The Sociolinguistics of Urban Vernaculars: Case Studies and their Evaluation 9783110857337, 9783110105346

Citation preview

The Sociolinguistics of Urban Vernaculars

Soziolinguistik und Sprachkontakt Sociolinguistics and Language Contact

Herausgegeben von / Edited by Norbert Dittmar

Band 1 / Volume 1

w DE

G_ Walter de Gruyter • Berlin • New York 1988

The Sociolinguistics of Urban Vernaculars Case Studies and their Evaluation Edited by Norbert Dittmar and Peter Schlobinski

w DE

Walter de Gruyter • Berlin • New York 1988

Printed on acid free paper (ageing-resistant — pH 7, neutral)

Library of Congress Cataloging-in- Publication Data

Sociolinguistics of urban vernaculars : case studies and their evaluation / edited by Norbert Dittmar and Peter Schlobinski. p. cm. — (Soziolinguistik und Sprachkontakt : Bd. 1 = Sociolinguistics and language contact ; v. 1) Bibliography: p. ISBN 0-89925-512-4 (U.S.) 1. Urban dialects—Case studies. 2. Sociolinguistics — Case Studies 3. German language — Dialects — Berlin (Germany) — Case studies. I. Dittmar, Norbert. II. Schlobinski, Peter, 1954 — III. Series: Soziolinguistik und Sprachkontakt ; Bd. 1. P40.5.U73S64 1988 401'.9-dc 19 88-18042 CIP

Deutsche Bibliothek Cataloguing in Publication Data

The sociolinguistics of urban vernaculars : case studies and their evaluation / ed. by Norbert Dittmar and Peter Schlobinski. — Berlin ; New York : de Gruyter, 1988 (Sociolinguistics and language contact ; Vol. 1) ISBN 3-11-010534-9 NE: Dittmar, Norbert [Hrsg.]; Soziolinguistik und Sprachkontakt

© Copyright 1988 by Walter de Gruyter & Co., Berlin 30. Alle Rechte, insbesondere das der Übersetzung in fremde Sprachen, vorbehalten. Ohne ausdrückliche Genehmigung des Verlages ist es auch nicht gestattet, dieses Buch oder Teile daraus auf fotomechanischem Wege (Fotokopie, Mikrokopie) zu vervielfältigen. Satz und Druck: Arthur Collignon GmbH, Berlin Bindearbeiten: Lüderitz & Bauer, Berlin

Vorwort zur Reihe „Soziolinguistik und Sprachkontakt"

Die Sprache ist ein Organismus, der sich den Lebensformen und -Verhältnissen in Sprach- und Kommunikationsgemeinschaften anpasst. Ihre so2iale und kulturelle Vielfalt ist die Folie und der Nährboden für Variation und Wandel, Erwerb und Verlust, stilistische Anpassung und Distanzierung. Formen und Funktionen der Einbettung von Sprache und Kommunikation in soziale und kulturelle Kontexte werden in dieser Reihe zum Gegenstand v o n Beschreibungen und Erklärungen gemacht. In diesem Sinne steht im Zentrum der Reihe Spachkontakt und Soziolinguistik die Erforschung formaler und funktionaler Leistungen sprachlicher Variation. Die Beschreibung sprachlicher Variation ist nicht nur eine Herausforderung, sondern das Maß der Validität, der Beschreibungsund Erklärungsadäquatheit linguistischer Theorien und Aussagen schlechthin, und zwar aus drei Gründen: 1. Unser Einblick in Variation liefert uns Kriterien für das „Maß" des Umfangs und der Beschränkungen linguistischer Beschreibungen und Theorien. Zwischen den extremen Polen der Stützung linguistischer Theorien und Beschreibungen (reduzierte, homogene Daten — z. B. „Idiolekte", „individuelle Intuitionen" — einerseits, auf sich komplex überlagernde heterogene sprachliche Fakten — z. B. „Soziolekte", „Stadtsprachen" — andererseits) muß das „rechte" Maß gefunden werden: es liegt etwa im gemeinsam geteilten Durchschnitt v o n Varietäten und diesen entsprechenden Normvorstellungen, ihrer zwischen „ K o n v e r g e n z " und „ D i v e r g e n z " ausgleichenden Mitte. Das „rechte Maß" (ein Maß der Mitte?) läßt sich natürlich nur auf der Grundlage breiter Beschreibung von Varietätenräumen bestimmen. N u r intuitive Belege marginalisieren daher die Signifikanz der Theorie: die Daten sind nicht valide, die Evidenz liegt außerhalb des „rechten Maßes". Dagegen können (empirische und theoretische) Einzelstudien, die die lernerspezifische, dialektale, soziolektale und kontaktspezifische Variation in eingegrenzten Bereichen der Einzelsprachen erfassen, Gütekriterien für die datenadäquaten Grundlagen einer Theorie liefern. Untersuchungen dieser A r t zeigen auf, welche Regeln mit welchem kognitiven, qualitativen, quantitativen und funktionalen Gewicht in welcher natürlichen

VI

Vorwort zur Reihe „Soziolinguistik und Sprachkontakt"

Ordnung benutzt werden, und stellen damit ein entscheidendes Verifizierungs- und Falsifizierungspotential von Sprachtheorien dar. 2. Sachverhalte, Einstellungen, Erzählungen, Wertungen, Argumente, Berichte etc. können in verschiedenen stilistischen Facetten formuliert werden. Der institutionelle und informelle Alltag fordert zweck- und zielgerechte exercises de style (Queneau) — die Kunst und Chance der „Sprachspieler" (aber auch die „Qual") liegt in der Wahl linguistischer Muster und ihrer Synchronisierung auf den verschiedenen sprachlichen Ebenen („Kookkurrenzrestriktionen"). „Sich auf einen Stuhl setzen" kann man durch sich hinsetzen, Plats^ nehmen, sich setzen, sich's gemütlich machen ausdrücken — je nach den situativen und pragmatischen Beschränkungen, die zur Geltung kommen sollen. Sprachliche Mittel und Formen erfüllen kommunikative Funktionen in einer Variation, die durch sozio- und psycholinguistische Faktoren eingeschränkt wird. Das polyfunktionale Zusammenspiel von Mitteln und Formen wird in der Erlernung erster und zweiter Sprachen und in der situativen, räumlichen und sozialen Ausprägung kommunikativer Funktionen in Varietäten im Vergleich sichtbar. Indem für verschiedene Handlungszwecke und relativ zu räumlichen, situativen und sozialen Umständen angemessene Wahlen aus einem Repertoire von Varianten getroffen werden können, sind durch Varietäten und Stile überhaupt erst Möglichkeiten „kommunikativen Überlebens" geschaffen; der Prozeß des Lernens wird einem Zweitspracherwerber so durch die Variation erleichtert — Wahlen können getroffen werden. Einsichten in die „Maßeinheiten", die die funktionale Auswahl und Anwendung von Regeln steuern, führen uns zu den Regelbeschränkungen, die tatsächlich in natürlichen Interaktionen und Diskursen wirksam sind. Sie korrigieren die Verzerrungen intuitiver Regelbelege in den apriorischen Grammatiken. 3. Die Variation zeigt, daß nicht mit einem, sondern mit vielerlei sprachlichem Maß gemessen wird. Variation wurde in der Vergangenheit oft mit „störend", „chaotisch", „verderbliche Mischung", „Komplikation" und „Belastung" gleichgesetzt. Demgegenüber widmet sich diese Reihe der Erforschung der ordnenden Kräfte der Variation, die Linguisten in den letzten Jahren in vielen Untersuchungen belegt haben. „Variation" wird als multifaktorieller soziolinguistischer Raum der Durchlässigkeit, der Flexibilität, der Angemessenheit für Austausch und Ausgleich betrachtet. Manchen „idealistischen" Linguisten galt „Variation" als Last und „sprachlicher Wildwuchs", dem man durch Normierung und Sprachplanung Einhalt gebieten soll. Dagegen sehe ich „Variation" vor allem als Chance: sie gibt genügend sprachlichen Raum für Austausch und Ausgleich, Aus-

