9, 13, 23). However, the apparent matrix clause Thus wrote Tacitus turns out to be in fact the speaker’s online name, i.e. the speaker presents her own online identity as congruent with Tacitus’s voice. Another exceptional feature of this contribution is that the quoted passages do not paint a heroic picture of ancient Germany and its inhabitants, in contrast to the great majority of Tacitus quotations. Within the forum debate on Kellerhoff’s article, example (6) forms the tenth posting and follows a comment by the contributor “Teiwswitan” who is outraged by the journalist’s suggestion that the Roman occupation of Germany also had its good aspects (contributions log no. 9). Without explicitly addressing Teiwswitan or referring to Kellerhoff, Thus wrote Tacitus takes the journalist’s side and invokes Tacitus as an authority to highlight the barbarian, uncivilised status of the ancient Germans. Thus wrote Tacitus’s intervention is followed by three strongly ‘pro-German’ postings which include another comment (log no. 12) by Teiwswitan and one posting by a presumably male speaker who calls himself German (“Germane”) (log no. 13), both of whom address Thus wrote Tacitus explicitly through the discussion forum’s “@”-answer function: (7)

Teiwswitan: @So schrieb Tacitus […]: schaurige Wälder und Sümpfe? Na und, so was schreckt nur verweichlichte Römer! Übrigens auch das schrieb Tacitus:  … Gleichwohl ist dort die Ehe streng und in keiner Beziehung möchte man ihre Sitten mehr loben; […] Na da hat uns das dekadente Rom ja richtig nach vorne gebracht!

136   

   Andreas Musolff

(‘Teiwswitan: @Thus wrote Tacitus […]: horrible forests and dreadful swamps: so what? That sort of thing only scares effeminate Romans. By the way, Tacitus also wrote this “… nevertheless, marriage is very strict among the Germans and their customs deserve the highest praise in that regard” – well, decadent Rome has really brought us progress!’) (8)

Germane: @So schrieb Tacitus […]: Gaius Julius Caesar, die Germanen sind sehr gross gebaut und kraeftig. P.S.: MMMMMmhhhhhhhh, schoene schaurige Waelder und Suempfe. …. (‘German: @Thus wrote Tacitus […]: Gaius Julius Caesar, the Germans are very tall and strong. P.S.: MMMMMmhhhhhhhh, good old dreadful forests and swamps! ….’)

In (7), Teiwswitan tries to neutralise Tacitus’s implied negative stance as highlighted in example (6) by allocating any deterring effect of ancient Germany’s features exclusively to the “decadent Romans”. The implied conclusion can only be that non-Romans, including Germans, are not that easily scared. This gives Teiwswitan a platform to provide a counter-quote, again taken from Tacitus (who thus continues to be used as an unquestioned authority), which focuses on the supposedly simple and strict marriage customs among Germanic tribes. Teiwswitan’s last sentence ironically praises the Romans for having brought “real progress” to the modern-day Germans (“us”) who, it is insinuated, follow the decadent Roman model more than the simple, unspoilt ways of their innocent Germanic forefathers. German’s posting in (8) contains further twists of (meta-)stance-taking: in the first place, it provides a counter-quotation that cites another, possibly even ‘higher’ and earlier historical authority than that of Tacitus, i.e. Gaius Julius Caesar who had indeed depicted his ancient German enemies as extraordinarily strong and dangerous fighters (Günnewig 2009: 31). German then adds a “post scriptum” that repeats a phrase from Tacitus’s uncomplimentary description of Germany as quoted by Thus wrote Tacitus, but also expresses a new (meta-)stance towards the proposed existence of ‘dreadful swamps and forests’ in Germany by reinterpreting them as positive features which evoke feelings of homeliness and comfort, symbolised in the interjection “MMMMMmhhhhhhhh”.¹⁴ German tries to both outdo and reassert Tacitus as an authority in order to refute Thus wrote Tacitus. He picks what he likes from Caesar’s writings and follows this ‘patriotic’ quotation by an expression of fond nostalgia for the wild forests and swamps of

14 For the concept of metastancing see Vandergriff 2012.

Quotation and Online Identity   

   137

ancient Germany, as described by Tacitus. The “PS” contains an echoic rendition of Thus wrote Tacitus’s Tacitus quotation (Sperber/Wilson 1995: 238–239) which contrasts with the cheery “MMMMMmhhhhhhhh”, thus creating a tongue-incheek and/or ironic meta-stance effect. Examples (7) and (8) illustrate two strategies of meta-stancing through quotation. One is to provide counter-quotations from the same authority (e.g. in (7)) or a higher authority (e.g. in (8)) which contradict or at least neutralise the stance expressed by the preceding quotation. The other strategy is to reinterpret an original quotation in a way that changes the evaluation (e.g., in (8), from negative to positive). Thus wrote Tacitus did not respond to Teiwswitan’s nor to German’s postings and appears to have left that particular online forum community for good. The last set of examples to be discussed shows how an attempted identity construction through quotation is first challenged and then defended. It is part of a forum discussion about an earlier article by Seewald from December 2008, which prepared the ground for further coverage of the Teutoburg Battle anniversary by reporting about a newly discovered battlefield from the third century AD near Northeim in Lower Saxony under the headline: ‘New Battlefield discovered: here the Romans massacred the Germans’ (“Schlachtfeld entdeckt. Hier metzelten Römer die Germanen nieder”, Seewald 2008). The forum debate on this article is one of the longest discussions: within a week it gathered 138 messages and lingered on well into 2009, with the last postings appearing as late as October 2009. It is also one of the most polemical debates in the corpus, with several postings having been deleted by the forum management. The list of adopted online names includes a host of pseudo-Germanic identities, e.g. Wotan, Hermann, Loki, Teiwswitan and Heinrich (the latter two are possibly identical with the Die Welt forum contributors of the same names mentioned above), as well as Thusnelda. This last contributor takes her name from Arminius’s wife who was captured by the Romans, as recorded in Tacitus’s Annals (Book I, chapters 57–60, see also example (5) above). Thusnelda first intervenes early on in the debate (contribution log no. 25) with a tongue-in-cheek response to Seewald’s ‘patriotic’ critics who are unhappy with his article’s focus on a Germanic defeat. She welcomes the discovery of a new battlefield find because it ‘will lead to the building of another museum’ which is good for Sunday afternoon outings and tourism (“Da gibt es bald nochn Museum und man kann da sonntags zum Kaffee und wenn man Besuch aus Amerika hat und mit Schulklassen hinfahren”). This tongue-in-cheek praise is followed up a day later (contribution log no. 114) by (presumably) the same contributor who now calls herself Thusnelda in heaven (“Thusnelda im Himmel”). In this second posting, she mocks the other participants’ serious discussion about Arminius’s heroism. In adopting a colloquial register by using expressions such as ‘my

138   

   Andreas Musolff

Daddy’, ‘dumb blonde’, and ‘Armin’ in place of ‘Arminius’, she begins to explain Thusnelda’s relationship with the Germanic warlord by describing her father’s reservations against the liaison: (9)

Thusnelda im Himmel: Also, mein Papa hat den Armin gar nicht leiden mögen, weil er gesagt hat, “das ist ein windiger Opportunist, den heiratest du nicht.” […] Aber der Armin und ich, wir hatten uns so lieb, wir haben trotzdem geheiratet […]. Was ich aber wirklich traurig finde, ist, daß ihr von mir gar nichts mehr wißt, nur noch die Abkürzung von meinem Namen, “Thussi”, und dabei an so ‘ne blonde, doofe denkt. Das ist nicht fair, bloß weil ich damals den Armin …” (‘Thusnelda in heaven: Well, my daddy couldn‘t stand that Armin guy, like, he said ‘that‘s an unreliable turncoat – you are not gonna marry him!’ Still, we were so much in love; we got married in spite of my father. But what I find really sad is that you don‘t remember anything about me, only my abbreviated name, “Tussi”, and then you only think of a dumb blonde. That‘s not fair, just because Armin and I …’)

Here, Thusnelda-in-heaven first relates her love-story in a mock-kitschy version that highlights its significance as an act of defiance against her father, Segestes, and then pointedly attacks her co-bloggers for viewing her original namesake as the mere side-kick of Germany’s “liberator” whose only historical legacy is the derogatory, sexist nickname “Tussi” in present-day German.¹⁵ Twenty minutes later (contribution log no. 116), one of the more ‘patriotic’ forum participants, who calls himself historiker2 (in contradistinction from another blogger, historiker), apparently feels the urge to correct this disrespectful tongue-in-cheek treatment of his hero: (10)

historiker2 sagt @ Thusnelda: Sie tun dem Arminius Unrecht. […] Arminius war ein großer Militärführer wie Vercingetorix bei den Galliern […]; und er wurde erst im Jahr 21 n. [= nur(?)] durch den Verrat seiner Familienmitglieder  – initiiert durch Rom  – gemordet und hätte fast eine germ. [=  germanische] Front gegen Rom zusammen mit Marbod und den Markomannen zustande gebracht. (‘historiker2 says@ Thusnelda: You are not fair to Arminius. He was a great military leader, just like Vercingetorix among the Gauls. He was only [?] murdered in 21 AD by a Rome-sponsored family cabal and he had almost

15 See http://web.archive.org/web/20090302081704/http://uni-giessen.de/~g41007/tussi.html (= Berhorst 2008; last accessed 5 Oct 2012).

Quotation and Online Identity   

   139

managed to create a [united] Germanic front against Rome with Marbod and the Marcomans’). Historiker2 does not just add another viewpoint to Thusnelda’s but tries to undermine her status as a competent contributor in the forum debate.¹⁶ He lectures her on omissions in her account and disregards the main focus of her posting, i.e. the relationship between Thusnelda and Arminius and their defiance of the Roman-sympathiser Segestes. Historiker2’s supposedly new information consists, however, mainly of allusive name-dropping, without any explanation of the argumentative grounds for his bold Arminius-Vercingetorix comparison or reference to a united Germanic front. Significantly, these speculations have no basis in Tacitus, whereas the main story-lines of the Arminius-Thusnelda-Segestes saga as retold by Thusnelda in heaven can be found in the Annals (I, 58). Fifteen minutes later in the online debate recorded by the Die Welt discussion-website, Thusnelda/Thusnelda-in-heaven, now re-renamed “Thusnelda in heaven – the whole truth” (“Thusnelda im Himmel: Die ganze Wahrheit“) returns to the debate¹⁷ by imagining an exchange between Thusnelda and her son, Thumelicus, who asks her to tell the complete story of her family’s tragedy: (11)

Thusnelda im Himmel: Die ganze Wahrheit sagt: Also, Thumelicus, der Sohn von Armin und mir, kam gerade zu mir, ich sollte euch doch die ganze Wahrheit sagen. Wollte ich eigentlich nicht, weil die noch viel schrecklicher ist: mein Papa Segestes hat […] mich […] dem römischen Kaisersohn Germanicus mitgegeben, als Sklavin. […] Armin hatte keine Chance (später hat ihn mein Papa hinterrücks umbringen lassen). So kam es, daß ich auf dem Triumphmarsch des Germanicus durch Rom geschleppt wurde, mit meinem kleinen Sohn. Der war inzwischen auf die Welt gekommen, aber nur, um dann in Ravenna zum Gladiator ausgebildet zu werden. Er war so ein starker Junge, aber es hat ihm nichts genützt … (‘Thusnelda in heaven – the whole truth – says: Well, Thumelicus − that’s Armin’s and my son − just came to me [and suggested that] I should tell you the whole truth. Didn’t want to, really, because it’s even worse: my daddy

16 The majority of historiker2’s contributions consist of criticisms and ‘corrections’ of other contributors’ postings; his claim to superior knowledge is thus not solely aimed at Thusnelda but also, for instance, at MS, Tiberius, Hermann, and Hiwi. 17 Between the postings by historiker 2 and Thusnelda in heaven – the whole truth (contribution log nos. 116 and 118) only one other posting intervenes, i.e. that by the blogger “Gerd” who emphatically asserts that (ancient) Germanic and (modern) German nationalities are identical in order to contradict an earlier contributor (“GEGEN BRAUNe”) who had dared to question this myth.

140   

   Andreas Musolff

Segestes gave me as a slave to the Roman Emperor’s son, Germanicus. Armin did not really have a chance; later on, Segestes even had him assassinated. I ended up being dragged along in Germanicus’s triumph in Rome with my small son. He had been born in the meantime but only to undergo training as a gladiator in Ravenna later. He was such a strong child but it did him no good …’) There are no further comments by other discussants about this intervention in the online discussion, so the reception of (11) in this community of Arminius admirers remains unknown. The most plausible intended effect of Thusnelda’s last posting is to reassert her credibility after historiker2’s face- (or, voice-)threatening attack against her. Whilst maintaining the informal tone, she seeks to have even more complete and up-to-date information than anyone else. By concentrating on Thumelicus’s tragic fate, Thusnelda-in-heaven – the whole truth tries to gain sympathy for him and for herself as his mother, thus again highlighting the personal and emotional aspects of the Arminius story. However, in the remainder of the online discussion among Teiwswitan, historiker 2, Heinrich et al., this aspect, and thus, Thusnelda’s voice, is drowned by debates about more ‘serious’ issues such as the question of whether the ancient Germans were really barbarians, the alleged decadence of the Romans, and the historical mistakes supposedly made by museums and exhibition organisers. Thusnelda apparently managed to defend her position by not ‘giving in’ to the patriotic experts but at the expense of being side-lined in the forum community.

4 Conclusions In the German media and online fora discussing the 2009 battle anniversary, quotations from Tacitus’s writings are generally used to enhance the speaker’s credibility in taking a stance on the evaluation of historical figures such as Arminius, Thusnelda and Varus, or the battle event and its significance for the long-term history of German national identity. Tacitus is invoked as the (most) famous classical author whose depiction of Romano-Germanic relations in the first century AD can largely be trusted and who happens to share the speaker’s evaluation. In terms of logical stringency, such an ‘anticipatory endorsement’ is of course imaginary, but its appeal lies in the extra-kudos conveyed onto the speaker/quoter’s stance by an argument from an (ancient) authority. Most articles in Die Welt attempt to buttress inferences from their Tacitus quotations through specific text references and explicit contextualisations, so that

Quotation and Online Identity   

   141

the circularity of argument is at least mitigated by some quasi-philological, textcritical information. Many contributors in the online forum debates, however, are not content with showing that they know ‘their’ Tacitus but also adopt his stance and voice it as their own. This stance-adoption can go so far as using a matrix clause, which introduces a direct quotation as a speaker’s online name (example (6): Thus wrote Tacitus). However, even without such explicit signalling of voicecongruence or identity, most forum quoters strive to gain credibility for their own specific stance and identity vis-à-vis other discussants in the forum. Like Thus wrote Tacitus, quoters Heinrich, wjbeckermann, Teiwswitan and Thusnelda(-in heaven) all compete with other online discussants for their own distinctive voice (and in addition, wjbeckermann has a vested interest in selling his special translation of Tacitus’s writings). They thus exploit the metarepresentational structure of quotations for their own discursive interests as members of the online forum. The reader is not only invited to recognise or assume a sufficient formal and interpretive resemblance between Tacitus’s original text and the quotation, as in the journalistic articles, but also to accept the referential and propositional contents of the quotation as largely correct in terms of historical reality, and therefore as vindicating the quoter’s stance. With regard to the conceptual architecture that is needed to account for the reader’s comprehension of these implied pragmatic inferences, we can unpack them in the form of a mini-syllogism: Premise 1 (explicit quotation text): Speaker/quoter X claims or assumes that her readers know that Tacitus said Y (about Arminius and/or ancient Germany) and she agrees with Y. Premise 2 (usually tacitly assumed): Tacitus as a famous classical historian can be trusted to give a correct account of Y. Conclusion: Therefore, X’s own proposition, i.e., ‘that Y’, can be trusted. The quoters’ identification with Tacitus’s voice does not even have to entail absolute congruence of stance to confer authority to her own voice: Thusnelda’s and German’s contributions, for instance, include register choices (e.g. colloquialisms, interjections), irony markers and propositional content that invite the reader to take their utterances with a pinch of salt. Heinrich, wjbeckermann, Teiwswitan and historiker’s quotations are at the other end of the spectrum of selfimportance, as they expect to be taken absolutely seriously as conveying Tacitus’s true meaning. Thus wrote Tacitus occupies a special position insofar as the quoted text in example (6) was selected for propositional content that contradicts the ‘patriotic’ quotation practice of others. By counter-selecting passages from

142   

   Andreas Musolff

Tacitus’s Germania that are neglected by most other quoters, Thus wrote Tacitus represents a subversive voice in the online discussion, utilising her Tacitus-selection to contradict the others’ stances. In conclusion, we have found that the quoters in the forum discussion use their Tacitus quotations not just to back up their own stance on the topic but also to assert a specific identity vis-à-vis other discussants in the online-discourse. By aligning their voices with that of Tacitus, they invite their readers to pay attention to the presupposed resemblance of their quotation and the Tacitean original not just as a philological exercise but as a constitutive aspect of their discursive identities. If a speaker fails to achieve this metarepresentational resemblance in the eyes of other contributors in the community of practice, she has not only made the mistake of quoting incorrectly but has undermined her own voice, and thus her online identity in the forum discussion, as is demonstrated in the disputes between Thus wrote Tacitus, Teiwswitan and German, and between Thusnelda and historiker2. Thus wrote Tacitus’s selection of passages from the Germania that criticise the ancient Germans is rejected and mocked by Teiwswitan and German for taking the wrong side in an (imagined) conflict between hardy German(ic) forest-dwellers and decadent Romans/present-day Germans. As a result, Thus wrote Tacitus was silenced in the forum debate. Thusnelda only fares marginally better. When historiker2 lectures her on alleged oversights in her retelling of the Arminius-Thusnelda-Thumelicus story, she tries to reassert her identity by trumping all accounts of Thusnelda’s fate through imagining her after-life insights ‘in heaven’. Whilst having the last word in this particular dispute, she is absent from the remainder of the online discussion. It would appear that quoting an ancient classical authority such as Tacitus in online debates is still a serious undertaking, despite the occasional insertion of irony.

References Angouri, Jo and Theodora Tseliga. 2010. “‘You have no idea what you are talking about!’ From e-disagreement to e-impoliteness in two online fora”. Journal of Politeness Research: Language, Behaviour, Culture 6/1, 57–82. Bendikowski, Tillmann. 2008. Der Tag, an dem Deutschland entstand. Geschichte der Varusschlacht. München: C. Bertelsmann. Berhorst, Robert. 2008. “Tussi – Thusnelda – Hermann”. http://web.archive.org/ web/20090302081704/http://uni-giessen.de/~g41007/tussi.html, last accessed 5 Oct 2012. Berke, Stephan. 2009. “‘Haud Procul’. Die Suche nach der Örtlichkeit der Varusschlacht”, in: Landesverband Lippe (ed.). 2000 Jahre Varusschlacht: Mythos. Darmstadt: Wissenschaftliche Buchgesellschaft, 133–138.

Quotation and Online Identity   

   143

Brown, Penelope and Stephen C. Levinson. 1987. Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Clunn, Anthon. 1999. The Quest for the Lost Legions. London: Minerva. Der Spiegel. 2008. “Die Geburt der Deutschen”. Der Spiegel 51/2008, 15 Dec 2008. Derks, Heidrun. 2009. “Die Varusschlacht”, in: Varusschlacht im Osnabrücker Land GmbH – Museum und Park Kalkriese (ed.). 2000 Jahre Varusschlacht: Konflikt. Darmstadt: Wissenschaftliche Buchgesellschaft, 37–55. Dörner, Andreas. 1995. Politischer Mythos und symbolische Politik. Sinnstiftung durch symbolische Formen am Beispiel des Hermannsmythos. Opladen: Westdeutscher Verlag. Du Bois, John W. 2007. “The stance triangle”, in: Robert Englebretson (ed.). Stancetaking in Discourse: Subjectivity, Evaluation, Interaction. Amsterdam: John Benjamins, 139–182. Goffman, Erving. 1972. “On face work: an analysis of ritual elements in social interaction”, in: John Laver and Sandy Hutcheson (eds). Communication in Face to Face Interaction. Harmondsworth: Penguin, 319–347. [First published 1955 in Psychiatry 18, 213–231.] Günnewig, Beatrix. 2009. “Zum Germanenbild der Römer aus literarischer Perspektive”, in: Landesverband Lippe (ed.). 2000 Jahre Varusschlacht: Mythos. Darmstadt: Wissenschaftliche Buchgesellschaft, 30–34. Halle, Uta. 2009. “Ideologisierung und Politisierung. Zur Vereinnahmung der prähistorischen Archäologie durch Ideologie und Politik im 19. und 20. Jahrhundert”, in: Landesverband Lippe (ed.). 2000 Jahre Varusschlacht: Mythos. Darmstadt: Wissenschaftliche Buchgesellschaft, 243–252. Hardaker, Claire. 2010. “Trolling in asynchronous computer-mediated communication: from user discussions to academic definitions”. Journal of Politeness Research: Language, Behaviour, Culture 6/2, 215–242. Holmes, Janet and Miriam Meyerhoff. 1999. “The community of practice: theories and methodologies”. Language in Society 28, 173–183. Husemann, Dirk. 2009. “Wie die Deutschen die Varusschlacht zurechtbogen”. Die Welt, 14 Jan 2009. Kellerhoff, Sven Felix. 2009. “In Varus’ Germanien war nicht alles schlecht”. Die Welt, 6 Jan 2009. Krebs, Christopher B. 2011. A Most Dangerous Book: Tacitus’s Germania from the Roman Empire to the Third Reich. New York: W.W. Norton and Company. Landesverband Lippe, Hermannsdenkmal-Stiftung (ed.). 2009. Das Hermannsdenkmal Geschichte erleben! http://www.hermannsdenkmal.de/index2.htm, last accessed 5 Oct 2012. Lippe Tourismus and Marketing AG (ed.). 2009. Das Internationale Kulturprogramm Hermann 2009. Detmold: Hermann Bösmann. Mellies, Dirk. 2009a. “‘Symbol deutscher Einheit’. Die Einweihungsfeier des Hermannsdenkmals 1875”, in: Landesverband Lippe (ed.). 2000 Jahre Varusschlacht: Mythos. Darmstadt: Wissenschaftliche Buchgesellschaft, 222–228. Mellies, Dirk. 2009b. “Politische Feiern am Hermannsdenkmal nach 1875”, in: Landesverband Lippe (ed.). 2000 Jahre Varusschlacht: Mythos. Darmstadt: Wissenschaftliche Buchgesellschaft, 263–272. Musolff, Andreas. 2010. “Hermann der Cherusker und das Internet – ein Fall für die Sprachkritik?”. Aptum. Zeitschrift für Sprachkritik und Sprachkultur 6/3, 214–229.

144   

   Andreas Musolff

Musolff, Andreas. 2012. “From teamchef Arminius to Hermann junior: glocalised discourse about a national foundation myth”. Language and Intercultural Communication 12/1, 24–36. Myers, Greg. 2010. The Discourse of Blogs and Wikis. London/New York: Continuum. Niebuhr, Hermann and Andreas Ruppert (eds). 1998. Nationalsozialismus in Detmold. Dokumentation eines stadtgeschichtlichen Projekts. Bielefeld: Aisthesis-Verlag. Niebuhr, Hermann and Andreas Ruppert (eds). 2007. Krieg – Revolution – Republik. Detmold 1914 – 1933. Dokumentation eines stadtgeschichtlichen Projekts. Bielefeld: AisthesisVerlag. Ottomeyer, Hans. 2009. “Die Erfindung der deutschen Nation”, in: Landesverband Lippe (ed.). 2000 Jahre Varusschlacht: Mythos. Darmstadt: Wissenschaftliche Buchgesellschaft, 140–148. Perelman, Chaim and Lucie Olbrechts-Tyceca. 1971. The New Rhetoric. A Treatise on Argumentation. Notre Dame/London: University of Notre Dame Press. Schlüter, Wolfgang (ed.). 1992. Archäologische Zeugnisse zur Varusschlacht? Die Untersuchungen in der Kalkrieser-Niewedder Senke bei Osnabrück. (Sonderdruck aus Germania 70/2 Halbband) .Mainz am Rhein: Verlag Philipp von Zabern. Seewald, Berthold. 2008. “Schlachtfeld entdeckt. Hier metzelten Römer die Germanen nieder”. Die Welt. 15 Dec 2008. Seewald, Berthold. 2009a. “Hermann und seine blonden Horden”. Die Welt. 15 May 2009. Seewald, Berthold. 2009b. “Als Urfaust untauglich”. Die Welt. 16 May 2009. Seewald, Berthold. 2009c. “Die Varusschlacht taugt uns nicht als Gedenktag”. Die Welt. 16 May 2009. Sperber, Dan and Deirdre Wilson. 1995. Relevance. Communication and Cognition. 2nd ed. Oxford: Blackwell. Strabon [= Strabon aus Amaseia]. Stefan Radt (ed.). 2003. Geographica. Griechisch/Deutsch. Göttingen: Vandenhoeck & Ruprecht. Tacitus, Publius Cornelius. 1999. Agricola, Germany. Translation, introduction and notes by Anthony R. Birley. Oxford: Oxford University Press. Tacitus Publius Cornelius. 2000. Germania. De origine et situ Germanorum liber. Lateinisch und Deutsch. Translation and editing by Manfred Fuhrmann. Stuttgart: Reclam. Tacitus, Publius Cornelius. 2005. “Annales/The Annals, Books I–III”, in: Clifford H. Moore and John Jackson. The Histories: Books IV–V, The Annals: Books I–III, tr. Cambridge: Harvard University Press, 225–634. Tannen, Deborah. 2007. Talking Voices: Repetition, Dialogue and Imagery in Conversational Discourse. 2nd ed. Cambridge: Cambridge University Press. Vandergriff, Ilona. 2012. “Taking a stance on stance: metastancing as legitimation”. Critical Approaches to Discourse Analysis Across Disciplines 6/1, 53–75. Varusschlacht im Osnabrücker Land (ed.). 2009. Museum und Park Kalkriese. www.kalkriese-varusschlacht.de, last accessed 5 Oct 2012. Watts, Richard J. 2003. Politeness. Cambridge: Cambridge University Press. Watts, Richard J., Sachiko Ide and Konrad Ehlich (eds). 2005. Politeness in Language. Studies in Its History, Theory and Practice. 2nd ed. Berlin: de Gruyter. Wells, Peter. 2001. Beyond Celts, Germans and Scythians: Archaeology and Identity in Iron Age Europe. London: Duckworth Academic. Wenger, Etienne. 1999. Communities of Practice: Learning, Meaning and Identity. Cambridge: Cambridge University Press.

Quotation and Online Identity   

   145

Werkhofer, Konrad T. 2005. “Traditional and modern views: the social constitution and power of politeness”, in: Richard J. Watts, Sachiko Ide and Konrad Ehlich (eds). Politeness in Language. Studies in its History, Theory and Practice. Berlin: de Gruyter, 155–199. Wilson, Deirdre. 2000. “Metarepresentation in linguistic communication”, in: Dan Sperber (ed.). Metaprepresentations: A Multidisciplinary Perspective. Oxford: Oxford University Press, 411–448. Wolters, Reinhard. 2008. Die Schlacht im Teutoburger Wald. Arminius, Varus und das römische Germanien. München: C. H. Beck. Yus, Francisco. 2011. Cyberpragmatics: Internet-mediated Communication in Context. Amsterdam: John Benjamins.

Rita Finkbeiner

6 “Ich kenne da so einen Jungen … kennen ist gut, wir waren halt mal zusammen weg.” On the Pragmatics and Metapragmatics of X ist gut in German¹ Abstract: The German construction X ist gut (‘X is good’), with X being a mentioned expression, is used meta-communicatively to negotiate situated word meaning. On the basis of examples from online discussion fora, this paper aims at a comprehensive pragmatic analysis of the formula. Taking the quotational framework of Gutzmann/Stei (2011) as a starting point, I argue that there are two pragmatic indicators in the construction: quotation marks enclosing X (optionally) and the formulaic predication ist gut. It is shown that both indicators can block the stereotypical interpretation of X, rendering pragmatic effects of contextshift and modalization. It is claimed that there is no interpretational difference between “X” ist gut and X ist gut. The analysis of the formulaic predication as a pragmatic indicator explains why the quotation marks may be abandoned. Keywords: conversational implicature, formulaic sequence, meta-communication, mentioned expression, quotation marks, pragmatic marker

1 Introduction Recent pragmatic, cognitive and interactional approaches to lexical meaning emphasize the dynamic character of meaning constitution in interaction (Linell 1998, Fauconnier/Turner 2003, Recanati 2004). In theoretical pragmatics, both Neo-Griceans and Relevance theorists agree that propositional structures are systematically underdetermined (Levinson 2000, Carston 2002). However, while these approaches try to keep semantics ‘minimal’, accounting for meaning constitution by processes of pragmatic enrichment, a common assumption of ‘maximalist’ approaches is that semantic entries involve a considerable extent of context-related information. In maximalist approaches, lexical meanings are perceived of as meaning potentials (Norén/Linell 2007) or ranges of ‘available’

1 I would like to thank Jenny Arendholz, Wolfram Bublitz, Franz D’Avis, Monika Kirner-Ludwig, Jörg Meibauer, and an anonymous reviewer for helpful comments on earlier versions of this paper. Thanks to John Weir for helping me with my English.

148   

   Rita Finkbeiner

senses (Recanati 2004: 20), including language users’ stored knowledge about recurrent specifications of meaning. Thus, a maximalist view of lexical meaning would assume that a lexical unit not only contains a (minimal) core meaning, but also vast experiential knowledge about what in the world usually is being described by this lexical unit under particular circumstances. If we adopt this potentiality view of lexical meaning, we can expect that interactive clarification of situated word meaning is a basic process in everyday communication. If one can mean different things by one and the same word in different situations, then it is evident that, at certain points in interaction, the need to make the choice of a word itself the topic of communication emerges. Indeed, clarification of situational meaning or, more generally, comprehensibility enhancement (‘Verständnissicherung’) has been described as a keystone of talk in interaction (Hagemann 1997, Bublitz 2001, Techtmeier 2001, Fiehler 2009). There are many different ways in which language users can comment on wordings. With respect to German, a range of conventional linguistic expressions have been described which can be used to prospectively comment on the subsequent utterance, e.g., offen gesagt (‘honestly’, ‘to tell you the truth’), mit anderen Worten (‘in other words’), vereinfachend ausgedrückt (‘to put it in simple terms’), etc. (e.g., Hindelang 1975, Gülich/Kotschi 1987, Hagemann 1997). However, there is rather little research on conventional linguistic expressions which are used retrospectively to comment on expressions already uttered in previous discourse (but see Hrncal 2011, Finkbeiner 2012). One example is the X ist gut (‘X is good’) construction, which to date has not been comprehensively described in the literature. Consider example (1). (1)

Max:

Aber Dir kann das doch egal sein, als Rentner kannste doch solange Urlaub machen wie Du willst. ‘But never mind, as a pensioner, you can take vacation as long as you like’

Bruno:

Moin Max, Rentner ist gut. Jetzt habe ich noch weniger Zeit, weil ich mein Hobby zum Beruf gemacht habe :>)) ‘Hi Max, pensioner is good [roughly: ‘what do you mean: “pensioner”?’]. Now I have even less time because I turned my hobby into a profession :>))’ (cf. URL 1)

In Example (1), Max refers to Bruno as a pensioner (als Rentner). In his response Rentner ist gut, Bruno conveys that he finds Max’s choice of the expression Rentner not quite appropriate in this situation. In the subsequent discourse, Bruno elabo-

Pragmatics and Metapragmatics of X ist gut   

   149

rates on the reasons for this evaluation. It becomes clear that in Bruno’s eyes, he does not fulfill a central condition of the expression Rentner to be used felicitously in the actual situation, namely, to be a person with a lot of leisure time. In other words, by using the construction, Bruno conveys a reservation against the applicability of the word Rentner to its denotation in this particular situation. The example points to several interesting semantic and pragmatic characteristics of the X ist gut-construction. First, it contains an element X which is repeated from an utterance in the previous turn. However, we do not find quotation marks or other forms of graphic marking which would signal that X is a mentioned expression. In other instances of X ist gut, in contrast, we do find quotation marks (e.g., cf. Example (4) below). Second, the predicate gut in the construction does not have its literal meaning ‘good’, it rather means (roughly) the opposite, namely, that the X-item in some (meta-linguistic) respect is ‘not good’. That is, although we have a positively evaluating lexeme gut, the speaker is partly or completely refusing the situational usage of X. Third, the construction seems to be highly formulaic or conventional, both in the sense that its meaning is not transparent (e.g., gut has a non-literal meaning), and in the sense that its usage is restricted to certain discursive settings (e.g., to a responsive sequential position). The overall goal of this paper is to give a detailed, empirically based pragmatic analysis of X ist gut which can account for its specific meta-communicative function. I will put forward the claim that both “X” ist gut (with quotation marks) and X ist gut (without quotation marks) generally render the same types of interpretation. They only differ in the kind of pragmatic markers which are at work, namely, quotation marks and/or the formula itself. The structure of the article is as follows: after a short description of the data (Section 2), I characterize the central formal features of the X ist gut construction (Section 3). In Section 4, I develop a pragmatic account of X ist gut. First, I describe its meaning potential in more detail (4.1). I then introduce the quotational framework (4.2), which is applied to instances of “X” ist gut (with quotation marks) (4.3). Section 4.4 extends the analysis to instances without quotation marks. It is shown that X ist gut can be regarded as a formulaic sequence in the sense of Wray (2002) (4.4.1), and it is argued that the formula has the status of a non-minimal pragmatic marker (4.4.2). Then, morphological restrictions on X are discussed in 4.4.3. Finally, in Section 4.5, I discuss the role of quotation marks in the construction more generally. Section 5 provides a short summary.

150   

   Rita Finkbeiner

2 Data The paper draws on a corpus of 54 examples of X ist gut, which I collected via specified searches in Google. The great majority of the examples stem from German-language online discussion fora. This collection of examples serves as illustration of the construction’s main usage patterns, which are analyzed qualitatively, and does not claim to be representative in a quantitative sense. Using the internet as a source addresses two assumptions: first, the assumption that the construction is relatively infrequent in actual performance, because it is bound to highly specific situational conditions, which ultimately are of a meta-linguistic nature (see Sections 3 and 4). The internet is large enough to yield a reasonable amount of hits. Second, the assumption that the construction appears mainly in informal spoken discourse. Internet forum communication is relatively close to orality, as it constitutes a form of (asynchronous) dialogical communication (cf. Dürscheid 2003). In addition to the internet examples, I occasionally draw on introspective examples when suitable for the line of argument.² For reasons of lucidity, the examples presented are typographically standardized. Henceforth, I will use X ist gut when referring to the construction in general. Only when relevant for the actual thread of discussion will I distinguish between “X” ist gut and X ist gut.

3 Formal Features of X ist gut The X ist gut construction can be described syntactically as a declarative sentence containing an open subject slot (X), which is followed by the third person singular present copula verb ist (‘is’) and the uninflected adjective gut (‘good’) in predicative function. Further formal features concern the formula’s sequential position, the possible X-categories, morphological/morpho-syntactic features of X and the use of quotation marks. The sequential position of the sentence is either turn-initial, as in (2) (other-quotational use), or turn-internal, as in (3) (selfquotational use). (2)

A: Da habe ich ja nen richtigen linux-professor erwischt um so besser ‘I must have bumped into a real linux professor, all the better’ [referring to B]

2 Throughout the article, I provide the URL with the internet examples. When no URL is provided, the example is an introspective one.

Pragmatics and Metapragmatics of X ist gut   

   151

B: Linux-Professor ist gut. Sag das mal meinem Prof in Unix-Werkzeuge. Der wird sich totlachen. ‘Linux-Professor is good. Tell that to my Unix-tools professor. He will die laughing’ (cf. URL 2) (3)

Nebenher (nebenher ist gut – hauptsächlich triffts wohl eher) drehe ich Lehrvideos ‘Alongside [my regular occupation] (alongside is good – main fits better) I shoot educational videos’ (cf. URL 3)

In self-quotational use, the construction either follows immediately after the first mention of X (cf. (3)) or with some distance in between (cf. (4)). (4)

Hatte ja gestern meinen Januar VU-Termin und es ist alles wunderbar und bestens mit der kleinen Maus. Wobei “klein” ist gut, sie ist ganz schön gut dabei ‘Yesterday, [I] had my January check-up appointment and everything is wonderful and great with the little sweetie pie. Well, “little” is good, she is actually quite chubby’ (cf. URL 4)

The X-slot can be filled by a range of different syntactic categories, which may also differ in complexity. As simple lexical categories, there are both nouns (cf. (1), (2)), adverbs (cf. (3)), adjectives (cf. (4)), verbs (cf. (5)), and particles (cf. (6)) in the corpus. (5)

Ich kenne da so einen Jungen … Kennen ist gut, wir waren halt mal zusammen weg ‘I know this guy … Know is good, we have [just] been on one date’ (cf. URL 5)

(6)

haballes: Voraussetzung ist hier nur eine Internetverbindung ‘The only precondition is an internet connection’ inpraxi:

“Nur” ist gut :D Aber es wird tatsächlich von Jahr zu Jahr besser mit WLAN überall ‘“Only” is good :D But actually, it does get better every year with WLAN everywhere’ (cf. URL 6)

152   

   Rita Finkbeiner

In addition, X may also be filled by complex categories. For example, there are noun phrases (cf. (7)), verb phrases (cf. (8)), prepositional phrases (cf. (9)), and adjective phrases in the corpus (cf. (10), where the AP functions as an adverbial). (7)

Aldiboz:

scheint ein kleines problem vorzuliegen… ‘We seem to have a little problem here…’

Gismaxx66: Hi, kleines Problem ist gut erreichen. ‘Hi, little problem is good the moment’ (cf. URL 7) (8)

Zitat SAZ12:

Der Server ist derzeit nicht zu The server is not accessible at

ich weiß nicht ob ich hier noch etwas ändern muss, ‘I don’t know whether I have to change anything else here’

Bewerbungshelfer: “Noch etwas ändern müssen” ist gut… mit dem, was den Leser interessiert, hast Du ja noch gar nicht angefangen ‘“Have to change anything else” is good… you haven’t even started with what the reader is really interested in’ (cf. URL 8) (9)

(10)

Zitat GeNeTiX:

Mit etwas Pech sind die eine Hälfte irgendwelche Manga-Freaks ‘If you are unlucky [‘with a little bad luck’] half of them are some kind of Manga freaks’

Shakkuri:

Mit ‘etwas Pech’ ist gut Ich würde eher sagen: ‘mit Sicherheit!’ ‘With ‘a little bad luck’ is good I would rather say: ‘with certainty’ [‘for sure’]’ (cf. URL 9)

michigrunf: also kannst du ganz normal Video for Windows (VfW) oder WIA oder WDM(?) nutzen ‘So you can use Video for Windows (VfW) or WIA or WDM(?) as usual [‘quite normally’]’

Pragmatics and Metapragmatics of X ist gut   

   153

dutschman: Ganz normal ist gut. Kannst du mir eine konkrete Anleitung geben? ‘Quite normally is good. Can you provide detailed instructions?’ (cf. URL 10) As to the morpho-syntactic features of the formula, one can observe specific differences between the original and the repeated X. In (4), the original inflected form of the adjective, kleinen (dat. sing. fem.) is changed into the uninflected form klein. In (8) and (11), the originally inflected verb forms muss (first person singular present tense, cf. (8)) and mußte (third person singular past tense, cf. (11)) are changed into the infinitive form müssen. In (12)-(13), the determiner phrase (die Eltern, der strenge Onkel) is changed into a bare noun phrase (Eltern, strenger Onkel). (11)

Zitat Chaos: und der mußte mit 5 sogar schon alleine einkaufen gehen ‘And he [the writer’s son] already had to go shopping by himself at the age of 5’ loewin68:

(12)

(13)

müssen ist gut – meine durften auch müssen – und waren unheimlich stolz, als sie dann auch noch das richtige nachhause gebracht haben. ‘Have to is good – mine were also allowed to go – and were extremely proud when they came home with the right stuff’ (cf. URL 11)

Heimü:

so das kommt davon. Man sollte die Eltern ich weiß nicht was ‘So that’s what you get from it. I don’t know what one should do with the parents …’

Ccorrel:

Naja, Eltern ist gut, der Trainer hat ihn adoptiert. ‘Well, parents is good, the coach adopted him.’ (cf. URL 12)

Jörg: Ich komme mir schon vor wie der “strenge Onkel”, den keiner mag ‘I already feel like the “strict uncle” nobody likes’ Shira: Hallo Jörg, “strenger Onkel” ist gut!! ‘Hi Jörg, “strict uncle” is good’ (cf. URL 13)

154   

   Rita Finkbeiner

As to the use of quotation marks, the examples mentioned so far show that the X-expression can be used either with (single or double) quotation marks (cf. (4), (6), (8), (9), (13)) or without (cf. (1)–(3), (5), (7), (10)–(12)). Moreover, there are cases where the quotation marks only enclose part of the (complex) X-expression, cf. (14) (see also (9)). (14)

Lightstorm: Genau deshalb frage ich erstaunt […] Das wäre schon etwas lahm. ‘That’s why I ask so astonishedly […] That would be quite slow indeed’ Thawra:

‘Etwas’ lahm ist gut… da wären ja Festplatten vor 15 Jahren schneller gewesen ‘‘Quite’ slow is good… hard disks 15 years ago would have been faster’ (cf. URL 14)

To sum up the formal features, X ist gut is a declarative sentence with an open X-slot sequentially used in reaction to the use of X in previous discourse. The X-slot can be filled by different syntactic categories, which may be simple lexical items or phrasal categories. The X-expression is lexically identical with some previously used item, but it may differ from it morphologically or morpho-syntactically in specific ways. Furthermore, the X-expression or parts of it may or may not be enclosed in quotation marks.

4 A Pragmatic Account In the following sections, I will argue that the formal characteristics of X ist gut can be accounted for pragmatically. I will first discuss the meaning of the construction in more detail. Then, I will apply the quotational framework of Gutzmann/Stei (2011) to account for the meaning constitution of X ist gut. Two types of pragmatic indicators will be discussed: quotation marks and the sentential formula itself. This will lead to the question of whether the quotation marks in the construction are redundant.

Pragmatics and Metapragmatics of X ist gut   

   155

4.1 Meaning What do speakers/writers convey by uttering X ist gut? Consider, once again, examples (2), (4), and (5), represented here as (15), (16), and (17). (15)

A: [mit Bezug auf B:] Da habe ich ja nen richtigen linux-professor erwischt um so besser [referring to B:] ‘I must have bumped into a real linux professor, all the better’ B: Linux-Professor ist gut. Sag das Mal meinem Prof in Unix-Werkzeuge. Der wird sich totlachen. Es mag scheinen, dass ich viel über Linux weiß, tatsächlich bin ich noch ziemlich unerfahren in GNU/Linux. ‘Linux-Professor is good. Tell that to my Unix-tools professor. He will die laughing. It may seem that I know a lot about Linux, but actually I’m rather unacquainted with GNU/Linux.’

(16)

Hatte ja gestern meinen Januar VU-Termin und es ist alles wunderbar und bestens mit der kleinen Maus. Wobei “klein” ist gut, sie ist ganz schön gut dabei ‘Yesterday, [I] had my January check-up appointment and everything is wonderful and great with the little sweetie pie. Well, “little” is good, she is actually quite chubby’

(17)

Ich kenne da so einen Jungen … Kennen ist gut, wir waren halt mal zusammen weg ‘I know this guy … Know is good, we have [just] been on one date’

In these examples, by uttering X ist gut, the writer conveys the attitude that the word in the X-slot is in mismatch with the situation it has been applied to by the original speaker. In (15), it becomes clear from the context that B is a competent user of Linux, but not an expert. In (16), we learn from the context that the little girl is little only by age, but not by weight or size. And in (17), which may be regarded as an instance of self-repair, the writer conveys that she does not know the guy in question very well, but rather superficially. The examples show that despite its making reference to some original utterance, the formula does not have a reporting function, in contrast to direct and indirect quotation. Rather, by uttering X ist gut, the speaker/writer mentions some previously used linguistic item, aiming at making this expression, or some of its linguistic properties, the topic of sub-

156   

   Rita Finkbeiner

sequent conversation.³ In this respect, the formula can be regarded as a metacommunicative construction. In terms of semantic extension and intension, one could say that by X ist gut, the speaker/writer conveys that the entity which was denoted by “X” in its previous usage should be regarded as a non-stereotypical member of the extension of X, because it does not meet all the requirements which are posed on X by the intension of X (cf. Klockow 1980: 178). As a first paraphrase, I will assume the following (where W stands for the writer): Paraphrase 1: ‘W wants to convey that the object / person / activity / state/ … denoted / characterized / described / alluded to / … by the expression “X” is not a proper / real X [for X = noun], or is not properly / really X [for X = adjective/verb].’

In other words, the writer conveys a reservation against the applicability of X to its denotation in the actual context (‘Applikationsvorbehalt’, cf. Klockow 1980: 177). In terms of stance, the writer distances himself or herself from the original writer’s choice of words. However, Paraphrase 1 does not fit all cases of X ist gut. Consider (18) – (19): (18)

Zitat David: Demnach würde ich mal behaupten, dass unser aller Onkel mit hoher Wahrscheinlichkeit etwas Ordentliches auf die Beine gestellt hat ‘I would claim that the uncle to us all most likely got some real stuff off the ground ’ Onkelfitty: “unser aller Onkel” ist gut! ‘“uncle to us all” is good! (cf. URL 15)

(19)

Zitat Sally_brown:



Das nächste Mal werd ich höflich aber bestimmt sagen, dass er ein bißchen Abstand halten soll, weil ich sonst Ausschlag kriege … ;-) ‘Next time I will tell him politely but firmly that he’d better keep his distance, as otherwise I will get a skin rash’

3 Cf., e.g., Recanati (2000) on the use/mention distinction.

Pragmatics and Metapragmatics of X ist gut   

Tinakind:

   157

Ausschlag ist gut ☺ stell mir grad eine mit Pusteln übersäte Sally vor … ☺ hehehe ‘Skin rash is good ☺ I am imagining a Sally littered with pustules right now … ☺ hehehe’ (cf. URL 16)

In (18), “Onkelfitty”, who has been called “unser aller Onkel” by David, does not want to convey a reservation against this naming. Rather, he seems to appreciate this choice of words, as is indicated by the laughing and clapping emoticons after the construction. Similarly, in (19), “Tinakind” does not distance herself from the (extensional) applicability of the expression Ausschlag to the state described by Sally. Rather, she conveys that there is something special, in a positive sense, with this choice of words: for instance, that it yields a funny pictorial association. Also in (19), there are smiling emoticons as well as verbalized laughing (“hehehe”) which support this reading. Thus, also in (18) and (19), the speaker conveys that X has been used in a non-stereotypical way. However, the markedness in these cases is not due to a perceived semantic mismatch between the expression and its denotation, but due to a perceived special stylistic or expressive value of the expression. In terms of speaker attitude, what is conveyed is a positive stance towards the original writer’s choice of words. As a paraphrase for these cases I will assume the following: Paraphrase 2: ‘W wants to convey that the expression “X” is more innovative/expressive/stylistically apt… in this utterance situation than a potential synonymous expression.’

In sum, the quotational formula X ist gut can be said to function as a metacommunicative comment on some kind of non-stereotypical word usage, either concerning the semantic relation between X and its denotation or concerning a special stylistic effect of X. As the examples show, in the former case, the original writer’s choice of words is evaluated negatively, while in the latter case, it is evaluated positively.

4.2 Quotational Framework In order to account for this general meaning potential of the construction, I will adopt the quotational framework of Gutzmann/Stei (2011). I will first introduce the framework and then develop an application to the case of “X” ist gut.

158   

   Rita Finkbeiner

In their quotational approach, Gutzmann/Stei (2011) develop a unified pragmatic analysis of quotation, starting from the assumption that quotation is “everything between a pair of quotation marks” (Gutzmann/Stei 2011: 2651). The analysis assigns quotation marks the role of minimal pragmatic indicators which, from the reader’s perspective, “block the stereotypical interpretation of an expression, and thereby indicate that some alternative meaning ought to be inferred” (Gutzmann/Stei 2011: 2651). From the writer’s perspective, quotation marks “are used to convey that a non-standard interpretation of the quoted expression is required” (Gutzmann/Stei 2011: 2653). Given a Gricean cooperative principle and Horn’s Q- and I-principle, a competent writer would not write “x” in case she wanted to convey only the standard interpretation of x. Analogously, a competent reader would not expect “x” to convey only the standard interpretation of x. (Gutzmann/Stei 2011: 2653)

Starting from there, Gutzmann/Stei sketch a two-staged derivational process. In the first stage, the quotation marks block the standard interpretation of the quoted expression. In the second stage, the specific contextual meaning of the sentence is inferred via conversational implicature. For example, in (20), the quotation marks first indicate to the reader that the expression enclosed by quotation marks, “theory”, is to be interpreted in a non-standard way (i.e., not in the same way as theory). (20)

Peter’s “theory” is difficult to understand.

In the second stage, the reader can infer the specific contextual meaning of the sentence via conversational implicature; for instance, the interpretation that Peter’s proposal (which Peter himself might have called a theory) is not a theory. In a similar fashion, Klockow (1980) characterizes the function of scare quotes (he calls them “modalisierende AZ”, ‘modalizing quotes’) (see also Meibauer, this volume): Der leser wird aufgefordert, aus dem gebrauch des ausdrucks unter den gegebenen situationsbedingungen nicht all die oder genau die schlüsse zu ziehen, die er normalerweise daraus ziehen würde. (Klockow 1980: 131 [original without capitalization])⁴

4 ‘The reader is required to draw not all or not exactly the same inferences from the usage of the expression under the given situational conditions that he normally would draw from it.’

Pragmatics and Metapragmatics of X ist gut   

   159

While Klockow restricts this description to scare quotes, Gutzmann/Stei claim that it is valid for all kinds of quotations.⁵ For example, in pure quotation such as (21), the marking effect of the quotation marks would draw the reader’s attention to the fact that the expression is to be interpreted in a non-standard way, namely, as a mentioned expression. (21)

“Boston” is disyllabic.

As becomes clear from this short outline, quotation marks are the central starting point of Gutzmann/Stei’s analysis of quotation. This may seem to pose a problem for the application of this framework to the case of X ist gut, because, as pointed out in Section 3, the construction is only optionally used with quotation marks (cf. Section 3). To resolve this problem, I will argue that there are two different types of pragmatic indicators, both of which can have the effect of blocking the stereotypical interpretation of an expression: quotation marks and the sentential formula. I will first discuss cases with quotation marks (“X” ist gut) and then turn to cases without quotation marks (X ist gut).

4.3 “X” ist gut Gutzmann/Stei’s (2011) approach perceives quotation marks as pragmatic devices which are used to convey that a non-standard interpretation of the quoted expression is required (given that the cooperative principle is at work, and given Horn’s Q- and I-principle). This analysis seems to fit well with the paraphrases for “X” ist gut developed above. Comparing the formula to the scare quote cases discussed in Gutzmann/Stei (2011), one finds similarities, but also interesting differences. Consider (22) and (23). (22)

Peters “Theorie” ist schwer zu verstehen. ‘Peter’s “theory” is difficult to understand.’

5 Gutzmann/Stei (2011: 2651) differentiate between five varieties of quotation, exemplified by the following examples: (a) “Boston” is disyllabic. (pure quotation); (b) “We may need it for taxes since we have no idea what we’re looking at next year”, Ms. Felder said. (direct quotation); (c) Harvey said an inquiry would not be ruled out, “should serious and systemic issues” emerge as a result of the MoD’s own investigations. (mixed quotation). (d) The “debate” resulted in three cracked heads and two broken noses. (scare quotes). (e) We sell “fresh” pastry. (emphatic quotes). See also Brendel et al. (2011) for an overview of philosophical and linguistic approaches to quotation.

160   

(23)

   Rita Finkbeiner

“Theorie” ist gut. ‘“Theory” is good.’

In both examples, the writer conveys that the expression Theorie (‘theory’) is to be interpreted in a modalized, non-standard way. However, while there is an obligatory context-shift in (23), i.e. reference is made to some previous usage of the quoted expression, in the scare quotation (22), context-shift is only one of several possible interpretations. Thus, (23) can be felicitously used only as a reaction to a certain previous usage of X. In contrast, (22) may be used without any context-shift, e.g., to convey that the writer/speaker is aware that Peter’s proposal is not a theory in the proper sense of the word, but that she just could not find a better word and therefore used theory as an approximation (see also Predelli 2003). Another difference is that in (23), the quoted expression does not contribute to the denotation of the sentence in which it is embedded, whereas in (22), it does. This can be shown by anaphora tests, which reflect the distinction between used (24) and mentioned expressions (25). (24)

Peters “Theorie”i ist schwer zu verstehen, aber Karla versteht siei trotzdem. ‘Peter’s “theory”i is difficult to understand, but Karla understands iti anyway.’

(25)

* “Theorie”i ist gut, aber siei ist schwer zu verstehen. ‘“Theory”i is good, but iti is difficult to understand.’

There are, intuitively, two ways of accounting for the context shift and modalization effects of “X” ist gut. Alternative 1 would be a division of labor account, assuming that the quotation marks indicate context-shift, whereas the predicate ist gut conveys the modalization, see therefore Table 1. Table 1: Division of labor account

“Theorie” ist gut

Quotation marks

Predication

Context-shift

Modalization

However, I would like to argue in favor of alternative 2, a reinforcement account, assuming that the quotation marks signal both context-shift and modalization, and that the predicate ist gut reinforces the modalization which is indicated by

Pragmatics and Metapragmatics of X ist gut   

   161

the quotation marks. That is, I would like to allow for an overlap of different functions for the quotation marks, cf. table 1 (for a refined version of the reinforcement account, see Section 4.4.2). This is in line with other meaning maximalist accounts, e.g., work on discourse markers (e.g., Schiffrin 1987, Aijmer 2002, cf. also Fischer 2006 for an overview). Table 2: Reinforcement account

“Theorie” ist gut

Quotation marks

Predication

Context-shift and modalization

Modalization

Example (26) provides evidence for the assumption that quotation marks can fulfill several functions. In order to explain the interpretation of (B2), which, at least in one usage⁶, is functionally equivalent to (B1), we need to assume that both context-shift and modalization are indicated by the quotation marks, as there is no predicate or other additional lexical material in (B2). (26)

A: Peters Theorie ist schwer zu verstehen. ‘Peter’s theory is difficult to understand.’ B1: “Theorie” ist gut. ‘“Theory” is good.’ B2: “Theorie”? ‘“Theory”?’

In order to account for the assumption that the quotation marks in “X” ist gut both indicate context-shift and modalization, we need to extend the first stage of Gutzmann/Stei’s two-staged inferential model so that the quotation marks trigger both an inference to context-shift and to modalization. Applied to example (27), we would get something like (28).

6 The utterance (B2) has at least the following two usages: (a) a usage as an “astonished question” (cf. Selting 1996); (b) a usage as a clarifying question (cf. Reis 2013; cf. also Bamford 2000). The interpretation (a) can be said to be roughly functionally equivalent to (B1).

162   

(27)

   Rita Finkbeiner

[Context: Manfred comments on a photograph posted by Aleks] Manfred: Gut getroffen… das [Foto] gefällt mir! like* LG, Manfred ‘Well captured… I like that [picture]! like* Best, Manfred’ Aleks:

(28)

Hallo Manfred, “getroffen” ist gut :-) Es ist mal wieder während der Fahrt aus dem Auto entstanden. ‘Hi Manfred, “captured” is good :-) It came again from the car while driving’, (cf. URL 17)

“Getroffen“ ist gut +> [context shift]:

W wants to convey that the word “getroffen” has been used in some original (previous) context with a certain denotation.

+> [modalization]:

W wants to convey that the word “getroffen” has been used in a non-stereotypical or marked way in the original context.

Both context-shift and modalization block the stereotypical interpretation of X. While the latter can be understood as marking some kind of deviance from the stereotypical lexical meaning (or from a stereotypical manner of speaking), the former can be understood as marking a “non-stereotypical context” in the sense that there is material from other writers/contexts within one writer’s utterance. As (29) and (30) show, these effects cannot be cancelled “without giving rise to the suspicion that CP [the cooperative principle, R.F.] is not met any more” (Gutzmann/Stei 2011: 2655). (29)

“Getroffen” ist gut. ?? “Getroffen” ist gut, but I don’t want to say that the word “getroffen” has been used in an original (previous) context with a certain denotation.

(30)

“Getroffen” ist gut. ?? “Getroffen” ist gut, but I don’t want to say that the word “getroffen” has been used in a non-stereotypical or marked way in the original context.

As a second stage in the inferential process, the actual interpretation of “X” ist gut in its specific context can be derived. This stage is represented by a context-

Pragmatics and Metapragmatics of X ist gut   

   163

dependent implicature which may be cancelled, giving rise to an alternative implicature. In the context of (27), a reader can derive the implicature which was introduced above as paraphrase 1. More specifically, in this context, the reader can infer that the writer wants to convey that “getroffen” is not situationally apt because the photographer has not been able to target the object carefully as he was sitting in a moving car. Here, targeting something carefully is regarded as a precondition of capturing something well with a camera. By contrast, in the context of (31), we get an implicature of the special stylistic value of the expression “getroffen”, as described above in paraphrase 2. (31)

[Context: M. and D. comment on a series of photographs taken at a party] Melanie84:

Nee wat haben se uns mal wieder gut getroffen… ;-) ‘They captured [‘struck’] us very well again… ;-)’

Dine:

“Getroffen” ist gut, ich seh irgendwie aus wie vom Blitz getroffen!!! So kalkwandmäßig!!! ☺ ‘“Captured” [‘struck’] is good, I somehow look like [I was] struck by lightning!!! Like a whitewashed wall!!! ☺’ (cf. URL 18)

More specifically, in the context of (31), the reader can infer that W wants to convey that “getroffen” is an extraordinarily apt expression, as it does not only fit the photography-situation (treffen ‘capture someone nicely with a camera’), but it also evokes the collocation of vom Blitz getroffen (‘struck by lightning’) which nicely meets the association of looking pale.

4.4 X ist gut Above, I proposed an analysis of “X” ist gut, sketching a two-staged inferential model in which the quotation marks were analyzed as pragmatic indicators that signal both context-shift and modalization. Now, how about cases of X ist gut where X is not marked by quotation marks? In what follows, I will argue that there is another type of pragmatic indicator associated with X ist gut which can block the stereotypical interpretation and which therefore may render quotation marks superfluous. This is in line with Gutzmann/Stei’s (2011, 2655) assumption that quotation marks are superfluous if there are already enough contextual hints [my emphasis, R.F.] that indicate that the literal meaning cannot be the intended meaning, given the writer is cooperative.

164   

   Rita Finkbeiner

What seems to be crucial for the omission of quotation marks to be felicitous, thus, is the context of the mentioned expression. For German, Klockow (1980: 65) lists a range of contexts which can render quotation marks highly redundant. Among others, there are “appositionsartige fügungen” (‘apposition-like constructions’) such as das Wort X (‘the word X’), der Begriff X (‘the notion X’), metalinguistic verbs such as bezeichnen als (‘call’), and “verbale modalisatoren” such as sogenannt, sozusagen (‘so-called’, ‘so to speak’) (cf. Klockow 1980: 127, original without capitalization). Whereas contexts such as das Wort X or verbs such as bezeichnen als typically mark the X-expression as mentioned (pure quotation), in the context of “verbal modalisators” such as das sogenannte X, X is typically marked as bearing some additional connotation (scare quotation). This is not to say, however, that in contexts such as das sogenannte X, no further implicatures may be indicated. For example, in many contexts, sogenannt also indicates context-shift. The different linguistic means of marking the need for further implicatures in (32) are functionally more or less interchangeable. Among them, we find both quotation marks, sogenannt (‘so-called’), parentheticals such as wenn man sie so nennen darf (‘X, if one may call it an X’), or certain word formation means such as compounding with Pseudo (‘pseudo’), Fake (‘fake’) or Möchtegern (‘wannabe’). (32)

a.

Peters “Theorie“ ist schwer zu verstehen.

b. Peters sogenannte Theorie ist schwer zu verstehen. c.

Peters Theorie, wenn man sie so nennen darf, ist schwer zu verstehen.

d. Peters Pseudo-/Fake-/Möchtegern-Theorie ist schwer zu verstehen. All of these utterances implicate that the proposal denoted by Theorie is not a theory in the proper sense of the word. In (32b-d), we could have additional quotation marks enclosing Theorie (cf. Klockow 1980: 126–127); however, there would not be a difference in interpretation. Rather, a sequence like Peters sogenannte “Theorie” in ordinary language would be perceived as a writer’s strategy to “play it safe” by double marking. In the following, I will argue that the formulaic context in which X is embedded is similar to the linguistic means illustrated in (32), indicating that X is to be interpreted in a non-stereotypical way. I will identify two ways in which X is marked as a mentioned expression in the local context of the formula. First, X occurs within a highly fixated, prefabricated predication. Second, X is restricted morphologically in specific ways which mark the X-expression as mentioned.

Pragmatics and Metapragmatics of X ist gut   

   165

4.4.1 X ist gut as a Formula X ist gut can be regarded as a formula in the sense of Wray (2002), who defines a formulaic sequence as a sequence that is, or appears to be, prefabricated: that is, stored and retrieved whole from memory at the time of use, rather than being subject to generation or analysis by the language grammar. (Wray 2002: 9)

In traditional phraseology research, the property of being “prefabricated” has been associated both with lexical-syntactic frozenness and the semantic idiomaticity of a sequence (e.g., Nunberg/Sag/Wasow 1994). Moreover, the role of situational-pragmatic boundedness of multiword units has been emphasized (e.g., Coulmas 1981, Feilke 2004, Kiefer 1996). As we will see, X ist gut shows characteristics of frozenness both with respect to syntax, semantics, and pragmatics. First, as to syntax, we can think of X ist gut as a “formal idiom” in the sense of Fillmore/Kay/O’Connor (1988). The constituents of the sentence which are lexically specified are syntactically highly fixed. For example, the copula is restricted to the singular, even if the X-expression is plural, cf. (12) above (Eltern ist gut, ‘Parents is good‘). Cf. also (33). (33)

[Context: About expensive watches] Zitat finch:

ganz geile shots, traurig das sich solche Bonzen wie ihr 2 nicht beide leisten könnt! ‘[These are] really cool watches, [it’s just] sad that fat cats like you guys cannot afford both of them!’

Cosmo1:

Bonzen ist gut, das sagt der richtige ‘Fat cats is good, look who’s talking’ (cf. URL 19)

The quotational readings in (12) and (33) vanish if we change the singular into a plural copula verb (this is indicated by #), cf. (34)–(35). (34)

#

Eltern sind gut. ‘Parents are good.’

(35)

#

Bonzen sind gut. ‘Fat cats are good.’

166   

   Rita Finkbeiner

Furthermore, the copula is fixed to the present. Whereas (36) can have both a quotational reading and a non-quotational reading, (37) can only have a nonquotational reading.⁷ (36)

Vertrauen ist gut. ‘Trust is good.’

(37)

#

Vertrauen war gut. ‘Trust was good.’

Semantically, one can say that the formula has a literal and a non-literal reading and thus behaves like an idiom. Both readings of an idiom may be potentially activated, depending on the context (cf. German blinder Passagier ‘blind passenger’: ‘stow-away’: (i) passenger who is not authorized to travel; (ii) passenger who cannot see). Similarly, a sentence like Linux-Professor ist gut (or: “Linux-Professor” ist gut) could be said to have the two readings (i) ‘the choice of the expression Linux-Professor in this particular situation is good’ and (ii) ‘the choice of the expression Linux-Professor in this particular situation is bad’. Reading (ii) can be regarded as ironical reading if we apply the classical definition of irony that what is implicated is the opposite of what is said (e.g., Grice 1975). While, in principle, both readings are available for utterances of X ist gut, the two readings are not conventionalized to the same extent. Rather, the ironical reading seems to be the standard interpretation because of its recurrent use. Even if my corpus is not a representative collection, it is striking that only 4 of 54 examples are non-ironical uses of X ist gut. Therefore, it seems reasonable to perceive X ist gut in terms of conventional irony (Dietz 1999). This does not imply, however, that we must abandon the assumption that the interpretation of the formula is calculated via a conversational implicature. As research on irony comprehension has shown, in ironical utterances, both the ironical and the literal reading are processed (e.g., Giora/Fein 1999). Thus, the literal reading does not seem to be completely abandoned.⁸ 7 One exception might be free indirect speech contexts, as one finds it in examples like the following: Schließlich seufzte sie und meinte: ‘Komm fürs Erste mit mir, okay? Vertrau mir.’ […] Langsam folgte er ihr […]. Vertrauen war gut. […] Er hatte niemals anderen vertraut. Dazu war er erzogen worden. (http://animefuture.net/index.php?page=Thread&postID=3081&highlight=vert rauen#post3081, 19 Dec 2012). (‘Finally, she sighed and said: ‘Just come with me, okay? Trust me.’ […] He followed her slowly […]. Trust was good. […] He never had trusted anyone. He had been brought up to be like that.’) 8 According to Morgan (1978: 275), who investigated conventionalized indirect speech acts, “an expression can be conventionalized and at the same time keep its literal meaning”. That is, even

Pragmatics and Metapragmatics of X ist gut   

   167

Finally, X ist gut is also a formula in the sense that it is pragmatically highly restricted. This aspect has been described as situational boundedness of phraseological units (cf. Kiefer 1996, Finkbeiner 2008). The formula can be used felicitously only in certain situational settings. In terms of contextualization cues (cf. Gumperz 1996), one could say that the formula contextualizes or creates a certain context of use. For example, it signals a certain sequential (responsive) position in discourse; it evokes the expectation that some kind of elaboration or motivation of the comment which it conveys will follow in subsequent context; it signals that speaker and hearer are emotionally more close than distant; and it indicates a rather informal context.

4.4.2 The Formula as a Non-minimal Pragmatic Indicator How does the fact that X ist gut is a formulaic sequence motivate its treatment as a pragmatic indicator? Gutzmann/Stei (2011: 5) define “minimal pragmatic indicators” as expressions which “do not have a proper semantic meaning”, and which “indicate something regarding the utterance or the context of utterance […] which has to be worked out by means of further pragmatic inferences”. As argued above, one can say that X ist gut indicates a certain context of use in which it fulfills a particular meta-communicative function (which has to be worked out by further pragmatic inferences). Its lexical-syntactic frozenness makes the formula highly recognizable for communicants. As a syntagmatic formula, it is highly distinctive and particularly suitable as a ‘pars pro toto’ indicator of a certain scheme or background (cf. Feilke 2004: 52). Thus, the formula qualifies as a pragmatic indicator in the sense of Gutzmann/Stei. At the same time, however, the formula clearly does have a proper semantic meaning. In this respect, it is not a minimal pragmatic indicator. Though, against the background of the description given so far, we can say that the formula, in addition to its propositional semantics, signals something regarding the utterance, namely, that X is to be interpreted as a mentioned expression which the speaker critically comments on. Following this line, one can describe the formula as a non-minimal pragmatic indicator (see also Fraser’s (1996) notion of pragmatic marker).

if it has become “common knowledge” in a linguistic community that by uttering X ist gut, a speaker standardly wants to critically comment on the applicability of the X-expression in its previous use, “a speaker of the language who lacks this bit of common knowledge will [nonetheless] understand what is intended […] by the original route of conversational implicature” (Morgan 1978: 275).

168   

   Rita Finkbeiner

The analysis of the formula as a non-minimal pragmatic indicator, signaling that X is to be interpreted in a non-stereotypical way, now allows us to refine the reinforcement account sketched above. Consider, once again, (26), represented here as (38). (38)

A: Peters Theorie ist schwer zu verstehen. ‘Peter’s theory is difficult to understand.’ B: “Theorie”? ‘“Theory”?’

Whereas in (38B), it is reasonable to analyze the quotation marks as indicating both context shift and modalization, in (39B), it is the formulaic predicate which must be analyzed as indicating both types of information. (39)

A: Peters Theorie ist schwer zu verstehen. ‘Peter’s theory is difficult to understand.’ B: Theorie ist gut. ‘Theory is good.’

Neither for (38) nor for (39), we can assume a division of labor between quotation marks and formulaic predication, because there is only one kind of pragmatic marker present in each case. In (38) and (39), there is no reinforcement of the modalization effect either; this effect (as the context shift effect) is solely brought about by the one pragmatic marker which is present. This is represented in Table 3. Table 3: Distribution of context shift and modification effects in “X”? and X ist gut

Quotation marks

Predication

“Theorie”?

Context-shift and modalization

--

Theorie ist gut.

--

Context-shift and modalization

However, if in cases like (39), the formulaic predication can function as a pragmatic indicator signaling both context-shift and modalization, we need to assume, even for cases like (23) (“Theorie” ist gut), that both potential effects are triggered not only by the quotation marks, but also by the formula. That is, both types of pragmatic markers may, independently of each other, signal both types of non-stereotypical interpretation effects. If the pragmatic markers occur

Pragmatics and Metapragmatics of X ist gut   

   169

together, they will reinforce each other. Consequently, we need to modify the reinforcement account (cf. Table 2) in the following way (cf. Table 4). Table 4: Reinforcement account, refined

“Theorie” ist gut.

Quotation marks

Predication

Context-shift and modalization

Context-shift and modalization

I will discuss the question of whether this makes quotation marks redundant in Section 4.5 below.

4.4.3 Morphological Marking of X It is not only its formulaic context which contributes to a quotational reading of X, but also its morphological features. As a generalization, we can say that if there is any morphological change between the original and the repeated X, the direction is always from morphologically richer towards morphologically poorer forms, i.e., inflectional marking on X may be deleted, but not added. Recall the examples in Section 3. In (4) and (8), represented here as (40) and (41), it is the grammatical inflection (agreement features) which is deleted, i.e. agreement within an NP in (40) and subject-verb-agreement in (41). (40)

… mit der kleinen Maus. Wobei “klein” ist gut, … ‘… with the little.Dat.Sg.Fem sweetie pie.Dat.Sg.Fem. Well, “little.Uninfl” is good …’

(41)

Zitat SAZ12:

ich weiß nicht ob ich hier noch etwas ändern muss ‘I don’t know whether I.1.Sg.Pres have to1.Sg.Pres change anything else here’

Bewerbungshelfer: “Noch etwas ändern müssen” ist gut… mit dem, was den Leser interessiert, hast Du ja noch gar nicht angefangen ‘“have toInfin change anything else” is good… you haven’t even started with what the reader is really interested in’ The deletion of agreement features can be seen as a reflection of the fact that X is a mentioned expression. Also in apposition-like constructions with a mentioned X-expression, such as der Ausdruck X (‘the expression X’) or die Bezeichnung

170   

   Rita Finkbeiner

X (‘the term X’) (Klockow 1980: 127), X regularly appears in uninflected form, cf. (42).⁹ (42)

a.

Der Ausdruck klein ist gut. ‘The expression little.Uninfl is good.’

b.

?

Der Ausdruck kleinen ist gut. ‘The expression little.Dat.Sg.Masc is good.’

However, not all kinds of inflection are deleted in the construction. In contrast to contextual inflection (agreement), inherent inflection (e.g., plural of nouns) is preserved, cf. (43).¹⁰ (43)

A: … solche Bonzen wie ihr … ‘… fat cats like you …’ B1: Bonzen ist gut ‘Fat cats is good’ B2:

??

Bonze ist gut ‘Fat cat is good’

This can be explained by a functional argument: If the function of the formula is to question whether a certain entity is a proper member of the extension of the X-expression, it follows that it is the referential potential of X that matters. Therefore, inflectional features on X that have a certain effect on its referential semantics (i.e., plural) need to be preserved. Regarding the phrasal level, I have shown that original determiner phrases such as die Eltern, der strenge Onkel, ein kleines Problem are typically reduced to bare NPs (Eltern, strenger Onkel, kleines Problem). This observation points to the fact that it is not primarily syntactic factors – i.e., the property of being part of a syntactic phrase – that license the appearance of a certain category on the X-slot, but rather lexical factors – i.e., the property of carrying lexical content. This would explain the deletion of articles on the X-slot. That is not to say, however, that it is strictly impossible to have determiner phrases as X (“Der strenge Onkel” 9 One might speculate that readers in the process of utterance interpretation replace ‘X’ in ‘X ist gut’ by ‘der Ausdruck X’. This would explain the number restriction on the copula, as the head N (Ausdruck) of the subject NP der Ausdruck X is always singular, no matter what number the appositional N (X) has (cf. Bonzen ist gut < ‘Der Ausdruck Bonzen ist gut’). 10 Cf., e.g., Booij (2005) on the distinction between contextual and inherent inflection in morphology.

Pragmatics and Metapragmatics of X ist gut   

   171

ist gut); neither would I strictly exclude the possibility of grammatical inflection on the X-slot, as in “Kleinen” ist gut. However, as the corpus examples indicate, speakers/writers seem to prefer to only copy categories which carry lexical information, as it is the negotiation of the meaning of X which is the primary function of the construction.

4.5 Role of Quotation Marks Above I have argued that in the case of “X” ist gut, both the quotation marks and the formulaic predication can be regarded as pragmatic markers blocking the stereotypical interpretation of X. This raises the question of whether the quotation marks in “X” ist gut are redundant. Generally, it has been pointed out that everyday writing practice “gestattet sich […] große freiheiten in [der] anwendung [von Anführungszeichen]” (‘behaves freely in the application of quotation marks’) (Klockow 1980: 69). However, there are at least two types of contexts in the corpus in which quotation marks are functional, namely, a) disambiguation and b) focus marking. In the other cases, it seems to me that the use or non-use of quotation marks is a question of individual preference, at least with respect to informal written language. As to the disambiguation function, it has been pointed out above that abandoning the quotation marks may render X ist gut ambiguous, cf. Vertrauen ist gut (‘Trust is good’). In cases like this, quotation marks can disambiguate the meaning of the sentence for a metapragmatic reading, cf. “Vertrauen” ist gut (‘“Trust” is good’). Note, however, that in most actual situations of use, co-textual and contextual cues will help readers/hearers to unambiguously identify the intended reading of the formula. For example, using the adjective schlecht (‘bad’) on X-position as in (44) blocks a non-quotational reading because of the otherwise occurring lexical mismatch between the antonyms gut (‘good’) and schlecht (‘bad’). (44)

Re: Schlechter Cewe Support? [= subject heading] Re: Bad Cewe Support?’ [= subject heading] Kkm3105: schlecht ist gut: die haben überhaupt keinen Support ‘bad is good: they don’t have any support’ (cf. URL 20)

Also, inflectional forms on the X-slot which are not compatible with the agreement features of the copula (cf. Eltern.Pl ist.Sg gut) or which violate determiner restrictions (cf. Idiot ist gut, ‘Idiot is good’) will suppress ambiguity. As bare sin-

172   

   Rita Finkbeiner

gular count nouns (Idiot) in German cannot usually refer to specific entities in the world, the only possible reading for an utterance like Idiot ist gut is a metapragmatic reading. Yet another disambiguating means is emoticons. For example, in the case of (18) or (19) above, the laughing emoticons after the construction clearly indicate a positively evaluating reading. As to the potential function of quotation marks as focus marking devices, consider, once again, (14), represented here as (45). (45)

Lightstorm: das wäre schon etwas lahm ‘That would be quite slow indeed’ Thawra:

‘Etwas’ lahm ist gut ‘‘Quite’ slow is good’

In (45), X is represented by an AP consisting of a modifier and an adjective (etwas lahm). Interestingly, however, the quotation marks only enclose the modifier (‘etwas’). Thus, it seems that in (45), we actually do have a genuine (but exceptional) instance of division of labor. The formulaic predication marks the whole pre-field phrase etwas lahm as mentioned (context-shift), while the quotation marks signal that only (or: in particular) the modifier etwas has been used in a non-standard way in the previous context (modalization). This means that the quotation marks in (45) are not superfluous. Rather, they fulfill the function of focus markers, while the formula fulfills the function of a scope marker. This may be represented as in Table 5. Table 5: Division of labor: Quotation marks as focus marking devices

“Etwas” lahm ist gut.

Quotation marks

Predication

Modalization (focus)

Context-shift (scope)

Evidence for this analysis can be found in differences concerning potential continuations of the sentences. Consider (46) and (47). (46)

a.

“Etwas” lahm ist gut, das war exTREM langsam. ‘“Quite” slow is good, that was exTREMEly slow.’

b.

??

“Etwas” lahm ist gut, das war schon relativ FLOTT. ‘“Quite” slow is good, that was actually rather QUICK.’

