Is reanimation of voices possible? Pragmatics of reported speech in selected Middle English texts.

ABSTRACT

In its title, the paper alludes to the monograph on mediaeval Russian reported speech by Daniel E. Collins (2001), a successful attempt at reanimating the voices silenced in the written record from the past by means of a historical-pragmatic analysis. The aims of the paper are the following: to test some aspects of this approach against Middle English data on the one hand, and to provide a pragmatic analysis of reported speech in selected Middle English texts from the Helsinki Corpus on the other.

For the purpose of the present analysis, Collins's model will have to undergo a major revision, as its original aim was to investigate a corpus of utilitarian texts (trial transcripts). The questions to be answered are: whether a historical-pragmatic analysis is at all feasible in non-utilitarian texts and if so, what kind of selection criteria have to be employed in order to identify most suitable data.

The romance genre will be analysed in the present paper. Investigation of reported speech does indeed turn out to be rewarding within the selected genre, as its usage in romances exceeds 50% in some of the samples included in the Helsinki Corpus.

The discussion of pragmatic aspects of reported speech proceeds along the lines of recent advances in historical dialogue analysis (Jucker, Fritz and Lebsanft 1999) and reported speech analysis (Couhnas 1986; Janssen and van der Wurff 1999).

"Given the fictional status of our sources, we must reckon with the possibility that what we are retrieving are renderings that may deviate strongly from what once constituted actual practice" Bax (2001: 36).

1. Introduction

Reported speech invariably involves an introduction of an additional perspective or voice into the ongoing discourse. Regardless of the form and complexity, the result of this introduction is a communicative act which is polyphonic in nature and, in order to be comprehensive and communicatively efficient, it requires setting clear boundaries between the constitutive voices (Clark and Gerrig 1990; compare Guldemann and von Roncador 2002a). In this way, the concept of voice becomes central to any report of speech and that inevitably brings spoken language into play. After all the natural property of voices is that they primarily function in the spoken medium. For this reason the limitations and conditions determined by the unavailability of this medium for linguistic history are the central theme of this paper.

Reanimation of voices from the past stages of the language has for the last decade become the focus of a new sub-discipline of historical linguistics, historical pragmatics. Posing questions about the spoken usage, this new area of studies has since its earliest attempts refused to acknowledge that speech of the past is unavailable to a modern researcher if preserved in the written medium exclusively. So far, historical pragmatic research has convincingly proven that the great majority of historical texts display at least some potential as to providing evidence on aspects of spoken language (e.g., Bax 1999, 2001; Schrott 1999, 2000; Onodera 1995, 2000; cf. Jucker 1998: 6).

2. A model for reanimating voices

A fairly recent comprehensive attempt at reanimation of the voices from the past is a monograph on mediaeval Russian reported speech (Collins 2001). This study belongs to the line of historical pragmatic research focusing on utilitarian speech-based (sermons, trial transcripts, witness depositions etc.) rather than on literary texts. The faithfulness of the former to real speech has been questioned and verified in a range of contributions (e.g. Culpeper and Semino 2000; Doty and Hiltunen 2002; Archer 2002) but the early statement (Rissanen and Kyto 1983; Rissanen 1986), i.e., that the texts recording or attempting to record speech are closer to spoken language than those which do not, still enjoys almost universal support. As far as Collins (2001) is concerned, an additional argument for the use of speech-based non-literary material in a historical pragmatic study is the institutionalized context of utilitarian genres (1) (i.e. a trial transcript). The insights gained from the conventionalized situation of the courtroom such as motivations, aims, intentions of speakers as well as functions of their utterances are valid for tracing pragmatic strategies of reported speech. In deed, the investigation into reported speech of courtroom discourse proves to be a fruitful enterprise as the growing number of contributions has lately shown (e.g. Culpeper and Kyto 1999; Archer 2002, 2003; Moore 2003; Hiltunen 2004).

3. Literary data in historical pragmatics

The issue I would like to address is whether a pragmatic analysis of reported speech in historical material has to be confined to speech-based utilitarian texts such as court records, or is it also to be accomplished in literary texts. In order to support the claim for feasibility of the latter sort of investigation I will first discuss selected arguments from historical pragmatic research which utilize literary texts. Secondly, my aim is an extension of my previous pragmatic analysis of selected Middle English romances on the basis of the samples from the Helsinki Corpus.