Vorwort zur Reihe „Soziolinguistik und Sprachkontakt"

VII

wähl und Abwechslung, Belastung und Entlastung in der Kommunikation. Der Traum vieler Linguisten von der „homogenen Endlösung der sprachlichen Frage" und der Schaffung von international stabilen, unveränderlichen Hilfssprachen wie das Esperanto als „Negation der Fülle" übersieht die selbstregulativen Funktionen sprachlicher Systeme, d. h. das Spektrum je nach Zielen und Zwecken variierbarer Mittel und Formen, um Austausch und Ausgleich überhaupt angemessen durchführen zu können. Sprachliche Variation ist Spiegel und Ausdruck der vielseitigen Gestaltung und Gestaltbarkeit menschlicher Beziehungen; sie ist Funktion von Austausch und Ausgleich — auf der Ebene der Interaktion von Individuen und Gruppen, von sprachlichen Varietäten und Systemen, muttersprachlichen und fremdsprachlichen Sprechern, von Sprachgemeinschaften, Minderheiten und Mehrheiten. Mit der Reihe Soziolinguistik und Sprachkontakt stelle ich die grundlegenden Funktionen sprachlicher Variation für die Gestaltung menschlicher Beziehungen in den Mittelpunkt: das Bedürfnis nach Austausch und Ausgleich, das Bedürfnis nach Zugehörigkeit und Trennung, das Bedürfnis nach dem rechten Rang, nach Verbindlichkeit und Ordnung. Sprachliche Variation dient der Regelung dieser Bedürfnisse — sie ist immer schon Ausdruck dieser Bedürfnisse und sie stellt die Mittel bereit, diese Bedürfnisse entsprechend den neuen Anforderungen und veränderten sozialen Bedingungen anders und angemessen zu gestalten. Die Sprache läßt uns wählen und sie gibt Wahlen vor — nur so können Austausch und Ausgleich, Distanz und Nähe, Belastung und Entlastung in der Kommunikation bewältigt werden. Bedingung für die angemessene Auswahl aus einem stilistischen Repertoire ist ein „Maßstab", der zur Lösung kommunikativer Aufgaben angelegt werden kann. Solche „Maßstäbe" existieren für Sprecher, Gruppen und Sprachgemeinschaften — sie sind eine Voraussetzung dafür, daß Austausch und der durch ihn notwendige Ausgleich stattfinden kann. In welchem Maße und welchen Grenzen finden Austausch- und Ausgleichsprozesse statt? Welche sprachlichen Mittel, Formen, Strukturen sind daran beteiligt? Welcher Nutzen oder Schaden ergibt sich aus extremen stilistischen Abweichungen und Divergenzen zwischen Varietäten — für Sprecher, Gruppen, Sprachgemeinschaften? Hier geht es um individuelle und soziale Konflikte aufgrund der verschiedenen Verwendung von Varietäten und um Konflikte in der Interaktion. Andererseits geht es auch um Ausgleich in Konflikten — um sprachliche Leistungen der Anpassung und Entlastung. Austausch und Ausgleich sind im übrigen auch an Prinzipien des sprachlichen Marktes gebunden. In den Kapiteln 2 und 4 des vorliegen-

VIII

Vorwort zur Reihe „Soziolinguistik und Sprachkontakt"

den ersten Bandes dieser Reihe werden die angesprochenen theoretischen Fragen und Probleme diskutiert. „Lebensweltlichen" und „systembezogenen", „grammatisch" und „pragmatisch" orientierten Ansätzen soll in dieser Reihe gleichermaßen in der Überzeugung Raum gegeben werden, daß eine Beschränkung soziolinguistischer Beschreibungen auf bestimmte linguistische Ebenen unfruchtbar und daß quantitative und qualitative Methoden sich in ihrem Erkenntniswert wechselseitig ergänzen. Die Reihe widmet sich daher Untersuchungen, die sich mit der soziolinguistischen Identität und Integration von Gruppen und Individuen in der Gesellschaft befassen. Hierzu gehören im weitesten Sinne die sozialen Formen alltagsweltlicher und institutioneller Diskurse, das Erlernen von ersten und zweiten Sprachen in mehrsprachigen und interkulturellen Kontexten, die verschiedenen Spielarten des raumbedingten, gruppenspezifischen, geschlechtsspezifischen und situationsbedingten Sprachgebrauchs, der in der Sprachverwendung offen zutage tretende Konflikt zwischen Minderheiten und Mehrheiten und die vielfaltigen durch Sprache in der Kommunikation gesteuerten symbolischen Abgrenzungen zwischen Gruppen in Kontakt. Beschränkungen auf bestimmte sprachliche Räume sind nicht vorgesehen. Einsichten in empirische Verhältnisse, Methodik und Theorie sind für die Auswahl entscheidendes Kriterium. Unser Verständnis von Sprache als einer vom sozialen und kulturellen Kontext abhängigen Variablen verlangt vornehmlich eine empirische Orientierung dieser Reihe. Ethnographische, variationslinguistische, diskursanalytische, kurz: an der gesprochenen Sprache orientierte Untersuchungen stehen im Vordergrund. Ganz im Sinne der Tradition europäischer Soziolingustik soll dabei jedoch den theoretischen Grundlagen der Disziplin und einer kontinuierlichen Theoretisierung der empirischen Ergebnisse nachdrücklich Rechnung getragen werden. Monographien stellen in der Reihe neue Forschungsergebnisse vor, Studienbücher sollen in Grundlagen und gesicherte Fundamente der Disziplin einführen. Berlin, im März 1988

Norbert Dittmar

Foreword to the Series "Sociolinguistics and Language Contact" Language is an organism which adapts to the forms and conditions of life in language and speech communities. Its social and cultural diversity is at once the framework and breedingground for variation and change, acquisition and loss, stylistic accommodation and divergence. The forms and functions of the embedding of language and communication in social and cultural contexts are the objects of descriptions and explanations in this series. In this sense the focal point of Sociolinguistics and Language Contact is the investigation of formal and functional accomplishments of linguistic variation. The description of linguistic variation is not only a challenge in and of itself, but is the measure of the validity and of the descriptive and explanatory adequacy of linguistic theories and assertions for the following three reasons: 1. Our insight into variation provides us with criteria for the measurement of the scope and the limitations of linguistic descriptions and theories. Between the extreme poles supporting linguistic theories and descriptions (on the one hand, reduced, homogeneous data (e. g. "idiolects", "individual intuitions"); on the other, complexly intertwined heterogeneous linguistic facts (e. g. "sociolects", "urban dialects")), the right balance must be found. It may be within the mutually shared average of varieties and the corresponding notions regarding norms — within the balance between "convergence" and "divergence". Of course, this "right balance" can only be determined on the basis of broad-based descriptions of variety spaces. Therefore, the exclusive use of intuitive evidence marginalizes the significance of theory: the data are invalid, the evidence lies outside the "right balance". On the other hand, (empirical and theoretical) individual investigations which include learner-specific, dialectal, sociolectal and contact-specific variation within delimited areas of individual languages can provide us with quality control criteria for the empirical foundations of a theory. Investigations of this type indicate which rules are used with which cognitive, qualitative, quantitative and functional weight in