Pragmatics and Metapragmatics of X ist gut   

(47)

a.

Etwas “lahm” ist gut, das war schon relativ FLOTT. ‘Quite “slow” is good, that was actually rather QUICK.’

b.

??

   173

Etwas “lahm” ist gut, das war exTREM langsam. ‘Quite “slow” is good, that was exTREMEly slow.’

Whereas quote-marked “etwas” projects a contrastive focus on the parallel modifier in the continuation (extrem), quote-marked “lahm” projects a contrastive focus on the parallel adjective (flott). As one can see, the quotation marks correlate with prosodic stress on the focused elements. As the respective b.-utterances in (46) and (47) violate these expectations, they are not felicitous.

5 Summary The overall aim of this paper was to give a pragmatic account of the German X ist gut (‘X is good’) construction. A central claim was that X is a mentioned expression, regardless of whether or not X is marked by quotation marks. Thus, I claimed that (in the majority of cases) there is no interpretational difference between “X” ist gut and X ist gut. More specifically, I argued that there are two potential pragmatic markers active in the construction: optional quotation marks and the formulaic predication (ist gut). Whereas the quotation marks were analyzed as minimal pragmatic indicators, the formula was analyzed as a nonminimal pragmatic indicator. Both types of pragmatic indicators may block the stereotypical interpretation of X, rendering a context-shift interpretation and a modalization effect. Analyzing the formulaic predication as a pragmatic indicator explains why the quotation marks may be abandoned. Regarding the question of how the context-shift and modalization information carried by the construction is brought about, I argued against a division of labor and in favor of a reinforcement account, in which both types of pragmatic markers may indicate both effects, reinforcing each other. However, I also showed that division of labor can occur in specific cases, where quotation marks are assigned the role of focus indicators and the formula the role of a scope indicator.

174   

   Rita Finkbeiner

References Aijmer, Karin. 2002. English Discourse Particles. Evidence from a Corpus. Amsterdam: Benjamins. Bamford, Julia. 2000. You Can Say That Again. Repetition in Discourse. Bologna: CLUEB. Brendel, Elke, Jörg Meibauer and Markus Steinbach. 2011. “Exploring the meaning of quotation”, in: Elke Brendel, Jörg Meibauer and Markus Steinbach (eds). Understanding Quotation. Berlin: de Gruyter, 1–33. Bublitz, Wolfram. 2001. “Formen der Verständnissicherung in Gesprächen”, in: Klaus Brinker and Herbert Ernst Wiegand (eds). Text- und Gesprächslinguistik. Ein internationales Handbuch zeitgenössischer Forschung. 2. Halbband. Berlin: de Gruyter, 1330–1340. Carston, Robyn. 2002. Thoughts and Utterances. The Pragmatics of Explicit Meaning. Oxford: Blackwell. Coulmas, Florian. 1981. Routine im Gespräch. Zur pragmatischen Fundierung der Idiomatik. Wiesbaden: Athenaion. Dietz, Hans-Ulrich. 1999. Rhetorik in der Phraseologie. Zur Bedeutung rhetorischer Stilelemente im idiomatischen Wortschatz des Deutschen. Tübingen: Niemeyer. Dürscheid, Christa. 2003. “Medienkommunikation im Kontinuum von Mündlichkeit und Schriftlichkeit. Theoretische und empirische Probleme”. Zeitschrift für Angewandte Linguistik 38, 37–56. Fauconnier, Gilles and Mark Turner. 2003. “Polysemy and Conceptual Blending”, in: Brigitte Nerlich et al. (eds). Polysemy: Flexible Patterns of Meaning in Mind and Language. Berlin: de Gruyter, 79–94. Feilke, Helmuth. 2004. “Kontext – Zeichen – Kompetenz. Wortverbindungen unter sprachtheoretischem Aspekt”“, in: Kathrin Steyer (ed.). Wortverbindungen – mehr oder weniger fest. Berlin: de Gruyter, 41–64. Fiehler, Reinhard. 2009. “Gesprochene Sprache”, in: Dudenredaktion (ed.). Duden. Die Grammatik. Vol. 8. überarbeitete Auflage. Mannheim: Dudenverlag, 1165–1246. Fillmore, Charles, Paul Kay, and Mary C. O’Connor. 1988. “Regularity and idiomaticity in grammatical constructions: the case of let alone”. Language 64/3, 501–538. Finkbeiner, Rita. 2008. Idiomatische Sätze im Deutschen. Syntaktische, semantische und pragmatische Studien und Untersuchung ihrer Produktivität. Stockholm: Acta Universitatis Stockholmiensis. Finkbeiner, Rita. 2012. “Naja, normal und normal. Zur Syntax, Semantik und Pragmatik der x-und-x-Konstruktion im Deutschen“. Zeitschrift für Sprachwissenschaft 31, 1–42. Fischer, Kerstin (ed.). 2006. Approaches to Discourse Particles. Boston: Brill. Fraser, Bruce. 1996. “Pragmatic markers”. Pragmatics 6/2, 167–190. Giora, Rachel and Ofer Fein. 1999. “Irony comprehension: the graded salience hypothesis”. Humor 12/4, 425–436. Grice, Herbert P. 1975. “Logic and Conversation”, in: Peter Cole and Jerry Morgan (eds). Syntax and Semantics 3: Speech Acts. New York: Academic Press, 41–58. Gülich, Elisabeth and Thomas Kotschi. 1987. “Reformulierungshandlungen als Mittel der Textkonstitution. Untersuchungen zu französischen Texten aus mündlicher Kommunikation”, in: Wolfgang Motsch (ed.). Satz, Text, sprachliche Handlung. Berlin: Akademie, 199–268.

Pragmatics and Metapragmatics of X ist gut   

   175

Gumperz, John J. 1996. “The linguistic and cultural relativity of inference”, in: John J. Gumperz and Stephen Levinson (eds). Rethinking Linguistic Relativity, Cambridge: Cambridge University Press, 374–406. Gutzmann, Daniel and Erik Stei. 2011. “How quotation marks what people do with words”. Journal of Pragmatics 43/10, 2650–2663. Hagemann, Jörg. 1997. Reflexiver Sprachgebrauch. Diktumscharakterisierung aus Gricescher Sicht. Opladen: Westdeutscher Verlag. Hindelang, Götz. 1975. “Äußerungskommentierende Gesprächsformeln. Offen gesagt, ein erster Schritt”, in: Veronika Ehrich and Peter Finke (eds). Beiträge zur Grammatik und Pragmatik, Kronberg: Scriptor, 253–263. Hrncal, Christine. 2011. Was heißt schon x? – Relativierungsverfahren im Sprachgebrauch des Deutschen. Studentische Arbeitspapiere zu Sprache und Interaktion 20. University of Münster. http://noam.uni-muenster.de/sasi/Hrncal_SASI.pdf, last accessed 21 Dec 2012. Kiefer, Ferenc. 1996. “Bound Utterances”. Language Sciences 18/1–2, 575–587. Klockow, Reinhard. 1980. Linguistik der Gänsefüßchen. Untersuchungen zum Gebrauch der Anführungszeichen im gegenwärtigen Deutsch. Frankfurt a, Main: Haag und Herchen. Levinson, Stephen. 2000. Presumptive Meanings. Cambridge: MIT Press. Linell, Per. 1998. Approaching Dialogue. Talk, Interaction and Contexts in Dialogical Perspectives. Amsterdam: John Benjamins. Morgan, Jerry L. 1978. “Two types of convention in indirect speech acts”, in: Peter Cole (ed.). Syntax and Semantics 9: Pragmatics. New York: Academic Press, 261–280. Norén, Kerstin and Per Linell. 2007. “Meaning potentials and the interaction between lexis and contexts: an empirical substantiation”. Pragmatics 17/3, 387–416. Nunberg, Geoffrey, Ivan A. Sag and Tom Wasow. 1994. “Idioms”. Language 70, 491–538. Predelli, Stefano. 2003. “Scare quotes and their relation to other semantic issues”. Linguistics and Philosophy 26, 1–28. Recanati, François. 2000. Oratio Obliqua, Oratio Recta: An Essay on Metarepresentation. Cambridge, Mass.: MIT Press. Recanati, François. 2004. Literal Meaning. Cambridge: Cambridge University Press. Reis, Marga. 2013. “Assertive Frage und Echofrage”, in: Hans Altmann, Jörg Meibauer and Markus Steinbach (eds). Satztypen des Deutschen. Berlin: de Gruyter, 105–119. Schiffrin, Deborah. 1987. Discourse markers. Cambridge: Cambridge University Press. Selting, Margret. 1996. “Prosody as an activity-type distinctive signalling cue in conversation. The case of so-called ‘astonished questions’ in repair-initiation”, in: Elisabeth CouperKuhlen and Margret Selting (eds). Prosody in Conversation. Interactional Studies. Cambridge: Cambridge University Press, 231 – 270. Techtmeier, Bärbel. 2001. “Form und Funktion von Metakommunikation im Gespräch”, in: Klaus Brinker and Herbert Ernst Wiegand (eds). Text- und Gesprächslinguistik. Ein internationales Handbuch zeitgenössischer Forschung. 2. Halbband. Berlin: de Gruyter, 1449–1463. Wray, Alison. 2002. Formulaic Language and the Lexicon. Cambridge: Cambridge University Press.

176   

   Rita Finkbeiner

Online References URL 1:http://www.haustechnikdialog.de/Forum/t/110603/Holzhackertreffen-Januar2010?page=12, (last accessed 5 Apr 2012) URL 2:www.nupo.de/kann-linux-besser-als-windows-17275-2.html, (last accessed 3 Feb 2012) URL 3:http://coderz-home.de/thread350/index.html, (last accessed 3 Feb 2012) URL 4:http://www.wunschkinder.net/forum/read.php?2,6644407, (last accessed 3 Feb 2012) URL 5:http://www.gutefrage.net/frage/verknallt-was-gibt-es-fuer-anzeichen, (last accessed 3 Feb 2012) URL 6:http://www.wohnmobilforum.de/w-t66588,start,30.html, (last accessed 6 Feb 2012) URL 7:http://board.myfreefarm.de/viewtopic.php?f=165&t=54162, (last accessed 6 Feb 2012) URL 8:http://www.bewerbung-forum.de/forum/bewerbung-eines-zeitsoldaten-t39840.html, (last accessed 5 Apr 2012) URL 9: http://www.japanisch-netzwerk.de/Thread-Japan-Ein-Jahr-Pers%C3%B6nlichesLernprogramm-erstellen?page=2, (last accessed 6 Feb 2012) URL 10:http://www.roboternetz.de/community/threads/19520-Webcam-Auslesen-mit-Delphi, (last accessed 6 Feb 2012) URL 11:http://diaet.abnehmen-forum.com/ich-bin-genervt/76519-von-ruecksichtsloseneltern-6.html, (last accessed 3 Feb 2012) URL 12:http://www.lauftipps.de/foren/viewtopic.php?p=59593&sid=a633ee20940f072c01743 d7d23ce30ed, (last accessed 3 Feb 2012) URL 13:http://www.natur-portrait.de/foto-77753-drei-auf-einen-streich.html, (last accessed 3 Feb 2012) URL 14:http://www.allmystery.de/themen/cp69914-31, (last accessed 3 Feb 2012) URL 15:http://www.hififorum.at/forum/showthread.php?p=126755, (last accessed 6 Feb 2012) URL 16: http://bfriends.brigitte.de/foren/tipps-fuer-ein-gluecklicheres-leben/81154-der-dreischoene-dinge-heute-strang-trifft-sich-258.html, (last accessed 6 Feb 2012) URL 17: http://www.panoramio.com/photo/47336554, (last accessed 6 Feb 2012) URL 18:http://annos.de/nt/Galleries/Gallery.aspx?do=img&gal=3282&op=349822, (last accessed 6 Feb 2012) URL 19:http://uhrforum.de/triton-black-bicolor-t60758, (last accessed 3 Feb 2012) URL 20:http://www.fotocommunity.de/forum/read.php?f=11&i=296172&t=296172, (last accessed 6 Feb 2012)

Jörg Meibauer

7 Only “nur”. Scare Quoted (Exclusive) Focus Particles at the Semantics/ Pragmatics Interface Abstract: This article discusses utterances in which the German focus particle nur (Engl. ‘only’) is used with quotations marks, e.g. in the newspaper headline Spears zahlt Mann “nur” 750.000 Euro (‘Spears pays husband “only” 750,000 euro.’) Part of the total signification of this utterance is that 750,000 euro is not a lot of money with respect to Spear’s income, although it is a lot with respect to the income of the average person. While it is obvious that this meaning effect comes about through the use of quotation marks (‘scare quotes’), it is not clear whether these contribute meaning in themselves, or whether they trigger pragmatic inferences such that the literal meaning (the sentence without the quotation marks) is enriched. After a discussion on earlier approaches to scare quotes by Predelli (2003), Cappelen/Lepore (2007), and Gutzmann/Stei (2011a, b), this article proposes an analysis using Levinson’s (2000) framework. Scare quotes trigger a conversational implicature based on the I-principle and often go with an ironical interpretation. Keywords: contextualism, conversational implicature, focus particle, irony, scare quote, semantics/pragmatics interface

1 Introduction There are numerous utterances in written language in which the German focus particle nur (Engl. ‘only’) is used with quotations marks: e.g. in the following newspaper headline (Wiesbadener Kurier, 31 Mar 2007) in the context of Britney Spears’ divorce from her husband Kevin Federline: (1)

Spears zahlt Mann „nur“ 750.000 Euro. ‘Spears pays husband “only” 750,000 euro.’

Part of the total signification of this utterance is that 750,000 euro is not a lot of money with respect to Spear’s income, while it is a lot with respect to the income of the average person. In the parlance of Predelli (2003), the quotation marks give rise to a certain “attachment” of the utterance, i.e. the specific meaning that is conveyed through the use of the quotation marks, as opposed to its “message”.

178   

   Jörg Meibauer

Such examples are by no means rare. A corpus search in DWDS (Digitales Wörterbuch der deutschen Sprache, www.dwds.de), based on the newspaper Berliner Zeitung, done on 1 August 2008, shows that there are quite a number of examples with “nur”; other scare quoted focus particles, such as “lediglich” in example (2), show up occasionally as well. (2)

Nach neuesten Berechnungen des BBF kostet der Neubau in Sperenberg 11,609 Milliarden Mark. In Schönefeld hingegen schlägt der Komplett-Ausbau „lediglich“ mit 10,914 Milliarden Mark zu Buche. [Berliner Zeitung, 17 Nov 1995] ‘According to latest calculations of the BBF, the new building in Sperenberg costs 11,609 billion marks. In Schönefeld, however, the completion adds up to “merely” 10,914 billion marks in the books.’

That (1) and (2) are related to numerical expressions is no mere accident: a quick glance at the results of the DWDS search shows that this is typical for most cases of the scare quoted “nur”. Typically, the focus constituents are noun phrases. But there are other maximal projections as well. In (3), the focus constituent is an adjectival phrase, in (4) a verbal phrase, and in (5) a prepositional phrase: (3)

Mit diesem Andrang habe niemand gerechnet, denn der Wettbewerb war „nur“ bundesweit ausgeschrieben worden. [Berliner Zeitung, 12 Aug 1994, B90] ‘Nobody expected such a rush, since the competition had “only” been put out to tender nationwide.’

(4)

Gleichzeitig aber inspirierten ihre Musik und ihre Dichtung berühmte Künstler, wie etwa Franz Liszt. Bauer bleibt ganz in der Haltung des distanzierten Wissenschaftlers, der „nur“ referiert, aber unter der Hand die Vorurteile der Täter abbildet. Er macht sich noch nicht einmal die Mühe, sie zu widerlegen. [Berliner Zeitung, 3 Feb 1994, B18] ‘At the same time, however, their music and poetry inspired famous artists, such as Franz Liszt for instance. Bauer remains completely in the position of a distanced scholar who “only” reports, but also represents the prejudices of the offenders underhand. He can’t even be bothered to refute them.’

Scare Quoted Focus Particles   

(5)

   179

Im Vorfeld hatte die FIS vollmundig Sanktionen angedroht, falls die 200m übertroffen werden sollten, denn der kritische Punkt der Schanze liegt „nur“ bei 185 Metern. (Berliner Zeitung, 19 Mar 1994, B38) ‘Previously, the FIS had pompously threatened to impose sanctions if the 200m was surpassed, since the critical point of the ski jump is “only” at 185 meters.’

The gist of this paper is to give a sketch of the use of quotation marks with respect to the meaning of the exclusive focus particle nur (Engl. only). The outline is as follows: In Section 2, I will analyze the meaning of the focus particle nur, drawing largely on the approach of Altmann (1976, 2007) and the discourse-based approach by Beaver/Clark (2008). In Section 3, I will discuss the general question of how quotation marks contribute to scare quoting, applying approaches of Predelli (2003), Gutzmann/Stei (2011a, b), and Cappelen/Lepore (2007). In Section 4, I will discuss how the phenomenon of scare quoted focus particles may be analyzed using Levinson’s (2000) presumptive meaning theory. In Section 5, I will point out the special properties of ironical scare quoting with respect to focus particles. Section 6 presents the conclusions.

2 On the Meaning of Nur and Only Since Horn’s (1969) seminal article, there has been much research into the semantics of focus particles, especially the semantics of only, even, and too.¹ Most researchers agree that there are two distinct meaning components with respect to a simple sentence like (6), cf. (7) and (8). (6)

Only Lucy came to the party.

(7)

[Meaning component 1] Lucy came to the party.

(8)

[Meaning component 2] No one other than Lucy came to the party.

1 A thorough review of the literature on only is beyond the scope of this paper, so the reader is referred to the papers of Atlas (1993, 1996), Beaver (2004), Beaver/Clark (2003), Bolinger (1972), Bonomi/Casalegno (1993), Geurts/van der Sandt (2004), Horn (1969, 1996, 2002), Ippolito (2007), König (1991), Roberts (2006), van Rooij/Schulz (2005), and the comprehensive discussion in Beaver/Clark (2008).

180   

   Jörg Meibauer

Research on only has largely been focused upon the semantic/pragmatic nature of these meaning components: for instance whether (7) has the status of a presupposition or an implicature. Following Horn’s (1996) lead, this meaning element is often called the prejacent. Roberts (2006), in a thorough review of the literature, cautiously calls the first meaning component the prejacent implication, and the second meaning component the exclusive implication. The second meaning component (alias exclusive implication) in (8) is often classified as an assertion (e.g., Altmann 1976, 2007). Since it is this meaning that enters into the truth conditions of (6), it may be considered as its truth-conditional content along the following lines (e.g. König 1991: 34): if it is true that only Lucy came to the party, it must be true that no one other than Lucy came to the party. In the following section, I summarize findings on the German focus particle nur, before sketching Beaver and Clark’s (2008) discourse-based approach to the focus particle only.

2.1 Findings on German Nur According to Altmann’s (1976, 2007) seminal work, focus particles (alias Gradpartikeln) may be classified along two dimensions, namely the inclusive/exclusive dimension and the scalar/quantifying dimension.² The first dimension refers to the meaning of a focus particle with respect to its property of either including the referent of its focus constituent or not. For instance, in Auch Ken kam zur Party (‘Ken came to the party, too.’), Ken is included in the set of party attenders, while in Nur Ken kam zur Party (‘Only Ken came to the party.’), the set of party attenders is restricted to Ken, i.e., all other party attenders are excluded. The second dimension has to do with typical readings of focus particles being salient in the context of an utterance. According to Altmann (1976, 2007), the focus particle nur is compatible with two readings, namely the quantifying reading and the scalar reading. The quantifying reading, for instance, may be seen in Ken besuchte nur Hamburg (‘Ken visited only Hamburg.’), where Hamburg is one city taken from a set of other cities, e.g. , which is not ranked according to the attractiveness of its members. The speaker only wants to convey that Ken did not manage to visit the other cities, without invoking an evaluation of these cities. In counter distinction, if Hamburg is ranked low on a subjective scale of attractive cities, e.g. , we

2 The focus of this article is on the semantics and pragmatics of nur. For matters of syntax, prosody, and information structure see Sudhoff (2010).

Scare Quoted Focus Particles   

   181

face the scalar reading, e.g. Ken besuchte nur Hamburg means that Ken did not manage to visit cities more attractive than Hamburg. According to the taxonomy proposed by Altmann (2007: 359f.), not all focus particles in the nur-group allow for both readings: nur, bloß, and lediglich do, but allein, ausschließlich, einzig, and einzig und allein allow for a quantifying reading only. It goes without saying that most sentences containing nur are ambiguous with respect to the quantifying/scalar dimension. When using scare quotes, it seems that an evaluation is imposed on both readings. Thus we might have: (9)

Ken besuchte „nur“ Hamburg (‘Ken visited ‘only’ Hamburg’). a. [Quantifying reading] … aber das ist ja besser als gar nichts besucht zu haben. (‘… but this is better than visiting no city at all’) b. [Scalar reading] … aber Hamburg ist ja auch eine ganz schöne Stadt (‘…but Hamburg is an attractive city after all’).

Besides the prejacent and the assertion, two additional meaning aspects of the German nur have been discussed. The first is the meaning component WENIG (‘little’), and the second is a generalized conversational implicature (GCI). Firstly, consider the meaning component of nur in association with scales. It is clear that the scales involved in scalar meaning are gained from the context of utterance. Moreover, the ordering relation between the elements on that scale needs to be established. On the basis of a specific ordering relation, utterances containing the scalar nur show an additional meaning aspect, such as for example  (1) without the quotation marks (Spears pays husband only 750, 000 euro): (10)

750,000 euro is not a lot of money.

In this case, the value of focus ranges low on the scale of income. It is not clear how that meaning aspect is captured. For Jacobs (1983: 161), this meaning component is part of the overall assertion and thus enters into the truth conditions. Note that Jacobs postulates an operator WENIG (‘little’) to model that meaning. Similarly, König (1991: 101) proposes an operator MinC, symbolizing that an item is ranked low on a contextually given scale. But for König, it is not exactly clear what the status of this meaning component is; thus, he occasionally speaks of “evaluative presuppositions” (1991: 44, 101). However, he is also aware of the problem of distinguishing between standard semantic presuppositions (being triggered by a certain lexical item, surviving the negation test, etc.) and these evaluative presuppositions. He concludes that these meaning elements are most

182   

   Jörg Meibauer

likely conventional implicatures (CI), claiming that they have an essentially deictic character (König 1991: 55–58). I am inclined to side with Jacobs on that matter. Thus, if a certain scale is active, as in example (1), a sentence like (11) does not make any sense. Only always marks a point on a scale that is ranked as low.³ (11)

# Spears pays husband only 750,000 euro and this is a lot of money.

In (11), the second clause may only be understood as a comment or denial of the first conjunct. Secondly, Altmann (1976, 2007) proposed that there is also a generalized conversational implicature (GCI) at play. He gives the following example (Altmann 2007: 374): (12)

Unsere Sekretärin bekommt nur 1.000 Euro brutto. ‘Our secretary only gets 1,000 euro before tax.’

(13)

Presupposition [= prejacent implication] of (12): Unsere Sekretärin bekommt 1.000 Euro brutto. ‘Our secretary gets 1,000 euro before tax.’

(14)

Assertion [= exclusive implication] of (12): Unsere Sekretärin bekommt nicht mehr als 1.000 Euro brutto. ‘Our secretary gets no more than 1,000 euro before tax.’

(15)

Generalized Conversational Implicature (GCI) of (12): Unsere Sekretärin bekommt nicht weniger als 1.000 Euro brutto.’ ‘Our secretary gets no less than 1,000 euro before tax.’

Why is (15), according to Altmann’s analysis, a GCI? The idea is that if the speaker had known that a lower value on the scale was correct, he would have said so. Assuming that he was observing the Gricean Maxim of Quantity, a lower value can be excluded. However, if (15) really is a GCI, it should be cancellable in principle. But this does not appear to be possible:⁴

3 The hash symbol indicates that an utterance is pragmatically unacceptable. 4 Note that Altmann (2007: 373) remarks that the GCI is not essentially bound to focus particles.

Scare Quoted Focus Particles   

(16)

   183

# Unsere Sekretärin bekommt nur 1.000 Euro brutto, und noch nicht einmal das. ‘Our secretary only gets 1,000 euro before tax, and not even that.’

Since (16) does not work as a cancellation, I doubt the status of (15) as a conversational implicature. For me, utterance (16) does not sound so much like a cancellation than it does a clarification. While a cancellation aims at suspending an implicature without any logical contradiction, a clarification makes the intended meaning more clear, and this might go together with revising or repairing the utterance under scrutiny (see Burton-Roberts 2006, 2010).⁵ Moreover, the assumption of such a generalized conversational implicature is not necessary, if the boundaries of scales are clearly marked. Thus, in Jacobs’ theory of scales, every scale has a lower and an upper boundary value. If a certain lower boundary value is given, e.g. 1.000 euro in (12), it makes no sense to assume a GCI of the type in (15). To sum up, I believe that the exclusive implication (Altmann: assertion) and the scalar assertion (Jacobs: WENIG operator) are together truth-conditionally relevant. Thus, the intuitive truth-conditional meaning of (17a) should be rendered as in (17b): (17)

a.

Spears pays husband only 750,000 euro.

b. Spears pays husband no more than 750,000 euro & 750,000 euro is not a lot of money. Although the second conjunct in (17b) seems to be part of the truth-conditional meaning of (17a), it is clear that it is likely to provoke its denial, since 750,000 euro is, for most of us, a lot of money. Here the need to make another perspective available comes in, since we have to ask for whom and under what circumstances is 750,000 euro not a lot of money (see § 2.3); e.g., the writer assumes that this is not a lot of money for Britney Spears, since s/he knows that she is wealthy.

5 Similarly, examples like Er besitzt nur fünf Mark, und möglicherweise sogar nur vier, that Altmann (1976: 81) finds acceptable, are not convincing to me. Note that möglicherweise leads the hearer to think of a possible situation in which the referent owns 4 mark. With regard to (16), Wolfram Bublitz provided the additional phrase eigentlich noch nicht einmal das – sie bekommt nur 900 (‘actually, not even that much – she only earns 900’) which is a clarification, since the assertion that she earns 1,000 mark is substituted by the assertion that she earns 900 mark.

184   

   Jörg Meibauer

2.2 A Discourse-Based Model of Exclusives Beaver and Clark (2008) developed a discourse-based account to the exclusive particle only, drawing on the earlier work of Craige Roberts. At the heart of their theory is a general approach to the meaning of exclusives that is rendered in (18) (original emphasis): (18)

Meaning of exclusives (Beaver/Clark 2008: 251) The lexical meaning of exclusives is exhaustively described by: Discourse function: to make a comment on the Current Question (CQ […]), a comment which weakens a salient or natural expectation. To achieve this function, the prejacent must be weaker than the expected answer of the CQ on a salient scale. Presupposition: the strongest true alternatives in the CQ are at least as strong as the prejacent. Descriptive Content: the strongest true alternatives in the CQ are at most as strong as the prejacent.

The main idea behind this approach is that utterances containing only are answers to a Current Question (CQ). According to the dynamic model favoured by Beaver/ Clark (2008: 249), the Current Question “associated with an utterance is the question which that utterance addresses, the most specific of the questions on the stack. If the utterance answers the question fully, and the answer is accepted, that question will then be popped off the stack.” For instance, Only Jane smokes is the answer to the Current Question Who smokes? in a given context. Current questions come with a salient or natural expectation on the part of the speaker. Such an expectation could be, in a given context, that there are several smokers. Yet, if the answer is Only Jane smokes, this is weaker on a given scale , where the strength of the information is ordered from left to right. Thus, the presupposition that Only Jane smokes is at least as strong as the prejacent Jane smokes. In addition, the descriptive content of Only Jane smokes is at most as strong as the prejacent Jane smokes. Note that “descriptive content” in (18) is another term for assertion/exclusive implication. The prejacent (prejacent implication) is said to be an entailed presupposition (Beaver/Clark 2008: 244); the authors follow Horn (2002) in this assumption.⁶ 6 Standard cases of presupposition, as in the case of presuppositions triggered by factive verbs, are entailed presuppositions. If they can be suspended, this happens through a rhetorical strategy called “letting the hearer down gently” (Beaver/Clark 2008: 246).

Scare Quoted Focus Particles   

   185

All in all, this approach is thought to spell out the idea that the function of exclusives “is partly mirative, to say that the true answer to the Current Question is surprisingly weak, and control the flow of discourse by resetting expectations about that answer” (Beaver/Clark 2008: 250). Beaver and Clark are not particularly clear about the question of why a speaker would entertain these expectations in the first place. But it seems safe to assume that these expectations are pragmatic in nature, being bound to general or personal experience. Interestingly, they claim that “the presence of an expectation that something is stronger than the prejacent is true is an essential part of the meaning of only” (Beaver/ Clark 2008: 252). I think that this observation fits to the meaning component WENIG that has been discussed with respect to the meaning of the German nur. Beaver/Clark (2008: 252) present examples, such as the following, in order to support their claim (presupposing that a suite is preferred to a single room): (19)

a.

I really expected a suite but only got a single room with 2 beds. [web example]

b. # I really expected a single room with 2 beds but only got a suite [constructed variant] With respect to the scales involved, Beaver and Clark draw a distinction between pre-orders (quasi-orders) and partial orders. While pre-orders are reflexive and transitive, partial orders are not.⁷ According to them, most scales discussed in the literature are pre-orders. In their sketch, they elaborate on three scenarios of scale use and the respective meaning components of only. These are, in turn, the objective invite scale, the subjective invite scale, and the subjective autograph scale. These cases are analysed with respect to the following mini-discourses (as usual, focus is marked by the index “F”): (20)

A: Who did Jim invite? B: Jim only invited [Mary and SamF].

7 Beaver and Clark (2008: 251, Fn. 6) explain: “A partial order is a pre-order which is antisymmetric, meaning that two distinct propositions cannot each be as strong as the other.” Note that Beaver/Clark (2008: 68–70) assume that all exclusives are scalar. For them, non-scalar uses are “simply” “more constrained” (p. 69).

186   

(21)

   Jörg Meibauer

A: What celebrity signatures did Brady get at the Philosophy of Language party? B: Brady only got a [SoamesF].

The exchange in (20) is compatible with the objective scale, where the information sought is “purely for information”, and the subjective scale where the questioner’s “primary concern is to evaluate whether the party will rock”. In the latter case, the expectation is that inviting many people is functional for the intended effect. Note that the distinction between these cases is similar to the distinction between quantifying and scalar readings in the traditional approach by Altmann (1976, 2007). The subjective autograph scale in (21) is also connected to an evaluation on the part of the hearer because Soames is ranked low on a scale which includes famous philosophers of language, like Putnam or Searle. According to Beaver/ Clark (2008: 258), the ordering is not a partial ordering, because “it is no better to get both Putnam’s and Schmuckski’s autographs than to get Putnam’s alone, because, after all, who’s ever heard of Schmuckski?” Hence “it is the irrelevance of Schmuckski that means that what we are dealing with here is not partial order but only a pre-order.” (This means that considerations of relevance may have an influence on the contextually adequate scale.) In the framework of Beaver/Clark (2008), our initial example in (1) would be analysed according to the subjective invite scale. The expectation would be that it is in Federline’s interest to get a lot of money. Then we may develop the following scheme: (22)

A discourse based analysis of Spears pays husband only 750,000 euro. [Current Question] How much does Spears give to her husband? Spears pays to her husband only 750,000 euro. Prejacent proposition: Spears pays X, with X = {750,000 euro}. Breakdown of alternative values of X: Stronger: any value significantly higher than 750,000 euro. Weaker: any value significantly lower than 750,000 euro. Presupposition: Spears pays more than 750,000 euro or less than 750,000 euro. Entailment: Spears pays no more than 750,000 euro. Overall implication: Spears pays 750,000 euro and no more.

Scare Quoted Focus Particles   

   187

However, we have to explain how the quotation marks in the scare-quoted version trigger the interpretation that 750,000 euro is not a lot of money for Britney Spears, while it is a lot of money for the average person.

2.3 Is 750,000 Euro a Lot of Money? Let us now have a closer look at example (1), which was actually taken from a newspaper headline; note that the meaning element ‘comparatively modest sum’ is explicated with the main text: (23)

Spears zahlt Mann „nur“ 750.000 Euro ‘Spears pays husband “only” 750, 000 euro.’ […] Dass der Ex-Background-Tänzer von Spears nur den vergleichsweise bescheidenen Betrag erhält, liegt an dem Ehevertrag, den Spears’ Anwältin vor dem Ja-Wort 2004 ausgehandelt hatte. […] ‘The fact that Spears’ ex background dancer only received this comparatively modest sum is due to the prenuptial agreement which Spear’s lawyer had negotiated before the wedding’ [Wiesbadener Kurier, 31 Mar 2007]

Supposing an average reader sees this article, there is likely some knowledge about Britney Spears stored in this reader’s long-term memory. Of particular salience is the information that Britney Spears is quite rich. Let us assume that the average reader tends to estimate that her fortune is roughly 25,000,000 euro.⁸ Furthermore, the average reader tends to think that in a typical divorce, the overall fortune of the couple should be shared. Thus, s/he would assume that Federline would get roughly half of Spear’s fortune, i.e. 12,500,000 euro. However, on reading the headline and the article, s/he is informed that Federline got only 750,000 euro. This is not a lot when taking into account that Spears is wealthy. However, there are the quotation marks on nur. These signal that 750,000 euro is quite a lot from one perspective: each of us would probably be happy to get 750,000 euro. Thus, in the interpretation of the headline, two scales play a role. The first scale is connected to the literal meaning of Spears pays husband only 750,000 euro. This scale ranges from 750,000 euro up to the value of the estimated fortune of Spears, say 25,000,000 euro.

8 Estimates to be found in the web resources differ; however, a fortune of 100,000,000 US dollars seems to be the minimum.

188   

(24)

   Jörg Meibauer

750,000 25,000,000 Scale 1

However, when noticing the scare quotes, another scale is activated by the reader, namely the one in (25): (25)

0 750,000 Scale 2

This has the effect that the amount of 750,000 euro is the right boundary of the scale in (25). Hence, what is not a lot when regarding Scale 1 is a lot when regarding Scale 2. And this relativization of scalar values is the intended effect. After all, for the average person 750,000 euro is a lot of money. Hence the way Scale 1 is represented does not reflect the perspective of the average reader (or of the writer); thus, to preclude that this is the only perspective induced, the quotation marks are used to signal the relevance of the second perspective, represented in Scale 2. Hence the complete content of the headline in (23) (= example (1)) may be represented as in (26). Note that it is left open here as to whether this content is the truth-conditional content; earlier it was argued that at least the first two conjuncts constitute the truth-conditional content. So we still have to consider where the conversational implicature (= the third conjunct) fits in; this will be done in §4. (26)

Spears pays husband no more than 750,000 euro & 750,000 euro is not a lot of money (for Spears) & 750,000 euro is a lot of money (for the average person).

The point being is that the information that 750,000 euro is a lot of money (for the average person) is somehow derived through the operation of pragmatic principles based on appropriate background knowledge. The quotation marks serve as a means to trigger this information.⁹ I will assume that this information is a conversational implicature. According to standard criteria for conversational implicatures (Grice 1989; see the overview in Meibauer 2006), they are cancellable, either by the addition of a cancelling phrase, or by assuming another (plausible)

9 Obviously, there is a contrast between the second and the third conjunct, so that coordination with but would be in order (Wolfram Bublitz, p.c.). I refrain from inserting but into (26), not only because notorious truth-conditional difficulties with but, as discussed broadly in the literature, but also because I assume that the contrast has to with the opposing perspectives, and hence it is pragmatic in nature.

Scare Quoted Focus Particles   

   189

context. In (27), it is shown that a denial of the (putative) implicature is possible in principle: (27)

Speaker A: Spears will pay her husband “only” [air quotes] 750,000 euro. +> 750,000 euro is a lot of money (for the average person). Speaker B: That’s not right./You’re wrong./Bullshit. 750,000 euro is not a lot of money (even for the average). or Actually/nowadays/where we live/in some neighbourhoods, 750,000 euro is not a lot of money (even for the average person).¹⁰

Thus, we can assume that the cancellability criterion for implicatures applies. Also, when the quotation marks are omitted, the additional meaning connected to the second scale and the related scale will not arise. Moreover, conversational implicatures are sensitive to conversational Maxims or Principles that play a role in a process of Gricean inferencing. I will go into this issue in §4 below. What about the bracketed material in (26)? Is this information about the relevant perspectives connected to the scales a proper part of the overall content, or does it merely serve as a hint at how to interpret the scales with respect to background knowledge? Note that König (1991: 45) said that the index “c” in his operator MinC “is also meant to indicate that the evaluation usually expresses the speaker’s point of view.” This means that the selector of a scale is sort of responsible for the evaluation connected with that scale. In the same vein, if (1), the headline of (23), is an answer to the Current Question How much did Spears pay her husband?, then this would imply the average reader’s expectation that she gave him, say, half of her fortune, i.e. 12,500,000 euro. In the headline of (23), the writer simultaneously makes two perspectives available to the reader. In the default reading without the quotation marks, the writer expresses that 750,000 euro is not a lot of money. (Here, the writer’s and Spears’ perspectives overlap)¹¹. With respect to another vantage point, namely that of the average person (or the alter ego of the journalist who penned (1)), it is expressed that 750,000 euro is a lot of money. This latter perspective is triggered by the quotation marks.

10 Thanks to Wolfram Bublitz for proposing these cancelling phrases. 11 At least with respect to the fact that 750,000 euro is not a lot in relation to Spears’ total fortune. Depending on Spears’ degree of avariciousness/generosity, standards with respect to similar divorces, a prenuptial agreement, etc., Spears may very well come to the conclusion that 750,000 euro is very generous.

190   

   Jörg Meibauer

In order to avoid a contradiction in (23), it should be made clear to whom the points of view are attributed. It is tempting to view these additional pieces of information as implicatures in the sense of Bach (1999, 2010). For instance, there are cases of missing constituents, unspecified scope of elements, underspecifity/ underspecification or weakness of encoded conceptual content, overspecifity/ overspecification or narrowness of encoded conceptual contents (Carston 2002). In all these cases, additional inferential steps are necessary to understand what the speaker intends to say.¹² Let us take stock. If it is assumed that the quotation marks trigger an additional scale that is connected with the point of view of someone else (the average person, the writer’s alter ego, etc.), and that the total signification of the sentence containing “nur” matters, then it must be asked which pragmatic principles are in play here.

3 On the Status of Quotation Marks in Scare Quotes From a general linguistic point of view, quotation marks are graphemes that have certain functions in the process of reading and writing (Nunberg 1990, Brendel 2008). As graphemes, they are not treated as lexemes, i.e. as lexical elements possessing a certain form-meaning pairing, though they are of course related to the various linguistic levels (including phonology, morphology, syntax, semantics, and pragmatics). There are several functions of quotation marks to be distinguished which may be illustrated by cases of direct quotation, mixed quotation, pure quotation, and scare quotation (cf. Brendel/Meibauer/Steinbach 2011):¹³ (28)

a.

Direct quotation. Ken said: “The theory is hard to understand.”

b. Mixed quotation. Ken said that the theory is “hard to understand”. c.

Pure quotation. “The theory is hard to understand” is a sentence.

d. Scare quotation. Ken said that the “theory” is hard to understand. 12 There are other concepts in the literature developed to cover types of (presemantic) enrichments, for instance explicature (Carston 2002), pragmatic intrusion (Levinson 2000), or modulation (Recanati 2010). For lack of space, I cannot go into a detailed discussion here. 13 Further uses of quotation marks include: free indirect quotation, book titles, and emphatic quotation (“greengrocer’s quotes”); indirect quotations, e.g., Ken said that the theory is hard to understand, generally lack quotation marks.

Scare Quoted Focus Particles   

   191

Since it is understood that bona fide cases of “nur” are scare quotations, we will focus on analyses of scare quotes in the following (see also Finkbeiner, this volume). It goes without saying that a general theory of the use of quotation marks should cover this case. Whether this use is inherently related with the other types of quotation shown in (28), is beyond the scope of this paper. With regard to the functioning of quotation marks in scare quotes, we may distinguish the following three hypotheses (see Levinson’s 2000 architecture in § 4 below): (H1)

Postsemantic pragmatics. Quotation marks in scare quotes are pragmatic indicators that trigger pragmatic inferences. They have no influence on the semantics of the construction that contains the quotation marks.

(H2)

Presemantic pragmatics. Quotation marks in scare quotes have an influence on the semantics of the construction that contains the quotation marks. This influence may come about indirectly, by shaping the truth conditions through pragmatic inferences.

(H3)

Semantics. Quotation marks in scare quotes have an influence on the semantics of the construction that contains the quotation marks. This influence may come about directly, e.g., by way of a semantic operator.

Taking into account recent discussions on the semantics-pragmatics interface, these hypotheses may turn out to be far too simple. Nevertheless, H1-H3 may be helpful in the understanding of the following sketch.¹⁴ Predelli (2003) proposed a theoretical approach to scare quotes that may be called the message-and-attachment theory. He discusses the following examples (Predelli 2003: 2–4) (scare-quoted expressions given in single quotes): (29)

in offset printing ‘printing proofs’ of illustrations come from the darkroom, not the proof press

(30)

had it not been for Bryce, the ‘coppers nark’, Collins would have made his escape

14 A close analysis, however, is beyond the scope of this paper, because that would imply an indepth discussion of recent disputes on minimalism vis-á-vis contextualism (cf. Borg 2010, Huang 2012). In the following section, I will shortly discuss two approaches to scare quotes, namely those of Predelli (2003), Gutzmann/Stei (2011a, b), and Cappelen/Lepore (2007).

192   

   Jörg Meibauer

(31)

national greed has disguised itself in mandates to govern ‘inferior’ races

(32)

the ‘debate’ resulted in three cracked heads and two broken noses

(33)

this remarkable piece of ‘art’ consists of a large canvas covered with mud and old bus transfers

(34)

myths of ‘paradise lost’ are common in folklore

(35)

life is ‘the farce which everybody has to perform’

(36)

life is ‘what happens while you are making other plans’

In (29), the writer assumes or knows that the word in quotation marks is not really adequate. In (30) the writer uses a slang expression, i.e. the writer excuses himself. In (31), the speaker wants do express a reserved attitude towards the content of the word in quotation marks. Sentences (32) and (33) display sarcasm. In (34), the quotation marks indicate a phrase that is presupposed to be mutually known, and in (35) and (36) the quotation marks indicate quotations whose authors are presupposed to be known (Arthur Rimbaud and John Lennon). In all these cases, the expressions in quotation marks are used, not mentioned. The quotation marks signal that the quoted expressions have certain specific properties, be it that they are used in an unusual way, or that they are attributed to a mutually known source. Let me add that I would not classify (34–36) as scare quotes at all. Note that Klockow (1980), who discusses scare quotes at length, does not include examples of this type either. Predelli (2003: 7) proposes drawing a distinction between the ‘message’ of an utterance and its ‘attachment’. The message is the content of the utterance, while the attachment is the specific meaning that is conveyed by the use of the quotation marks. Hence, the quotation marks in (29)-(36) function as ‘attachment triggers’. These are compared with expressions like but and therefore, which, according to Bach (1999), are also connected with information that surmounts the message of an utterance. For Predelli, it is important that the attachment be a matter of an utterance’s semantic profile: (37)

Truth-functionality of attachments “[…] the content triggered by the quotes is truth-functionally relevant, in the sense that, at least with respect to some contexts, sentences such as [(29)–(36)] display a truth-functionally different profile than their counterparts without quotation marks.” (Predelli 2003: 18).

Scare Quoted Focus Particles   

   193

The reason for that claim is that negations may relate to attachments. For instance, if an American speaker utters We don’t watch ‘colour TV’ here (thereby insisting on the difference between English and American spelling), it is not the message that is negated, but the attachment. However, since this is a case of metalinguistic negation, it does not really prove the need for the notion of attachment. Furthermore, the quoted elements are mentioned, not used. It appears to me that the major drawback of such an analysis is the introduction of new terminology. Firstly, it is not really clear what the message and the attachment are in terms of the notions used in recent debates about the semantics/pragmatics interface, or what their differentia specifica are. Thus, one could suspect that the notion ‘message’ is similar to ‘what is said’, while the notion ‘attachment’ is similar to ‘conventional implicature’ (Horn 2008: 24). The latter is also suggested with respect to the comparison of information triggered by but and therefore, but this is not clear. At least in a Gricean framework (Grice 1989; see also the articles in Gibbs/Colston 2007) irony and sarcasm are conversational implicatures on the basis of an apparent violation of the Maxim of Quality. They are definitely not conventional implicatures because the cancellability criterion for conversational implicatures applies. Secondly, as Gutzmann/Stei (2011a, b) point out, there is the problem that in ironical scare quotes the total semantic profile (as a unit that collapses message and attachment) might contain two contradictory pieces of information, e.g., in Peter’s bagels are “fresh”, the informations that Peter’s bagels are fresh and not fresh holds at the same time. That is, Predelli has to account for the substitutivity of ironies: the fact that ironical implicatures are intended to substitute the ironical utterance, and not, as other (additive) implicatures, to be added to utterances. We can conclude that, as far as scare quotes influence the so-called semantic profile, Predelli’s theory fits with H3, at least when “semantics” is equated with “semantic profile”. However, instead of treating scare quotes as “attachments”, their treatment as conversational implicatures is far more promising. This idea has already been proposed by Klockow (1978, 1980) and was elaborated on, among others, in Meibauer (2007) and Gutzmann/Stei (2011a, b). Pragmatic approaches to quotation marks hold that quotation marks per se do not have a meaning on their own. Instead, they function as pragmatic indicators, telling the readers that they should interpret the material within the quotation marks in a certain way (cf. Klockow 1980, Clark/Gerrig 1990, Saka 1998, Recanati 2001). There is a long tradition of analysing quotation marks as pragmatic indicators, a tradition opposed to genuine semantic approaches. Looking at the different uses of quotation marks, we find that in (28a) and (28b), the quotation marks signal that the material within them is taken to rep-

194   

   Jörg Meibauer

resent an original utterance or a part of that utterance; in (28c), the quotation marks signal that the material within the quotation marks is to be taken as an example of a sentence; in (28d), the quotation marks signal the ironical attitude of the speaker, and finally, in (28e), the quotation marks have an emphatic or highlighting function. The main advantage of a pragmatic approach to quotation marks is that one is not forced to assume the polysemy of quotation marks, or to ignore certain uses of quotation marks altogether – as Cappelen/Lepore (2007) do with respect to scare quotes (but see the next section). On the other hand, the pragmatic approach receives the burden of explaining all the effects of quotation marks with respect to pragmatic principles. In particular, there are two major problems for a pragmatic approach to quotation marks. Firstly, the observation that quotation marks signal, at least in the case of direct quotation, that a certain string of elements has been uttered by an original speaker. Secondly, the observation that the use of quotation marks in direct quotation regularly goes with a context shift (see Bublitz/Hoffmann 2011, Bublitz this volume). For instance, in the sentence Rory said: “I love blues rock.”, the first person pronoun refers to Rory, not to the speaker of the sentence. In contrast, in indirect quotation, e.g. in Rory said that I love blues rock, the first person pronoun refers to the speaker. Thus, the quotation marks have an influence on the computation of indexicals, and hence on the truth-conditions of the sentences containing the quotation. As for the first problem, pragmatic approaches to quotation marks argue that the verbatim assumption – the assumption that a direct quotation renders an original utterance verbatim – cannot be upheld anyhow (Clark/Gerrig 1990), and that similar suggestions of strictly verbatim quotations are possible without any use of quotation marks.¹⁵ As for the second problem mentioned above, the case of deixis (context shift), it might be argued that the distinction between direct quotation (with context shift) versus indirect quotation (without context shift) has nothing to do with quotation marks, for, if other graphemes are used instead of quotation marks, e.g. My friend Peter said – my bagles are very fresh, the same effects show up as in direct quotation (Gutzmann/Stei 2011a). However, this argument hinges on the acceptability of the given examples. A typical property of pragmatic approaches to quotation marks is to treat certain effects of quotation marks as conversational implicatures. This is also 15 I think, however, that Johnson/Lepore (2011), pace Clark/Gerrig (1990), are right in pointing out that quotation is regularly connected to the hearer’s expectation that the quoted material had been uttered verbatim.

Scare Quoted Focus Particles   

   195

the case with Gutzmann/Stei (2011a). Thus, their theory fits with H1 above. The triggering force of quotation marks is due to the operation of the I-principle (in Levinson’s framework, see Levinson 2000: 114), which is related to Grice’s second Maxim of Quantity “Do not make your contribution more informative than is required”. Since quotation marks are, strictly speaking, superfluous devices, they give rise to the derivation of conversational implicatures. Below, I will discuss in more detail whether this is plausible with respect to scare quotes. The crucial observation with respect to the use of quotation marks in writing is of course that, if they are omitted, there is no clear signal towards interpreting the utterance as ironical. Thus, the quotation marks are used in order to block a possible literal interpretation on the part of the hearer. This is in line with traditional approaches to irony where either an explicit signal (such as an ironical tone or air quotes in spoken language, a certain evaluative predicate, etc.) or context information are necessary to trigger of ironical interpretation Proponents of a semantic approach, such as Cappelen/Lepore (2007), tend to neglect scare quotes. However, in an “Appendix” (to their second chapter) called “Scare-quoting”, we read that “a slightly less implausible view would be that scare-quoting and mixed-quotation are, somehow, intimately connected.” However, Cappelen/Lepore deny that this is the case. They argue that “the quotes in mixed quotation cannot be dropped” while “in scare-quoting they can be dropped without a loss of semantic content […]”. (Cappelen/Lepore 2007: 17) In contrast, I would like to argue that scare-quoting and mixed-quotation may be “intimately connected”. Look at the following quotes from the DWDS (Digitales Wörterbuch der deutschen Sprache, see www.dwds.de), in which we face a mixed-quotation/scare quote ambiguity: (38)

Bei einem Ionenstrahl-Versuchsaufbau sei Kühlwasser ausgelaufen und habe in einem Transformator einen starkrauchenden Schwelbrand ausgelöst, der aber von selbst erstickt sei, sagte HMI-Sprecher Thomas Robertson. Der Schaden betrage „nur“ rund 1000 Mark. [Berliner Zeitung, 21 Oct 1995, B160) ‘During an experiment with ion radiation, coolant leaked and set off a strongly smoking, smoldering fire in a transformer, which then smothered itself, said HMI speaker Thomas Robertson. The damage “only” amounts to roughly 1,000 Mark.’

196   

(39)

   Jörg Meibauer

Der Mörder selbst hofft bei der türkischen Justiz auf „Verständnis“, indem er sein Opfer zum Staatsfeind macht: Er habe in Deutschland „ja nur“ einen Mann der verbotenen kurdischen PKK liquidiert. Im Klartext, so schlimm könne nicht sein, was er getan habe. [Wiesbadener Kurier, 22 Dec 2009] ‘The murderer himself hopes for some understanding from the Turkish judiciary by making his victim a public enemy. He “[MP¹⁶] only” liquidated one man from the prohibited Kurdish PKK in Germany. In plain English, what he did can’t be so bad.

I take it that the material in quotation marks is part of what the original speaker said. If mixed quoted, the writer invites the reader to a scare quote reading. More generally, there has to be a motive for the writer to choose only the elements from an original utterance to go into a mixed quotation. This motive is, typically, to draw attention to a typical wording, or to bring in ironical attitudes. Pace Cappelen/Lepore, I think that (like in ironies) the cancellability of scare quote readings is not plausible. For instance, in (47) a possible scare quote reading associated with “theory” is cancelled by the content of the adversative sentence. Consequently, the mixed quote reading of “theory” is forced: (47)

?? Paula insisted that his “theory” is wrong, but I think it is a very respectable approach.

Finally, there is an argument that to my knowledge has not been discussed in the literature yet. Note that in written as well as in spoken language quoted material can be marked by adding the phrase in Anführungszeichen (‘in quotation marks’). The use of quotation marks can even be found in addition to this prepositional phrase, e.g. Das ist in Anführungszeichen „absolut“ falsch (‘This is, in quotation marks, “absolutely” false’).¹⁷ The prepositional phrase in Anführungszeichen has a literal (compositional) meaning, and this literal meaning certainly has to do with giving an instruction on how to read material in its scope. This shows that there may be a need for explicitly marking scare quotation, be it by quotation marks or be it by lexical constructions. I conclude from the foregoing sketch that the function of the quotation marks in scare quotes is to trigger a conversational implicature. But we still do not know how this function is related to the meaning of nur, and on which level of seman-

16 “MP” abbreviates “modal particle”. Roughly, the use of ja signals that it is well-known to the hearer that p. If not, the hearer has to accommodate p. 17 http://www.photovoltaikforum.com/energiepolitik-f90/schwimmendes-lng-kraftwerk-fuerhamburg-alternativ-t81746.html, last accessed 5 Apr 2013.

Scare Quoted Focus Particles   

   197

tic/pragmatic analysis conversational implicatures come into play. In the next section, I will concentrate on H2, and discuss it with regard to Levinson’s (2000) presumptive meaning framework.

4 “Nur” and Presumptive Meaning There are many rivalling approaches to the semantics-pragmatics interface (see Borg 2010, Huang 2012, Meibauer 2012 for reviews). In this section, I will go into the (Neo-Gricean) approach of Levinson (2000), also dubbed the presumptive meaning model. This model contains three pragmatic components, namely Indexical Pragmatics, Gricean Pragmatics 1 and Gricean Pragmatics 2, and two semantic components, namely Compositional Semantics and Semantic Interpretation (modeltheoretic interpretation). The output of Compositional Semantics and Indexical Pragmatics is the input for Gricean Pragmatics 1. The output of Gricean Pragmatics 1 is the input for Semantic Interpretation, and its output (“sentence meaning, proposition expressed”) is the input for Gricean Pragmatics 2, whose output is “speaker meaning, proposition meant by the speaker”. The following diagram offers a visualization of the overall architecture of Levinson’s (2000: 188) framework: Table 1: Presemantic and postsemantic pragmatics (Levinson 2000)

Compositional Semantics

Indexical Pragmatics

Gricean pragmatics 1

‘pragmatic intrusion’

Semantic Interpretation

Gricean pragmatics 2

Whereas Indexical Pragmatics and Gricean Pragmatics 1 are presemantic pragmatic components, Gricean Pragmatics 2 is a postsemantic pragmatic component. It seems that Gricean Pragmatics 1 deals with Generalized Conversational

198   

   Jörg Meibauer

Implicatures (GCIs) (“disambiguation, fixing reference, generality-narrowing, etc.”), while Gricean Pragmatics 2 deals with Particularised Conversational Implicatures (PCIs) (“indirection, irony and tropes, etc.”), but Levinson is not very explicit here. The overall impact of Levinson’s (2000) Neo-Gricean approach is the assumption that GCIs intrude on the presemantic level. This is of course one variant of an underdeterminacy model that is widespread in recent contextualist pragmatics (see Meibauer 2012 for a review). In this section, I will first discuss in more detail the kind of pragmatic principles that apply in the derivation of scare quotes, and then, whether the third conjunct in (26) – ‘750,000 euro is a lot of money (for the average person)’ – is of the presemantic or postsemantic kind. To begin with, let us discuss the M-principle, the I-principle, and the Q-principle, as outlined in Levinson (2000) (cf. Meibauer 2006 for an overview).¹⁸ The M-principle, according to Levinson (2000: 136–137), requires that the speaker “indicate an abnormal, nonstereotypical situation by using marked expressions that contrast with those you would use to describe the corresponding normal, stereotypical situation”, or, as in the expression (“recipient’s corollary”): “What is said in an abnormal way indicates an abnormal situation, or marked messages indicate marked situations […].” The M-principle is said to cover a number of cases, among them lexical doublets, rival word formations, nominal compounds, litotes, certain genitive and zero morpheme constructions, periphrasis and repetitions (Levinson 2000: 138–153). Arguably, “nur” is more marked than nur. In fact, the addition of the quotation marks may be seen as an act of marking. However, because the quotation marks are lexically void (they are not lexemes), the marked version (with the quotation marks) cannot be seen as a lexical alternative having the same denotation as the unmarked version (without the quotation marks). Next, let us have a look at the I-principle (Levinson 2000: 114–115) that is intended to cover, among others, cases like and-coordination (Conjunction buttressing), bridging, and inference to stereotype. Here, it is required that the speaker “say as little as necessary” and “produce the minimal linguistic information sufficient to achieve your communicational ends (bearing Q in mind)”, whereas the recipient needs to “amplify the informational content of the speaker’s utterance, by finding the most specific interpretation […].” If this principle is applied to our case of the scare quoted “nur”, it can be argued that the speaker says Spears only pays her husband 750,000 euro, this being “the minimal linguistic information”. Furthermore, by using quotation marks, the writer invites the recipient to enrich this information such that the

18 “M” relates to modality, “I” to informativeness, “Q” to quantity.

Scare Quoted Focus Particles   

   199

implicature that 750,000 euro is a lot of money (seen from the perspective of the average person), comes into play. Recall that Gutzmann/Stei (2011a, b) hold the view that the operation of quotation marks is generally connected to the I-principle. So in essence, it would be fitting that in the special case of scare quoted nur the I-principle also is at work, such that the implicature “750,000 euro is a lot of money (for the average person)” may be derived. The Q-Principle (Levinson 2000: 76) is devised to cover scalar and clausal implicatures. The Speaker’s maxim is: “Do not provide a statement that is informationally weaker than your knowledge of the world allows, unless providing an informationally stronger statement would contravene the I-principle. Specifically, select the informationally strongest paradigmatic alternate that is consistent with the facts.”, and the recipient’s corollary demands: “Take it that the speaker made the strongest statement consistent with what he knows […].” With regard to our Britney Spears example, the additional activation of Scale 2 may be said to be due to the compliance with the Q-principle (Meibauer 2007). In fact, the writer is more informative here than without the use of quotation marks. Specifically, if the relevance of another perspective is informative for the reader, the writer should give this information instead of suppressing it: what is a lot of money for the average person, is not a lot of money for Britney Spears. In fact, this information may be very important for the reader. Obviously, if s/he does not understand that information, s/he has not understood how the quotation marks work at all. However, while it is plausible to assume that informativity is enhanced, this process does not fit nicely into Levinson’s framework. We simply have not a scale or the like. When we compare Q-inferences with I-inferences, as Levinson (2000: 119) does, we find the following characteristics (see Table 2). In the third column, the properties of “nur” are listed in comparison to standard inferences related to the I-principle and the Q-principle. I conclude, then, that – at least in Levinson’s framework – the conversational implicature triggered by the quotation marks is most likely an I-inference. (In a classical Gricean approach, however, observing the Maxim of Quantity remains plausible.) Let us now ask whether the third conjunct in (26), repeated here for convenience, is of the presemantic or postsemantic kind. (26)

Spears pays husband no more than 750,000 euro & 750,000 euro is not a lot of money (for Spears) & 750,000 euro is a lot of money (for the average person).

200   

   Jörg Meibauer

Table 2: Characteristics of I- versus Q-inferences (Levinson 2000: 119)¹⁹

I-inferences

Q-inferences

“nur”

More specific interpretations (e.g., inference to stereotype)

more precise interpretations (e.g., scalar implicature is a subcase of what is said)

the addition of the second scale is more specific

positive in character

negative in character

the second scale is positively added

guided by stereotypical assumptions

no reference to stereotypical or indeed any background (nonlinguistic) knowledge

stereotypical assumptions about scales play a role

absence of a metalinguistic element

metalinguistic in character: they make reference to something that might have been said but wasn’t

there is no Horn scale¹⁹ or contrast set involved

An argument pro presemantic analysis would hinge on the assumption that focus particles are regularly connected with scales and that the derivation of the necessary scales should be modelled in a localist fashion (all content-related information at one place). However, if one sticks to a localist approach, one has to show that the case of scare quoted exclusives really patterns with the other intrusive (GCI-based) phenomena discussed by Levinson (2000). Or, alternatively, that (26) is identical to what Predelli (2003) calls the “semantic profile”, as explained above. Furthermore, since we are dealing with scare quotes, the analysis should extend to scare quotes in general. So a postsemantic pragmatic analysis with respect to the conversational implicature is simpler. Here, we can model this in a globalist (or modular) way, where the truth conditions of the sentence containing nur are just like in (17b) above, and the pragmatic inferences at the Gricean pragmatics 2 level do all the rest. Data on language processing might help to gain further support, but this is beyond the scope of this paper. Thus, the reader is referred to the works of Noveck/Sperber (2007), Katsos (2012), and Doran et al. (2012) where experimental findings related to the GCI-PCI-distinction, the distinction between ‘what is said’ and ‘what is implicated’, and the context-dependency of implicatures in general are discussed.

19 Scale named after Laurence Horn where the left (strong) element entails the right (weak) element, while the negation of the right element implicates the left element, e.g. .

Scare Quoted Focus Particles   

   201

5 A Note on Ironical Interpretation As seen in (32) and (33), scare quotes often serve to indicate irony or sarcasm. As Klockow (1980: 247–249) has already observed, the writer may have the aim of dissociating herself/himself from a possible speaker (be it an original speaker or an alter ego) who uses the sentence without quotation marks. Telling examples can be given in the context of reported speech; for instance in the following report on a boxing event: (48)

Das Tänzeln im Stile des „Größten“ könnte bei einem echten Puncher noch mit Schmerzen enden. Culcay fand diesen Abend in Schwerin „super“ und war „rundum zufrieden“ mit seinem Debüt. [Frankfurter Rundschau, 21 Dec 2009] ‘Prancing in the style of the “Greatest” [= Muhammad Ali] could – against a true puncher – end in pain. Culcay found this evening in Schwerin to be “super” and was “totally happy” with his debut.’

While the first quote is a scare quote (Applikationsvorbehalt ‘reservation of application’) according to Klockow 1980), the other two are mixed quotes. In the context of the critique in the first sentence, however, these mixed quotes may also be interpreted as ironical scare quotes, since Culcay naively portrays what was actually quite risky as “super” – dancing in the style of Muhammad Ali when not possessing his outstanding boxing abilities. Hence, as already pointed out above, there is sometimes an interesting overlap between three aspects of a scare quoted utterance, namely (i) the mixed quote interpretation, (ii) the scare quote interpretation, and (iii) the ironical attitude of the writer towards the utterer of the mixed quote content. Let us now discuss in more detail whether our example concerning Spear’s divorce can be interpreted in an ironical way. I will briefly allude to aspects of irony discussed in the recent literature; for a more comprehensive account, see Gibbs/ Colston (2007). To begin with, scare quoted focus particles are not used ironically per se. Irony requires a special context. While scales are typically connected with evaluations, more or less neutral scare quote uses are possible. For instance, in Bayern München wurde diesmal “nur” zweiter (‘Bayern München came in “only” second’), this use does not necessarily evoke an ironical interpretation, because a specific context supporting the construction of Bayern München as the ironical target would be needed. (a) Most theories agree that irony is often used to criticize some behaviour or norm. In addition, an ironical target may exist. In our case, several targets come into mind: Britney Spears, Kevin Federline, or VIPs in general. The

202   

   Jörg Meibauer

irony may be targeted at Britney Spears, who seemed stingy on the occasion of her divorce; at Kevin Federline for whom the amount of 750,000 euro is not enough; the rich and famous in general for whom hypocrisy is believed to be a typical trait; someone who holds in earnest that 750,000 euro is not a lot of money without even considering the perspective of people who do not belong to the small group of the very rich, etc. The extent to which an ironical target is selected, will depend on the attitudes of the writer and the reader. This is one important function of scare quoting: to invite the reader to evaluate an event from his or her own point of view. (b) There are several major theories of irony on the market (e.g., pretense theory, echo theory, allusional pretense theory, etc.; see Gibbs/Colston 2007), but there is general acceptance of the idea that contrast between a norm, an expectation, a recent utterance, etc., and the content of an ironical utterance play an important role (cf. Giora 2012). In our example, contrast is built into the description of the truth-conditional content, according to which 750,000 euro is a lot for the average person, whereas it is little for Britney Spears (conceived of as a wealthy pop star). For example, one particular expectation is that her ex-husband would get 50 % of her income. (Note that in the main text in (23), “the comparatively modest sum” is explained with reference to the prenuptial agreement.) (c) The distinction between ironical blame (positive utterance used to convey a negative attitude) and ironical praise (negative utterance used to convey a positive attitude) is much disputed. There is some evidence that the second type of irony is harder to achieve. It appears that both types of irony are available with the scare quoted focus particle “nur”.²⁰ (49)

Ironical blame: [Peter aspires to be Number One, but he comes in third.] [Maurice comments:] Oh, du bist „nur“ Dritter geworden! ‘Oh, you “only” came in third.’

(50)

Ironical praise: [Peter would have been happy, if he had managed to come in third. In the end, he is Number One.] [Maurice comments:] Oh, du bist „nur“ Erster geworden! ‘Oh, you “only” came in first.’

20 A more complete scenario would provide sentences with and without the quotation marks, and would also provide the respective utterances containing the focus particle sogar. This cannot be done here for reasons of space.

Scare Quoted Focus Particles   

   203

In (49), Maurice wants to cheer up Peter insofar as he reminds him that the third place is a relative success. And in (50) he praises Peter’s success while alluding to his restricted expectations. The coming about of an ironical interpretation builds on literal meaning (involving a contrastive perspective), but is not automatically triggered by it. While in standard cases like What a piece of art! or There were many “pieces of art” to be seen., there is a substitutive effect, having roughly to do with the “opposite” of what is said; in the case of scare quoted focus particle “nur” the literal meaning remains intact. In (1) Spears zahlt Mann „nur“ 750.000 Euro. (‘Spears “only” pays husband 750,000 euro.’), it remains a fact that Spears pays her husband no more than 750,000 euro & 750,000 euro is not a lot of money (for Spears) & 750,000 euro is a lot of money (for the average person). Irony comes into play when further aspects of the context are observed. Thus, the ironical implicature adds an additional, contrastive perspective. While for Predelli (2003) ironical meaning goes into the attachment and hence is part of the semantic profile, Levinson (2000: 386, Fn. 2) argues that irony and sarcasm are cases of particularized conversational implicatures (PCIs) (Levinson 2000: 386, Fn. 2). Following this proposal, the ironical interpretation of scare quoted focus particles is delegated to postsemantic pragmatics (Gricean pragmatics 2).²¹ Scare quoted focus particles are not inherently ironical, but their special scalar meaning along with a necessary evaluation may give rise to ironical interpretations in a fitting context. It goes without saying that there are many further questions to be asked, e.g., what the role of contextual clues is (accent on the focus particle, air quotes, lexical indicators like in Anführungszeichen ‘in quotation marks’, and other elements in the context, such as for instance evaluating predicates, see Chapter X in this volume), whether sincerity conditions or conversational maxims are being violated or exploited (cf. Colston 2007), and what the social motives for using this kind of irony are (Dews/Kaplan/Winner 2007).

21 Conventionalized ironies and the indication of irony through special prosodic clues (ironical tone, mocking voice, special accents, etc.) are problematical for this claim, cf. Anolli/Ciceri/ Giaele Infantino (2007).

204   

   Jörg Meibauer

6 Summary The gist of this paper was to explore the idea that, in the case of the scare quoted focus particle “nur”, the addition of a further perspective signalled by the use of quotation marks is crucial. After reviewing the meaning components of nur, it was assumed that the literal meaning of nur in the case of our example Britney Spears only pays husband 750,000 euro is “Spears pays husband no more than 750,000 euro & 750,000 euro is not a lot of money”. The additional information “750,000 euro is a lot of money (for the average person)” was argued to be a conversational implicature, most probably derived by an I-inference. Moreover, examples of mixed quote/scare quote ambiguity and cases of additional interpretation were discussed. Drawing on the earlier work of Klockow (1978, 1980), Meibauer (2007), and Gutzmann/Stei (2011a, b), scare quotes were analysed as pragmatic indicators. This is in line with Gutzmann/Stei (2011a, b), who analyze quotation marks as pragmatic indicators across the board. Cappelen/Lepore (2007), in contrast, defend the view that quotation marks contribute to the semantics of an utterance. Predelli’s (2003) account of scare quotes, which argues that scare quotes contribute to the semantic profile of an utterance and thus have a truth-conditional impact, was dismissed because of the unclear status of the notions message and attachment. What I hope to have achieved in this article, is to have shown that an apparently minor fact like scare quoted focus particles can give rise to ‘bigger’ questions of the semantics/pragmatics distinction.

References Altmann, Hans. 1976. Die Gradpartikeln im Deutschen. Untersuchungen zu ihrer Syntax, Semantik und Pragmatik. Tübingen: Niemeyer. Altmann, Hans. 2007. “Gradpartikel”, in: Ludger Hoffmann (ed.). Handbuch der deutschen Wortarten. Berlin: de Gruyter, 357–385. Anolli, Luigi, Rita Ciceri and Maria Giale Infantino. 2007. “From ‘blame by praise’ to ‘praise by blame’: analysis of vocal patterns in ironic communication”, in: Raymond W. Gibbs Jr. and Herbert L. Colston (eds). Irony in Language and Thought. New York, London: Erlbaum, 361–380. Atlas, Jay David. 1993. “The importance of being ‘only’: testing the Neo-Gricean versus Neo-Entailment Paradigms”. Journal of Semantics 10, 301–318. Atlas, Jay David. 1996. “‘Only’ noun phrases, pseudo-negative polarity items, and monotonicity”. Journal of Semantics 13, 265–328. Bach, Kent. 1999. “The myth of conventional implicature”. Linguistics and Philosophy 22, 327–366.

Scare Quoted Focus Particles   

   205

Bach, Kent. 2010. “Impliciture vs explicature: what’s the difference?”, in: Belén Soria and Esther Romero (eds). Explicit Communication. Robyn Carston’s Pragmatics. Basingstoke: Palgrave Macmillan, 126–137. Beaver, David I. 2004. “Five only pieces”. Theoretical Linguistics 30, 45–64. Beaver, David I. and Brady Z. Clark. 2003. “Always and only: why not all focus-sensitive operators are alike”. Natural Language Semantics 11, 323–362. Beaver, David I. and Brady Z. Clark. 2008. Sense and Sensitivity. How Focus Determines Meaning. Chichester: Wiley. Bolinger, Dwight. 1972. Degree Words. The Hague: Mouton. Bonomi, Andrea and Paolo Casalegno. 1993. “Only: association with focus in event semantics”. Natural Language Semantics 2, 1–45. Borg, Emma. 2010. “Meaning and context: a survey of a contemporary debate”, in: Daniel Whiting (ed.). The Later Wittgenstein on Language. Basingstoke: Palgrave Macmillan, 96–113. Bredel, Ursula. 2008. Die Interpunktion des Deutschen. Ein kompositionelles System zur Online-Steuerung des Lesens. Tübingen: Niemeyer. Brendel, Elke, Jörg Meibauer and Markus Steinbach. 2011. “Exploring the meaning of quotation”, in: Brendel/Meibauer/Steinbach (eds), 1–34. Brendel, Elke, Jörg Meibauer and Markus Steinbach (eds). 2011. Understanding Quotation. Berlin: de Gruyter Bublitz, Wolfram and Christian Hoffmann. 2011. “‘Three Men Using our Toilet all Day Without Flushing – This May Be One of the Worst Sentences I’ve Ever Read’: Quoting in CMC”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Tier, 433–447. Burton-Roberts, Noël. 2006. “Cancellation and intention”. Newcastle Working Papers in Linguistics 12/13, 1–12. Burton-Roberts, Noël. 2010. “Cancellation and intention”, in: Belén Soria and Esther Romero (eds). Explicit Communication. Robyn Carston’s Pragmatics. Basingstoke: Palgrave Macmillan, 138–155. Cappelen, Herman and Ernie Lepore. 2007. Language Turned on Itself. The Semantics and Pragmatics of Metalinguistic Discourse. Oxford: Blackwell. Carston, Robyn. 2002. Thoughts and Utterances. The Pragmatics of Explicit Communication. Oxford: Blackwell. Clark, Herbert H. and Richard J. Gerrig. 1990. “Quotations as demonstrations”. Language 66, 764–805. Colston, Herbert L. 2007. “On necessary conditions for verbal irony comprehension”, in: Raymond W. Gibbs Jr. and Herbert L. Colston (eds). Irony in Language and Thought. New York/London: Erlbaum, 97–134. Dews, Shelley, Joan Kaplan and Ellen Winner. 2007. “Why not say it directly? The social functions of irony”, in: Raymond W. Gibbs Jr. and Herbert L. Colston (eds). Irony in Language and Thought. New York/London: Erlbaum, 297–317. Doran, Ryan, Gregory Ward, Meredith Larson, Yaron McNabb and Rachel E. Baker. 2012. “A novel experimental paradigm for distinguishing between what is said and what is implicated”. Language 88, 124–154. Geurts, Bart and Rob van der Sandt. 2004. “Interpreting focus”. Theoretical Linguistics 30, 1–44.

206   

   Jörg Meibauer

Gibbs, Raymond W. Jr. and Herbert L. Colston (eds). 2007. Irony in Language and Thought. New York/London: Lawrence Erlbaum. Giora, Rachel. 2012. “The psychology of utterance processing: context vs salience”, in: Keith Allan and Kasia M. Jaszczolt (eds). The Cambridge Handbook of Pragmatics. Cambridge: Cambridge University Press, 151–167. Grice, Paul. 1989. “Logic and conversation”, in: Paul Grice. Studies in the Way of Words. Harvard: Harvard University Press, 22–56. Gutzmann, Daniel and Erik Stei. 2011a. “Quotation marks and kinds of meaning. Arguments in favour of a pragmatic account”, in: Brendel/Meibauer/Steinbach (eds), 161–194. Gutzmann, Daniel and Erik Stei. 2011b. “How quotation marks what people do with words”. Journal of Pragmatics 43, 2650–2663. Horn, Laurence R. 1969. “A presuppositional analysis of only and even”. Papers from the 5th Regional Meeting of the Chicago Linguistics Society 5, 98–107. Horn, Laurence R. 1996. “Exclusive company: only and the dynamics of vertical inference”. Journal of Semantics 13, 1–40. Horn, Laurence R. 2002. “Assertoric inertia and NPI licensing”, in: Papers from the 38th Regional Meeting of the Chicago Linguistics Society 38, 55–82. Horn, Laurence R. 2008. “On F-implicature: Myth-analysis and Rehabilitation”. Unpublished Manuscript. Huang, Yan. 2012. “Introduction: what is pragmatics?”, in: Yan Huang. The Oxford Dictionary of Pragmatics. Oxford: Oxford University Press, 1–19. Ippolito, Michaela. 2007. “On the meaning of only”. Journal of Semantics 25, 45–91. Jacobs, Joachim. 1983. Fokus und Skalen. Zur Syntax und Semantik der Gradpartikeln im Deutschen. Tübingen: Niemeyer. Johnson, Michael and Ernie Lepore. 2011. “Misrepresenting misrepresentation”, in: Brendel/ Meibauer/Steinbach (eds), 231–248. Katsos, Napoleon. 2012. “Experimental investigations and pragmatic theorising”, in: Keith Allan and Kasia M. Jaszczolt (eds). The Cambridge Handbook of Pragmatics. Cambridge: Cambridge University Press, 275–290. Klockow, Reinhard. 1978. “Anführungszeichen, Norm und Abweichung”. Linguistische Berichte 57, 14–24. Klockow, Reinhard. 1980. Linguistik der Gänsefüßchen. Untersuchungen zum Gebrauch der Anführungszeichen im gegenwärtigen Deutsch. Frankfurt am Main: Haag und Herchen. König, Ekkehard. 1991. The Meaning of Focus Particles. A Comparative Perspective. London/ New York: Routledge. Levinson, Stephen C. 2000. Presumptive Meanings. The Theory of Generalized Conversational Implicature. Cambridge: MIT Press. Meibauer, Jörg. 2006. “Implicature”, in: Keith Brown (ed.). Encyclopedia of Language and Linguistics. 2nd ed. Vol. 5. Oxford: Elsevier, 568–580. Meibauer, Jörg. 2007. “Syngrapheme als Pragmatische Indikatoren: Anführung und Auslassung”, in: Sabine Döring and Jochen Geilfuß-Wolfgang (eds). Von der Pragmatik zur Grammatik. Leipzig: Universitätsverlag, 21–37. Meibauer, Jörg. 2012. “What is a context? Theoretical and empirical evidence”, in: Rita Finkbeiner, Jörg Meibauer and Petra B. Schumacher (eds). What is a Context? Linguistic Approaches and Challenges. Amsterdam: John Benjamins, 9–32. Noveck, Ira and Dan Sperber. 2007. “The why and how of experimental pragmatics”, in: Noël Burton-Roberts (ed.). Pragmatics. Basingstoke: Palgrave Macmillan, 184–212.