According to Rissanen, literary data may serve the purpose of spoken language analysis as they contain imagined speech with a "number of features with which an author hopes to create an illusion of spoken idiom" (1986: 99). Bax (1999) in turn is able to show that some literary genres (a Middle Dutch romance) are indeed mimetic of authentic language use. Applying historical explanations to the analysis of ritual precombat exchange between mediaeval knights, the author proves that their dialogue "fully answered particular realistic needs" (1999: 53). (2) In a slightly different vein, Bax looks at the literary material in the study of a 17th century Dutch farce (3) (2001). Reiterating his earlier concepts (1981, 1991) he claims that what the researcher may see in literary texts is "simulated spoken interaction" (emphasis original; Bax 2001: 37) obscure to a modern reader who is unable to come up with an adequate understanding of its covert meaning aspects. These aspects, through a sort of "translation" (4) may nevertheless be uncovered on condition that their communicative functions within their temporal and social contexts are taken into consideration. As Bax has it: "Much of what happens to be distinctly 'functional' about historical texts is a matter of interpretation and reconstruction--'translation' as it were" (2001: 34). The additional factors complicating this "translation" in the case of literary genres are the production-reception issues and the uncertainty as to whether one is dealing with real speech/life practice or literary motives. The latter problem is connected to the issue of authenticity or factual status of speech events preserved in the utilitarian genres, which is lacking from literary texts.

A sound argument against the alleged authenticity of communicative events of utilitarian in comparison to literary genres is put forward by Schrott (2000). As she rightly emphasizes, the relation of all texts to everyday spoken language is invariably determined by individual generic models and that in turn undermines the concept of authenticity (2000: 294). Rather than sustain the validity of the notion of imagined or simulated spoken language, the author believes that representations not reproductions are provided by literary sources, and she sees speech acts which she analyses as models of human interaction in a given community (2000: 266). Indeed, bearing in mind the filtering processes involved in the transition from the oral to the written medium, which are particularly valid for the orally composed literature, one is never too cautious in assessing the multilayered relation of texts to the reality of the language.

A position similar to Schrott (2000) is taken by Fritz (1995) in a paper that may be considered an introduction to historical dialogue analysis. Dialogues are the core of linguistic interaction being the most common genre of everyday communication. One can nevertheless not deny that recorded in written form, dressed in literary conventions, left speakerless (cf. Fleischman 2002) dialogues of the past are no longer more than the representations of dialogues and "products of intentional action" (5) (1995: 472), as Fritz has it. Moreover, their interpretation is even further blurred by the lack of competence of readers/researchers to become active participants of verbal interaction of the past. With these limitations in mind, not unlike Schrott (2000) above, Fritz does not find reasons for favouring either kind of material be it literary or non-literary, speech-based or non-speech-based genres in research into dialogue forms, although he points to a certain explanatory force of the institutionalization process for the evolution of utilitarian dialogue forms (1995: 486). Lebsanft also strongly supports the equal relevance of various sources for the study of dialogue forms seeing both utilitarian and literary data not as imitations of speech but "as a representation of how, in the opinion of the writers, medieval speakers tried to arrange and construct their discourse" (1999: 272).

An empirical justification for the use of literary sources in historical pragmatic studies is contributed by Culpeper and Kyto (1999), in a study of hedges in Early Modern English dialogues of two non-literary (trials and depositions) and two literary (drama and prose fiction) genres. The dialogues in the latter texts are treated as constructed and imaginary while the former provide what, in the opinion of the authors, is recorded material. Closer to the real speech event than literary data, supposedly recorded dialogues with their underlying authenticity should hypothetically be more faithful reflexes of spoken language. Thus utilitarian texts in question should contain more orality (6) features, such as hedges, than fictional constructed data. Surprisingly, the results of a statistical analysis show the opposite pattern. As the authors explain, the relatively low frequency of hedges in trials and depositions may be put down to the formality of courtroom situation as well as to the numerous filtering processes involved in the recording procedure (1999: 302).