X

Foreword to the series "Sociolinguistics and Language Contact"

which natural order, and thereby represent a decisive potential for the verification and falsification of linguistic theories. 2. States of affairs, attitudes, narratives, appraisals, arguments, reports, etc. all can be formulated in stylistically different ways. Institutional and informal everyday life demands exercises de style (Queneau) which are suited to its purposes and goals. The art and opportunity of the "language player" (but also the "agony") lies within the choice of linguistic patterns and its synchronization on the various linguistic levels ("cooccurrence restrictions"). One can express the notion "to sit down on a chair" with to take one's seat, to plop down, to make oneself comfortable — according to the situational and pragmatic restrictions which are in effect at the time of the utterance. Linguistic means and forms fulfill communicative functions in a variation which is delimited by socio- and psycholinguistic factors. The multi-functional interaction of means and forms makes itself clear in the comparison of varieties found in the acquisition of first and second languages and in the situational, spatial and social expression of communicative functions. Because choices can be made from a repertoire of variants to suit various purposes and spatial, situational and social conditions, it is through varieties and styles that the possibilities of "communicative survival" exist at all. For example, variation facilitates the learning process for the second-language learner because choices can be made. Insights into the "units of measurement" which guide the functional selection and application of rules lead us to the restrictions on rules which are actually in effect in natural interactions and discourse. They correct the distortions of intuitive evidence for such rules in the a priori grammars. 3. Variation shows that measurements are made not with one but with many linguistic measures. In the past, variation was often described with terms such as "disturbing", "chaotic", "corrupted mixture", "complication" and "burden". In contrast, this series is dedicated to the investigation of the ordering powers of variation for which linguists have provided evidence in many investigations over the past few years. "Variation" is viewed as a multi-factored sociolinguistic space of permeability, flexibility, and suitability for exchange and balance. For many "idealistic" linguists, variation was seen as burdensome linguistic proliferation which should be brought to a halt by standardization and language planning. In opposition to this view, I understand "variation" first and foremost as opportunity: it provides enough linguistic space for exchange and balance, choice and change, burden and relief in communication. The dream of many linguists of the "homogeneous final solution to the linguistic question" and the creation of internationally stable, invariable auxiliary languages

Foreword to the series "Sociolinguistics and Language Contact"

XI

such as Esperanto as the "negation of the abundance" overlooks the selfregulating functions of linguistic systems, i. e. the spectrum of means and forms according to goals and purposes in order to be able to carry out exchange and balance suitably for all. Linguistic variation is the reflection and expression of the multifaceted shape and shaping of human relationships. It is the function of exchange and balance — on the level of individual as well as group interaction, of linguistic varieties and systems, native speakers and foreign language speakers, of speech communities, minorities and majorities. With the series Sociolinguistics and Language Contact I am making the basic functions of linguistic variation for the configuration of human relationships the focal point: the need for exchange and balance, for belonging and separation, for the right position, and for attachment and order. Linguistic variation serves to regulate these needs. It has always been the expression of these needs, and it provides the means to shape these needs in a different and suitable way according to new demands and changed social conditions. Language lets us choose and it pretends to give us choices — only in that way can exchange and balance, distance and proximity, burden and relief be managed in communication. The condition for appropriate selection from a stylistic repertoire is one "measuring stick" which can be used to solve communicative tasks. Such "measuring sticks" exist for speakers, groups and speech communities. They are a prerequisite for the occurrence of exchange and the balance which is a necessary result of such an exchange. To what degree and with what limits do exchange and balancing processes take place? Which linguistic means, forms, and structures take part in such processes? Which gains and losses result from extreme stylistic deviations and divergences between varieties — for speakers, groups, speech communities? Here it has to do with individual and social conflicts based on the different ways of using varieties and with conflicts within the interaction. On the other hand, it has to do with balance within conflicts — with linguistic accomplishments of accommodation and relief. Exchange and balance are, incidentally, bound to the principles of the linguistic market. Such theoretical questions and problems are discussed in chapters 2 and 4 of this first volume of the series. "Life-world" and "system-oriented", "grammatically-oriented" and "pragmatically-oriented" approaches will be represented equally in this series in the conviction that a limitation of sociolinguistic descriptions to certain linguistic levels is unproductive and that quantitative and qualitative methods complement each other in their epistemic value. The series

XII

Foreword to the series "Sociolinguistics and Language Contact"

is therefore dedicated to investigations which deal with sociolinguistic identity and integration of groups and individuals in society. To these investigations belong in the broadest sense the social forms of everyday discourse as well as institutional discourse, the acquisition of first and second languages in multilingual and intercultural contexts, the various types of spatially-dependent, group-specific, sex-specific and situationallydependent language use, the conflict between minorities and majorities which comes to light in language use, and the multi-faceted symbolic demarcation between groups in contact which is steered by language in communication. Limitations as to the representation of certain linguistic areas in the series are not planned. Insights into empirical relationships, methodology and theory are the deciding selection criterium. Our understanding of language as a variable dependent on social and cultural context demands the empirical orientation of this series. Investigations which are oriented to spoken language, whether they be ethnographic, linguistic variationist, or discourse analytic in approach, are the focus of the series. In the tradition of European sociolinguistics, the theoretical foundations of the discipline and a continuing theorization of the empirical results will most definitely be taken into account. Monographs in the series will present new research results; textbooks will introduce the foundations of the discipline. Berlin, March 1988

Norbert Dittmar

Introduction 1. The "external" occasion which resulted in this book was a workshop involving the contributors to this volume which took place in September 1986 at the Wissenschaftskolleg (Institute for Advanced Study) of the Free University of Berlin1. The Berlin Urban Vernacular Project, which had been supported by the Free University as well as by the Berlin Senate, had reached its conclusion and was to be evaluated by outside experts. In consultation with the Commission for Research and Junior Faculty (Kommission für Forschung und Nachivuchs) of the Free University, the committee of experts was expanded to include a group of internationally-recognized sociolinguists. The three-part division of this volume is a result of this constellation: 1. Presentation of the research results of the Berlin Urban Vernacular Project (chapters 1—4); 2. Methodological considerations regarding analysis of urban vernaculars (chapters 5 — 8); and 3. Research perspectives in the examination of urban vernaculars (chapters 9-12). The research concerns of the workshop were clearly outlined with these three points. The starting point was to be a concrete case description of 1

I wish to thank the Wissenschaftskolleg (Institute for Advanced Study), especially Professor Wapnewski, for the generous permission to carry out the workshop in the quiet and productive atmosphere of the Institute. The meeting location and the technical facilitation contributed greatly to the success of the workshop. I also thank the Free University for its helpful, unhesitating financial support, without which this workshop could not have been carried out. My colleagues in the German Department supported me in every possible way during the planning and execution of the workshop. Without the stimulating research of the Berlin Urban Vernacular Project, this workshop would have been unthinkable: Thanks to the good work of my assistants, Inge Wachs, Peter Schlobinski, Detlev Kruse and Olaf Eckert, we were able to present a series of empirical results regarding the Berlin vernacular for a fruitful discussion with colleagues outside the Free University. Last but not least, I would like to thank David Antal and Heidi Hamilton for their fine contribution as translators.