Scare Quoted Focus Particles   

   207

Nunberg, Geoffrey. 1990. The Linguistics of Punctuation. Stanford: CSLI. Predelli, Stefano. 2003. “Scare quotes and their relation to other semantic issues”. Linguistics and Philosophy 26, 1–28. Recanati, François. 2001. “Open quotation”. Mind 110, 637–687. Roberts, Craige. 2006. “Only, Presupposition and Implicature”. Unpublished Manuscript. Rooij, Robert van and Katrin Schulz. 2005. “Only: Meaning and Implicatures”. Unpublished Manuscript. Saka, Paul. 1998. “Quotation and the use-mention-distinction”. Mind 107, 113–135. Sudhoff, Stefan. 2010. Focus Particles in German. Syntax, Prosody, and Information Structure. Amsterdam: John Benjamins.

Klaus P. Schneider

8 Manufacturing Credibility: Academic Quoting Across Cultures Abstract: The present paper reports on “Quoting Across Cultures”, an on-going empirical research project focused on the analysis of academic quoting practices across languages and cultures, and also across academic disciplines and communities of practice. In accordance with the dual perspective of this volume, the analysis involves synchronic and diachronic perspectives. It further includes an examination of the development of quoting competence in university students, comparing their practices with those employed by professional academics. Additionally, the description of academic quoting is combined with an analysis of prescriptive sources such as style guides and research manuals. The current focus is on present-day quoting practices in international journals of linguistics. Information from style guides and research manuals is contrasted with answers given by students in a questionnaire. Interestingly, the students’ responses provide a more differentiated picture of the functions of quoting than prescriptive sources, which, as a rule, deal predominantly with the formal and technical aspects of academic quoting. Against this background, it is argued that apart from the commonly acknowledged supporting and illustrating functions, quoting can also be used as a positioning strategy and for building professional identity. Strategies employed to these ends include practices which are labeled e.g. as ‘doing agreement’ and ‘doing disagreement’ or ‘doing criticism’. Finally, the question is raised of whether cross-cultural variation in quoting practices can actually be expected to occur in academic texts. Considering the high degree of standardization in the scientific community world-wide, which includes subscribing to such values as objectivity as well as using English as an international lingua franca, cross-cultural variation may seem unlikely. However, in the light of findings from such areas as contrastive rhetoric, which is mainly concerned with contrasting the organization of research articles authored by scholars from different linguistic and cultural backgrounds, cross-cultural differences can be expected in the employment of quoting as a positioning strategy. Illustrative analyses suggest that variation in quoting practices may occur not only between speakers of different languages and cultures, but also between older and younger researchers (‘doing seniority’ versus ‘doing juniority’), and between famous and less famous scholars (‘doing superiority or eminence’ versus ‘doing modesty’). We have found that the relationship between other-quoting and also

210   

   Klaus P. Schneider

other-referencing, on the one hand, and self-quoting and self-referencing on the other, may play a crucial role in this type of identity building. Keywords: functions of quoting, quoting practices, positioning strategies, identity building, self-quotations, modesty.

1 Introduction When recently yet another German politician was found guilty of plagiarism in his PhD dissertation, his initial reaction was to claim that he had used an innovative style of quoting newly developed at Harvard.¹ What he did was this: on almost every page he provided no more than one source in a footnote, without giving the original page numbers and without using quotation marks in his text, which often included verbatim repetitions. This sad but at the same time amusing case gives rise to at least two questions. First, do German and American quoting styles really differ? And second, are old styles replaced by new styles? In a more general vein, these two questions concern cross-cultural variation and diachronic change in academic quoting practices. While research in variational pragmatics has shown that many communicative practices differ not only across languages, but also across cultures sharing the same language (e.g., Schneider 2010), it could be argued that academic texts are much more standardized than other discourse genres, notably those of spoken discourse, and that therefore quoting should be universal. It has, however, been found that scientific writing also differs across cultures, ever since Kaplan (1966) postulated, in his ‘contrastive rhetoric’, the existence of cultural thought patterns in speech and writing. For instance, Clyne (1987) identified a number of differences in English and German academic texts which he attributed to diverging argumentative styles. Further differences were found in the macrostructural organization of English and German research articles (among others, Oldenburg 1992), and also across disciplinary cultures, e.g. linguistics and education (among others, Gnutzmann/Lange 1990). The question is whether these differences have ceased to exist in an increasingly globalized world and whether quoting is subject to variation and change.

1 By taking refuge in appealing to (Harvard as) a distinguished authority in order to legitimise his work, he adopted a practice common enough in the Middle Ages; but even then, text creators of medieval texts almost certainly had an understanding of plagiarism and were well aware of what it meant to quote or not to quote (despite a lack of means for indicating acts of quoting), as is shown in Kirner-Ludwig & Zimmermann, this volume.

Manufacturing Credibility: Academic Quoting Across Cultures   

   211

The present paper reports on a new research project aimed at establishing quoting practices in academic texts. Some of the topics addressed in this project include the differences between quoting practices across languages, cultures and academic disciplines, the development of quoting practices through time, particularly over the past sixty years, and also the acquisition of quoting competence. A special focus has been placed on the functions of quoting. It is argued that quoting is not only employed e.g. for supporting an argument (cf. Aijmer, Lutzky, Moore, Johnson, Fetzer/Reber, and Landert, this volume), but also for building identity (cf. Arendholz, Arendholz/Kirner-Ludwig, Bös/Kleinke, and Landert, this volume). In other words, quoting is analyzed as a strategy for self-positioning in the academic community. In section 2, the research project is introduced and its aims are discussed. Furthermore, relevant theories and frameworks are outlined and a working definition of quoting along the lines laid down by Bublitz (this volume) is provided. Section 3 focuses on the functions of quoting. This section is informed by several sources, including style manuals and handbooks, as well as a questionnaire survey among students. Furthermore, an approach is posited in which quoting is understood as a positioning strategy. Section 4 explores cross-cultural variation in academic writing. Against this background we hypothesize that quoting practices may differ not only across languages and cultures, but also across disciplines and communities of practice, and possibly also across generations and between male and female authors. In section 5, a number of examples taken from an international journal of linguistics are discussed which illustrate a range of positioning strategies.

2 The QuAC Project QuAC stands for Quoting Across Cultures, which is the name of an on-going empirical research project carried out at Bonn Applied English Linguistics (BAEL) at the University of Bonn, Germany, in cooperation with the Writing Centre (Kompetenzzentrum Schreiben) at the University of Cologne. This project started in the winter term of 2011/2012. The project has four aims. The first and most general aim is to identify quoting practices used in academic writing. The second aim consists in establishing variation in quoting practices across cultures. Comparison may involve speakers of different languages or speakers sharing the same native language, but not the same practices. Differences may be found between communities of practice in different disciplinary cultures within academia. A further aim is to track changes in

212   

   Klaus P. Schneider

quoting practices through time, which is the diachronic perspective of the project. Finally, the fourth aim is to investigate the development of quoting practices in academic writers, in other words, to examine how novices, i.e. undergraduate and postgraduate students, acquire quoting competence for writing essays, term papers and a thesis in the course of their studies at university. The research questions about quoting in academic writing currently addressed in QuAC include the following: 1. What should writers do? 2. What do novices think they should do? 3. What do novices actually do? 4. What do professionals do? 5. How is what professionals do perceived? To answer these questions, a range of different methods is employed. Research addressing the first question (‘What should writers do?’) is based on the systematic analysis of prescriptive works such as style guides and manuals, e.g. The Oxford Guide to Style (Ritter 2002). Research on the second question (‘What do novices think they should do?’) makes use of a Questionnaire on Quoting, which was developed for the purposes of QuAC and tested in a pilot study at the University of Bonn in the winter term of 2012/2013. After this pilot study, a revised version of the questionnaire (QQ 1.1) was used in a first survey conducted at the Writing Centre of Cologne University in the spring of 2012 (N=94). What novices actually do, on the other hand (Question 3 above), i.e. who, what, when and how they quote, can be observed in their unpublished work, e.g. in assignments, term papers and BA theses. What, by contrast, professionals do (Question 4) is revealed in an analysis of their publications, e.g. research articles and monographs. The last of the five above questions, i.e. how readers perceive what academic writers do, requires an answer which is informed by experimental work on reader perception, for instance by appropriateness ratings (cf. also Schneider 2013a). Several frameworks are relevant for this project which includes text analysis; particularly approaches which are focused on non-fictional texts (e.g. Brinker 2010; for an overview cf. Titscher et al. 2000), and approaches developed in genre analysis of the Swalesian type (e.g. Swales 1990), originally intended for practical purposes in applied contexts such as teaching foreign students at American universities how to write research articles in English as a second language (cf. also Tardy/Swales 2014). Contrastive studies on Language for Special Purposes (LSP) and specifically Language for Academic Purposes (LAP) are also immediately relevant. For instance, Gnutzmann and his team published a number of studies comparing research articles in English and in German in various academic disciplines (e.g. Gnutzmann/Lange 1990, Oldenburg 1992). Contrastive rhetoric,

Manufacturing Credibility: Academic Quoting Across Cultures   

   213

which is older than the other approaches mentioned here, also has its origin in the context of teaching academic writing to students at American colleges and universities whose native language is not English (cf. Kaplan 1966). However, in this approach a more general perspective is adopted in comparing ‘intellectual styles’ across so-called super-cultures (cf., e.g., Galtung 1981). Against the background of these earlier studies, Clyne (1987) contrasts academic texts written in English and in German (cf. section 4 below). These more practical approaches are supplemented by a range of theoretical frameworks from different traditions. One of these frameworks is variational pragmatics, which deals with intralingual pragmatic variation, i.e. variation in language-use conventions across varieties of the same language (cf., e.g., Schneider/Barron 2008/2011). While most studies in this field concentrate on the impact of such macrosocial factors as nation, region, gender and age, the influence of microsocial factors such as distance and power is also acknowledged (e.g. Barron/Schneider 2009) as well as the influence of discourse types and genres (e.g. Schneider 2007). Another relevant framework is evaluation theory, which provides the tools for the systematic analysis of evaluative expressions and assessments (e.g. Thompson/Hunston 2000). The analysis of the interactional and interpersonal aspects of academic writing and academic quoting, on the other hand, is grounded in current theories of (im)politeness and relational work (cf., e.g., Locher/Watts 2005, Culpeper 2011), whereas research into social cognition helps shed some light on the cognitive underpinnings of these interactional and interpersonal aspects (cf., e.g., Langlotz 2010). Any analysis of quoting practices has to be based on a definition of quotation. In accordance with the conceptual frame of the present volume, we adopt the definition of quoting presented in the introduction as the springboard for our investigation. Hence, quoting is regarded as a complex meta-communicative act of a quoter re-contextualizing, re-focusing and reflecting upon another person’s or their own prior communicative act (cf. Bublitz, this volume). This definition is general enough in that it is not restricted to the academic domain, it refers to both written and spoken discourse, and it does not exclude self-quotation, which is a relatively frequent feature of academic texts and especially of research articles (cf. section 5). For the purposes of the QuAC project, we prefer a more specific and static working definition which focuses not on the act but on the outcome of quoting and captures the typical features of quoting in academic writing: A quotation is a group of words prototypically (i.e. most frequently) taken from a previously written (and usually published) text (or, much less frequently, from presentations or personal communication) and repeated in some form (verbatim or as a paraphrase) in a more recent text.

214   

   Klaus P. Schneider

This definition also covers self-quoting, as reference is made not to people (as in this volume’s introduction), but to texts. It might be added here that in academic writing quotations have to be unambiguously marked as having been adopted from an earlier source. Where this is not made sufficiently clear, we are faced with cases of intentional or unintended plagiarism (cf. Kirner-Ludwig & Zimmermann, this volume).

3 Functions of Quoting The Chicago Manual of Style (1993: 356) offers the following guideline: “Ideally, authors of works of original scholarship present their arguments in their own words […].” To the innocent mind, this guideline may suggest that quoting in academic writing is undesirable, should be avoided, and may give rise to such questions as ‘Why don’t writers simply say what they have to say?’, ‘Why should they quote and resort to the words of other authors?’ or ‘What are the functions of quoting in scientific texts and research articles anyway?’ In the present section of this paper, the question about the functions of academic quoting will be answered from three perspectives. First, I will briefly consider what style guides, manuals and handbooks have to say on this topic. Then, I will examine what university students think the functions of quoting are, and finally I will offer a more specific view on this issue which has been developed in the framework of the QuAC project. In this view, quoting is seen as a positioning strategy, which is, of course, a familiar feature of other, modern-day genres such as online message board presentations, comments in online discussion fora on TV or in weekly journals, and online news articles (as shown in Arendholz, Bös/Kleinke, and Landert, respectively, this volume). Style guides, manuals and handbooks such as The Oxford Guide to Style (Ritter 2002), The Chicago Manual of Style (1993) and the MLA Handbook for Writers of Research Papers (2009) do not have very much to say about the functions of quoting in academic texts. They are predominantly focused on the “Mechanics of Writing” (title of the respective chapter in the MLA Handbook) and concentrate more or less exclusively on the how and not the why of quoting, i.e. on the formal and technical aspects and not on its functions. Student guides to doing research, however, may contain some information about the functions of quoting. For instance, in The Map: A Beginner’s Guide to Doing Research in Translation Studies (Williams/Chesterman 2002: 104) at least some functions are mentioned, albeit in passing, e.g. that quoting can be employed for supporting an argument or for illustrating a claim.

Manufacturing Credibility: Academic Quoting Across Cultures   

   215

Students, on the other hand, both postgraduates and undergraduates, show some awareness of a range of different functions of academic quoting. The following illustrative examples are taken from answers to the first question in the Questionnaire on Quoting used in the original QuAC pilot study (cf. section 2 above): 1. What are the functions of quoting in academic texts? The answers were given by undergraduate students in their final year at the University of Bonn. The examples listed below appear in their original wording, errors of spelling and grammar etc. have not been corrected. – make texts more academic and trustworthy – showing that other people thought something/have worked on a subject too – situating in the progression of scientific discourse – clearly differentiate own thought & work from those of previous authors – to present some examples of different authors According to the views expressed in these and similar examples, quoting is a typical feature of academic texts and serves the purpose of making texts “trustworthy”, i.e. reliable or convincing. Also, students highlight the processual nature of research. According to them, quoting demonstrates that one’s own work does not exist in a vacuum and shows that previous research is available on the given subject. Moreover, quoting situates one’s own work in the subject area, with its debates and developments. It is further emphasized that one’s own ideas and those of other researchers must be clearly kept apart, thus displaying an awareness of the dangers of plagiarism. Finally, the last of the above examples refers to the illustrative function of quoting also mentioned in Williams/Chesterman (2002). There is no denying that supporting a writer’s ideas, theory, argumentation or findings is one of the main functions, if not the most important function, of quoting previous work and other authors in academic texts. Yet, in the QuAC project we advocate the view that quoting is also a positioning strategy. In this context, positioning is understood as identity building (cf. Davies/Harré 1990). Following De Fina, who observes that “[p]ositions are acquired through social and discursive practices” (2010: 212), we suggest that quoting practices are among the social and discursive practices through which academic writers build a professional or scholarly identity for themselves. This type of identity building is called ‘reflexive positioning’ (as opposed to ‘interactional positioning’) in the terminology introduced by Davies/Harré (1990: 48); in QuAC it is called, more specifically, ‘academic self-positioning’ (cf. section 5 below). Essentially, acts of quoting position the quoter in two ways, namely first in the respective field or subject area, and second vis-à-vis the reader and in the community. By displaying (profound) knowledge, or at least awareness, of previ-

216   

   Klaus P. Schneider

ous relevant research, quoting positions the quoter in the field. More specifically, in discussing previous relevant research, the quoter associates with or dissociates from particular ideas, theories, approaches, schools, traditions, etc. In other words, quoters project the image of competent researchers with an adequate overview of relevant issues and also a marked stance regarding these issues. This first subtype of academic self-positioning might be called ‘illocutionary’ positioning (to use speech act-theory terminology). That is to say, quoters build this type of identity in quoting. The other subtype of academic self-positioning, i.e. positioning the quoter vis-à-vis the reader and in the community, is perhaps a more explicit strategy of self-marketing which could be glossed as ‘impress the reader’ (including reviewers and colleagues), or, in the case of student writers, ‘impress the supervisor’. This can for instance be done by quoting a large amount of publications and/or publications from a wide range of disciplines or subdisciplines and/or the latest and most recent publications. By employing these and similar practices, novices can build an identity as good, hard-working, bright, original, up-to-date, promising etc. students, and the same applies, mutatis mutandis, for professional academics. Moreover, this subtype of academic self-positioning can be used as an anticipatory strategy of self-defense. Quoting not just any researcher who has worked on the subject but specifically associating with eminent scholars and authorities in the field can be used as a strategy to preclude criticism (cf. for a similar motivation, Kirner-Ludwig/Zimmermann, Musolff, and, in medieval texts, Rudolf, this volume). Referring to a different, non-academic context, Kettemann/ Marko (2011: 451), in their CDA framework, call this type of strategy “intertextual immunisation” (cf. also Kettemann/Marko 2011: 454). Furthermore, quoting not published, but unpublished sources or not yet published texts authored by eminent scholars, or even quoting “personal communications” with these scholars may be employed to claim insider status and membership to a powerful circle and also, perhaps, eminence and authority for oneself. This second subtype of academic self-positioning might be termed ‘relational’, ‘interpersonal’ or ‘perlocutionary’ positioning. By quoting, the said claims are made and the said aims are achieved. Obviously, the distinction between “in quoting” and “by quoting” goes back to Austin’s distinction between “in saying” and “by saying” to differentiate between illocutionary and perlocutionary acts (cf. Austin 1962). One aim of the QuAC project is to establish how academic self-positioning is actually done, i.e. which particular quoting practices are employed for doing agreement, disagreement, evaluation, criticism, and so on. As some of these acts are (inherently) face-threatening and as notions of politeness, rudeness and appropriateness have been found to differ across languages and cultures (cf., e.g., Schneider 2012), quoting practices, rather than the functions of quoting, can

Manufacturing Credibility: Academic Quoting Across Cultures   

   217

be expected to differ as well. The extent to which such practices diverge interlingually and interculturally is the subject of the next section.

4 Cross-Cultural Variation in Academic Practices The question is whether variation in quoting practices can really be expected. While it is well known that participant practices in conversation and other types of spoken discourse differ in many ways across languages and cultures (cf. Buchstaller & van Alphen 2012 for a recent overview and, more specifically, Finkbeiner, this volume), this does not necessarily mean that the same holds for written discourse and, more particularly, for academic writing. Written discourse in general is much less spontaneous and also more formal than spoken discourse; especially conversation. This applies to an even greater extent to academic texts, which are, moreover, standardized to a high degree; cf., for instance, the so-called “IMRD” structure of research articles (i.e. Introduction – Methods – Results – Discussion) accepted in many scientific disciplines world-wide (cf. Swales 1990). The global scientific community seems to be a community of practice whose members subscribe to the same international standards, not least because science is considered objective and culture-independent, and most researchers today use one and the same language, namely English, as their lingua franca. From this perspective, variation in quoting seems unlikely. There is, however, evidence available that strongly suggests that there are, after all, cultural influences on academic work and thinking. As early as 1966, Kaplan published a paper which is generally acknowledged to mark the beginning of the field known as ‘contrastive rhetoric’. Kaplan’s study was based on the observation that foreign students at American universities were found incapable of writing “proper” essays, term papers, research reports, etc., “proper” academic texts by American standards in any case. The explanation offered by Kaplan was that these foreign students were unaware of the American standards and followed the patterns of their native languages and cultures instead. Furthermore, Kaplan found these patterns to differ systematically across cultures, in other words, each culture has its own “thought pattern” which is reflected in the argumentation of academic texts. Based on a corpus of several hundred essays written by international students at American universities, Kaplan postulated five such cultural thought patterns, which he named English, Semitic, Oriental, Romance and Russian. These five he described in some detail. He characterized the English thought pattern, prevailing in the English-speaking world and specifically in the United States of

218   

   Klaus P. Schneider

America, as linear, i.e. as straight-forward and without digression. The Semitic thought pattern, by contrast, includes many parallel constructions. This type of thought pattern is typical in the Arabic world. The thought pattern Kaplan termed Oriental was identified in texts authored by students from China and Korea, but not in texts produced by Japanese students. Texts following this particular thought pattern were found to circle around the topic. The two remaining types of thought patterns are called Romance and Russian, with the label ‘Romance’ pertaining to the speakers of such languages as French or Italian. Both these patterns, the Romance as well as the Russian, are referred to as digressive, yet the kinds of digression differ. It is worth noting in this context that these five thought patterns were not conceptualized as national patterns, but as patterns characteristic of larger, super-national cultural spaces. Kaplan also made an attempt to visualize the five pattern types he identified in his now well-known and widely quoted drawings (Figure 1).

Figure 1: Kaplan’s five ‘cultural thought patterns’ (Kaplan 1966)

A similar approach was adopted by Galtung in a paper published in 1981, in which he contrasts ‘intellectual styles’. An intellectual style is defined as how “we process impressions into expressions” (Galtung 1981: 818). The author claims that these styles vary across so-called macro-cultures, of which only very few are assumed to exist world-wide. He posits that under the umbrella of each macroculture, there are several ‘sub-civilizations’. Galtung sets out to determine the specific nature of the intellectual styles in four such sub-civilizations, which he names Saxonic, Teutonic, Gallic and Nipponic. According to Galtung, Saxonic style, which is typical of the Anglo-American sub-civilization, is, in essence, empirical description. Teutonic style, on the other hand, which is found e.g. in Germany, is said to favour theoretical explanation. Gallic style, typically employed in a French speaking context, is also characterized by theoretical explanation, which is, however, specified as dialectic, whereas theoretical explanation in Teutonic style is specified as non-dialectic. Finally, the distinctive feature of Nipponic

Manufacturing Credibility: Academic Quoting Across Cultures   

   219

style, predominantly adopted, as the name suggests, in Japan, is commentary on other intellectuals. Galtung, whose classification is based on his own experience in various cultures and not on empirical work, also provides visualizations of these intellectual styles, but, unlike Kaplan, he cautions his readers to take these visualizations cum grano salis (Galtung 1981: 839). Following Kaplan’s empirical tradition, Clyne (1987) developed contrastive rhetoric further by adopting a more linguistic and more concrete descriptive approach. Using a corpus of research articles written in (Australian) English and (German) German, he analysed the differences between these texts which he attributes to cultural differences. His main findings can be summarized as follows. English texts are organized in a linear fashion (confirming Kaplan’s – slightly normative and ethnocentric – findings) and German texts include digressive passages. Moreover, English texts are symmetrical and German texts asymmetrical, i.e. displaying an unbalanced structure. While English texts include a large number of so-called advance organizers, e.g. connectors and linking words guiding the reader by making the structure and organization of an article explicit, such elements are largely absent from German texts. A further difference concerns the status of examples and statistics, which are fully integrated into English texts, while they are relegated to footnotes or an appendix in German texts. Overall, English research articles are reader-oriented and German articles content-oriented. Clyne examined these and similar parameters, but did not note any differences pertaining to quoting. Against this general background, quoting practices can be expected to vary across languages and cultures. Furthermore, there is evidence to suggest that variation occurs across academic disciplines and disciplinary cultures. For instance, Gnutzmann/Lange (1990), who compared English and German research articles in a Swalesian framework (Swales 1990), observed differences between the disciplines of linguistics and education. They noted that more language-specific macro-structural features are found in educational articles and concluded that disciplines vary regarding the proportion of such language-specific and language-independent features, depending on their level of abstraction. There is reason to believe that there are also differences across disciplinary subcultures within the same discipline. Apart from such general distinctions as theoretical versus applied, or quantitative versus qualitative research, there are various communities of practice in the same discipline rooted in particular traditions and schools of thinking. While there are certainly marked differences between the approaches normally adopted in e.g. phonology and syntax, on the one hand, and pragmatics and anthropological linguistics, on the other hand, to take the example of linguistics, differences also exist in each of these areas; e.g. Gricean (including neo- and post-Gricean) approaches versus sociolinguistic and applied

220   

   Klaus P. Schneider

approaches in pragmatics, or generativists versus proponents of construction grammar in syntax. Finally, it is perfectly conceivable that within these communities of practice, characteristic differences may appear between senior and junior researchers and perhaps also between male and female researchers. Whether or not quoting practices vary across these parameters can only be established with empirical work. For this purpose, a suitable text corpus has to be designed. To warrant maximal comparability of the texts, a number of variables would have to be controlled for. These variables include publication date, field of research, text genre, and author identity. Ideally, all texts in the corpus would have to have been published within the same few years, or at least within a very limited time span (e.g. within five years), thus providing a synchronic snapshot. The more limited the field of research, the more reliable the results. As a text genre, research articles could be chosen, i.e. articles reporting on genuine (empirical) research, which may be considered a prototypical academic genre. Alternatively, overview articles or chapters in handbooks or in textbooks could be examined, but not, e.g., book reviews, as quoting is rare in this genre. The identity of authors is relevant as well, especially their native language and culture. Establishing this may, however, not be trivial. Authors may have been born and raised in a language and culture significantly different from the language and culture in which they were educated at university and in which they received their qualifications, e.g. their bachelor’s, master’s or doctoral degree. It may be easier to establish age and gender. In this context, the fact that generational and gender differences also count as cross-cultural differences has to be emphasized, at least in the framework provided by variational pragmatics (cf., e.g., Schneider/ Barron 2008/2011, and Barron/Schneider 2009).

5 Quoting as Positioning in Journal Articles: Some Illustrative Examples In the present section, the corpus-based analysis of quoting practices outlined above is illustrated with a few examples. Given the spatial limitations of this paper, and also the current stage of the QuAC project, it is not possible to provide a more systematic demonstration. All examples discussed below are taken from the same issue of the journal Intercultural Pragmatics. As can be seen from the journal title, the area covered is fairly specific and relatively narrow, e.g. by comparison to the areas covered in such journals as Language or Journal of English Linguistics. The issue of Intercultural Pragmatics which was chosen is issue 1 of volume 2, published in 2005. In the discussion of the examples taken from this

Manufacturing Credibility: Academic Quoting Across Cultures   

   221

journal issue, we will return to the idea developed in section 3 that quoting is an act of positioning. The first example illustrates a positioning strategy termed ‘Doing Criticism’. The following passage was taken from the article “Linguistic Pragmatics: A Discipline Bedeviled by its Own History? Implications for Intercultural Studies”, authored by Kanavillil Rajagopalan: Attempts to break with the past are often met with rearguard action aimed at putting research back on the beaten track. A highly successful move in this direction is the work of Sperber and Wilson (1986), aptly characterized by Levinson as ‘an ambitious bid for a paradigm-change in pragmatics’ whereby ‘pragmatics is reduced to a single mental reflex,[…]’ (Levinson 1989: 469). (Rajagopalan 2005: 96)

The claim made in this passage is that new developments (“attempts to break with the past”) in pragmatics are frequently countered by scholars wishing to turn back the wheel (“rearguard action aimed at putting research back on the beaten track”). The author then refers to Sperber/Wilson and their 1986 book on Relevance, which he considers a particularly good example of such retrogressive action (“a highly successful move in this direction”). This attack on Sperber and Wilson is supported by a quote from Levinson (1989), a publication in which Relevance Theory is described in openly negative terms. The key words in the two quotations from Levinson (1989) are ambitious in “an ambitious bid for a paradigm-change in pragmatics”, and reduced in “pragmatics is reduced to a single mental reflex, […]”, and also single in this second quotation, a specification which indicates that further reduction is not possible. Thus, for doing criticism, Rajagopalan employs three strategies simultaneously. First, he quotes from another writer to support his argument, which seems to be the most common use of quotation in academic texts (cf. section 3 above). Second, the quoted author is more outspoken in his negative assessment than the quoting author, i.e. the negative evaluation which is quoted is more detailed and direct and, hence, more facethreatening, which is, however, the quotee’s, and not the quoter’s, responsibility. This strategy could, therefore, be termed ‘shifting responsibility’. Third, Rajagopalan criticizes two colleagues who are better known than him by supporting his attack with a quotation from a colleague who is also better known (and, arguably, perhaps even more famous than those attacked). In other words, Rajagopalan employs a strategy which Kettemann/Marko (2011) termed ‘intertextual immunisation’ (cf. section 3). The relationship between these three strategies is that of inclusion. They are distinguished here for purely analytical purposes, but otherwise work like Russian dolls (matryoshki). One last point about this first example: not only is Levinson quoted by Rajagopalan to support his opinion that Sperber and Wilson are good examples of backward-looking scholars, Rajagopalan also

222   

   Klaus P. Schneider

explicitly affiliates with Levinson’s negative evaluation by using aptly in his phrase “aptly characterized by Levinson as”. It would be interesting to examine aptly in large corpora of academic discourse to identify the patterns in which it occurs in connection with quotations and to see how it is used in associating with the authors quoted, directly or indirectly, and their opinions and assessments. Further expressions which could also be studied in this way include the adverbs succinctly and convincingly. The second example used for this illustration is Wilson and Sperber’s response to Rajagopalan’s criticism, which appeared in the same issue of the journal. What Wilson and Sperber do in their response could be called ‘Doing counter-attack’. Here is a representative passage from their response: The polemical tone which leads him, for example, to describe our work imaginatively as a ‘rearguard action’ of a type ‘carefully designed […]’ is best ignored. (Wilson/Sperber 2005: 99)

In this passage, in which Wilson and Sperber explicitly refer to Rajagopalan’s categorization of their work as “rearguard action” as quoted above, they accuse him of using a “polemical tone” by adopting a polemical tone themselves, thus paying him back in his own coin. In other words, they employ a retaliation strategy. In an endnote they add: “The polemic is, in any case, quite hard to keep track of.” (Wilson/Sperber 2005: 102), suggesting that Rajagopalan’s paper is not well written or not coherently argued and, therefore, not to be taken seriously. To make their counter-attack more effective, they use a further strategy which could be called ‘Doing Seniority’. This consists in referring to 22 publications from their own ‘camp’, i.e. works authored by scholars adopting their framework, to provide evidence for the claims which were attacked by Rajagopalan. The reference section in Wilson/Sperber’s contribution includes only these 22 publications plus two self-references, i.e. two of their own publications. The strategy ‘Doing Seniority’ could alternatively be called ‘Doing Superiority’ or ‘Doing Eminence’. It is also used in a paper by Gumperz/Cook-Gumperz in the same 2005 issue of the journal Intercultural Pragmatics. The opening sentence of their paper starts as follows: “In an earlier paper, ‘Language Standardization and the Complexity of Communicative Practice’ (Gumperz/Cook-Gumperz 2004), we argued […].” While it may not be considered unusual to write a follow-up to an earlier article and to refer to this earlier publication at the very beginning of the follow-up publication, what could be considered unusual, however, is seen when examining the reference section of the paper. Ten out of 26 works listed in this section are self-references, which amounts to a share of 39 percent. As this example demonstrates, references can be quantified and the share of self-

Manufacturing Credibility: Academic Quoting Across Cultures   

   223

references can be taken as a measure of identity building in terms of eminence (or seniority or superiority). Needless to say, the same applies to quotations and self-quotations as well. In our times, in which ‘publish or perish’ is the dominant ideology in the academic world around the globe, and ratings, rankings and citation indexes are gaining an increasing influence, this purely quantitative dimension is of growing importance. In a diachronic perspective, it would be interesting to see how self-referencing and self-quoting have developed over time, given that modesty appears to be a traditional academic virtue. It would also be interesting to carry out experimental studies to establish which degree of self-referencing and self-quoting is perceived as inappropriate boasting among academics from various cultural backgrounds. In quite different contexts, it has been found that modesty seems to be a universal value. For instance, it has been observed that diminutives are used to play down one’s own achievements or possessions (cf. Staverman’s ‘diminutivum modestum’; Staverman 1953: 409) in such languages as English, German and Dutch as well as Italian and Greek, but also in typologically completely unrelated languages such as Snohomish (a Salishan language spoken in Washington State, USA) and Selee (a Niger-Congo language spoken in Ghana) (cf. Schneider 2013b for examples and references). On the other hand, in work on responses to compliments it has been found that the weight of Leech’s Modesty maxim of politeness (“Minimize praise of self”; Leech 1983: 132, original emphasis) varies across cultures. For example, Chen (1993) reports that in their compliment responses, speakers of Chinese orient predominantly towards the Modesty maxim, whereas speakers of American English orient more towards Leech’s Agreement maxim (“Maximize agreement between self and other”; Leech 1983: 32, original emphasis). However, in a later replication of the Chinese part of this study, Chen/Yang (2010) found that speakers of Chinese orient significantly less to the Modesty maxim than they did in the past. This finding suggests that political and economic developments may have an impact on attitudes towards modesty even in the short term. It seems that modesty is valued more by younger scholars. In the concluding section of his paper in Intercultural Pragmatics 2:1 (2005), Haugh includes an overt display of modesty. He writes: “The aim of the analysis in this paper has thus been rather more modest, […]” (Haugh 2005: 61). His point of comparison is Wierzbicka’s Natural Semantic Metalanguage, which he characterizes as “perhaps the most ambitious attempt to formulate a universal metalanguage for cross-cultural and intercultural analyses” (Haugh 2005: 60).² Given that Wierzbicka is a 2 Cf. the use of ambitious in Rajagopalan’s quotation from Levinson (1989) cited at the beginning of the present section; both ambitious and modest would merit a corpus-based study of their uses in academic discourse.

224   

   Klaus P. Schneider

senior and eminent scholar, Haugh’s display of modesty could be interpreted as ‘Doing Juniority’. This interpretation is supported by an act of thanking included in Haugh’s acknowledgement, where he writes: “I would like to thank Professor Robert Arundale for […]” (Haugh 2005: 64). It is not the act of thanking as such which indicates juniority, but the fact that the title of the individual thanked is used in this context. This choice emphasizes that appreciation is expressed not from professor to professor, but from junior to senior. Such acts of thanking, generally employed not only to avoid (covert) plagiarism, but also, needless to say, to convey genuine gratitude, can be regarded as a further (or embedded) type of other-referencing which is very similar to the ‘personal communication’ type mentioned in section 3 (i.e. claiming insider status and membership to a powerful circle) and may, hence, serve as an affiliation strategy. This strategy may even be considered a defensive strategy or a special case of ‘intertextual immunisation’ (cf. section 3 above), and would perhaps be more aptly termed as ‘interpersonal immunisation’. The last example which illustrates ‘Doing Juniority’ is also taken from the acknowledgements, in this case those in a paper by Robbins. Robbins first of all thanks her “mentor and friend”, the eminent A. A. Leontiev (who had just passed away), and then the anonymous reviewers of her paper. Finally, she adds the following sentence: “And, although I speak out against the Chomskyan model of linguistics, I have deep admiration for Noam Chomsky” (Robbins 2005: 37). Thus, after challenging Chomsky’s position in her paper and clearly dissociating from his work, Robbins expresses general admiration for the person in order to mitigate her criticism of this eminent senior colleague and his famous ideas. Thus, ‘Doing Juniority’ involves other-referencing and other-quoting, especially when referring to and quoting famous scholars, and also such defensive strategies as explicitly mitigating disagreement and criticism. ‘Doing Seniority’, by contrast, involves a high degree of self-referencing and self-quoting, and also such offensive strategies as counter-attacking.

6 Conclusion Quoting is a prototypical feature of academic texts, which fulfills a number of specific functions which can be summarized as ‘manufacturing credibility’. Two major functions include supporting an argument and illustrating a claim. These functions are known to students, but only rarely mentioned in style guides and research manuals. In this paper it is argued that quoting in academic texts is also a positioning strategy. In quoting, writers position themselves in the field, i.e. vis-

Manufacturing Credibility: Academic Quoting Across Cultures   

   225

à-vis traditions, theories, frameworks and approaches (‘illocutionary positioning’). At the same time, by quoting, writers position themselves in the academic community, i.e. on the interpersonal level as senior or as junior members of this community, affiliating with (other) eminent colleagues (‘perlocutionary positioning’). How is quoting used to perform acts of relational work? This is the current focus of the ongoing research project Quoting Across Cultures (QuAC) carried out at the University of Bonn. As ‘thought patterns’, ‘intellectual styles’ and academic texts have been found to differ across languages and cultures (cf. section 4), cross-cultural variation can also be expected in quoting practices such as doing agreement or doing criticism which concern issues of politeness, appropriateness and facework. Questions to be addressed in this perspective include who quotes whom, where and why. Of particular interest is how authors assess the work of other authors by employing such adjectives as ambitious or such adverbs as aptly and convincingly. The quantification of self-references and self-quotations also provides relevant insights, especially about which identities of themselves writers wish to project, and not least their attitude towards professional modesty. These and similar issues are best addressed by using comparable corpora of academic discourse. This is a task for future research, and the next step that will be taken in the QuAC project. Further topics which will be addressed in this project include the acquisition of quoting competence by academic novices.

Acknowledgements I would like to thank Wolfram Bublitz, Christian Hoffmann, Jenny Arendholz und Monika Kirner-Ludwig for giving me the opportunity to present my ideas at the Third International Conference on Quoting and Meaning (ICQM3), which they organized at the University of Augsburg, Germany, in April 2012. Also, I would like to thank the audience for their discussion of my talk. I am particularly indebted to Alison Johnson for her detailed feedback. Finally, I am grateful to an anonymous reviewer for many useful comments. And last but not least, special thanks go to Wolfram Bublitz for his support in the final stages of getting this paper published. Needless to say, the responsibility for any shortcomings is entirely my own.

226   

   Klaus P. Schneider

References Austin, John L. 1962. How to Do Things With Words. Oxford: Oxford University Press. Barron, Anne and Klaus P. Schneider 2009. “Variational pragmatics: studying the impact of social factors on language use in interaction”. Interlanguage Pragmatics 6/4, 425–442. Brinker, Klaus. 2010. Linguistische Textanalyse. 7th ed. Berlin: Erich Schmidt. Buchstaller, Isabelle and Ingrid van Alphen (eds). 2012. Quotatives. Cross-linguistic and Crossdisciplinary Perspectives. Amsterdam: John Benjamins. Chen, Rong. 1993. “Responding to compliments: a contrastive study of politeness strategies between American English and Chinese speakers”. Journal of Pragmatics 20, 49–75. Chen, Rong and Dafu Yang. 2010. “Responding to compliments in Chinese: has it changed?”. Journal of Pragmatics 42, 1951–1963. Clyne, Michael. 1987. “Cultural differences in the organization of academic texts”. Journal of Pragmatics 11, 211–247. Culpeper, Jonathan. 2011. Impoliteness. Using Language to Cause Offence. Cambridge: Cambridge University Press. Davies, Bronwyn and Rom Harré. 1990. “Positioning: the discursive production of selves”. Journal for the Theory of Social Behaviour 20, 43–63. De Fina, Anna. 2010. “The negotiation of identities”, in: Miriam A. Locher and Graham L. Sage (eds). Interpersonal Pragmatics. Handbooks of Pragmatics, vol. 6. Berlin: De Gruyter Mouton, 205–224. Galtung, Johan. 1981. “Structure, culture, and intellectual style”. Social Science Information 20, 817–856. Gnutzmann, Claus and Regina Lange. 1990. “Kontrastive Textsortenlinguistik und Fachsprachenanalyse”, in: Claus Gnutzmann (ed.). Kontrastive Linguistik. Frankfurt: Lang, 85–116. Gumperz, John and Jenny Cook-Gumperz. 2005. “Making space for bilingual communicative practice”. Intercultural Pragmatics 2/1, 1–23. Haugh, Michael. 2005. “The importance of ‘place’ in Japanese politeness: implications for cross-cultural and intercultural analyses”. Intercultural Pragmatics 2/1, 41–68. Kaplan, Robert B. 1966. “Cultural thought patterns in intercultural education”. Language Learning 16, 1–20. Kettemann, Bernhard and Georg Marko. 2011. “A critical analysis of American Christian fundamentalist discourse”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 449–455. Langlotz, Andreas. 2010. “Social cognition”, in: Miriam A. Locher and Graham L. Sage (eds). Interpersonal Pragmatics. Handbooks of Pragmatics, vol. 6. Berlin: De Gruyter Mouton, 167–202. Leech, Geoffrey. 1983. Principles of Pragmatics. London: Longman. Levinson, Stephen L. 1989. “A review of relevance”. Journal of Linguistics 25/2, 455–472. Locher, Miriam A. and Richard Watts. 2005. “Politeness theory and relational work”. Journal of Politeness Research 1, 9–33. MLA Handbook for Writers of Research Papers. 2009. 7th ed. New York: The Modern Language Association of America. NODE = Pearsall, Judy (ed.). 2001. The New Oxford Dictionary of English. Oxford: Oxford University Press.

Manufacturing Credibility: Academic Quoting Across Cultures   

   227

Oldenburg, Hermann. 1992. Angewandte Fachtextlinguistik: “Conclusions” und Zusammenfassungen. Tübingen: Narr. Rajagopalan, Kanavillil. 2005. “Linguistic pragmatics: a discipline bedeviled by its own history? Implications for intercultural studies”. Intercultural Pragmatics 2/1, 93–97. Ritter, Robert M. 2002. The Oxford Guide to Style. Oxford: Oxford University Press. Robbins, Dorothy. 2005. “Generalized holographic visions of language in Vygotsky, Luria, Pribram, Eisenstein, and Vološinov”. Intercultural Pragmatics 2/1, 25–39. Schneider, Klaus P. 2007. “Genre matters: textual and contextual constraints on contemporary English speech behaviour”. Anglia 125/1, 59–83. Schneider, Klaus P. 2010. “Variational pragmatics”, in: Mirjam Fried, Jan-Ola Östman and Jef Verschueren (eds). Variation and Change: Pragmatic Perspectives. Handbook of Pragmatics Highlights, vol. 6. Amsterdam: John Benjamins, 239–267. Schneider, Klaus P. 2012. “Appropriateness across varieties of English”. Journal of Pragmatics 44, 1022–1037. Schneider, Klaus P. 2013a. “Emerging e-mail etiquette: lay perceptions of appropriateness in electronic discourse”, in: Katrin Röder and Ilse Wischer (eds). Anglistentag Potsdam 2012. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 329–340. Schneider, Klaus P. 2013b. “The truth about diminutives, and how we can find it: some theoretical and methodological considerations”. SKASE Journal of Theoretical Linguistics 10/1, 137–151. Schneider, Klaus P. and Anne Barron. 2008. “Where pragmatics and dialectology meet: introducing variational pragmatics”, in: Klaus P. Schneider and Anne Barron (eds). Variational Pragmatics. Amsterdam: John Benjamins, 1–32. (Also in: Asa Kasher (ed.). 2011. Pragmatics II. Vol. 5. Routledge Major Works Series: Critical Concepts in Linguistics. London/New York: Routledge, 369–401.) Staverman, Werner H. 1953. “Diminutivitis Neerlandica”. De Gids 116/7, 407–418. Swales, John. 1990. Genre Analysis. Cambridge: Cambridge University Press. Tardy, Christine M. and John M. Swales. 2014. “Genre analysis”, in: Klaus P. Schneider and Anne Barron (eds). Pragmatics of Discourse. Handbooks of Pragmatics, vol. 3. Berlin: De Gruyter Mouton, 165–187. The Chicago Manual of Style 1993. 14th ed. Chicago/London: University of Chicago Press. Thompson, Geoff and Susan Hunston. 2000. “Evaluation: an introduction”, in: Susan Hunston and Geoff Thompson (eds). Evaluation in Text. Authorial Stance and the Construction of Discourse. Oxford: Oxford University Press, 1–27. Titscher, Stefan, Michael Meyer, Ruth Wodak and Eva Vetter. 2000. Methods of Text and Discourse Analysis. London: Sage. Williams, Jenny and Andrew Chesterman. 2002. The Map: A Beginner’s Guide to Doing Research in Translation Studies. Manchester: St. Jerome. Wilson, Deirdre and Dan Sperber. 2005. “Reply to Rajagopalan”. Intercultural Pragmatics 2/1, 99–103.

Part II: Quoting Then

Karin Aijmer

9 Quotative Markers in A Corpus of English Dialogues 1560–1760 Abstract: Innovative ways of quoting such as go and be like in some varieties of English have drawn attention to on-going changes in the domain of quotation (Buchstaller/van Alphen 2012). Studying quotation markers at an older stage of the language means focusing on devices that are no longer current in present-day English rather than on the emergence of new markers. The present article investigates the kinds of devices which are used to convey quotation in witness depositions in A Corpus of English Dialogues. It is shown that grammaticalization plays an important role for the development of quotative markers. Say as a speaking verb can for instance become semantically weakened and be used with new quotative functions. Quotatives can also be established from other sources than say. This is illustrated by deictic pointers such as as followeth, namelye, to this effect. Keywords: quotation, Early Modern English, witness deposition, grammaticalization, A Corpus of English Dialogues

1 Introduction Quoting what other people said seems to be a general feature of language. This is not only carried out by means of a reporting clause such as he said followed by an assertion. According to Güldemann (2008: 2), “even a cursory look at the available cross-linguistic data reveals that quotative indexes often do not have a sentential structure of the form X said to Y, suggesting that they are more than simple predicative assertions about a speech event.” Such observations have resulted in more interest in the ways in which quotative markers (quotative indexes) are realized and the preconditions for their further developments. Innovative ways of quoting such as go and be like in some varieties of English have drawn attention to on-going changes in the domain of quotation (Buchstaller/van Alphen 2012). Studying quotation markers at an older stage of the language, on the other hand, means focusing on devices that are no longer current in present-day English. In present-day English quotation is easy to recognize when quotation marks are used to frame what is quoted. In earlier English, such typographic conventions were less firmly established although quotation marks are usually added in the published editions of earlier work (cf. Arendholz/Kirner-Ludwig; Moore, this volume).

232   

   Karin Aijmer

The aim of the present article is to go back in time and study devices used to signal quotation in historical corpus data representing witness depositions in Early Modern English (see also Lutzky, this volume). The following research questions will be asked: is quotation always marked by a verb of speaking? What other techniques are used? In what ways are the methods used to convey quotation in earlier English different from those characteristic of present-day English? To what extent are quotative devices grammaticalized? This study will be carried out against the background of what is known about quoting in present-day English, but the results can also be compared with other studies of quoting in earlier English. The analysis is both quantitative and qualitative. The availability of corpus data makes it possible to study the distribution and frequencies of quotative markers as well as studying their use in context. The article is structured as follows. Section 2 presents the material and method. Section 3 provides an overview of the different devices with a quotative function which will be discussed. Section 4 deals with verbs of speaking, in particular with the occurrence of double quotatives. I also discuss the use of deictic pro-forms referring cataphorically to the quotation and the pragmatic functions taken on by said he (she) as an insert in the quoted utterance. Section 5 complements the picture of quotative markers with a discussion of quote-internal vocatives, interjections and pragmatic markers. Section 6 contains the conclusion.

2 Material and Method The present study is based on the witness depositions in A Corpus of English Dialogues 1560–1760 (CED) (Kytö/Walker 2006). The texts in the corpus are written, but have been selected because the depositions contain passages where the clerk used direct speech in order to render as precisely as possible the actual words used by the witness reporting the alleged offence. The texts in the corpus represent five periods each consisting of 40 years covering the period between 1560– 1760 (Table 1):

Quotative Markers in A Corpus of English Dialogues 1560–1760   

   233

Table 1: Witness depositions in five periods in the CED

Period

Number of words

1 1560–1599

 42,080

2 1600–1639

 39,930

3 1640–1679

 46,820

4 1680–1719

 26,500

5 1720–1760

 17,610

Total

172,949

Witness depositions were chosen for this study because they employ “reported discourse for purposes essential to the genre” (Moore 2011:  88).¹ Depositions “present the record of the oral testimony of a witness, defendant or plaintiff taken down by a scribe” (Culpeper/Kytö 2010: 54). They were subsequently read aloud in court. Among the crimes or offences involved in the depositions are, for example, slander, defamation, physical or verbal abuse, witchcraft and high treason (see further Culpeper/Kytö 2010: 58). According to Culpeper/Kytö (2010: 54), the witness depositions “contain a broader dialogue which takes place between the witness and the court officials”. They usually consist of answers to a question by a court official (or the magistrate). What makes the data interesting for the present study is that the depositions can also contain reports of direct speech that the witness heard (Culpeper/ Kytö 2010: 54). These passages have been selected for a study of quotation. An advantage with using this data is that direct speech was coded apart from the rest of the text (outside the [$…$] code which marks “running text other than direct speech”) (Kytö/Walker 2006: 37). We can therefore use the ‘$]’ code to collect examples of quotation in the data.² We have recently witnessed a change in the study of quotation from a more philosophical approach to the empirical analysis of the use of quotation in authentic data. Methodologically such approaches are embedded in a functional and pragmatic perspective on language. The focus is on the different ways in which an orientation to quotation can be expressed and how the strategies can be explained in a grammaticalization perspective.

1 Lutzky (2012: 52), however, states that “the focus was often not on a verbatim report but on confirming the criteria necessary for the legal charge of slander”. 2 For a description of the characters used for encoding in the CED, see Kytö/Walker (2006: 36– 37). Examples where $] was not (immediately) followed by direct speech were deleted manually.

234   

   Karin Aijmer

The articles in Buchstaller/Alphen (2012) discuss new quotatives and their semantic sources. Innovative quotative forms can for example be derived from lexical items that mean similarity (be like) or from generic verbs of motion (go). Approaches to the study of quotation have also been enriched by a typological perspective (e.g. Güldemann 2008). Cross-linguistic comparisons suggest that a range of typologically related and unrelated languages underwent similar developments. Recently we have also seen more interest in how quotative practices change over time. “The tangling and untangling of quotation” needs also to be examined for earlier periods of language (Moore 2011: 2). Moore (ibid.), for example, examined the pragmatic methods of reporting speech in late medieval manuscripts and texts (1245–1645).

3 Lexical Markers of Direct Quotation According to Lucy (1993: 92), the direct quotation “maximizes the integrity of the original and conveys the secondary message that the reporter is ‘merely’ reporting the form and is not personally interpreting or predicating anything about the content of the reported speech”.³ In the witness deposition, the original message refers to the context in which the speech offence has taken place. In the secondary message, the original text (the words uttered by the presumed offender) has been transposed to a text giving a verbatim representation of what was said (the witness’ testimony). The following example of a slander deposition from A Corpus of English Dialogues 1560–1760 (http://www.engelska.uu.se/Research/English_ Language/Research_Areas/Electronic_Resource_Projects/A_Corpus_of_English_ Dialogues/) illustrates some methods used to mark the reported speech as quoted: (Raffe (Raph Wilson) has been charged with defaming Isabell by implying that her goods have not been honestly acquired) (1)

and by comme one Janett Percivall, a young woman, and said to the aforesaid Isabell,$] “Yonder goith your good man,” [$meaninge the said Raffe. And Isabell annswered,$] “Our good is to well wonne to comme emongest thers.” [$Which wordes the said Raiff hard and said,$] “Thou giglott, thou knowest best whither your goods be well woon or noo.” [$And the said Isabell annswerde and said,$] “Aither you or some of yours so reports.”

3 Ideally direct speech does not include any interference from the ’reporter’. The witness depositions in the CED, however, can contain interventions from the scribe (Culpeper/Kytö 2010: 54).

Quotative Markers in A Corpus of English Dialogues 1560–1760   

   235

[$And then spoke the said Raiffe and said,$] “I will thou know that yf I had stolne 2 yawdes, as thi father dyd, I wold nott have thought that good well woon.”⁴ [and one Janet Percivall comes by, a young woman and said to the aforementioned Isabell “There goes your good man,” meaning the afore-mentioned Raffe. And Isabell answered “Our possessions have not been won to come among theirs.” Which words the afore-mentioned Raiff heard and said, “You wanton woman, you know best whether your goods are well won or not.” And the afore-mentioned Isabell answered and said, “Either you or some of yours report this.” And then spoke the afore-mentioned Raffe and said, “I want you to know if I had stolen 2 yards, as your father did, I would not have considered that goods well won.]” (D1WDURHA Courts of Durham 1560–1588)⁵ The passages in direct speech can be identified by their being enclosed by quotation marks (added in the edited text), but they are also marked in other ways. The verbal markers have in common that “they signal strongly and clearly that the quote is in fact, a quote” (Güldemann 2008: 134). There are a large number of such markers. Verbs of speaking are dedicated to the function of marking direct speech. They are external to the direct speech itself, unlike features inside the quoted speech referring deictically to the original speech situation (cf. Moore 2011: 44; see also Lutzky, this volume). Deictic ‘shifters’ refer to “the hic-et-nunc’ of the SP [speaker] and AD [addressee] in the non-immediate speech environment” (Güldemann 2008: 8). They are exemplified by first and second person pronouns referring to the speaker and the hearer (I, you, thou) and other deictic forms (yonder, thither, this, that). Deictic elements can also have the function of referring forwards to the quotation (as followeth, thus, namelye, etc). Verbs of speaking can also be the source of new quotative structures. Say can be used with a weakened meaning in a bipartite construction with another speaking verb (answered and said). Moreover it occurs as an ‘insert’ in the middle of the quoted sentence (said he, quoth he, saying) with a discourse marking or clarifying function. Interjections (oh, ah) and expletives (e.g. by Godes blode) can also have an additional quotative function in the beginning of the quoted clause (cf. thou giglott in example 1).

4 The formatting of the examples is the same as in the manuscript. 5 The year refers to the speech event and not to the year of publication of the edited volume. Full bibliographical information about the source texts and the editions they are taken from is given in Kytö/Walker (2006).

236   

   Karin Aijmer

4 Verbs of Speaking as Quotative Markers Verbs of speaking are the most common way of reporting what another person said. In the witness depositions, a verb of speaking seems to be the rule although quotation can be marked in additional ways. Table 2 shows the distribution of the most frequent speaking verbs in five different periods in the CED. In cases of ‘double quotatives’ (answered and said) as an alternative to answered, only the first verb has been counted. Table 2: Most frequent speaking verbs in witness depositions in five 40-year periods in the CED. Frequencies have been normalized to 100,000 words (within parentheses).

Witness depositions

DW1 1560–1599

DW2 1600–1639

DW3 1640–1679

DW4 1680–1719

DW5 1720–1760

Total

say

175 (416)

65 (163)

88 (187)

68 (257)

44 (249)

440

answer*

 74 (176)

17 (43)

15 (32)

14 (53)

 8 (45)

128

reply

  2 (5)

23 (58)

12 (26)

 9 (34)

 8 (45)

 54

quoth

 19 (45)

34 (85)







 53

speak

  6 (14)

18 (45)





 1 (6)

 25

ask

  2 (5)





 3 (11)



  5

other verbs

 17 (40)

 6 (15)

16 (34)

19 (72)

 4 (23)

 62

Size of corpus  42,080

39,930

46,820

26,500

17,610

172,940

* Including answered in this sorte, his answer was, she made answer

The frequent verbs are comparatively few. Say was by far the most frequent speaking verb followed by answer. See Table 3 showing the relative frequencies of the speaking verbs. Quoth was not found with direct speech after 1639. In comparison with present-day English there are few verbs of quotation. Only ask, cry, cry out occurred more than once. It can be seen from Appendix 1 that speak was only found with a speech noun pointing forwards to the direct quotation (e.g. spake the following words). Other examples of verbs followed by a speech noun are put these words to sb, give the above speeches as, say these words, use speaches as, repeat these words (see also Section 4.3). In the CED data there were no examples without a verb of speaking (‘zero-quotatives’). This may be due to the text type. Moore (2011: 46) for example noted several examples in Troilus and Criseyde where the onset of quoted speech had not been explicitly marked by a verb of speaking (however

Quotative Markers in A Corpus of English Dialogues 1560–1760   

   237

Table 3: The relative frequencies of the speaking verbs in witness depositions in five 40-year periods in the CED.

Witness depositions

DW1 1560–1599

DW2 1600–1639

DW3 1640–1679

DW4 1680–1719

DW5 1720–1760

say

175 (59.3 %)

 65 (39.9 %)

 88 (67.2 %)

 68 (60.2 %)

44 (67.7 %)

answer

 74 (25.1 %)

 17 (10.4 %)

 15 (11.5 %)

 14 (12.4 %)

 8 (12.3 %)

reply

  2 (0.7 %)

 23 (14.1 %)

 12 (9.2 %)

  9 (8.0 %)

 8 (12.3 %)

quoth

 19 (6.4 %)

 34 (20.9 %)







speak

  6 (2.0 %)

 18 (11.0 %)





 1 (1.5 %)

ask

  2 (0.7 %)





  3 (2.7 %)



other verbs

 17 (5.8 %)

  6 (3.7 %)

 16 (12.2 %)

 19 (16.8 %)

 4 (6.2 %)

Total number of verbs

295 (100 %)

163 (100 %)

131 (100.1 %) 113 (100.1 %) 65 (100 %)

92 % of the total number of passages in direct speech contained a verb of speaking). On the other hand, there were “no cases in the text of speech onsets with no markers at all” (Moore 2011: 46).

4.1 The Use of and said as a Lexical Marker of Quotation Say could also be co-ordinated with another verb. This is illustrated in (2) where we find a division of labour between answered with a specific speech act meaning and the semantically weakened and said: (2)

[$About Whitsondaye last this examinate, being at his work at a coole pitt in Dunston, nigh Whickham, did see the said Janet comeing towards him; whereupon she, beinge nigh him, he called hir to him, and asked hir where she had bene. She answered and said,$] “I would God that eyther you or any honest man had been with me, where I have bene.” [$This examinate answered,$] “Why, where have you bene?” [about last Whit Sunday this examinate being at his work in a coal pit in Dunston, near Whickham, saw the afore-mentioned Janet coming towards him; whereupon, she being near him, he summoned her to him and asked her where she had been. She answered and said, “I wish to God that either you or any honest man had been with me, where I have been.” The exami-

238   

   Karin Aijmer

nate answered, “Why, where have you been ?”] (D1WDURHA Courts of Durham 1560–1588)

The speaker has a choice between the simple answered and a conjoined structure (answered and said). Table 4 shows the distribution of answer as a simple verb and with a form of say in the CED: Table 4: Answer as a simple verb or together with ‘and said’ in five periods in the CED⁶

answeren* (aunswer, answer)

simple verb

occurring with say

Total

128

26

15

*Including make an answer (6 examples in the simple form; 3 examples of made answere and said).

Out of 154 examples of answer, 16.9 % occurred with a form of say. And said is used in the second part of the construction. In the first part we find more variation (see Table 5). There were 18 different verbs of speaking, the majority of which occurred only once. The table also shows that the construction with and said occurs only sporadically after 1639. Double quotatives in earlier English have been studied by Herlyn (1999) and by Moore (2011). Herlyn refers to them as ‘multiple dialogue introducers’ and compares their use in Middle English romances with their use in present-day English. Moore (2011) used a large corpus of late medieval texts to study verbs of speaking and combinations such as answered and said. In Moore’s medieval English data, 46 % of the usages of the verb answeren (usually in the form answered) were used together with seien (usually said). The texts in my data contained fewer examples of double quotatives, which makes the usage of verbs of speaking more similar to the situation in present-day English. This phenomenon has also been observed cross-linguistically. The data presented by Güldemann (2008: 119) show for example that the phenomenon (‘double quotation index’) is frequent in African languages which suggests that the situation in older English “could well represent a variation on a universal theme”. Answered and said and similar constructions can be described as a frame where the first part can be filled by a descriptive verb and the second part has been routinized: [VP + and said]. The weakened meaning of and said in the bipar6 Twenty-six of the examples come from the first period (1560–1599). Doublets of this kind have also been noted in Old English and Middle English (see Moore 2011: 58 and the references there).

Quotative Markers in A Corpus of English Dialogues 1560–1760   

   239

Table 5: The first part in double quotatives in five forty-year periods in the CED

Witness depositions

DW1 DW2 DW3 DW4 DW5 Total 1560–1599 1600–1639 1640–1679 1680–1719 1720–1760

answer*

22

1



3



26

cry out

 3





1



 4

call (sb)

 1



3





 4

reply

 2

1







 3

speak

 1

2







 3

confess**

 1

2







 3

ask

 1









 1

revile and curse

 –

1







 1

pray

 1









 1

call on sb

 –





1



 1

laugh

 –





1



 1

reprove

 –

1







 1

fall into bitter and  – passionate words

1







 1

curse

 –





1



 1

swear and curse

 –





1



 1

depose***

 –

1







 1

hard and said thes wordes

 1









 1

beganne (and spoake)

 –

1







 –

* one example had the form aunswerith and saith ** the examples had the form confesseth and saith *** the example had the form deposeth and saith

tite structure can be explained as the result of grammaticalization. And said can be shown to be highly routinized: – say cannot be replaced by another verb of speaking – and said has lost its ideational meaning and marks the transition to the quoted speech

240   

– –

   Karin Aijmer

and said is optional (answered and said is in variation with the simple answered) and said is not normally used with an object and nothing is inserted between answered and said

As a result of weakening and grammaticalization, and said acquires a quotative function. According to Moore (2011: 60), “the frequent co-occurrence of both verbs in Middle English indicates just how far seien had shifted in the direction of a grammatical marker”. Frajzyngier (1996: 156) suggests that “the X verb ‘say’ is simply used by the speakers as the marker of the de dicto domain, including the hypothetical and other semantic extensions of the de dicto domain…” (quoted from Güldemann 2008: 133). In other words, the speaker uses and said in order to mark a piece of discourse as a representation of the words as they were spoken.⁷ The clearest examples of the grammaticalization of and said are those where the double quotative has a corresponding simple verb with the same meaning (answer, reply, ask, cry, cry out, call on, swear). Cry out was for instance also found both with and without and said (3 examples cried out and said; 20 examples cried out; in addition 2 examples contained cried out saying). In (3) and said is tautological: (3)

time his brother cryed out and said,$] Father, Father, come helpe mee, there is a blacke thing y=t= hath me by y=e= legge, as big as my sister: [his brother cried out and said, Father, come and help me, there is a black thing that has me by the leg, as big as my sister] (D3WESSEX Witches… in the County of Essex 1645)

Less grammaticalized examples are also of interest. When cryed out was further qualified by a manner adverbial, it was followed by and said: (4)

immediately after the said Lords and this Deponent were in the Room; the Queen cryed out extremely, and said,$] (^Oh, I die; you kill me, you kill me^) [immediately afterwards the afore-mentioned lords and this deponent were in the room; the Queen cried out to an extreme degree and said, (Oh, I die; you kill me, you kill me] (D4WWALES Concerning the birth of the Prince of Wales 1688)

7 On the distinction between a ‘de re’ and a ‘de dicto’ reading of reported speech, see e.g. Moore (2011: 84).

Quotative Markers in A Corpus of English Dialogues 1560–1760   

   241

Similarly, in (5) the descriptive verb is modified by a manner adverbial (spoke allowde unto examinate in the time of the administracion). And said has the function of pointing forward to the quotation: (5)

[$Mr.Richardson spoke allowde in the time of the said administracion unto examinate, and said$] “I thinke yow will have done before me.” [Mr.Richardson spoke aloud at the time of the afore-mentioned administration [of oaths] to the examinate and said “I think you will be done before me.”] (D2WDIOCE High Commission Court…Diocese of Durham 1627–1637)

In (6), the descriptive part consists of coordinated verbs (did revile and curse). And said makes it possible to use the descriptive verbs before a quotation: (6)

[$Aboute August or September last Isabell, in the townegaite of Heddonn and neere unto the doore of Reede, did revile, and curse, and said unto him,$] “Godes plague and Godes curse light of the and thine beastes, and God lett never they nor anie thing thou hast prosper nor doe well.” [about last August or September Isabell at the town gate of Heddon and near Reed’s door reviled and cursed and said to him, “God’s plague and God’s curse light on you and your beasts and may God not allow that they or any other thing you have ever prosper or do well.”] (D1WDURHA Courts of Durham 1560–1588)

Reproved and said is used to characterize the speaker as ‘reproving’: (7)

Soone after they came into the milne, Carrock acquainted her that she did utter the wordes articulate. Reproved Isabell, and saide$] “Fie upon yow, Isabell, whie doe yow speake anie such Wordes?” [Soon after they came into the mill, Carrock told her that she did utter the words. [He] reproved Isabell, and said “Fie upon you, Isabell, why do you speak any such words?”] (D2WDIOCE High Commission Court…Diocese of Durham 1627–1637)

The quoted speech is additionally marked as a reproof by the initial exclamation and vocative (Fie upon yow, Isabell):

242   

   Karin Aijmer

The descriptive part of the bipartite structure can be quite complex and need not be adjacent to the speaking verb. In (8), the first part describes the speaker’s feelings (did fall into bitter and passionate words) to introduce the quoted speech. And said makes it possible to point forwards to the direct quotation. (8)

whereupon Carr did fall into bitter and passionate wordes against Mr. Macklewyan and amongst others did tearme him “Thiefe, carle, and Gallowaie knave,” and said$] “Sirra, if yow and I live I will remember yow,” [whereupon Carr broke out in bitter and passionate words against Mr. Macklewyan and among other things called him “Thief, churl, and Gallowaie knave” and said “Sir, if you and I live I will remember you,”] (D2WDIOCE High Commission Court…Diocese of Durham 1627–1637)

Most of the verbs in the first part are speaking verbs. However, and said is also found after sound verbs (Güldemann 2008: 133) such as laugh: (9)

you, I will have the moneys I have won,” [$and offred to strike him but did not. Whereupon Maddox laughed and said,$] “Your Lordshipp may make me do anythinge,” [you, I will have the money I have won,” and prepared to strike him but did not. Whereupon Maddox laughed and said, “Your Lordship may make me do anything,”] (D3WYORK Depositions from the Castle of York 3 1655–1664)

(10) illustrates the use of inchoative verbs as a part of the quotative structure. This is the only example where and spoke (spoake) is found instead of a form of say: (10)

[$Hodgshon beganne and spoake unto examinate as followeth,$] “Did yow observe Doctor Cosins his gesture in time of Divine service?” [$Examinate answeared he did a litle. Hodgshon then againe said$] “Doe yow know how they catch apes?” [Hodgshon began and spoke to the examinate as follows, “Did you observe Doctor Cosins’ deportment at the time of the divine service?” The examinate answered he did a little. Hodgshon then again said “Do you know how they catch apes?”] (D2WDIOCE High Commission Court…Diocese of Durham 1627–1637)

Quotative Markers in A Corpus of English Dialogues 1560–1760   

   243

In the next example started up and said (to him) introduces a direct quotation: (11)

Sir (^Walter^) thereupon furiously started up, and said to him,$] Master Atturney, you must not thinke that all that maketh for me is policy, and all that maketh against me is plain, and God revealeth it. I were well fitted for justice, if you should come to be my Judg, what indifferency is ther [Sir Walter thereupon furiously started up, and said to him, Master Attorney, you must not think that all that operates in favour of me is trick, and all that speaks against me is clear, and God reveals it. If I were well fitted for justice, if you should come to be my judge, what freedom of choice is there] (D2WRALEI Arraignment of Sir Walter Raleigh 1603)

Previous work (Herlyn 1999, Moore 2011) has only discussed double quotatives with and said. Answered saying represents a different type of the bipartite structure where saying refers to the upcoming quotation. (12) illustrates the type: (12)

and the vicar William Mellerbye came to the said William Kirkus house; at which tyme the said vicar asked the said Kirkus of his welfair, who aunswered saing,$] “I thank God, never worse, for I am trobled in all partes of my body;” [$to whom the vicar said then,$] “Will ye make a will?” [$and he, the said testator, aunswered and said,$] “Ye, I will make a will.” “And what will ye gyve to the poore man box?” [$qooth the vicar; and he, the said Kirkus, aunswered and said,$] “I gyve dayly to the poore, as other neighbours doith; and therefore I will nothing to the poore man box.” [and the vicar William Mellerbye came to the afore-mentioned William Kirkus’ house; at which time the afore-mentioned vicar asked the aforementioned Kirkus about his welfare, who answered saying, “I thank God, never worse, for I am troubled in all parts of my body: to whom the vicar said then, “Will you make a will.” “And what will you give to the poor man’s box?” quoth the vicar; and he, the afore-mentioned Kirkus, answered and said,” I give daily to the poor, as other neighbours do; and therefore I will give nothing to the poor man’s box.”] (D1WDURHA Courts of Durham 1560–1588)

Saying (like and said) functions as a grammatical marker of direct quotation together with a specific speaking verb. Other examples are murmured… saying,

244   

   Karin Aijmer

denyed… saying, cried saying, cried out saying, told … saying, spake… saying, interrupted… saying, where saying is a non-finite clause.

4.2 Deictic Pointers Verbs of speaking are the most common way of reporting what another person said. However speakers have additional ways of marking a forthcoming quotation. They can for instance use “quote pro-forms like speech nouns, pronouns, deictics, and adverbs” (Güldemann 2008: 187). The use of a speech noun is illustrated in (13): (13)

Before Robert Shaftoe, Esq. and Mark Milbank, Esq. (^Sarah, wife of Oswald Walker, yeo.^), did say on the 13th of July these wordes (to witt):$] “There was never a King in England that was a chimney sweaper but this,” [Before Robert Shaftoe, Esq. and Mark Milbank, Esq, Sarah, wife of Oscar Walker, yeoman uttered these words namely “There was never a King in England that was a chimney sweep but this one,”] (D3W Depositions from the Castle of York 3 1655–1664)

Say can take an object (words) as a pro-form deictically pointing forward to the quotation. Other frequent lexical markers occurring at a boundary before the quotation are adverbials (typically with a manner or similarity meaning), such as to this effect, in this manner, after this sort(e), namelye, to witt, as followeth, thus, videlicet. The pro-forms can be exploited as sources for grammaticalization. Moore (2002, 2011) has for instance discussed the grammaticalization of videlicet in the context of code-switching from Latin into English in a sample of witness depositions dating between 1245–1645. According to Moore (2011: 65), videlicet lost its concrete meaning and became a lexical marker with the function of introducing reported speech: “The increased use of videlicet in the sixteenth century, and the increased correlation of Latin videlicet with code-switching suggest that videlicet has acquired a grammaticalized function in these records as a conventional marker to introduce reported speech”. Videlicet was also found in my data as a quotation marker (before direct speech):

Quotative Markers in A Corpus of English Dialogues 1560–1760   

(14)

   245

to whom the said Maior answerid thus; (\videlicet\) / $] “Roger, it is best to take deliberacion in the # matter, and knowe what frendes the said Ellin hath, and what they will do for her” [to whom the afore-mentioned Mayor answered this; namely “Roger, it is best to deliberate on this matter and learn what friends the afore-mentioned Ellin has, and what they will do for her”] (D1WCHEST Bishop’s Court, Chester 1561–1566)

In the CED witness depositions, videlicet was also used in texts which were fully in English which suggests that it has ‘a quotative sense’ also without other evidence for its grammaticalization such as code-switching into English. Videlicet is similar to namelye (or to witt) which has also developed into a quotative marker: (15)

setling his fiery eyes vpon her eyes, spake and pronounced vnto her these woords following, namelye:$] Joan Prentice giue me thy soule, [settling his fiery eyes upon her eyes, spoke and pronounced unto her the following words, namely Joan Prentice give me your soul] (D1WNORWI Affray at Norwich 1583)

4.3 Quotative Inserts (said he) Verbs of speaking can be referred to as quotative ‘inserts’ when they occur in the middle of direct speech. This is exemplified by said he in (16): (16)

Vvho ansvvered nothing to those speeches vvhich vvere personall, having said before, that they vvere used onely to bring him into detestation of the vvorld, but spake to this effect.$] Novv it shall appeare that my Lord (^Cobham^) , is an unworthy, base, silly, simple poore soule. [$Master Atturney said,$] Is my Lord so poore? Yea, [$ (said he) $] in Spirit. Would to God you were so [$ (quoth Master Atturney:) Sir (^Walter^) proceeded;$] I will tell you the troth. [who answered nothing to those speeches which were personal, having said before that they were used only to make him detested by the world, but spoke to this effect. Now it shall appear that my Lord Cobham, is an

246   

   Karin Aijmer

unworthy, base, silly, simple poor soul. Master Attorney said, Is my Lord so poor? Yeah, said he, in spirit. I wish to God you were so (said Master Attorney: Sir Walter proceeded I will tell you the truth.)] (D2WSOUTH Earles of Essex and Southampton 1600) Said he can be regarded as a (fairly) fixed phrase. It does not, for example, occur with an indirect object and the subject is usually a pronoun (he or she). The inserted form occurs shortly after the beginning of an utterance (e.g. after yea), usually as the second element in the clause. Only a restricted number of speaking verbs were found as inserts. In addition to say, the only other verb was quod. Quod (quoth),⁸ unlike say, cannot occur outside the context of direct speech. Both quod and say are illustrated in example (17) in their use as quotative markers. (17)

Baker, ‘saies, that in certen talk apon a tyme had betwixe William Ball Articulate, and this deponent, Custance Frost came vnto them; and after salutacions done, she said vnto the said William:$] # “ar you one of them that bringe word from the Counsell that I have two husbandes?” [$he said$] “yea, mary, that I am.” “then,” [$quod # she,$] “will you stand to it?” “yea, mary,” [$said [{William Ball{] ,$] “that, I # will stand it, that I hard it.” “then,” [$quod she,$] “you may better bringe them furth that spake.” “nay,” [$quod William Ball,$] “I will not bring # them furth: bring you them furth, & you will have them.” [$quod she,$] “perchaunce I shall make you bring them furth, or els the lawe shall # faile” [Baker says that in certain talk between William Ball and this deponent, Custance Frost came up to them; and after the greetings had been performed she said to the afore-mentioned William: “are you one of those that bring words from the Council that I have two husbands? He said “yeah marry I am one of them.” “then” she said, “will you stand by it?” ” “yeah, marry,” said William Bull, “that I will stand by, that I heard it. “then” said she, you had better bring them forth that spoke.” Nay.” said William Ball, I will not bring them forth: you bring them forth and you will have them.” Said she, “perhaps I shall make you bring them forth or else the law will fail”] (D1WCHEST Bishop’s Court, Chester 1561–1566)

8 It was also found in my data in abbreviated form (q=th=).

   247

Quotative Markers in A Corpus of English Dialogues 1560–1760   

One hundred and nineteen examples with say or quoth were used as inserts with a pragmatic boundary function. As shown in Table 6, they were frequent above all in the two earliest periods in my material (1560–1639): Table 6: Say and quoth as pragmatic ‘inserts’ in five 40-year periods in the CED. Figures have been normalized to 100,000 words (within parentheses)

Witness depositions

D1W D2W D 3W D4W D5W Total 1560–1599 1600–1639 1640–1679 1680–1719 1720–1760

pragmatic insert

59 (140)

55 (138)

 5 (11)





119

Size of corpus

42,080

39,930

46,820

26,500

17,610

172, 949

‘Inserts’ are particularly frequent as strategies in long dialogic sequences characterized by speaker change. They are typical of direct speech and serve to identify the speaker. The finite verb can occupy the second position of the clause, both with a nominal and pronominal subject. However, verb-second in general is optional and rare with pronominal subjects. There were only two examples in the period between 1560–1599 and six examples in the period between 1600–1640. Direct speech is not marked by the inserts alone but also by pragmatic markers such as yea mary, nay as shown in example (17). In (18), the speaker’s identity is already known from the text where the passages of direct speech are inserted and need not be explicit. However, an important function of the insert is to re-identify a previously mentioned speaker. Mr Bacon spake to this effect describes the speaking event. Quoth he was added in the quoted utterance with a clarifying function: (18)

Then M=r= Bacon spake to this effect:$] I expected not [$ (q=th= he) $] that the matter of defence shold have # bene [Then Mr Bacon spoke to this effect: I expected not said he that the matter of defence should have been] (D2W Earles of Essex and Southampton 1600)

In (19), quoth (without a subject) is used as a quotative marker with the function to reassert the speaker’s identity: (19)

[$He answeared,$] “I care not what I spoke, soe as I might but speake ill enough to shew my malice against him [$ (meaneing Mr. Raph Blaikston) ;$] but [$ (quoth) $] good Sir, lett this matter goe noe further and I will never speake the like againe, and give yow any content yow will have for your good name.”