As has been shown above, historical pragmatics has indeed found ways to overcome the ubiquitous "bad data" problem (7) (Labov 1994; cf. Fries 1998: 85; Nevalainen 1999; Kyto and Walker 2003) as well as to effectively describe aspects of spoken language of the past relying on literary material. Since, however, sources do not exhibit equal potential as to describing speech of the past, it seems in order to carry out a systematic data selection. In the present research, three Middle English romances are investigated on the evidence that this genre bears affinity to the language of immediacy (Koch 1999) as has been shown by Taavitsainen (1993). As I have further determined in my previous research, this genre seems suitable for reported speech analysis as the reported speech percentages in the total word count exceed 50% (Table 1).

4. Aspects of reported speech analysis in Middle English romances

The first and foremost issue in reported speech analysis is to provide a categorisation of various reports. Here, reported speech is seen as a continuum of categories (Leech and Short 1981) ranging from those controlled by the narrator or reporter: narrative reports of speech acts (NRSA), indirect speech (IS), through free indirect speech (FIS) where this control diminishes to direct speech (DS) and free direct speech (FDS) which are largely independent of authorial/narratorial discourse (Figure 1):

NRSA IS FIS DS FDS

Figure 1. The continuum of reported speech categories (based on Short--Semino--Culpeper 1996: 114)

Linguistic features of speech reports are an outcome of choices made by reporting speakers striving to fulfill their own speech plans (Collins 2001; Sternberg 1982: 109 (8)). Therefore an analysis of these features may provide some information as to the intention of the reporters (9) whose main function within the communicative act of reporting is mediation or interference, i.e. their part is creative and intentional. It is assumed that distinct categories of reported speech and linguistic properties of reports (e.g. formal indexing, choice of a tag, the position of a tag against the report) may reveal certain functional patterns or pragmatic strategies of speakers.

Table 2 presents the percentages of different categories of speech reports on different levels of embedding in the discourse (i.e. reports within narration or dialogue), the smallest unit being a structure with a single predicate. The statistical data reveal certain similarities of the three texts, which allows posing some further questions as to the generic conventions determining the occurrence of reported speech strategies. First of all, the predominance of direct speech, the prototypical strategy, is common to all texts and will therefore be the focus of discussion. One further aspect of RS analysed below will be the position of the tag against the report.

5. Prototypical strategy: Direct speech

As I have proposed elsewhere, direct speech as the dominant reporting strategy may be viewed as one of the conventions of the genre (Wlodarczyk-Golka 2004a). This conjecture may further be supported by the findings of Clark and Gerrig (1990: 793) who see direct quotations as a means of creating direct experience (compare Wierzbicka 1974 on the theatrical aspect of direct speech, the Lakoff 1984 and Tannen 1986:311 who see it as an involvement-creating device). This direct experience is linked to two phenomena: ineffability connected with the need to minimize communication effort and secondly, engrossment ("[o]n the addressee's side, to become engrossed in the event is to experience it vividly", Clark and Gerrig 1990: 794).

The general assumption of the authors (Clark and Gerrig 1990) is that direct speech is a "demonstration of what a person did in saying something" (1990: 769). (10) Just as demonstrations, quotations are seen as non-serious actions (Goffman 1974) that have to be essentially separated from serious actions for the sense-making process not to become critically interrrupted. (11) That this separation is linguistically crucial (12) not only in the case of direct mode has been rightly pointed out by Guldemann and Roncador (2002a) in one of the latest monographs on reported speech: "Many languages employ specialized quotative constructions for signaling the presence of reported discourse or setting off from the co-text" (2002a: ix). Clear demarcation of all reported discourse is particularly important for medieval writings which lack graphic cues for different levels of discours (13) and which, moreover, are designed for reading aloud (14) (i.e. oral transmission, cf. del Lungo Camiciotti 2000: 152).

As far as direct speech is concerned, it is, by virtue of its formal properties, always foregrounded as is noticed by Collins (2001:115). Similarly, del Lungo Camiciotti in the study of orality in the Book of Margery Kempe, emphasizes "the textual organizing function ... of marking structural and thematic salience" (2000: 154) of direct reports. Furthermore, she sees a specific kind of direct speech, dialogue, as conventional for the religious genre under discussion as well as "an involvement device rooted in late medieval religious culture which focused on stimuli for meditation which actively involved the reader/listener" (2000: 156).