XIV

Introduction

an urban vernacular (in this case, the Berlin vernacular). Chapters 1—3 present qualitative and quantitative descriptions and explanations of the Berlin vernacular based on the structural distribution of variants and conversational styles in natural contexts. Our methodological considerations are reflected in a theoretical fragment regarding variation in chapter 4. The empirical descriptions and the methodological and theoretical considerations which underlie them served as the starting point for two major points of discussion in the workshop: first, the comparison of urban vernaculars within European and extra-European areas and, second, the evaluation of the appropriateness of methodological instruments for the description of variation. At the same time, this workshop in the heart of Europe was also a "first lesson" in European sociolinguistics. The British, French and Italian presence highlights this fact. The Canadian linguists were invited not only because they are at present the most important mediators between European and American cultures, but more importantly because their research is derived from a context within which urban vernacular research is carried out most intensively in the world.

2.

The methodological and theoretical directions of the contributions in this volume cannot be assigned definite attributes, such as interpretative, variantionist, ethnographic, ethnomethodological, correlative, qualitative, etc. Rather, the papers begin with concrete problems and discuss possible solutions to these problems. At the center of all of the contributions is the question whether variation of urban spaces is the same as or different than urban linguistic styles. Various geographic, social and historical conditions are responsible for the considerable differences in type and degree of variation as well as in the explanation of the discovered variation through extralinguistic parameters. Cities such as Berlin have a highly changeable political and social history and reveal variation shaped by linguistic history and language contact. Cities such as Lecce and Montréal, on the other hand, find themselves in a situation of exchange between city centers and the areas surrounding the city. This aspect is especially emphasized in the study by Klaus Mattheier regarding Erp in the area surrounding Cologne. Despite the unique socio-historical matrix in which each of the urban varieties is embedded, universal tendencies can be postulated based on small comparisons. The Berlin "Schnauze" ("bluster") which at first seems to be a distinctive repertoire of style shows similarities to "Mannheimer

Introduction

XV

Gosch" and "gueule montréalaise". The speaker's fast tongue, monotonous intonation, denseness of jokes and quick-wittedness (hypothetically) appear to be characteristics of urban vernaculars which seem to develop naturally out of the necessity to survive communicatively in complex cities. Methodological consolidation and innovative research desiderata draw a picture of the sociolinguistics of urban vernaculars in transition from case studies. Random sample studies and ethnographic observations of particular cases complement each other. Quantitative and qualitative descriptions serve as the basis for different assumptions regarding observational and descriptive adequacy of variation. "Conversational" code-switching and quick-wittedness of urban styles deserve special attention. A number of methodological problems await solution: Can quick-wittedness be described sequentially? Here, Levinson's ethnomethodological concept presents itself. Are large amounts of data necessary for a convincing demonstration of quickwittedness? If so, qualitative studies are, according to Thibault, mere euphemisms for "not enough data." Are urban styles principally describable as variety switching? In that case, they may best be described as "code switches", as Kallmeyer/Keim and Sobrero show. What if urban styles are seen to be characterized less by conversational techniques than by special intonation? Then Wolfgang Klein's notion of the suprasegmental structure of the urban variety is a variable which affords the "cloud" of an urban vernacular a more definite contour. In this volume, language and speech communities, networks, city districts and group identity are discussed as explanatory variables. A comparison of the individual urban language projects gives us the hope of isolating overarching explanatory parameters for urban language variation.

3. comprised This first volume of the series Sociolinguistics and Language Contact of three parts. Part one (chapters 1—4) reports the results of the Berlin Urban Vernacular Project. Part two (chapters 5 — 8) discusses methodological problems of sociolinguistic descriptions of urban language varieties with special reference to the Berlin study. Part three (chapters 9 — 12) contains reports of ongoing investigations of urban varieties and is therefore dedicated to work in progress. The volume as a whole provides a representative look at the present state of urban language research and the methodological problems of sociolinguistic description of urban varieties. The diversity of methods represented mirrors the complexity of urban spaces.

XVI

Introduction

Chapter 1 summarizes the linguistic variation results of the urban language study of the Berlin variety. The dependence of phonological variables on lexical input is discussed. The historical dimensions of the genesis of dialect(s) are called on to explain the present sociolectal substratum in urban variety space. Results for selected phonetic-phonological variables, which were gained through use of the log-linear model of variation analysis based on a random sample survey, are discussed regarding their social meaning for the speech community. The quantitative linguistic variationist study confirms and further differentiates Agathe Lasch's (1928) very precise, although in some areas clearly intuitivelybased, grammar of the Berlin vernacular. Chapter 2 attempts to explain the striking linguistic differences found between East and West Berlin. "Divided City — Divided Language" is the sociolinguistic diagnosis of this speech community. The methodological problems which arise with the divided urban vernacular lead to a new definition of the concepts language and speech community. With the assistance of Bourdieu's theory of language as symbolic capital, the variety space in the western part of the city can be identified with the status-based model while the variety space in the eastern part of the city can be identified with the solidarity-based model. This differentiation can be supported by attitudes towards the Berlin vernacular. Chapter 3 deals with lexical and discourse patterns of the Berliner style register. The quick-wittedness and narrative competence of Berliners in everyday life are illustrated by examples. The narratives about conflicts in the neighborhood show how skillfully Berliners can represent their selfimage. The features of a style characterized by quick-wittedness, joking, and directness are collected. The analysis of the lexical field "schlagen" (to hit) yields new insights into the description of syntactic-semantic variation. Concluding part one, chapter 4 draws up a theoretical perspective of sociolinguistics. Since Charles-James Bailey's theoretical writings at the beginning of the 1970s the theory of variation has been completely neglected in favor of empirical descriptions. The basic dimensions of a theory of variation are elucidated within their connection to broader European linguistic traditions. The perspectives worked out in this chapter serve at the same time as a guide to the entire series Sociolinguistics and Language Contact. Chapters 5 — 8 in part two single out individual methodological problems of urban language description. In chapter 5, Wolfgang Klein deals with the role of intonation in urban language varieties and the intertwining of