248   

   Karin Aijmer

[He answered, “I care not what I spoke as long as I can speak unfavourably enough to inflict any injury on him (meaning M. Raph Blaikson); but said he, good Sir, don’t let this matter go any further and I will never speak the same again, and I will give you any satisfaction you want to have for your good name”]. (D3WESSEX Witches … in the County of Essex 1645) Another type of quotative insert in witness depositions is saying. Saying (like said he) marks a break in the quoted speech after which the same speaker continues: (20)

[$and this examinat said vnto her,$] I am glad you are here you vield strumpet, [$saying,$] I do think you haue bewitched my wife, and as truly as God doth liue, if I can perceiue y=t= she be troubled any more as she hath been, I will not leaue a whole bone about thee, & besides I will seeke to haue thee hanged: [and this examinate said to her, I am glad you are here you vile harlot, saying I do think you have bewitched my wife, and as truly as God lives, if I can perceive that she is troubled any more than she has been I will not leave you with a whole bone in your body and besides I will seek to have you hanged] (D3WESSEX Witches … in the County of Essex 1645)

In (21), saying re-identifies the speaker as ‘the said Glouer’: (21)

and this Examinate then said before the said Glouer,$] I would to God that I could drinke, [$where upon the said Glouer said to this Examinate,$] take that drinke, and in the name of the (^Father^) , the (^Sonne^) , and the (^Holy Ghost^) , drinke it, [$saying;$] The Deuill and Witches are not able to preuaile against GOD and his Word, [and this Examinate then said before the afore-mentioned Glouer, I would to God that I could drink whereupon the afore-mentioned Glouer said to this Examinate, take that drink, and in the name of the Father, the Son and the Holy Ghost, drink it, saying , the devil and witches are not able to prevail against God and his word] (D1WBARKS Witches… in the Countie of Barks 1579)

The common feature of inserts with say and quoth is that they have been routinized and occur in a fairly fixed form. They have a pronoun subject (or no subject at

Quotative Markers in A Corpus of English Dialogues 1560–1760   

   249

all) and they are restricted to the past tense (or less frequently the present tense). Their usage is closely associated with the text type. The frequent use of the inserts with a verb of saying in witness depositions can be explained by the need to identify or re-identify the speaker of the direct quotation. Inserts are also common in present-day English and have correspondences in other languages. They have a parallel in French where the form il dit [idi:] can be used as a generalized marker of direct speech. Haßler (2002: 151) describes its function as follows, “For the recipient of the message the function of [idi:] is reduced to a disclaimer not to consider the producer of the utterance as responsible for the truth of its contents.”

5 Pragmatic Markers with a Quotative Function Pragmatic markers are lexical elements characteristic of spoken language which can co-occur with quotative verbs (see also Lutzky, this volume). However, unlike verbs of speaking or deictic pro-forms, they are internal to the quoted speech. Pragmatic markers can express the speaker’s attitudes and emotions. They may therefore be important in representing what was said as precisely as possible in the witness depositions. Table 7 indicates the relative frequencies of lexical features such as vocatives, interjections and exclamations and other pragmatic markers. Table 7: Relative frequencies of vocatives, interjections and pragmatic markers in five 40-year periods in the CED.

Witness depositions DW1 1560– 1599

DW2 1600– 1639

DW3 1640– 1679

DW4 1680– 1719

DW5 1720– 1760

Total

initial vocatives

33 (11.2 %)

21 (12.9 %)

14 (10.5 %)

14 (12.2 %)

6 (9.2 %)

 88

initial interjections and exclamations

22 (7.5 %)

7 (4.3 %)

1 (0.7 %)

6 (5.2 %)

3 (4.6 %)

 39

Initial (other) pragmatic markers

30 (10.2 %)

18 (11.0 %)

8 (6.0 %)

7 (6.1 %)

3 (4.6 %)

 66

number of examples 210 without a pragmatic (71.2 %) marker

117 (71.8 %)

111 (82.8 %)

88 (76.5 %)

53 (81.5 %)

579

total number of verbs of speaking

163 (100 %)

134 (100 %)

115 (100 %)

65 (99.9 %)

772

295 (100.1 %)

250   

   Karin Aijmer

Vocatives were the most frequent type. They are not obligatory but do contribute in marking the text as an example of a quotation of direct speech. Vocatives are, for example, personal names, family names (good husband, wife, father, daughter, sister). A small number of vocatives such as Lord, Lady, sir can be associated with politeness and deference in addition to the quotative function. Evaluative exclamations have been distinguished from vocatives in that their chief function is to express a strong feeling about the person referred to (Leech 1999: 109). These can consist of combinations with thou or ye (thou vilan, thou drouken villan, thou giglott, noghty hore, thou droucken horemonger priest, ye stinking hore). Typical interjections are, for example, alas, beware, tush, in faith, fie, marry as well as what Culpeper/Kytö (2010: 199) refer to as ‘pragmatic noise’ (o(h), ah, o ho). The witness depositions also contained some swearing (classified as interjections) (commother, for God’s sake, by Gode’s harte, by Gode’s blod, by the holy contents of the Bible, God’s my life, in the name of God). A distinction has been made between interjections and other pragmatic markers such as no, nay, yes (yeah, ye, yees, yeis), well, why, what, now, so, and I pray. Example (22) illustrates the co-occurrence of well with a quotative verb: (22)

well [$sayde the quenes attourney$] well, can you make it come before vs nowe, if ye can we will [well said the queen’s attorney, well can you make it come before us now, if you can we will] (D1WCHENS Wytches at Chensforde I566)

The data also contained pragmatic markers which are no longer found in present day English. In example (23), nay marry (mayry) as an answer-preface marks the beginning of the reporting of direct speech. The pragmatic marker has the additional function to evoke the dramatic quality of the verbal exchange in the original speech situation:

Quotative Markers in A Corpus of English Dialogues 1560–1760   

(23)

   251

[$And then Bullman’s wyff said$] “What, noughty hoore, caull thou me goose steiler?” “Nay, mayry, I know thee for no such” [$saith Stillinge wyffe,$] “but I thank you for your good reporte, whills you and I talk further.” [And then Bullman’s wife said “what, naughty whore, do you call me goose stealer?” “No, marry, I don’t know you for such” says Stilling’s wife, but I thank you for your good report, while you and I talk further.”] (D1WDURHA Courts of Durham 1560–1588)

Pragmatic markers along with verbs of speaking are fairly peripheral as markers of quotation in witness depositions. Culpeper/Kytö (2010: 267) remark “that the figures [instances of ‘pragmatic noise’ K.A] obtained for Witness Depositions and Trials are considerably lower than those obtained for the text types containing constructed ‘spoken’ language.” This was confirmed by Lutzky (2012) who conducted a detailed study of marry, well and why across text types in the CED (Lutzky 2012: 96).

6 Conclusion This study has shown that witness depositions do not rely only on one type of strategy to mark direct quotation but can use many different devices which, like quotation marks, have scope over the quoted utterance. Studying the conventions for marking quotation in earlier periods of language draws attention to the role of grammaticalization and the semantic sources of the quotative devices. Say as a speaking verb can, for instance, become semantically weakened as a result of grammaticalization and be used with new functions. This is illustrated by combinations such as answered and said where say functions as a quotative device. The inserts quoth (he) and said he can also be associated with grammaticalization. They have a general or neutral meaning and a ‘new’ quotative function of re-identifying the speaker. Quotatives can also be established from other sources than the generic verb say. This is illustrated by deictic pointers such as as followeth, namelye, and to this effect, which are used to draw the hearer’s attention to the direct quotation. In my data from witness depositions, there were no examples where the quotation was not marked at all. On the other hand, several different devices can be used together to draw attention to the status of an utterance as a direct quotation. The list of lexical ‘quote-internal’ markers includes interjections, vocatives and

252   

   Karin Aijmer

pragmatic markers at the beginning of a quoted utterance which can co-occur with a quotative verb of speaking. Recently, the focus in research on quotation has been on innovative forms of quotation ‘world-wide’. As a result, we now know much about the formal properties, functions and social distribution of newcomers such as be like. Studying quotative markers in the past can also contribute to what we know about the functions of quotation markers and their semantic or lexical sources. By studying the different ways in which quoting is expressed from a different perspective, we can also come closer to answering the ‘big question’ (Buchstaller/van Alphen 2012: xiii): “What do speakers do when they report?”. Appendix 1. The verbs of speaking in the witness depositions in five forty-year periods in the CED (including singletons).

DW1

DW2

DW3

DW4

DW5

say

175

65

88

82

44

answer

 74

17

15

14

 8

reply

  2

23

12

 9

 8

quoth

 19

34





speak

  6

18





 1

ask

  2





 3



cry

  4



 6

 2

 1

cry out

  5



 8

 6

 1

call out

  1





 1



bid

  1









swear

  1









put these words to someone

  1









report as

  1









give the above speeches as

  1









say these words

  1









use speaches as

  1









repeat (these words)

  1







 1

call on someone and said







 1



conclude with these words







 1



tell







 1



   253

Quotative Markers in A Corpus of English Dialogues 1560–1760   

DW1

DW2

DW3

DW4

DW5

whisper sb in the ear







 1



say in answer







 2



NPs answer was







 1



say with several oaths







 1



conclude with these words







 1



adding these further words vuz.







 1



say these words to witt











contradict with words to this purpose





 1





proceed to this effect



 1







demand of somebody



 1







begin to this effect



 1







curse as followeth



 1







enter into a speech to this effect



 1







protest judgement against someone in this manner



 1







continue









 1

References Buchstaller, Isabelle and Ingrid van Alphen. 2012. Quotatives: Cross-linguistic and Crossdisciplinary Perspectives. Amsterdam: John Benjamins. Culpeper, Jonathan and Merja Kytö. 2010. Early Modern English Dialogues. Spoken Interaction as Writing. Cambridge: Cambridge University Press. Frajzyngier, Zygmunt. 1996. Grammaticalization of the Complex Sentence: A Case Study in Chadic. Amsterdam: John Benjamins. Güldemann, Tom. 2008. Quotative Indexes in African Languages: A Synchronic and Diachronic Survey. Berlin and New York: Mouton de Gruyter. Haßler, Gerda. 2002. “Evidentiality and reported apeech in Romance languages”, in: Tom Güldemann and Manfred von Roncador (eds). Reported Discourse. A Meeting Ground for Different Linguistic Domains. Amsterdam: John Benjamins, 143–172. Herlyn, Anne. 1999. “So he says to her, he says, ‘Well,’ he says. …: multiple dialogue introducers from a historical perspective”, in: Andreas H. Jucker, Gerd Fritz and Franz Lebsanft (eds). Historical Dialogue Analysis. Amsterdam: John Benjamins, 313–330.

254   

   Karin Aijmer

Kytö, Merja and Terry Walker. 2006. Guide to a Corpus of English Dialogues 1560–1760. Acta Universitatis Upsaliensis. Studia Anglistica Upsaliensia 130. Leech, Geoffrey. 1999. “The distribution and function of vocatives in American and British English conversation“, in: Hilde Hasselgård and Signe Oksefjell (eds). Out of Corpora: Studies in Honour of Stig Johansson. Amsterdam: Rodopi, 107–118. Lucy, John A. 1993. “Metapragmatic presentationals: reporting speech with quotatives in Yucatec Maya”, in: John A. Lucy, (ed.). Reflexive Language. Reported Speech and Metapragmatics. Cambridge: Cambridge University Press, 91–126. Lutzky, Ursula. 2012. Discourse Markers in Early Modern English. Amsterdam: John Benjamins. Moore, Colette. 2002. “Reporting direct speech in early modern slander depositions”, in: Donka Minkova and Robert Stockwell (eds). Studies in the History of the English Language; A Millennial Perspective. Berlin: Mouton de Gruyter, 339–416. Moore, Colette. 2011. Quoting Speech in Early English. Cambridge: Cambridge University Press.

Colette Moore

10 Histories of Talking about Talk: Quethen, Quoth, Quote Abstract: This study examines the histories of the two verbs quoth and quote as they illustrate the ways that English speakers have talked about talk. Quoth derives from the Middle English verb quethen, which syntactically narrowed in Early Modern English to be used only in the inverted constructions quoth he/she or quoth I. The process is one of different stages of grammaticalization: the verb quethen took on a greater role in organizing discourse, and then, ultimately, as the verb was restricted syntactically to quoth-constructions, it was also restricted pragmatically as a tag for direct speech in literary narrative. The verb quoth then declined in use in the eighteenth and nineteenth centuries as the use of quotation marks carried the day as the primary method for marking direct discourse. At the same time that the verb quethen/quoth faded, the verb quote was rising. Quote is etymologically unrelated to quoth, but their phonetic and orthographic similarity led to a perception of them as related words. This perceived relationship stems in part from the coterminous increase in the usage of one and decrease in the usage of the other. The words quoth and quote, however, represent different pragmatic aims for language: the first a dialogue marker that served as a grammaticalized organizational word (in a language that had no punctuation conventions to do this work), and the second a credentializing word that creates authority within a text by asserting the accuracy of the reported words. Keywords: reported speech, quotation, quoth, grammaticalization, inquit verbs

1 Introduction Norms for understanding quotation and conventions of indicating quoted passages have altered dramatically from premodern English to present-day English. As societal organization and communicative needs shift, so, necessarily, do the pragmatic and functional structures of language. Even the words for talking about quotation differ between premodern English and the present-day: the verb quote, for instance, only enters English in the modern period, while early English used other terms and strategies for indicating a quotation. Middle English did employ terms like reherse to indicate the quotation of speech, but more often, quoting was simply talked about with verba dicendi (verbs of speaking) like the term quethen, whose forms (especially the later form quoth) resemble quote. These words are

256   

   Colette Moore

false cognates, however. Indeed, the difference in their usage  – quote refers to the act of citing and is often used with a narrative report of speech or writing, and quoth is a verbum dicendi that introduces directly reported speech  – indicates alternate perspectives on how to understand the act of representing another person’s words. The contrast between the histories and functions of quoth and quote can therefore help us to understand quotation as a pragmatic impulse and a cultural product. The process of quotation fundamentally overlaps with the direct reporting of speech (oratio recta): the mechanisms through which one narrative includes and represents the interjection of another voice. It also overlaps textually with the related pragmatic impulse to mark the voice of recognized experts whom one is citing for authorization. This essay will explore these productive overlaps to discuss the historical development of quotative practice from premodern English. In present-day English texts, we typically employ quotation marks that clearly demarcate the onset and offset of quotations. Premodern English has no set punctuation conventions for the marking of quoted speech, so this marking is more fluid and less conventionalized in early English manuscripts. Indicating quotations was done not with quotation marks, but through linguistic and scribal means: through the words of the texts (particularly the verbs of speaking) and through scribal marks on the manuscript page (Moore 2011: 18–68). The development of the verb quoth, for this reason, sheds light on the process of grammaticalization in verbs of speaking. In previous work, I pointed out the pragmatic and textual-organizing function of verbs of speaking (verba dicendi) in a minimally-punctuated writing system (Moore 2011). There, I focused on the dominant verb say, but the ME verb quethen that becomes quoth is another case of a grammaticalized verb of speaking, particularly interesting because of its similarity to the later verb quote. Because the quotative verb quoth is always part of a clause, it is also illuminating for the consideration of how constructions are grammaticalized. (Noël 2007, Trousdale 2012, and Traugott 2003 have connected grammaticalization theory with construction grammar, and Vandelanotte 2009 showed what a constructionist approach to speech presentation might look like). Section 2 explores the afterlife of the Middle English quotative verb quethen (‘say’) and its descendant quoth, and the emergence of the verb quote. Section 3 pursues the strategies that premodern texts utilized to signal the voices of external authorities.

Histories of Talking about Talk: Quethen, Quoth, Quote   

   257

2 Talking about Quotation: Verbs of Speaking and Quoting Quotation and speech reporting are related phenomena. Even the words for speaking and quoting overlap: the verbs quoth and quote, for example, have long encroached associatively upon one other’s semantic territory, though they are in fact etymologically distinct. An internet search for quoth, for example, reveals several popular assertions that it is an archaic form of quote, possibly reasoning by analogy to the archaic third person singular indicative verb ending -eth. The third person -eth endings were characteristic of southern dialects of Middle English (maketh, sayeth, knoweth, etc.) and were used into the Early Modern period (see e.g. Lass 1994: 162–166, Nevalainen 2006: 90–92). Since the -eth endings were preserved in the morphosyntax of the King James Bible and Shakespeare’s plays, they are widely emblematic of archaic English (so, for instance, if goeth is the old form of go, then quoth must be the old form of quote). This is perhaps the reason, then, that the words quote and quoth are linked to one another in the folk etymological imagination and mistaken for one another. One example: searching Google for “quote the raven nevermore” provides 344,000 hits (Google gently suggests, however, that I might mean “quoth the raven,” and I was obliged to restrict my search to the exact phrase lest it “help” by including the correctly quoted instances of Poe’s poem). Though they came into English by different paths, the words quote and quoth are popularly assumed to be related, and this folk etymological misprision affects the way that the words are perceived, making them an example of what Elizabeth Traugott terms an “attractor set” of similar constructions whose histories are interconnected in ways that motivate and constrain their functionality (Traugott 2003: 645). The Oxford English Dictionary entry for quote (v.) does not acknowledge any mutual influence between the two, however; it refrains from mentioning any associative connection to quoth, quethen, and cweþan, indicating only that quote derives partly from post-classical Latin quotare (cotare) ‘to mark (a book) with numbers’ and Middle French quoter ‘to cite, to mark’. Defining the second sense of the word as: “To reproduce or repeat a passage from (a book, author, etc.); to repeat a statement by (a person); to give (a specified person, body, etc.) as the source of a statement,” it dates this usage from the sixteenth century. Similarly, the OED entry for quoth (v.) does not mention quote or acknowledge any connection even though it points out that one branch of the orthographic forms of quoth in Middle English is quod and that this is sometimes devoiced to -t. Quod is a very common usage in Middle English, in part because manuscripts often employ the abbreviation qd or a letter q with a bar through the descender, prac-

258   

   Colette Moore

tices which are borrowed from Latin abbreviations of the word quod. This results in the written practice of orthographically depicting the word as quod – even in regions where the final sound might have been an interdental fricative. The frequent use of quod, though, means that the forms of quethen often resemble forms of quote. Final devoicing gives, for example, the following usage cited from Cursor Mundi (ca. 1400): ‘Say me,’ quot iacob, ‘hou es þis?’. Cursor Mundi’s incarnation of quethen looks very much like quote, and it is easy to see how, given their phonetic and orthographic similarity, speakers might assume that these words are related and how quoth and quote could be connected associatively in ways that influenced the conception of both.

2.1 Quethen The decline of Middle English quethen and its now-archaic modern descendant quoth indicate shifting methods of marking and conceiving of reported speech. In my larger study of speech reporting in Middle English, I found that quethen is the second most frequent verb of speaking (following the dominant seien ‘say’). Data from a selection of speech onsets (a collection of 3993 onsets of directly reported speech) randomly sampled from the Corpus of Middle English Prose and Verse¹ shows the breakdown of verbs of speaking as follows: Table 1: Middle English verbs of speaking indicating passages of direct speech in the CMEPV (Moore 2011: 57)

verb (forms of)

Occurrences n = 3993

percentage of total usages

seien

2223

56 %

quethen

 596

15 %

answeren

 171

 4 %

speken

  77

 2 %

other (crien, callen, etc.)

 304

 8 %

total verbs of speaking

3371

84 %

total lines of DS containing verbs of speaking

3193

80 %

1 For further details on the corpus or on the study, see Moore 2011: 54–61.

Histories of Talking about Talk: Quethen, Quoth, Quote   

   259

Although forms of the verbs seien and quethen do compete for some of the same environments in Middle English, they are not interchangeable in usage. Forms of seien are by far the most common and can be used for both directly and indirectly reported speech, as well as narrative descriptions of speech. Although it is possible for quethen to indicate indirect speech, this is an early usage that becomes less frequent and does not occur often in late Middle English. By later Middle English, quethen typically marks direct speech, occurs in an inverted position followed by the subject, and occurs in a clause following an initial onset of reported direct speech (boldface added): (1) ¶ To holy chirche quod he • for to here masse (Piers Plowman, Cambridge Trinity College B.15.17.) (2) ¶ Petur qd a ploughman & putte forth his heed (Piers Plowman, BL Addl 35157) Notice the similar sentence structure of the two: an initial word or clause followed by the inquit clause (Latin for “he said”) with inverted word order and then the resumption of the direct speech. This inverted word order together with its morphosyntactic restriction to the third or first person form indicate a level of morphosyntactic fixing that is consonant with the verb’s development as a grammaticalized quotative marker (for discussion of the inverted word order of quotative clauses, see Buchstaller 2014: 39–41). By the late medieval period, quethen has become narrower in its syntactic range, and is used generally as an inquit clause that interjects into a passage of directly reported speech to mark the different speaker. As such, it has developed functionality as a grammaticalized phrase, a fixed inquit formula (see Moore 2011: 49–54). The inquit clause appears differently in different manuscripts, and looking at manuscript witnesses for the sentence in (2) gives evidence of different visual manifestations of the phrase and of how backgrounded it appears to be (boldface of the verb of speaking is mine, and special characters in these transcriptions are attempts to render the manuscript characters). (3) a. peter coth a plowman • and put forþ his hede (London, British Library, MS Cotton Caligula A xi) b. petɔ • ꝗ a ploumā z putte forþ hs hed (Oxford, Bodleian Library, MS Laud Misc. 656) c. petr qd a ploumā @ put forth his hede (Oxford, Bodleian Library, MS Digby 145)

260   

   Colette Moore

d. ¶ Peter quod a plouhman • and putte forþ his heed (London, British Library, MS Cotton Vespasian B xvi) e. Peyter qd a plowman t put forþ hys hefd (London, British Library, MS Harley 2376) f. ¶ Petr ꝗ a plouʒman ǀ and put forth his 8 hede (London, British Library, MS Harley 6041) g. Petyr ꝗ a plowmā z put forth hys hedˑ (London, British Library, MS Harley 3954) h. Petre qd a ploughmā: and put forþe his hede (Oxford, Bodleian Library, MS Douce 102) In some of these, the verb is reduced to an abbreviation, even a symbol, making it appear almost as a form of punctuation. This kind of visual truncating is somewhat analogous in written texts to the phonetic reduction that characterizes grammaticalization. Grammaticalized constructions, therefore, are part of the textual organization of medieval texts. In the absence of clear punctuation conventions for delineating reported speech consistently, the words of the text had to assume some of this functionality (Moore 2011: 42).

2.2 Quoth Quethen ultimately survives into modern English only in the fixed form quoth, the past indicative form, with formulaic inquit usage. Searching for quoth in the Corpus of Historical American English (COHA), for instance, gives 991 hits, all of which occur in the third or first person (most in the third person), and 987 (99.6% of them) occur with inverted word order. A few typical examples from the beginning of the list look like this: (4) Sir, “ quoth Honorius, “we presume You guess from… (1820, FIC) (5) “A novel! “ quoth I – “when Waverly is galloping,…. (1824, FIC) (6) “I felt weakened in spirit, “ quoth he in return, “and…. (1827, FIC) These citations show the typical form of the construction: third or first person and inverted word order. More often, also, the clause interjects medially into the reported speech, and it always occurs in a clause directly adjacent to the pur-

Histories of Talking about Talk: Quethen, Quoth, Quote   

   261

ported utterance. It is solely a reporting verb, in other words, and is not used for narrative descriptions of speech. Early grammars describe the restricted usage of quoth: R. G. Latham in his Elementary English Grammar (1875) lists it among the “defective verbs.” He points out that We sometimes say quoth I. We never say quoth thou. We rarely if ever say quoth we, quoth ye or quoth they. The Infinitive and Participial forms are unknown. No one living has ever said – to quoth or quothing. On the other hand, however, the compound bequeath is perfectly regular. To bequeath; bequeathed, I bequeathed, &c. are ordinary Verbs with the usual inflection. The word quoth, then, so far as it is not obsolete, is a signal, typical, or representative instance of Defect in its most limited form. Nor is it this as a single word only. It impressed its character on the Syntax. We rarely, if ever, say he quoth, but quoth he; in other words, it is followed by its Pronoun which in other Verbs, as a rule, comes first. Add to this that the Pronoun itself is, from a comparatively late period, not given in full, but abbreviated (quoth-a), and to a great extent, incorporated with the main word as a Postpositive Affix. (Latham 1875: 134)

Latham’s invocation of “defect” as a feature of certain verbs is an interesting characterization – he assesses quoth as defective because we no longer have access to the full paradigm. This negative valuation privileges a syntactic perspective, of course; from a pragmatic angle, we could just as easily focus on the verb as positively “enriched” by its morphosyntactic fixing which permits it greater communicative force. Perhaps this fixing in the usage presages the declining usage of the form, however: the COHA frequency data also paint a picture of a term that becomes marginalized after the nineteenth century. Table 2: Occurrences of quoth in the Corpus of Historical American English (COHA)

262   

   Colette Moore

The vast majority of the hits for quoth occur before the twentieth century (peaking at 11.4 hits per million in the 1820s), and the frequency of hits per million falls below 2 after 1910. Even this is perhaps not wholly representative of the speed at which attitudes evolved; Samuel Johnson’s dictionary entry for quoth in 1755 already disparagingly remarks that it is “now only used in ludicrous language” (Johnson 1755: 1627). The later twentieth century citations are visibly more restricted – the construction is used only in purposeful archaisms or markedly stylized expressions. The sparse recent usages are either from science fiction/ fantasy novels or are explicit references to Poe’s “The Raven”. The same generalizations in usage hold true for the Google Books database, which contains half a million hits for the word quoth. Looking through the hits indicates that here also the word is used only in the past tense indicative, associated with the third and first person, and occurs with inverted V-S word order. Both the American and British English datasets show quoth in use in the books published in the nineteenth century (fluctuating between 4 and 5 hits per million in the American data and remaining above 12 hits per million in the British) and hanging on through the mid twentieth century (dropping below 2 hits per million in 1940 in the American data, and dropping below 2 hits per million in 1970 in the British data). Table 3: Occurrences of quoth in the Google Books (U.S.) database

These data should not suggest that speakers are still using quoth in the midtwentienth century, of course; the dates here denote the publication dates for texts rather than their dates of composition, so it is reflecting the usage of printed books which are preserving archaic usages. Clicking on the British data from the 1970s, for instance, shows that hits for quoth are primarily from editions of earlier texts (often Shakespeare). The use of quoth in its syntactically restricted constructions: quoth he/she, quoth I, shows what Joan Bybee describes as the “automatization of frequently occurring sequences” (Bybee 2003: 153). The forms are frequently repeated in the first and third person past tense as markers of narrative dialogue, and the verb

Histories of Talking about Talk: Quethen, Quoth, Quote   

   263

otherwise vanishes from use, preserving these constructions only as grammaticalized dialogue markers.

2.3 Quoth and Quote The decline in the usage of quoth is coincident with the rise in the occurrence of quote. These usages are not competing forms, of course (since one is used in the sense of ‘cite’ and the other is used in the sense of ‘say’), so the rise of quote does not come at the expense of quoth – at least not directly – but it is an interesting cooccurrence, given their associative connection. Figure 1 shows the two forms in Google Books’ ngram viewer.² The ngram search engine could not successfully manage to isolate the verbal form of quote, so I searched for quoted, which can only be a verb and which has the advantage of matching the past tense of quoth. We cannot place too much confidence in Google Books’ data, especially from the early periods which contain many OCR errors (errors introduced by optical character recognition software) and patchy sampling. But even given the limitations of Google Books ngram viewer, I think that the general picture is probably accurate. Examining the respective frequency patterns of quoth and quoted in the British English section of Google Books shows that quoth was quite frequent through the mid-seventeenth century. The terms then coexisted in the language through the mid to late eighteenth century before quoted rose in popularity while quoth dwindled as a construction.

2 Google Books ngram viewer is a searchable interface for Google Books that permits comparisons between different words over time (Michel et al. 2011). It is a limited tool, as I explain, but provides access to heretofore unrealized amounts of data.

264   

   Colette Moore

Figure 1: Google ngrams for quoth and quoted in British English (smoothing = 10 years)

Figure 2: Google ngrams for quoth and quoted in English fiction (smoothing = 10 years)

Figure 2 examines the two constructions in the part of the database that Google tags as “English fiction.” Again, Google Books’ search engine is too blunt a tool to produce conclusive results about genres, but it reinforces our impression that quoth serves as a particularly literary narrative marker. Quoth is a construction that is particularly useful for representing dialogue in narrative, and its syntactic narrowing (restriction to first and third person and the past tense) is consonant with a functional narrowing for this purpose (since narrative is typically in the past tense and oriented from the first or third person). Through the mid-nineteenth century, the English fiction part of Google Books shows usage of quoth with a higher frequency than quoted (in contradistinction to Fig. 1). The dates

Histories of Talking about Talk: Quethen, Quoth, Quote   

   265

are dates of publication, certainly, and not dates of composition, so the hits for quoth are an artifact of reprinted editions and are certainly preserving the usage of older literary works in the printed record. Comparing figures 1 and 2, we see that while the word quoth is overrepresented in the fiction section proportional to the broader sample, the word quoted is underrepresented. Fiction does not find as much use for the modern verb quote to organize discourse in the way that it depended upon the older verb quoth for the purpose of organizing dialogue. Quoted enters the scene in the eighteenth and nineteenth centuries, therefore, but it does not show the kind of frequencies associated with a discourseorganizing form (as quoth had been). Quoth, for its part, shows precipitous decline, suggesting that its discourse-organizing function is no longer needed. These trajectories are not directly related paths (though they might go some way to creating the impression that quoth is the archaic form of quote). If the heavy lifting of marking represented speech is done by quotation marks, the two main quotative verbs, said and quoth, are no longer functionally necessary for this purpose. They might, however, both be part of larger textual shifts in the marking and conception of quotation. The decline in quoth coincides generally with the spread of quotation marks as a convention for marking reported speech. As written language settled on the convention of using punctuation to mark reported speech, the grammaticalized verbs of speaking became a redundant strategy for indicating the shifts in speaker.

2.4 Coda: quotha Evidence for the further grammaticalization of the construction comes from the appearance of the fused form quotha, from ‘quoth he’, which has 21 hits in COHA and several thousand hits in Google Books. This conjoined spelling (mentioned by Latham 1875, cited above) suggests that the construction was phonetically reduced in rapid speech; it also appears as cohee. (Compare the parallel pronunciation spelling or eye-dialect form sezee in Joel Chandler Harris: “‘Mawnin’!’ sez Brer Rabbit, sezee – ‘nice wedder dis mawnin’,’ sezee.” [Harris 2003–4].) Another related construction, cohee, in fact, was apparently such a characteristic usage of the speech of Scots-Irish settlers to the Virginia/Carolina region of the United States that they were referred to as “Cohees.” Texts of the period explain that the name for the group “Cohee” comes from Scotch-Irish dialectal “quo’ he” (quoth he) (Caruthers 1834: 192). The OED entry for quotha shows that the form had moved even further on the cline of grammaticalization. It gives QUOTH v. + a, variant of HE pronoun

266   

   Colette Moore

as the etymology and describes the usage as follows: “Used with contemptuous, ironic, or sarcastic force after repeating words said by someone else: ‘he said?’, ‘she said?’; (hence) ‘indeed!’ COHA gives examples like this one: (7) …we should take our proper places among the Old Masters. Young, quotha? Why, we were thirty-four! (1879, MAG) This ironic meaning suggests that the phrase had grammaticalized even further, intensifying its function as an indicator of speaker attitude. If the use of quoth as a grammaticalized quotative marker is an example of what Traugott and König term semantic-pragmatic tendency II: “Meanings based in the described external or internal situation > meanings based in the textual situation” (Traugott/König 1991: 208), then the further development of quotha could be taken as an example of further grammaticalizing, their semantic-pragmatic tendency III: “Meanings tend to become increasingly situated in the speaker’s subjective belief-state/ attitude toward the situation” (Traugott/König 1991: 209, see also Traugott 1989: 34–5, Traugott/Dasher 2002: 95). The contemptuous or sarcastic sense of quotha picks up on the subjective angle: that one might report the speech of another in order to dismiss it. This commentary function is an additional subjectification of the term: a further pragmatic development.

3 Quethen vs. Quote: Perspectives on Quoting The functions of the verb quethen in premodern English and the verb quote in present-day English are indicative of different frameworks for organizing speech presentation and different sets of priorities about what it was important to indicate. The verb quethen prepares the reader for the narrative to switch to the orientation of another speaker; the verb quote insists that reported words are a precise copy. This is a very different kind of textual work. The second kind of textual labor, the marking of faithful reporting, is partly an invocation of alternate authority, a claim that the present speaker or writer is deferring responsibility for the words to another source. This can be a way of shifting the blame (“I’m just quoting!”) or a way of assuming authority from an acknowledged expert – sometimes called the credentializing function of direct speech (Bednarek/Bublitz 2006: 551). Premodern English texts – though they did not seem as eager to claim precision of representation – did have means for indicating important sources. These methods for marking recognized auctores are known as auctoritas, and are typically seen as the purview of scribes.

Histories of Talking about Talk: Quethen, Quoth, Quote   

   267

One of the classic examples of auctoritas marking, for instance, is the Ellesmere manuscript of the Canterbury Tales, where the word auctor appears in the margin beside some sentences. This marginal note was used by the Ellesmere scribe to mark citations worthy of particular attention or utterances credited to a classical or patristic figure. Such notae are a medieval way of flagging utterances that are acknowledged to be notable for truth and wisdom (Minnis 1988: 10). Minnis cites Hugotio of Pisa’s definition of an auctoritas as a sententia digna imitatione, a ‘profound saying worth of imitation or implementation’. Auctor notes, then, were not primarily markers of quotation, but rather markers for sententiae written by acknowledged auctores, and therefore worthy of being noted and imitated. Manuscript marking is linked to the scribal responsibility for ordinatio and compilatio: particular roles associated with the production of texts. St. Bonaventura’s writings detail the medieval taxonomy of textual production: a scriptor copies text, a compilator orders a text by rearranging or adding marginal notae, a commentator supplements texts with other arguments (sometimes his own), and an auctor composes texts of his own (Parkes 1976: 127). The compiler’s role in production was to create an interpretational frame to assist in reading a manuscript (compilatio), and to perform an ordering of text (ordinatio). Most of the revision and use of notae by scribes, then, was completely consistent with the medieval models of compilatio and ordinatio (Smith 2001: 42–46). Manuscript markers serve as the compiler’s tools for interpretation, and the page records his construction of markings to eludicate the text (Parkes 1997: 58–9; 1999: 338). Scribes, then, employed the mise-en-page (the layout on the page) in different ways to assist in the interpretation of texts. There are some markers that are shared between texts and scribes, but no punctuation conventions of marking reported speech that were so widespread and consistently applied as to be expected in a text. Individual studies of speeches and voices in Confessio Amantis and Roman de la Rose, for example, have both shown inconsistency of methods and narrative confusion of levels (Hult 1984: 260–261 and Echard 2001: 61–62). The scribal methods of organizing information in texts were not conventionalized in a broad way, but created local coherences – sometimes within a single manuscript or group of manuscripts and sometimes only within a section of a manuscript. It was, however, these scribal methods of flagging auctoritas that primarily performed the function of credentializing. Direct speech was not so clearly demarcated or so precisely conceived as a faithful representation that it could be relied upon for that textual work. Only in the ensuing centuries in which print conventions like the quotation mark developed, could reported speech effectively take over from scribal notae.

268   

   Colette Moore

A verb like quote, therefore, is a product of a later model for speech presentation, one in which faithfulness in presentation is understood to be closer to a verbatim ideal. It is a way of tapping into a similar kind of cultural authority as the auctoritas marks in manuscripts. The fact that the rise of quote as a credentializing pragmatic element loosely corresponds to the decline of quoth as a quotative pragmatic element is an interesting circumstance. It is perhaps not too much to suppose that their phonetic and orthographic similarity may have caused them to be associated, and that quote benefitted from the former prominence of quoth.

4 Conclusion The grammaticalization of the constructions quoth he or quoth I, then, is a pragmatic solution to the imprecisions in indicating direct speech. As the form syntactically constricted from the early ME fully functional verb quethen to a more restricted late ME verb that appeared in numerous orthographical forms (quod, quoth, etc.) to the even further restricted form quoth, it assumed a greater pragmatic role in textual organization at the expense of its original ideational meaning ‘to say’. This grammaticalizing process allowed the construction to become a major strategy for marking dialogue turns in narrative (especially in fictional narrative). This function was eroded, however, by the development and wider conventionalizing of quotation marks, which occurred in the eighteenth century. By the nineteenth century, quotation marks are widely used in broadly consistent ways, and verba dicendi like quoth have become less necessary to the narrative organization. At the same time that quoth was becoming less frequent (because less central as a pragmatic feature of textual organization), the similar-appearing quote was becoming more frequent. In a more text-based and literate society, the credentializing function of reported speech in written texts becomes more important, and more of the pragmatic energy of a text must be devoted towards pointing to other texts and figures to establish its authority. The development and increased frequency of the verb quote, therefore, a fully functional verb that can be used in any number of constructions, does not surprise us. Perhaps piggybacking on the prestige of quoth, which in its narrative use in literary narrative and its increasingly archaic sound was becoming a word that indexed conservative ornate literary writing, quote takes on a new function for new textual needs. Where premodern English needed quethen as a quotative verb to organize discourse and early modern English kept quoth as a dialogue marker, later English no longer needed a quotative verb in such a grammatical way. Instead, later English makes more use

Histories of Talking about Talk: Quethen, Quoth, Quote   

   269

of a verb like quote that indicated sources and authorities. We might note, then, that the words and the marks on the page share the work in both cases, but the labor is divided along different lines: in Middle English, scribes used notae for credentializing and verbs to indicate shifts of orientation, in present-day English, writers use verbs (among other strategies) for credentializing and punctuation marks to indicate shifts of orientation. In the histories of the words quoth and quote, therefore, we can see the shifts in pragmatic goals for English. At the same time that a discourse-organizing quotation function (quoth) was becoming less helpful, a credentializing assessment of reported discourse (quote) was becoming more helpful. The associative connection and folk etymological linking of the two verbs makes it all the more appropriate to consider these two trajectories together.

References Bednarek, Monika and Wolfram Bublitz. 2006. “Reported speech: pragmatic aspects”, in: Keith Brown (ed.). Encyclopedia of Language and Linguistics, vol. 2. Oxford: Elsevier, 550–553. Buchstaller, Isabelle. 2014. Quotatives: New Trends and Sociolinguistic Implications. Chichester, U.K.: Wiley Blackwell. Bybee, Joan. 2003. “Mechanisms of change in grammaticization: the role of frequency”, in: Brian D. Joseph and Richard D. Janda (eds). The Handbook of Historical Linguistics. Oxford: Blackwell, 602–623. Caruthers, William. 1834. The Kentuckian in New York; or, The Adventures of Three Southerns. By a Virginian; Volume 2. New York: Harper & Brothers. Davies, Mark. 2008. Corpus of Historical American English (COHA): 400 million words, 1810–2009. http://corpus.byu.edu/coha/, accessed 30 Sep 2013. Davies, Mark. 2011- Google Books Corpus. (Based on Google Books n-grams). Available online at http://googlebooks.byu.edu/. Based on: Jean-Baptiste Michel, Yuan Kui Shen, Aviva Presser Aiden, Adrian Veres, Matthew K. Gray, The Google Books Team, Joseph P. Pickett, Dale Hoiberg, Dan Clancy, Peter Norvig, Jon Orwant, Steven Pinker, Martin A. Nowak, and Erez Lieberman Aiden. “Quantitative Analysis of Culture Using Millions of Digitized Books.” Science 331 (2011) [Published online ahead of print 12/16/2010]. Echard, Siân. 2001. “Dialogues and monologues: manuscript representations of the conversation of the Confessio Amantis”, in: Alastair J. Minnis (ed.). Middle English Poetry: Texts and Traditions: Essays in Honour of Derek Pearsall. York: York Medieval Press. http://books.google.com/ngrams, accessed 30 Sep 2013. Harris, Joel Chandler [2003–2004]. Uncle Remus: His Songs and His Sayings. University of Virginia American Studies Program. http://xroads.virginia.edu/~HYPER/Harris2/cover. html, accessed 30 Sep 2013. Hult, David F. 1984. “Closed quotations: the speaking voice in the Roman de la Rose”. Yale French Studies 67, 248–269.

270   

   Colette Moore

Johnson, Samuel. 1755. A Dictionary of the English Language: A Digital Edition of the 1755 Classic by Samuel Johnson. http://johnsonsdictionaryonline.com/, last modified 5 May 2013. Lass, Roger. 1994. “Phonology and morphology”, in: Roger Lass (ed.). The Cambridge History of the English Language, vol. 3. Cambridge: Cambridge University Press, 56–186. Latham, Robert Gordon. 1875. An Elementary English Grammar for the Use of Schools. London: Longmans, Green, and Co. Michel, Jean-Baptiste, Yuan Kui Shen, Aviva Presser Aiden, Adrian Veres, Matthew K. Gray, William Brockman, The Google Books Team, Joseph P. Pickett, Dale Hoiberg, Dan Clancy, Peter Norvig, Jon Orwant, Steven Pinker, Martin A. Nowak, and Erez Lieberman Aiden. 2011. “Quantitative Analysis of Culture Using Millions of Digitized Books.” Science 331: 6014. 176–182. (Published online ahead of print: 12/16/2010) Moore, Colette. 2011. Quoting Speech in Early English. Cambridge: Cambridge University Press. Nevalainen, Terttu. 2006. An Introduction to Early Modern English. Oxford: Oxford University Press. Noël, Dirk. 2007. “Diachronic construction grammar and grammaticalization theory”. Functions of Language 14, 177–202. Parkes, Malcolm B. 1976. “The influence of the concepts of ordinatio and compilatio on the development of the book”, in: Jonathan J. G. Alexander and Margaret T. Gibson (eds). Medieval Learning and Literature: Essays Presented to Richard William Hunt. Oxford: Clarendon Press, 115–141. Parkes, Malcolm B. 1997. “Punctuation in copies of Nicholas Love’s Mirror of the Blessed Life of Jesus Christ”, in: Shoichi Oguro, Richard Beadle, and Michael G. Sargent (eds). Nicholas Love at Waseda: Proceedings of the International Conference. Cambridge: D. S. Brewer, 47–60. Smith, D. Vance. 2001. The Book of Incipit: Beginnings in the Fourteenth Century. Minneapolis: University of Minnesota Press. Traugott, Elizabeth Closs. 2003. “Constructions in grammaticalization”, in: Brian D Joseph and Richard D. Janda (eds). The Handbook of Historical Linguistics, Malden: Blackwell, 624–647. Traugott, Elizabeth Closs and Ekkehard König. 1991. “Semantics-pragmatics of grammaticalization revisited”, in: Elizabeth Closs Traugott and Bernd Heine (eds). Approaches to Grammaticalization. Amsterdam: John Benjamins, vol. I, 189–218. Trousdale, Graeme. 2012. “Grammaticalization, constructions, and the grammaticalization of constructions”, in: Kristin Davidse, Tine Breban, Lieselotte Brems, and Tanja Mortelmans (eds). Grammaticalization and Language Change: New Reflections. Amsterdam: John Benjamins, 167–198. Vandelanotte, Lieven. 2009. Speech and Thought Representation in English: A CognitiveFunctional Approach. Mouton de Gruyter.

Winfried Rudolf

11 Quoting and Translating Latin in the Old English Homilies of the Vercelli Book Abstract: Many of the anonymous Old English homilies are eclectic compositions that were translated and compiled from a variety of Latin sources. Sometimes traces of these sources remain visible in the vernacular text in the form of short Latin quotations, often comprising scriptural content. Combining palaeographical, linguistic and literary analysis, this paper analyses the remaining Latin citations in the Vercelli Homilies, which belong to the oldest vernacular preaching texts from Anglo-Saxon England. The evidence presented here demonstrates how a close study of the bilingual passages can inform theories of transmission, revision and performance of these impressive pastoral addresses. The preliminary results invite a holistic study of the Latin vestiges in Old English homilies, which may provide a gateway towards a new understanding of the origins of vernacular preaching in England. Keywords: quotes, quoting, homiletics, source studies, palaeography, bilingualism, Anglo-Latin

1 Introduction The great majority of vernacular homilies preached by the Anglo-Saxon church depends on Latin sources (cf. Wright 2007; Fontes Anglo-Saxonici 2013). This does not only testify to the established doctrinal status of the Bible and the works of the most influential Christian auctores, but also to the authority of Latin as one of the three holy languages of Christianity (Berschin 1980: 31–38). In his famous instructions in Book IV of De Doctrina Christiana (ed. Green 1995: 280–281), Augustine declared the compilation of homiletic texts from various sources to be a legitimate procedure, as long as such textual concoctions maintained a respectable level of style and wisdom and did not lead to major doctrinal conflict. This eclectic compositional method of pastoral addresses (cf. Tristram 1995) would rarely involve the visible demarcation of the distinct textual units drawn from the various Latin sources, which could sometimes gravely differ in style, authority and orthodoxy, nor would it always care to mention the original authors, for example the Fathers of the Church. Some Old English homilies, once translated and constructed from such patristic sources, or indeed from any set of Latin excerpts, could sometimes appear more coherent in style and diction to the extent that the compilational

272   

   Winfried Rudolf

seams in the text(ile) became completely opaque. Others, such as many of the earliest surviving pieces, still mirror the linguistic, stylistic and theological inconsistencies of their varying source materials, which caused the elite Anglo-Saxon homilist Ælfric of Eynsham to dismiss these texts as heterodox (ed. Clemoes 1997: 174) and answer this formal problem by developing his own well-tempered prose style (Pope 1967: 105–36). By the late tenth century, the period to which the Vercelli Book (Vercelli, Biblioteca Capitolare, MS CXVII), one of the two oldest surviving collections of Old English homilies, can be dated, homilists were no doubt already drawing on Latin as well as Old English translations of Latin materials to compile their texts.¹ The Vercelli collection contains pieces written by a variety of anonymous homilists at different times and therefore shows no pattern of a specific relationship to one specific source (cf. Scragg 1973 and 1992: xxxvii). The exact age as well as the original place of composition of the single items is still uncertain. Although there are only few homilies or selected passages in the collection for which direct or antecedent sources have not been identified (yet), the influence of Latin texts can be traced with relative precision through the language and contents of most of the pieces (Scragg 1992: xxxvii–xxxviii). A number of the source passages used in the compilation of the Vercelli homilies derive from the older homiletic tradition of pre-Carolingian times, such as the works of John Chrysostom, Caesarius of Arles, Isidore of Seville and Gregory the Great, while other portions were translated from the anonymous deutero-canonical tradition of Christian writings, such as for example the Catechesis Celtica, the Apocalypse of Thomas, the Homiliary of St-Père-de-Chartres, Pseudo-Augustinian sermons and the Visio Pauli (cf. Fontes Anglo-Saxonici 2013).² Many of these texts were transmitted in various Latin redactions during the Middle Ages, and it is not always possible to prove with absolute certainty to which of these an Old English homiletic text is directly indebted, nor how many unknown antecedent Latin or Old English versions participated in the textual state of a version that survives today. Younger sources of some Vercelli homilies include the works of Paulinus of Aquilea (homily X) or the Capitula Theodulfi (homily III) from the beginning of the ninth century, which serve as a terminus post quem for the composition of these pieces (Scragg 1992: 1 The other collection is the so-called Blickling Homilies, preserved in the Princeton University Library, Scheide Collection MS 71. For a facsimile of the Vercelli manuscript see Sisam (1976). The best edition of the homilies, including references to their sources, is Scragg (1992). All subsequent references to manuscript readings and edited text (by homily and line number) follow Sisam and Scragg. Zacher (2009) has even identified an antecedent Greek source for Vercelli homily VII. 2 A full bibliography on the Vercelli homilies including scholarship on their specific sources, as far as they are known, can be found in Remley (2009).

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   273

72). It goes without saying that all sources mentioned contain quotations from the Bible which equally made their way into the Old English composite homilies through translation. The twenty-three homilies of the Vercelli Book do not represent a fullyfledged collection analogous to those found in the Carolingian liturgical tradition (Barré 1962). The series of homilies is incomplete with regard to the temporale, as it contains a disordered sequence of addresses for the Passion, Lent, Christmas, Rogationtide and prose accounts of the lives of selected saints such as Martin and Guthlac, but also some catechetical pieces that could be preached at any given occasion.³ This mixture, complemented and enriched by the well-known Christian poems of the manuscript, does not give the impression of a preconceived idea for the manuscript that would fulfil the liturgical needs of a community, but rather points towards the tastes of an individual (O’Carragain 1975, Treharne 2007). Even so, the liturgical status of the vernacular collection remains as mysterious as its potential audiences (Zacher 2009: 32–42),⁴ let alone its transport to Italy before the middle of the twelfth century, which is still steeped in speculation (Halsall 1969, Sisam 1976: 44–50, McBrine 2009). Yet there is certainly no risk in assuming that these Old English homilies were delivered to people who were not always sufficiently educated to understand similar addresses in Latin (Rudolf 2011: 186). Against this background it makes sense to ask why twelve of the Old English Vercelli homilies still include about sixty instances of short quotations from Latin.⁵ This article examines the nature of these quotations and tries to evaluate their functions with regard to the source situation, compositional procedures, quotation, and translation practice.

2 Palaeographical Aspects of the Latin Quotations in the Vercelli Manuscript The analysis of the quoted Latin within the vernacular English pieces must begin with a look at the material representation in the original manuscript. The Vercelli 3 For the last category note the rubric of homily XIV: LARSPEL TO SWYLCERE TIDE SWA MAN WILE (‘Sermon suited for any occasion’), 76v. 4 On the various types of homiliaries and the difficulty of determining their audiences see Clayton (1985) and Gatch (1989). 5 These are homilies I–V, VIII, XI–XVI. Here I do not consider the Latin rubrics of five of the Vercelli homilies (VI, XV–XVIII) to be quotations, nor single Latin words which occur in the prose, such as coronan (I:144), rex (I:212), euuangeliste (I:58, 83, 227), militie (I:143, 214, 223) or tribunus (XVIII:11).

274   

   Winfried Rudolf

codex is traditionally dated to the second half of the tenth century (Förster 1913: 28; Ker 1957: 460). It shows aspects in its layout which are typical of the time, yet markedly different from modern expectations of any form of a consistent demarcation of quotations. Like almost all manuscripts containing more than just glosses or annotations in Old English, the Vercelli Book presents the vernacular text in a single main column. This arrangement of the page is one reason why the immediate identification of an embedded Latin passage by an Anglo-Saxon reader may have been difficult. Nonetheless, it is legitimate to ask – in absence of unmissable markers in the layout – whether the quotations stand out graphically in any particular way at all. As a corollary, two aspects take centre stage when tracing the possible visual highlighting of these passages: on the one hand, the type of script used for the Latin, on the other, any punctuation framing these quotations. Around the turn of the millennium Anglo-Saxon scribes were only gradually beginning to segregate Old English and Latin in bilingual manuscripts by introducing different types of script for each language (Dumville 1993: 101). The early landmark for a strict distinction in this regard is Oxford, Corpus Christi College, MS 197, a bilingual translation of The Rule of St Benedict which served as a palaeographical benchmark for Neil Ker’s seminal Catalogue of manuscripts containing Anglo-Saxon (1957: xxvi). Conversely, Vercelli can be regarded as a late representative of an almost indistinct palaeographical treatment of both languages by the scribe, as this manuscript hardly employs any letterforms for the Latin which cannot normally be found in Anglo-Saxon minuscule script.⁶ Hardly surprising among the differences is the letter (e.g. fol. 2v/8, Numquid), since it is absent from the Anglo-Saxon alphabet. One would assume the same for the tailed e-caudata (e.g. 27v/2, cęlo), but this letter can occasionally be found in Old English spellings of the Vercelli Book too (e.g. 28r/17, ęfne). In turn, unmistakable Anglo-Saxon minuscule letterforms occur in the spellings of the Latin in some cases (e.g. 72r/1, seð for sed and 72v/1, seminaþ for seminant; see plate 1). While writing for in seð could be explained as an unintentional slip by the scribe, the substitution of the Latin plural ending -nt with the equivalent plural Old English ending -þ in seminaþ (‘they sow’) in line one of the above image (plate 1) looks intriguing enough. Giving this spelling the benefit of the doubt, the form could at best be called a bilingual portmanteau or vulgar Anglo-Latin form,

6 For an incomprehensive analysis of those see Sisam (1976: 20–23).

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   275

Plate 1: Vercelli, Biblioteca Capitolare, MS CXVII, fol. 72v (detail)⁷

or, otherwise, an error.⁸ Whether it must be attributed to the scribe or an antecedent exemplar remains unclear, but it is possible here that linguistic interference of the Old English translation (ða ðe sawað on tearum hie eft on gefean ripað), which directly follows the Latin sentence (Qui seminaþ in lacrimis in gaudio mete[n]t)⁹ in the manuscript, played a part in this mixture of graphemes. Likewise possible is the use of a Latin exemplar with Old English interlinear glosses; the parallel representation of both languages in such manuscripts could no doubt cause confusion when copying and translating, especially because a largely indistinct code of writing was used for both languages in the Vercelli manuscript. The evidence could indicate that the original compilation and translation of the Old English versions involved bilingual manuscripts. Further evidence for bilingual hybridity can be found in a short passage at the beginning of homily XVI, giving the pericope for Epiphany (Matthew 3:13–17),¹⁰ in which the text switches six times between both languages in short succession, similar to the alternating bilingual text of the Homiliary of Angers in the Taunton Fragments (cf. Gretsch 2004). In the example below (85v/21, plate 2), the scribe wrote the form prohibebað¹¹ for prohibebat (‘forbad’, see Matthew 3:14) in the Latin,¹² this time turning the into the of the Anglo-Saxon minuscule alphabet. The form is thus reminiscent of a weak class 2 verb third person singular present indicative in Old English.

7 All photographs are reproduced by the kind permission of the Archivio e Biblioteca Capitolare, Vercelli (Italy). I am greatly indebted to Timoty Leonardi for providing these images. 8 Note that the erasure of the in metent indicates that a reader of this manucript considered seminaþ to be a singular form, in contradiction to the biblical plural. 9 The quotation is taken from the Roman version of Psalm 125:5 (‘They that sow in tears shall reap in joy’). All translations according to the King James Bible (eds. Marks & Hammond 2012). 10 All subsequent references to the Bible are according to the Vulgate in the edition of Weber (1969). 11 Abbreviations are expanded in straight letters throughout this paper. 12 Other examples occur in homily XI (72r/1, seð) and homily XIII (75v/14, quið).

276   

   Winfried Rudolf

Plate 2: Vercelli, Biblioteca Capitolare MS CXVII, fol. 85v (detail)

What follows then is the translation in Old English extended by another sentence before the text quotes Latin again. This short intermezzo contains the unusual spelling of gedauenlicre (instead of the more regular gedafenlicre (‘fitting, suitable’)), which seems to employ the Latin spelling for the voiced /v/. How exactly the scribe arrived at these rare spellings is again hard to tell. If they were not already included in his exemplar, then they could reflect his own (dialectal) pronunciation. In the case of the verb endings we could then indeed be looking at traces of anglicised Latin forms, whilst gedauenlic would offer just one acceptable spelling variant that would simply exploit the ambiguous meaning of the letter in Latin in an untypical context. However, the fact that the scribe was struggling with the -spelling of gedauenlic becomes clear in a second instance of this form in the same homily (gedau\f/enlicra, 86r/11) where he corrects it by adding a superscript . In a third instance, he must have spelled the word with the more regular right away (gedafenað, 86v/14). Although this progression indicates the conscious identification and adaptation of a rare spelling to a more familiar one by the Vercelli scribe, other textual cruces in both the Latin and Old English stints suggest that he operated in a relatively mechanical way, as has been pointed out before (cf. Szarmach 1979; Scragg 1992: lxxi–lxxiv). Not all errors may have their origin in the potentially flawed spelling, grammar or material state of the different exemplars that the scribe used; some errors may have been introduced by him independently, since there is no sign that he was able to discover and correct any mistakes in the Latin that he copied. Of course, it is notoriously difficult to distinguish errors simply reproduced by the scribe from those he added himself, but some of the evidence may paint a more precise picture of the scribe’s expertise. One way of gauging the scribe’s attitude towards Latin is to analyse the shape and frequency of abbreviations. These often contain unusual graphemes which allow the Latin text on the page to be optically distinguished from the Old English, at least to some extent, because they reflect the comparatively longer history of codification of Latin as a holy language and its growing scribal economy during

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   277

the early Middle Ages. In the Vercelli Book we can find contractions of the very common nomina sacra (e.g. christus, deus, dominus, ihesus, sanctus etc., cf. Sisam 1976: 25) and their inflected forms, and those of other biblical names (Dauid, Iohannes) or titles (Confessor, apostol). Most of these forms commonly occur in Old English sentences too, so that they are not necessarily highlighting a longer Latin passage to a reader skimming a page in the manuscript. Equally common is the use of a tilde, a superscript stroke most often standing for the multi-minim letter m. More representative of the potential states of the exemplars and of the scribal expertise is the spectrum of the following abbreviations: eet + superscript stroke for esset’ (homily I, 4r/13, ‘Si non esset [hic] malefactor…’) ur + superscript stroke for uester (homily I, 6v/16, ‘Ecce rex uester…’) uob + superscript stroke for uobis (homily II, 11v/3, ‘Dimitt[it]e et dimit[t]etur uobis’) p + top stroke for pr(a)e- (homily V, 26r/7, ‘Euntes [in] uniuersum mundum, predicate’) quo + superscript stroke for quoniam (homily V, 26v/21, ‘Beati pacifici quoniam filii dei uocabuntur.’) q + superscript i for qui (homily V, 27v/2, ‘Ego sum panis uiuus qui de celo descendit’) -b; for -bus (homily V, 28v/7; ‘Pax hominibus bone uoluntatis’) scdm + superscript stroke for secundum (homily VIII, 59r/22, ‘tunc reddet unicuique secundum opera sua’) crossed p for per- (homily XI, 72r/1, ‘Non qui ceperit, [sed] qui perseuerauerit…’) aut + superscript stroke for autem (homily XIII, 75v/14; ‘[Qui] autem perseuerauerit…’) horned h for autem (homily XVI, 85v/24; Respondit autem ihesus et dixit…) p + left slope for pro- (homily XVI, 85v/22; Iohannes autem prohibeba[t] dicens…) Many of these abbreviations were common in Latin manuscripts by the late tenth century (cf. Cappelli 2008). There is no statistically significant tendency of these forms occurring more towards the right side of the text column in the Vercelli Book, which could suggest that most of the abbreviations were copied in more or less the same way the scribe found them in the exemplars, and not newly introduced by him with the line or page length in mind.¹³ In this specific case on fol. 26v/21 (plate 3), for example, the line would have provided affordable space for dissolving the abbreviations of ‘quoniam’ and ‘dei’ in this quotation of Matthew 5:9 (beati pacifici quoniam filii Dei vocabuntur)¹⁴ into five more letters: 13 Cf. Sisam (1976: 23), who suggests that the scribe might only rarely have introduced his own abbreviations for his personal convenience. 14 ‘Blessed are the peacemakers: for they shall be called the children of God’.

278   

   Winfried Rudolf

Plate 3: Vercelli, Biblioteca Capitolare MS CXVII, fol. 26v (detail)

Stronger evidence for the potentially dominating influence of the exemplars on the graphic shape of the copied Latin comes from the beginning of homily XVI (85v/21 & 23; plate 4) in which the horned was used two times in close proximity to abbreviate the word autem (‘however’):

Plate 4: Vercelli, Biblioteca Capitolare MS CXVII, fol. 85v (detail)

This particular abbreviation belongs to the set of notae antiquae, probably of Irish origin, used in the early scriptoria of Anglo-Saxon England (Lindsay 1915: 13, Brown 1993: 186–87). According to scholarly consensus, these contractions became unfashionable during the ninth century (Bischoff 2004: 116–17), but a few later manuscripts still display these graphemes (Rudolf 2013: 197). In some of these tenth- and eleventh-century books, the amount and density of the outdated abbreviations could suggest that very old exemplars participated in the composition of the versions found in them, or at least the versions from which these later manuscripts were copied. This seems particularly plausible for direct biblical quotations which were probably drawn unaltered from authoritative manuscripts, comparable to the Chad Gospels or the Lindisfarne Gospels.¹⁵

15 Lichfield, Cathedral Library, s.n. and London, British Library, MS Cotton Nero D.iv.

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   279

Admittedly, the sample of Latin in Vercelli homily XVI is short, but the double occurrence of the horned -autem as well as the age of this palaeographical marker poses the important question as to the date of this text and the state(s) of its antecedent version(s). Homily XVI is unique in the Old English corpus and any attempt to identify a direct source for it has failed so far. However, it seems clear that the piece must have suffered in transmission, and, Donald Scragg notes, “hypercorrection and misunderstanding of syntax” (1992: 266). It would seem then that the version found in the Vercelli Book was drawn from an exemplar that already included a number of problematic, possibly even illegible readings. In theory these could have occurred due to damage of the text in a potentially old exemplar, which might have stood close to the first vernacular version of this text. Next to the identification of a source of homily XVI, it would therefore be paramount to the study of the origin and composition of the Vercelli Book or its vernacular exemplars, if a gospel book or in fact any other Latin quotation of Matthew 3:13–17 with this exact sequence of abbreviations could be identified.¹⁶ That scribal interference in the Latin passages may have been very limited overall seems clear from the quotation of John 19:36–37 in Vercelli homily I (plate 5):

Plate 5: Vercelli, Biblioteca Capitolare, MS CXVII, fol. 8v (detail)

16 The use of the -shaped autem is relatively rare in the Rushworth Gospels (Tamoto 2013: lxxvi-lxxvii), but a full survey of other surviving materials (cf. Lowe 1934–71, Gneuss 2001) may produce valuable results.

280   

   Winfried Rudolf

The Vulgate would read Os non comminuetis ex eo,¹⁷ which the Vercelli version turns into Omnis enim communio estis ex eo (8v/9). This transformation would certainly require the misunderstanding of the noun Os (‘bone’) as an abbreviation of Omnis (‘all’), a probably abbreviated non (‘not’) as enim (‘namely’), and the segregation of comminuetis (‘shall not be broken’) into the two separate words communio estis (‘you are the community’). The sentence gained, although possible, is certainly implausible within this semantic context. More conclusive evidence can be found two lines later in the inappropriate Videbitur in quem trans felix (8v/11, for Videbunt in quem transfixerunt).¹⁸ The changes necessary to result in these Vercelli readings seem too numerous to assign all of them to the scribe alone. What is more likely in view of these examples is that he copied the quotations from his exemplar largely unaltered. The obvious contradiction between the erroneous Latin clauses and their adjacent translation does not seem to have occurred to the scribe. This may indeed reveal him as a copyist who had no interest in the redaction of the holy language, or, more likely, lacked the necessary competence or authority to do so. Most of the Latin quotations in the Vercelli Book are enclosed by punctuation. The scribe normally used the punctus simplex (e.g. 85v/17, before Uenit) before and after the quoted text, with one exception (85v/18, after ab eo) where he put a punctus versus (plate 6).¹⁹

Plate 6: Vercelli, Biblioteca Capitolare, MS CXVII, fol. 85v (detail)

In a few cases the end of the quotation coincides with the end of a line or the page; no punctuation is used in these cases. Only once does the scribe fail to demarcate an entire Latin tag (7v/3–4, partiti sunt) and only once is the end of a 17 ‘A bone of him shall not be broken’. 18 ‘They shall look on him whom they pierced’. 19 For medieval forms of punctuation and their technical terms and meaning see Parkes (1993: esp. 302–307). Puncti versi are often used in Latin manuscripts to indicate the end of psalm verses or the completion of a sententia. This singular case on fol. 85v/18 may therefore well be taken as a graphic remnant of the Latin exemplar.

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   281

quote not adequately marked in the layout (84v/18, after mundi). The majority of the quotations begin with a capital letter, but this cannot be declared a general principle, and therefore has no exceptional effect on the visual demarcation of the Latin clauses in the layout of the page. What we can sense, nonetheless, is an awareness of the necessity to set both languages apart, if not by means of script, then at least by punctuation.

3 The Latin Quotations in Relation to Their Old English Translations The larger part of homilies preserved in the Vercelli Book are, as far as we can say, straight translations from Latin texts, which show relatively little adaptation of the source material (Scragg 1992: xxxvii). For some pieces, especially when the complete ultimate source material is known (such as for homily III), we can recognize a very direct relation to the quoted text. These Latin source homilies often contain biblical quotations, and it is such quotations exclusively which are retained in Latin by the vernacular versions in the Vercelli Book. This means that the original compilers and translators of the source homilies had already recognized the biblical passages as distinct quotations in the Latin context. This was certainly not too complicated, as all surviving quotations are in oratio recta (cf. Moore 2011: 3) and show corresponding inquit or scripsit phrases before they quote (e.g. 3v/2, þa cwædon hie ealle; plate 7).²⁰

Plate 7: Vercelli, Biblioteca Capitolare, MS CXVII, fol. 3v (detail)

This technique provides the quoted Latin sentences with more significance and authority in the vernacular environment; they stand out simply due to the con-

20 ‘And then they all said’.

282   

   Winfried Rudolf

trast between both languages and therefore lend definitive rhetorical gravity to the word of the Bible. The Old English translation, adhering to this hierarchy, generally follows the Latin text instead of preceding it. Most of the Latin quotations in the Vercelli homilies hardly give more than a single verse or sentence before being translated, a technique which could suggest that glossed Latin manuscripts or even complete bilingual sentence-by-sentence versions were part of the compilation process. However, many of the Latin quotations give incomplete clauses, with a full sentence added only in the ensuing Old English translation. When read aloud, these elliptic Latin clauses may have acted as triggers for the delivery of the complete memorized Latin sentence, but this would depend on the education of the reader. At worst, these remnants might have worked as plain authoritative fillers by means of which the homilist could display (or feign) some command of Latin in front of uneducated audiences. A number of such short tags occur in Homily I, a narrative account of the passion of Christ, which mainly follows John 18 and 19, with frequent references to Matthew 26 and 27 (e.g. Numquid…(26), Ego palam… (33), Si[c] respondi[s]… (41; plate 8), Si male locutus sum… (42–3; plate 8) etc.).²¹

Plate 8: Vercelli, Biblioteca Capitolare, MS CXVII, fol. 3r (detail: Si respondit […] Si male locutus sum)

Almost all the Latin in this homily marks the direct speech of the passion’s dramatis personae (Christ, Pilate, the Jews etc.). All of these turns are preceded by a typical inquit formula in Old English (e.g. þa cwæð he…).²² This bilingual staging 21 ‘Yet…, Do you answer thus…, I spoke…, If I have spoken evil…’. 22 ‘Then he said…’. In one case the plaque on the cross is quoted after a scripsit formula: Wæs ðus awriten: ‘Iesus Nazarenus […]’, 206–7; (It was written thus: ‘Iesusu Nazarenus…’). In two

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   283

of the passion does not only foreground the memorable speeches of its protagonists, it could also suggest a reading of this account in which the different roles were performed by various lectors. The layout and the marginalia of later Old English homilies indicate that such dramatized readings of the passion on Good Friday were possible, as they still are nowadays in certain congregations of the Catholic church.²³ Whilst homily I contains a good number of such tags, the use of Latin for the characters’ direct speech is by no means consistent in this homily, as the following quotation shows: Þa eode he, Pilatus, eft in his domern ⁊ het hine, urne dryhten Crist, to him lædan, ⁊ him to cwæð: ‘Tu es rex Iudeorum? Gesaga me, la, eart ðu Iudea cining?’ Þa cwæð Crist to him: ‘Cwist ðu þæt þines þances þe þe oðre men be me sægdon?’ (I:101–4) [Then Pilate went into his judgement hall and commanded that our Lord Christ be brought to him, and he said to him: ‘Tu es rex Iudeorum? O tell me, are you the king of the Jews?’ Then Christ answered him: ‘Do you speak from your own thoughts or what other men said about me?’]

Here Pilate’s Latin question is extended in Old English by an imperative and apostrophe, rendering the vernacular more effective in oral delivery. Jesus, however, does not answer in Latin in this case. This lack seems to indicate that consistency in Latin was no longer important to this version of the homily, especially not with regard to performance, when an earlier version of the text may well have used the dramatic effect. It is difficult to judge why Latin was used in some cases and omitted in others, but with regard to some homilies the scribe of the Vercelli Book or some intermediary copyist seems to have avoided quotations on purpose, as the parallel versions of Vercelli homilies I and XVIII prove. Oxford, Bodleian Library, MS Bodley 340, a parallel version of the account of the passion in Vercelli homily I, shows the Latin clause Adducunt ergo Iesum ad Pilatum in pretorium (John 18:28; Scragg 1992: 25),²⁴ a sentence which has been omitted in the latter version, possibly because it just gives a descriptive sentence of the pericope, representing neither direct speech nor a prophecy. Vercelli homily XVIII, an account of the life of St

other cases former prophecies are mentioned: ⁊ awitegod wæs be þyssum ylcan ðinge: ‘In siti me[a] potauerunt me […]’ (‘And it was prophecized about this matter: “In my thirst they gave me…”’, 246–7); Swylce eft oðer gewrit cwæð: ‘Uideb[unt] in quem transf[ixerunt] […]’ (‘Another scripture says: “They shall look on him whom they pierced”’, 270). 23 See for example the marginal ‘eli eli lama asabthani’ with neumes in Cambridge, Corpus Christi College, MS 178, p. 219. 24 ‘Then led they Jesus from Caiaphas unto the hall of judgment’.

284   

   Winfried Rudolf

Martin of Tours, avoids any Latin quotations from its sources, despite the fact that a parallel version in the Blickling Manuscript (Princeton, University Library, Scheide Collection, MS 71, item 17) contains ten of these (Morris 1880: 210–27).²⁵ Although Vercelli homily XVIII is lacking some text due to missing leaves in the manuscript after fols. 97 and 100, the overall impression of the surviving text, when compared to its parallel versions, strongly suggests that Vercelli XVIII deliberately excluded any use of Latin. This might have happened because the source texts for the life of Martin, unlike all other Latin quotations in the Vercelli manuscript, were identified as non-biblical, or the composer of this version did not see any point in using Latin considering the education of his audience. Notwithstanding such observations, Scragg’s edition (1992: 291–309) of Vercelli XVIII supplies the missing parts from the Blickling manuscript, including its Latin quotations, therefore quite possibly distorting the idea of the textual state as it may originally have existed in the Vercelli manuscript. In reverse fashion, there are two instances in which the subsequent English translation of the Latin quotation is missing. Homily II contains the sentence Dim[it]te et dimit[t]etur uobis (a rendering of Luke 6:37), which can be found in Alcuin’s Vulgate and was popular in homilies at the time (Scragg 1992: 69), in the following context: Ond þeah ure hwylc wið oðerne gegylte on worde oððe on worce, forbere he him þæt liðelice, þe læs him God þæt yrre witnige. Swa sylfa cwæð: Dim[it]te et dimit[t]etur uobis. (II: 79–81) [And still whosoever becomes guilty against another in word or deed, let him gently refrain, lest God punish him with wrath for that. As he himself says: Dim[it]te et dimit[t]etur uobis. (Forgive and you will be forgiven.)]

While Scragg (1992: 69) considers the lack of a translation after the Latin to be a copying error, without further explanation, it is important to note that the inquit phrase (swa sylfa cwæð)²⁶ includes an anaphoric reference to the preceding clause, in which the gist of the Latin quote is already expressed.²⁷ An extra translation may therefore have become redundant. On the other hand, the sentence is the only use of Latin in Vercelli homily II whose parallel versions in Napier homily XL (cf. Scragg 1992: 53 and 55) include two more Latin quotations with subsequent translations. Again, as in the case of homily XVIII on St Martin, this could point towards a conscious omission of Latin in this Vercelli version. What could be considered as a copying error in Vercelli II, then, is the preservation of the Latin rather than the lack of its English translation. This conclusion is substantiated

25 A new critical edition of this manuscript is a desideratum of Anglo-Saxon studies. 26 ‘As he himself said’. 27 A similar case occurs in Vercelli homily V: 104–5.

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   285

by Vercelli homily XXI, which contains a long parallel of Vercelli II, including the above sentence (⁊ þeah ure hwylc wið oðerne agylte … þæt yrre God witnie, XXI: 229–31), but omits the Latin (and its potential translation) entirely. The Vercelli homilies further incorporate a few Latin quotations which show notable deviations in their Old English translation and so permit a tentative evaluation of parts of the compositional process. Two further cases may illustrate this issue. Homily I shows the following passage: Sceolde se witedom beon gefylled se þe ær in þam sealme awriten wæs ⁊ awitgod, ⁊ se halga gast dihtode of Cristes sylfes muðe ⁊ þus cwæð: ‘Partiti sunt … Hie todældon him mine hrægl, ⁊ hie ofer minne gegyrlan tanas sendon and tohluton.’ (I: 222–24) [So the prophecy should be fulfilled, that which was written in the psalm and foretold, and which the Holy Spirit dictated from Christ’s own mouth, thus saying: ‘Partiti sunt … They parted my garments among them, and upon my clothes did they cast lots (lit. sent twigs and cast lots)’]

Scragg (1992: 46) observes here that the Latin tag was taken from John 19:24, but that the translation follows Matthew 27:35, which quotes Psalm 21:19 (Diuiserunt sibi uestimenta mea et super uestem meam miserunt sortem).²⁸ The difference between both gospels is only slight – John has partiti sunt vs. Matthew’s diuiserunt, the sentences otherwise being identical. No doubt both gospels were used in the compilation of this homily, but the context of this passage shows material primarily drawn from John. If translating partiti sunt as hie todældon (‘they divided’) was really due to a change of source, then the compiler could even have consulted the psalm directly, because the preceding Old English sentence betrays this source, while neither of the two gospels makes reference to the psalms in this context. However, it is important to note here that the Lindisfarne gospels show the Old English double gloss gedaelde uoeron vel todældon (technically the passive ‘they have been divided’ or the active ‘they divided’) in the respective passage from John (ed. Kendrick 1956: I, 254v), which acknowledges partiti sunt in its quality as a possible deponens. The confusion in Vercelli homily I could therefore derive from this alternative in a similar exemplar used during compilation. An opposite example can be found in Vercelli homily III, a penitential homily for Lent, whose Latin source has been identified in the Homiliary of St-Père-deChartres,²⁹ where we read:

28 ‘They part my garments among them, and cast lots upon my vesture’. 29 For a description and partial edition of this important homiletic collection see Cross (1987).

286   

   Winfried Rudolf

Be þære wæccan eac swylce se hælend þara tolysendra mod aweahte, þus cweðende: ‘Beatus ille seruus quem … Eadige beoð þa þeowas [þa þe] þonne se hlaford cym[ð] he hie wæccende gemet. To soðan ic eow secge, ofer eall [h]is god [he hie] geset.’ (III: 70–4) [Concerning the watch the Saviour in the same manner awakened the minds of the desolate, thus speaking: ‘Beatus ille seruus quem… Happy the servants whom the Lord finds watching when he comes. Truly I say to you, he will put them in charge of all his possessions.’]

Although the homily quotes the Latin singular (beatus ille servus), it translates to plural (eadige beoð þa þeowas) in Old English. This plural would be the correct translation of Luke 12:37 (Beati serui illi…inuenerit uigilantes.), which can indeed be found in manuscripts containing the potential Latin source (e.g. Cambridge, Pembroke College, MS 25, item 22). However, Luke 12:37 parallels Matthew 24:46, which is singular, and in the Latin source Matthew 24:47 directly follows (Amen, dico uobis, super omina bona sua constituet eos),³⁰ which is adequately translated into Old English by Vercelli III (To soðan… geset, cf. Scragg 1992: 77 and 85). This seems to suggest that the Latin tag in Vercelli III derives from the gospel of Matthew rather than from Luke. This muddling up of sources could have occurred in the course of a reincorporation of Latin tags after an earlier omission – the homily holds only two Latin quotes in short succession – during which a reviser turned to Matthew, inspired by the second sentence of the Old English translation.³¹ Alternatively, as in the previous example, an erroneous gloss in an exemplar could also have been responsible, or the sentence from Matthew was quoted from memory and then confused with Luke because of its unmistakable similarity to the latter. These examples elucidate how the wording of individual Vercelli homilies can reveal very specific relations to potential source manuscripts, most particularly versions of the gospels and their varying glosses. It is evident, for example, that the gospel manuscript which contributed to the composition of the ancestor of Vercelli homily IV cannot be identical with that used for homily XV, as the quotation of Matthew 25:41 in both homilies and their varying translations demonstrate. Vercelli IV reads: þonne eft cwyð se dema to þam synfullum sawlum: ‘Discedite a me, maledicti, in ignem æternum…’. He cwið: ‘Gewitaþ ge, awyrgede, on þæt ece fyr, þæt gegearwode min fæder diofle 7 his englum’. (IV: 304–307)

30 ‘Verily I say unto you, that he shall make him ruler over all his goods’. 31 Oxford, Bodleian Library, MS Bodley 340, item 19 corrects the Latin, while MS Bodley 343, item 28 preserves the error.