To sum up, the dominance of direct speech in romances analysed in this paper not only serves to create the impression of a direct perceptual experience but standing out from narration forces the reader/hearer to appreciate the boundaries of different discourse levels. Thus the onset of direct reporting cannot pass unnoticed by the audience and its inherent abruptness may facilitate the receivers' creative participation in the communicative event.

6. Position of the tag against the report: Functions of intercalation 

In the majority of reports, tags are preposed although intercalation (putting the reporting verb in a syntactically medial position) is not infrequent while post-posed tags and untagged reports are few and far between. Table 3 shows the distribution of the unprototypical tag positions:

Table 3. Percentages of the unprototypical tag positions

                         BEVIS     HAVELOK        HORN

VD intercalated DS      33.33%        8.4%      17.51%
VD intercalated IS       0.66%          --          --
Postposed                   2%        4.2%          --

In the studied material, intercalation occurs in direct speech and in many cases in the adjoining turns within dialogue. Putting tags in a syntactically interruptive medial position rather than preposing them creates the impression of a dynamic verbal exchange whose pace is not unnecessarily slowed down by the initial reporting verb (Collins 2001). The cohesive function of intercalation pointed out by Collins (2001: 238) is corroborated in this analysis: in as many as 31.25 % of all cases, intercalation occurs in dialogic turns which are unintroduced rapid responses to previous turns, though the distribution of this phenomenon is not equal in the texts under study (see Table 4 for the distribution of this function in individual texts).

This analysis cannot, however, substantiate some further findings of Collins (2001: 240-241) and others (e.g., cf. Short--Semino--Culpeper 1996:117), like the disclaimer and distancing functions of intercalations that could not be established in the material under study. Some functions central to Collins' analysis, like for instance "conveying heteroglossic information" (Collins 2001: 239) in layered reports were found to be marginal or even negligible (2.08%). Still, a closer look at intercalation in romances shows a surprisingly consistent pattern: with few exceptions, it occurs consistently in dialogic turns in the first line and in 58.33% of cases it follows a vocative form of a noun, usually a personal name, the choice of reporting verbs is limited to seggen (54.17%) and cwepen (43.75%) with the negligible 2.08% of other verbs. The following is a typical example:

1) DS VOCATIVE 'Felawe,' a TAG seide, 'par amur: Whar mai ich finde pemperur? Dow me tel!'

(M2 NI ROM Bevis: 6 (editor's punctuation)).

Furthermore, this pattern (X + tag (seggen or cwepen), the first line) occurs even more frequently with a variable first element which may also be an interjection (11.46%, example 2), an imperative form of the verb (9.38 %, example 3), or a question word (4.17%, example 4). Other patterns (example 5) occur only in 14.58 % of cases.

2) INTERJEC 'Allas,' TAG quep Beues, 'pat ilche stounde!

(M2 NI ROM Bevis: 83 (editor's punctuation)).

3)

IMPERAT "Go nu," TAG quap heo, "sone, And send him after none Whane pe kyng arise, On a squieres wise.

(M2 NI ROM Horn: 16 (editor's punctuation)).

4)

QUESTION WORD 'Whanne,' TAG quep Beues, 'schel pis be don?'

(M2 NI ROM Bevis: 102 (editor's punctuation)).

5)

OTHER 'Dat ne wile Ich neueremo', TAG Quoth erl Godrich, 'for Ich shal slo De, and hire forhenge heye!

(M2 NI ROM Havelok: 74 (editor's punctuation)).

The repetitive nature of this usage points to its particular status in the genre in question. Tannen (1989: 50) emphasizes the cohesive function of repetition which in the present study may be seen as a way of strengthening the basic cohesive function of intercalation. As the author further claims, repetition may fulfill other functions as well. (15) Interestingly, when repetition is combined with variation, as in the material investigated here, the discourse becomes "semantically less dense" (Tannen 1989: 49). Moreover, a conventionalized discourse strategy, repetition may even be seen as "a verbal analogue of the pleasure associated with the familiar physical surroundings" (Tannen 1989: 52). Also, as Bauman (1993: 190) puts it, parallelism in oral literature functions as a key to performance (to use Goffman's 1974 term).