Introduction

XVII

several linguistic levels. — In order to free ourselves from the helplessness of "the more we know, the less we know", a greater degree of theoretical clarity must be created. In chapter 6, Pierrette Thibault comes out in favor of the use of quantitative methods even in stylistic descriptions. Using the Montréal corpus, she shows that there are similar strategies of "quick-wittedness" in the subcultures of both Montréal and Berlin. This contribution provides a good overview of work in progress in the urban language project in Montréal. Among other things, it gives us enough material to attend to David Sankoffs provocative sentence "qualitative is only a euphemism for 'not enough data' ". In chapter 7, Levinson contrasts a linguistic variationist concept of style with an ethnomethodologically-based notion of style. According to Levinson, "quick-wittedness" can be described within the sequential analysis framework of conversation analysis. The hypothesis is that urban speakers are more likely to break rules of indirectness and of sequential closing of initiated patterns. When initiated moves in the conversation are not brought to a close and, in addition, no so-called "formulations" announce the type of conversational activity (joke, negotiation, etc.), we are confronted with a direct urban style register. This hypothesis was verified recently in an empirical study which appears within the framework of the anthology, Wandlungen einer Stadtsprache (Dittmar and Schlobinski, 1988). In chapter 8, Normand Labrie discusses the sociolinguistic explanatory category "social network". According to Labrie, it is an explanatory variable which can be applied successfully to all urban contexts. Labrie shows that, in principle, the application of network analysis would also be possible for Berlin and could be tied to the Bourdieu approach of "linguistic market." The papers in part three of the volume report research in progress. In chapter 9, Klaus Mattheier examines the evaluation of the dialect in the area surrounding Cologne (Erp). A constant exchange of linguistic norms takes place between city and surrounding areas. According to Mattheier it is not enough just to set up variation profiles of cities and their surrounding areas. Evaluations and attitudes of local citizens and commuters are decisive explanatory dimensions of a variety continuum to be understood historically. In dialectal conversations in Salento, code switching is more or less used according to situation and conversational partner. Speakers of the same or similar networks show a high degree of dialectal expressions and

XVIII

Introduction

only a small degree of code switching. If, however, for example, speakers of rural networks come together with speakers from outside these networks, intensified switching toward standard Italian occurs, although the local dialect cannot be given up entirely. Sobrero provides evidence for the role the social network plays in code switching and can also trace linguistic innovations back to certain types of code switching. In Chapter 11, Normand Labrie examines the relationship between code switching and social network based on Italians in Montréal. The explanation of his own model for the establishment of a typology of code switches follows a critical review of relevant literature. What makes this model unusual is that situational and discursive factors are set in relationship to each other. Labrie's paper is based on his dissertation which has since been completed at the Université de Laval (Québec). The volume comes to a close with an examination in chapter 12 by Werner Kallmeyer and Inken Keim of the "symbolization of social identity" based on the urban vernacular in Mannheim. The Mannheim variety is described etbnographically according to city districts. Through participant observation, the social norms of groups in the individual city districts are discovered. Tape-recordings from club and group meetings can then be linguistically evaluated according to this new information. The expression of social identity is especially evident in code-shiftings. Especially in narratives, self-identity is marked by dialect and set apart from the behavior of out-group speakers, who are cited accordingly with closer-to-standard linguistic forms. In this way, convergence and divergence in discourse can be understood as forms of symbolization of social identity. Although the contributions to this volume cannot be collected under a unified methodological or theoretical "umbrella" of sociolinguistics (which would be detrimental to fruitful scientific discussion in any case), variation in all its specific expressions and conditioning factors is the focal point of all the papers. The forms and functions of variation (cf. foreword to this series) require fundamental research and empirical studies. This series will thrive on their interaction.

Contents Vorwort zur Reihe "Soziolinguistik und Sprachkontakt"

V

Foreword to the Series "Sociolinguistics and Language Contact" . . IX Introduction

XIII

Part 1: Berlin Urban Vernacular Studies: Contributions to Sociolinguistics b y N O R B E R T D I T T M A R , PETER S C H L O B I N S K I , INGE W A C H S

1

Chapter 1: Variation in a Divided Speech Community: The Urban Vernacular of Berlin Chapter 2: The Social Significance of the Berlin Urban Vernacular .

3 19

Chapter 3: Berlin Style and Register 44 Chapter 4: Components for an Overarching Theoretical Perspective in Sociolinguistics 114

Part 2: The Analysis of Urban Varieties: Some Methodological Considerations

145

Chapter 5: The Unity of a Vernacular. Some Remarks on "Berliner Stadtsprache" b y WOLFGANG KLEIN

147

Chapter 6: Discourse Analysis in Sociolinguistics b y PIERETTE T H I B A U L T

154

Chapter 7: Conceptual Problems in the Study of Regional and Cultural Style b y STEPHEN LEVINSON

161

XX

Contents

Chapter 8: Comments on Berlin Urban Vernacular Studies b y NORMAND LABRIE

191

Part 3: Investigations of Urban Varieties Research in Progress

195

Chapter 9: Attitudes in Communication Profiles. Some Remarks on the 'Erp-Project' b y KLAUS MATTHEIER

197

Chapter 10: Villages and Towns in Salento: the Way Code Switching Switches b y A L B E R T O SOBRERO

207

Chapter 11: Social Networks and Code-Switching: a Sociolinguistic Investigation of Italians in Montreal b y NORMAND LABRIE

217

Chapter 12: The Symbolization of Social Identity. Ethnography and Analysis of Linguistic Variation in a Project about Urban Communication in Mannheim b y W E R N E R K A L L M E Y E R , INKEN K E I M

232

Bibliography

259

Contributors

276

Part 1 Berlin Urban Vernacular Studies: Contributions to Sociolinguistics by Norbert Dittmar, Peter Schlobinski, Inge Wachs

Chapter 1 NORBERT DITTMAR/PETER SCHLOBINSKI/INGE W A C H S

Variation in a Divided Speech Community: The Urban Vernacular of Berlin 1. Introduction This is a report on the sociolinguistic variety space in the politically divided speech community of Berlin (Germany). The following results were gained in a project on Berlin Urban Vernacular (BUV) which was carried out by PETER SCHLOBINSKI and INGE W A C H S in 1 9 8 3 / 8 4 under the direction of NORBERT D I T T M A R . The project focused on three topics: (1) Description and explanation of the variation of BUV, in particular the effect of the division of the speech community into social and politically distinct subcommunities; connected to this particular issue is the question as to whether there are two speech communities or one. On the other hand, BUV-variation has been described within the speech community of West-Berlin according to sex, age, social class, and social network. (2) Attitudes towards BUV. (3) Description of the stylistic and rhetorical devices of Berlin vernacular register which is called "Berliner Schnauze" ('geule berlinoise'). For a detailed analysis see DITTMAR/SCHLOBINSKI/WACHS ( 1 9 8 6 ) and SCHLOBINSKI ( 1 9 8 7 ) . Although the primary outcome of the project will be to contribute to an understanding of urban dialects of German which have not been taken into account until recently, the present discussion will focus on some general problems of sociolinguistic description and explanation with special reference to our data of BUV. 2. The Case of Berlin Like many other big cities Berlin has had a typically polycentric development over the centuries. But Berlin's development was suddenly inter-

4

Norbert Dittmar/Peter Schlobinski/Inge Wachs

rupted by the Second World War and its consequences. Berlin was occupied by the military forces of the Western Allies on the one hand, and of the Soviet Union on the other. As a consequence the Federal Republic of Germany (FRG) was proclaimed in 1949; some months later the German Democratic Republic ( G D R ) was proclaimed in East-Berlin. This division was finally cemented in 1961 by the building of the wall. Running for a length of 45 kilometers through the city center and for 120 kilometers round the city the wall still separates the western sectors from the eastern sector and f r o m the G D R . Today it is possible under specific conditions for West-Berliners to visit East-Berlin; East-Berliners on the other hand are allowed to visit the western part of Berlin only after they reach the age of 65 (62 for women).