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   287

[Thereafter the judge says to the sinful souls: ‘Discedite a me, maledicti, in ignem æternum…’. He says: ‘Depart from me, you accursed ones, into the eternal fire which my father prepared for the devil and his angels’.]

Here the phrase þæt gegearwode min fæder (‘that my father prepared’) corresponds to the variant quem praeparauit pater meus which can be found in the late tenth-century Canterbury Bible (London, British Library, Royal MS 1.E.viiviii; Scragg 1992: 107) and might, perhaps, provide a further hint at the possible Kentish origin of some of the Vercelli homilies (Scragg 2010). In the repetition of the quote in homily XV (179–184), however, we find the Vulgate’s qui praeparatus est diabulo (‘that has been prepared for the devil’), correctly translated there as ðe ðam diofle wæs geearwod. Despite this, other non-Vulgate readings (e.g. omnes potestas in homily V: 42–43 vs. omnis potestas in Matthew 28:18; Scragg 1992: 122) seem to connect the homilies with Insular gospel books.

4 Conclusion Even a preliminary examination of the Latin quotations in the Vercelli homilies, such as this, yields some important, if only sketchy information on the composition and transmission of these texts. We may confirm from the varying degrees of frequency, linguistic accuracy, and handwritten representation of the Latin that the pieces were drawn from various exemplar manuscripts of mixed quality and age, arguably corresponding to some extent to the grouping which Scragg (1993, 2009) suggested on the basis of his analysis of Old English vocabulary and its spelling variants. However, the original composition of the single homilies could, at least in theory, predate these homogenous groups of versions by a very long time. The original composition may sometimes have started with the compilation of purely Latin versions which made use of a range of sources, probably preserved in manuscripts which predated the tenth century. These pieces were then translated, occasionally retaining the important biblical quotations in Latin. The reason for this may have been twofold: apart from the general authority that the Latin would lend an Old English homily, it could also have been essential at the initial stage of composition because the vernacular translations of the biblical verses might have required that the Latin precede them for reasons of verification. Gradually then, due to the insufficient command of Latin or general ignorance of copyists and preachers before the Benedictine Reform, these quotations then deteriorated to authoritative rudiments, in some cases with no value for perfor-

288   

   Winfried Rudolf

mance, working only on a graphic level. It is therefore hardly surprising that they were even dropped entirely in certain instances. In some cases Old English versions or parts of them may have been composed directly from glossed bilingual manuscripts in which the Latin predated the vernacular gloss by several centuries. Finally, some evidence suggests the reincorporation of Latin quotations into versions that had been systematically anglicised in the process of transmission, possibly through comparison of older and younger versions of the purportedly ‘same’³² homily. More evidence is needed to establish if such re-Latinization was primarily the result of the Benedictine Reform and could ipso facto be considered as an attempt to teach priests the basic and most important verses of the New Testament in Latin. Some of the palaeographical and linguistic evidence suggests that Insular gospel books written before the year 900 AD took part in the composition of antecedent versions of the Vercelli homilies. Of course, these books may have enjoyed exceptional textual and material authority and – regardless of any vernacular glosses – could have had an impact on the compilation of both Latin and Old English homilies at any given point in time. Nevertheless, the concision as well as the deficient state of numerous Latin sentences in the Vercelli Book indicate at least several textual predecessors so that the original composition of some of these homilies may well date to the first half of the tenth century, if not before. Yet even if these thoughts must remain merely conjectural, the evidence produced by this preliminary study encourages a general investigation of all Latin quotations in Old English homilies. The palaeographical and linguistic data collected from such a survey would certainly enable us to identify new codicological cues in specific Latin gospel books or even homiliaries produced in England which may prove to be invaluable for the contextualization and tracing of the possible origins of single Old English homilies and their manuscript witnesses. They may also be able to gauge to what extent and in which ways the quoting of Latin became more standardized in form and contents during the eleventh and twelfth centuries.

References Augustine. 1995. De Doctrina Christiana. Roger P.H. Green (ed.) (transl.). Oxford Early Christian Texts. Oxford: Clarendon Press. Barré, Henri. 1962. Les homéliaires Carolingiens de l’école d’Auxerre. Studi e testi 225. Città del Vaticano: Bibliotheca apostolica Vaticana.

32 It is problematic to use such a label with respect to a genre marked by frequent use and reuse of texts which leads to comprehensive textual intervention.

Quoting and Translating Latin in the Old English Homilies of the Vercelli Book   

   289

Berschin, Walter. 1980. Griechisch-lateinisches Mittelalter: von Hieronymus zu Nikolaus von Kues. Tübingen: Francke. Biblia sacra iuxta vulgatam versionem. 1969. Robert Weber (ed.). Vol. 2. Stuttgart: Württembergische Bibelanstalt. Bischoff, Bernhard. 2004. Paläographie des römischen Altertums und des abendländischen Mittelalters. Vol. 3. Berlin: Schmidt. Brown, T. Julian. 1993. A Palaeographer’s View: The Selected Writings of Julian Brown. Janet Bately, Michelle P. Brown and Jane Roberts (eds). London: Harvey Miller. Cappelli, Adriano. 2008. Lexicon Abbreviaturarum. Dizionario di Abbreviature Latine ed Italiane. Vol. 6. Milan: Hoepli. Clayton, Mary. 1985. “Homiliaries and preaching in Anglo-Saxon England”, Peritia 4. 207–42. Clemoes, Peter (ed). 1997. Ælfric’s Catholic Homilies. The First Series Text. EETS s.s. 17. London: Oxford University Press. Cross, James E. 1987. Cambridge, Pembroke College MS 25: A Carolingian Sermonary Used by Anglo-Saxon Preachers. King’s College London Medieval Studies 1. London: King’s College. Dumville, David N. 1993. English Caroline Minuscule and Monastic History: Studies in Benedictinism, A.D. 950–1030. Cambridge: Boydell and Brewer. The English Bible. King James Version. 2012. Herbert Marks and Gerald Hammond (eds). New York: Norton. Fontes Anglo-Saxonici: A Register of Written Sources Used by Anglo-Saxon Authors. Oxford: Fontes Anglo-Saxonici Project. 2013. CD-ROM, 1.1. Förster, Max. 1913. “Der Vercelli-Codex CXVII nebst Abdruck einiger altenglischer Homilien der Handschrift”, in: Ferdinand Holthausen and Heinrich Spies (eds). Festschrift für Lorenz Morsbach: dargebracht von Freunden und Schülern. Studien zur Englischen Philologie 50. Halle an der Saale: Niemeyer, 21–79. Gatch, Milton McC. 1989. “The unknowable audience of the Blickling Homilies”. ASE 18. 99–115. Gretsch, Mechthild. 2004. “The Taunton Fragment: a new text from Anglo-Saxon England”. ASE 33. 145–93. Gneuss, Helmut. 2001. “Handlist of Anglo-Saxon manuscripts: a list of manuscripts and manuscript fragments written or owned in England up to 1100”. Medieval and Renaissance Texts and Studies 241. Tempe: Arizona Centre for Medieval and Renaissance Studies. Halsall, Maureen. 1969. “Vercelli and the Vercelli Book”. PMLA 84, 1545–50. The Lindisfarne Gospels. Evangeliorum quattuor Codex Lindisfarnensis. 1956–1960. Thomas D. Kendrick et al. (eds). Olten: Urs-Graf. Lindsay, Wallace M. 1915. Notae Latinae: An Account of Abbreviation in Latin MSS. of the Early Minuscule Period (c. 700–850). Cambridge: Cambridge University Press. Lowe, Ernest A. (ed). 1934–1971. Codices Latini Antiquiores: A Palaeographical Guide to Latin Manuscripts Prior to the Ninth Century, 12 Vol. Oxford: Clarendon Press. McBrine, Patrick. 2009. “The journey motif in the poems of the Vercelli Book”, in: Samantha Zacher and Andy Orchard (eds). New Readings in the Vercelli Book. Toronto Anglo-Saxon Series 4. Toronto: University of Toronto Press, 298–317. Moore, Colette. 2011. Quoting Speech in Early English. Cambridge: Cambridge University Press. Morris, Richard (ed). 1880. The Blickling Homilies of the Tenth Century. EETS o.s. 73. London: Trübner. O’Carragain, Eamonn A. M. 1975. “The Vercelli Book as an Ascetic Florilegium”. PhD Diss., Queen’s University of Belfast.

290   

   Winfried Rudolf

Parkes, Malcolm. 1993. Pause and Effect: An Introduction to the History of Punctuation in the West. Berkeley: University of California Press. Pope, John C. (ed). 1967. Homilies of Ælfric. A Supplementary Collection. Vol. 1. EETS o.s. 259. London: Oxford University Press. Remley, Paul G. 2009. “The Vercelli Book and Its Texts: A Guide to Scholarship”, in: Samantha Zacher and Andy Orchard (eds). New Readings in the Vercelli Book. Toronto Anglo-Saxon Series 4. Toronto: University of Toronto Press, 318–415. Rudolf, Winfried. 2011. “The Homiliary of Angers in Tenth-century England”, ASE 39.163–192. Rudolf, Winfried. 2013. “The Selection and Compilation of the Verba Seniorum in Worcester, Cathedral Library, MS F.48”, in: Loredana Lazzari and Claudia di Sciacca (eds). Adopting and Adapting Saints’ Lives into Old English Prose (c. 950- 1150). Textes et Études du Moyen Âge of the Fédération Internationale des Instituts d’Études Médiévales. Turnhout: Brepols, 183–226. Scragg, Donald G. 1973. “The compilation of the Vercelli Book”, ASE 2, 189–207. Scragg, Donald G. (ed). 1992. The Vercelli Homilies and Related Texts. EETS o.s. 300. Oxford: Oxford University Press. Scragg, Donald G. 2008. “The Vercelli Homilies and Kent”, in: Virginia Blanton and Helene Scheck (eds). Intertexts: Studies in Anglo-Saxon Culture Presented to Paul Szarmach. Tempe: Arizona State University, Arizona Center for Medieval and Renaissance Studies, 369–80. Scragg, Donald G. 2009. “Studies in the language of copyists of the Vercelli Homilies”, in: Samantha Zacher and Andy Orchard (eds). New Readings in the Vercelli Book. Toronto Anglo-Saxon Series 4. Toronto: University of Toronto Press, 41–61. Sisam, Celia (ed). 1976. The Vercelli Book: A Late Tenth-Century Manuscript Containing Prose and Verse, Vercelli Biblioteca Capitolare CXVII. EEMF 19. Copenhagen: Rosenkilde and Bagger. Szarmach, Paul E. 1979. “The scribe of the Vercelli Book”, SN 51, 179–88. Treharne, Elaine. 2007. “The form and function of the Vercelli Book”, in: Alastaire Minnis and Jane Roberts (eds). Text, Image, Interpretation: Studies in Anglo-Saxon Literature and Its Insular Context in Honour of Éamonn Ó Carragán. Turnhout: Brepols, 253–66. Tristram, Hildegard L.C. 1995. Early Insular Preaching: Verbal Artistry and Method of Composition. Österreichische Akademie der Wissenschaften, Philosophisch-historische Klasse, Sitzungsberichte 623. Veröffentlichungen der Keltischen Kommission 11. Vienna: Verlag der österreichischen Akademie der Wissenschaften. Wright, Charles D. 2007. “Old English Homilies and Latin sources”, in: Aaron J. Kleist (ed.). The Old English Homily: Precedent, Practice and Appropriation. Studies in the Early Middle Ages 17. Turnhout: Brepols, 15–66. Zacher, Samantha. 2009. Preaching the Converted: The Style and Rhetoric of the Vercelli Book Homilies. Toronto Anglo-Saxon Series 1. Toronto: University of Toronto Press. Zacher, Samantha. 2009. “The Source of Vercelli VII: An Address to Women.” in: Samantha Zacher and Andy Orchard (eds). New Readings in the Vercelli Book. Toronto Anglo-Saxon Series 4. Toronto: University of Toronto Press, 98–150.

Monika Kirner-Ludwig and Iris Zimmermann

12 Quoting and Plagiarising – Concepts of Both Now and Then? Abstract: Our paper seeks to carve out facets of the medieval concepts of quoting, citing and referring and will, in doing so, also consider the question of to what extent there was a medieval consciousness of originality and plagiarism at all. By presenting case studies from both Middle High German as well as texts produced in medieval England (both in Anglo-Latin and Middle English), we are not only choosing a diachronic but also a crosslinguistic perspective: we will argue that the acts of plagiarising and quoting are not inventions of our modern time, but were deeply enrooted in the minds of medieval text producers – seemingly as a concept which was even handed down from classical times in the first place. Taking into account the general unfixedness of texts in the Middle Ages, we will discuss the high potential of intentions and functions behind the act of quoting in medieval texts in order to shed light on the pragmatic reverse of the medal. Keywords: plagiarism, citation, reference, diachronic, creator, scribe, prologoi, praefatio

1 Terminological and Conceptual Framing of the Objective Zitat, Entlehnung, Topos und Formular sind legitime Formen der Aussage in einem Zeitalter, dem Autorität alles, Originalität nichts bedeutete. Das heißt gewiß Begrenzung der geistigen Individualität, nicht aber Auslöschung. (Beumann 1962: 13f.)¹

The English lexico-semantic inventory contains a number of terms belonging to the field of intertextual reference: The most prominent and frequently occurring hyponymic proponents in this umbrella category are citations and quotations, which primarily denote explicit ways of establishing reference in its widest sense. In modern academic contexts, citing and quoting are applied by authors in order to e.g. sustain their argument, provide further reading material or to simply let

1 [Quote, loan, topos and formula – these all are legitimate notions pertaining to an era which was all about authority, not originality. This certainly meant a limitation of intellectual individuality, however not its obliteration.].

292   

   Monika Kirner-Ludwig and Iris Zimmermann

the reader know that they are acquainted with recent or generally acknowledged authorities and previously conducted research. Today’s opportunities to research and obtain information have certainly changed the way people refer to their sources of information: The continuously growing network of data which is constantly fed into the World Wide Web as well as the vagueness of intellectual properties as a non-neglectable result thereof may be regarded as both a blessing and a curse. As the evolution of an omnia-omnibus attitude towards data that is both freely accessible and can be copy-and-pasted effortlessly has been facilitated if not nurtured by various offers on the internet, it has also led to misuse and even unethical self-ascription of findings acquired by others in the first place. This phenomenon is not new and has been a bone of contention from time immemorial, generally referred to by the notion of plagiarism. By now it has long outgrown its youthful status of a peccadillo and is, due to a growing fixation on the authority of texts and their producers – a circumstance fostered by the invention of printing and the mass-distribution of texts – increasingly considered an offense against righteousness and respect for honest work, at least in the western hemisphere. Such recent associations with plagiarism, however, clearly bear the impress of a present-day point of view, which must be revisited and possibly amended when it comes to medieval references to authorities with regard to motives and modi operandi. Starting from this line of thought, we will attend to two major types of reference: For one thing we speak of citing and quoting, these being fairly clear and comprehensible; second, we refer to ambiguous references or allusions.The latter are distinguished from the first two with regard to their generally high degree of implicitness, regularly rendering a reference vague enough to potentially make it untraceable and inevitably incomprehensible to a recipient (cf. Figure 1).² Plagiarism then will be considered the counter-phenomenon to quoting and citing and will be under focus in Section 4. Figure 1 serves as a visualisation and explanation of the key terminology as well as the relational concepts we will base our observations on.

2 Whether or not a reference is perceived as ambiguous or opaque will always depend on the degree of knowledge a recipient shares with the text producer and their cognitive potential in comprehending the latter’s point.

Quoting and Plagiarising – Concepts of Both Now and Then?   

reference

citations

exact and thus traceable references

vague and thus untraceable references

quotations

direct

indirect

   293

plagiarism

other intertextual references /allusions

ambiguous references /allusions

opaque references /allusions

Figure 1: The relational concept of reference pursued in this paper

As our focus is a diachronic one concerned with the conceptual and semantic shifts identifiable in text reception, elaboration and usage, we will dedicate the whole of Section 2 to a discussion on the semantic etymology of the conventional concepts and understandings of quoting, citing and referring. Since all of these concepts and their application solely depend on the category of ‘text’ and its perception as e.g. fixed vs. unfixed, private vs. public etc., we will discuss the major divergences between today’s conception of ‘text’ from the one encountered in the medieval frame in Section 3. To sustain our arguments, a selection of Middle High German and Middle English examples from the period between 13th to the 15th century will be provided and arranged according to their degree of referential vagueness. Section 4 will focus on the phenomenon of plagiarism, again starting with an etymological sketch of the semantic development of the term itself. This will be followed by texts from the German as well as the English Middle Ages (with a focus on the latter), ranging from the 8th to the 15th century, which may reveal the extent of a text recipient’s awareness of another text producer’s intellectual property and the evolving categorisation of right vs. wrong in this particular respect. With Section 5 we will provide a critical summary to contrast both diachronic and synchronic perspectives on the phenomena of reference and plagiarism, along with the perceived awareness of these categories.

294   

   Monika Kirner-Ludwig and Iris Zimmermann

2 The Act of Referring: A Brief Sketch of the Etymological and Conceptual History of ModE Reference, Citation and Quotation Let us first try to distinguish the terms quotation, citation and reference semantically.³ The recent online version of the Oxford English Dictionary (henceforth OEDo, URL 1) provides numerous text samples to sketch a fairly continuous line of development,⁴ which is selectively displayed in Table 1.⁵ Tables 2 and 3 contrastively display the selective semantic evolution of citation and quotation.⁶ Table 1: Relevant semantic stages of reference n.

Earliest Primary sense attested year of attestation 1579

the action or an act of sending a matter to an authority for decision or consideration

1581

relation or regard to a thing or person

1591

a mention of or allusion to a person or thing

1591

the action or process of mentioning or alluding to someone or something

1599

a mention or citation of a source of information in a book or article

a1627

an authority or standard referred to

1774

the action of consulting of a source of information in order to ascertain something

1799

the action or an act of referring one person to another for information or testimonial

1774

book of reference n.

1792

work of reference n.

1883

the relation between a word or expression and that which it denotes (chiefly Philos. and Linguistics); the entity or entities denoted by a word or expression, a referent (freq. contrasted with sense)

3 The Modern German (ModG) inventory, for instance, does not commonly make a distinction parallel to the one in Modern English (ModE). The ModE noun quotation and the verb quote will generally be translated as ModG Zitat/zitieren, whereas a citation of an authority equals ModG Verweis (~ ModE reference). 4 For the notional distinction between ‘continuous’ and ‘discontinuous’ etymological developments, see Durkin 2011. 5 The bold script in Table 1 for the year 1599 will be explained below. 6 See the OEDo for respective quotations.

Quoting and Plagiarising – Concepts of Both Now and Then?   

   295

Table 2: Relevant semantic stages of citation n.

Earliest year of attestation

Primary sense attested

1297

a citing or summoning to a court of justice, a summons

a1387 ↓ 1520 1548

concr. a passage cited, a quotation

1651

the action of citing or quoting any words or written passage, quotation

Table 3: Relevant semantic stages of quotation n. Earliest year Primary sense attested of attestation c1485

a numbering, a number Obs.

1532

a reference (usually in a margin) to a passage of text by page, chapter Obs.

1592

an observation; a matter noted. Obs.

1618

a passage quoted from a book, speech, or other source

1646

the action or an act of quoting

1683

typogr. More fully quotation quadrat. A large (usually hollow) quadrat used for filling blanks in letterpress printing. Now hist.

1715

quotation mark

1890

a short musical passage or visual image taken from one piece of music or work of art and used in another.

As may be seen from Tables 1, 2 and 3, reference n. is the youngest term to have been introduced into the English inventory as a morphological adoption of pclassL referentia (< L re-ferre ‘to report’ [both literally and metaphorically < ‘to bring back’]) during the 16th century. For 1599 the OEDo identifies the ModE sense for the first time.⁷ The OEDo references suggest that its denotation has been fairly stable for over 400 years, displaying a minimally broken, continuous line of development between a1627⁸ and 1774. 7 Thus presented in bold script, as in Tables 2 and 3. 8 The preceding years is used in the sense ‘ante’ and following both MEDo and OEDo, i.e. ‘before the respective year, but probably not earlier than Y-25 years’.

296   

   Monika Kirner-Ludwig and Iris Zimmermann

The noun quotation is a borrowed lexeme from Latin, viz. from pclassL quotatio-n- (< quotat-, the past participial stem of quotare v. ‘to count; to recount’) as well. The extensional reference to a ‘passage quoted from a written or spoken source’ does apparently not occur before the 17th century, some 50 years later than in the noun citation, which was borrowed from French citation n. (< ad. L. citatio-n- infl. n. < citare v. ‘to summon’). While the lexical items were ultimately all derived from the Latin lexical inventory, it is remarkable that citation and quotation, which had been adopted earlier than reference, were introduced bearing distinct semantic features that had coalesced by c.1600. The distinctive feature [+pertaining to the law frame] in citation had gradually been weakened from essential to occasional by the middle of the 16th century, shifting from the ‘act of summoning someone (to court)’ to the ‘written form of summons’ and thence developing the common extensional reference to ‘any piece of written or spoken speech’. Thereby citation reached a semantic makeup fairly even to quotation, which was originally introduced into English in the sense of ‘numbering; calculation’ in the middle of the 15th century. In 1532 we have an example of an occurrence which is synonymous to today’s denotation of citation, viz. ‘a reference to a passage of text by page, chapter’, however with the specification [+usually in a margin]. This sense, though, became obsolete again after 1700, most likely due to citation beginning to express this sense, too, from the middle of the 17th century, and thus possibly to avoid total synonymy. As an interim conclusion, we can state a specific kind of cross-semantic shift within the nouns citation and quotation: While citation once used to refer to the content of a text passage, its extension was narrowed to a mere elliptic pointing to the site where a specific passage or attestation could be found. Quite in contrast, it is the concept of quotation that entails the textual contents one refers to in today’s use. The term reference, although the youngest of these three notions, has displayed the most flexible and varying semantic behaviour since its first appearances during the 16th century. The extensional range has been rather wide from its first attestations onwards: Whereas today the primary meanings of quotation and citation can generally be considered fairly clear-cut and distinct, reference remains the most generally applicable notion of these three, capable of pointing allusive indications to any piece of written or spoken speech just as well as to a person or an object. This semantic broadening is attested for by the end of the 16th century and has been retained up until today. What all three notions still have in common, in the end, is of course the capability of linking intertextual sites and data.

Quoting and Plagiarising – Concepts of Both Now and Then?   

   297

3 Quoting in the Middle Ages 3.1 Uncovering Source and Course of a Medieval Text – Applying Some Variables According to its entry in the OEDo (cf. URL 1), the ModE term text goes back to the medieval Latin textus (< texere ‘to weave’).⁹ What exactly was conveyed by the term in medieval times can only be guessed at, but certainly the medieval understanding of the concept text must not be considered congruent with its meaning today (cf. Bein 2001: 15–36; Müller 1996: xv; Wachinger 2011: 1–23; Wolf 2002: 180). On the contrary, we insinuate a general unfixedness and thus a high potential of malleability in any medieval text that was passed down, which necessarily impinges on the conceptual understanding of both reference and plagiarism. Figure 2 provides a visualisation of the intricate complexities that arise from the endeavour of sketching the changes a medieval text likely underwent during the process of being passed down in history. Stage 1¹⁰ portrays an incorrupt version of the medieval text as it is still directly linked to its producer (i.e. creator/author):¹¹ it is yet in a status in which it is neither expressed in writing nor speech. As soon, however, as it is brought to the public in oral or written form, the (internal) dynamics of a medieval text are launched by being uttered, received, processed and passed on. In order to ensure a text’s outreach, a text producer would e.g. have copies produced himself,¹² let alone that any text could potentially be reproduced by others without him having any say or knowledge of it. Such reproductions – each and every one biased by the idiosyncrasies of their makers¹³ – would then diverge with regard to e.g. regional, temporal, social and political factors.¹⁴

9 How well this literally fits the structure of a medieval text will be shown in Figure 2. 10 Stages are separated by horizontal dashed lines and numbered vertically downwards from 1 to x. 11 We decided to give preference to the term ‘creator’ over ‘author’ bearing in mind that ‘author’ would always convey the modern idea of originality, which does not necessarily apply to medieval texts, as these often were mere translations or transformations of existing pretexts. Those ‘source’- or ‘influencing’ texts are referred to in Figure 2 as ‘source’ and ‘influence’. 12 Here and in the following, we decided in favour of the male personal pronoun to refer to (re-) producers of a medieval text. We did so with regard to the fact that despite some few female exceptions – such as Marie de France – most medieval text (re-)producers are known to be male. 13 t1, t2, … tx; see Figure 2. 14 Such variation, for one thing, can be observed on the phonological/graphemical level and, secondly, on the morphological/inflectional level when it comes to lexical choices (variance) and the rebuilding of greater structures, such as syntax or stanza (mouvance).

298   

   Monika Kirner-Ludwig and Iris Zimmermann

source influence producer (creator/ author)

stage 1

text t1, t2, … tx = recipient-oriented variants authorized by the creator/ author

(1) creator/ author mentioned? (2) attribution to imitator‘s œuvre? (3) misattribution?

text/ s as an autograph (axtx) corruption of the text at that stage already possible: slips

text/ s written down by commissioned scribe (sxtx) corruption of the text at that stage already possible: creative scribe, mistakes due to mishearings, slips

orally performed text/ s (oxtx)

copy/ ies (cxaxtx) / (cxsxtx) further corruption: [see before] plus censorship, varieties (political, regional, social, temporal)

oral tradition (by imitator/ s) (oxaxtx) / (oxsxtx) / (oxoxtx) further corruption: [see before] plus censorship, varieties (political, regional, social, temporal)

notes (nxoxtx) possible corruption of the text: slips, mishearings, bad memory, misunderstandings evoked by regional varieties

copy/ ies (cxcxaxtx) / (cxcxsxtx) / (cxnxoxtx)

oral tradition (oxcxaxtx) / (oxcxsxtx) / (oxoxaxtx) / (oxoxosxtx) / oxoxoxtx) / (oxnxoxtx)

notes (nxoxaxtx) / nxoxsxtx) / (nxoxoxtx)

stage 2

stage 3

stage 4

(…) textual version/ s of today

stage x

Figure 2: Potential developmental stages of a medieval text being passed down

Stage 2 focuses on three ways of passing on texts that are still creator-related in their regional/ temporal/ social/ political variants (tx).¹⁵ Option one (stage 2: box to the left) depicts the text creator as a learned scribe, able to write down his texts himself (autograph). While option two (stage 2: box in the middle) includes the possibility of a creator commissioning a scribe to put down the words dictated to him. The third option (stage 2: box to the right) possible, a text creator uttering the words himself before an audience, rounds off stage 2. Only in the latter case would a text be potentially left untouched by any corruptions inflicted by others than the producer (cf. Figure 2). A text reaching stage 3 has already been completely disconnected from its creator’s version and must be regarded as being handed down without his supervision or opportunity for intervention, as the inter-mediator(s) (e.g. scribes or

15 On non-equal estimation of textual variants see e.g. Wolf (2002: 184f.).

Quoting and Plagiarising – Concepts of Both Now and Then?   

   299

imitators) pass(es) on the text.¹⁶ Ideally, copies would directly relate to an autograph master copy, especially with respect to possible deviations from a text’s original appearance (cf. corruptions in Figure 2). After all, the level of corruption in a text can be estimated as slightly higher if it has been linked to a master copy written by a commissioned scribe.¹⁷ If it is the case that a text entered the oral tradition – either once more due to the creator himself¹⁸ or to imitators of his works (cf. arrows in Figure 2), this situation might have been influenced by all the three options in stage 2 (oxaxtx, oxsxtx, oxoxtx). The corrupted text – and thus all of its corrupted copies – would consequently reproduce further corrupted texts.¹⁹ Stage 4 serves as a logical continuation of stage 3, which potentially passes over into further stages of corruption (cf. stage x (…)). In a nutshell, Figure 2 suggests that the trajectory of a medieval text (tx) – from a modern-day viewpoint – is almost irretraceable due to the undiscernable number of variables involved in a text’s development. Taking into account that the notion quotation can only be regarded as applicable or useful when referring to a text that hands down its producer’s name (citation) and, as a surplus, can be classified as relatively stable and fixed, we come to an understanding of the hindrances or limitations we quoters face with in medieval times.

16 Various alternatives of textual attribution are displayed in the box far left of Figure 2: some texts would give a creator’s name, some would not. The case that the work is passed on – be it willfully or not – under an imitator’s name or misattributed without any obvious reason, is not unusual either. 17 The complex formulae bracketed in stage 3 need explaining. When looking at variant cxaxtx the following must be kept in mind: one variant of an autograph (ax) must be estimated as having originated from one of x textual variants (tx). This axtx is then handed down as one copy out of x (cx) (again variants), potentially produced from the master copy. In stage 3, the text and its variants have therefore spread like the branches of a tree; already at this stage, trying to trace a text back to its starting point must be deemed impossible. 18 This is the best-case-scenario, as textual changes would be idiosyncratically cohesive. However, it must also be regarded as the most improbable case: not only because our figure’s stages are ordered chronologically and, with respect to the average medieval life expectancy, our creator might have been dead at that time, but also because there is no sense in a text creator consulting his or others’ handwritings in order to be able to perform his own work. 19 Just like copies, notes and annotations to a text may be regarded as the least corrupted versions of an original text. Supposedly, notes would be taken e.g. while listening to an oral performance by the creator (stage 2): listening and writing simultaneously in possibly bad writing posture and surrounded by a rather high level of noise could have led to greater or lesser deviations from the text performed (nxoxtx).

300   

   Monika Kirner-Ludwig and Iris Zimmermann

3.2 Quoting in Medieval Texts: Cultural and Historical Background Assumptions So far, we have dealt with some of the risks of falling for anachronistic classifications and conceptualisations with concern to the phenomenon of texts then and now²⁰ – and, in consequence, of references as well. Unlike today, a medieval text was regarded and treated as common property and was passed on like mere modelling clay to be shaped and reshaped by an indeterminable number of hands. Attributional separation from a text’s creator as well as corruptional deviations from the original version therefore may be regarded as a given (cf. Figure 2: stage 2). Any second or further user of a text – more specifically a quoter – therefore became a producer himself whilst revising the text that had been passed down to him. Scribes, copyists and imitators were free to adjust the text they produced to their tastes and needs (compare Figure 2: creative scribe/s). In other words, they were basically free to add, leave out or change anything according to their subjective estimation. All three kinds of producers were capable of adding or deleting quotes in unspecified stages of a text. When we set out to search for medieval cases of quotations, we are very likely to find examples in the religious, political and legal spheres. Sometimes even texts belonging to courtly literature contain respective references (as will be shown in Section 3.3).²¹ In terms of classification, a medieval quote is generally limited in its referential preciseness, as compared to today’s rather elaborate quotes. This is, for one thing, because in medieval times only first names (and cognomina) where used, not surnames (cf. Sections 3.2, and 3.3). What is more, the source of a text was often unknown as well. If the manuscript was innominate, one could only refer to the library one found it in or to the manuscript’s owner. Not even a work’s title was always easy at hand as most texts were left untitled or their titles changed in their different textual variants. Still, referring back to an authority by citing or quoting them, appears to have been a common practice in the Middle Ages as it had also been in Greek and Roman antiquity. Quoting not only from Greek and Latin texts, but also from recent ones, proved a producer’s knowledge and established the reliability and trustworthiness of his work. Moreover, it gave a text user/producer the foundation for construing a professional identity by positioning themselves in a line with former and maybe more advanced professionals in their trade. 20 Also cf. de Beaugrande/Dressler’s definition of text (URL 2). 21 Generally, one would tend to call this phenomenon intertextuality or cross-reference (cf. Section 1) rather than quotation in this case.

Quoting and Plagiarising – Concepts of Both Now and Then?   

   301

Nevertheless, explicit quotational or citational references are relatively small in number when drawing a comparison with modern texts, and this is on the grounds of obvious reasons: For one thing, there is the fact that textual contents were quite closely connected to space in the Middle Ages, meaning that space in medieval parchment manuscripts or flyleaves was highly limited due to their costly production. Diligent planning was needed to arrange and convey as much content as possible, and leaving out the dispensable was a logical consequence. Additionally, the enormously high production costs must be seen as one of the main reasons for the exclusiveness and rareness of texts: only monasteries and a small group of well-off literates could and would afford them. As such, documents were then often stored in libraries or kept in private possession; and access to them was absolutely restricted. A text producer would thus, at best, have a limited chance to quote or cite with any accuracy, as he could not actually lay hands on most source texts. A look back at Figure 2 reminds us of another difficulty when it comes to quoting: the instability of texts. With uncountable variants of a text having drifted apart over years and regions, they were at times rendered incomparable to each other and quoting was almost impossible. As a consequence, chances were very low that identical wording (as required in the modern concept of quotes) would be found in any other variant of a text quoted, assuming that someone made a successful attempt to look it up.

3.3 What Makes a Quote a Quote? Having investigated the classificational challenges posed by medieval quotes so far, it can definitely be stated that even if they cannot be pushed into any modern delimitational framework, medieval quotes still share a number of formal and functional features with quotes in our modern understanding of them. Three functional assets have to be highlighted in particular and will be of interest for the upcoming sections of this paper, both then and now (cf. also Bublitz, this volume), → a quote can only function as such, when intended, indicated and recognised as a quote by its producer and recipient; → a quoted text is always shifted from its source context into another; → the act of quoting entails an evaluation of the quoted text and the new context it is posed in. Surely, though, these features are much harder to pin down in medieval texts and from a modern stance. Recognising a quote as such certainly is not easy. A

302   

   Monika Kirner-Ludwig and Iris Zimmermann

lack of conceptually shared knowledge with the text producer(s) due to the time gap is most probably the major reason for not spotting quotational references in manuscripts. This cognitive gap is even deepened by the fact that punctuational devices indicating quotes as such or even marking the endings of sentences are absent: Among the few helpful signals we may rely on in that matter are verba dicendi/cantandi, often introducing direct or indirect quotations.²² As for the evaluative criterion which comes with quotes, we proceed from the assumption that any producer’s decision to re-use another man’s ideas or words in his own work is based on his positive or negative (sometimes neutral, if you will) evaluation of the quoted excerpt. The outcome of this evaluation can either be a discussion and correction of the thoughts criticised, or the employment of arguments regarded as useful in supporting his own claims.²³ The following samples are meant to serve as examples for what has just been argued regarding quotes and their classificational criteria²⁴ and were excerpted from the corpus of Middle High German courtly literature (ca. 1290–1230) – with the exemption of a religious text referring to Meister Eckhart’s ‘Mystik’ (cf. (2)) as well as an extract from Geoffrey Chaucer’s Canterbury Tales dating back to the late 14th century (ed. Benson 1988).²⁵ (1)

des muoz hêr walther singen ‘guoten tac, bœs unde guot.’ [Consequently the honorable Walther must sing ‘Good day, be you bad or good.’] (Wolfram 1998: 301)

In (1) the narrator of the Parzival relates to an extra-narrative context by addressing landgrave Hermann von Thüringen, who suffered at the hands of untrustworthy companions at his court (“bœs unde guot”). In doing this he also directly quotes (cf. verbum cantandi “singen”) the honourable Walther (“hêr walther”), supposedly the Middle High German singer-composer Walther von der Vogelweide, with an excerpt of one of his lyrics, which was not passed on and might even be fictitious (cf. Strohschneider 2002: 102). 22 Also cf. Colette Moore’s introductory discussion of formal aspects of quotes in Early English manuscripts (2011: 1–17). 23 Compare Bublitz/Hoffmann’s definition of quoting as being “the act of taking up a source text (T1) of an author (A1) and shifting it from its context (C1) to another context (C2) as a target text (T2)” (2011: 434). 24 All highlighting underlinings are ours. 25 All translations from MHG in this paper have, if not indicated otherwise, been produced by Iris Zimmermann, all renditions from Middle English and Latin are by Monika Kirner-Ludwig.

Quoting and Plagiarising – Concepts of Both Now and Then?   

(2)

   303

hie von spricht maister eckhart: ettlich lit nement got (…) [Meister Eckhart speaks about the following: Many people take God (…)] (Cod. III. 1. 8° 42 (Eckhart-Sigle Mai6): Zimmermann 2011: 152)

Sample (2) was taken from a paper manuscript presumably produced in 1477 (University Library Augsburg, Cod. III. 1. 8° 42; Eckhart-Sigle Mai6) as part of the tract ‘Von den drîn fragen’ [On the three questions], and contains once more the direct quote (cf. verbum dicendi “sprichet”) from Meister Eckhart, which again closely resembles a part of one of his German sermons (Sermon 5a, ‘In hoc apparuit’, Meister Eckhart, DW 1, p. 77–82.). As for a parallel case of direct quoting in Middle English, sample (3) was chosen from Chaucer’s Tale of Melibee, which displays a fairly congruent structure just as (1) and (2), thus entailing a rather specific reference to both the source²⁶ and the quoter as well as a verbum dicendi that points the text out as a quote: (3)

This noble wyf Prudence remembred hire upon the sentence of Ovide, in his book that cleped is the Remedie of Love, where as he seith, […] He is a fool that destourbeth the mooder wepen in the deeth of hire child […] (ed. Benson 1988: 217 [ll. 976ff.]) [The noble wife [of Melibee] Prudence remembered a sentence uttered by Ovid in his book called Remedy of Love, saying: ‘He who disturbs a mother weeping over the death of her child is a fool’]

Although samples (1) to (3) do display the phenomenon of re-embedding text into a new context, examples (4) to (7) are more apt in carving out the rather functional aspects of doing so. We will focus on the evaluative factor here in particular. (4)

Ahî, wie kristenlîche nû der bâbest lachet, swanne er sînen Walhen seit: ‘ich hânz alsô gemachet!’ [Hah, the pope laughs like a Christian whenever he tells his Romans: ‘I did it that way!’] (Walther 1996: 64)

Excerpt (4) describes how Walther von der Vogelweide in one of his political songs makes use of another direct quote, the source context of which we do not know. Walther instrumentalises this utterance not only to reinforce his polemics against

26 In fact, Benson verified this quote, pointing the reader to Ovid’s Remedia Amoris, 127 (1988: 924).

304   

   Monika Kirner-Ludwig and Iris Zimmermann

Pope Innocent III by inserting the Pope’s utterance into a rather fierce context, but also in order to raise his own credibility by feigning closeness to the Pope. (5)

und hetest anders niht wan eine rede gesungen – sô wol dir wîp, wie reine [ein] nam! [And if you had sung these words only – Woman, you may be estimated so highly, what a virtuous name!] (Walther 1996: 182)

“So wol dir wîp, wie reine [ein] nam” in (5) is part of Reinmar’s ‘programmatical’ songs. Walther von der Vogelweide picked this up in the obituary he wrote to his colleague: this time using the words not in honour of ladies (as Reinmar did) but in honour of the singer Reinmar himself. (6)

In dem dône: Ich wirbe umb allez daz ein man [In the melody: I strive for everything that a man] (Walther 1996: 234)

Walther’s song is a parodistic contrafactum to Reinmar’s ‘Ich wirbe umbe allez, daz ein man’, Moser/Tervooren (eds.) 1988: 305–307). The melody (“done”) of Reinmar’s song (“Ich wirbe umbe allez daz ein man”) seems to have been re-used by Walther (cf. (6)). The indirect quotation thereby created from Reinmar’s source text (in the honour of ladies) allowed Walther to pervert his colleague’s intention and to denounce him at the same time (as the song’s stanzas show). Another intricate case of reference, presented from Chaucer’s Miller’s Tale, borders on being but an allusion rather than a quote: sample (7) does not follow the formal expectations of what a recipient would lay into a quote. Chaucer wove several references into the ME poem and song commonly known as Alysoun (cf. Harley 2253) into his own text, neither giving the source nor any verbum dicendi to indicate his act of quoting. Nevertheless, Chaucer made sure that his intertextual references were comprehended as such: not only did he name the main female character of this story Alysoun, but he also introduced her by inserting a beauty catalogue similar to the one in Harley 2253. Most importantly, though, Chaucer put a phrase found in the poem (“Ich libbe in love-longinge”, ed. Luria/Hoffman 1974: 23 [l. 5]) into her admirer’s mouth, and repeated it three times to make sure it got picked up:

Quoting and Plagiarising – Concepts of Both Now and Then?   

(7)

   305

This parissh clerk, this joly Absolon, Hath in his herte swich a love-longynge (ll. 3349f.) […] [The jolly parish clerk Absolon had such longing for love in his heast] ‘[…] To Alison now wol I tellen al My love-longynge […]’ (ll. 3678f.) […] [‘I want to tell Alyson all about my longing for [her] love’] ‘[…] Ywis, lemman, I have swich love-longynge […]’ (l. 3705) [Truly, Sweetheart, I have such love-longing.]

The transfer of Alysoun, the immaculate object of an anonymous man’s “lovelongynge” as encountered in Harley 2253, into a fabliau-context, where she gets involved in a sexual affair with a young student whilst alluringly leading on the clerk Absolon and insidiously tricking and betraying her husband, bears several effects: not only would Chaucer’s audience find this intertextual transmission hilariously funny, possibly even picking up on Chaucer’s intention to lead the whole idea of courtly love ad absurdum, but also would this witty move of his reflect positively on Chaucer’s humorous finesse and literacy.

3.4 An Interim Summary As has been shown, it is almost impossible to state how and at which stage of its textual development a text variant passed on in a manuscript actually entered it (cf. Figure 2). What is more, written documents that are still accessible nowadays must generally be regarded as mere fragments of their medieval distribution. Figure 2 tries to illustrate the high potential dynamics of a text in the Middle Ages, which was certainly enhanced by limited written fixation as well as far reaching illiteracy. Furthermore, we cannot assume that our modern-day concept of an ‘author’ existed. Medieval text creators most probably helped themselves from a freely accessible repertoire of ideas. Not very many texts that have come down to us contain references to their ‘singer/teller’ (who were not necessarily the creators of the texts they presented; cf. Figure 2). The lack of clear-cut concepts of author and text results in the conceptual vagueness of quotation – which is consequently true for plagiarism as well. Presumably with the introduction of printing in the late 15th century and the resulting mass production and circulation of documents, the number of literates slowly grew and an understanding of author, text and quotation began to gain shape.

306   

   Monika Kirner-Ludwig and Iris Zimmermann

4 Sounding out Conceptual Features of Plagiarism in Medieval Praefationes and Prologoi 4.1 Introductory Remarks Concerning the Etymology and Semantic Makeup of Plagiarism As should already have become obvious from our observations made on ‘text’ as well as from the numerous examples presented above, the medieval concept of quoting naturally differs from our modern view for the reason alone that the understanding and usage of texts differed immensely in the first place. Much more than today, texts were considered ‘public domain’,²⁷ free to be recited and amended. Thereby, it appears that utterances were generally not considered ‘intellectual property’ as they are today. This, however, does not mean that there was no awareness of the major importance and relevance of referring to the authority of a text, e.g. in order to express one’s acknowledgement. This begs the question of when and how the idea of text as someone’s intellectual property (with ‘copyright’) evolved between the Middle Ages and today. Since, for instance, genealogical and historical data represent one of the major domains regarded as common property and shared knowledge by the co-members of a cultural group, a diachronic case study into the intentions of their compilers seems most promising when trying to retrace this conceptual shift. In historiographic works, praefationes and prologoi are most potent in granting glimpses into the personal profile of a text compiler and thus revealing a potential growth in their awareness of plagiarism (cf. 4.2). This is why a selection of such forewords will be investigated in the following in order to complement our observations on the medieval concept of text and reference. A brief etymological outline of the term plagiarism illustrates that the 1st century AD already knew the concept of plagiarism as we use it today. An utterance by Marcus Valerius Martialis explicitly expresses this: (8)

I, lii Commendo tibi, Quintiane, nostros – nostros dicere si tamen libellos possum, quos recitat tuus poeta:

I entrust you, Quintian, with my little writings – I say ‘my’, although I cannot anymore call them thus, since your poet recites them.

27 Note, of course, that the general modern understanding of the notion public domain pertains to internet sites in particular.

Quoting and Plagiarising – Concepts of Both Now and Then?   

   307

si de seruitio gravi queruntur,

If they lament this hard servile service, adsertor venias satisque praestes, please come to their rescue and see them through et, cum se dominum vocabit ille, and, when he refers to himself as their master, dicas esse meos manuque missos. please tell that they are mine and have been released by my hand. hoc si terque quaterque clamitaris, If you shout that out three or four times, inpones plagiario pudorem. you will expose the thief in shame. (9)

I, liii Vna est in nostris tua, Fidentine, libellis pagina, sed certa domini signata figura quae tua traducit manifesto carmina furto. […] indice non opus est nostris nec iudice libris,

One page in my little volume is yours, but it is marked by the unique seal of its master, which consigns your poems to theft.

it is not my work indicated by a seal and on grounds of a judicial decision stat contra dicitque tibi tua pagina it is not in my volumes, but your ‘Fur es.’ page calls out to you: “You Thief!”

As these two epigrams (quoted from ed. Lindsay 1903: no page numbers) show, Martial experienced cases of plagiarism first hand, i.e. the theft of intellectual property that he had noted down but not sealed as being his own straight away, which was taken advantage of by a plagiarius, a ‘thief’.²⁸ We may thus state that our modern concept of plagiarism is not modern at all, but rather describes a condemned and specific category of theft already extant as early as in Martial’s days. If we want to continuously trace both term and concept from their beginnings up until the modern English inventory, it becomes evident, however, that neither plagiarism nor any derivative form of it was borrowed into the OE or ME lexical inventory according to both OEDo and MEDo. The same can be said for MHG.

28 This term itself derives from L plagium n. ‘abduction’ (< classL plagium n. ‘netting of game’ ? < classL plaga n. ‘net, sling’).

308   

   Monika Kirner-Ludwig and Iris Zimmermann

At first glance, this could suggest that the English Middle Ages neither knew nor applied the concept of plagiarism and therefore saw no need to borrow this notion. However, as we will show in the upcoming section, it seems that the concept was simply expressed differently. The same idea of a plagiator as a ‘thief’ can be found in a song from the Middle High German singer Der Marner when he denunciates Reinmar von Zweter, a fellow singer of his, with the following verses: (10)

Wê dir von Zweter Regimâr! dû niuwest mangen alten funt, dû speltest als ein milwe ein hâr [.] (…) dû dœnediep[!] (…) [Shame on you, Reinmar von Zweter! you sell old things as new ones. you split hair as an acarian does[.] (…) you thief of melodies[!] (…)] (Burkhard 2012: 145)

In his invective against Der Meißner, the medieval German singer Fegfeuer also picks up the idea of stealing and giving back (literally) the melodies that were stolen before: (11)

Er [Der Meißner] gebe den pfaffen ir dœne wider und singe, swaz er welle. [He [Der Meißner] should give the melodies back to the clerics and shall sing whatever he wants to sing.] (Burkhard 2012: 74)

4.2 Praefationes and Prologoi to Historiographical Texts of the English Middle Ages – and What They Reveal about the Concept of Plagiarism With regard to their function, the notions prologoi and praefationes are synonymously used to refer to forewords preceding the actual body of a main text. They even have a parallel etymological history: while the Greek notion prologos made its way through Old French and entered the English inventory via the AngloNorman prologe (cf. ModE prologue n.), the Latin noun praefatio (< praefari v. ‘to say in advance’) was borrowed into the English inventory as the Anglo-Norman/ Middle French preface (cf. ModE preface n.). We are going to observe what these forewords reveal about their producers’ attitudes.

Quoting and Plagiarising – Concepts of Both Now and Then?   

   309

First, to state the most obvious, praefationes and prologoi, due to their lead position alone, categorically share the crucial task of introducing the main text. In medieval terms this would generally include a number of formulaic patterns which were conventionally expected and provided, such as devices of a) authentification of both the material and the intention of the text producer as being trustworthy and b) the authorisation of his work as e.g. being a work commissioned or inspired by a specific person/sponsor. These points also imply c) the matter of relevance to the reader, which is also linked to a fourth and highly common component frequently of an elaborate stylistic nature, viz. d) the endeavor to earn the reader’s benignity even before the main deliberations begin (captatio benevolentiae). The textual aspects a) to d) are features inherent to the ‘text-type’ preface/ foreward both then and still now. What may at first glance seem as mere formulae inserted into any preface on grounds of convention, does fulfil a number of much more implicit functions going beyond any formal criteria that was simply expected by the recipient: all the devices featuring in a preface are applied in order to compose and convey a trustworthy image of the text producer and of his text. The appeal of authorities, therefore, must have occurred as an imperative asset of each preface, which also matches Beumann’s observation (cf. above, 1962: 13f.) concerning the dismissal of originality but absolute reliance upon authority in medieval texts. Both aspects, viz. the citational reference to textual authorities as well as the alleged non-intention of being original in one’s own ‘new’ version of events retold is under diachronically descriptive focus in the following. The examples chosen for this purpose range chronologically from the first half of the 8th century (cf. (12)), Bede’s praefatio to his Historia Ecclesiastica Gentis Anglorum, 731 AD (quoted from URL 3; transl. quoted from McClure & Collins 1999: 3), to two samples from the 12th century, viz. Layamon’s Brut (cf. (13), quoted from British Museum Ms. Cotton Caligula A.IX, URL 4) and Geoffrey of Monmouth’s De gestis Britonum (cf. (14), quoted from ed. Reeve 2007: 4f.), to one further sample (15) from the late 15th century, namely William Caxton’s preface to his print edition of Thomas Malory’s Morte Darture (URL 5). The Historia Ecclesiastica must certainly be considered the most prominent and most tone-setting work in early English historiography, both with regard to content as well as to style, and this is the main reason for including a section of Bede’s praefatio in the upcoming analysis. (12)

Historiam Gentis Anglorum Ecclesiasticam quam nuper edideram, libentissime tibi desiderant, rex [Ceoluulfe], et prius ad legendum ac probandum transmisi, et nunc ad transcribendum ac plenius ex tempore meditandum retransmitto: satisque studium tuæ sinceritatis amplector, quo

310   

   Monika Kirner-Ludwig and Iris Zimmermann

non solum audiendis Scripturæ sanctæ verbis aurem sedulus accomodas, verum etiam noscendis priorum gestis sive dictis et maxime nostræ gentis virorum illustrium, curam vigilanter impendis. […] Auctor ante omnes atque adjutor opusculi hujus Albinus abba reverentissimus vir per omnia doctissimus extitit.[…] [Your Majesty has asked me to see the History of the English Church and Nation which I have lately published. It was with pleasure, sire, that I submitted it for your perusal and criticism on a former occasion; and with pleasure I now send it once again, for copying and fuller study, as time may permit. I gladly acknowledge the unfeigned enthusiasm with which, not content merely to lend an attentive ear to hear the words of Holy Scripture, you devote yourself to learn the sayings and doings of the men of old. […] My principal authority and helper in this modest work has been the revered Abbot Albinus, a man of universal learning.] Bede’s pioneering role with regard to the history of the English people automatically rules out the claim of insinuated non-originality: writing down history may have been quite a conventional thing to do, but in this case it was something new indeed and a case where the text producer would certainly have been unable to draw on much previously compiled material. Thus, Bede seeks to authenticate his material by e.g. explicitly addressing and involving King Ceolwulf as his critical and learned reader. Since Bede could not have had too many sources at hand for his undertaking, he could also have neglected this issue, but instead he did the opposite: he referred to his contemporary Albinus, abbot of St. Augustine, Canterbury, whose death is recorded as having taken place in 732, i.e. one year after the publication of the Historia Ecclesiastica. By giving gratias to Albinus – it is needless to point out the parallel to modern (academic) writing – Bede not only openly acknowledges his value to his work, but also explicitly declares that he received help, instead of pretending that he did not and thus committing an act of plagiarism. The two 12th century samples (13) and (14) paint a diverse picture concerning the awareness of intellectual property and the criteria a) to d): In (14), Geoffrey of Monmouth’s prologus (c.1135), we find the same components as in (12): authentification of the material is achieved by the reference to a “librum vetustissimum” handed to him personally by a trustworthy man of the Church who asked him (“rogatu”) to translate it, which thereby also fulfills criterion b). Apart from sophisticatedly begging for his readers’ favours, particularly of his direct addressee Robert of Gloucester (cf. e.g. “sale mineruae tuae”), both the reliability of the true nature of Geoffrey’s intention as well as the explanation of his work’s relevance to the reader are described from his first person singular perspective.

Quoting and Plagiarising – Concepts of Both Now and Then?   

   311

This stance was also chosen by Bede in his praefatio and should clearly be read as an indicator for the intention of pointing to his individually valuable performance in the first place. Geoffrey expresses both his personal deliberations (“Cum mecum multa et de multis saepius animo reuoluens”) as well as his surprise (“mirum contuli”) when, after doing research, he realises that there had been a historiographical gap with regard to King Arthur in the works by Bede and Gildas. The most remarkable utterance, however, is “tametsi infra alienos ortulos falerata uerba non collegerim” [while I have not gathered flowery words from the gardens of others], which in elegant wording presents a direct reference to Geoffrey’s awareness of plagiarism and his explicit refrainment from any such action. Again, this 12th century prologue gives a very modern impression with regard to both references to earlier authorities as well as to the near-academic argumentative derivation of the objective of Geoffrey’s work, which is doubtlessly relevant due to its newsworthiness. The slightly earlier text by Laȝamon (cf. (13), quoted from Bzdyl 1989) seems to apply a different strategy. Whereas (12) and (14) focus upon the reduction of distance between the reader and their creator, Laȝamon seems to take other priorities: whilst referring to himself in the 3rd person sg. instead of using the 1st person and elaborate captatio-devices like Bede and Geoffrey, he tries to win his audience’s trust by describing how his chains of thought (“Hit com him on mode; & on his mern Þonke” etc.) led his careful research into standard references by Bede, Augustine (“þe feire Austin”) and Wace’s Roman de Brut (ca. 1155). What appears to be deliberately new in Laȝamon’s approach is the gathering of all known-tobe-relevant reference works mentioned in his version. Thus, his work, too, is of an enterprising nature – just like his metaphorical journey of getting hold of the books he mentions. (13)

An preost wes on leoden; Laȝamon wes ihoten. […] Hit com him on mode; & on his mern þonke. þet he wolde of Engle; þa æðelæn tellen. […] Laȝamon gon liðen; wide ȝond þas leode. & bi-won þa æðela boc; þa he to bisne nom. He nom þa Englisca boc; þa makede Seint Beda.

[There was a priest among the people who was called Layamon. It came to his mind and in his chief thought to tell the noble deeds of the English. Layamon began to journey wide over this land and procured the noble books, which he took for authority. He took the English book made by St Bede.

312   

   Monika Kirner-Ludwig and Iris Zimmermann

An-oþer he nom on Latin; þe makede Seinte Albin. & þe feire Austin; þe fulluht broute hider in. Boc he nom þe þridde; leide þer amidden. þa makede a Frenchis clerc; Wace wes ihoten; þe wel couþe writen. & he hoe ȝef þare æðelen; Ælienor þe wes Henries quene; þes heȝes kinges. Laȝamon leide þeos boc; & þa leaf wende. he heom leofliche bi-heold. liþe him beo Drihten.

Another book in Latin by St Albin and the fair Austin, who brought baptism hither. The third book he took laid there amidst. it was made by a French clerk called Wace, who could write well. He gave it to the noble Eleanor Who was the high King Henry’s queen. Layamon laid before him these books and turned the leaves he tenderly beheld them. May the Lord aid him.]

One particularly curious aspect with regard to the phenomenon of plagiarism under focus here must briefly be highlighted before we move on to Caxton’s text (15): Laȝamon turns out to have indeed committed plagiarism from a modern point of view. A number of studies, e.g. by Madden (1847) and Le Saux (1989), have called attention to the fact that the two sources referred to by Laȝamon as “Seinte Albin” and “Austin” are highly conflictual on grounds of temporal connection. Whereas St Austin died in 604, Albinus, who is generally identified as the Abbot of St Austin’s at Canterbury by Madden (1847: xii [introd.]), died in 732, which renders it hardly plausible to believe that these “should [have been] conjoined in the same work” (Madden 1847: xii [introd.], quoted from Le Saux 1989: 17). This inconsistency is generally assumed to be the reason that the scribe of the Otho version of the text departed from his master copy by citing Albin as the creator of the second and Austin as the one of the third book instead (Le Saux 1989: 17; cf. Madden 1847: xii [introd.]). Laȝamon’s (deliberately?) false reference, however, is not the only one to be found in his preface since his entire appeal to Bede’s work is inaccurate. The line “He nom þa Englisca boc; þa makede Seint Beda”) is misleading too, since Bede’s work was definitely written in Latin, not English. It even seems, as Madden (ibd.) suggests, that Laȝamon’s “Seinte Albin” may just have been copied from Bede’s praefatio (“Albinus abba”, cf. (12)) simply in consideration of (the need of?) naming another legitimate source . (14)

Cum mecum multa et de multis saepius animo reuoluens in hystoriam regum Britanniae inciderem, in mirum contuli quod infra mentionem

Quoting and Plagiarising – Concepts of Both Now and Then?   

   313

quam de eis Gildas et Beda luculento tractatu fecerant nichil de regibus qui ante incarnationem Christi inhabitauerant, nichil etiam de Arturo ceterisque compluribus qui post incarnationem successerunt repperissem, cum et gesta eorum digna aeternitate laudis constarent et a multis populis quasi inscripta iocunde et memoriter praedicentur. Talia michi et de talibus multociens cogitanti optulit Walterus Oxenefordensis archidiaconus […] quendam Britannici sermonis librum uetustissimum […] proponebat. Rogatu itaque illius ductus, tametsi infra alienos ortulos falerata uerba non collegerim, agresti tamen stilo propriisque calamis contentus codicem illum in Latinum sermonem transferre curaui. […] Opusculo igitur meo, Roberte dux Claudiocestriae, faueas, ut sic te doctore te monitore corrigatur quod […] censeatur […] sale mineruae tuae²⁹ conditum […] [While my mind was often pondering many things in many ways, my thoughts turned to the history of the kings of Britain, and I was surprised that, among the references to them in the fine works of Gildas and Bede, I had found nothing concerning the kings who lived here before Christ’s Incarnation, and nothing about Arthur and the many others who succeeded after it, even though their deeds were worthy of eternal praise and are proclaimed by many people as if they had been entertainingly and memorably written down. I frequently thought the matter over in this way until Walter archdeacon of Oxford […] brought me a very old book in the British tongue […]. Though I have never gathered showy words from the gardens of others, I was persuaded by his request to translate the book into Latin in a rustic style, reliant on my own reed pipe. Therefore, earl Robert of Gloucester, look favourably on my little work; let it be corrected by your instruction and advice so that it seems duly seasoned with the genius of your wit.] Caxton’s preface to his print edition of Thomas Malory’s Morte Darthure (15) completes the diachronic set of our samples and provides the exact same devices of persuasion of the reader as samples (13) and (14): it also displays the personal experience of cognitively developing the thoughts leading to the decision to write the work. Caxton describes this process in all of its details, not only taking into account other opinions but citing them most precisely: he refers to “the v book the syxte chappytre” in Ranulph Higden’s Polychronicon (mid 14th century) and to Geoffrey’s Historia Britonum (“galfrydus in his brutysshe book”), thereby providing authentification of both the material gathered “vnder the fauour and correctyon of al noble lordes and gentylmen” (also cf. d)) and his trustworthy “entente

29 Metonymical use of Minerva, the Goddess of wisdom ~ ‘wisdom, genius’.

314   

   Monika Kirner-Ludwig and Iris Zimmermann

that noble men may see and lerne the noble actes of chyualrye”, thereby expressing his hope for the reader’s “fayth and byleue that al is trewe that is conteyned herin” (cf. a)). Caxton ultimately traces back his authorisation for printing this work to God’s will (cf. b)), which led to him being provided with a “copye […] Syr Thomas Malorye dyd take oute of certeyn bookes of frensshe and reduced it in to Englysshe” [a copy that Sir Thomas Malory made from certain French books and then translated into English]. Not only his reasoning for the existence of Arthur (for revealing the truth about him even), but also his provision of Malory’s English work, which covers the contents of the majority of earlier and partly nonvernacular accounts, highlights the relevance of his print to the reader. (15)

[…] dyuers men holde oppynyon / that there was no suche Arthur / and that alle suche bookes as been maad of hym / ben fayned and fables / by cause that somme cronycles make of hym no mencyon ne remembre hym noo thynge ne of his knyghtes […] / and one if specyal sayd / that in hym that shold say or thynke / that there was neuer suche a kynge callyd Arthur / myght wel be aretted grete folye and blyndenesse / For he sayd that there were many euydences of the contrarye / Fyrst ye may see his sepulture in the monasterye of Glastyngburye / And also in polycronycon in the v book the syxte chappytre / and in the seuenth book the xxiij chappytre/ […] Also galfrydus in his brutysshe book recounteth his lyf […] Thenne al these thynges consydered there can no man resonably gaynsaye but there was a kyng of thys lande named Arthur […] & many noble volumes be made of hym & of his noble knyȝtes in frensshe which I haue seen & redde beyonde the see / which been not had in our maternal tongue / but in walsshe ben many & also in frensshe /& somme in englysshe but no wher nygh alle / […] I haue after the symple connynge that god hath sente to me / vnder the fauour and correctyon of al noble lordes and gentylmen enprysed to enprynte a book of the noble hystoryes of the sayd kynge Arthur / and of certeyn of his knyghtes after a copye vnto me delyuerd / whyche copye Syr Thomas Malorye dyd take oute of certeyn bookes of frensshe and reduced it in to Englysshe / And I accordyng to my copye haue doon sette it in enprynte / to the entente that noble men may see and lerne the noble actes of chyualrye […] And for to passe the tyme thys boook shal be plesaunte to rede in / but for to gyue fayth and byleue that al is trewe that is conteyned herin […] Amen / [Many people do not believe that there was a [king called] Arthur, but rather that all those books about him have been contrived and are fictitious; this [the reason for their unbelief] is because some chronicles neither mention, nor remember him or his knights. One extraordinary man [however once] claimed that anyone who would not believe that there ever

Quoting and Plagiarising – Concepts of Both Now and Then?   

   315

was a king called Arthur might well be called foolish and blind, as many pieces of counterevidence existed. For one thing, one may see his tomb in Glastonbury and consult the sixth chapter of Book V and the 23rd chapter of Book VII of the Polychronicon […]. Galfredus [Geoffrey of Monmouth], too, recounts his [Arthur’s] life in his British book […]. All these aspects considered, there is no reasonable man who could possibly state that there was no king of this country named Arthur. […] Beyond the sea, I have seen and read many valuable volumes that have been composed about him and his noble knights in the French tongue and which have not been translated into our mother tongue, but many have been into Welsh and French. Some, but not nearly all, have been translated into English. With the little skill God has granted me, and the benefit of benevolent criticism provided by all noble lords and gentlemen, I have embarked on the enterprise of printing a book containing the noble stories of the aforementioned Arthur and some of his knights. My printed version is based on a copy, which was handed to me by Sir Thomas Malory, who had translated into English what he had compiled from certain French books. I have put this text into print according to my copy, hoping that noble men would learn about noble actions of chivalry. May this book be a pleasant read as a pastime, but may it also persuade [the reader] that all its contents are true. Amen.] As can be seen, all four texts, chosen at various points on the diachronic timeline, contain the features that are typical in forewords as their authors felt the obligation to highlight their true intentions of providing the reader with both reliable and newsworthy information. For this purpose we find them not only recurrently stressing the fact that they themselves are trustworthy writers building their argumentation upon traditional authorities, but even sustaining their honest intentions by building a construct of (fictional) causalities which put them in the exact position of doing so. The greatest authority of course is God, who made them write down the text in the first place; this is particularly explicit in (15), where Caxton finishes his preface with the word “Amen”, and implicitly expressed in (12), (13) and (14), which all stem from a clerical perspective: Bede writes an ecclesiastical history and Laȝamon introduces himself as a “preost”, who, just like Geoffrey of Monmouth, uses reference works by authorities of the Church. The observations made in the samples selected for this paper will also find congruent affirmation in other prefaces, e.g. by Nennius to his Historia Brittonum (8th/9th century) and William of Malmesbury to his Gesta regum anglorum (ca. 1120).

316   

   Monika Kirner-Ludwig and Iris Zimmermann

5 Reference and Plagiarism Revisited – a Then-and-Now Perspective By presenting a wide range of textual examples, we have demonstrated that the Middle Ages certainly knew how to express an explicit respect for authorities and how to refer to them. Each creator of our samples seems to have been more or less interested in legitimising his work and locating himself and his statements in a certain tradition, which again holds true for scientific quotations of today. And despite the instability of medieval texts (see Figure 2) (and many other hindrances), we still have a great amount of textual witnesses for quoting in the Middle Ages. If we now hypothesise that there was an understanding of quotation, it is only a small step to assume that there must have also been an understanding of plagiarism, which, due to textual diversity and intraceability, is hard to grasp from the modern viewpoint. So despite a lack of lexemes denoting acts of referring, quoting or citing, text creators of the English and German Middle Ages seem to have been well aware of what it meant to quote or not to quote. The medieval concept of originality might, due to the environment of medial tradition, not have been quite the same as today, but the importance of respectful reference to distinguished authorities and the creation of something that never existed before must in some way or another have been valued to a certain degree. This is also expressed in the following stanza from the Middle High German singer/composer Walther von der Vogelweide: (16)

Ir sult sprechen willekomen: der iu mære bringet, daz bin ich. allez, daz ir habt vernomen, dest gâr ein wint, nû vrâgtet mich. Ich wil aber miete. wirt mîn lôn iht guot, ich sage vil lîhte, daz iu sanfte tuot. seht, waz man mir êren biete. (Walther 1996: 117)

[You shall welcome me: I’m the one who brings news to you. Whatever you have heard before, is nothing. Ask me now. Yet I ask for a reward. If it is good enough, I will possibly tell you pleasant things. Make sure to do me honour.]

Quoting and Plagiarising – Concepts of Both Now and Then?   

   317

References Primary Sources Beda Venerabilis. Historia Ecclesiastica Gentis Anglorum. Günter Spitzbart (ed.). 1982. Darmstadt: Wissenschaftliche Buchgesellschaft. Geoffrey Chaucer. The Canterbury Tales. Larry D. Benson (ed.). 1988. Oxford: Oxford University Press. 3rd edition. Geoffrey of Monmouth. The History of the Kings of Britain. An Edition and Translation of De gestis Britonum. Michael D. Reeve (ed.) & Neil Wright (transl.). 2007. Suffolk: Boydell Press. Gildas. De Excidio Britanniae. Joseph Stevenson (ed.). 1964. Vaduz: Kraus. Layamon. Brut. A History of the Britons. Donald G. Bzdyl (ed.). 1989. Binghampton, New York: Medieval & Renaissance texts & studies. Lindsay, W. M. (ed.). 1903. M. Val. Martialis Epigrammata Selecta. Oxford: Clarendon Press. Luria, Maxwell S. & Richard L. Hoffman (eds.). 1974. Middle English Lyrics. Authoritative Texts. Critical and Historical Backgrounds. Perspectives on Six Poems. New York: Norton. Nennius. Historia Britonum. Joseph Stevenson (ed.). 1964. Vaduz: Kraus. Moser, Hugo & Helmut Tervooren (eds.). 1988. Des Minnesangs Frühling. I. Stuttgart: Hirzel. Reinmar. Lieder. Günther Schweikle (ed.). 2002. Stuttgart: Reclam. Wace. Roman de Brut. A History of the British. Text and Translation. Judith Weiss (ed.). 1999. Exeter: University of Exeter Press. Walther von der Vogelweide. Leich, Lieder, Sangsprüche. Christoph Cormeau (ed.). 1996. Berlin: de Gruyter. Wolfram von Eschenbach. Parzival. Peter Knecht (ed.). 1998. Berlin: de Gruyter. Zimmermann, Iris. 2011. “Aus der Bibliothek des Fürsten Kraft Ernst: Eine Sammelhandschrift mit anonymen Traktaten der Eckhart-Nachfolge und reihenweise Auszügen”, in: Freimut Löser (ed.). Meister Eckhart in Augsburg. Deutsche Mystik des Mittelalters in Kloster, Stadt und Schule. Universitätsbibliothek Augsburg, 151–156.

Secondary Sources Bein, Thomas. 2001. “Zum Umgang mit handschriftlichen Autorzuweisungen: Bilanz und Vorschläge für eine literarhistoriographische Handhabe”, in: Anton Schwob, András Vizkelety (eds.). Entstehung und Typen mittelalterlicher Lyrikhandschriften (Jahrbuch für Internationale Germanistik, Reihe A, Band 52). Bern et al: Lang, 15–36. Beumann, Helmut. 1962. Ideengeschichtliche Studien zu Einhard und anderen Geschichtsschreibern des früheren Mittelalters. Darmstadt: Wissenschaftliche Buchgesellschaft. Bublitz, Wolfram & Christian Hoffmann. 2011. “‘Three men using our toilet all day without flushing – This may be one of the worst sentences I’ve ever read’: Quoting in CMC”, in: Frenk, J. & L. Steveker (eds.). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 433–447. Burkhard, Mirjam. 2012. Sangspruchdichter unter sich. Namentliche Erwähnungen in den Sprüchen des 12., 13. und 14. Jahrhunderts. Heidelberg: Universitätsverlag Winter.

318   

   Monika Kirner-Ludwig and Iris Zimmermann

Durkin, Philip. 2011. The Oxford Guide to Etymology. New York: Oxford University Press. Online Etymological Dictionary. Le Saux, Françoise. 1989. Layamon’s Brut. The Poem and its Sources. Cambridge: Brewer. Madden, Frederick (ed.). 1847. Brut, or Chronicle of Britain. A Poetical Semi-Saxon Paraphrase of the Brut of Wace; Now First Published from the Cottonian Manuscripts in the British Museum, Accompanied by a Literal Translation, Notes, and a Grammatical Glossary. London: Society of Antiquaries of London. Moore, Colette. 2011. Quoting Speech in Early English. Cambridge: Cambridge University Press. Müller, Jan-Dirk. 1996. “Vorbemerkung”, in: Jan-Dirk Müller (ed.). ‘Aufführung’ und ‘Schrift’ in Mittelalter und Früher Neuzeit. (DFG-Symposion 1994. Germ. Symposion-Berichtsbände 17). Stuttgart/Weimar: Metzler, xiii-xviii. Strohschneider, Peter. 2002. “Fürst und Sänger. Zur Institutionalisierung höfischer Kunst, anläßlich von Walthers Thüringer Sangspruch 9,V [L. 20,4]*”, in: Ernst Hellgardt, Stephan Müller & Peter Strohschneider (eds.). Literatur und Macht im mittelalterlichen Thüringen. Köln: Böhlau, 85–107. Wachinger, Burghart. 2011. “Autorschaft und Überlieferung (1991)”, in: Burghart Wachinger (ed.). Lieder und Liederbücher. Gesammelte Aufsätze zur mittelhochdeutschen Lyrik. Berlin/New York: de Gruyter, 1–23. Wolf, Jürgen. 2002. “New Philology/Textkritik”, in: Claudia Benthien & Hans Rudolf Velten. Germanistik als Kulturwissenschaft. Eine Einführung in neue Theoriekonzepte. Reinbek bei Hamburg: Rowohlt-Taschenbuch-Verlag, 175–195.

Online references URL 1: http://www.oed.com/, last accessed November 23, 2012. URL 2: http://beaugrande.com/introduction_to_text_linguistics.htm, last accessed December 19, 2012. URL 3: http://www.thelatinlibrary.com/bede/bedepraef.shtml, last accessed January 8, 2013. URL 4: http://quod.lib.umich.edu/c/cme/LayCal/1:1?rgn=div1;view=fulltext, last accessed November 23, 2012. URL 5: http://quod.lib.umich.edu/c/cme/MaloryWks2/1:1?rgn=div1;view=fulltext, last accessed November 23, 2012.

Jenny Arendholz and Monika Kirner-Ludwig

13 In-between Cognitively Isolated Quotes and References: Looking for Answers Lurking in Textual Margins Abstract: More often than not, spotting and interpreting quotes is a quite doable endeavor for hearers/readers. On closer inspection, however, we find that the mere classification, let alone the interpretation of an utterance as a proper quote, is not at all self-evident in every case. In fact, there are instances of communication in which we, as recipients, rely on our subjective intuition when classifying an utterance as a quote – especially since formal conventions, e.g. inverted commas, verba dicendi, etc., are not always given. In other cases, the formal indicators that usually come with a quote are inserted, but interpreting hearers/ readers are unable to make sense of the quote within the particular context in which it is uttered, as they lack the necessary knowledge to ascribe meaning to it. They may not even find a satisfying answer to what the author really wanted to convey by placing the respective quote into a particular context. It is these cases – we will call them cognitively isolated quotes/references – which will take center stage in the following. By looking at various samples from both online message boards (The Student Room) as well as medieval manuscripts, we will investigate under which conditions a reference may even be classified as a quote proper. Although this textual scope is rather broad, we found that both these text groups, which occur 600 years apart, share one phenomenon: references and quotes being deliberately inserted in the margins as part of message board signatures on the one hand, and as marginal annotations in medieval manuscripts on the other. A diachronic comparison will be drawn between quote-like references taken from these two textual realms, which, at first glance, could not differ more – be it in terms of their chronological classification as well as their multimodal make-up. Keywords: signatures, manuscript marginalia, online message boards, quotes/ quotations, Chaucer, references

1 Introduction It is a truism that the terminology and definitions in any area of research tend to blur right before our eyes once we look at them a little closer. This is also true for quoting. Still, it is not too hard to agree on some basic components which make a quote a quote. In fact, it takes an author A1 in a context C1, (part of)

320   

   Jenny Arendholz and Monika Kirner-Ludwig

whose content is then transferred into another context (C2) by a quoter (A2¹) (cf. Bublitz, this volume). The result of this process is a quote, usually consisting of the quoted content and some sort of indication about the origin and/or the originator of the said content. Sometimes, formal characteristics such as quotation marks, italics, a change of font and the like help us identify quotes in various communicative encounters in our everyday lives. It is in this way that quotes can be distinguished from mere repetitions. However, what might at first look like a quote from a strictly formal point of view does not always qualify as such under closer examination. In some cases, we lack crucial information to distinguish between a quote and a non-quote, let alone to pinpoint the true nature of a nonquote. Thus, all that we are left with in terms of labeling such phenomena is the notion of reference, which we would like to use in the broadest sense of the term and with Lyons’ (1995: 294, original emphasis) definition in mind: […] a relation that holds between speakers […] and what they are talking about on particular occasions. The referential range of referring expressions is fixed by their meaning in the language (i.e., by their sense and denotation). But their actual reference depends upon a variety of contextual factors.

Based on our observations, we developed the following scale, ranging from rather unambiguous cases of quotes to vague references: 1. generally identifiable and comprehensible quotes; 2. generally identifiable quotes, comprehensible, however, only by peers; 3. fragments of text lending themselves to be read as quotes; 4. completely opaque fragments of text. To find instances of questionable quote specimens, which might as well be labeled references in some cases, consider, for instance, the three examples dis-

a)

b)

c)

Figure 1: Questionable specimen of quotes 1 This does not, however, mean that the quoter cannot also be the person quoted. In this case, the quoter quotes themselves, which is why they would also have to be labeled A1.

Cognitively Isolated Quotes and References   

   321

played in Figure 1. While examples a) and b) stem from our own office, example c) was taken from an advertizing pillar in Cologne, Germany. Admittedly, only example a) is formally indicated as a conventional quote, the utterance being embraced by quotation marks (“Neulich habe ich den dicken, fetten Chaucer mit ins Bett genommen”, ‘I recently took fat, old Chaucer with me to bed’). Also, the source is named (“G. Reinhold 23.7.1985”). But how can we justify categorizing b) and c) as quotes as well? We need to rely on hints: In b) a pseudonym source is given in brackets, viz. “(die Archivdrachen)” [literally: the dragons of the archives]. This and the fact that the utterance is presented in ‘eye-dialect’² imply that we are faced with a text quoted from a Swabian dialect speaker. Example c) is even less quote-like and thus prone to be classified as a reference. The German text inserted into the advertizing poster reads: “Und Hammer so ein Schitt zu haben! Fragt Thor!” (literally: Awesome having this shit! Ask Thor!). The only hints that point to this text actually being quoted are, for one thing, the second sentence referencing, most likely, to a pseudonym (“Fragt Thor!”), as well as the grammatical deficiencies and the poor lexical register. The latter suggests that it is a transcript of an utterance that was just heard by the quoter (or just referencer?) and deemed adequate and worthy to be noted down for everyone passing by to read. It remains, however, unclear what utterance had preceded the supposed quote beginning with the German coordinator und (and). What is more, the extra-linguistic and deictically indicated referent of ‘this shit’ is equally opaque, with drinking water as displayed on the advertisement being just one of the most obvious interpretations. Although all three examples present varying degrees of cognitive isolation, partly missing formal indicators for quotes, all of them share a prominent position as they were obviously put out there to be perceived as quotes. The deliberate allocation of text to a marginal but prominent and conspicuous place is one of the main criteria we will take into account in our study, when it comes to quotes and quote-like utterances. Obviously, not every reader will agree with us in interpreting every example as a quote. In fact, this will be the basis of our line of argumentation, viz. that quotes are recognizable only to those who perceive them as such in any given utterance (‘X is a quote if I understand X as a quote’). We often lack knowledge about the source context C1, the quoter and most of the time also about the author A1, including episodic knowledge of how these utter-

2 This notion, coined by George P. Krapp (1926), is generally used to refer to a literary technique deliberately applied by an author to render a speech situation more vivid and real. Since we infer this intention in the quoter of this utterance and also believe to understand the additional evaluative attitude in it, the term eye-dialect seems appropriate in this particular context.