The above-mentioned functions of intercalation and repetition, which are closely related in the texts under study, are all speaker-based (Collins 2001 : 296-297), and at the same time compliant with the needs of the reader/hearer. In the specific case of romances under investigation (oral transmission, the lack of consistent punctuation), this tagging strategy, in my opinion, is also an important basic demarcation device whose efficiency is unquestionable due to its interruptive nature. The repetitiveness and consistency of its pattern even further fortify this function of intercalation. One cannot ignore the fact that the interruptive nature and the role of marking boundaries between different voices or levels of discourse do not exactly agree with the fact that intercalation and repetition are also cohesion-promoting. Indeed, in the material under study, the former function is dominant, taken its significance for the reader/hearer and bearing in mind the issues connected with grasping oral literature or its written, punctuation-free counterpart.

6. Summary

The focus of attention in this study has been the theoretical status of literary sources and their relevance for a historical pragmatic investigation into reported speech. Having shown possible theoretical limitations and possibilities of such research, I have proceeded with a preliminary analysis of selected pragmatic aspects of reported speech in three Middle English romances. The statistically discerned prototypical reporting strategy, direct speech and one of the unprototypical tag positions--intercalation--have revealed a common function of the two aspects of reported speech, namely that they serve to demarcate different voices in speech reports as well as to set boundaries between different levels of discourse. Far from being conclusive, this paper has been a step towards the reanimation of the voices from the past echoed in a range of linguistic aspects of reported speech.

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.

2003 The role of the question in Early Modern English trial proceedings: A corpus-based approach. [Unpublished PhD dissertation, Lancaster University.]

Bauman, Richard

1993 "Disclaimers of performance", in: Jane H. Hill--Judith T. Irvine (eds.), 182-196. Bax, Marcel

1981 "Rules for ritual challenges; a speech convention among mediaeval knights", Journal of Pragmatics 5: 423-444.

1991 "Historische Pragmatik; Eine Herausforderung fuer die Zukunft. Diachrone Untersuchungen zu pragmatischen Aspekten ritueller Herausforderungen in mittealterlicher Literatur", in: Dietrich Busse (cd.), 197-215.

1999 "Ritual leveling: The balance between the eristic and the contractual motive in hostile verbal encounters in medieval romance and Early Modern drama", in: Andreas Jucker--Gerd Fritz--Franz Lebsanft (eds.), 35-80.

2001 "Historical frame analysis. Hoaxing and make-believe in seventeenth century Dutch play", Journal of Historical Pragmatics 2/1: 33-67.

Borgmeier, Raimund--Herbert Grabes--Andreas H. Jucker (eds.)

1998 Historical pragmatics. Reprinted from: Anglistentag 1997 Giessen. Proceedings. Wissenschaftlicher Verlag: Trier.

Boeder, Winfried

2002 "Speech and thought presentation in the Kartvelian (South Caucasian) languages", in: Tom Guldeman--Manfred von Roncador (eds.), 3-48.

Brugman, Claudia--Monica Macaulay (eds.)

1984 Proceedings of the Tenth Annual Meeting of the Berkeley Linguistic Society. Berkeley: Berkeley Linguistic Society.

Busse, Dietrich

1991 Diachrone Semantik und Pragmatik. Untersuchungen zur Erklarung und Beschreibung des Sprachwandels. (Reihe Germanistische Linguistik 113.) Tubingen: Niemeyer.

Clark, Herbert H.--Richard J. Gerrig

1990 "Quotations as demonstrations", Language 66/4: 764-805.

Collins, Daniel E.

2001 Reanimated voices. Speech reporting in a historical-pragmatic perspective. Amsterdam--Philadelphia: John Benjamins.

Coulmas, Florian (ed.)

1986 Direct and indirect speech. Berlin: Mouton de Gruyter.

Culpeper, Jonathan--Merja Kyto

1999 "Modifying pragmatic force. Hedges in Early Modern English dialogues", in: Andreas Jucker Gerd Fritz--Franz Lebsanft (eds.), 293-312.

Culpeper, Jonathan--Elena Semino

2000 "Constructing witches and spells: Speech acts and activity types in Early Modern England", Journal of Historical Pragmatics 1/1: 97-116.

Del Lungo Camiciotti, Gabriella

2000 "Orality and written texts: The representation of discourse in the Book of Margery Kempe", in: Gabriella di Martino--Maria Lima (eds.), 143-157.

Di Martino, Gabriela--Maria Lima (eds.) 2000 English diachronic pragmatics. Naples: CUEN.

Doty, Kathleen L.--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.