Figure 1.1: Districts of Berlin

3. Historical Aspects of BUV In a sense, the description of BUV is a test case for German sociolinguistics; it does not belong to the extremes of the German variety space, i.e. dialect and standard, which have been investigated almost exclusively in the history of more than a century of German dialectology. Although BUV has typical dialect features, it seems to be more Umgangssprache (vernacular) than a

The Urban Vernacular of Berlin

5

consistent dialect. As M U N S K E ( 1 9 8 3 ) points out, German dialectology developed strategies immunizing it from the description of complex heterogeneous varieties of German. BUV was therefore long considered to be "irregular" and therefore "improper". These qualities of BUV stem from its history as a spoken variety of High German with Low German substratum. Grosso modo there are four periods which produced the present linguistic situation of BUV: (1) From the seventh to the twelfth centuries Slavic was spoken in the Berlin area. (2) From the thirteenth to the fifteenth centuries Berliners spoke Middle Low German. (3) In the sixteenth century Berlin was heavily influenced by the written language of eastern central Germany (the Luther Bible) and the spoken vernacular called obersächsische Umgangssprache. (4) In the eighteenth century High German became the prestige variety and was generally codified in the nineteenth century. BUV has become finally a kind of slang (Berliner Jargon).

4. Some linguistic features of BUV As can be seen from fig. 1.2, Berlin and the surrounding area belongs to the dialectal variety "Brandenburgisch". In general, this bundle of local dialects conserves the Low German plosives [t], [p], [k] (in contrast to High German [s], [pf], [?]). The BUV is marked by the following cluster of variants (see Table 1.1). Variable

Realization

(g)

g

J

Phonologically restricted to: (i) w o r d initial position: [Jamain], [Je:bç], [Ju:t] etc. (ii) w o r d medial position: [le:Jç], [li:Jç], [morjij] etc.

(ai)

ai

(au

(S-10)

(S-7)

zm o

o^ — tI 2— t 5I o z

'k, sach: [I say:] "Aber eins schwoer ik dir: [But I swear you one thing:] Du kannst Kinder anfassen, [You can touch children] wen und wat du willst", [whoever and whatever you want,] ick sage: [I say:] "Und seh ich das, [and if I see it,]

N,

79

Berlin Style and Register

ob dit meine sind (.) [whether they're mine] oder auch fremde Kinder", [or someone else's,]

Y ( S - 8 ) - (S-9)

ick sage: [I say:] "Then hebt sich ALLES u f f [nothin's gonna hold me back] oock von 'na Verwaltung!" [not even the management!] Ick sage: [I say:] "Ick hau da mit'n Kopp [I'll slam your head] durch de Betonwand durch! [through that cement wall!] Wir brauchen kein Bohrer!" [We won't need any drill!] Sag' ick, [I say,] "merk da dit!" [remember that!] So, isser blass jewordn, [Well, he went pale,] und nu isser ruhich. [and now he's cooled it.]

m

C

5

80

Norbert Dittmar/Peter Schlobinski/Inge Wachs

In example (e) the dispute is provoked by a threat of the adversary. The threat expresses his confidence that certain personal traits of his put him in a position to declare sanctions and thereby move the speaker to execute or abandon a certain action. He thereby defines a relationship of forces to his own advantage. Normally used only at the climax of a dispute, the threat, when resorted to at the beginning of an argument, very quickly introduces the final stage of escalation (see APELTAUER, 1 9 7 7 : 1 0 4 ) . With his threat, O also arrogates competences to which his status does not entitle him, for the right to administer physical punishment to children belongs only to their parents. Thus the threat is an arrogant claim to status also because the opponent tries to usurp rights reserved for the speaker alone. The opponent's attempt to dominate is thus parried with an act of compensation. Following the presequence of the imperious-address type is an utterance that cannot be classified precisely at this time ( D u kannst machen, wat de willst [Ya c'n do what ya want]), and then by an insult (du Affe [you ape]) and a threat. This countermove is characterized by the tactic of staying calm and by the countertactic of answering a threat with a threat. The imperturbability is expressed in the fact that the speaker first lets his antagonist "advance" prior to the actual counterblow. As a reaction to a threat, his Weeste, du kannst machen, wat de willst communicates that the perlocutionary effect of the threat has not been achieved. This negates the opponent's claims to status and underscores the speaker's own strength. But the tactic of "letting the opponent advance" is simultaneously part of a rhetorical feint ("avoidance strategy"). The speaker makes clear that he is in no way prepared to accept the opponent's threat. In the next moment, the antagonist is caught off guard by a countertactic of the speaker. The adversary's defeat is clearly assured with a counterthreat. The comparatively colorless ick hau demnaechst deine Tochter oder dein Sohn (I'll spank you daughter or your son before long") is trumped by a monstrous verbal tour de force (see also section 1 . 2 . 4 ) . The hyperbole that distorts objective possibility into something impossible in no way damages the speaker's credibility, for the effect of the threat is not in the realization of that which is announced but in intimidation and deterrence. Thus, the speaker cannot threaten "too much" (SCHELLING, 1972: 261). Unlike the opponent's threat, the speaker achieves the intended perlocutionary effect, a fact that receives extra mention in the concluding sequence: So, isser blass jeworden, und nu isser ruhichl [Well, he went pale, and now he's cooled it!]).

81

Berlin Style and Register

2.2 The "Hypothetical" Conflict The Berliner comes across as an even bigger loudmouth in the hypothetical conflict — the construed conflict situation — than in the real one. The hypothetical conflict consists of a premise (PREM), an opponent's possible action that the speaker judges to be intolerable, and an act of compensation (COMP). The premise and negotiation are linked as a causal relationship. The hypothetical conflict assumes the form of a conditional speech act as defined by WUNDERLICH (1976: 272). In the majority of our examples, it has the form of "if x, then y". Example (f) depicts a speaker's reaction to a person who has parked illegally on the sidewalk. (f)

Und wenn da eena, [And if someone,] PREM

verstehste, [ya understand,] mit Auto ruffjeht, [goes up there with a car] dit hat u f f s Damm stehn, [that is supposed to be on the street,] dem schlag ick uff t Jehuern. [I'll knock his block off.]

—•

COMP

The conditional does not have to be formulated explicitly in every case, however. It can also follow implicitly from the context, for the conditional speech act depends on context; in other words, it is determined by pragmatics, not by semantics (see WUNDERLICH, 1 9 7 6 : 2 7 6 ) . This is also clear in example (g), which concerns the consequences for the person w h o dares to incite big children to Jiit smaller ones. (g) Ick hau den vor de Schnauze, der, verstehste, uergendwie 'n Aeltren uff putsch oder wat oder uergndwat machen will.

\

COMP PREM

82

Norbert Dittmar/Peter Schlobinski/Inge Wachs

[The guy who gets an older kid to beat up on someone else or somethin' like that, I'll belt 'im in his trap, ya know?]