322   

   Jenny Arendholz and Monika Kirner-Ludwig

ances came into being in the first place and why someone deemed them valuable to quote. Nevertheless, the value we see in classifying such utterances as quotes may allow for particularly functional or pragmatic aims on the quoter’s side, assuming that an act of quoting will generally be conducted on grounds of (a) certain intention(s). But will a quoted utterance that can no longer be assigned to either quoter, quoted source or, particularly, communicative purpose, not lose the pragmatic function originally intended by A1? These preliminary thoughts indicate some of the problems and peculiarities that the present paper tries to tackle with regard to this challenging subject that we decided to call cognitively isolated quotes, beginning with the subjective perception of any such utterance as more or less quote-like. As of yet, these kinds of quotes have not attracted much academic interest, which is why this contribution investigates cognitively isolated quotes in two textual areas. We juxtapose instances of computer-mediated communication, i.e. signatures found in the online message board The Student Room (TSR for short, URL 1), with a selection of marginalia inserted into medieval manuscripts. The following explorative, qualitative and diachronic comparative study sets out to examine cognitively isolated quotes, with the aim of answering the following questions: What are their forms and functions? What are the diachronic parallels with regard to appearance and aims conventionally expected in and from ‘classic’ quotes? A brief outline of the forms and functions of ‘conventional’ quotes in section 2 provides a starting point for contrastive observations. Section 3 will present two sets of samples for the ‘now’ and ‘then’ and discuss the varying degrees of cognitive accessibility of these quotes. Section 4 will then focus on cognitive strategies for the establishment of common ground between a quoter and their audience(s), thereby drawing on various kinds of building blocks for reconstructing context and, thus, establishing present-day coherence. Our case studies will close with a functional reassessment of potential motives behind cognitively isolated quotes (section 5).

2 Quotes and the Act of Quoting: What We Expect to Find in Online Message Boards In order to showcase cognitively isolated quotes in message boards, we need to review regular quotes and complete acts of quoting. A first sample taken from TSR may serve to demonstrate the elementary components and processes within this particular act of online quoting while at the same time introducing one of the two objects of study.

Cognitively Isolated Quotes and References   

   323

The Student Room, which serves as the source for our present-day examples, is a UK-based message board system that attracts British and international students, future students as well as wannabe students (see Arendholz this volume). As one of the “world’s largest and fastest growing student community, with over 250,000 members and more than 16,000,000 posts on our forum” (URL 2), TSR deals with university, relationships, life in general and every other conceivable topic that users feel inclined to discuss with like-minded others. More often than not, heated discussions turn into multilayered and branched polylogues that call for a simple yet unambiguous way to establish reference between single contributions. Since TSR does not feature a tree-like structure as online fora do, it is not always clear which comment refers to which. In TSR, posts are displayed in a flush-left arrangement. Similarly to other message boards and other forms of computer-mediated communication, TSR offers a quote-button to facilitate referencing previous contributions. The following screenshot (Figure 2) shows two random, subsequent messages (post #23 and #24³) from a Student Room discussion.

Figure 2: Use of the quote button to establish topical reference between two subsequent posts

As can be seen in Figure 2, Steeps.’ original post is quoted by cpj1987 by means of clicking on the quote-button on the right hand side. This results in Steeps.’ contribution reappearing as a formally marked quote in cpj1987’s subsequent post, thereby clearly linking both contributions to each other. We can therefore state 3 Note that all of the online examples are taken from the Message Board Corpus compiled by Arendholz (2013). Differences with regard to the most recent TSR design are only marginal and do not affect the basic mechanisms of quoting, let alone the content of quotes.

324   

   Jenny Arendholz and Monika Kirner-Ludwig

that a source text of an author A1 (Steeps.) is shifted from its context C1 (post #23) to another context C2 (post #24) by a second author A2 (cpj1987), who is also called the quoter. Thus, the act of quoting is set in motion by author A2’s clicking on the aforementioned button. Formally turning a contribution into a quote, i.e. adding a frame and quotation marks, is then automatically implemented by the message board system. The act of quoting, however, involves one important additional step when compared to mere quotes: As indicated by the text underneath the framed box in Figure 2, A2 (cpj1987) comments and thus evaluates the quote, be it explicitly or implicitly. In this sense, quotes are an integral part of the more complex act of quoting, with both quotes and acts of quoting being vitally important for the argumentative structure of texts and the progression of the ongoing conversation. As can be seen in the example in Figure 2, quotes are embedded in different layers of communicative co- and context, turning their interpretation into an ordinary step in conversation for readers, or, in our case, users. While the act of quoting can frequently and typically be observed in entries, as shown in Figure 2, not every quote in message boards comes with an evaluation that turns it into a full-blown act of quoting. In fact, when looking at signatures, which can be compared to creatively and individually crafted business cards at the end of some message board contributions, all we have are plain quotes without any trace of evaluation by A2s⁴ (see Figure 3).

Figure 3: Message board signature containing a quote

4 It should go without saying, though, that the mere presence of a quote without further evaluative commentary does not stand on its own. On the contrary, the quote alone can be taken as a springboard for interpretation by its interlocutors.

Cognitively Isolated Quotes and References   

   325

Again judging the quote in the signature displayed in Figure 3 by its formal appearance, we can be almost certain that we are in fact dealing with a quote, as it bears typical trademarks such as inverted commas and a hint about the A1, i.e. the source William Shakespeare. This is, however, not always the case. There are also signatures that enclose content that could be interpreted as quotes for some formal reasons (e.g. tildes, change of color and fonts and the like) and – as we will show in sections to come – heavily rely on potential, interpreting interlocutors for the establishment of reference to some original source C1.

3 … and What We Sometimes Find Instead: Cognitively Isolated Quotes Now and Then Signatures in TSR contributions are a rich source of all kinds of verbal and visual quotes and quote-like utterances borrowed from a wide range of C1 sources such as literature, movies, music, politics and both historical and contemporary A1 originators, including fellow TSR-members. However, contrary to the formal and functional delimitations of the quotes outlined above, the following quotes lack quote-specific indications and are thus recognizable as quotes only to a certain degree. The quoter’s original communicative aim may (have) become opaque. Thus, even if a recipient recognizes an utterance as a quote despite its formal unmarkedness, the cognitive connection between this utterance (T2) and its new context C2 can at best be interpreted only vaguely. Examples (1)–(5) are random samples taken from TSR as well as from medieval manuscripts: (1)

~ Life isn’t about waiting for the storm to end… it’s about learning to dance in the rain ~ [originally set in red letters]

(2)

If you strike me down, I shall become more powerful than you could possibly imagine.

(3)

326   

   Jenny Arendholz and Monika Kirner-Ludwig

(4)

“Maybe now I know how robots feel, because the emotion is more like a feeling that anything else. Not like a feeling that is a synonym FOR emotion, but feeling like a physical sensation. Only it’s not physical. It’s somewhere in bet n physical and mental. Maybe even emotional.” [originally set in purple letters]

(5)

Mr Jelly That’s with an “E”, not an “O“! [Jelly, E and O originally set in red letters]

Most of us would agree that examples (1) to (5) all bear at least one formal feature which justifies interpreting them as quotes. While none of these examples indicate the original source, we still frame them as quote-like contents based on formal elements such as tildes, quotation marks, italics or changes of colors. Yet, examples (2), (3) and (5) almost entirely escape this formal classification. The only reason for this kind of interpretation can be found in their positioning, viz. as part of signatures, which is commonly associated with the use of quotes. From a functional point of view, however, there is a noticeable increase in opaqueness, or, as we would phrase it, cognitive isolation, when interpreting examples (1) through (5) and when looking for meaning, let alone originators: While the content of the first three examples can be understood perfectly, only readers who dispose of the necessary background knowledge of Star Wars and of The Hitchhiker’s Guide to the Galaxy respectively will be able to interpret examples (2) and (3) as quotes. Note, however, that this kind of knowledge does not reveal the purpose of the image in which the quoted passage (in (3)) is embedded. The last two examples are entirely opaque in terms of their meaning as well as their source context C1 and author A1. Although we categorize all five examples as cognitively isolated, examples (4) and (5) are by far the most cryptic examples, thus begging the question: What is the point of these references? And more importantly: Do all of these deserve to be classified as quotes? Before trying to provide answers to these questions, we will shift our focus to written discourse found in marginalia inserted into medieval manuscripts. The diachronic text samples in the following, (6) to (10), show a strikingly high degree of correspondence to the modern ones just outlined. (6)

For the book seith, ‘Axe alwey thy conseil of hem that been wise’ (CT, ToM ll. 1152f., in: Benson 1988: 223)

In sample (6), we find a quote which appears to be accompanied by formal markers. The quoted utterance, literally saying ‘Always ask those for advice who

Cognitively Isolated Quotes and References   

   327

are wise’, is indicated as such by the verbum dicendi “seith” ([lit.] ‘says’). The quotation marks and all further indications of punctuation, however, have only been provided by present-days editors of this 14th century text. Although they have become conventional nowadays, this was definitely not the case in Chaucer’s lifetime or in scribal practice in general up to the late Middle Ages or the beginnings of the Early Modern English Period (see Aijmer, see Moore, both this volume).⁵ Still, there is no doubt that we are faced with a quote here, despite the fact that the origin of or authority behind it cannot be identified. It may have been a widely-known saying or a traditional piece of knowledge that we are just not familiar with (anymore). The reference to “the book” could be a reference to the book, i.e. the Bible, as the highest possible authority, however no concrete piece of evidence seems to have been identified so far. What is more, sample (6) is but one of about 242 quotes and references appearing in Chaucer’s Tale of Melibee (Canterbury Tales). This particular tale is basically a conglomerate of quotes, one following after the other – partly formally indicated by verba dicendi – without leading to any sensible conclusion. And although the content of each quote is understandable and can be embedded into the common ground held between modern readers and the medieval quoter, it seems that the act of quoting is only conducted for the sake of quoting itself, not for the sake of the message contained in each quote. Nevertheless, a certain context is given, such as intertextual linkages to other texts by Chaucer and to pragmaphilological characteristics of Chaucer’s usage of quotes (cf. Pakkala-Weckström 2010), supposedly inserted as means of self-reflection regarding his literacy. Due to the lack of clues available for this example, the degree of its cognitive isolation is fairly low. Then again, examples (7) and (8) show quotes that are not only isolated from the contents of the main text – as they are found in the margins of manuscript leaves – but that are also formally unmarked as quotes, which points to a much higher degree of cognitive isolation. (7)

Loquere Domine quia audit seruus tuus/ Audiam quid loquatur in me Dominus Deus (Red Ink Annotator of the Book of Margery Kempe: 71r/30–31, interlinear/ lower margin; from I Samuel iii, 8–10)

(8)

e cum lapsu temporum labantur actiones homi (MS KL Herrenchiemsee 13, fol. 9r, Bavarian State Archives, Munich)

5 Cf. e.g. Parkes (1992). Other scholars have argued, though, that modern punctuation inserted in text editions originally unpunctuated has the high potential of falsifying the original sense by disregarding sentence borders (cf. e.g. Robinson (1971), Chap. 7 on “Manuscript Punctuation”).

328   

   Jenny Arendholz and Monika Kirner-Ludwig

With regard to the Latin quote in sample (7), modern-day technologies, viz. database and internet searches, shed light on the actual source and support our intuitive classification of it as a quote. In fact, it can be identified as a passage appearing in the Bible, I Samuel iii, 8–10. Although the quote itself is generally ascribed to the Red Ink Annotator, knowing this still does not indicate to whom we owe this quote. The formulaic invocation calling for a word from God ([lit.]: ‘Speak Lord, as your servant is listening. I will hear what my Lord speaks’) was inserted by a male scribe. The gender of the scribe can easily be identified considering the fact that we are dealing with a medieval scribe. Moreover, this is even made explicit through the wording “seruus tuus”. This also means that no reference whatsoever is intended to the first-person narrator of the main text, the female Margery Kempe.⁶ At the same time, the content of the quote is quite general and therefore does not clash with the actual main text on the same folio. Still, the immediate coexistence between the two parts of the text remains arcane. While sample (7) is a piece of text inserted in an interlinear fashion into the lower margin of MS Add. 61823 fol.71r/30–31 (British Library) by the so-called Red Ink Annotator (16th century) of the Book of Margery Kempe (see 4.2.1), sample (8) displays a text which was noted down in the lower margin of fol. 9r of the MS KL Herrenchiemsee 13 (Bavarian State Archives, Munich) around the middle of the 12th century. Of all the folios of the charter, sample (8) was placed right next to the main text, which refers to Konrad I of Salzburg’s founding of the Abbey Herrenchiemsee, which is particularly noteworthy since this part of the main text (predominantly in Latin) is clearly one of the most prominent passages in the whole charter.⁷ The two lines of classical Latin read “e cum lapsu temporum labantur actiones homi”, which can be understood in various ways, e.g. ‘In the course of time the actions of men pass into oblivion’ or ‘In the course of time the actions of men are subject to variation’. Whichever of the two readings or any other was truly intended, we may certainly assume all kinds of connections with the main text, but the communicative purpose remains highly ambiguous. In fact, it may even be nothing more than a pen trial, which was inserted into the margin just to probe the ink, the pen or one’s own hand-writing. Again, the prominent position in the margin, as well as the solemnly metrical wording of philosophical content which seemingly expresses a piece of the scribe’s mind, leads us to believe that we are dealing with a quote.

6 Margery Kempe was a female mysticist (died after 1438), who had a scribe write down her autobiography, which also entailed her pilgrimage to foreign countries and cultures, e.g. the Saracens. 7 We thank Birgit Gilcher from the Bavarian State Library Munich for this suggestion.

Cognitively Isolated Quotes and References   

   329

Sample (9), another annotation by the Red Ink Annotator (cf. example (7)) in Brit. Lib. MS. Add. 61823, fol. 43v, and sample (10), the so-called Canterbury Spell in Brit. Lib. MS Cotton Caligula A.xv fols. 123v and 124r, are particularly curious for the following reasons: not only do they present themselves as multimodal and visual quotes, but we, as recipients, also fail to comprehend their contents, their reference(s) and thus the aim of their quoter/referencer. In order to do justice to their complex makeup, we will discuss them separately in section 4.2 below.⁸ (9)

(Red Ink Annotator of the Book of Margery Kempe: MS Add. 61823 fol.71r/30–31, British Library, London) (10) (Canterbury Spell, MS Cotton Caligula A.xv, fols. 123v and 124r, British Library, London) In summing up this section, we can state that in both of the textual spheres of investigation compared here, cognitively isolated quotes cannot be integrated into the argumentative structure of texts and do not contribute to the progression of the ongoing conversation (used here in the widest sense of the word) – at least not in an obvious and cognitively salient way. We therefore face two interconnected problems: Firstly, we cannot be entirely sure if we are being presented with a quote at all; and secondly, we cannot be entirely sure of the purpose of this text passage. Against this backdrop, the following sections set out to find (sources of) evidence suggesting that we are in fact dealing with quotes and the kinds of communicative functions that can be ascribed to them. In this context, it is of course helpful to find out as much as possible about the quoter of the content that we believe to have identified as a (cognitively isolated) quote.

8 Both images in (9) and (10) as well as those in section 4.2 are here displayed by kind permission of the British Library.

330   

   Jenny Arendholz and Monika Kirner-Ludwig

4 Assessing the Surrounding Co(n)text of Cognitively Isolated Quotes 4.1 Approaching the Quoter’s Identity As has already been shown, referring to quotes as isolated mostly pertains to their formal separation and unmarkedness. When it comes to the aims and functions of a quote that was (un)deliberately rendered cognitively inaccessible, there is a cognitive gap between ourselves as the audience and the originator. Although this gap can hardly ever, if at all, be bridged, this section still looks at the co(n)text of cognitively isolated quotes. Whereas on a medieval manuscript leaf, the quote can be found in any of the four margins that usually frame the main text or even in interline position, the signatures in TSR are positioned exclusively below the main text, i.e. below the typed message board entry that conveys the actual communicative content. Despite their obvious inflexibility in terms of position, TSR-signatures turn out to be a rich source of co- and contextual input, as virtually every conceivable content can be inserted in and transmitted by the section underneath the main part of the message. Ideally, we find a full display of personal data, including pictures, information about the user’s school, courses, even grades, as well as implicit or explicit information about hobbies and friends. Even if we are only given snippets of the user’s identity, these might be enough to identify the user, i.e. the quoter, and thus to learn something about the original context C1 and/or the original author A1 of the cognitively isolated quote under investigation. Other ways of trying to find out about the person behind the nickname – thus gaining more information about the quote-like passage – include investigating various kinds of message board templates (see box in the upper half of Figure 4) and profile information (see Figure 5). Each and every post comes furnished with a range of additional information about the user such as facts about his/ her present location, gender, membership in TSR and present status within this community of practice. As this info is neatly presented in templates surrounding the entry, readers do not have to click or leave the current website to learn more about a certain user. Note that, in contrast, with profile information, interested users have to click on a link to be directed to another site separate from the ongoing discussion in order to access more information about the quoter. Still, both templates and profile may provide valuable insights into the background of the user and thus contain clues about the true nature, meaning and purpose of the (cognitively isolated) quote.

Cognitively Isolated Quotes and References   

   331

It goes without saying that the information given there does not necessarily have to be correct and/or authentic – “errors”, be they honest, humorous or even strategic, do occur. But even when we are given detailed information within (one of) these categories, i.e. if the user filled in these templates, we cannot expect this contextual input to be helpful and the quote to automatically make sense. Coming back to sample (4) of cognitively isolated quotes (“Maybe now I know how robots feel, because…”, see section 3, above), the following screenshots bear witness to this point:

Figure 4: Cognitively isolated quote in context

Figure 5: Profile site of lilythrash21, creator of cognitively isolated quote

332   

   Jenny Arendholz and Monika Kirner-Ludwig

Despite the amount of given information, provided either automatically by the message board system or by the users themselves, we are still left in the dark about the origin and the meaning of the quote contained in the signature, which is why it remains cognitively isolated. Thus, introspection seems to be the main tool for shedding some light on cognitively isolated quotes – be it in medieval manuscripts or in modern computer-mediated textual frames. Our experience and our background knowledge concerning e.g. message board interaction might also help us make sense of such quotes. This is especially true if we have, for some reason or another, formed a relationship with the quoter and thus share some common ground with them. Note that, for obvious reasons, this is not an option with respect to medieval quoters. Still, two case studies with highly intriguing specimens of quotes which are embedded in a multimodal context will be presented in the following.

4.2 Two Medieval Case Studies 4.2.1 The Red Ink Annotator in Brit. Lib. MS. Add. 61823, fol. 43v In medieval manuscripts, the identity of a scribe is at best hard to pinpoint, since he is not the actual creator of the text but only a person who documents content someone else conceived before him. Accordingly, there is only so much that we can learn about a quoter-scribe, e.g. by determining the age, kind and origin of the ink he used and by looking at his handwriting, which might parallel others’ classified before.⁹ Apart from estimating the timeframe and location of the creation/compilation of a medieval manuscript, trying to identify its quoting scribe principally means using the wrong approach, as it cannot serve as a foundation for interpretation. This problem becomes particularly evident when looking into the 16th century Red Ink Annotator (cf. examples (7) and (9), section 3) and into one of numerous manuscript leaves in which he left traces – the first of two medieval examples for restricted cognitive accessibility, presumably both to a medieval as well as to a modern recipient. When trying to illuminate our quoter-scribe’s background, viz. the Red Ink Annotator’s, we fail because our knowledge is limited: Judging by his critical annotations, he must have been well-read and was most likely a man of the church (cf. Fredell 2009: 18). Based upon the dating of an ownership indication on the binding leaf of the manuscript, he is commonly allocated to 9 See, for instance, the characteristic handwriting of the scribe commonly referred to by ‘the tremulous hand of Worcester’ (13th century, e.g. Bodl. MS. Hatton 113, f. 68; cf. Franzen 1991).

Cognitively Isolated Quotes and References   

   333

the Carthusian priory of Mount Grace in Yorkshire around the late 15th/early 16th century (Meech/Allen 1940: xxxii, xxxvi). This and the red ink he constantly uses in the only extant manuscript of the Book of Margery Kempe (a1450, according to Meech/Allen 1940: xxxiv) are the only obvious identity-specific hints he has left us. However, the marginal annotations he inserted do indeed seem to sketch certain intentional patterns: Not only did the Red Ink Annotator refer to the main text for reasons of emphasis, summary, comment and emendation, he also drew directly on older annotations by earlier scribes, emphasizing or even agreeing with them. This kind of scribal interaction provides an additional base for dating him (Fredell 2009: 8ff.).¹⁰ It also suggests that he wanted to ascertain that his personal opinion, even when already expressed by an earlier annotator, was received. Fredell (2009: 11ff.) points out that not all annotations in red were produced by one and the same ‘Red Ink Annotator’,¹¹ an assumption already stated with reservations by Meech/Allen (1940: xxxvii), whereas Parsons (2001) and Erwin (2006) leave this matter untouched. For that reason, intertextual links between the red ink marginalia in the manuscript remain somewhat spurious. Nevertheless, it seems more than likely that the numerous annotational cross-references from Margery Kempe’s mystical experiences described in the main body of the text to respective utterances by medieval mysticists such as Richard Methley and Richard Rolle have been inserted by one and the same annotator. In the following, we will argue from this assumption. Figure 6a displays the complex conglomerate of references presented by the Red Ink Annotator. Although the scribe remains only a name, we are nevertheless capable of resolving most of the cognitive isolation. The drawing of the “Flames of Divine Love” (Parsons 2001: 159) is explicitly described at the bottom of the flame, reading “ignis D[e]/Divini amoris”. This visual quote, which picks up the accompanying text passage, forms part of a multimodal body of reference. As can be seen, the drawing itself extends over about eight lines of the text written on fol. 43v of MS. Add. 61823 (London, British Library), and ten lines when also taking into account the written annotation below the flame, making this insertion an

10 Fredell provides examples showing that the Red Ink Annotator tended to comment on annotations and emphases made by other scribes (cf. 2009: 16f.). 11 In his 2009 paper, Fredell distinguishes three annotators using red ink, viz. “Ruby Paraph”, identifiable by the recurrent repetition of red paragraph markers (2009: 11), “Big Red N” and “Red-Ink Annotator” (2009: 12f.). Apart from that, he also names three more scribes involved in annotations of various kinds, viz. “Little Brown”, assigned to Scribe Salthows, who produced the manuscript in the 1440s (2009: 3ff., 18), “Big Brown” and “Big Black”.

334   

   Jenny Arendholz and Monika Kirner-Ludwig

Figure 6a: MS. Add. 61823 fol. 43v (London, British Library)

Figure 6b: Magnified display of the Red Ink Annotator’s visual reference

unmissable eyecatcher.¹² Throughout these lines of the main text body, instances of lexico-semantic cohesion, e.g. referring to “fyer” (‘fire’), the “fyer of loue brennyng” (‘fire of love burning’) and Margery “feeling” all this (see first word in line next to “hampall”), matches the visual of drawn flames and their respective associations. The reference beneath the drawing, “so s. R. hampall” (cf. Figure 6b), is more problematic to demystify. This reference must have been a “cognitive pointer” to Richard Rolle (see Meech/Allen (1940: xl) and Parsons (2001: 146)), since “hampall” points to Hampole, in Doncaster, South Yorkshire, where this 14th century writer and mysticist came from. Even if a medieval recipient had been 12 We very much wish to thank Joel Fredell for his valuable advice and insightful suggestions for this section.

Cognitively Isolated Quotes and References   

   335

able to decipher this reference and had known about the connection between Hampole and “R.” (short for ‘Richard’), the “so” in the comment remains unclear. Is it possible that Rolle’s experience with godly love had been a personal one that he himself wrote about and could thus be paralleled to Margery’s visions? Could the Red Ink Annotator have been convinced that the multimodal display of drawing and text would be easy to link cognitively due to the conventionality of this symbol and the abbreviated text added? Or did he insert this cryptic reference deliberately, intending for only a limited circle of users capable of comprehending his message? Although some of these questions must remain unanswered at this point, we can state that the Red Ink Annotator clearly bore a particular interest in and experience with mystical texts (cf. Fredell 2009: 18), providing a number of clues for the recipient of both the main text and his annotations to link the complex multimodal information. It is thus possible to conclude that the annotator made intentional use of cognitively isolated quotes: We can assume that he wanted us to pick up on the intertextual pointer in order to increase our knowledge. At the same time, however, he must have been aware that this elaborate reference would not be cognitively accessible to just everyone, but certainly only to those ‘on the same page’ as him.

4.2.2 The Canterbury Spell in Brit. Lib. MS Cotton Caligula A.xv fols. 123v and 124r The second example (see example (9), section 3) that we have chosen from medieval marginalia in manuscripts may be the epitome of the cognitively isolated quote. In this case, we not only have problems trying to identify the source of the quote (C1) and the quoter-scribe, but also his intentions and target readership. The so-called Canterbury Spell (Lindquist 1923: 28f.),¹³ which we find in the bottom margins of folios 123v and 124r in MS. Cotton Caligula A.xv, commonly assigned to the Canterbury Cathedral (thus the name), has not been looked into much since the beginning of the 20th century (cf. Sperber 1911, Lindquist 1932), most likely because of evident difficulties in making any sense of it.¹⁴

13 Frankis (2000: 1f.) prefers to refer to this text by the name “Canterbury Runic Charm”. 14 There is, however, one recent exception. In his paper, Schiltz (2004) approaches this particular spell in order to locate the textual criteria laid out by de Beaugrande/Dressler in 1981. From the perspective adopted in the present paper, it becomes clear that this heterogeneous text bears far too many complexities as to classify it according to the rigid template of such a model.

336   

   Jenny Arendholz and Monika Kirner-Ludwig

Figure 7a: Brit. Lib. MS Cotton Caligula A.xv fols. 123v and 124r

The main text on the two folios in Figure 7a presents a computistic table used to calculate Easter. Most of the folios 120r to 139r, which make up the second of three fragments of the codex MS Cotton Caligula A.xv,¹⁵ display similar tables, some prognostications and further calculatory entries concerning the liturgical year. The table on folios 123v and 124r, to be discussed in more detail in this section, fits perfectly into this wider context. After the codex’ production, which can quite safely be dated to 1075 (Schiltz 2004: 133), its contents saw a period of approximately 400 years of constant clerical usage, resulting in various annotations and extensions of the computistic tables. According to Schiltz, the runic inscription must have been inserted somewhere between 1075 and 1109, the latter being the year of the computistic annals switching from Old English to Latin (2004: 133). The blackened erasure at the bottom left beneath the table, which separates the table from the runic sequence on fol. 123v, must have been inserted by a later, 16th century reviser of the manuscript, who used the same kind of felt pen to paint over five lines of Jerome’s prologue to De Viris Illustribus on fol. 3 (Schiltz 2004: 128). What lies beneath that layer of ink has not yet been revealed. For this

15 The first fragment (fols. 1–119) displays Latin texts from the 8th and 9th century and was once possessed by St. Augustine Abbey, Canterbury (Schiltz 2004: 124), and the third one (fols. 142– 153) contains an excerpt of an Old English translation of Bede’s De Temporibus Anni (ibd.: 125).

Cognitively Isolated Quotes and References   

   337

reason, it cannot be ruled out completely that whatever is covered might help us draw a connection between the main text and the runic spell. This hypothesis is supported by the fact that the blackened entry appears to have begun further left than the table above, but still flush with the charm below (cf. Schiltz 2004: 128). It is, however, the runic inscription that deserves our particular focus. Sperber (1911) identified this line of text as an incantation of Cyril, the Norse God of wounds, whom the scribe implores to leave his or someone else’s body (cf. Schiltz 2004; Lindquist 1932). The fact that we face an unbroken, linear line of graphemes is noteworthy. It might either point to the scribe having copied the text from an object also displaying these symbols, or he might even have been aware of the importance of runes being written in an unbroken, continuous unspaced flow – a prerequisite if one wanted the charm to work its magic. This would suggest that he actually knew what he was noting down. Figure 7b displays the erasure, the magnified inscription, its transcription and a ModE translative approach following Moltke (1985: 360; cf. Schiltz 2004: 121).

Kuril sarþuara far þu nu funtin is tu þur uigi þik [þ]orsa trutin iuril sarþuara uiþr aþrauari Figure 7b: Magnified image of the Canterbury Spell and its transcription Kuril sarþuara Cyril

with your wounding staff

far þu nu funtin is tu

out you drive, you are found

þur uigi þik [þ]orsa trutin

Thor may bless you, lord of Thurses

iuril sarþuara

Cyril with your wounding staff

uiþr aþrauari

Against atter [pus] in the veins

In contrast to the Red Ink Annotator’s case, the Canterbury Spell could be called a cognitive one way street which poses, amongst others, the following unanswerable questions: If the rune scribe was a cleric, why would he mutilate one of the most prominent and important folios in this manuscript by adding an incantation to a heathen god? He must have known that the computistic table was used regularly. If he bore any such superstitious or heathen thoughts in mind, was his fear of being blamed for these the reason for his writing in runes which presumably no one in the 14th century would be able to read or understand? And how did he himself come to write these symbols, which are not runes from the English inventory but from the Danish (cf. Schiltz 2004: 121)? Could it, after all, just have been meant as a simple pen trial?

338   

   Jenny Arendholz and Monika Kirner-Ludwig

Most probably, the rune-scribe copied this text from somewhere else, possibly not even understanding its content and thus not being aware of its blasphemous message. If he did understand the meaning of the spell, he must have been pursuing some communicative goal with it rather than just trying his ink and pen. In any event, we cannot comprehend or reconstruct the meaning of the spell and neither would have anyone unacquainted with the scribe. Only those who knew the scribe could have been informed about e.g. his particular interest in runes or the illness which he might have wanted to cure by copying or maybe even reciting this spell. The most important point we would like to deduce from this curious example may come as a surprise: we argue that in many respects, the Canterbury Spell as a cognitively isolated quote comes rather close to message board signatures. After all, in both cases we face formally isolated texts which are still tentatively classifiable as quotes and exhibit traces of clipped-off cognitive paths towards comprehending their relation to the main text and the underlying quoter’s intention. The last section will put forward some speculations about the potential communicative aims and intentions of using cognitively isolated quotes.

5 Why Use Cognitively Isolated Quotes? An Assessment of Potential Functions As has become obvious by now, little can be said about the motivations behind cognitively isolated quotes, even less so if the quoter is unknown to the (present day) audience. All we can do is speculate about possible reasons for inserting opaque content into otherwise valuable and most of all purposeful communicative material. It should go without saying that this assessment of potential functions only considers a limited range of deliberate and conscious choices made by the quoter. We cannot, however, rule out mishaps, pen trials, or any passing thoughts considered noteworthy at that time, hence in no way aiming for a reaction from a potential audience. Inserting a quote into a text in order to express one’s level of sophistication to the audience seems to be a timeless yet effective strategy. Adopting Goffman’s (1955) perspective, a quoter makes use of quotes in order to be seen in a positive light, i.e. to be given a desired, favorable face by his audience. Using quotes is thus comparable to making face claims (cf. Arendholz 2013). The quoter can also show his/her affiliation with a certain group. With regard to examples (2) and (3), the quoters thus position themselves as connoisseurs, maybe insiders or even fans of Star Wars and The Hitchhiker’s Guide to the Galaxy respectively – simply

Cognitively Isolated Quotes and References   

   339

by disposing of lexical knowledge typically shared among members of these particular communities of practice. The same can be said about the anonymous Red Ink Annotator’s deep interest in and expertise regarding e.g. Richard Rolle’s mystical experiences (cf. 4.2.1). Even example (10), the Canterbury Spell, may be the outcome of a scribe’s desire to show his secret affinity for runes and/or heathen superstitions. The effects of such quotes are hence twofold: With regard to the speaker, they place the quoter within a certain group. With regard to the recipient, however, they also bear the potential of differentiating between peers and non-peers. While peers dispose of the necessary frame knowledge to situate the (not so) cognitively isolated quote immediately, non-peers lack this common ground. For them, cognitively isolated quotes remain incoherent passages of text and, as such, at most vague references but not quotes. Unfortunately, most of the above examples have marked us as non-peers incapable of cracking the code.

6 Conclusion: Quotes, References and Degrees of Cognitive Isolation In order to establish a diachronically coherent bridge between the modern examples taken from the TSR message board corpus and the medieval marginalia, we would like to point to the parallels observed for cognitively isolated quotes both now and then. The most remarkable similarities of these two textual areas are their position as accompanying elements in the margins of main bodies of texts, as well as their veiled contents. From a functional point of view, we assumed similar intentions for both medieval scribes and contemporary writers. Considering that we were not able to make sense of many of the medieval, or some of the TSR samples of isolated quotes/references, it seems to us that the chances of belonging to an in-group capable of cognitively de-isolating such quotes was and still is equally slim both then and now. Time may thus be but one influential factor in restricting a recipient’s cognitive access to quoted material, even though it does not seem to be the most decisive one. As we pointed out in the case of the Canterbury Spell, the linguistic code used seems to be just as important and influential in the deciphering process. From a formal point of view, we find a number of apparently timeless parallels between the two textual areas under investigation. Due to their relatively prominent position, both sets of cognitively isolated quotes naturally catch the text recipient’s eye. This can further be enhanced by means of visual highlighters and graphical elements, both in medieval as well as contemporary cases. Apart from this, the seemingly conventionalized formal indication of quotes (from a

340   

   Jenny Arendholz and Monika Kirner-Ludwig

modern-day perspective), i.e. quotation marks, sometimes appears to be regarded as simply optional and even neglectable in message board signatures. Instead, we found idiosyncratic and creative ways to indicate the insertion of quoted material, among these the use of italics, different coloring and changed fonts. However, we need to remind ourselves that medieval quoting conventions – if we may even be so anachronistic as to use this notion – do not display any such indicators, but may instead only provide a verbum dicendi as the introducing element of a quote (see Aijmer, Moore, this volume). We may conclude that a quote that does not reveal itself as such to the recipient, be it formally or functionally, has certainly lost much of its communicative purpose, including the one originally intended by the quoter. However, as soon as a quote, irrespective of how isolated it is perceived in the eyes of the recipient, is recognized as such, as in our approach, then it is assumed to be one. It will thus trigger cognitive processes, even if these have nothing to do with the ones originally intended by the quoter. This even holds true if its classification as a quote was a false assumption in the first place. The functional potential is therefore only partly dependent on the formal recognition of a quote. Based on this observation, recipients will automatically estimate the degree of cognitive isolation of the utterance at hand on a highly subjective scale, opening up potential stages of accessibility. Note that all of our examples (1) to (10) can be integrated into the following scale: 1. generally identifiable and comprehensible quotes (1), (6); 2. generally identifiable quotes, comprehensible, however, only by peergroups (2), (3), (7); 3. fragments of text lending themselves to be read as quotes (4), (8), (9); 4. completely opaque fragments of text (5), (8), (10). Revisiting our examples, we cannot help but wonder if the notion of (cognitively isolated) quote is really the appropriate term to use for all four points on our scale. Considering the fact that we cannot be sure if we are in fact dealing with a quote, at least for the phenomena grouped in 3. and 4. on the scale, it seems sensible at this point to use another term. We propose two terms for each pole of our scale, viz. 1. cognitively isolated quotes and 4. cognitively isolated intertextual references, with a lot of room for discussion in between. After all, the act of quoting and quotes as one of its essential components may be regarded as a particular strategy within the frame of reference. In addition to Lyons’ definition given above (see section 1), we would like to stress that by references, or the act of referring, we also mean the “cognitive, collaborative, constructive and dynamic action performed by someone” as defined by Bublitz/Hoffmann (2011: 435, original emphases). In our case, referring both pertains to a quoter laying an associative and intertextual

Cognitively Isolated Quotes and References   

   341

trail as well as the recipient, who will, based on their individual interpretation, try to cognitively retrace it. These two cognitive and communicative processes are prone to remain the same, both in synchronic and diachronic analyses.

References Arendholz, Jenny. 2013. (In)Appropriate Online Behavior. A Pragmatic Analysis of Message Board Relations. Amsterdam: John Benjamins. Beaugrande, Robert-Alain de and Wolfgang Ulrich Dressler. 1981. Einführung in die Textlinguistik. Konzepte der Sprach- und Literaturwissenschaft 28. Tübingen: Niemeyer. Benson, Larry D. 1988. The Riverside Chaucer. 3rd Edition. Oxford: Oxford University Press. Bublitz, Wolfram and Christian Hoffmann. 2011. “‘Three Men Using our Toilet all Day Without Flushing – This May Be One of the Worst Sentences I’ve Ever Read’: Quoting in CMC”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 433–447. Erwin, Rebecca Schoff. 2006. “Early editing of Margery Kempe in manuscript and print”. Journal of the Early Book Society 9, 75–94. Frankis, John. 2000. “Sidelights on post-conquest Canterbury: towards a context for an Old Norse Runic Charm (DR 419)”. Nottingham Medieval Studies XLIV, 1–27. Franzen, Christine. 1991. The Tremulous Hand of Worcester. A Study of Old English in the Thirteenth Century. Oxford: Clarendon Press. Fredell, Joel. 2009. “Design and authorship in the Book of Margery Kempe”. Journal of the Early Book Society 12, 1–28. Goffman, Erving. 1955. “On face-work: an analysis of ritual elements in social interaction”. Psychiatry 18, 213–231. Jucker, Andreas H. and Irma Taavistainen (eds). 2010. Historical Pragmatics. Handbooks of Pragmatics, vol. 8. Berlin: de Gruyter Mouton. Kerby-Fulton, Kathryn and Maidie Hilmo. 2001. The Medieval Professional Reader at Work: Evidence from Manuscripts of Chaucer, Langland, Kempe, and Gower. Victoria: University of Victoria. Kerby-Fulton, Kathryn. 2001. “The medieval professional reader and reception history, 1292–1641”, in: Kathryn Kerby-Fulton and Maidie Hilmo (eds), 7–13. Krapp, George P. 1926. “The psychology of dialect writing”. The Bookman 63, 522–527. Lindquist, Ivar. 1932. Religiösa Runtexter I. Signtunga-Galdern. Skrifter Utgivna av VetenskapsSocieteten I. Lund: Ohlssons. Lyons, John. 1995. Linguistic Semantics. An Introduction. Cambridge: Cambridge University Press. Meech, Sanford Brown and Hope Emily Allen (eds). 1940. The Book of Margery Kempe. The Text from the Unique MS. Owned by Colonel W. Butler-Bowdon. Volume 1 (EETS: 212). London: Oxford University Press. Moltke, Erik. 1985. Runes and Their Origin. Denmark and Elsewhere. [transl. from Danish by Peter G. Foote]. Copenhagen: Nationalmuseets Forl. Pakkala-Weckström, Mari. 2010. “Chaucer”, in: Andreas Jucker and Irma Taavitsainen (eds), 219–245.

342   

   Jenny Arendholz and Monika Kirner-Ludwig

Parkes, Malcolm B. 1992. Pause and Effect: An Introduction to the History of Punctuation in the West. Aldershot, Hants: Scolar Press. Parsons, Kelly. 2001. “The Red Ink Annotator of the Book of Margery Kempe and his lay audience”, in: Kathryn Kerby-Fulton and Maidie Hilmo (eds), 143–216. Robinson, Ian. 1971. Chaucer’s Prosody: A Study of the Middle English Verse Tradition. Cambridge: Cambridge University Press. Schiltz, Guillaume. 2004. “Der Canterburyspruch oder ‚Wie finden dänische Runen und englische Komputistik zusammen?‘ Ein Beitrag zur historischen Textlinguistik”, in: Thomas Honegger (ed.). Riddles, Knights and Cross-dressing Saints. Essays on Medieval English Language and Literature. Bern: Lang, 115–138. Sperber, Hans. 1911. “Exegetische Miscellen”. Beiträge zur Geschichte der deutschen Sprache und Literatur 37/1, 148–156.

Internet sources URL 1: http://www.thestudentroom.co.uk/, last accessed 26 Oct 2012 URL 2: http://www.thestudentroom.co.uk/faq.php?faq=about_tsr#faq_welcome, last accessed 26 Oct 2012

Ursula Lutzky

14 Quotations in Early Modern English Witness Depositions Abstract: This chapter discusses the use of direct speech quotations in Early Modern English witness depositions, which are the official records of a witness’s oral testimony. In particular, it focuses on the preface position in direct speech quotations in this text type, which typically takes the form of third person narratives in the past tense, reporting previous speech events in indirect form. The witness depositions included in A Corpus of Early English Dialogues 1560–1760 will be studied to discover which lexical means are attested at the beginning of direct speech quotations to signal shifts in speech reporting, considering that punctuation was not consistently used to this end at this time. The analysis will first focus on the preface position in general in order to investigate how direct speech quotations are introduced in the corpus and to discover to what extent attestations in this position are explicitly associated with direct speech reporting (and potentially spoken language). Then, a closer look will be taken at the role pragmatic markers play in this position and how their structural and interpersonal functions may be linked to the change in speech presentation. The aim is to gain further insights into the linguistic means attested in preface position in direct speech quotations of Early Modern English witness depositions. Keywords: witness depositions, direct speech, preface position, pragmatic markers, Early Modern English

1 Introduction This study explores the attestation of direct speech quotations in Early Modern English (EModE) witness depositions (see also Aijmer, this volume). That is to say that it focuses on those sections in witness depositions that can be regarded as direct quotations reconstructing or representing previous speech acts. Witness depositions from this period are mainly third person narratives of previous speech events with the use of direct speech quotations being the exception rather than the rule. It is their comparative infrequence that calls for further study of their use and raises questions as to the means used to introduce these shifts from indirect to direct speech presentation. For the analysis, a corpus-based approach to EModE witness depositions is used. In particular, it is based on the depositions included in A Corpus of Early

344   

   Ursula Lutzky

English Dialogues 1560–1760 (CED), a corpus comprising five different speechrelated text types that was designed specifically “for research within such areas as variationist studies, historical pragmatics, and the study of speech presentation” (Kytö/Walker 2006: 11). For the present study, only the subsection of the CED comprising witness depositions will be used, which includes a total of almost 172,000 words. The analysis of the witness depositions from the CED will draw on corpus linguistic tools and will provide some quantitative insights, while at the same time focusing on the close study of examples from a qualitative perspective. In this study, I will investigate how direct speech quotations in EModE witness depositions are introduced, and specifically what role pragmatic markers play in these cases. My main focus will be on the preface position of these direct quotations, and I will consider what pragmatic means are used in this position to signal shifts in speech reporting. This will ultimately also involve a discussion as to why these shifts occur and how this is reflected in the functions served by the turn initial pragmatic markers.¹

2  Witness Depositions Witness depositions are the official records of a witness’s oral testimony. Hence they “are statements from court cases, where witnesses, or occasionally the accused person, give an account of what they saw, heard or did” (Cusack 1998: 92). They thus represent spoken text that was recorded by a scribe, usually before the actual proceedings were held in court. They were subsequently read aloud in court and confirmed by the witness (see also Kytö/Walker/Grund 2007: 66). Contrary to, for example, trial proceedings, witness depositions are monologic in nature and show a considerable amount of narratorial intervention, both on the part of the witness and the scribe. As they report what had happened or what someone had said, they comprise narrative devices and features of reporting speech, for example in the form of reporting clauses (Culpeper/Kytö 1997: 67, 2010: 54ff.; Kytö/Walker 2006: 21). Witness depositions fall into the category of authentic data that is based on an original speech event, i.e. spoken language. They have therefore been described as speech-based or speech-related text types, which represent the actual spoken language to a certain extent but cannot be regarded as exact records. As Archer

1 Contrary to Aijmer (this volume), whose study focuses on the discussion of speech external linguistic structures, such as verbs of speaking, this chapter is exclusively concerned with speech internal markers.

Quotations in Early Modern English Witness Depositions   

   345

(2002: 5) mentions with reference to the EModE courtroom, “the demands of converting one medium (speech) into another (writing) in real-time meant that seventeenth-century scribes often shifted between direct and indirect recording, summarised parts of the discourse or reconstructed a record of the trial at a later time from hastily scribbled notes”. As a consequence, EModE witness depositions show considerable scribal intervention, which is reflected clearly in the disambiguation of references through the use of legal formulae or formulaic phrases (e.g. the said X) (Grund/Walker 2011: 47–49; Kytö/Walker 2006: 21). Furthermore, scribes seem to have been concerned more with recording the content rather than the language and reproduced texts based on their notes, which were often taken down in shorthand in the EModE period. They are reconstructions of original speech events comprising what were deemed to be the main points by the recorder, probably also involving a degree of construction in that scribes may have constructed parts of the original speech event based on their perception of the spoken language of the time (Grund 2007: 119ff.). Indeed, EModE depositions are a mixture of different uses of language and represent different voices, involving the witness, the scribe and potential editors of the manuscripts. Thus, after recording a deposition, the scribe himself may have edited his work but also the printer, editor or publisher may have introduced linguistic emendations and changes to the content for religious or political reasons.² As a consequence, Grund/Walker (2011: 42ff.; see also Culpeper/Kytö 2010: 69–73) identify a minimum of five discourse levels in witness depositions: the level of the modern editor, the scribe, the deponent, the level of the speech event reported by the deponent and within that a potential further embedded level of speech reporting.³ The words attested can therefore not necessarily be clearly attributed to a particular participant or referent, which entails that it may be difficult to make claims about whose language is represented, although legal and formulaic phrases are most certainly the scribe’s and informal language attestations most likely not. However, when discussing the faithfulness of these reports, it has been pointed out that “no transcription is an exact copy of the original spoken language, but a subjective interpretation” (Culpeper/Kytö 2000: 178). This is not only the case for EModE transcriptions, which had to be done in the absence of modern technologies like audio or video recording, but also Present Day English (PDE) studies have found that direct speech presentations are rarely verbatim and

2 See also Culpeper/Kytö 2010: 52, 82f.; Huber 2007; Kryk-Kastovsky 2002: 212f., 2009: 441; Kytö/ Walker 2003: 228ff. 3 See also Walker/Grund (forthcoming) for a discussion of speech (re)presentation on these different discourse levels.

346   

   Ursula Lutzky

involve a subjective element.⁴ As Collins (2001: 290) points out, “[direct speech] typically functions not as an analysis but as a quasi-dramatic reenactment of [reported speech] events, even though under ordinary circumstances it cannot be considered a true facsimile”. Thus, while witness depositions provide access to authentic spoken language of the past, they cannot be regarded as completely faithful to the original but as (re-)constructed to a certain extent (Doty/Hiltunen 2002: 300; see also Claridge 2008: 247; Taavitsainen 1995: 442). This is also due to the historical and socio-cultural context in which these depositions were produced. With reference to pre-modern texts, Moore (2011: 184) notes that “strict faithfulness was not a viable prerequisite for direct speech reporting”. Moore points out that the expectations of speakers and listeners or writers and readers at the time did not necessarily entail a verbatim interpretation for a direct speech representation, as the focus was more on the content than the exact words spoken. This was also due to the fact that the distinction between de dicto (word-for-word) and de re (not word-for-word) was not clear-cut in medieval court documents (see also Moore 2002). Nevertheless, Moore (2002: 412) concludes that witness depositions can “serve as ‘speech-based’ texts in linguistic historical research” as they are “examples of a variety ‘originating in speech,’ even if their conformity to speech differs from our present-day construals of direct speech”. EModE witness depositions are typically third person narratives in the past tense. They are mainly presented in indirect speech, and while direct speech occurs as well, “it is important to note that direct speech is given as quoted statements in the depositions” (Grund/Kytö/Rissanen 2004: 155). The quoted speech can represent something the witness themselves said, it may be attributable to another person, or it may present earlier speech events in the form of a dialogue which was recounted by the witness and transcribed in this form (Culpeper/Kytö 1997: 67; Grund/Walker 2011: 45; Kytö/Walker 2003: 223). However, transitions in voice were not consistently signalled and so direct speech is, for instance, not clearly marked off by punctuation from other levels of discourse (Moore 2002; see also Huber 2007). While punctuation can therefore not be referred to as a reliable distinguishing criterion for direct and indirect speech, other means on the lexical level may signal switches between direct and indirect forms of speech reporting. This is also pointed out by Moore (2011: 183), who notes that “early writers of English had to separate modes of discourse with the words of the texts themselves. Lexical items such as vocatives, interjections and especially verbs of speaking were coopted for the purpose of marking reported speech”. It is these 4 See further Culpeper/Kytö 2010: 78ff. and sources cited therein; see also Bucholtz 2000; Collins 2001: 52; Moore 2011: 80ff.

Quotations in Early Modern English Witness Depositions   

   347

lexical items that will be studied in the following analysis, with a special focus on the use of pragmatic markers.

3 The Data The present analysis is based on A Corpus of English Dialogues 1560–1760 (CED). This, as has been said above, is a corpus that was specifically designed for the study of historical pragmatics and speech presentation, including the study of spoken interaction in the EModE period. To this end, this corpus was designed to include dialogic texts and comprises a total of 1.2 million words in 177 text samples (Culpeper/Kytö 2010: 23; Kytö/Walker 2006: 11ff.). The current study focuses on the subsection of the CED including witness depositions only, which comprises a total of almost 172,000 words and 33 text samples. The individual texts span the period 1560–1760 and are divided into five forty-year subperiods (D1 1560–1599, D2 1600–1639, D3 1640–1679, D4 1680–1719, D5 1720–1760). Witness depositions fall into the ‘authentic’ text category of the CED, as they can be traced to real speech events that were captured in written form. At the same time, witness depositions show considerable narratorial intervention as dialogic passages are framed by first or third person narrative passages including reporting clauses (Kytö/Walker 2006: 12; see also Culpeper/Kytö 2000: 176ff., 2010: 23ff.). Of particular relevance to the current study is the fact that it was a prerequisite for the text samples compiled for the CED to contain direct speech, which is not necessarily the case for all EModE witness depositions, and the compilers specifically included witness depositions “in which the witness reports conversations he or she has heard” (Culpeper/Kytö 1997: 67). An advantage of the text-level coding used in the CED is that the running text or non-speech is marked off from the dialogue or direct speech passages; nonspeech is defined by the corpus compilers as “the words a courtroom witness uttered rendered into indirect speech report by an individual recording the event in writing” (Culpeper/Kytö 2010: 66). The distinction between direct speech and other levels of discourse is of course not always straightforward, especially in EModE text types and witness depositions in particular. This was also pointed out by the corpus compilers, who claim that their “coding is one interpretation only” (Kytö/Walker 2006: 39). This study is based on the interpretation arrived at by the CED compilers as I use the CED coding to distinguish direct from indirect forms of speech and to extract the total number of words for both of these categories from the XML tagged files using the Multi-Lingual Corpus Toolkit (MLCT) developed

348   

   Ursula Lutzky

by Scott Piao. The results are given in Table 1, which shows the total number of words for dialogue and non-speech split up according to subperiod. Table 1: Dialogue and non-speech word counts for the CED witness depositions

CED witness deposition

dialogue

non-speech

D1 (1560–1599)

 6,043

14 %

 36,645

86 %

D2 (1600–1639)

10,091

26 %

 28,790

74 %

D3 (1640–1679)

 8,325

18 %

 38,355

82 %

D4 (1680–1719)

 2,537

10 %

 23,459

90 %

D5 (1720–1760)

 2,518

14 %

 14,944

86 %

Total

29,514

17 %

142,193

83 %

As can be seen in Table 1, non-speech predominates in the corpus with the dialogue or direct speech section amounting to less than one fifth of the overall word count. Furthermore, one can note variation not only in the total number of words but also in the distribution of dialogue and non-speech across the different subperiods. When taking a closer look at the individual files, it becomes obvious that some of them comprise a relatively low percentage of direct speech, such as the Indictments for Witchcraft D3WSUFFO (1 %) or the Witchcrafts of Margaret and Phillip Flower D2WFLOWE (2 %), while the text sample Late Design of the Papists D3WBROOK (99 %) consists almost entirely of dialogue. However, there are also examples where the amount of dialogue and non-speech is more or less balanced, as in the Tryal of the Memorable Jonathan Wilde (D5WWILDE) which contains 42 % direct and 58 % non-speech, or the Earles of Essex and Southampton (D2WSOUTH), which has an exactly equal distribution of both forms. In addition to the total number of words, the subsequent analysis requires the total number of directly quoted speaker turns. As mentioned above, the text level coding of the CED marks off direct speech from the running text. However, some of these direct speech passages are interrupted by scribal or editorial comments, either in the form of a (repeated) quotative or an addition for clarification. This is why the total number of direct speech attestations does not equate to the total number of speaker turns exactly, as illustrated by example (1).⁵

5 Note that non-speech is here marked off from direct speech quotations through square brackets and dollar signs, which is the coding adopted by the corpus compilers for the text file version of the CED.

Quotations in Early Modern English Witness Depositions   

   349

(1) Well, [$saith the Doctor, in the hearing of this Examinate,$] though the Plot be discovered, here’s my Major [$ (meaning Henshaw) $] is not discouraged, the business may go on yet; [Well, says the doctor, in the hearing of this examinate, although the plot is discovered, here my Major (meaning Henshaw) is not discouraged, the business may go on yet;] (CED: D3WCROMW, Inhumane Conspiracy against … Lord Protector, 1654, p. 82) In Example (1), the doctor’s turn is interrupted twice: first by the quotative saith the Doctor and the clarifying addition of in the hearing of this Examinate and second by the scribal comment meaning Henshaw. That is to say that a single speaker turn is here coded as three instances of direct speech. In order to arrive at the actual number of speaker turns, a concordance of all direct speech attestations in the witness depositions of the CED was combed for interruptions similar to those exemplified in (1) and resulted in a total of 943 directly quoted turns.

4 The Analysis The following analysis will be concerned with the lexical means that are used to introduce direct speech quotations in the witness depositions of the CED. The focus will thus be on the initial position of directly quoted speaker turns in the data and I will, in particular, study the role that pragmatic markers play in this position.

4.1 Direct speech initiators While the CED offers the advantage of text level coding added by its compilers, “[i]n EModE writing, direct forms are not generally signalled by quotation marks” (McIntyre/Walker 2011: 112; see also Culpeper/Kytö 2010: 66f.). They can, nevertheless, be identified through reference to particular linguistic forms that characterise spoken language. As Moore (2011: 46) points out, readers and, in the case of EModE witness depositions, listeners “require some form of marking to conclude that a passage is direct speech, and the passages must employ some clue indicating their direct speech. These markings are most likely in the near environment of the speech onset”. In her account of quoting speech in different types of early English texts, Moore (2011: 43ff.) discusses lexical markers that commonly signal the onset of direct speech. Among them, she distinguishes between speech inter-

350   

   Ursula Lutzky

nal ‘perspective shifters’ and speech external linguistic structures. The speech internal group comprises forms like interjections (alas!), vocatives (Sir), deictic pronouns (thou, you), spatio-temporal deictics (here, now), tense switching and other pragmatic markers. These features are also mentioned by Wright (2000: 93–97) in her discussion of attestations that are associated with speech in EModE witness depositions, in addition to some further examples such as the clause-initial coordinators and or but; several of these features have been related to the creation of involvement and said to characterise interactive speech situations (see Biber 1988; Biber/Finegan 1992, 1997). Speech external forms, on the other hand, include simple inquits, like seyde she, as well as explicit quotatives or meta-discourse, as in he spoke in these words following. Additionally, Moore (2011: 43ff.) distinguishes between conventional social interaction routines (e.g. politeness formulae) and conventional narrative interaction structures (e.g. formulae for public address), which also function as markers of direct speech onset. These markers were of particular importance for EModE witness depositions not only because of the absence of punctuation to mark off direct speech in writing but also because of their spoken form of delivery in the courtroom. Witness depositions, as was pointed out above, were usually read out during the actual proceedings in the courtroom, which means that their intended audience will have consisted primarily of “hearers rather than readers” (Collins 2001: 47). In the following analysis, I will mainly address the first group of speech internal forms, which are also regarded as signals indicating that a speaker is “embarking on direct speech quotation” (Biber et al. 1999: 1118) in the grammar of PDE conversation. I will study the linguistic means that the witnesses, scribes and subsequent editors of the depositions could use to signal the beginning of direct speech reporting, and which themselves formed part of the direct speech quotation. In particular, the discussion will focus on the role pragmatic markers play in direct speech reporting in EModE witness depositions (see also Aijmer, this volume). The main direct speech initiators that are attested in the EModE witness depositions of the CED are given in Table 2, with the plain token numbers weighted to the total number of 943 direct speech turns in brackets. As can be seen in Table 2, pragmatic markers and first person pronouns are the most frequently attested type of internal direct speech initiator in the data studied. ‘Pragmatic marker’ here serves as a cover term, comprising both discourse markers and interjections, as will be discussed in more detail below.

Quotations in Early Modern English Witness Depositions   

   351

Table 2: Direct speech initiators

Form

I/we

interrogative

Total

139 (15 %) 74 (8 %)

imperative vocative

thou/you

yes/no

pragmatic markers

71 (8 %)

61 (6 %)

52 (6 %)

143 (15 %)

69 (7 %)

The second most frequent category introducing direct speech quotations includes the first person pronouns I and we, with I (129) being significantly more frequent than we (10). The use of this deictic pronoun in preface position indicates the shift to a directly reported speech event by moving the focus from the scribe and the indirect form of narrative reporting, in which parts are summed up, to the perspective of the witness and what they said or witnessed someone saying. This is also true of the use of the second person deictic pronouns thou (30) and you (31) which signal a speech act that was addressed to an interlocutor (e.g. the witness, the attorney or a third party) and was recorded in direct form. Related to the use of pronominal forms is the attestation of vocatives (69) in turn initial position, that is forms of address (e.g. sir, my lord, mother), which indicate to whom the following words were directed and that are a feature of direct speech only. In addition to these, direct speech quotations are also repeatedly introduced by imperative (71) and interrogative constructions (74), for example signalled through the inversion of subject and verb. Finally, the particles yes (24) and no (28, including spelling variations like yea or nay) are associated with direct quotations and are frequently attested in answers recorded as direct speech to questions given in indirect form in the EModE witness deposition samples studied. As noted above, Table 2 only provides the main direct speech initiators, i.e. those which appear with the highest frequencies in the data, whereas other forms appearing with rather low token frequencies in turn initial position were not included here (e.g. spatial deictics like this, yonder or here). Overall, a total of 659 direct speech attestations are introduced by what Moore (2011: 43ff.) refers to as speech internal forms in first position, which equals 70 % of all direct speech turns. In the remaining 30 % of direct speech quotations, these features do not occur in preface position, although they may form part of them in a different position. These quotations are introduced, for example, by third person pronouns (i.e. he, she, it, they), names, noun phrases with direct or indirect articles, or adverbials, which are not exclusively regarded as speech internal criteria. Example (2) illustrates the use of some of the direct speech features mentioned above. The example is taken from the text sample “Witches in the Covntie of Lancaster” and comprises an extract which starts out being reported as indirect speech, with a dialogic passage appearing towards the end.

352   

   Ursula Lutzky

(2) which said Blacke-Dogge did not appeare to this Examinate, vntill the eighteenth day of March last: at which time this Examinate met with a Pedler on the high-way, called Colne-field, neere vnto Colne: and this Examinate demanded of the said Pedler to buy some pinnes of him; but the said Pedler sturdily answered this Examinate that he would not loose his Packe; and so this Examinate parting with him: presently there appeared to this Examinate the Blacke-Dogge, which appeared vnto her as before: which Black Dogge spake vnto this Examinate in English, saying;$] What wouldst thou haue me to do vnto yonder man? [$to whom this Examinate said,$] What canst thou do at him? [$and the Dogge answered againe,$] I can lame him: [$whereupon this Examinat answered, and said to the said Black Dogge,$] Lame him: [$and before the Pedler was gone fortie Roddes further, he fell downe Lame:… $] [the afore-mentioned black dog did not appear to this examinate until the eighteenth day of last March, at which time this examinate met with a pedlar on the highway called Colnefield, near Colne. And this examinate demanded of the afore-mentioned pedlar to buy some pins from him, but the aforementioned pedlar firmly answered this examinate that he would not unpack his bundle. And so when this examinate parted from him, there presently appeared to this examinate the black dog, which appeared to her as before. This black dog spoke to this examinate in English, saying, What would you like me do to this man? To which this examinate said, What can you do to him? And the dog answered again, I can make him lame, whereupon this examinate answered, and said to the afore-mentioned black dog, Lame him. And before the pedlar had gone forty rods further, he fell down lame…] (CED: D2WPENDL, Witches in the Covntie of Lancaster, 1612, p. R3Vf.) The extract recounts a time when the examined Alizon Device met a pedlar from whom she wanted to buy pins. However, as the pedlar refused to sell her any, a black dog appeared to Alizon and asked her what she wanted him to do to the pedlar. This conversation between the dog and the examined is given in the form of direct speech, with each quotation being marked by speech internal forms. Thus, the first two direct speech quotations are introduced by the interrogative pronoun what, followed by inversion of subject and verb, a switch in pronoun to the first and second person and the use of the deictic yonder. Furthermore, the answer to the second question, in which Alizon asks the dog how he could harm the pedlar, is initiated by the first person pronoun I and Alizon subsequently commands the dog to do as he suggests through the use of the imperative construction lame him. In Example (2), the shifts from indirect to direct speech reporting are thus clearly marked by features that are associated with direct speech.

Quotations in Early Modern English Witness Depositions   

   353

4.2 Pragmatic markers in EModE witness depositions One of the most frequently occurring direct speech initiators in the EModE witness depositions of the CED are pragmatic markers, as shown in Table 2. In previous work (see e.g. Lutzky 2012a), however, I found that, in general, pragmatic markers were not very prominently represented in witness depositions when compared to other speech-related EModE text types. My results showed that they were particularly frequent in constructed text types such as prose fiction, drama comedy or didactic works and appeared with reduced frequencies in authentic text types recording real speech events such as trial proceedings, witness depositions or letters. Additionally, they were attested with higher frequencies in primarily dialogic text types than in text types comprising mainly monologic as well as some dialogic parts. In short, my previous research found pragmatic markers to predominate in fictional and dialogic as opposed to authentic and monologic data. Reasons that may explain the less frequent attestation of pragmatic markers in trial proceedings and witness depositions include the formal and authoritarian context of the court in which documents needed “to look like an official courtroom record, and […] many ‘performance features’ or interpersonal features […] were deliberately not represented” (Culpeper/Kytö 2000: 187; see also Culpeper/ Kytö 2010: 376f.). Pragmatic markers may create “a highly ‘oral’ style” and a “high degree of involvement of the participants” (Culpeper/Kytö 1999: 307), which is also not compatible with the indirect form in which the witnesses’ testimony was mainly recorded. This is why, as will be seen below, several of the pragmatic markers studied appear primarily in direct speech passages in witness depositions. Furthermore, the interpersonal and performance features may have been lost in the process of transferring spoken text into written form. As has been stated above, courtroom scribes were primarily concerned with recording the content of proceedings and depositions, which may have led to the omission of particles serving a pragmatic function and contributing little to the semantic content of an utterance. At the same time, however, those pragmatic markers that are attested in EModE witness depositions could have been included deliberately by scribes. For example, Grund (2007: 133) points out with reference to the Salem witchcraft records that their attestation possibly “reflects the recorder’s sprinkling of these features into the record to make it more speech-like, and hence sound more verbatim”. Culpeper (2010: 86; see also Culpeper/Kytö 2010: 84f.) discusses this in the context of selection and interpretation, noting that the decision to include, for example, a surprise marker in a transcription “may be part of a design to engender a particular interpretation, and that interpretation may or may not be the one that the original speaker intended and/or hearer construed”.

354   

   Ursula Lutzky

Trial proceedings are of a relatively structured nature with sequences of question and answer pairs reducing the need for structural signals. In witness depositions, on the other hand, the switch from indirect to direct speech reporting is in more immediate need of being clearly marked, because direct speech passages are often the exception rather than the rule, as discussed above with regard to the depositions included in the CED (see section 3). There will, therefore, have been a higher incentive for the signals introducing direct speech to be recorded accurately, not leaving out semantically empty or reduced forms used by the witness, or if necessary to (re)construct them for this purpose. One way of marking this transition is through the use of a pragmatic marker in turn initial position. Indeed, discourse markers and interjections are regarded as prefatory expressions by Biber et al. (1999: 1074) in their PDE grammar of written and spoken English, which means that they have “the function of orienting the listener to the following utterance, especially in relation to what has preceded”.

4.3 Pragmatic markers as direct speech initiators in the witness depositions of the CED Table 3 lists the discourse markers and interjections that occur more than once in initial position in direct speech quotations in the witness depositions of the CED.⁶ Overall, 143 turns (15 %) out of 943 are introduced by a pragmatic marker and in Table 3 the plain token frequency is weighted to the total number of direct speech turns. Table 3: Turn initial interjections and discourse markers

Form

ah

oh

and

but

now

then

marry pray

well

why

what

Total

3 30 11 13 10 10 9 10 16 6 19 (0.3 %) (3.2 %) (1.2 %) (1.4 %) (1.1 %) (1.1 %) (1.0 %) (1.1 %) (1.7 %) (0.6 %) (2.0 %)

Some of the forms cited in Table 3, such as and (7,736 tokens in total), but (1,093), now (131) or then (861), are attested with considerably higher overall token frequencies in the corpus, serving both structural as well as temporal functions in different positions and both direct and indirect speech reporting (for and in EModE witness depositions see also Culpeper/Kytö 2010: 177ff.). Furthermore,

6 The forms alas, whough, hush hush, how now, pho pho, and so then appear only once in turn initial position.

Quotations in Early Modern English Witness Depositions   

   355

Table 3 shows that the frequencies with which discourse markers and interjections appear in initial position are not particularly high when considering the individual forms separately, but taken together they account for a considerable amount of direct quote initiating signals. Several of them in fact only or primarily appear in this position in the current data. Thus, the frequency of ah in Table 3 equates to its overall frequency in the corpus. The interjection o(h), on the other hand, predominates in initial position (83 %) in direct speech quotations, with a few examples (17 %) appearing in medial position. Examples (3) to (5) illustrate the use of o(h) in initial position in a direct speech quotation. All of these examples are taken from the same text sample, “Contemporary depositions respecting an affray at Norwich in the year 1583 in which Queen Elizabeths company of players, then acting at the Red Lion Inn, were involved” (CED: D1WNORWI 1583). The direct speech quotation more or less conveys the same content in each case, although certain differences in wording can be discerned. What is crucial from the point of view of pragmatic marker usage is that the transition from indirect to direct speech reporting is in all three instances marked by the interjection o(h). (3) [$Thomas Holland of Norwiche, caryer, examyned the sayd day and yere, sayeth that on Satturdaye last in the afternoone, he, beinge without the Red Lyon gate, dyd see one of the Quenes players in his playing apparell in the gate-house … but the player overtooke hym that had the blewe cote at the Cockey, nere Mr. Davyes howse, with his raper drawn, and thrust at hym that had the blew cote into the legg, whereat hee that had the blew cote cryed,$] “Oh! you have mayned me,” [$and at the Cockey tooke up a stone and therwe at the quenes servaunt, but whether he dyd hurt hym or not he knoweth not; [Thomas Holland of Norwich, carrier, examined the afore-mentioned day and year, says that last Saturday in the afternoon, he, being outside the Red Lion gate, saw one of the Queen’s players in his costume in the gate-house … but the player overtook the one wearing the blue coat at the Cockey, near Mr. Davy’s house, with his rapier drawn, and thrust at the one wearing the blue coat, hitting his leg, which made the one wearing the blue coat cry, “Oh! You have maimed me,” and at the Cockey he picked up a stone and threw it at the Queen’s servant, but whether he hurt him or not he does not know;] (CED: D1WNORWI, Affray at Norwich, 1583, p. 13–14)

356   

   Ursula Lutzky

(4) [$Thomas Holland confesseth that one of the quenes men runninge out of there play, for that there was a quarrelling at the gate, the Quenes man drew his raper at one that stood a lyttle from the gate, wiche he perceving run awaye, and the quenes man following hym, thrust hym into the legg, and the fellow sayed,$] “O! thou hast mayned mee;” [$but recovering hymselfe agayne, threw a stone at the quenes man, and hyt hym, and… [Thomas Holland confesses that one of the Queen’s men ran out of their play, because there was a quarrel at the gate. The Queen’s man drew his rapier at one who stood a little from the gate whom he perceived run away, and the Queen’s man followed him, thrust him at the leg, and the fellow said, “O! You have maimed me,” but recovering again, threw a stone at the Queen’s man and hit him, and…] (CED: D1WNORWI, Affray at Norwich, 1583, p. 20–21) (5) [$Thomas Crowe of Horton confesseth that one in a tawny cote and a cognoscence on his sleve stroke at hym that is deade, and hit hym on the knee, and after that blud followed, and after that one of the quens men hit hym on the back, and thrust hym twyce or thryce under the syde, and thereupon the fellow cryed,$] “O! Lorde, I am mayned!” [Thomas Crowe of Horton confesses that one in a tawny coat and with an insignia on his sleeve struck at the one who is dead and hit him on the knee, and after that blood followed, and after that one of the Queen’s men hit him on the back and thrust him twice or thrice under the side, and thereupon the fellow cried, “O! Lord, I am maimed!”] (CED: D1WNORWI, Affray at Norwich, 1583, p. 25) Following previous discussions on the reliability of EModE depositions (see in particular Kytö/Walker 2003: 227–8), it seems that the speaker will most likely have made an exclamation of this kind as all of the three witnesses introduced the quotation of the words spoken by the wounded Wynsdon before he died with the same interjection, o(h). This is supported, on the one hand, by the fact that this interjection functions as an exclamation with emotional colouring (see also Taavitsainen 1995), in this case possibly of shock or pain, and on the other hand, by its exclusive occurrence in direct speech passages in the data studied. The use of the interjection in initial position serves an important structural function which is related to the general infrequence of direct speech quotes in the depositions of this text sample. It clearly signals the beginning of direct speech in a text primarily comprising non-speech (2,853 words vs. 105 words of speech) and the necessity for such signals is reinforced by the remaining quotations, which are predominantly introduced by other direct speech features such as vocatives, imperatives or first and second person pronouns. While o(h) is followed by further

Quotations in Early Modern English Witness Depositions   

   357

direct speech clues, such as the use of first and second person pronouns and the shift to the present tense, it is the very first clue, considering that the quotative sayed/cryed could also have been followed by indirect speech.⁷ Overall, the number of direct speech quotations in this text sample (CED: D1WNORWI) is not very high, comprising only 15 direct speech turns. Considering that the three quotations appearing in Examples (3) to (5) make up one fifth of all direct speech quotations in the sample, one can assume that they carried a considerable significance for the case in question. Indeed, they imply which of the men attacking Wynsdon was responsible for his death by causing the deadly wound, a man called Henrye Browne striking his leg above the knee with a rapier, and this at least partly explains why the shift from indirect to direct speech reporting occurs. Further evidence supporting this conclusion was given by other witnesses who claimed that “the wounde wherof […] the man dyeth was a thrust above his knee” and that they “never sawe man bleed so much as hee dyd after [Browne] had pricked hym” (CED: D1WNORWI 1583). The discourse marker marry is more frequent in initial (46 %) than in medial position (23 %) in direct speech quotations. Contrary to ah and oh, it is also attested in indirectly reported passages (31 %). The discourse markers well and why appear exclusively in direct speech quotations in witness depositions with the clear majority (89 % for well, 67 % for why) attested in initial position. While in Examples (3) to (5) above o(h) followed the reporting clause, Example (6) illustrates that pragmatic markers and other direct speech features may also precede the reporting clause, which then appears embedded parenthetically in the quotation (see Aijmer, this volume, on quotative inserts). Indeed, in the corpus of EModE witness depositions studied, well appears most frequently in this position preceding the reporting clause at the beginning of a new turn, when compared to all pragmatic markers attested in turn initial position. The text extract given in (6) is taken from “The examination and confession of certaine wytches at Chensforde in the countie of Essex” (CED: D1WCHENS 1566), which contains a higher proportion of dialogue (632 words, 20 %, compared to 2,585 words or 80 % of non-speech). As a consequence, there are some direct speech

7 Note that I am not commenting on the use of quotation or exclamation marks as the edition used in the CED is what the compilers refer to as a type 2 edition, which differs “from the manuscript in that capitalization and punctuation are modernized (i.e. reflecting the usage of the time of the edition not that of the manuscript)” (Kytö/Walker 2006: 29). When later text editions were used, the compilers consulted the source manuscript for these editions, as explained in detail in the corpus guide (Kytö/Walker 2006: 27–31), and they state that “the text is reproduced as closely as possible to the source manuscript” (Kytö/Walker 2006: 28).

358   

   Ursula Lutzky

passages, like the one quoted in (6), where direct speech turns follow each other immediately. (6) yea yea, my lorde, [$quoth Ione waterhouse$] she lieth in that she saith that it hadde a face like an ape, for this that came to mee was like a dogge, well [$sayde the quenes attourney$] well, can you make it come before vs nowe, if ye can we will dyspatche you out of pryson by and by, no [$saith said Agnes Waterhouse$] I can not, for in faith if I had let hym go as my daughter did I could make hym come by and by, but now I haue no more power ouer him, [Yes, yes, my Lord, said Joan Waterhouse, she lies when she says that it had a face like an ape, as the one that came to me looked like a dog. Well, said the Queen’s attorney, well, can you make it come before us now? If you can, we will release you from prison soon. No, says the afore-mentioned Agnes Waterhouse, I cannot, for in faith if I had let him go as my daughter did, I could make him come back soon, but now I have no more power over him.] (CED: D1WCHENS, Wytches at Chensforde, 1566, pp. 27–28) As can be seen in (6), in all three of the directly quoted turns given, the reporting clause appears parenthetically, being preceded by the particles yes/no, the vocative my lorde and the discourse marker well. All of these forms share one function despite their otherwise distinctive uses (e.g. emphasis, negative answer to a question) and that is their turn initiating function signalling the onset of direct speech and a change of speaker before it is confirmed by the reporting clause. As the reporting clause appears turn medially, these other means are used to mark this switch (see also Moore 2011: 47). The discourse marker well in particular signals acknowledgement on the part of the Queen’s attorney of what has been said by Ione Waterhouse, who claims that Agnes Browne has lied in her description of the creature that Agnes Waterhouse allegedly sent to her. In fact, the discourse marker well is attested twice, once before and once after the reporting clause, and while this repetition may be due to the intervening parenthetical, well as a discourse marker is also used in the construction well, well. Previous studies found this construction to act as a comment on misbehaviour (Bolinger 1989: 306), to denote “surprise, resignation, or acquiescence” (OED: s.v. well, adv. VI. 24.a.), as well as to signal reassurance or acknowledgement (Defour 2009: 165ff.). The Example given in (6) combines all of these aspects: well, well conveys acknowledgement of what has been said (positive appraisal), while at the same time expressing the attorney’s surprise and possibly carrying some reprimanding features, considering that the witness allegedly lied. Furthermore, well, well functions as a delay device indicating consideration on the part of the speaker and it has a hedging function due to its

Quotations in Early Modern English Witness Depositions   

   359

remnant positive meaning (see also Defour 2009). The repetitive use of well by the Queen’s attorney thus mitigates the following request for information and action, asking Agnes Waterhouse if she could make the said creature appear before them, which she denies. When comparing the forms why and what, it turns out that overall what (461 tokens) is relatively more frequent than why (63 tokens) in the witness depositions of the CED. Out of this total number of attestations, 19 occurrences of what (4 %) appear with a discourse marking function, whereas it is 6 for why (10 %). Thus, why has a higher density of attestation as a discourse marker, whereas what appears more frequently in its pronominal and adverbial uses.⁸ Although further research into other ways of asking for reasons would be needed, an additional implication could be that questions relating to what someone had done, said, or heard could be more common than questions relating to the reasons why something had been done in the context of this sample of EModE witness depositions. Example (7) is a deposition by John Sterne, who gives evidence as part of the proceedings against alleged witches in the county of Essex (CED: D3WESSEX 1645). In his testimony he recounts an encounter he had with Elizabeth Clark, who is suspected of being a witch and told him and several others about her interaction with the devil and her imps. Sterne’s whole deposition is reported in indirect speech apart from the one direct speech quotation introduced by the discourse marker what in (7). As Collins (2001: 297) points out in his study of medieval Russian trial transcripts, direct speech reporting has a hearer-based orientation which gives “the intended interpreters the greatest share in the sensemaking process”. As a consequence of involving a greater demand on interpretation, these passages are usually perceived as having more importance. (7) [$This Informant saith, That watching with Elizabeth Clarke, (suspected for Witchcraft, as aforesaid) shee confessed that the Devill had had carnall copulation with her in the likenesse of a man; and that the said Elizabeth desired this Informant, and the rest that were in the roome with her to sit downe, and said, shee would shew this Informant and the rest some of her Impes: And within halfe an houre there appeared a white thing in the likeness of a Cat, but not altogether so big: And being asked if she would not be afraid of her Impes; the said Elizabeth answered,$] What doe yee thinke I am afraid of my children? 8 Note that I arrived at a similar result which showed that why had a higher incidence as a discourse marker than what in EModE drama comedy and concluded that “why seems to have had a greater tendency to function as a discourse marker in EModE than what” (Lutzky 2012b: 187).