Dylewski, Radoslaw--Piotr Cap (eds.)

2004 History and present-day pragmatics of the English language. Lodz: Wydawnictwo Wyzszej Szkoly Humanistyczno-Ekonomicznej.

Fleischman, Suzanne

2002 "Methodologies and ideologies in historical linguistics: On working with older languages", in: Susan Herring--Pieter van Reenen--Lene Schosler (eds.), 33-58.

Fries, Udo

1998 "Dialogue in instructional texts", in: Raimund Borgmeier--Herbert Grabes Andreas H. Jucker (eds.), 85-96.

Fritz, Gerd

1995 "Topics in the history of dialogue forms", in: Andreas Jucker (ed.), 469-498.

Goffman, Erving 

1974 Frame analysis. An essay on the organisation of experience. New York: Harper and Row.

Guldeman, Tom--Manfred von Roncador

2002 "Preface", in: Tom Guldeman--Manfred von Roncador (eds.), vii-ix.

Guldeman, Tom--Manfred von Roncador (eds.)

2002 Reported discourse: A meeting ground for different linguistic domains. Amsterdam--Philadelphia: John Benjamins.

Herring, Susan, C.--Pieter van Reenen--Lene Schosler (eds.)

2002 Textual parameters in older languages. Amsterdam--Philadelphia: John Benjamins.

Hill, Jane H.--Judith T. Irvine (eds.)

1993 Responsibility and evidence in oral discourse. Cambridge: Cambridge University Press.

Hiltunen, Risto

2004 "Salem, 1692: A case of courtroom discourse in a historical perspective", in: Risto Hiltunen--Shinichiro Watanabe (eds.), 3-26.

Hiltunen, Risto--Janne Skaffari (eds.)

2003 Discourse perspectives on English: Medieval to Modern. Amsterdam--Philadelphia: John Benjamins.

Hiltunen, Risto--Shinichiro Watanabe (eds.)

2004 Approaches to style and discourse in English. Osaka: Osaka University Press.

Holt, Elizabeth

2000 "Reporting and reacting: Concurrent responses to reported speech", Research on Language and Social Interaction 33/4: 425-454.

Jucker, H. Andreas

1998 "Historical pragmatics: An interdisciplinary approach", in: Raimund Borgmeier--Herbert Grabes--Andreas H. Jucker (eds.), 3-7.

Jucker, H. Andreas (ed.)

1995 Historical Pragmatics. Pragmatic developments in the history of English. Amsterdam--Philadelphia: John Benjamins.

Jucker, H. Andreas--Gerd Fritz--Franz Lebsanft

1999 "Historical dialogue analysis. Roots and traditions in the study of Romance languages, German and English", in: Andreas Jucker--Gerd Fritz--Franz Lebsanft (eds.), 1-33.

Jucker, H. Andreas--Gerd Fritz--Franz Lebsanft (eds.)

1999 Historical dialogue analysis. Amsterdam--Philadelphia: John Benjamins.

Koch, Peter

1999 "Court records and cartoons. Reflections of spontaneous dialogue in early romance texts", in: Andreas Jucker Gerd Fritz--Franz Lebsanft (eds.), 399-429.

Krygier, Marcin--Liliana Sikorska (eds.)

2004 Medieval English Mirror. Vol 1. For the Love of Inglis Lede. Frankfurt am Main: Peter Lang Verlag.

Kyto, Merja--Matti Rissanen

1983 "The syntactic study of early American English: The variationist at the mercy of his corpus?", Neuphilologische Mitteilungen 84: 470-490.

Kyto, Merja--Terry Walker

2003 "The linguistic study of Early Modern English speech-related texts. How 'bad' can 'bad' data be?", Journal of English Linguistics 31/3: 221-248.

Labov, William 

1972 Language in the inner city: Studies in the Black American vernacular. Philadelphia: University of Pennsylvania Press.

1994 Principles of linguistic change: Internal factors. Oxford: Blackwell.

Lakoff, Robin Tolmach

1984 "The pragmatics of subordination", in: Claudia Brugman--Monica Macaulay (eds.), 481-492.

Lebsanft, Franz

1999 "A late medieval French bargain dialogue (Pahetlin II), or: Further remarks on the history of dialogue forms", in: Andreas Jucker--Gerd Fritz--Franz Lebsanft (eds.), 269-292.