Construed conflicts generally offer more opportunity for ascribing attributes of strength to oneself than do descriptions of "real" disputes, in which various narrative constraints (see SCHUTZE, 1 9 7 6 : 1 6 3 ) hold the speaker to the sequence of events as they actually transpired. Without objective facts to keep to, the speaker in a hypothetical conflict can choose a type of negotiation especially appealing to him. Much as with the threat, it thus seems that there cannot be "too many" sanctions. Compared to the usual pattern of behavior, premise and compensation are typically way out of proportion with each other, monstrous sanctions being imposed even for minor offenses. In example (h), for instance, the hypothetical screaming of a small child is followed by a violent act.

(h) Wenn dit Ding anfaengt bruellen = = dann komm ick hoch mit'n Knueppel: NAH! BOM! BOM! [If that thing starts screamin', I'll go up there with a club: NAH! BAM! BAM!]

PREM L

COMP

To be sure, physical factors and social conventions make it unlikely that the announced acts of violence would be executed. In some cases, though, the speaker does not forego detailing even the most improbable things. Example (i) presents the description of what awaits an opponent who strikes the speaker's children.

(i)

Und jeht ER runter [And if HE comes down] PREM und faesst eens von meine Kinder an, [and lays hold o' one o' my kids,]

83

Berlin Style and Register

den nehm ick [I'll take 'im] und drehn Hals um, [and break his neck,] >

COMP

und denn is Ende mit ihm! (.) [and that'll be the end of 'im!] Dit schwoer'k da! [I swear it!]

The speaker in this example is not content with wanting to wring the adversary's neck in the verbal sense. Emphasizing resolve with the performative formula dit schwoer 'k, da (I swear it), the speaker describes the act of violence as a sequence of action in three "frames": Den nehm ick (I'll take 'im) — und drehn Hals urn (and break his neck) — und denn is Ende mit ihm (and that'll be the end of 'im). A more detailed comparison is found in example (h), too. The three frames described are (1) the speaker, armed with a club, hurrying up the stairs, (2) the opponent's door being swung open, the speaker taking a hostile stance ( N A H ! ) , and (3) the violent act itself being executed {BOM! BOM!). The act of compensation in example (k), below, is a take-off on the obligatory "pie fights" of the silent film. The fate of a delivery man w h o delivers poor-quality meat is described.

(k) Wenn da so'n Fleischer mit mir so wat versuchn wuerde, dem wuerde ick diese Por^jon knallhart ^urueckschmeissen, ueber'n Tresen — ins Jesicht! [If a butcher ever tried something like that with me, I'd chuck this cut over the counter — right smack back in his face!]

PREM COMP

In this example, the act of compensation is detailed by two paraphrases (,ueber'n Tresen — ins Gesicht).

84

Norbert Dittmar/Peter Schlobinski/Inge Wachs

(1) Aber wenn eener fremde Kinda anfasst [But when a stranger lays hold of kids] >

PREM

^

COMP

undsoweiter, [and so on,] denn werd ick a f f i c h , [then I get uppish,] und denn looft dit bei mir jamj rigoros. [and then I really come down hard.] Denn jeh ick mit dem durch de Wand [Then I'll go through the wall with 'im — ] bloss dit den sein Kopp natuerlich vor mein is, ['cept, o' course, that his head goes first,] wa? [right?] Hat er Pech jehabt! [He's outa luck!]

In example (1), the compensation {denn werd ick a f f i c h , und denn looft dit bei mir jan% rigoros), which is indicated only vaguely at first, is escalated to be a hyperbolic act of violence by additional information made available afterwards. On the basis of our observations, we can evolve the thesis that the "big-mouthed" act of compensation in the hypothetical conflict is characterized by the most detailed depiction possible of a reaction by the speaker that, given physical limits and conventional norms, is extremely unlikely. Vague declarations of sanctions like denn loof dit bei mir jan% rigoros, dann kann er wat erlebm, or denn f l i p p ick ja nu jan% aus are elaborated in more detail where possible. By outlining a grotesque act of violence, the speaker absolves himself of any obligation to carry out the deed just announced. Whereas the "realistic", credible act of compensation creates obligations to carry out what has been announced, the grotesque does not need to be

Berlin Style and Register

85

acted upon because it cannot be carried out. This, however, in no way keeps the speaker from underscoring his resolve to make the improbable come true. The preference for the indicative verb instead of the subjunctive customarily used in hypothetical formulations (see GRUNDZUEGE, 1 9 8 1 : 532) is an expression of a self-assured belief in one's own grossly exaggerated capabilities. 2.3 Summary Living in a city encourages competitive struggle between its inhabitants ( W I R T H , 1974). "Big-mouth" cockiness is an important component of selfassertion in this struggle. It is an expression of adaption to the specific ways of life in the big city. In rural areas and provincial towns, where most people still know each other, the social hierarchy defined by signs of status like occupation or affiliation with a specific social class is not only relatively stable but readily intelligible for everyone. The social rank of communicating partners, one's position in the societal structure is a basic factor of communication for the members of the local community. It regulates not only the choice of topic but the style of interaction as well. In the anonymity of the big city, however, knowledge about one's neighbor is comparatively limited. Herein lie opportunities and risks in the negotiation of social rank through interaction. The handling of conflicts, the struggle between two parties, becomes an important touchstone of their strength (COSER, 1 9 7 2 : 1 5 8 159), with the self-assured belief in one's own superiority being an important source of power as a self-fulfilling prophecy in successfully asserting claims to prestige. L A S C H ( 1 9 2 8 ) has already pointed out the Berliner's characteristic belief in his own superiority. By labeling the style of speech in Berlin as "bigmouthed", the members of our speech community acknowledge this state of affairs. As there are no studies with a comparable thesis, our analysis represents an initial step toward describing a style of speech that we intuitively feel is "big-mouthed". In analyzing five examples of conflict descriptions, we found that the disputes involved power struggles over the identity of the participants — regardless of the issue actually being contested. The main point of the description was the negotiation of positions, an activity manifested in the establishment of an asymmetrical relationship between forces. The speaker's claim to domination was advanced in various ways — depending on the opponent's potential for aggression.