360   

   Ursula Lutzky

[This informant says that remaining awake with Elizabeth Clark, suspected of witchcraft as afore-mentioned, she confessed that the devil had had carnal copulation with her in the likeness of a man; and that the afore-mentioned Elizabeth desired this informant and the rest that were in the room with her to sit down and said she would show this informant and the rest some of her imps. And within half an hour there appeared a white thing in the likeness of a cat, but not altogether so big. And being asked if she was not afraid of her imps, the afore-mentioned Elizabeth answered, What, do you think I am afraid of my children?] (CED: D3WESSEX, Witches … in the County of Essex, 1645, pp. 3–4) The direct speech quoted is the answer to a question reported in indirect form, asking Elizabeth Clark if she was not afraid of her imps. In addition to its indirect nature and the ensuing third person pronouns, the question was also “put into the passive so that the syntax stresses the person questioned rather than the unspecified questioner” (Cusack 1998: 96). Elizabeth’s answer is introduced by the exclamation of the discourse marker what, expressing her surprise at being asked this question as she is obviously not afraid of her imps whom she regards as her children. The one directly quoted utterance is thus one that emphatically expresses Elizabeth’s astonishment at what has been said through the use of an exclamative what and by answering in the form of a counter-question. This is a context in which what seems to have occurred typically as a surprise marker, as one of my previous studies on EModE drama comedy showed that what predominantly prefaced questions when attested as a discourse marker (see Lutzky 2012b: 182f.). Example (8) illustrates the use of the discourse marker why introducing a direct speech quotation in the corpus. Again, the depositions relate to witchcraft proceedings in the county of Essex but they stem from an earlier subperiod in the corpus (E1 1560–1599). The extract given in (8) is, like the one in (7), concerned with the topic of imps and children, only the two are not equated in this case. (8) [$The saide Henrie saith, that hee is of the age of ix. yeeres, and that sithence Candlemas last, one night about midnight, there came to his brother Iohn a spirite, and tooke him by the left legge, and also by the litle Toe, which was like his sister, but that it was al blacke: at which time his brother cryed out and said,$] Father, Father, come helpe mee, there is a blacke thing yt hath me by ye legge, as big as my sister: [$whereat his father said to his mother,$] why thou whore cannot you keepe your impes from my childrẽ? [$whereat shee presently called it away frõ her sonne, saying,$] come away, come away, [$At which speech it did depart.$]

Quotations in Early Modern English Witness Depositions   

   361

[The afore-mentioned Henry says that he is of the age of nine years and that since last Candlemas, one night at around midnight, there came a spirit to his brother John and took him by the left leg, and also by the little toe; it was like his sister only that it was all black. At this time his brother cried out and said, Father, Father, come and help me, there is a black thing that has me by the leg, as big as my sister. Then his father said to his mother, Why you whore! Can you not keep your imps from my children? Whereupon she presently called it away from her son, by saying, Come away, come away. At which words it departed.] (CED: D1WDARCY, Witches, Taken at S. Oses, 1582, p. D1R) The deposition cited is given by the nine-year-old Henrie, who witnessed a black spirit grab his brother Iohn by the leg. While one might question the accuracy of the evidence given by a nine-year-old, especially when it comes to reporting previous speech events as direct speech, the use of why at the beginning of his father’s turn is nevertheless revealing in terms of its discourse function, despite the fact that it may be of a constructed nature. This is because it illustrates a way of indicating the beginning of direct speech which conveys the father’s negative emotions of surprise and anger. These are then also reflected in the following vocative thou whore, which addresses the mother in a derogatory manner and introduces a request phrased in interrogative form. As can be seen in (8), all three speaker turns cited are introduced by a direct speech feature, including in addition to the exclamation of the discourse marker why also the vocative father and the imperative construction come away. The lexeme pray appears more frequently with a discourse marking function (20 tokens) than a literal meaning (13 tokens) in the witness depositions, with half of the discourse marker attestations appearing in initial position in direct speech quotes. Both of these uses of pray are illustrated in example (9), extracted from the same text sample as (8), where the first attestation of pray functions as a discourse marker and the second one has the literal meaning of ‘beseeching God through prayer’. (9) [$This examinat saith, yt within two moneths after his said wife said vnto him,$] I pray you as euer there was loue betweene vs, (as I hope there hath been for I haue v. pretie children by you I thanke God) seeke som remedie for me against yonder wicked beast [$ (meaning the saide Annis Herd).$] And if you will not I will complaine to my father, and I thinke he wil see som remedie for me, for [$ (said she) $] if I haue no remedie, she will vtterly consume me, [$whereupo~ this examinat did exhort his said wife as hee had before, & desired her to pray to God, and yt he wold hang her the said Annis Herd if he could proue any such matter…$]

362   

   Ursula Lutzky

[This examinate says that within two months after his afore-mentioned wife said to him, I pray you if ever there was love between us, as I hope there has been because I have five pretty children with you, thank God, seek some remedy for me against this wicked beast (meaning the afore-mentioned Annis Herd). And if you will not, I will complain to my father, and I think he will find some remedy for me. Because, said she, if I have no remedy, she will utterly consume me. Thereupon this examinate exhorted his afore-mentioned wife as he had before and desired her to pray to God, and that he would hang the afore-mentioned Annis Herd if he could prove any such matter…] (CED: D1WDARCY, Witches, Taken at S. Oses, 1582, p. F2Vf.) The discourse marker attestations of pray appear in two forms: either pray appears on its own or it is attested in the construction I pray you, as in (9). Other forms, such as pray thee or prithee, are absent from this data set. While this may be due to the specific text samples studied, it may also be the case that pray was not used in combination with the more informal form of the second person pronoun by witnesses when giving their testimony and reporting on previous speech events (see also Walker 2007: 271, Table 8.13). When looking at the examples in more detail, it turns out that the discourse marker pray always collocates with a request, asking an interlocutor to do or say something (see also Lutzky/Demmen 2013 for the functions of the discourse marker pray in EModE drama, and Włodarczyk 2007 for trial proceedings). As a consequence of its co-occurrence with imperatives, the discourse marker attestations of pray are restricted to the direct speech passages in the corpus, whereas the non-discourse marker uses are not. This is illustrated in example (9), where the discourse marker I pray you precedes the request seeke som remedie for me against yonder wicked beast in a direct speech turn, while the literal use of the verb pray is attested in indirect speech.

5 Conclusion This chapter investigated the use of direct speech quotations in the EModE witness depositions forming part of A Corpus of English Dialogues 1560–1760. In particular, the focus of this study was on the preface position in direct speech quotations, which marks the transition from indirect to direct speech reporting. Witness depositions from the EModE period were mainly recorded in the form of third person narratives; in the CED, direct speech accounts for less than one fifth of the overall word count. Previous speech events were thus primarily reproduced

Quotations in Early Modern English Witness Depositions   

   363

in indirect form and were subsequently read aloud in the courtroom. Therefore, witness depositions were not only records of spoken reports of previous speech events but also reconstructed in spoken form, making their immediate audience listeners rather than readers. As a consequence the preface position in direct speech quotations was one of the main means of signalling a switch in the type of speech reporting; this was due to the inherently spoken nature of the text type and, when considering the written record, to the absence of consistently indicating transitions in voice through the use of punctuation (McIntyre/Walker 2011: 112; Moore 2002; see also Huber 2007). Previous research has shown that EModE witness depositions were mainly concerned with recording the content of what was said, rather than being accurate reflections of the exact words used by a witness. EModE witness depositions are therefore regarded as a combination of construction and reconstruction, which could be influenced by several participants including the editor, publisher, scribe but also the deponent (Culpeper/Kytö 2010: 82f.; Grund 2007: 119ff.; Kytö/ Walker 2003: 228ff.). Furthermore, it has been pointed out that as a consequence of real-time pressures and the equipment available at the time (see e.g. Archer 2002: 5), scribes may have omitted certain features typical of spoken language, e.g. repetitions, false starts or semantically empty expressions. Additionally, this may have been due to the formal and authoritarian context in which these documents were produced, which could have entailed the omission of, for instance, pragmatic particles creating an oral and involved style (Culpeper/Kytö 1999: 307, 2000: 187; 2010: 376f.). These are possible reasons why pragmatic markers are rather infrequently attested in witness depositions when compared to other, especially fictional and dialogic, EModE text types (see my previous findings in Lutzky 2012a). However, the current chapter has shown that many of the pragmatic markers appear exclusively or primarily in direct speech passages in the data, with the majority attested in turn initial position. This is specifically the case for those pragmatic markers which show both discourse structuring and interactive or interpersonal functions, such as for example ah, oh, what or well; other pragmatic markers whose primary function is to structure discourse appear more readily also in non-direct speech, such as for example and, but, now or then. It is in particular the members of the former group which can create an involved and interactive style and this is why scribes may have (re)constructed them, based on their own perception of EModE speech, when writing up the depositions from their notes, to add an oral feel to direct speech passages (see e.g. Culpeper 2010; Grund 2007). Additionally, their function of signalling direct speech onset will have contributed to the significance of recording or constructing these forms.

364   

   Ursula Lutzky

Out of 943 instances of direct speech quotations which are represented in the text samples of the CED, 659 are introduced by a speech internal perspective shifter (Moore 2011: 43ff.). That is to say that 70 % of all direct speech turns are introduced by a form that is characteristic of direct speech and the transition from indirect to direct speech is thus clearly marked. In the current data, pragmatic markers and first person pronouns appear in this preface position most frequently, followed by interrogative and imperative constructions, vocatives, second person pronouns, and yes/no particles. While these forms may have been used by the witnesses themselves, constructed by the scribes or inserted by publishers or editors at the time, they seem to have been associated with spoken language in the history of English. Thus, they function as clear signals of the onset of direct speech and will have been used and perceived as such by speakers of EModE. When attested in initial position, a pragmatic marker may even precede a reporting clause which is inserted parenthetically into a quotation and thus signals a change in speaker or the beginning of a new turn before this is confirmed by the reporting clause. Among the pragmatic markers studied, well appears most frequently in this position in the witness depositions of the CED. Overall, the most frequently attested pragmatic markers in preface position are the interjection oh and the discourse marker what, with the interjection ah and the discourse marker why appearing least frequently. As the examples discussed in this chapter have shown, pragmatic markers with interactive or interpersonal potential can express different types of emotional colouring in an emphatic manner, ranging from surprise and anger to acknowledgement and positive appraisal. These markers are exclamative expressions which are typically used in direct speech and their functions are often tied to the reason why a shift in voice reporting occurs. As Collins (2001) argues in his study of medieval Russian trial proceedings, direct speech reporting is hearerbased and demands a greater effort at interpretation compared to indirect reporting. In depositions which mainly consist of indirect reporting, direct speech quotations may therefore be perceived as important and they usually contain evidence that is directly relevant to a case, indicating for example what an alleged witch had said and how this contributed to the construction of her witch identity (see Example 7). The pragmatic markers attested in preface position require the hearer or reader to interpret their pragmatic function at the beginning of a quotation, identifying it as direct speech. At the same time, their emphatic nature and orientation towards the addressee have an attention-catching function, operating as a further incentive for the hearer or reader to discover the significance of the quotation for the case in question.

Quotations in Early Modern English Witness Depositions   

   365

References Archer, Dawn. 2002. “‘Can innocent people be guilty?’ A sociopragmatic analysis of examination transcripts from the Salem witchcraft trials”. Journal of Historical Pragmatics 3/1, 1–30. Biber, Douglas. 1988. Variation Across Speech and Writing. Cambridge: Cambridge University Press. Biber, Douglas and Edward Finegan. 1992. “The linguistic evolution of five written and speech-based English genres from the 17th to the 20th centuries”, in: Matti Rissanen, Ossi Ihalainen, Terttu Nevalainen and Irma Taavitsainen (eds). History of Englishes: New Methods and Interpretations in Historical Linguistics. Berlin: de Gruyter, 688–704. Biber, Douglas and Edward Finegan. 1997. “Diachronic relations among speech-based and written registers in English”, in: Terttu Nevalainen and Leena Kahlas-Tarkka (eds). To Explain the Present. Studies in the Changing English Language in Honour of Matti Rissanen. Helsinki: Société Néophilologique, 253–275. Biber, Douglas, Stig Johansson, Geoffrey Leech, Susan Conrad and Edward Finegan. 1999. Longman Grammar of Spoken and Written English. Harlow: Longman. Bolinger, Dwight. 1989. Intonation and Its Uses. Melody in Grammar and Discourse. London: Edward Arnold. Bucholtz, Mary. 2000. “The politics of transcription”. Journal of Pragmatics 32/10, 1439–1465. CED = A Corpus of English Dialogues 1560–1760. 2006. Compiled under the supervision of Merja Kytö (Uppsala University) and Jonathan Culpeper (Lancaster University). Claridge, Claudia. 2008. “Historical corpora”, in: Anke Lüdeling and Merja Kytö (eds). Corpus Linguistics. An International Handbook. vol. 1. Berlin: de Gruyter, 242–259. Collins, Daniel E. 2001. Reanimated Voices: Speech Reporting in a Historical-Pragmatic Perspective. Amsterdam: John Benjamins. Culpeper, Jonathan. 2010. “Historical sociopragmatics”, in: Andreas H. Jucker and Irma Taavitsainen (eds). Historical Pragmatics. Handbooks of Pragmatics, vol. 9. Berlin: de Gruyter Mouton, 69–94. Culpeper, Jonathan and Merja Kytö. 1997. “Towards a corpus of dialogues, 1550–1750”, in: Heinrich Ramisch and Kenneth Wynne (eds). Language in Time and Space. Studies in Honour of Wolfgang Viereck on the Occasion of his 60th Birthday. Stuttgart: Franz Steiner Verlag, 60–73. Culpeper, Jonathan and Merja Kytö. 1999. “Modifying pragmatic force. Hedges in Early Modern English dialogues”, in: Andreas H. Jucker, Gerd Fritz and Franz Lebsanft (eds). Historical Dialogue Analysis. Amsterdam: John Benjamins, 293–312. Culpeper, Jonathan and Merja Kytö. 2000. “Data in historical pragmatics. Spoken interaction (re)cast as writing”. Journal of Historical Pragmatics 1/2, 175–199. Culpeper, Jonathan and Merja Kytö. 2010. Early Modern English Dialogues. Spoken Interaction as Writing. Cambridge: Cambridge University Press. Cusack, Bridget. 1998. Everyday English, 1500–1700: A Reader. Edinburgh: Edinburgh University Press. Defour, Tine. 2009. “Well, well! What a surprise! A diachronic look at the relation between well, well and the pragmatic marker well”, in: Stef Slembrouck, Miriam Taverniers and Mieke Van Herreweghe (eds). From Will to Well. Studies in Linguistics Offered to Anne-Marie Simon-Vandenbergen. Gent: Academia Press, 161–172.

366   

   Ursula Lutzky

Doty, Kathleen and Risto Hiltunen. 2002. “‘I will tell, I will tell’. Confessional patterns in the Salem witchcraft trials, 1692”. Journal of Historical Pragmatics 3/2, 299–335. Grund, Peter J. 2007. “From tongue to text: the transmission of the Salem witchcraft examination records”. American Speech 82/2, 119–150. Grund, Peter J., Merja Kytö and Matti Rissanen. 2004. “Editing the Salem witchcraft records: an exploration of a linguistic treasury”. American Speech 79/2, 146–166. Grund, Peter J. and Terry Walker. 2011. “Genre characteristics”, in: Merja Kytö, Peter J. Grund and Terry Walker (eds). Testifying to Language and Life in Early Modern England. Amsterdam: John Benjamins, 15–56. Grund, Peter J. and Terry Walker. Forthcoming. “‘wth many vnsemely woorde’: speech (re)presentation in 16th-century depositions”. Huber, Magnus. 2007. “The Old Bailey Proceedings, 1674–1834. Evaluating and annotating a corpus of 18th- and 19th-century spoken English”, in: Anneli Meurman-Solin and Arja Nurmi (eds). Annotating Variation and Change. Helsinki: University of Helsinki. http://www. helsinki.fi/varieng/journal/volumes/01/huber/, last accessed 10 Apr 2015. Kryk-Kastovsky, Barbara. 2002. Synchronic and Diachronic Investigations in Pragmatics. Poznan: Motivex. Kryk-Kastovsky, Barbara. 2009. “Speech acts in Early Modern English court trials”. Journal of Pragmatics 41/3, 440–457. Kytö, Merja and Terry Walker. 2003. “The linguistic study of Early Modern English speechrelated texts. How ‘bad’ can ‘bad’ data be?” Journal of English Linguistics 31/3, 221–248. Kytö, Merja and Terry Walker. 2006. Guide to A Corpus of English Dialogues 1560–1760. Uppsala: Uppsala University. Kytö, Merja, Terry Walker and Peter J. Grund. 2007. “English witness depositions 1560–1760: an electronic text edition”. Icame Journal 31, 65–85. Lutzky, Ursula. 2012a. Discourse Markers in Early Modern English. Amsterdam: John Benjamins. Lutzky, Ursula. 2012b. “Why and what in Early Modern English drama”, in: Manfred Markus, Yoko Iyeiri, Reinhard Heuberger and Emil Chamson (eds). Middle and Modern English Corpus Linguistics. A Multi-dimensional Approach. Amsterdam: John Benjamins, 177–189. Lutzky, Ursula and Jane Demmen. 2013. “Pray in Early Modern English drama”. Journal of Historical Pragmatics 14/2, 263–284. McIntyre, Dan and Brian Walker. 2011. “Discourse presentation in Early Modern English writing”. International Journal of Corpus Linguistics 16/1, 101–130. MLCT = Multi-Lingual Corpus Toolkit. 2002. Designed by Scott Songlin Piao. https://sites. google.com/site/scottpiaosite/software/mlct, last accessed 10 Apr 2015. Moore, Colette. 2002. “Reporting direct speech in Early Modern slander depositions”, in: Donka Minkova and Robert Stockwell (eds). Studies in the History of the English Language: A Millennial Perspective. Berlin: de Gruyter, 399–416. Moore, Colette. 2011. Quoting Speech in Early English. Cambridge: Cambridge University Press. OED = The Oxford English Dictionary. Online Edition. Oxford: Oxford University Press. www.oed. com, last accessed 10 Apr 2015. Taavitsainen, Irma. 1995. “Interjections in Early Modern English. From imitation of spoken to conventions of written language”, in: Andreas H. Jucker (ed.). Historical Pragmatics. Pragmatic Developments in the History of English. Amsterdam: John Benjamins, 439–465. Walker, Terry. 2007. Thou and You in Early Modern English Dialogues. Trials, Depositions and Drama Comedy. Amsterdam: John Benjamins.

Quotations in Early Modern English Witness Depositions   

   367

Włodarczyk, Matylda. 2007. Pragmatic Aspects of Reported Speech. The Case of Early Modern English Courtroom Discourse. Frankfurt am Main: Peter Lang. Wright, Laura. 2000. “On the construction of some Early Modern English courtroom narratives”, in: Gabriella di Martino and Maria Lima (eds). English Diachronic Pragmatics. Napoli: CUEN, 79–102.

Alison Johnson

15 Haunting Evidence: Quoting the Prisoner in 19th Century Old Bailey Trial Discourse. The Defences of Cooper (1842) and McNaughten (1843) Abstract: This paper focuses on the representation of quoted speech in a corpus of Old Bailey trials from Victorian London, with detailed examination of two of them: Cooper (1842) and McNaughten (1843). These trials involve an insanity defence. As the trials were recorded by hand, they entail scribal choices about whether to record and how to represent quotation, making choices about direct and indirect quotation pragmatically interesting, since the illocutionary force of the recorded utterances is selected as important by the scribe. My corpus, the Monomania Corpus, is collected from the larger Proceedings of the Old Bailey, 1674–1913 [available online: http://www.oldbaileyonline.org/], which contains 197,745 criminal trials over more than two centuries. Criminal trial hearings are communicative events which are densely intertextually structured. Prior questioning of witnesses and records of what they say are extensively quoted, requoted, and recontextualized in the course of the trial and quotation is used to demonstrate the defendant’s criminal liability or his incapacity to commit a criminal offence. The establishment of a binding legal reality crucially hinges on the differential weights of prosecution and defence evidence, so the selective power of quotation is an important resource for defending the prisoner. In the case of Cooper, the prosecution’s use of direct and indirect quotation is incriminating for the defendant and the witnesses seem well ‘rehearsed’ in supporting the prosecution case, whereas the defence picture, drawn out through family members, presents the defendant’s insane utterances from childhood to early manhood, with the miserable, poor, young man being seen as easy prey by a constabulary looking for results. In McNaughten’s case, we see the two institutions of the law and medicine fighting over culpability and insanity, as the defence quotes from medical interviews with the accused. The analysis shows the powerful use of hypothetical quotation and establishes a new category of quotation: metatalk for suppressing quotation. The “haunting evidence” (Eigen 1995: 160) of prisoner testimony is given voice through the powerful alliance of advocate and witness (lay, and medical), as they tell the court, through the prisoner’s words, what he is incapable of putting together himself. The prisoner is prosecuted in his own words. What he says on arrest, in gaol, and before the magistrate, is recorded and replayed to self-incriminate, as he is ventriloquized by the prosecution, and it is

370   

   Alison Johnson

therefore a masterful adversarial reversal to defend with the same voice, but from a different perspective. Keywords: quotation, reported speech, insanity, Victorian trial, defence, corpus

1 Introduction At 120 million words the Old Bailey Proceedings 1674–1913 (Hitchcock et al. 2012) (hereafter OBP), containing 197,000 trials, is a seemingly homogenous, large corpus. For forensic discourse analysts, the two and a half centuries of the OBP is too wide a period to examine as a whole, since statutes, trial practice and lawyer and witness involvement in trials change substantially over this time (Hostettler 2009) and, as Hunston (2011) points out, “a corpus, unlike a text, cannot be analysed”. I therefore focus on two complete trials from the Victorian period (1837– 1901), a period which could be described as a crucible of legal change, since the Old Bailey Sessions House sees the growth and rise of three professions within its walls. Police officers from the newly-formed Metropolitan Police force (from 1829), barristers from a “newly emerging advocacy bar whose ambitious efforts to fashion a defense and prosecution ‘case’ would have profound consequences for the tone and texture of witness examination”, and the “first forensic-psychiatric witnesses” (Eigen 1995: 133), though these medics did not call themselves ‘psychiatrists’ until the 20th century,¹ form a new cast of participants in the trial courtroom. Prior to the Prisoners’ Counsel Act (1836), defendants had not been entitled to a full defence by counsel and defendants largely defended themselves (Cairns 1998; Archer 2013). Outside the courtroom the Industrial Revolution had seen the expansion of “urban areas where poverty and crime were growing” (Hostettler 2009: 167) and “the invisible plague”, an “epidemic of insanity […] haunted the age” (Torrey/Miller 2001: 3). The OBP website’s statistics function shows a sharp rise in non compos mentis (a post-classical Latin legal term, meaning ‘not of sound mind’ or insane) verdicts at the end of the 18th and beginning of the 19th century, which continued throughout the 1900s. This is reflected in the broader international picture provided by Torrey/Miller (2001) across England and Wales, Ireland, Canadian and Atlantic Provinces, and the United States, rising from less than 1 to over 3 insane persons per 1,000 in England and Wales between 1807 and 1907.

1 There are no occurrences of psychiatr -ist -y -ic in the OBP and the lemmata psychology -y -ical -ist -ists only exist after 1882.

Quoting the prisoner   

   371

There was a uniquely Victorian word for insanity, viz. monomania (today diagnosed as “psychosis or paranoid schizophrenia” (Bewley 2008), which, according to the Oxford English Dictionary (OED Online 2012), spans the period 1815 to 1883. Jean Étienne Esquirol’s treatise on insanity, which introduced the concept of monomania, is described in Taylor/Shuttleworth (1998: 256) as “a ‘disease of the sensibility’, which changes in accordance with the prevailing forms of society”, or as Esquirol (1838/1845) put it “a disease of civilization” where, following the decline of religion, governments, in order to maintain authority over men, have had recourse to a police. Since that period, it is the police that troubles feeble imaginations, and establishments for the insane are peopled with monomaniacs who, fearing this authority, are delirious respecting the influence which it exercises, and by which they think themselves pursued. This monomaniac, who would formerly have been delirious with respect to magic, sorcery and the infernal regions; is now delirious, thinking himself threatened, pursued, and ready to be incarcerated by agents of the police. (Esquirol [1838, trans Hunt 1845] Taylor/Shuttleworth 1998: 257)

The condition and circumstances described by Esquirol of feeble people feeling threatened and pursued fit both of the cases we examine here. The trials occur within twelve months: Cooper (June 1842) and McNaughten (February 1843). In Cooper’s case he is literally pursued by the police, as well as fearing and threatening them. When cornered and arrested after a chase, he is recorded as saying “don’t ill-use me” (Testimony of Stephen Turnbull), and McNaughten is quoted as saying he is “pursued” by “Tories” when he is before the Magistrate after his arrest for shooting the Prime Minister, Sir Robert Peel’s private secretary: The Tories in my native city have driven me to this, and have followed me to France, Scotland, and other parts; I can get no sleep from the system they pursue towards me; I believe I am driven into a consumption by them; they wish to murder me. (Testimony of Inspector Tierney)

This paper investigates the Victorian insanity defence in what I have called the Monomania Corpus, since all trials contain the word. This deliberate bias allows us to examine the role of medical evidence of insanity. The corpus contains 15 trials from 1833 to 1907, almost identically coinciding with Queen Victoria’s reign (1837–1901), with the final two trials containing references to the disappearance of the diagnosis (“monomania is not much believed in now”, trial of Sando 1898; “the old idea of monomania is not now accepted”, trial of Church 1907). My focus here is on two of the largest trials and, in particular, on the way the prisoner is prosecuted and defended through quoting his own words. Quotation by each side creates very different views of the offender and the offence, the pros-

372   

   Alison Johnson

ecution presenting a picture of a rational and sane individual, and the defence one of a mentally incapacitated person incapable of knowing right from wrong. We examine the advocacy work of the defence barristers, Mr Horry (for Thomas Cooper) and Mr Cockburn (for Daniel McNaughten). As they cross-examine prosecution witnesses and quote their client’s conversations with prosecution and defence witnesses over the course of the trial, they demonstrate the power of advocacy. The prisoner is thus made to speak, while being ventriloquized (Tannen 2007: 21; Johnson 2013: 152) by the professional (lawyer and medic), as he sits silently in the court. The detailed record in the OBP and in the newspapers of the day makes use of direct quotation, indirect quotation (with that and zero that), and narrative reports of speech acts (henceforth NRSA, Leech/Short 2007) (examples i, ii, and iii, from the Cooper trial), which all contribute to a representation of the prisoner’s miserable position. In addition, we have hypothetical speech (example iv), since not all quotation is reporting of speech said on an earlier occasion, and this can be used “to suggest what someone might say/have said on some (often hypothetical) occasion” (Holt 2007: 47). In this paper I look at who quotes and who is quoted, where it is in the trial, and when it occurs in relation to the offence and its alleged commission. i. The prisoner said, “The first one who dares to attack me shall have the contents” [of a pistol] ii. I accompanied him to the station, and he gave his name as Thomas Cooper, and that he was a bricklayer. iii. he did not offer to get out while I was looking at him iv. Supposing an individual in apparent health exhibited great sleeplessness, was excessively ravenous in his appetite, very dirty in his habits, said he was converted and a child of God, called himself Dick Turpin and King Richard, and wanted to dig his father out of his grave, because it was no use he should lie there, would you be prepared to say he was in a sound state of mind?

2 The Old Bailey Proceedings Online and the Monomania Corpus The OBP consists almost entirely of trial discourse. However, it is underexplored by (forensic) linguists and used much more by historians (including medical and legal historians e.g. Eigen 1995), a gap which is identified and exploited in the present research. OBP trial discourse consists of “spoken interaction (re)cast as writing” (Culpeper/Kytö 2000: 175), since the oral mode of lawyer, witness and

Quoting the prisoner   

   373

judicial talk is rendered into writing to produce a monthly publication for sale: The Proceedings. The methods, skills, and scribes vary across the two-and-ahalf centuries of the OBP and across individual trial records. By the 19th century Pitman shorthand had been invented and the adversarial trial started to resemble modern trial discourse more closely. Defence lawyers were present in the Crown Court (Archer 2010) and advocacy, rather than individuals prosecuting their own cases, was the norm (Archer 2005). By the 19th century, the defendant rarely speaks in the trial. Individual trials contain two structural parts – prosecution and defence cases – and both professional and lay voices: barristers, police, medical and lay witnesses. The 19th century Monomania Corpus is a “small corpus” (Cameron/Deignan 2003) downloaded from the OBP website. It contains 15 trials of between 300 and 32,000 words (a total corpus of around 133,000 words). The trials of Cooper (19,200 words) and McNaughten (31,744 words) are two of the three largest records in the Monomania Corpus. Quotation is a central activity in trial discourse, seen in the frequency of reporting verbs in the whole Monomania Corpus. Quotation can be introduced by a range of verbs, but in my corpus SAY is the most frequent verb (said is 22nd in a frequency list of 6,239 types) and therefore quoting with SAY is focused on in this paper. There are 1,258 occurrences of the lemma SAY, accounting for nearly 1 % (0.95 %) of the vocabulary, making speech reporting a central activity of trial discourse. The most frequent form is the past tense: said (0.68 %); then say (0.23 %), saying (0.03) and says (less than 0.01, only 7 occurrences), making uses of says ‘marked’ (Jakobson 1972), so that they stand out against a background of non-use when they occur. Say is not always used for quotation, of course. A frequent pattern is I cannot say X, indicating a common use to indicate a witness’s reluctance to agree with a lawyer question. It is also used for self-quotation by physicians in the corpus, as in I made a report – I say “Having to-day seen the above…” and by witnesses, as in I heard mother say to father on Thursday night, “I am…”. Note that the word that is the 11th most frequent word in the corpus, its prominent position stemming from the large proportion of indirect speech reporting, as we shall see. The inclusion of OBP trial discourse in news reports of the day highlights the relative incompleteness of the OBP (lawyers’ opening speeches and judges’ sentencing remarks are omitted from the OBP, but can be found in news reports) and points to the advantages of using additional sources. News reporting from both trials is used where it adds details absent from the trial data.

374   

   Alison Johnson

3 Quotation and Quoting in Prosecutions and Insanity Defences Bell (1991: 207–209) outlines the functions of direct quotation in news discourse: to produce “reportable facts” that are “valued as particularly incontrovertible”, to “distance and disown”, and to produce “soundbites” which add “the flavour of the newsmaker’s own words”. And Clift/Holt (2007: 6) point out the “dramaturgical quality of DRS [direct reported speech]”. These are all important factors to consider, alongside the Labovian insight of the evaluative aspect of quotation in storytelling sources. Quotation produces what Bublitz/Hoffmann (2011: 434) describe as an evaluative surplus (cf. also Bublitz, this volume). In quoting, the quoter identifies both a text and its context and relocates T[ext]1-in-C[ontext]1 as T2-in-C2 in order to reflect upon it […] in an evaluative manner […] Quoting as a speech act yields an evaluating surplus. (Bublitz/Hoffmann 2011: 434)

Clift/Holt (2007: 6) also note this phenomenon, drawing our attention to the simultaneous “replay” of an interaction with the opportunity for the quoter to “convey his or her attitude towards the reported utterance” (Clift/Holt 2007: 6). Quoting in the criminal trial context, where the interaction is replayed for a jury as primary audience, provides an occasion for these listeners to accept or reject the meanings produced as they weigh the evidence and come to their decision about the innocence or guilt of the defendant. Embedding reported speech is therefore a “highly selective and thus powerful resource of institutional meaning-making”, both legal and medical, and of reproduction, in the historical courtroom (Johnson 2013: 148) and in the contemporary one, constructing “intertextual authority” (Matoesian 2000: 879). Since legal discourse is conservative (Tiersma 1999: 95–97), rather than varying widely in lexis and structure over time, it is appropriate to consider Reisigl/Wodak’s (2009) discourse-historical approach, which suggests that discursive practices are embedded in history. Using this approach, we can test to what extent current defence advocacy and attitudes towards the police, expert witnesses and defendants are evident in 19th century trials. In the case of Cooper, the prosecution’s use of direct and indirect quotation is damning and incriminating for the defendant, since the witnesses seem well ‘rehearsed’ in supporting the prosecution case, whereas the defence picture, drawn out through family members, presents the working-class misery of the time and a young man seen as easy prey by a constabulary looking for results. In McNaughten’s case, we see the two institutions of the law and medicine fighting over culpability and insanity, as the defence quotes from medical interviews with the accused.

Quoting the prisoner   

   375

4 The Cases of Cooper and McNaughten Cooper is on trial for the murder of a policeman in Islington and McNaughten is on trial for the murder of Prime Minister Sir Robert Peel’s private secretary, Edward Drummond. Both occasions involve a literal smoking gun; each accused is armed with two pistols, seen to fire at and kill the murdered man, and in both cases the accused was arrested at the scene of the crime by police witnesses. In both trials insanity defences are offered. However, the McNaughten trial is replete with medical testimony, unlike in the case of Cooper, where medical evidence of his sanity or insanity is limited to direct observations from those who came into contact with the case in the course of the investigation, with the judge refusing to admit direct medical experts for the prosecution. In Cooper’s case, The Morning Chronicle (18 June 1842) reports: Dr Sutherland was then called; but Mr. Justice Pattison objected to Mr. Bodkin [for the prosecution] that calling medical gentlemen to prove sanity or insanity was not at all a desirable course. It was, in fact, placing the medical witnesses in the position which belonged solely to the jury. Mr Bodkin is said to have “acquiesced in [sic] the learned judge’s opinion”. The Morning Chronicle (18 June 1842)

The McNaughten trial (one year later) makes history, as no fewer than seven medical experts are called in a trial that contains 47 participants. In order to look at what the defence achieves through quotation in each case, we will first look at how the prosecution presents what the prisoner says.

4.1 Cooper’s Prosecution (13th/18th June 1842 OBP Trial Reference: t18420613–1766)² Thomas Cooper, aged 23, an unemployed bricklayer of 1 Rawnstorne Street, Clerkenwell (in the London Borough of Islington), lived about a mile north of the Old Bailey, just off the A1, the “Great North Road” out of London, the haunt of notorious 17th and 18th century highwaymen. He was charged with the “wilful murder of Timothy Daly” (a policeman, also spelled Daley) in the district of Highbury and Islington (Metropolitan Police N Division), London. The prosecution case, conducted by Messrs. Bodkin and Chambers, presents him as a notorious petty criminal of the sort described by Harper (1908: 4) as a “footpad”, a nineteenth century foot version of the mounted highwayman, but without “the slightest inkling of 2 The OBP records 13th June 1842 as the date of Thomas Cooper’s trial, but the microfiche of newspapers of the time records 18th June 1842 as the date of the trial.

376   

   Alison Johnson

romance” of his legendary predecessors, “a miserable, muddy, cowardly figure, for whom no one ever had a good word”, as reflected in the newspaper headlines of time, from his arrest, 5th May 1842, to his trial and conviction, 18th June 1842 (for example: 1 and 2). (1)

Horrible murder by a highwayman. The Standard (London, England), Friday May 6 1842.

(2)

Murder of a policeman, and two other persons shot, by a foot-pad. The Essex Standard, and General Advertiser for the Eastern Counties (Colchester, England), Friday May 13 1842.

On the afternoon of 5th May 1842, Cooper was arrested for three offences: the shooting and wounding of Police Constable Moss, the shooting and wounding of a baker, and the fatal shooting of a second policeman: PC Timothy Daly. The crimes were the result of an unfortunate set of coincidences: a uniformed police constable on patrol, a well-to-do gentleman out for a walk and Cooper, an unemployed and disenchanted young bricklayer, out on his own, carrying a couple of pistols and powder, probably under the influence of arsenic and laudanum,³ which he took in six penny amounts (“he said he had taken sixpenny worth of laudanum and sixpenny worth of arsenic,” Figure 1, line 48). PC Moss was on patrol in the rural and affluent area of Hornsey, having “previously […] received information of robberies having been committed in that neighbourhood and a description of the man suspected had been read at the station” (testimony of Moss). He encountered “a gentleman” walking in the area and had “observed that he had a large bunch of seals” (OED: “an ornamental appendage to a watch chain”). In saying this Moss’s testimony self-constructs a policeman alert to potential criminality by the unruly, poor, criminal class preying upon the rich gentleman. The coincidence of Cooper arriving on the scene makes PC Moss suspicious. The wellto-do gentleman does not figure any more in the story and no attempt is made to rob him. Moss testifies: “when he [Cooper] saw me he ran to the opposite side of the road, and went down a little bank, into the adjoining field by the road-side”. When Moss asked Cooper “what he did there” and when he responded “nothing particular”, Moss “then told him if he did not give a better account of himself than that” he would “take him into custody”. He does not say what for, but the Vagrancy Act of 1824 (Legislation.gov.uk) had created the offence of “loitering

3 Arsenic and laudanum were both easily obtainable at the time from an apothecary. Arsenic was considered by early Victorians as something of “a panacea” (Doyle 2009: 311)

Quoting the prisoner   

   377

with intent to commit an indictable offence” and “wandering abroad […] and not giving a good account of himself”. The story unravels sadly but not entirely unpredictably, given the coincidences reported and Cooper’s past history with the police. Cooper “sprang from the hedge […] aimed the pistol in his right hand […] fired […] and something from it went into my arm, and came out just above the elbow” (testimony of Moss). A two mile chase ended with Cooper cornered “in Highbury Park south” and “against the paling” (testimony of Mott, the baker). A stand-off ensued with the still-armed Cooper facing a collection of his pursuers, variously armed with a gun, a broomstick and a brick, and Daly, the policeman, who had joined the chase after being summoned by Young the waiter. Cross-examination questions, as we will see, suggest a defence of provocation, which resulted in him firing the two pistols at Daly, fatally shooting him, with a large number of eye witnesses ready to give evidence in court. His actions and direct and indirect speech are recorded in detail as evidence against him (Figure 1). We see his speech and conversation at all the stages of the story: the chase and the shooting (lines 1–8 and 12–23); the arrest (lines 9, 24, and 25); on the way to the station (lines 25 and 26); to the Magistrate (line 11); whilst in the custody of Waddington, the gaoler at the Police Court (lines 27–32). None of the examples in Figure 1 shows Cooper to be contrite, making what was said and recorded a powerful indictment against him in the prosecution case. The witnesses seem to be well versed in their ‘lines’, each confirming the incriminating words Cooper uses (lines 4, 16 and 20) and the police investigation interviews have produced a seemingly watertight case in respect of their murdered colleague. Two previous robberies are unproblematically assigned to him at the end of the trial (lines 33–34), with the identification corroborated in part by his apparently using the same modus operandi on both occasions: the notorious highwaymen’s threat (Harper 1908: 49) made famous by Dick Turpin:⁴ “Your money, or your life” (lines 33–34). Even while in custody, Waddington, gaoler at Clerkenwell Police Court, reports a threat to his life (lines 31–32). The prosecution presents a case strengthened by direct and indirect quotation of incriminating

4 Dick Turpin was a notorious highwayman who robbed rich travellers and mail coaches on the roads out of London in the 18th century. He frequented the Great North Road, the A1 north out of London and is described by Harper (1908: 246) as “the greatest of all the rascals who made the name of the Great North Road a name of dread”. He was executed for his crimes in 1739 (Seal 2011: 128), after being imprisoned in York castle. His crimes were romanticised after his death in “penny lives of Turpin”, popular publications of the early 19th century, sold by pedlars and street-sellers (Harper 1908: 128–9). Thomas Cooper seems to be a young man ‘haunted’ by this notorious highwayman.

Moss

Moss

Young

Young

Young

Howard

Turnbull

Turnbull

Turnbull

Turnbull

Turnbull

Smith

Smith

Smith

Smith

Smith

Smith

Smith

Smith

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

knife were taken from him, and he said he was done then, they had got all his weapons from

hand, and said, “You ought to have this” – he presented a pistol, and said, “You shall have this”

they are” – he was holding them out, one in each hand, at the time

Daly said he thought the pistols were not loaded – the prisoner said, “You may take my word

said no more, but sidled down the field – I saw him go through a motion as if loading a pistol

man in Hornsey-wood” - he said he had done nothing – he pulled out two pistols and presented

I went after him – I said, “The people are hallooing after you and, and saying you have shot a

I was coming up towards him – he said, “What do you want with me? I have done nothing”

eenwood, the Magistrate, asked me which he said – I was never in hesitation about his words

I am certain the prisoner said, “I would serve you the same,” not “I will serve you the same”

asked him how he could do such a thing as this – he said, “I would have served you the same”

ting the knife from him, he said, “I am done, you have got all my weapons, but don’t ill-use me”

are loaded” the prisoner said, “The first one who dares to attack me shall have the contents”

the pistols – on their being taken he said, “I am done” – I said, “It is pretty near time you had”

hat to me at the station – I showed it to the prisoner, and asked him if it was his – he said, yes

Daly said, “Surrender, I don’t think those pistols are loaded” – then he said they were, and he

not think those pistols are loaded” – the prisoner said they were, and he would shoot two more

when he drew that knife out, he said if I followed him any further, he certainly would do for me

and he said if I attempted to follow him, or make any alarm, he certainly would shoot me dead

378       Alison Johnson

Mott

Grover

Simmonds

Wheeler

Wheeler

Wheeler

Wadding

Wadding

Wadding

Wadding

Wadding

Wadding

Roach

Swears

21

22

23

24

25

26

27

28

29

30

31

32

33

34

presented a pistol at me, and said, “Your money, or your life” – I told him I had got very little

he turned round, presented a pistol in Mrs. Cooks’ face, and said, “Your money, or your life”

about there?” – he said, “If I could have got hold of it, I meant to have done away with myself

towards where my cutlass hung – he did not get off the box – I said, “Mr Tom, what are you

how came the grass to be in your pistol?” – he said he had no paper, and he wadded the pistol

in my custody that some grass was issuing from Mott’s wound, and I said to the prisoner, “Tom,

he said he wished Penny had been there instead […], he should have shot him more freely

when I told him it was a bad job, he said, yes it was, but he should have shot any man that

a thing? – he said he gave them notice, he gave them all warning, and told them if they attempt

as he was being conveyed to the station, one of the by standers said, “How could you do such

knife was taken from him, he said, “It is all done now, I will surrender, I have got nothing more”

I heard Daly ask him to surrender, and to give the pistols up – he said something in reply

after the weapons were taken from the Prisoner, he said, “I will surrender, don’t ill-use me”

Daly asked him what he had been doing – he said he had been cutting turf – I asked him what

prisoner told him to keep off, he said he did not believe the pistols were loaded – before that

Figure 1: Concordance of he/the prisoner said in the trial of Cooper, produced using Wordsmith (Scott 2011), showing a range of events leading up to and after arrest and whilst in custody.

Mott

20

Quoting the prisoner       379

380   

   Alison Johnson

words from the prisoner and the news coverage adds to the prosecution view the negative public threat of the highwayman or foot-pad.

4.2 McNaughten’s Prosecution (27–28 February 1843 OBP Trial Reference t18430227-874) McNaughten’s murder trial is the largest monomania trial in the corpus and is famous for creating the McNaughten rules,⁵ the rules on insanity in criminal cases in England and Wales (and in a number of US states, New Zealand etc.), and for the successful, but highly controversial, defence that saw McNaughten acquitted of murder by reason of his insanity at the time. The prosecution was conducted by The Solicitor General and Messrs. Aldolphus, Waddington and Gurney. Following his arrest on 20 January 1843 and between McNaughten’s Magistrate’s Court appearance (21 January 1843) and his trial (27–28 February 1843), there was widespread reporting of the case in the news, an examination of which adds details to the OBP. As in the Cooper case, the first prosecution witness is the police officer who encountered him. In the McNaughten case, Police Constable James Silver witnesses McNaughten shooting his victim, as a passer-by in Whitehall. He intervenes and arrests McNaughten, and is recorded as giving evidence saying: “I heard the report of a pistol on the opposite side, which attracted my attention – I saw a gentleman reeling, with his left hand under his left side – I saw his coat on fire – he was near Mr. Tatham, the gunsmith’s, on the opposite side of the street to where I was” (Testimony of PC Silver).

The only words PC Silver reports as uttered by McNaughten are insignificant: “the prisoner gave his name and address, ‘Daniel M’Naughten, 7, Poplar-row, Newington’”. But Police Inspector Tierney, the Inspector in charge of the police station where McNaughten was held overnight before going before the magistrate, gives evidence that is densely filled with quotation, and his testimony is the focus of detailed cross-examination as part of the defence work, to which we now turn.

5 There are many spellings of McNaughten’s name: M’Naghten, M’Naughten, McNaughtan, across the documents I have consulted. I have adopted McNaughten here throughout, being the most frequent.

Quoting the prisoner   

   381

4.3 Defending Cooper (1842) and McNaughten (1843) – Mr Horry and Mr Cockburn Mr Horry (18–071881) was an experienced barrister by the time he acted as Cooper’s counsel in 1842, having acted in over 100 Old Bailey trials by that time. He is present in a total of 1,010 OBP trial records over 36 years, his first trial in 1837, two years after being called to the bar, after being admitted to Gray’s Inn in 1831, aged 24 (Foster 1889: 441), and his final trial in 1875, six years before his death. Cooper’s case was his 127th at the Old Bailey. All of his cases (he most often acted for the defendant) gave him intimate knowledge of the Victorian poor, with the majority of trials (816 out of 1,010) for theft, including: simple larceny (255 cases), robbery (53), stealing from a master (142), and pocketpicking (126). Sir Alexander Cockburn (1802–1880) of the Middle Temple, McNaughten’s counsel, is described as “highly courteous, witty and eloquent in company, and keen to share anecdotes with his friends, who included Charles Dickens”. He had a “taste for high-profile cases”, both as a barrister and later as a judge (Lobban 2004 online). McNaughten’s case was certainly high profile, because, as the newspapers said, he had attempted to assassinate the Prime Minister, Sir Robert Peel, and had shot his private secretary, Mr Edward Drummond instead. For both defence barristers, as is the case in the contemporary adversarial trial, the first primary mode of defending is during the prosecution case as they cross-examine prosecution witnesses. What they suggest to the jury in cross-examination questions and what is elicited in answers attempts to undermine the prosecution case and starts to form the defence case. In the next section we examine the defence work that is done in this cross-examination of prosecution witnesses. As we saw in Cooper’s prosecution, the testimony of witnesses, strengthened by direct evidence in Cooper’s own threatening words throughout the course of events, from his encounter with PC Moss to shooting PC Daly dead, presents a convincing prosecution case. And McNaughten, too, is quoted as self-incriminating. How then were Mr Cockburn and Mr Horry to defend these men seemingly already incriminated by their own words?

4.3.1 Defending in Cross-examination of Prosecution Witnesses – Direct and Indirect Speech, and Narrative Report of Speech Act (NRSA) Mr Horry adopts three strategies in his cross-examination of prosecution witnesses: to attack the police evidence, suggesting that the police were on the lookout for any suitable suspect for a spate of recent robberies; to attack the pursuers’ evidence, suggesting that they hounded and intimidated Cooper leading him to fire the pistol killing PC Daly; and, using the opportunity of the prosecu-

382   

   Alison Johnson

tion calling Mr Drewry, the doctor who certified PC Daly’s death, to elicit from him a lesson on insanity for the jury. He begins his cross-examination of the first prosecution witness, PC Moss, with the suggestion that he was under “instruction”, a particularly indirect form of indirect speech (Example 3), described by Leech and Short (2007: 259) as “narrative report of speech acts” (NRSA), where sentences “merely report that a speech act (or a number of speech acts) has occurred, but where the narrator does not have to commit himself entirely to giving the sense of what was said, let alone the form of words in which they were uttered” (Leech/Short 2007: 259–260). The unreported speech act is implied to have gone something like: “Apprehend someone for these robberies in the area.” (3)

Horry to PC Moss Q. I believe you had instruction to apprehend somebody, not anybody by name? A. No.

His opening argument is therefore one that suggests that Cooper might have been a likely person to take into the station for questioning. He then moves to his second argument: one of intimidation, made by inference. His questioning (Examples 4–7) contributes to a narrative of blame on the pursuers, not on the pursued, and presents a counter-narrative to the prosecution account. In (4) the use in indirect speech within the “when you say”-question forms the first part of a two-clause question that contrasts a prosecution view (looking from right to left) with a defence view (throwing his head backward and forward as fast as he could). One is rational and vigilant and the other is desperate. In (5) Horry draws attention to the enclosed space that Cooper is trapped in. We do not have his next question (the dashes indicate where questions have been asked but not recorded), but from Howard’s answer we can suggest that Horry used a NRSA within a question such as: “Did he not offer to get out?”, implying Cooper’s willingness to surrender. (4)

Horry to Young (a waiter at Hornsey-wood-house tavern who found PC Moss lying shot) Q. When you say he was looking from right to left, was not he throwing his head backward and forward as fast as he could? A. No. Q. Was it not just about that time that he fired when he saw the gun in your hands? A. Yes.

Quoting the prisoner   

   383

(5)

Horry to Howard (a baker in his cart on Hornsey Road, spoken to by Young) Q. The wooden enclosure is spiked, I believe? A. No, it goes up sharp – it is not six or eight feet high […] – he did not offer to get out while I was looking at him

(6)

Horry to Turnbull (a gardener at work in his master’s garden in Highbury) Q. You had a brick in your hand? A. Yes – I was not at the enclosure – Mott had a stick in his hand – I did not hear him offer to strike the prisoner, nor know of a stick being broken over the paling […] I am certain the prisoner said, “I would serve You the same,” not “I will serve you the same” – Mr. Greenwood, the Magistrate, asked me which he said  – I was never in hesitation about his words […] Daly did not call the people to come up and help – all he said was, “I don’t believe those pistols are loaded.”

(7)

Horry to Smith (a gardener on his way home near Highbury-park west) Q. Try and recollect whether the stick Mott had was not broken over a paling? A. I do not know – […] I heard the prisoner say he did not mean to shoot any one, but the first who came near him should have it

Turnbull, a gardener, is most carefully questioned, his epistemic certainty (Huddleston and Pullum 2002) marked by “I am certain” in the reporting clause, followed by direct speech in the record (“I am certain the prisoner said, ‘I would serve you the same’”) (6). Though Turnbull provides negative answers (“I did not hear him [Mott] offer to strike the prisoner”; “Daly [the policeman] did not call the people to come up and help”), the suggestion that Cooper was threatened by the policeman and goaded and intimidated by the crowd is, nevertheless, successfully planted in the jury’s mind. In (7) Smith gives evidence via indirect speech representation (underlined) that adds weight to a defence argument that emerges out of the questions and answers: Cooper only shot after provocation from the crowd. The most careful cross-examination is reserved for the medical witness, Edward Drewry, who appears at the end of the prosecution case. His testimony consists of 68 words, in which he describes attending and examining PC Daly’s dead body, finding a pistol shot and confirming that as the cause of death. There follow, however, 1,100 words of cross-examination question and answer record on the nature of Cooper’s sanity or insanity. Given that Drewry has appeared as a prosecution witness to establish the cause of death of a serving police officer and has “heard the evidence in this case, and the conduct and demeanor of the pris-

384   

   Alison Johnson

oner deposed to throughout the transaction” (question by Mr Bodkin, prosecution barrister, to Drewry), it is unsurprising that he resists the cross-examination questions that might undermine the prosecution case, although he has a duty to the court to give his opinion. Example (8a) is a particularly striking cross-examination question, which includes hypothetical indirect quotation (underlined). (8a)

Horry to Drewry Q. Supposing an individual in apparent health exhibited great sleeplessness, was excessively ravenous in his appetite, very dirty in his habits, said he was converted and a child of God, called himself Dick Turpin and King Richard, and wanted to dig his father out of his grave, because it was no use he should lie there, would you be prepared to say he was in a sound state of mind?

Drewry’s answer is not the preferred one, built into the controlling, closed yes/ no question’s design (Archer 2005). Horry has filled the question with a hypothesis that contains no less than six symptoms of insanity, including the defendant calling “himself Dick Turpin”, making his question vivid with behavioural and indirect speech details (Tannen 2007) that make it difficult for Drewry not to say “no”. What Drewry replies, however, is: (8b)

continued from 8a A. It would depend very much on the circumstances

And, as Horry continues to press him, Drewry replies as in (8c) (dashes indicate further cross-examination questions that are not transcribed by the short-hand notetaker). (8c)

continued from 8b 1 if satisfied those things existed, I should say there might be some 2 peculiarity about him – he might have some of those symptoms without 3 having the mind affected at all – if they all existed, I should say he was not 4 in a sound state of mind – an alienation of the natural affections, a person 5 fond of his relatives at one time, and subsequently exhibiting great violence 6 to them, is one symptom of insanity – […] – an acute state of fever generally

Quoting the prisoner   

7 8 9 10 11 12 13 14 15 16 17 18 19 20

   385

affects the brain, as putrid fever, and a malignant form of typhus – the brain is peculiarly liable to be acted on in that disease. Q. And sometimes the effects of the disease do not pass rapidly away? A. No, they last for a considerable time – I have known that disease, and disorders of that character, affect the brain for some years. Q. Supposing a person affected in that way, and a year or two after wards had another severe illness of an inflammatory kind, that might increase the effect produced by the fever? A. Certainly […] – in a person of unsound mind it is quite uncertain at what time the disease will reach the period of mania – any circumstance may bring it forth. Q. It may come out at any time, and particularly when acts of violence are threatened? A. Yes – a great many individuals only exhibit unsound ness of mind on certain subjects, that is monomania.

From Drewry’s answers, we can see that Horry tenaciously pursues an acknowledgement that the hypothetical actions of “a person” do amount to insanity. The epistemic modality in the uncertain “might” (line 1) and conditional “if” (line 1) show that Drewry is not to be persuaded, as he continues to resist in cross-examination, preserving his professional opinion, though it is made to look increasingly untenable. When Horry introduces a similar level of weak epistemic certainty into his last three recorded questions (“sometimes” line 9, “might” line 13, and “may” line 18) Drewry is eventually forced to concede (line 19). Crucially the concession comes in relation to the likelihood of mania occurring “when acts of violence are threatened” (lines 18–19), this being important in the case of Cooper, since Horry’s cross-examination of the prosecution witnesses has established that, when Cooper was cornered, Turnbull the gardener had a brick in his hand, and Mott the baker had a broomstick. Horry’s persistent questioning continues to ask about a theoretical case, refusing to apply the symptoms to Cooper, making Drewry confirm theory, before Horry deals with Cooper’s case in his defence. The force of the initial question, “Supposing an individual in apparent health exhibited great sleeplessness, was excessively ravenous […] said he was converted […] called himself Dick Turpin […] would you be prepared to say he was in a sound state of mind?” (8a), with the hypothetical indirect speech report, signalled by the

386   

   Alison Johnson

verb “supposing” and followed by indirect quotation, forces the jury to consider a medical theory of insanity in advance of hearing from the defence witnesses. The cross-examination work that is engaged in is therefore damaging for the prosecution, because, despite Drewry’s strength of resistance to cross-examination on the subject of monomania, and his insistence on his opinion in the face of overwhelming symptoms in the hypothetical case, his eventual concession will not be lost on the jury. This makes hypothetical indirect quotation a powerful defence strategy, as it anticipates the defence witness testimony. In the McNaughten case, Mr Cockburn, like Mr Horry for Cooper, has a difficult advocacy job to do for his client, given the weight of prosecution evidence. As well as being arrested in possession of two pistols, after being seen to shoot the Prime Minister’s Private Secretary in broad daylight in the public thoroughfare of Whitehall, the prosecution has a number of incriminating statements made by McNaughten in the cells after caution to add weight to their case. Figure 2 shows Police Inspector Tierney’s (the inspector in charge of the police station where McNaughten was held after arrest) evidence-in-chief (in the left hand column) and his answers in cross-examination (on the right). Uses of direct and indirect speech are underlined. The Figure puts the two activities side-by-side, showing them simultaneously, though the cross-examination was of course later in the trial record than the examination. During examination-in-chief, Tierney is recorded as having given detailed accounts of his own and McNaughten’s speech (Figure 2, left-hand column). Tierney uses direct speech to indicate verbatim accuracy (Leech and Short 2007) and indirect speech (underlined in left-hand column), though, when cross-examined, he claims that he “cannot give the whole conversation” (Figure 2, lines 23–24, right). As the activity changes from examination to cross-examination, the mode of reporting changes from direct and indirect speech (underlined in column 1) to the metalanguage of speaking (underlined in column 2). Metatalk is used for excuses. Defence barrister Mr Cockburn’s tough cross-examination of Tierney amounts to an accusation of the policeman abusing his duty (Q: “Is it part of your duty to visit the prisoner in his cell?”, lines 1–2; “What was your motive…?”, lines 16–17; “Do you mean you had no motive…?”, lines 25–27) and interfering with the course of justice. The cross-examination questions “assert a version of events which conflicts” with Tierney’s prosecution evidence version (Heffer 2005: 137). In preparing his case, Mr Cockburn was probably aware of adverse reporting on police behaviour in the case by the press. The excerpt in (9) appeared in the newspaper John Bull on Monday 13 February, two weeks before the trial. It consists of a letter to the editor, signed “a lawyer”, and quotes verbatim and at length from the earlier Magistrates’ Court hearing. In the John Bull news text, phrases and sen-

Quoting the prisoner   

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

   387

Police Inspector Tierney’s testimony for the Prosecution

Police Inspector Tierney cross-examined by Mr. Horry

when the constable left the cell, I remarked to him [McNaughten], “I suppose you will assign some reason this morning to the Magistrate for the act you have committed?” he said he would, a short one I said, “For that matter, you might have stated anything you chose last night after the caution I have given you” he then said he was the object of a persecution by the Tories, or Tory persecution, or something of that sort – that they followed him from place to place with their persecution – he seemed inclined to go on I said, (merely to turn the conversation, I did not want to hear the confession,) “I suppose you are aware of the gentleman you have shot at?” he said, “It is Sir Robert Peel, is it not?” I said, for the moment, “No,” and then retracted, and said, “We don’t exactly know who it is yet, but recollect the caution I gave you last night, not to say anything that might be used in evidence against you” he replied, “But you won’t use this against me?” I said, “I will make no promise, I gave you a caution” – I then left the cell – I took him up to Bow-street [Magistrates’ Court]

Q. Is it part of your duty to visit the prisoner in his cell? A. Yes — if I put interrogatories to the prisoners under my care as long as I do not interfere with the case in point I do not see any harm in it — I was not directed by any one to put questions to this man — when I visited him I had my uniform on — I cautioned him more than once not to say anything that might criminate himself […] I put the question to him because I wanted to get some information relative to the man himself, his former life, not to make use of in the course of any investigation that might take place — I had no intention of bringing this forward as evidence, until he mentioned the name of Sir Robert Peel […] Q. What was your motive in wanting to get information relative to his past life? A. I suppose nothing but the anxiety of human nature to know, under such revolting circumstances, who and what the man was — I cannot give a verbatim account of the whole conversation — the first conversation was the caution and where he came from — I cannot give the whole conversation […] Q. Do you mean you had no motive for asking him whether he wanted to make a statement to the Magistrate? A. I will tell you my motive — I wished him to understand that I was quite at liberty to hear anything he wished to say voluntarily — I thought the responsibility was removed from my shoulders after giving him the caution the night before

Figure 2: Quoting the prisoner in Police Inspector Tierney’s testimony: examination (left) and cross-examination (right).

tences are highlighted through the use of italics, drawing attention to them (and its formatting is reproduced here). Having drawn attention to these phrases (9a), the lawyer writing to the newspaper performs a critical analysis of the reported testimony (9b).

388   

(9)

   Alison Johnson

a.

‘The Police Inspector and M’Naughten’ – letter to the editor in John Bull. Police Inspector Murphy [sic] examined – I saw the prisoner on the evening he was apprehended, at Gardiner’s-lane station-house; it was in the outer room. I afterwards visited him four or five times in his cell about six o’clock. I had a communication with him. […] I gave him a strict caution not to say anything to criminate himself, as anything that might pass between us would be made evidence against him. […] I saw the prisoner again the next morning about a quarter past nine o’clock. He asked me for some water to wash with […] When the constable was gone out of the cell, I said to the prisoner, “I suppose you will assign some reason to the Magistrates this morning for the act you have committed;” he answered, “I will, but it will be a short one.” I then told him after the caution I had given him, he might say anything he pleased to me. He then said that the Tories had adopted a system of persecution against him; they followed him from place to place with that persecution. I then said, “I suppose you are aware who the gentleman is that you shot at?” He replied, “It is Sir Robert Peel, is it not?” I said, “No;” but immediately retracted the word, and said, “We are not exactly aware who the gentleman is.”

b. Now, Sir, we have here a cunning, busy, meddling police-inspector, in contravention of the law, and in violation of his duty, visiting the prisoner at intervals, six or seven times, affecting to give him “a strict caution” at each interview “not to say anything that may criminate himself,” at the very time that he is artfully putting questions that assume his guilt, and that manifestly intended to worm out an admission of it. Mark the conduct of this police-inspector, as exhibited in his own statement. “I suppose you are aware who the gentleman was that you shot at?” What answer did he expect? What answer was he fishing for, but one by which, if given, the accused must criminate himself? The prisoner replies, “It is Sir Robert Peel, is it not.” The question was successful – it brings an admission that he fired the pistol, but has shot the wrong person. […] There is something dark and treacherous in all this. The material here provides a rich cross-examination resource. We do not know whether Cockburn saw this letter (he might even have been the anonymous lawyer), but this critical approach to the Inspector’s actions was the one he took

Quoting the prisoner   

   389

in the trial. After such damaging cross-examination, the Solicitor General, for the prosecution, re-examines Tierney and gets him to read out McNaughten’s confession to the Magistrate (10). (10)

Solicitor General re-examines Inspector Tierney 1 Tierney: I was present when the prisoner was asked what he had to say by 2 Mr. Hall [the Magistrate], at Bow-street – I heard the statement he made – it 3 was read over to him afterwards, and I saw him sign it – this is the state4 ment he made on the 21st – it is the exact words – (read) – “The prisoner 5 after being cautioned as to his statement, says, ‘The Tories in my native city 6 have driven me to this, and have followed me to France, Scotland, and other 7 parts; I can get no sleep from the system they pursue towards me; I believe 8 I am driven into a consumption by them; they wish to murder me. That is 9 all I wish to say at present; they have completely disordered me, and I am 10 quite a different man before they commenced this annoyance towards me.’”

Here the prisoner speaks through the Police inspector. The Inspector ventriloquizes him (Tannen 2007) while the accused stands silent in the court. These words are appropriated by the prosecution (“I heard” line 2; “I saw” line 3) as evidence of his guilt and it is particularly notable that the record marks McNaughten’s words with the metadiscursive frame (Hyland 2005: 51) “it is the exact words  – (read)”(line 4), and then directly quotes from the written statement, marking the quoted words with the infrequent present tense form “says”, rhetorically appealing to the jury’s logos or reason (Ramage and Bean 1998: 81). The defence case follows and here history is made.

4.3.2 Defending through Defence Witness Examination – the Vividness and Power of Direct and Indirect Speech Representation Allocating the McNaughten trial data to 47 different files by witness name, we find that the prosecution and defence have different vocabularies. This is starkly seen in the concordance line examples for “insane” and “insanity” (Figures 3 and 4), all of which come from the medical witnesses for the defence, apart from one, which is Mr. Solicitor General’s cross-examination of the defence witness Jane Drummond Patterson, McNaughten’s landlady for two years before the trial (Figure 3, line 7). On the right hand side of each concordance the eminent defence medical witnesses are lined up: Hutchinson, Morrison, Monroe, Winslow, Key. “Insane” and “insanity” are defence words; they are not in the prosecution lexicon or in their witness accounts.

gay, that you considered the prisoner insane, to be labouring under delusion Edward Thomas Monroe

5

I can testify myself that the man is insane,” and he urged me to take him he knew quite well the young man was insane I would not let him go and he

8 9

Jane Drummond Patterson

Jane Drummond Patterson

Mr. Solicitor-General

Not guilty.

Rev. Alexander Turner

Sir Alexander Morrison

Figure 3: Concordance of “insane” by participant in the McNaughten trial.

10 led me to the conclusion that he was insane coupling that with the history of Edward Thomas Monroe

Allison’s to speak to the prisoner being insane, were you not?

7

insane.

certainly thought at the time that he was insane my impression was such

4

6

act with which he is charged, he was of insane mind in my judgment,

3

William Hutchinson, Esq

Commissioner of police I think he was insane at that time I do not think that

2

William Hutchinson, Esq

heard to-day I suppose that he was insane at Glasgow eighteen months ago

1

390       Alison Johnson

Looking at the trial record, we can see that a combination of indirect speech quotation and the work of defence advocacy produces a powerful language of defence (11). In (11) Mr Cockburn’s elicited evidence of the medical questioning allows Monroe, who was physician to Bethlem asylum, according to Eigen (1995: 201), to tell “the substance” of his “notes” taken “in gaol” (testimony of Edward Monroe). The multiple indirect speech that-clauses (plus zero-that, signalled by and) and small number of “I asked” and “he said” reporting clauses, produce a ventriloquized (Tannen 2007: 22) testimony, animating McNaughten’s voice in his presence, through the professional voices of advocate and expert doctor. In court,

(11) Marcs? Q. Making a nice definition of insanity, which he calls “moral mania” are some recent theories on the subject of insanity, are there not? Have you doubt of the existence of the prisoner’s insanity, and that the act with which

7 8 9

Forbes Winslow, Esq

Edward Thomas Monroe

Edward Thomas Monroe

Edward Thomas Monroe

Edward Thomas Monroe

Edward Thomas Monroe Aston Key, Esq

17 may exist in various minor forms I think insanity involves delusion generally 18 turned my attention particularly to cases of insanity, but I have been occasio Figure 4: Concordance of “insanity” by participant in the McNaughten trial.

Edward Thomas Monroe

16 terra “moral insanity” I have heard of an insanity which irresistibly compels

15 I understand what is intended my moral insanity, a perversion of the passion Edward Thomas Monroe

14 to say whether I thought he was assuming insanity or not, and I came to the

13 his life, I have not the remotest doubt of his insanity I am quite satisfied of it Edward Thomas Monroe

Edward Thomas Monroe

to avoid persecution? Q. Do you consider insanity may exist without the

6

Edward Thomas Monroe

12 self control I have no doubt that this partial insanity may exist, and the facul

an established principle in the pathology of insanity that there may exist a

5

Edward Thomas Monroe

Edward Thomas Monroe

for his actions, exciting a partial insanity only, although the rest of the

4

Sir Alexander Morrison

11 not at all impair my conviction as to his insanity I have known many lunatics

labouring under different descriptions of insanity. Q. Do you consider the

3

Sir Alexander Morrison

Forbes Winslow, Esq

mind in my judgment, the nature of his insanity is a morbid delusion on the

2

William Hutchinson, Esq

10 and am author of a work on the subject of insanity. I have been in Court

were persecuting him that is a symptom of insanity which we see frequently

1

Quoting the prisoner       391

as a result of a series of unrecorded, but clearly enabling wh-questions, lawyer and expert witness work together to produce elegant and eloquent evidence of the prisoner’s speech: his narrative of delusion.

Medical testimony for the defence EDWARD THOMAS MONROE [also spelt Monro]: it was on the 18th of Feb. – he commenced by stating that he was persecuted by a system, by a crew at Glasgow, Edinburgh, Liverpool, London, and Boulogne – that they pursued or followed him wherever he went, that he had no peace – he said

392   

   Alison Johnson

it would kill him – that it was a grinding of the mind – I asked him if he had taken any medical advice – he said physicians would be of no use – that tons of drugs could not benefit him – that in Glasgow he had observed people in the street speaking of him – they said, “That is the man,” that he was considered a murderer, and the worst of characters [ 22 that clauses and 3 and + zero that clauses follow] – that he imagined the person at whom he fired at Charing-cross to be one of the crew – one of the system destroying his health. Eigen (1995: 133) notes that lawyers and physicians formed, at this time, a “mutual admiration society,” and their combined forces, here, bring the science of asylum medicine and the rationality of law together, to produce a persuasive professional discourse in which to present the multiplicity of McNaughten’s insane delusions to the court. An array of people and places, fears and delusions, plague McNaughten’s life and his story is told with the eloquence of a learned doctor using powerful pairs of nouns (a murderer, and the worst of characters), verbs (pursued or followed him), adjectives and the rhetoric of binomials (watched, and followed 12a; beastly and atrocious; untrue, and insufferable 12b), figurative similes (tossed like a cork on the sea 12a), and contrastive, patterned speech (whether in town or country, or by the sea shore 12a). (12)

a.

that he was tossed like a cork on the sea, that wherever he went, whether in town or country, or by the sea shore, he was perpetually watched, and followed

b. that he had seen things in the Glasgow Herald, beastly and atrocious, insinuating things untrue, and insufferable. The transformative power of indirect quoting organises and professionalises the possibly incoherent and definitely fragmented narrative, which was elicited by the five doctors “in turns” in their interviews with McNaughten in prison and “the notes” made to record these interviews (reported in Monroe’s testimony). Cockburn’s “melodious voice”, as defence barrister, was no doubt exuding the “warm charm”, “distinguished air”, and courteousness and eloquence that his biographers describe (Lobban 2004 online), and Monroe, “fourth generation of his illustrious family to find himself in ‘the mad business’” (Eigen 1995: 145), was one of the eminent band of “mad-doctors” who combined the professional roles of asylum keeper, author, and witness, giving him expertise in speaking for the insane. These “early forensic-psychiatric witnesses” who “were attempting to become more than mere custodians, to become scientific observers and chroniclers of

Quoting the prisoner   

   393

the essence of insanity” (Eigen 1995: 132), Monroe in McNaughten, and Drewry in Cooper, made powerful spokesmen for or against the prisoner. Sir Alexander Morrison, M.D., corroborates Monroe’s evidence as he “concur[s] with him in the conclusions to which he has come”: that the “nature of his insanity is a morbid delusion on the subject of persecution” that “acted on his mind so as to deprive him of the exercise of all restraint against the act” (Evidence of Morrison). The prosecution and defence views are, as in the case of Cooper, entirely different, as seen in the way the words of the prisoner do different work in the mouths of the prosecution and defence witnesses as elicited by the skilful advocacy of the barristers. As the first, key, medical defence witness, Edward Monroe, states in his testimony, two doctors interviewed McNaughten at Newgate prison on “the part of the Crown” (Dr. Bright and Dr. Sutherland junior), and three doctors on behalf of the prisoner’s friends (Monroe with Sir Alexander Morrison and Mr M’Clewer). Once the medical experts had given testimony concerning their conversations with McNaughten in gaol, and based on their own interviews, the prosecution doctors agreed with the defence experts and the prosecution, in the voice of Mr. Solicitor General, as reported in the newspaper, John Bull, but not recorded in the trial record, conceded that after the strong intimation received from their Lordships (Lord Chief Justice Tindal said both himself and his Learned Brothers felt that a very strong case of insanity had been made out), he could not think of asking them [the jury] for a verdict against the prisoner. […] he [Mr. Solicitor General] felt bound to admit that a most powerful defence had been made on behalf of the prisoner. (John Bull February 13 1843)

Cooper’s defence, though rigorous, did not fare so well. Even after the masterful strategy of exploiting the prosecution physician, Mr Drewry, for defence work and though Mr Horry mounts a vigorous defence, his advocacy is unsuccessful in presenting Cooper to the jury as insane. The trial record does not contain Mr Horry’s defence speech to the jury, but the newspapers do, saying that he delivered it “in a quiet voice” and “addressed the jury [at] great length, and after dwelling upon the difficulties the prisoner’s mother had to contend against in procuring witnesses with her limited means, called upon the jury to return a verdict in accordance with the evidence he should lay before them” (Leeds Times, Saturday 25 June 1843). When he then calls witnesses for the defence, he designs his defence around Cooper’s life and speech, as recalled by his mother, brothers, and various lodgers with the family: people who knew his habits well from intimate observation over a lifetime. These people and their evidence are presented by Horry as ‘experts’ (in the absence of funds to procure medical experts) on Cooper’s state of mind. Most

394   

   Alison Johnson

compelling of that evidence was the testimony of Isabella Cooper, Cooper’s mother. Horry elicits multiple ‘he said’ quotations which establish detailed evidence of insane speech and behaviour for the court, as well as NRSA (13 and 14). The hypothetical speech of an insane “person” in Horry’s question to the doctor, Mr Drewry (8a above) in the prosecution case becomes real in his/the mother’s voicing of the prisoner’s speech. Although “the resources of his [Cooper’s] relatives […] were unfortunately very limited” (The Morning Chronicle 18 June 1842), the mother’s evidence makes real and vivid the hypothetical speech of an insane person, used in Horry’s cross-examination of Drewry, and thereby presents a compelling picture of the prisoner as having periods of mania. (13)

Isabella Cooper – Indirect speech quotation a. frequently after I have been to bed, I have been disturbed by the prisoner’s getting up, and walking about the room – he has many times said that there has been something lifting his bed up and down, and he could not sleep in it, and I have frequently made him up a bed in my back parlour where I sleep now b. he has often jumped up, and said he was King Richard, and Dick Turpin, when he was sitting in a chair – at that time he had no weapons – some times he had a pistol in his hand – that has happened more than once – he would often say he was King Richard when he had nothing in his hand

(14)

NRSA a. he has threatened me with violence when I have put him out, when I made him angry, when I wished to have any control over him b. he has always expressed an indifference to life, always wishing to make away with himself – he was always tired of his life, from being quite a boy – he has said so many times, more than 100 times

The vividness of the re-enacted speech in the defence case, coupled with a defence opening speech “of nearly two hours’ duration” (The Morning Chronicle 18 June 1842) makes the defendant speak in his defence in the trial, though he is not invited to speak until the verdict has been delivered and he is “asked by the clerk of the arraigns […] what he had to say why sentence of death should not be passed upon him according to law” (The Morning Chronicle 18 June 1842). After the defence witness evidence, Horry makes a closing speech (again not reported in the OBP, but in the news), which is reported using indirect speech and NRSA forms (15).

Quoting the prisoner   

(15)

   395

He compared the case set up by him for the defence with that endeavoured to be set up by the prosecution. All that had been proved was, that instead of being quiet he had been outrageous; there was not a circumstance that was inconsistent with the idea of insanity. The question was not whether he was insane after the act, but at the very moment the murder was committed. With regard to the means of knowledge of the medical men, they were deficient in the point of examining those who had been in the habit of knowing the man for years” (The York Herald, and General Advertiser, Saturday June 25 1842).