Leech, Geoffrey N.--Mick Short

1981 Style in fiction. A linguistic introduction to English fictional prose. London--New York: Longman.

Moore, Collette

2003 "Reporting direct speech in early Modern slander depositions", in: Robert Stockwell --Donka Minkova (eds.), 399-416.

Nevalainen, Terttu

1999 "Making the best use of 'bad' data: Evidence for sociolinguistic variation in Early Modern English", Neuphilologische Mitteilungen 100/4: 499-533.

Onodera, Noriko

1995 "Diachronic analysis of Japanese discourse markers", in: Andreas Jucker (ed.), 393-437.

2000 "Development of demo type connectives and na elements: Two extremes of Japanese discourse markers", Journal of Historical Pragmatics 1/1: 25-57.

Pakkala-Wekstrom, Mari

2004 "Discourse strategies in the marriage dialogue of Chaucer's Canterbury Tales", Neuphilologische Mitteilungen 2: 153-175.

Rissanen, Matti

1986 "Variation and the study of English historical syntax", in: David Sankoff (ed.), 97-110.

Rissanen, Matti--Merja Kyto--Minna Palander-Collin (eds.)

1993 Early English in the Computer Age: Explorations through the Helsinki Corpus. Berlin: Mouton de Gruyter.

Sankoff, David

1986 Diversity and diachrony. (Current Issues In Linguistic Theory 53.) Amsterdam Philadelphia: John Benjamins.

Schrott, Angela

1999 "Que fais, Adam?: Questions and seduction in the Jeu d'Adam", in: Andreas Jucker -- Gerd Fritz Franz Lebsanft (eds.), 331-370.

2000 "?Qui los podrie contar? Interrogative acts in the Cantar de mio Cid: Some examples from Old Spanish on asking questions", Journal of Historical Pragmatics 1/2: 263-299.

Short, Mick--Elena Semino--Jonathan Culpeper

1996 "Using a corpus for stylistic research: Speech and thought presentation", in: Jenny Thomas Mick Short (eds.), 110-131.

Steinberg, Meir

1982 "Proteus in Quotation-Land. Mimesis and the forms of reported discourse", Poetics Today 3: 107-156.

Stockwell, Robert--Donka Minkova (eds.)

2003 Studies in the history of the English language. A millennial perspective. Berlin New York: Mouton de Gruyter.

Stratmann, Francis H.

1891 A Middle English dictionary. A new edition, re-arranged, revised and enlarged by Henry Bradley. Oxford: Clarendon Press.

Tannen, Deborah

1986 "Introducing constructed dialogue in Greek and American conversation and literary narrative", in: Florian Coulmas (ed.), 311-332.

1989 Talking voices: Repetition, dialogue and imagery in conversational discourse. Cambridge: Cambridge University Press.

Taavitsainen, Irma

1993 "Genre/subgenre styles in Late Middle English?", in: Matti Rissanen--Merja Kyto--Minna Palander-Collin (eds.), 171-200.

Taavitsainen, Irma--Gunnel Melchers--Paivi Pahta (eds.)

1999 Writing in nonstandard English. (Pragmatics & Beyond New Series 67.) Amsterdam --Philadelphia: John Benjamins.

Thomas, Jenny--Mick Short (eds.)

1996 Using corpora for language research: Studies in the honour of Geoffrey Leech. London: Longman.

Vincent, Diane--Laurent Perrin

1999 "On the narrative vs. non-narrative functions of reported speech: A socio-pragmatic study", Journal of Sociolinguistics 3/3:291-313.

Warvik, Brita

2003 "Issues of orality and literacy in Old English texts", in: Risto Hiltunen--Janne Skaffari (eds.), 13-55.

Wierzbicka, Anna

1974 "The semantics of direct and indirect discourse", Papers in Linguistics 7: 267-307.

Wtodarczyk-Golka, Matylda

2004a "Reported speech analysis in a historical pragmatic perspective--in search of a database", in: Marcin Krygier--Liliana Sikorska (eds.), 77-96.

2004b "A costly compromise?--The 'bad' data problem in historical pragmatics", in: Radoslaw Dylewski--Piotr Cap (eds.), 75-110.

MATYLDA WLODARCZYK

Adam Mickiewicz University, Poznah

(1) Compare Moore (2003: 399-402) on a different justification of the suitability of slander depositions for the study of reported speech ("switching" in discourse as indicative of the relationship between spoken and written discourse, text and reported speech, code-switching and discourse organization). See also her stance on literary material: "Court records, on the other hand, presumably have no aesthetic purpose, and so serve as a better source in some ways for understanding the presentation of reported speech ... than Visser's literary examples" (2003: 401).

(2) Cf. Labov's study of insults in present day Black American English (1972).

(3) Compare Pakkala-Wekstrom (2004:153) on the relationship between actual communication and fictional dialogue in plays (quoting V. Herman. 1995. Dramatic discourse. London and New York: Routledge).

(4) Compare a similar notion in Collins, i.e. "capacity for a truly empathetic reading" (2001 : 18). He further makes a rather cautious remark as to this capacity: "While conclusive verification may ultimately be impossible, the validity of an analysis can be appraised, at the very least, for its plausibility on culture-specific and typological grounds".

(5) Compare Lebsanft's view: "Instead of obtaining a faithful picture of how people "really" talked to one another, we can get a description of how people intended to interact orally" (emphasis original; 1999: 272).

(6) Compare Moore (2003: 400), a polemic: "[Culpeper and Kyto] took as their texts the entire deposition, which blends together different styles of discourse. Considering separately the passages of reported speech from the body of the depositions might produce different results" (2003: 400).

(7) I have discussed the "bad data" problem elsewhere: Wlodarczyk-Golka (2004b).

(8) "This perspectival montage indeed renders the inset a complex, but nothing like a democratic coalition, of voices and viewpoints. As a necessary result of the subordination of part to the whole, the local perspective of the quotee always subserves the global perspective of the quoter, who adapts it to his own goals and needs" (Sternberg 1982: 109).

(9) Taking into consideration the fact that literary works are under discussion here, it is necessary to distinguish the author's speech wills visible on the level of global discourse organization from possible intentions of character's on a different level of discourse.

(10) Compare Holt (2000).

(11) Cf. the decoupling principle (Clark and Gerrig 1990: 768-769).

(12) Cf. Boeder (2002: 37): "Georgian and Svan share some formal features of speech reporting that are quite common in the languages of Near East and of Europe: reported speech tends to be preceded by a conjunction both with direct and indirect speech".

(13) Compare Moore (2003: 409) who notices that even the presence of punctuation in later utilitarian texts does not clearly disambiguate the levels of the discourse for its conventions may be obscure to the modern reader or it may not be used consistently.

(14) Cf. Vincent and Perrin claiming that there is "a clear preference for direct style in oral discourse" (1999: 306).

(15) "By facilitating production, comprehension, connection and interaction ... repetition serves an over-arching purpose of creating interpersonal involvement" (Tannen 1989: 50). Compare also Warvik (2003: 26) who sees repetition as an interactive feature and a device of contextual involvement in oral genres.

Table 1. Reported speech percentages in the total word count

Beues of Hamtoun   53.65
King Horn          59.60
Havelok            51.92

Table 2. Distribution of reported speech categories

     BEVIS               HORN                HAVELOK

TOTAL TOKENS 132    TOTAL TOKENS 132    TOTAL TOKENS 123

DS         50.75%   DS         52.27%   DS         43.09%
IS          6.07%   IS         16.67%   IS          2.44%
NRSA       35.61%   NRSA       23.48%   NRSA       34.96%
IT          4.54%   IT          7.58%   IT         16.26%
DT          3.03%   DT             --   DT          3.25%

Table 4. Intercalation as a cohesive device
in rapid verbal exchanges (individual texts)

BEVIS             HAVELOK           HORN

30.90%            60.00%            15.39%

Table 5: General and individual distribution of
variation in the major intercalation pattern

FUNCTION                    GENERAL   BEVIS    HAVELOK    HORN

REPETITION: vocative         58.33%   45.45%    53.33%   88.46%
REPETITION: interjection      11.6%   18.18%     6.67%      --
REPETITION: imperative        9.38%   12.73%       --     7.69%
REPETITION: question word     4.17%    7.27%       --       --
REPETITION: other            14.58%    7.27%       40%    3.85%
LAYERED REPORTS:
heteroglossic info            2.08%    3.64%       --       --