This comparative method, used by Horry in (15) has also been our approach here. Comparing what each prisoner is quoted as saying within the prosecution case with what he says, through his advocate, in his defence, we see two very different views: one rational, evil, murderous, defiant, threatening and the other manic, insane, ill, suicidal, incapacitated, feeble. At the end of each case, medical experts are brought in from the gaol, to add their observations on the prisoner’s behaviour there, while awaiting trial. In Cooper’s case the medical expert, Dr Fisher, surgeon at Newgate gaol (Eigen 1995: 195), contradicts the defence witnesses. Here we see a different kind of speech representation. Even though he has had a lengthy interview with the defendant, the speech reporting is metadiscursive (Hyland 2005): conversation(s), questions, answered, using the metalanguage of quotation, rather than quotation itself. After one “hour’s conversation” between Fisher and Cooper, Fisher “consider[s] [him] self capable of pronouncing an opinion of the prisoner’s sanity” (testimony of Fisher). Isabella Cooper’s compelling evidence of her son’s lifelong behaviour, made vivid by dense quotation of his speech (in nearly 10 % of her testimony), is contrasted with Fisher’s suppression of quotation through metadiscursive speech report (conversation, the conversation, the hour’s conversation). Cooper is found guilty and sentenced to death, the insanity defence failing due to Drewry and Fisher’s strong opinions of sanity. In McNaughten’s case, after the five medical experts for the prosecution and defence have agreed, a further doctor adds to the medical evidence. The record shows that Forbes Winslow (who went on to own two private asylums and was perhaps in court researching for his 1843 book The Plea of Insanity in Criminal Cases (Andrews 2004 online)), was present “during the two days of this trial”. He stands up and gives an opinion on McNaughten’s “insanity” though he “was not summoned on either side”: I have not the least doubt of the existence of the prisoner’s insanity, and that the act with which he is charged was intimately connected with the delusion of mind under which he appears to have been labouring for some considerable length of time.

396   

   Alison Johnson

Advocacy and medical testimony can work together or in opposition, using quotation, to present alternative perspectives and different versions of the story from the defendant’s perspective. Hypothetical quotation by Mr Horry with the prosecution doctor in the Cooper trial anticipates the insanity defence. In relation to police witnesses, quotation is vividly deployed by the prosecution to portray the self-incrimination in Cooper’s words and McNaughten’s conversation with Inspector Tierney, but in cross-examination in the McNaughten trial metatalk is used for excuses and lack of recall about what was said, undermining the prosecution case.

5 Conclusion The paper offers a number of conclusions relating to methodological and analytical concerns. A large and complex “specialized corpus” (Flowerdew 2004) of professional discourse, such as the OBP requires informed and discipline-specific thinking in order to extract and develop sub-corpora of use to discourse analysts. There are almost unlimited possibilities, and this paper offers a discussion of what can be done with a small sub-corpus of the whole Proceedings and two specific trials, while suggesting that there is a wealth to explore. In addition to the modus operandi for doing socio-legal discourse studies within the OBP that is illustrated, there are a number of benefits revealed. Through extraction of small and specialized corpora from the larger whole, the linguistic researcher is able to produce new socio-legal and socio-historical linguistic knowledge. The quotation phenomenon can be seen in the context of practices of text-production (Fairclough 2001: 20), and, in doing so, we can see that the record selects this vivid representation mode, capable of speaking for and through the prisoner. The fact that the record includes such rich verbatim and indirect reported speech gives us unique access to the words and worlds of speakers in trials across time. We see that what, who, and where words are quoted are matters for record, and the result is the production of a powerful record, suggesting that the trials embodied evidentially powerful testimony for the prosecution and defence and for the jury’s decision-making. In particular, at this point in legal history when defence advocacy was just beginning and when expert medical witnesses first entered the courtroom to testify to defendants’ insanity, these combined forces acutely affected the tone of criminal trials. Advocacy and medical voices worked in concert to speak for the insane defendant in words more eloquent and powerful than they could manage themselves, altering juries’ perceptions of the crime and creating a richly textured trial, with prosecu-

Quoting the prisoner   

   397

tion and defence voices intertwined in examination and cross-examination. For the first time in history, following the introduction of the Prisoners’ Defence Act (1836), the defence voice was more equally and fairly matched with the prosecution, creating the conditions for change. Defence strategies of quoting the defendant’s voice, rather than getting him to speak became key. Quotation is used in the record to demonstrate the defendant’s criminal liability or his incapacity to commit a criminal offence and the establishment of a binding legal reality crucially hinges on the differential weights of prosecution and defence evidence. We have seen that prosecution and defence cases select different words from the same person and the same story, making quotation a highly selective and powerful advocacy resource. We have seen particularly how vivid direct quotation is contrasted with suppression of speech to favour the prosecution case in McNaughten, and then how the defence uses the power of indirect speech in the mouth of the medical expert to speak more eloquently for the defendant than he can for himself. In the Cooper case we have seen how the density of quoted speech in the testimony of Cooper’s mother represents his whole life as one of insane utterances, but, even so, this lay witness does not have the power of medical witnesses to convince the jury. The analysis shows the power of hypothetical quotation and establishes a new category that operates alongside quotation: metatalk for suppressing quotation. The “haunting evidence” (Eigen 1995: 160) of prisoner testimony is given voice through the powerful alliance of defence advocate and witness (in Cooper’s trial through his advocate and the prisoner’s mother and in McNaughten’s through the advocate and the medical experts), as they tell the court, through the prisoner’s words, what he is incapable of putting together himself. The prisoner is also prosecuted through his own words. What he says on arrest, in gaol, and before the magistrate, is recorded and replayed to self-incriminate, as he is ventriloquized by the prosecution, and it is therefore a masterful adversarial reversal to defend with the same voice, but from a different perspective. These trials suggest that the Victorian policeman, medic and barrister knew a great deal about the power of quotation, and, whether it was effective in persuading the jury or not, the advocate’s skill in eliciting these voices of the insane gives us a unique insight into the Victorian courtroom’s interest in medicine and the gradual development of what we now call (forensic) psychiatry. The voices of Cooper, who was haunted by Dick Turpin, and McNaughten, who was “haunted by devils” (testimony of Patterson), are replayed to the jury for evaluative work, by the new cast of characters in the courtroom, particularly the defence advocate and the medical witness. The crucible of change that was the early 19th century courtroom is seen in these residues of talk from the Old Bailey, thus providing lasting testimony, from the dead, to the lives of the Victorian poor.

398   

   Alison Johnson

References Andrews, Jonathan. 2004. “Forbes Benignus Winslow (1810 – 1874)”. Oxford Dictionary of National Biography. Oxford: Oxford University Press. http://0-www.oxforddnb.com./ article/29752, last accessed 27 Nov 2012. Archer, Dawn. 2013. “Tracing crime narratives in the Palmer trial (1856)”, in: John Conley, Chris Heffer and Frances Rock (eds). Legal-Lay Communication: Textual Travels in the Law. Oxford: Oxford University Press, 168–186. Archer, Dawn. 2010. “The historical courtroom. A diachronic investigation of English courtroom practice”, in: Malcolm Coulthard and Alison Johnson (eds). The Routledge Handbook of Forensic Linguistics. Abingdon: Routledge, 185–198. Archer, Dawn. 2005. Historical Sociopragmatics: Questions and Answers in the English Courtroom (1640–1760). Amsterdam: John Benjamins. Bell, Allan. 1991. The Language of News Media. Oxford: Blackwell. Bewley, Thomas. 2008. “Daniel McNaughten (the McNaughten Rules)”. Madness to Mental Illness. A History of the Royal College of Psychiatrists. Online archive 19. http://www. rcpsych.ac.uk/, last accessed 22 Nov 2012. Brendel, Elke, Jörg Meibauer and Markus Steinbach (eds). 2011. Understanding Quotation. Berlin: de Gruyter. Bublitz, Wolfram and Christian Hoffmann. 2011. “‘Three Men Using our Toilet all Day Without Flushing – This May Be One of the Worst Sentences I’ve Ever Read’: Quoting in CMC”, in: Joachim Frenk and Lena Steveker (eds). Anglistentag Saarbrücken 2010. Proceedings. Trier: Wissenschaftlicher Verlag Trier, 433–447. Busby, Siân. 2009. McNaughten. A Novel. London: Short Books. Cairns, David J.A. 1998. Advocacy and the Making of the Adversarial Criminal Trial 1800–1865. Oxford: Clarendon Press. Cameron, Lynne and Alice Deignan. 2003. “Combining large and small corpora to investigate tuning devices around metaphor in spoken discourse”. Metaphor and Symbol 18/3, 149–160. Clift, Rebecca and Elizabeth Holt. 2007. “Introduction”, in: Elizabeth Holt and Rebecca Clift (eds). Reported Talk. Reported Speech in Interaction. Cambridge: Cambridge University Press, 1–15. Culpeper, Jonathan and Merja Kytö. 2000. “Data in historical pragmatics: spoken interaction (re)cast as writing”. Journal of Historical Pragmatics ½, 175–199. Eigen, Joel P. 1995. Witnessing Insanity. Madness and Mad-Doctors in the English Court. New Haven: Yale University Press. Esquirol, Jean Étienne. 1838 [trans. E.K.Hunt 1845] Mental Maladies: A Treatise on Insanity. Extract: “Monomania.”, in: Jenny Bourne Taylor and Sally Shuttleworth (eds). 1998. Embodied Selves. An Anthology of Psychological Texts 1830–1890. Oxford: Oxford University Press, 256–262. Fairclough, Norman. 2001. Language and Power. Vol. 2. Harlow: Pearson Education Limited. Flowerdew, Lynne. 2004. “The argument for using English specialized corpora to understand academic and professional language”, in: Ulla Connor and Thomas Albin Upton (eds). Discourse in the Professions. Amsterdam: John Benjamins, 11–33.

Quoting the prisoner   

   399

Foster, Joseph. 1889. The Register of Admissions to Gray’s Inn, 1521–1889, Together with the Register of Marriages in Gray’s Inn Chapel, 1695–1754. London: The Hansard Publishing Union Limited. http://www.archive.org, last accessed 15 Jan 2013. Harper, Charles G. 1908. Half-hours with the Highwaymen; Picturesque Biographies and Traditions of the “Knights of the Road”. Vol. 1. London: Chapman and Hall. http://www. archive.org/details/halfhourwithhig01harpiala, last accessed 15 Jan 2013. Heffer, Chris. 2005. The Language of Jury Trial: A Corpus-Aided Analysis of Legal-Lay Discourse. Basingstoke: Palgrave Macmillan. Hitchcock, Tim, Robert Shoemaker, Clive Emsley, Sharon Howard and Jamie McLaughlin. 2012. The Old Bailey Proceedings Online, 1674–191. www.oldbaileyonline.org version 7.0, 24 March 2012, last accessed 14 Jan 2013. Holt, Elizabeth. 2007. “‘I’m eyeing your chop up mind’: reporting and enacting”, in: Elizabeth Holt and Rebecca Clift (eds). Reporting Talk. Reported Speech in Interaction. Cambridge: Cambridge University Press, 47–80. Hostettler, John. 2009. A History of Criminal Justice in England and Wales. Hook: Waterside Press. Huddleston, Rodney and Geoffrey K. Pullum. 2002. The Cambridge Grammar of English. Cambridge: Cambridge University Press. Hunston, Susan. 2011. Doing Analysis in Discourse and Corpus: The Case of Evaluative Language. Plenary paper presented at Corpus Linguistics 2011. Birmingham, 20–22. Hyland, Ken. 2005. Metadiscourse. London: Continuum. Jakobson, Roman. 1972. “Verbal communication”. Scientific American 227, 72–80. Johnson, Alison J. 2013. “Embedding police interviews in the prosecution case in the Shipman trial”, in: John Conley, Chris Heffer and Frances Rock (eds). Legal-Lay Communication: Textual Travels in the Law. Oxford: Oxford University Press, 147–167. Leech, Geoffrey and Mick Short. 2007. Style in Fiction: A Linguistic Introduction to English Fictional Prose. Vol. 2. Harlow: Pearson Education. Lobban, Michael. 2004. Cockburn, Sir Alexander James Edmund, twelfth baronet (1802–1880). Oxford Dictionary of National Biography. Oxford: Oxford University Press. http://www. oxforddnb.com, last accessed 15 Jan 2013. Matoesian, Greg. 2000. “Intertexual authority in reported speech: production media in the Kennedy Smith rape trial”. Journal of Pragmatics 32, 879–914. Old Bailey Proceedings Online. January 1907. “Trial of Frederick George Church” (t19070128–29). www.oldbaileyonline.org version 7.0, 24 March 2012, last accessed 14 Jan 2013. Old Bailey Proceedings Online. May 1898. “Trial of John William Sando” (t18980516–396). www.oldbaileyonline.org version 7.0, 24 March 2012, last accessed 14 Jan 2013. Old Bailey Proceedings Online. February 1843. “Trial of Daniel M’Naughten” (t18430227–874). www.oldbaileyonline.org version 7.0, 24 March 2012, last accessed 14 Jan 2013. Old Bailey Proceedings Online. June 1842. “Trial of Thomas Cooper” (t18420613–1766). www. oldbaileyonline.org version 7.0, 24 March 2012, last accessed 14 Jan 2013. Oxford English Dictionary Online. 2012. John Simpson (ed.). www.oed.com, last accessed 6 Jan 2013. Ramage, John D. and John C. Bean. 1998. Writing Arguments. 4th edn. Needham, MA: Allyn and Bacon.

400   

   Alison Johnson

Reisigl, Martin and Ruth Wodak. 2009. “The discourse-historical approach”, in: Ruth Wodak and Michael Meyer (eds). Methods of Critical Discourse Analysis, Vol. 2. London: Sage, 87–121. Scott, Mike. 2011. Wordsmith Tools, Version 5. Oxford: Oxford University Press. Seal, Graham. 2011. Outlaw Heroes in Myth and History. London: Anthem Press. Tannen, Deborah. 2007. Talking Voices: Repetition, Dialogue, and Imagery in Conversational Discourse. Vol. 2. Cambridge: Cambridge University Press. Taylor, Jenny Bourne and Sally Shuttleworth (eds). 1998. Embodied Selves. An Anthology of Psychological Texts 1830–1890. Oxford: Oxford University Press. Tiersma, Peter. 1999. Legal Language. Chicago: The University of Chicago Press. Torrey, E. Fuller and Judy Miller. 2001. The Invisible Plague. The Rise of Mental Illness from 1750 to the Present. New Brunswick: Rutgers University Press. Vagrancy Act 1824, The. www.legislation.gov.uk/ukpga/Geo4/5/83/section/4, last accessed 27 Jan 2013.

19th Century British Library Newspapers via Gale Online The Morning Chronicle (London, England), Saturday 18 June 1842 The Standard (London, England), Friday May 6 1842. The Essex Standard, and General Advertiser for the Eastern Counties (Colchester, England), Friday May 13 1842. John Bull (London, England), Monday February 13 1843 Leeds Times (Leeds, England), Saturday 25 June 1843 The Morning Chronicle (London, England), Saturday 18 June 1842 The York Herald, and General Advertiser (York, England), Saturday June 25 1842

Bettina Lindner

16 Quotations from 17th and 18th Century Medical Case Reports¹ Abstract: As recent research shows, 29 % of quotations featured in modern surgical publications are inaccurately quoted, and it is estimated that misquotation in medical literature amounts to 17 % of all quotes (Regier 2010). This is remarkable since quotation is one of the characteristic and often defining features of scientific texts (Jakobs 1999, Gläser 1990). However, to the best of this author’s knowledge, the historical development of quoting in medical contexts has not been investigated at the time of writing, either in linguistics or in the philosophy of language. This diachronic study describes some aspects of the development of quotation forms and functions in Early Modern German scientific medical literature and investigates to what extent modern expectations on quotation can be applied to historical scientific medical literature. The focus is on medical case studies which came into use towards the end of the 17th century. The results are heterogeneous: different forms of quotation were used and the way in which they were reproduced varied, but all in all most medical authors preferred reported speech. It is only when physicians quoted colleagues that varying forms of quotation were utilized, direct speech and reported speech. In general, quotation was not as common in the 17th century as it was in the 18th century, nor was it an obligatory feature of case reports. It may become evident through the discussion of the data that quotation customs only begin to meet modern expectations by the end of the first half of the 18th century, as standard conventions of scientific presentation were not well established prior to this. Keywords: medico-legal history, medical literature, case reports, quoting authorities

1 Introduction Quotation is undoubtedly one of the characteristic features of contemporary scientific texts (Jakobs 1999, Gläser 1990). Direct, indirect or mixed quotes, as well as pure references, are used to refer to the research of the scientific community.

1 From conception to redaction, this essay greatly benefited from the advice and support of Peter Brayshay, Christine Ganslmayer and Anita Lasch, who helped me most generously with their useful comments.

402   

   Bettina Lindner

However, the historical development of quoting in medical contexts has not yet been investigated, either in linguistics or in the philosophy of language. In the present paper, I present some results from ongoing research on medical case reports from 1600 to 1800. I will endeavour to determine the extent that modern expectations on quotation can be applied to Early Modern German scientific medical literature. Therefore, I will focus on the following questions: 1. Which forms of quoting were used? 2. Who was quoted? 3. What functions did the quotations of Early Modern medical scientists have? 4. Was quotation an obligatory feature of medical case reports?

2 Corpus The research is based on a corpus of German medical case reports dating from 1600 to 1800. Consequently, the focus is on German compilations of medical case reports, which came into use towards the end of the 17th century. The corpus contains 97 medical case reports which are taken from printed, edited compilations. Because of the Thirty Years’ War it is not easy to find German reports from the 17th century and this is why only 33 of the 97 investigated reports are from this period. Since many of these reports were lost in the war during the first half of that century, the earlier texts date back to the time from 1650 to 1700. Case reports have played a leading role in medicine since antiquity (e.g. “Epidemics” by Hippocrates), as practical application has always been a relevant aspect of this discipline (Stolberg 2007: 81). Since the end of the 16th century, many physicians – especially famous authorities such as Wilhelm Fabry or later on Georg Ernst Stahl – have published examples from their own experience either in compilations or, increasingly since the 17th century, in periodicals (Stolberg 2007: 81). Crawford demonstrated that this development was typical of continental Europe, whereas, due to the differing status of medical experts in trial procedures, Britain lagged behind by nearly two centuries in developing a science of forensic medicine (Crawford 1994). Since the 1980s, several studies have investigated compilations of case reports and their importance for developments in forensic medicine, mainly from a medico-historical point of view: Esther Fischer-Homberger (1983) published a history of legal medicine based on medico-legal case report compilations. Johanna Geyer-Kordesch (1990) highlighted the importance of case reports for the Early Modern reform of knowledge in medicine. Maren Lorenz (1999) analysed case reports with respect to both bodily experience and bodily perception.

Quotations from 17th and 18th Century Medical Case Reports   

   403

The forms and functions of different types of case reports were investigated by Michael Stolberg (2007). Through an analysis of visa reperta,² Müller/Fangerau (2010) showed that, due to their special structure and mode of representation, visa reperta not only shaped the practice of forensic medicine, but also of standardised examination and documentation in pathological anatomy. The conference transcript “Der ärztliche Fallbericht”, edited by Rudolf Behrens and Carsten Zelle (2012), sheds light on different aspects of these texts, mainly from a literary, cultural and medico-historical scientific point of view. These compilations of case reports were meant for a variety of readers. Many of them were written for members of the courts, judges or advocates for instance, as a collection of precedents. Most of them were intended for medical education at universities (Stolberg 2007). A small percentage of these reports were published guidelines for a larger audience giving advice on how to prevent infection and disease. The collections contain different subcategories of reports. Some are so-called Consilia, which either advised a single patient on how to cure his/her ailment or were directed at a larger audience to prevent or to counter epidemics. Most of the investigated reports are legal reports which were written by medical experts such as physicians, surgeons or midwives. These were often requested by judges to shed light on violent felonies. Topics typical of these latter texts include infanticide, lethality, epidemics, insanity, and suicide.

3 Which forms of quotation were used? Six kinds of quotation are typically discussed in linguistic literature: direct quotations, indirect quotations, mixed quotations, pure quotations, scare quotations and emphatic quotations (see Brendel/Meibauer/Steinbach 2011: 2–7). The latter three do not play a role in the medical case reports of the 17th and 18th centuries and will not be discussed here. Direct quotation is regarded as the prototypical form of quotation (Brendel/Meibauer/Steinbach 2011: 2). In modern texts, the use of colons and quotation marks signal that the utterance is to be read as a direct quote. The utterance is usually reproduced ad verbatim (but see Brendel/ Meibauer/Steinbach 2011: 12–14 for exceptions). As direct quotation is not the most commonly used form of reporting another speaker’s utterance in modern scientific texts, it is no surprise that we find only very few marked direct quotations in the case reports from between 1600 and 1800. This might be due to the

2 Visum repertum is a special form of Early Modern medical case report.

404   

   Bettina Lindner

literary customs of the medieval and Early Modern period: verbatim reports were not a common procedure and quotation marks and punctuation in general had not yet been subject to standardisation and convention. The example for quotation (1) below is taken from a Southern German compilation of medical case reports that was intended for judges. The report is from 1736 and the members of the local so called Collegium Medicum were asked to discuss the effect of a special medicine’s ingredient, namely Coloqinten, which a young girl was accused of having taken to abort her baby. (1)

Fernelius in Methodi medendi Lib. III. Cap. IV. p.m. 96. ſetʒet ſie unter die purgantia maligna & venenoſa; und der alte Herr Doctor Wedel in ſeinen Amoenitatibus materiæ medicæ pag. 249. nennet ſie ein infame medicamentum, und ſchreibet demſelben die Würkung einer raſionis inteſtinorum torminum immanium & prolectionis ſanguinis ʒu; ja er erʒehlet aus dem Tulpio, daß jemand in grauſames Reiſſen und Blutſturʒ und ʒugleich in die gröſte Lebensgefahr gerathen, als er von einem Trank, in welchem nur 3. Coloquinten Aepfel (die doch kaum 2. Loth, geſchweige dann 8. Loth ausmachten) gekochet worden, getrunken. (Hasenest I 1755: 4) (In Fernelius’ Methodi medendi Lib. III. Cap. IV … they are classified as poisonous and malignant emetics; and old Doctor Wedel calls them an infamous medicine in his Amoenitatibus materiae medicae …, and ascribes to them the effects of strong intestinal pain and irritation of blood; he relates an episode from Tulpio, in which someone was in agony and nearly died after drinking a potion which was prepared with only three of the “Coloquinten” apples ….)

The physicians catalogued and compiled the different professional opinions presented in the contemporary medical literature and evidently employed modern methodologies: they used indirect speech and gave precise references for the sources of their information. In (2), we find one of the few examples of marked direct quoting in a case report from 1724. It is taken from a compilation by Troppanneger, published in 1733. The faculty of medicine in Halle was asked to produce a statement to settle a dispute between two physicians about the appropriate treatment for a patient’s ailment. For this purpose, the scientists from Halle directly quoted Johann Bohn, a famous coroner from Leipzig. As in modern direct quoting, the colon signals a change in the text and also that the subsequent utterance is taken from another context.

Quotations from 17th and 18th Century Medical Case Reports   

   405

Original utterance

Quote

Ordinarie equidem (4) jejunis vomitoria exhibentur, illis tamen, qui difficulter & cum ſingulari moleſtia hinc moventur poſt paſtum paulo propinari valent, pro evacuatione facilitanda.

Bohnius de off. med. dupl. Part.I. Cap. 14. de Purgatione. hiervon ſaget, da dieſer das letʒtere Tempus matutinum ad ordinarium modum propinandi emetica rechnet; allwo er ſaget:

Bohn (1704: 279)

Ordinarie Jejunis vomitoria exhibentur, illis tamen qui difficulter & cum ſingulari moleſtia hinc moventur, poſt paſtum paulo propinari valent, pro evacuatione facilitanda. (Troppanneger 1733: 68) (Bohnius in De off.med. dupl. Part. I. Cap. 14. on emetics, who considers morning time as the norm for taking emetics, says:

(Normally, emetics are prescribed to be taken on an empty stomach; those patients, who cannot be convinced/persuaded to do so, should ingest emetics shortly after a meal to facilitate emisis.)

(Normally, emetics are prescribed to be taken on an empty stomach; those patients, who cannot be convinced/persuaded to do so, should ingest emetics shortly after a meal to facilitate emisis.)

(2) Example of direct quotation in a medical case report Quotation marks were not used, but the quote was set from Gothic to Roman type and indentation was used to set off the quote which, according to Weyers (1992: 19), is the established and most common form of highlighting in bilingual editions from between 1600 and 1800. Quoting Latin utterances in this way was commonplace in medical case reports from 1600 to 1800, as Latin remained the lingua franca in science and at universities. At the time many of the reports were written entirely in Latin, and even in German reports, Latin words and phrases were frequently incorporated – especially for medical terms. As we have already seen in example (1), physicians of the 17th and 18th century also used indirect quotation. This form of quotation is very frequent in the corpus and especially popular with authors from the 17th century. In fact, Modern High German has developed specific grammatical markers to signal that certain content constitutes indirect quotation. According to Brendel/Meibauer/ Steinbach (2011: 11) and Duden (2009: 523–543), German uses complementizers like ʻdassʼ (that), verb position, and the subjunctive in indirect quotation. These grammatical markers can be utilised in different syntactical contexts. With this in mind, let us return to example (1). The authors gave detailed information about

406   

   Bettina Lindner

the sources of the original utterances, i.e. Fernelius’ “Methodi medendi” and Dr. Wedel’s “Amoenitates materiae medicae”. They used verba dicendi to indicate indirect speech: nennet, schreibet zu, and erzählet. Since the finite parts of the verb complex were left out, it is not possible to say if the indicative or the subjunctive was used. Furthermore, we find the complementizer daß. However, there are also a lot of ambiguous cases – especially when experts were mentioned – as seen in the following example (3) from the 17th century: (3)

Ob dieſe Wunde abſolute & perſe oder per accidens tödtlich geweſen ſey? melden wollen / wie ich nehmlich nicht davor halte / daß man nach angeführten Umſtänden das erſte mit beſtand mainteniren könne / nam quod vulnus in aliis aliquando curatum eſt, id pro ſimpliciter lethali haberi nequit, ſed illud demum, quod neceſſario mortem vulnerato infert, nec ulla arte ſanari poteſt. Vid. Sennert. Lib. 5.p.4.c.3. (Fischer 1704: 272) (Regarding the question of whether this wound is absolutely and per se or just accidentally lethal, the former possibility is not clear, taking the individual circumstances into account, because “as long as the lesion is somehow healed eventually, it is not presumed to be lethal. This can only be applied [to wounds], which effect death inevitably and could not be cured by any skill.” See Sennert. Lib. 5.p.4.c.3.)

This is a Latin statement, set in Roman letters, but it is difficult to determine whether it is an indirect quotation from the source the authors mentioned below, a direct quotation, or if it only gives the example of another coroner who was of the same opinion. Typographically, the statement appears the same as the direct quotation from 1724 discussed above (2). Regarding the 17th century, Schulz (2012) shows that a change of type was frequently used to indicate foreign, “zitathafte” (similar to quotation) parts, as well as to set off the names of non-indigenous people and places in texts. This highlighting of foreign elements had increased in different types of printed texts from the second half of the 17th century. According to Schulz, this development was caused by an increased sensibility to foreign elements assimilated into the vernacular. Further investigation of the cited source in (3) reveals that it is not a direct quote but in fact that it summarizes important information from the Latin chapter (see above). Readers are expected to pay attention to the indication of the source and to remember the structure of Sennert’s works: the abbreviation “Vid. Sennert. Lib. 5.p.4.c.3.” indicates that the passage above was taken from one of his more extensive works, either “Institutiones medicae” or “Practica medicina”. Both books have five subdivisions, so called “liber”, (but only liber five of “Practica

Quotations from 17th and 18th Century Medical Case Reports   

   407

medicina” has four parts). It is quite obvious that Fischer wrote for an audience who was familiar with the cited works. Precise details about the sources were not necessary as long as most of the readers were assumed to know the mentioned texts. At the time, standard forms of scientific quotation were neither well established nor adhered to, and distinct, recognisable ways of highlighting the various kinds of quotation had not yet been developed. In order to reach a definite conclusion, every single indication of source needs to be critically examined. At this point some general remarks about the graphical characteristics of quotations in case reports are in order. As can be seen in (4), changing the type from Gothic to Roman and italicizing the letters marked the indication of source. It was the general method of highlighting a passage in the text. (4)

Source: Troppanneger (1733: 298) Sometimes indentation was used both to highlight the quote and to indicate the source. Moreover, abbreviations are very typical for quotes from that time. Each part of the indication of the source could be shortened, such as, for example, the author, especially when he was well known, like the ancient authors, or when he had previously been mentioned in the text. In the latter case, a version of L.C. would be added, which stands for the Latin loco citato. Alternatively, the acronym for the title of the work would be given, as in (5): (5)

[…] ja wenn auch beyde humores heraus, und die Læſion durch das Auge ſo beſchaffen, daß ſelbige hinten im Nacken wieder heraus gehet, von ſelbigen aber keine groſſen vaſa lædiret worden ſind, der Menſch doch bey dem Leben bleibet, dergleichen Exempel ebenfalls die M.N.C.D.I.A.3.Obſ. 283. geben. (Troppanneger 1733: 33) ([This periodical] M.N.C.D.I.A.3.Obſ. 283. also gives examples [of cases] in which the patients survive, even though […] both humours leak out, and the lesion of the eye goes through to the neck, but none of the larger vessels are injured.)

This habit of abbreviating is probably due to the long titles which were fashionable in the baroque period, such as “Miscellanea Curiosa Medico-Physica Academiae Naturae Curiosorum sive Ephemeridum Medico-Physicarum Germanicarum Curiosarum Annus Primus Anni scilicet M.DC.LXXmi: continens Celeberrimo-

408   

   Bettina Lindner

rum Medicorum in & extra Germaniam Observationes Medicas & Physicas, vel Anatomicas, vel Botanicas, vel Pathologicas, vel Chirurgicas, vel Therapeuticas, vel Chymicas; Praefixa Epistola Invitatoria ad Celeberrimos Medicos Europae”, which is the extended and rather unwieldy title of the acronym just cited. Most collections of medico-legal cases are divided into so called Centuriae (100 cases), also abbreviated in (4). In a very few cases, the indication of the source was shortened in such an extreme way that it is not possible to identify the cited work. The third kind of quotation, the mixed quotation, is in general very rare. In linguistic literature, mixed quotation is defined as a combination of direct and indirect quotation, which means that an indirect quotation contains a directly quoted segment as part of the indirectly reported utterance. Thus, a segment is conveyed verbatim and constitutes a proper part of the utterance (see Brendel/ Meibauer/Steinbach 2011: 3–4). In my corpus there are no examples which fully comply with the characteristics of mixed quotation, but there are some which are mixed in another interesting way. In the case report from 1737 (6), physicians discussed the issue of whether the pharmaceutical product that a woman had taken could cause an abortion. For this purpose, they quoted two medical case reports from the faculty of medicine in Halle which had been published in Friedrich Hofmann’s “Medicina Consultatoria” and which discussed similar questions. (6)

Wie dann auch eben dieſe Negativa in ʒweyen von der mediciniſchen Facultät ʒu Halle in ſuſpicione tentatæ provocationis abortus, geſtellten und in Fr. Hoffmanni Medicinæ Conſultatoriæ 5. p. 211 und 6. Theil p. 281 ſeqq. befindlichen Reſponſis behauptet und daſelbſt angemerket wird, „daß es an und vor ſich ſelbſt in der „Natur keine dergleichen allgemeine und ausgemachte Mittel gebe, da „durch man ʒuverläßig, bey allen und jeden, einen Abortum procuriren, „oder die Frucht abtreiben könne, maſſen ſonſt ſolches weit mehr practici „ret und wohl wenige oder gar keine uneheliche Kinder ʒur Welt geboh= „ren werden würden. Sondern daß ſolches nicht eher angehe, als nur in „dem Fall, wenn eine gewiſſe Diſpoſition ʒu dem Abortu, entweder „in Vtero ſelbſt, oder auch ſonſt bey der ſchwangeren Perſohn, vorhanden „iſt rc. (Hasenest I 1755: 7) (These disadvantages were claimed in two reports on alleged abortions from the faculty of medicine in Halle, published in Friedrich Hofmann’s Medicina consultatoria …: “There is no such general medicine in nature which could reliably cause an abortion – otherwise illegitimate children

Quotations from 17th and 18th Century Medical Case Reports   

   409

would no longer be born. Rather there has to be a predisposition to an abortion in the uterus or in the pregnant women herself.”) The quoted passage is marked by quotation marks and verba dicendi, and partly subjunctive partly indicative forms were used. We find the complementizer “daß” and accordingly the finite verbs in the last position. According to Weyers, the form of quotation marks found in this quotation tallies with the instruction given in Hieronymus Freyers “Anweisung zur Teutschen Orthographie”, first published in 1721. Freyer recommended the use of quotation marks at the beginning of each line (Weyers 1992: 19). In examining the source of the passage, we find that it is mixed in another way as well: the first part of the directly quoted text (“daß … gebohren werden würden”) comes from Casus VI of the sixth part of Friedrich Hofmann’s “Medicina Consultatoria”, page 212 (7), the second part (“Sondern … iſt”) comes from Casus IV of the fifth part of Hofmann’s “Medicina Consultatoria” (page 282) (8). (7)

daß es an und vor ſich ſelbſt in der Natur keine dergleichen allgemeine und ausgemachte Mittel gebe, da durch man ʒuverläßig, bey allen und jeden, einen Abortum procuriren, oder die Frucht abtreiben könne, maſſen ſonſt ſolches weit mehr practiciret und wohl wenige oder gar keine uneheliche Kinder ʒur Welt gebohren werden würden. (Hoffmann VI 1728: 212.)

(8)

[…], ſondern daß ſolches nicht eher angehe, als nur in dem Fall, wenn eine gewiſſe Diſpoſition ʒu dem Abortu, entweder in Vtero ſelbſt, oder auch ſonſt bey der ſchwangeren Perſohn, vorhanden iſt rc. (Hoffmann V 1726: 282.)

The example in (7) is quoted literally, whereas in the second part of (6) a new sentence starts which is part of a sentence in the original text (8). Furthermore, the authors deleted a short Latin passage that is in the original, “abortu und præmatura expulſione fœtus”. Thus the authors of this case report combined two different statements and merged them into one. To my knowledge this form of quotation has not yet been described in the linguistic literature. A systematic study and meta-analysis of historical quotes would resolve the issue of whether this example is an exception or whether this was an established form of historical quoting. Incidentally the source citation is not correct: the authors confused the parts of Hofmann’s “Medicina Consultatoria”. They quoted page 282 from part 5 and page 212 from part 6. This is worthy of mention because research has shown that 29 % of quotations in today’s surgical publications are misquoted and it is esti-

410   

   Bettina Lindner

mated that misquotation in modern medical literature amounts altogether to 17 % (Regier 2010: 17).

4 Who was quoted in medical case reports? This leads us to the next question: Who was quoted in medical case reports? I am especially interested in whether the demographic profile and prestige of the quoted individuals had any influence upon the form of quotation which could be comparable to the results of several pragmatic historical studies on politeness strategies since the 1960s. Is there a hierarchy of form, for instance, such that ancient authorities were mostly quoted in direct speech and non-academic persons in an indirect manner? There were in fact four groups of quotable individuals: patients and witnesses, non-academic medical practitioners, academic physicians and ancient authorities. So far, it seems that there is no difference in style between the quoting of patients, witnesses and non-academic practitioners. They are all mainly quoted using indirect speech. Depending on their role in the case, however, the speech indicating verbs differ. In the following consilia, for instance, the individuals quoted were the commissioners of the report and therefore the reported speech is introduced with rather demanding expressions: (9)

Dahero ſie darüber und inſonderheit über folgende Fragen unſer collegialiſches Judicium cum allegatis Doctorum teſtimoniis verlanget [‘request‘]: (Alberti 1721: 9.) (They therefore request our collegial advice together with the Doctor’s testimony concerning the following questions:)

(10)

Weil denn nun Herr Patient von mir verlanget ʒu wiſſen I) was der Chirurgus äuſſerlich vor medicamenta adhibiren ſolle? 2) wie etwa die vielflüßige weiße materie durch innerliche medicamenta könte ʒertheilet, und alſo der Wachsthum des wilden Fleiſches gäntʒlich könte gedämpffet werden. (Gohl 1735: 48.) (Because Mr Patient requested to be told 1) what kind of medicine the surgeon should administer externally? 2) how for instance the multi-fluid white matter could be dissected by internal medicine, and how thus the growth of the wild flesh could be constrained totally.)

Quotations from 17th and 18th Century Medical Case Reports   

   411

In the following example the utterance of a prisoner is reported and consequently the verb confess is used: (11)

Ja, Inquiſit hat ſelbſten: Er ſey nur nach dem letʒten Urthel, welches ihm den Tod mitgebracht, in ſolche Alteration geſetʒet worden, daß er darüber gantʒ deſperat und auſſer ſich geworden, ſonderlich, weil ſeine Defenſion ihm nichts geholffen, geſtanden [‘confess‘]. Vid. Regl. Sol. 293. (Troppanneger 1733: 68) (Yes, the inquisit/patient/inmate himself has confessed: After the last sentence, which brought him death, he had a mood swing, that he was beside himself with despair, because his defence did not help him.)

One reason why these people were quoted indirectly could be the fact that legal reports were rarely written up directly after a conversation with a patient/witness, but rather later in the doctor’s office, based on transcripts of interviews or from memory. Consequently, the authors did not use direct speech, but reported instead only the gist of utterances due to the fallibility of cognitive processes. When quoting academic experts however, all forms of quotation were used and the quoted authors were often introduced with positive attributes to establish a sense of authority, as in the following example: (12)

Es iſt ſolches ſchon vorlängſtens im Druck recommendiret, alſo gar nichts neues; inmaßen der berühmte Profeſſor Medicinae in Jena Dr. G.W. Wedelius dergleichen in einem Scripto unter dem Titul: Experimentum Curioſum de Colchico Veneno & alexipharmaco, welches er in einem Manuſcript am Hof gefunden, propaliret […] (Hasenest II 1756: 179) (This was published a long time ago and is not a recent finding; for instance this was shown by the famous professor in Jena, Dr. G.W. Wedel, who published a transcript entitled “Experimentum Curioſum de Colchico Veneno & alexipharmaco”.)

The quoted scientist is described as a famous professor of medicine in Jena and his name is supplemented by the Latin suffix -ius, which humanistic scholars in the 16th century added to personal names to express their affinity with antiquity, the elite and the esoteric. The fourth group of quotable individuals were authorities. In my corpus these were mainly ancient medical scientists like Hippocrates and Galen. However, other ancient authorities, like Plato, were also quoted. The most often quoted ancient authority was certainly Hippocrates. From the 17th to the 18th century, however, the number of quotations from ancient authors declined. This reflects

412   

   Bettina Lindner

a gradual avoidance of antique authorities from the 17th century onward, when innovations in science, and later on the principles of the enlightenment, led to an emphasis on the personal and contemporary observations of scientists, thus diminishing the need to refer to an ancient contextual framework. In the 18th century, quotations were nevertheless often introduced by a positive statement about ancient authorities. On the whole though, it is not possible to identify distinctive forms of quotation which depend on the category of person quoted. Most of the quotations in medical case reports were taken from the contemporary medical literature. Lorenz pointed out that many of the case reports referred to medical discourse through comments, footnotes and references (Lorenz 1999: 38). A closer examination of the indications of source in the reports has revealed that the number of references to medical literature increased between the 17th and the 18th century. As quotation in general was not as common in the 17th century as it was in the 18th century (see chapter 6), only a few references to the literature in reports from the 17th century (20 examples) can be found. Of these, nearly half of the counted items (eight examples) were quotations from Hippocrates’ works, particularly his “Aphorismi”. Daniel Sennert’s works were frequently mentioned as well; four times in fact. The remaining indications of source refer to different physicians’ works from the 16th and 17th centuries, such as Domenico Leoni’s “Ars medendi humanos, particularesque morbos a vertice usque ad pedes” from 1583, or Thomas Willis’ “Pathologiæ cerebri, et nervosi generis specimen in quo agitur de morbis convulsivis, et de scorbuto” from 1667. In the 18th century, it was more common to incorporate contemporary subjective experiential data and opinion into medical discourse. Consequently, we find 143 references to medical literature in the 18th century reports. Most of them refer to literature from the 17th (44 examples) and the 18th century (59 examples). There are also references to literature from the 16th and the 13th centuries and from ancient times. A small percentage of the references could not be identified at all (ten examples) due to the Early Modern physicianly practice of extensively shortening the indication of source, as previously noted. Unlike their 17th century colleagues, authors from the later period surprisingly often consulted medical periodicals such as “Miscellanea Curiosa MedicoPhysica Academiae Naturae Curiosorum […]” (nine examples). This might be due to the fact that physicians increasingly tried to consider more up-to-date findings in their reports. Simultaneously, they also valued the established works of famous coroners such as Paul Zacchias’ “Quaestiones medico legales” from 1557, Paul Amman’s “Medicina Critica” from 1670 and Friedrich Hofmann’s “Medicina Consultatoria” from 1721–1733. Interestingly, although authors mentioned the doctoral theses of young physicians, it was never the candidates themselves who

Quotations from 17th and 18th Century Medical Case Reports   

   413

were named, but instead the examiners at the thesis defence, as in the following example: (13)

Dergleichen ſehr notable Exempel in einer ʒu Wittenberg AO. 1726. unter dem Prœſidio Dr. Lœſchens gehaltenen Diſputation: de Judicio circa abortum concitatum ferendo, viele angeführet werden, und in der ganʒen Diſſertation die bißhero erörterte Negativa, ex ratione, experientia, auctoritate, mit Allegirung gar vieler auctorum, und derer dahin gehörigen Stellen, weitläuftig ausgeführet wird. (Hasenest I 1737: 6/7) (Plenty of such notable examples were mentioned in a thesis defence entitled “de Judicio circa abortum concitatum ferendo“ presided over by Dr. Loesch at Wittenberg university in 1726, where all disadvantages were discussed in detail.)

The doctoral candidate is thus not mentioned by name and instead the person with the higher social rank, in this case the head of department of medicine in Wittenberg, is the one named. In view of the fact that it was an established procedure to pay for one’s degree at Early Modern universities, and as the real author of a thesis was not easily identified (Marti 2001: 13–17), quoting the chairperson was an understandable modus operandi. The modern scientific habit of naming everyone who was involved in a publication, including the supervisor, is probably based on this practice of quoting.

5 What functions did Early Modern medical scientists’ quotations have? Having discussed the relationship between social rank and the form of quoting, we will now look at the typical functions of quoting in medical case reports from 1600 to 1800, which are interesting from a historical point of view. These functions of quoting are: – structuring the medical case reports – pointing out the supremacy of academic medicine – self-praising and self-promoting The first fairly inconspicuous function of quotations is an interesting indicator of the authors’ awareness of their own writing strategies. The following example is taken from one of the most famous compilations of medico-legal reports by

414   

   Bettina Lindner

Michael Alberti, entitled “Systema Jurisprudentiae medicae”. The legal report was once again written by the faculty of medicine in Halle. (14)

I) Ob nicht Impotentia eines Ehemannes ad coeundum per ocularem inſpectionem denen Rechten nach bewieſen werden müſſe? So iſt es dieſes nach Beypflichtung aller Caſuiſten in foro medico bey […] (Alberti 1721: 13) (I) Is it not the case that a husband’s impotence must be proven by examination within a consultation/convention? And therefore according to all medical casuists’ opinion it is the case […])

This use of quotation (14) is very frequent in my corpus of medical case reports. It is evident that most of these physicians were expert writers with a strong background in rhetoric, who closely monitored their own text production. They quoted the client’s question to structure their text by dividing it into sections. The question was used as a topic introduction, which was followed by the physicians’ answer. According to Taavitsainen (1999), this question-answer form is a classical model based on Aristotelian treatises, which embedded theological, medical and scientific problems, which were dealt with in the form of so called quaestiones (Taavitsainen 1999: 245). This form had been intended and used for scientific writing through the ages and was thus very suitable for this text type. It is likely that only a very small percentage of these questions had really been asked by patients and prosecutors, and that the authors had just adapted the original requests to this question-answer form. Since the Middle Ages, academically educated physicians have had to compete with a variety of medical practitioners such as surgeons, apothecaries, midwives, herbalists and homeopaths. Affected by the requirements of public authorities, medical practitioners with any kind of licence were able to gain more and more influence on public health and consequently the conflict among medical professionals in society grew. The medical discourse of the 17th and 18th centuries, and with it its medico-legal reports, was therefore strongly influenced by the competition between academic and non-academic medical practitioners. Thus quoting other academic practitioners and ancient authorities had the function of showing the superior establishment of academic medicine. As Latin was the preferred language of science until the 18th century, academically educated physicians often continued to quote in Latin to distinguish themselves from nonacademic German-speaking medical practitioners. This might be one reason why some authors used medico-legal reports to praise their own achievements by referencing themselves, even though self-praise and self-promotion were condemned by the rules of politeness of the 17th and 18th

Quotations from 17th and 18th Century Medical Case Reports   

   415

centuries. A case in point is Johann Georg Hasenest’s reference to his own article in “Commercium Litterarium Noricum”, a southern German medical periodical. The physicians of the Collegium Medicum Onoldinum had to judge whether the use of a flower bulb, which was said to have been newly discovered, prevented infection. They explained that this was not a novel method, and that they had already utilized it successfully in therapy: (15)

[…] unter welchen auch unſer Collega der Rath Haſeneſt, der es Anno 1735, bey denen damahlig unter den Soldaten graſſirenden hitʒigen petechialFiebern, als ein Amuletum gebrauchet, u. in des Commercii Litterarii Norici Anno 1736. pag. 107. mit folgenden Worten anführet: Ich habe die Zwiebel an dem Tag und Stunde, da das æquinoctium autumnale oder der Herbſt eintritt, graben laſſen, und darbey gefunden, daß alle ſo nebſt mir bey denen Kranken ſeyn müſſen,[…], dadurch ſie verwahret gefunden, […]; (Hasenest II 1756: 179–180.) (One of these persons is our colleague Hasenest, who used it [the flower bulb] as an amulet in 1735, when many soldiers ran a fever; he wrote about this usage in Commercii Litterarii Norici Anno 1736. page 107: I prompted the digging of this bulb at the beginning of autumn and noticed that everybody who used it and stayed with the sick persons like I did remained healthy, […].)

It is quite clear why Hasenest decided to publish this medico-legal report in his compilation: he is described as a very modern and successful physician, who applied the latest therapies - another reason for quoting himself might have been the opportunity to emphasize the importance of his own experience and personal observations as an instrument of knowledge.

6 Was quoting an obligatory feature of medical case reports from the 17th and 18th centuries? Finally I want to present some data which is relevant to the issue of whether quotation was an obligatory feature of medical case reports in the 17th and 18th century. For this purpose I counted all of the quotations used in the reports from my corpus. As quoting is such a common feature of modern scientific texts, it might come as a surprise that it was not obligatory in medical case reports from the 17th and 18th century. Only eight out of 33 investigated reports from the 17th century feature direct, indirect or mixed quotes, which equates to 24 % of all

416   

   Bettina Lindner

cases. However, as already mentioned, this might be due to the configuration of the corpus. In the 18th century, on the other hand, more than half of the reports, 36 out of 64 reports to be precise, contain direct, indirect and mixed quotes (56 %). This means that the percentage had almost doubled in less than a century.

7 Conclusion The authors of 17th and 18th century medico-legal case reports used various forms of quotation. Most of these quotations were reported in indirect speech. Very few were directly quoted and there were no examples of mixed quotes as defined in modern linguistic literature. Nevertheless, there seems to have been a certain degree of combination as different statements were occasionally compiled into one. Overall, it has become evident that modern standards of quotation had not yet been formally or informally established. The examination of indications of source in the reports has shown that the number of references to medical literature increased between the 17th and the 18th centuries. As quotation in general was not as common in the 17th century as it was in the 18th century, we can find only very few references to literature in the reports from the 17th century. In the 18th century, it was more common to incorporate subjective experiential data and opinion into medical discourse. Most of these citations referred to literature from the 17th and the 18th centuries. While physicians quoted patients, witnesses, non-academic and academic medical practitioners as well as ancient authorities, it was not possible to identify a link between characteristic forms of quotation and the person quoted in each instance. The function of quotations ranged from text structuring to selfpromotion. Finally, it has become clear that quoting was not an obligatory feature of medical case reports between 1600 and 1800. Over the course of time, modern standards of scientific presentation and quoting have gradually established themselves and have led to the forms we know today. In fact, a remarkable continuity between historical medical case reports and modern case reports can be found.

Quotations from 17th and 18th Century Medical Case Reports   

   417

References Primary sources Michael Alberti: Systema jurisprudentiae medicae, quo casus forenses, a jctis et medicus decidendi, explicantur omniumque facultatum sententus confirmantur, in partem dogmaticam et practicam partitum, casibus, relationibus jurididis et medicis forensibus specialibus illustratum, […] Halle 1721. Johann Bohn: De officio medici duplici, clinici nimirum ac forensis, hoc est Qua ratione ille se gerere debeat penes Infirmos pariter, ac in Foro, ut Medici eruditi, prudentis ac ingenui nomen utrinque tueatur. Lipsiae 1704. Johann Andreas Fischer: Consilia medica continuata, que in usum Practicum & Forensem proscopo curandi & renunciandi adornata sunt. Francofurti ad Moenum 1704. Johann Daniel Gohl: Medicina practica, clinica et forensis, sive collectio casvvm rariorvm ac notabiliorvm medico-clinicorvm, chirurgicorvm ac forensivm, partim responsis ac epicrisibvs, partim sectionibvs anatomicis et dispositionibvs illustratorvm […]. Lipsiae 1735. Johann Georg Hasenest: Der Medicinische Richter oder Acta Physico-Medico Forensia Collegii Medici Onoldini ; Von Anno 1735 biß auf dermalige Zeiten zusammen getragen, hier und dar mit Anmerckungen, Dann Mit einer deutlichen Erläuterung der medicinischen Kunstwörter und vollständigen Register versehen. 4 Theile. Onolzbach 1755–1759. Friderich Hoffmann: Medicina Consultatoria, Worinnen Unterschiedliche über einige schwehre Casus ausgearbeitete Consilia und Responsa Facultatis Medicae enthalten, Und in Fünf Decurien eingetheilet, Dem Publico zum Besten herausgegeben. Zehn Theile. Halle im Magdeburgischen 1721–1733. Andreas Petermann: Casuum medico-legalium Decas II. Herausgegeben von Dessen Sohne/ Benj. Petermannen / Practico daselbst. Leipzig 1709. Christian Gottlieb Troppanneger: Decisiones Medico-Forenses. Sowohl dessen eigene, und zwar die meisten Judicis, also auch anderer, und unterschiedlicher Juristisch und Medicinischer Facultäten Urthel und Responsa, Uber Siebentzig rare und zum Theil schwere Casus, Sonderlich De Lethalitate Vulnerum. […] Dresden 1733.

Secondary sources Behrens, Rudolf & Carsten Zelle (eds.). 2012. Der ärztliche Fallbericht. Epistemische Grundlagen und textuelle Strukturen dargestellter Beobachtung. Wiesbaden: Harrassowitz. Brendel, Elke, Jörg Meibauer & Markus Steinbach. 2011. “Exploring the Meaning of quotation”. In: Brendel, Elke, Jörg Meibauer & Markus Steinbach (eds.). Understanding quotation. Berlin: De Gruyter. 1–34. Crawford, Catherine. 1994. “Legalizing medicine: early modern legal systems and the growth of medico-legal knowledge”. In: Clark, Michael & Catherine Crawford (eds.). Legal medicine in history. Cambridge: Cambridge University Press. 89–116. Dudenredaktion. (eds.). 2009. Die Grammatik. Mannheim: Duden. 529–542.

418   

   Bettina Lindner

Geyer-Kordesch, Johanna. 1990. “Medizinische Fallbeschreibungen und ihre Bedeutung für die Wissensreform des 17. und 18. Jahrhunderts”. Medizin, Gesellschaft und Geschichte 9. 7–19. Gläser, Rosemarie. 1990. Fachtextsorten im Englischen. Tübingen: Narr. Jakobs, Eva-Maria. 1999. Textvernetzung in den Wissenschaften. Zitat und Verweis als Ergebnis rezeptiven, reproduktiven und produktiven Handelns. Tübingen: Niemeyer. Lorenz, Maren. 1999. Kriminelle Körper – Gestörte Gemüter. Die Normierung des Individuums in Gerichtsmedizin und Psychiatrie der Aufklärung. Hamburg: Hamburger Edition. Marti, Hanspeter. 2001. “Dissertation und Promotionen an frühneuzeitlichen Universitäten des deutschen Sprachraums. Versuch eines skizzenhaften Überblicks”. In: Müller, Rainer. (ed). Promotionen und Promotionswesen an deutschen Hochschulen der Frühmoderne. Köln: SH-Verlag. 1–20. Müller, Irmgard & Heiner Fangerau. 2010. “Protokolle des Unsichtbaren. Visa reperta in der gerichtsmedizinischen Praxis des 18. und 19. Jahrhunderts und ihre Rolle als Promotoren pathologisch-anatomischen Wissens”. Medizinhistorisches Journal 45. 265–292. Regier, Willis G. 2010. Quotology. Nebraska: University of Nebraska Press. Schulz, Matthias. 2012. “Fraktur und Antiqua in deutschsprachigen gedruckten Texten des 17. Jahrhunderts”. Sprachwissenschaft 37/4. 423–456. Stolberg, Michael. 2007. “Formen und Funktionen medizinischer Fallberichte in der Frühen Neuzeit (1500–1800)”. In: Süßmann, Johannes, Susanne Scholz & Gisela Engel (eds.). Fallstudien: Theorie – Geschichte – Methode. Berlin: Trafo Verlag, 81–95. Taavitsainen, Irma. 1999. “Dialogues in English Medical Writing”. In: Jucker Andreas, Gerd Fritz & Franz Lebsanft (eds.). Historical Dialogue Analysis. Amsterdam: John Benjamins, 243–268. Weyers, Christian. 1992. “Zur Entwicklung der Anführungszeichen”. Zeitschrift für Semiotik 14. 17–28.

About the Authors Karin Aijmer is professor emerita in English linguistics at the University of Gothenburg (Sweden). Her research interests focus on pragmatics, discourse analysis, modality, corpus linguistics and contrastive analysis. Her books include Conversational Routines in English: Convention and Creativity (1996), English Discourse Particles. Evidence from a Corpus (2002), The Semantic Field of Modal Certainty: a Study of Adverbs in English (with co-author) (2007), Understanding Pragmatic Markers. A Variational Pragmatic Analysis (2013). She is co-editor of Pragmatics of Society (Handbook of Pragmatics, de Gruyter Mouton, 2011), and of A Handbook of Corpus Pragmatics (Cambridge University Press 2014) and coauthor of Pragmatics. An Advanced Resource Book for Students (Routledge 2012). Jenny Arendholz is a senior lecturer at the Department of Anglistik and Amerikanistik at Ludwig-Maximilians University Munich (Germany). She holds a Ph.D. in English Linguistics, which she received at Augsburg University in 2011 with her monograph (In)Appropriate Online Behavior. A Pragmatic Analysis of Message Board Relations published at Benjamins (2013). Her main research interests include (interpersonal) pragmatics, identity construction, computer-mediated communication, multimodality, (hyper-)text linguistics and syntax. Birte Bös is Professor of English Linguistics at the University of Duisburg-Essen (Germany). Her research interests include synchronic and diachronic pragmatics, discourse analysis and media linguistics. She has investigated the communicative practices of historical and modern news discourse. She is the co-author of News as Changing Texts. Corpora, Methodologies and Analysis (2012) and the coeditor of Changing Genre Conventions in Historical English News Discourse (2015). Wolfram Bublitz holds the Chair of English Linguistics at the University of Augsburg (Germany). His research interests concern most areas of text analysis, (constructivist) pragmatics and CMC. His books include Foundations of Pragmatics (ed. with N. Norrick, Handbooks of Pragmatics 1. De Gruyter, 2011), Englische Pragmatik (2nd ed., Schmidt, 2009), Metapragmatics in Use (ed. with A. Hübler, Benjamins, 2007), Coherence in Spoken and Written Discourse (ed. with U. Lenk, E. Ventola, Benjamins, 1999). Together with Andreas Jucker (Zürich, Switzerland) and Klaus Schneider (Bonn, Germany) he is the editor of the thirteen-volume series Handbooks of Pragmatics (de Gruyter Mouton, 2010 ff). Anita Fetzer is a full professor of Applied Linguistics at the University of Augsburg (Germany). Her research interests focus on pragmatics, discourse analysis

420   

   About the Authors

and functional grammar. She has had a series of articles published on context, political discourse, discourse relations, and the communicative act of rejection. Her most recent publications are The Pragmatics of Political Discourse (2013), Contexts and Context: Part Meets Whole (2011, with Etsuko Oishi), and Context and Appropriateness (2007). She is editor of the book series Pragmatics & Beyond: New Series (John Benjamins). She is a member of several editorial boards, including Pragmatics & Cognition, Journal of Language and Politics (John Benjamins), Research on Language and Social Interaction (Taylor and Francis), Text & Talk (de Gruyter) and Studies in Pragmatics (Elsevier), and she is an elected member of the Consultation Board of the International Pragmatics Association. Rita Finkbeiner is a lecturer in the German Department, Johannes Gutenberg University Mainz (Germany). In 2008 she received her PhD at Stockholm University with a thesis on ‘Sentential idioms in German. Syntactic, semantic and pragmatic studies and investigation of their productivity’. She has published on a variety of topics in German linguistics, her main research areas being phraseology, constructions, sentence types, and the grammar/pragmatics interface. She is the author of Einführung in die Pragmatik (‘Introduction to Pragmatics’) (2015) and the co-editor of the volumes What is a Context? Linguistic Approaches and Challenges (2012) and Satztypen und Konstruktionen (‘Sentence types and constructions’, to appear 2015). Her current research project investigates the grammar and pragmatics of reduplicative constructions in German. Alison Johnson is a Lecturer in English Language in the School of English, University of Leeds (UK). She researches in corpus-based forensic linguistics, drawing on discourse and conversational analysis, interactional sociolinguistics, stylistics, and pragmatics. Formerly a police officer for six years, Dr. Johnson’s doctoral research explored the pragmatic uses of questions in police interviews with both adults and children. She is author of articles on narrative evaluation in police interrogation, impoliteness and quotation in contemporary courtroom discourse, processes of making evidence in legal discourse, plagiarism, and identifying idiolect in forensic authorship attribution (with David Wright). She is author of An Introduction to Forensic Linguistics: Language in Evidence (2007) and editor of The Routledge Handbook of Forensic Linguistics (2010) (both with Malcolm Coulthard). Monika Kirner-Ludwig is a postdoctoral associate in the Department of Educational Theory and Practice at the State University of New York at Albany (USA). She earned her Ph.D. in Historical English Linguistics and Medieval Literature from the Ludwig-Maximilians-University, Munich (Germany) in 2013, with her

About the Authors   

   421

monograph titled Heathens, pagans, misbelievers. A lexico-semantic field study and its historio-pragmatic reflections in texts from the English Middle Ages (Winter, 2015). Her teaching and research interests focus on the area of Intercultural Pragmatics, which she approaches from both synchronic and historical viewpoints. These have led to interdisciplinary and interlinguistic projects on e.g. J.R.R. Tolkien’s works and language, studies into pseudo-archaisms, as well as the pragmatics of today’s youth language. Sonja Kleinke is a Professor in English Linguistics at the English Department of Heidelberg University. She has worked in the fields of Cognitive Linguistics, English verb complementation, and the pragmatics of computer-mediated discourse. Her current research interests include Cognitive Linguistics, Pragmatics, and Internet-Communication. She has published on English verb complementation, cognitive pragmatics, cognitive metonymy and pragmatic aspects of Internet-communication. Bettine Lindner is research assistant at the Universities of Erlangen and Augsburg (Germany), where she currently finalizes her PhD thesis on 17th and 18th century medical case reports. She studied history and German literature and language at the University of Erlangen. Her broader research interests cover text linguistics and historical academic languages. Ursula Lutzky is Senior Lecturer in English Language Studies at Birmingham City University (UK). Her research interests lie in the study of pragmatics, sociolinguistics, historical linguistics and corpus linguistics, and in finding innovative ways in which they can be combined. Her research focus on the study of discourse markers in the history of English culminated in the publication of her monograph Discourse Markers in Early Modern English (2012, John Benjamins), which won the ESSE Book Award for 2014 in English Language and Linguistics. This work involved the sociopragmatic annotation of corpus data to arrive at new insights into the use of discourse markers, such as their distribution according to social rank (see also Lutzky and Demmen 2013) and gender (Lutzky 2013). Jörg Meibauer is professor of German language and linguistics at the Johannes Gutenberg University in Mainz (Germany), and affiliated professor at the University of Stockholm (Sweden). His research focuses on the semantics-pragmatics interface, the grammar of German, word formation, lexical acquisition, and the linguistics of children’s literature. His monographs include Rhetorische Fragen (1986), Modaler Kontrast und konzeptuelle Verschiebung (1994), Pragmatik (1999), and Lying at the Semantics-Pragmatics Interface (2014). Among his many (co)-

422   

   About the Authors

edited books is the handbook Satztypen des Deutschen (2013). Currently, he is editing The Oxford Handbook of Lying. Colette Moore is Associate Professor of English at the University of Washington (USA). Her areas of specialization are English language studies, history of the English language, and late medieval literature, and her published research, including her 2011 book, (Quoting Speech in Early English), reflects her combined interest in historical pragmatics, sociolinguistics, and stylistics. Andreas Musolff is Professor of Intercultural Communication at the University of East Anglia in Norwich (UK). His research interests include Pragmatics of Intercultural and Multicultural Communication, Metaphor Studies, Public Discourse and Metarepresentation Theory. He has published monographs on Metaphor, Nation and the Holocaust (2010), Metaphor and Political Discourse (2004) and Mirror Images of Europe (2000) and eight co-edited volumes, including Metaphor and Intercultural Communication (2014). Elisabeth Reber is a postdoctoral researcher and lecturer for English linguistics, Institute of Modern Languages, at the University of Würzburg (Germany). Her research is in the area of Interactional Linguistics and focuses on prosody in interaction, particularly affect-laden minimal responses; the construction of evidence in political interaction; and multimodality and embodiment. Her publications include the monograph Affectivity in Interaction: Sound Objects in English (John Benjamins, 2012) and several articles in peer-reviewed journals and edited volumes. She co-edited (with Dagmar Barth-Weingarten and Margret Selting) Prosody in Interaction (John Benjamins, 2010) and (with Pia Bergmann, Jana Brenning, and Martin Pfeiffer) Prosody and Embodiment in Interactional Grammar (Mouton de Gruyter, 2012). Winfried Rudolf is professor for medieval English language and literature at the University of Göttingen (Germany). His research interests include the palaeography and editing of medieval English manuscripts, especially those of the AngloSaxon period, homiletics, medieval semiotics as well as childhood, adolescence and education in Anglo-Saxon England. He has published widely on Old English homilies and their textual instability and their Latin sources, especially on the earliest witness of the Homiliary of Angers in Europe. He has also discovered a hitherto unknown English prayer book preserved in Italy and co-edited a volume on its palaeography and its Middle English texts.

About the Authors   

   423

Klaus P. Schneider holds the Chair in Applied English Linguistics at the University of Bonn (Germany). Before, he taught at the universities of Marburg, Hamburg, and Rostock, and at University College Dublin. His current research focuses on pragmatic variation, pragmatic competence, perceptions of (im)politeness, and evaluative morphology. Publications include the monographs Small Talk: Analysing Phatic Discourse (Hitzeroth, 1988) and Diminutives in English (Niemeyer, 2003), the edited volumes The Pragmatics of Irish English (Mouton de Gruyter, 2005), Variational Pragmatics (Benjamins, 2008) and Pragmatics of Discourse (De Gruyter Mouton, 2014), with Anne Barron, and the special issues Variational Pragmatics (Intercultural Pragmatics, 2009, also with Anne Barron) and Im/ politeness across Englishes (Journal of Pragmatics, 2012, with Michael Haugh). He is co-editor of the handbook series Handbooks of Pragmatics (De Gruyter Mouton) and the book series Language in Cultural Contexts (V&R Academic), and he serves on the editorial board of the Journal of Pragmatics. Iris Zimmermann has been research assistant in both the Department of German and English at the University of Augsburg (Germany) for several years, where she compared ME love lyrics and MHG ‘Minnesang’ in her doctoral dissertation Codex Manesse und Harley Lyrics – Kompositionsprinzipien, Selektionsprinzipien, Minneauffassung. At the moment, she is completing a two year stint in teacher training.

Index of Subjects adjacency 12, 32, 71, 78, 80–81, 84, 93 alignment 17, 62–64, 82, 97, 100, 118, 120–121, 127; see also dis~ Alysoun (cf. Harley 2253) 304–305 Anglo-Latin 22, 271, 274, 291 Applikationsvorbehalt 156, 201 as followeth see quotative marker auctor, auctores, auctoritas 266–268, 271, 310 authentification 15, 309–310, 313 authority, authorization 12, 17, 36, 125, 127, 309, 314 be like see verbum dicendi Bede 309–313, 315 biblical quotation see quotation Book of Margery Kempe 327–329, 333 Canterbury Spell 329, 335, 337–339 Canterbury Tales 267, 302 case reports, medical ~ 1, 16, 24, 401–418 Caxton, William 309, 312–315 citation, citing 2, 22, 256, 260, 262, 271, 291–296, 299–300, 316 coherence 17, 19, 53, 66, 71–72, 74, 78, 91, 93, 97, 322 cohesion 1, 53, 57, 66, 79–80, 299, 334 colon see quotative marker commentator 102–103, 113–114, 117–118, 267 common ground 53, 89, 104, 121, 322, 327, 332, 339 community of practice 19, 67, 125, 132–133, 142, 217, 330 computer mediated discourse 2, 12, 17, 32, 54, 71–72, 93, 99–100, 102, 104, 116, 133, 332–323 context – change of ~, shift of ~ 2, 4, 5, 7, 10, 24, 55, 68, 147, 160–164, 168–169, 172–173, 194 – ~ual embeddedness 71, 104 – ~ual factors 36–37, 320 – original, prior ~ 4–6, 20, 40, 56, 74, 162

contextualization – ~ cues 114, 116, 120, 167, 171, 203 – see also re~ Corpus – ~ of Historical American English (COHA) 260–261, 265–266 – ~ of Middle English Prose and Verse 258 – Monomania ~ 369, 371–373 – ~ of Early English Dialogues 1560–1760 (CED) 21, 231–234, 236–239, 245, 247, 249, 251–252, 343–344, 347–350, 352–362, 364 cry see verbum dicendi cweþan see verbum dicendi deictic – ~ shifter 11, 235 – ~ pointer 21, 231, 244, 251 direct speech see speech disalignment 97, 100, 117–118, 120–121; see also alignment discussion forum see forum discourse marker 350, 354–355, 357–362, 364 disruption, communicative ~ 4, 8 Early Modern English 347, 353, 363 echo quote see quotation erzählet see verbum dicendi evaluation 13, 17, 105, 181 face – negotiation of ~ 54, 58–59 – ~ enhancing 63, 66 – ~ manipulating 18, 53, 62 – ~ threatening 63, 216 – ~ work 17, 132 face-to-face 101–102, 105, 122, 132 focus – shift of ~ 4 – ~ particle 20, 177–182, 200–204 – ~ marker 172

426   

   Index of Subjects

forum, discussion ~ 1, 11, 16, 19, 71–72, 78, 80, 84–86, 97–100, 104, 111, 114, 116, 119–122, 125–127, 131–132, 135, 147, 150, 214 function – coherence-related ~ 71–72, 78 – credentializing ~ 12, 266–268 – distancing ~ 17, 37, 43 – interpersonal ~ 62, 86–87, 343, 363 – mental access ~ 71, 83 – metacommunicative ~ 149, 167 – ~ of quoting 1, 10, 14–15, 17, 19–20, 29–32, 34–37, 39, 42–43, 46, 49–50, 53, 62, 72, 80, 82, 84–85, 97, 99, 103–105, 116, 209–211, 214–216, 413–414, 416 – pragmatic ~ 21, 29, 232, 322, 353, 364 – relational ~ 71–72, 78, 82, 84–85, 93 – trigger ~ 82, 84 Geoffrey of Monmouth 309–311, 313, 315 go see verbum dicendi go like see verbum dicendi grammaticalization 231, 233, 239–240, 244–245, 251, 255–256, 260, 265, 268 Homilies 1, 22, 271–273, 275–276, 279, 281–288 identity – discursive ~ 19, 125, 142 – group ~ 17, 86 – ~ building 209–211, 215–216, 223 – ~ construction 85–86, 127, 132–134, 137, 364 – online ~ 127, 132, 134–135, 142 – professional ~ 20, 209, 215, 300 implicature – conversational ~ 100, 147, 158, 163, 166, 177, 182–183, 188–189, 193–197, 200, 204 – generalized conversational ~ 181–183, 197–198, 200 – particularized conversational ~ 198, 200, 203 inquit (scripsit) 21, 255, 259–260, 281–282, 284, 350 intellectual property 293, 306–307, 310

interpersonal – ~ alignment 62 – ~ functions 62, 86–87, 343, 363 – ~ pragmatics 16, 53–54 – ~ relations 18, 53, 57, 62, 84 inverted commas see quotative marker irony, ironical – ~ praise 136, 202 – ~ scare quotes 179, 193, 201–203 knowledge see serial knowledge Lindisfarne gospels 278, 285 marker – discourse ~ 161, 350, 354–355, 357–362, 364 – focus ~ 172 – lexical ~ 21, 234, 237, 244, 349 – pragmatic ~ 21, 23, 103, 147, 149, 167–168, 171, 173, 232, 247, 249–252, 343–344, 347, 349–351, 353–355, 357, 363–364 – speech internal ~ 344 – turn initial pragmatic ~ 344, 354, 357, 363–364 meta-communicative – ~ acts of re-contextualizing, re-focusing and reflecting 4, 213 – ~ comment 125, 156–157 – ~ function 149, 167 metalanguage 223, 369, 386, 395–397 metalinguistic quote see pure quote metonymy, metonymic – cognitive ~ 82, 92 – ~ elaboration 71–72, 82–83, 90–93 – ~ implicature 83 – ~ inferencing 83, 92 Middle English 1, 22, 238, 240, 255–259, 269, 291, 293, 303 Monomania see Corpus namelye see quotative marker nennet see verbum dicendi newspapers – British ~ 18, 35, 43 – down-market ~ 35, 48, 50

Index of Subjects   

nickname 55, 100, 138, 330 nomen dicendi 112 Old Bailey – ~ trials 23, 369, 381 – Proceedings of the ~ 369–370, 372 oratio recta 256, 281; see also direct speech originality 22, 291, 309–310, 316 other-quoting see quoting palaeography 271, 273–274, 279, 288 perspective, change of~, shift of ~ 2, 4, 8, 10 plagiarism 20, 22, 210, 214–215, 224, 291–293, 297, 305–308, 310–312, 316 positioning strategies 20, 209–211, 214–215, 221, 224 praefatio 291, 306, 308–309, 311–312 pragmatics, pragmatic – ~ enrichment 104, 147 – ~ functions 21, 29, 232, 256, 322, 353, 364 – ~ indicator 147, 154, 158–159, 163, 167–168, 173, 191, 193, 204 – ~ inferences 141, 167, 177, 191, 200 – interpersonal 18, 53–54 – ~ markers 21, 23, 103, 147, 149, 167–168, 171, 173, 232, 247, 249–252, 343–344, 347, 349–351, 353–355, 357, 363–364 – ~ principles 188, 190, 194, 198 – variational ~ 210, 213, 220 preface position 23, 343–344, 351, 362–364 presumptive meaning 20, 179, 197 profile information 55–57, 68, 330 prologoi 291, 306, 308–309 pure quote see quote quethen, quoth, quod see verbum dicendi quotation, quote – biblical ~ 278, 281, 287 – cognitively isolated ~ 22, 319–342 – direct ~ 7, 9–10, 16, 18, 21, 23, 29–52, 73, 97–124, 127, 129, 190, 194, 231–254, 302–303, 343–344, 351, 355, 369, 372, 374, 377, 397, 401, 403, 405–406, 415–416 – echo ~ 9 – extended ~ 128 – hypothetical ~ 369, 396–397

   427

– indirect ~ 9–11, 16, 23, 29, 32, 73, 97–124, 127, 129, 194, 302, 304, 369, 372, 374, 377, 384, 386, 401, 403, 405–406, 415–416 – Latin ~ 22, 271–290 – matrix-clause-plus-direct ~ 134–135, 141 – mis ~ 401, 410 – mixed ~ 9–10, 16, 32, 111, 128, 190, 195–196, 201, 204, 401, 403, 408, 415–416 – multiple ~ 75, 77 – other ~ 2, 98, 105–109, 111, 119–121, 150 – pure ~ 10, 159, 164, 190, 403 – scare ~ 9–10, 16, 20, 158–160, 164, 177–207, 403 – self ~ 2, 9, 98, 105–109, 120–121, 150–151, 209–230, 373 – ~ type 9, 16, 104, 191 – ventriloquizing ~ 53–54, 66–67 – visual ~ 325, 329, 333 quotative – ~ verbs 16, 21, 249–250, 252, 256, 265, 268 – zero ~ 103, 116, 236 – see also quotative marker quotative marker – as followeth 21, 231, 235, 242, 244, 251, 253 – colon 11, 24, 403–404 – inverted commas 6, 11, 21, 23, 194, 319, 325 – namelye 21, 231, 235, 244–245, 251 – videlicet 244–245 – witt 244–245, 253 quote see quotation quote button 54–55, 57, 60–62, 65, 73, 68, 323–324 quotee 17, 135, 221 quote-like (utterances) 319, 321–322, 325–326, 330 quoter 4–19, 327, 332, 335 quoting – ~ as a metacommunicative act 2–9, 53 – automatized (semi, fully) ~ 11, 32, 74 – ~ competence 209, 211–212, 225 – ~ conventions (in written discourse) 3, 340 – functions of ~ 1–28, 32, 62, 72, 82, 84–85, 103, 209–211, 214–216, 413

428   

   Index of Subjects

– incorporated ~ 22, 73 – marked ~ 21, 55, 57, 60–61, 75, 77, 112–117, 164, 173, 196, 198, 231–254, 323, 343–368, 403–409 – medieval concept of ~ 291, 306 – multi-layered ~ 97, 111 – other ~ 209, 224 – ~ practices 17, 20–21, 72, 78, 93, 122, 209–230 – self ~ 210, 214, 223–224 – thread-internal ~ 18, 71–72, 78, 85 – types of ~ 9–10, 68 – unmarked ~ 21, 61, 64, 73, 75, 103, 115, 163, 173, 198, 281, 325, 327, 330

– indirect ~ 259, 343–368, 369–400, 404, 406, 410, 416 – reported ~ 15, 30, 201, 234, 244, 255–270, 346, 351, 373–374, 395–396, 401, 410 stance 9, 97–124, 116–120, 125–146, 156–157

re-contextualization 4, 23, 58, 72, 77, 101, 104–105, 213, 369; see also contextualization Red Ink Annotator 327–339 reference – intertextual ~ 291, 293, 304, 340 – self ~ 120, 222, 225 repetition 1, 4, 79–80, 93, 198, 210, 287, 320, 358, 363 reportable fact 18, 29–52, 374 reported speech see speech responsibility 32–33, 37, 120, 221, 266–267

use-mention distinction 10, 104, 156, 160

say, seien see verbum dicendi scare quote see quotation schreibet zu see verbum dicendi self-positioning see positioning strategies signatures 22–23, 53–70, 319–342 source text see text speak see verbum dicendi speaker authority see authority speech – direct ~ 15, 21, 23, 29, 104, 231–254, 255–270, 282–283, 343–368, 377, 383, 386, 401, 410–411

target text see text tense shift 11, 111, 350, 357 text – source ~ 4–5, 17, 55, 57, 77, 133, 324 – target ~ 4–5, 55, 57 – ~ type 236, 249, 251, 309, 344, 353, 363

verbum dicendi 10, 21, 55, 57, 103, 112, 120, 129, 256, 303–304, 327, 340 – be like 11, 15, 231, 234, 252 – cry 236, 239–240, 252, 355–357, 360 – cweþan 257, 286 – erzählet 406 – go 11, 15, 231, 234 – go like 11, 15 – nennet 404, 406 – quethen, quoth, quod 11, 16, 21, 231–254, 255–270 – schreibet zu 406 – speak 236–237, 239, 252 – see also quotative verbs verbum sentiendi 11, 55 Vercelli Book 22, 271–290 videlicet see quotative marker Vulgate 280, 284, 287 Walther von der Vogelweide 22, 302–304, 316 witt see quotative marker