86

Norbert Dittmar/Peter Schlobinski/Inge Wachs

An adversary who was not overtly aggressive was bluffed with aggrandizing variations of the request. In this way the speaker succeeded in tentatively arrogating a high social status while according the opponent a subordinate position, thereby trusting that self-confident claims to presige had the best chance of being accepted. The overtly offensive-minded opponent, the one who openly tried to dominate, was met with studied calm. With the absolute control of someone who knows he has the upper hand, the speaker allowed his antagonist to "advance", frustrating him with the information that he was not an adversary to be taken seriously. The relationship between forces as defined by the opponent was thereby not only negated but turned in the speaker's favor. Letting the opponent "advance" can also be a preparation for the real counterblow, which consists in the reply. Already frustrated, the adversary was caught off guard and outflanked by the speaker. Letting the opponent "advance" was a clever technique to negate and go the adversary one better in the bid for status. The speakers executed both a greater number and a greater variety of speech acts than did their opponents, whose utterances were restricted primarily to refusals and threats. The speakers held both the insult and the imperious address in reserve for their own moves. Whereas the opponent's claim to domination was advanced solely through threats, the speaker was able to choose between the threat and the intensifying request. Accordingly, these pretensions were negated by the opponent through refusal, which, in turn, was nullified by the speaker with the rhetorical technique of letting the opponent "advance". Unlike the opponent's threats, which ricocheted off the speaker's calmness, the speaker's threats fully achieved their perlocutionary effect. This was expressed in a brief description of their impact. In the passages of dialogue, the adversaries generally did not often get a word in edgewise; when they did, their verbal contributions were shorter for the most part. Whereas the opponents were granted as little speaking time as possible, the protagonists took comparatively large amounts of time for the appropriate formulation of the utterances, whose importance was stressed by the verbum dicendi {ick sage/sag' ick) introducing and concluding the passages. This established a difference in rhetorical competence that aided the effort to establish an asymmetrical structures of forces. The adversary's defeat was assured once and for all by the speaker's final move, which brought the opponent to give up in resignation because he was no longer in a position to execute further moves to gain the victory himself. This can be expressed through the use of two techniques.

Berlin Style and Register

87

(1) The speaker concludes his description of the conflict by quoting himself. By not allowing his opponent get a word in edgewise, he gives the impression that the adversary was speechless. (2) The speaker's victory is expressly highlighted in a resolution part following the compensation. This makes the effect of the final move on the adversary explicit. For example: Denn war Ruhe, wa? (Then it was quiet, eh?), Na hat er'n Flunsch jeyogen (Well, he pouted), er is nur blass, wenn er mich sieht (He goes pale whenever he sees me), So, issa blass jewordn, und nu isser ruhik (Well, he went pale, and now he's cooled it), Und denn is Ende mit ihm (And that'll be the end of 'im), or die sind so kleen sind se jeworden (They really got cut down to size). 3. The Semantic Field of "schlagen" Beyond the realm of conflict descriptions, physical violence is an extraordinarily rich linguistic domain. The following utterances are intended to convey an idea of the range of expressions within the semantic field of the word schlagen (to beat, to hit). Picking up on SANDIG ( 1 9 7 8 ) and FILLMORE ( 1 9 6 8 ) , we will follow this list with a description of the data in terms of two characteristic types of references — "the target of attack" and "the blow" — that is, the verbal predicate and the predications concerning the blow. We will also study two patterns of formulation often used. The object is to present a scheme of description that can serve as the basis for the contrastive study of dialects. The type of sentence and the illocution of the utterances are two other important components to be considered, but they will have to be ignored in the present context for the time being. 3.1 The Data 1. Pass uff, sonst krieste heisse Ohrn! ( ) [Watch it, or you're gonna get a slap!] 2. Die kriejn so een rinjeschmokt, dat's ja nich anders jeht! (22a/199) [they'll get clobbered so hard they won't know what hit 'em.] 3. Dem schlag' ick u f f t Jehuern! (22a/207) [Then I'll knock his block off!] 4. Ick nimm allet und denn KNICK KNACK is allet zu spaet fuer den! (22a/217) I'll take everything and then SLAM, BANG it'll be all over for him.]

88

Norbert Dittmar/Peter Schlobinski/Inge Wachs

5. Dir hau'k vor de Fresse! (22a/231) [I'll belt you in the kisser!] 6. Ick hau demnaechst deine Tochter oder dein Sohn! (22a/242) [I'm gonna slap your daughter or your son before long.] 7. Ick hau da mit'n Kopp durch de Betonwand durch! Wir brauchn kein Bohrer! (22a/257) [I'll slam your head through that cement wall! We won't need no drill!] 8. Denn jeh ick mit dem durch de Wand, bloss dit den sein Kopp natuerlich vor mein is, wa? (22a/279) [Then I'll take 'im through the wall, 'cept, o'course, that his head goes first, eh?] 9. Ick schlach da windelweich! (22a/302) [I'll beat 'im to a pulp!] 10. Die kricht so'ne Oelfmeter, dass die ja nich aus de Oogn kieken kann. (22a/329) [I'll blast her so hard she won't see straight.] 11. Halt die Schnauze, sonst pell ick dir eene! (22a/354) [Shut your trap, or I'll let you have one!] (a) Der haut sich mit Micha, [He's fighting with Micha] (b) der haut ihm, oder [He's hitting him] or (c) er haut ihn. (22a/401) [He's hitting him] 13. "Pass ma uff", sag' ick, "denn kriste von mir reichlich, wat de brauchst!" (22a/418) ["Watch it," I said, "or I'll give you plenty o' what you're askin for!" 14. Denn kriegt er wieder eine. (22a/432) [Then he'll get another one.] 15. Den nehm ick und drehn Hals um, und denn is Ende mit ihm. (22a/ 434) [I'll take 'im and break his neck, and that'll be the end of 'im.] 16. Ick hau se vor'n Kopp. (22a/442) [I'll knock her silly.] 17. Jochen schlaegt keine fremden Kinder. (22a/448) [Joachim doesn't hit other people's kids.] 18. Ick hau den vor die Schnauze. (22a/452) [I'll belt ya right in the mouth.] 19. Ick fass keine fremden Kinder an. (22a/449) [I don't touch other people's kids.]

Berlin Style and Register

89

20. Ick wollt ihm uff de Schnauze haun. (15a/90) [I wanted to slug 'im in the mouth.] 21. Ick haett ihm beinah eene rinjehaun. (15a/92) [I almost let 'im have one.] 22. Petra, jetzt hauste Sonja! (22a/414) [Petra, now you go belt Sonja!] 23. Den werd ick mir irjendwann krallen. (15a/108) [I'm gonna nail him sometime.] 24. Harry darf ja ooch nich so viel sich erlauben, sonst kricht er oben wat uff de Schnauze. (22a/358) [Harry mustn't go too far, either, or he'll get one in the mouth.] 25. (a) Wenn die beide sich nu keilen, is dit jut. [If they're only roughing each other up, that's O.K..] (b) Aber Sonja haut Harry, denn hat aeh meine Tochter [But if Sonja hit Harry, then, uh, my daughter has] (c) dit Recht, die Petra zu schlagen. (22a/424) [the right to hit Petra.] 26. Gestern habm sich zwei Brueder hier drin gepruegelt. (4b/35) [Two brothers had a scrap in here yesterday.] 27. Ick geh mit Dir durch de Wand. (22a/292) [I'll take ya through the wall.] 28. Der kricht wat jegen Kopp! ( ) [He'll get one on the noggin.] 29. Wenn die ruffkommt, kricht se wat u f f t Hackbrett. ( ) [If she comes up here, she's gonna get it.] 30. Dann kriegt der eine jedroehnt. ( ) [Then he's gonna get lambasted one.] 31. Wat, die habm sich jepruejelt? (4b/38) [What? They got into a fight?] 32. Ick hab die beede pruegeln lassen. (4b/40) [I had 'em both get whipped.] 33. Krist gleich'n Barhocker a u f n Schaedel da! (4b/33) [You're about to get busted over your skull with a bar stool!] 34. Uff eenmal habm sich hier jepruejelt. (4b/186) [All of a sudden they had a fight.] 35. Eena Sekunde spaeter K L A T S C H K L A T S C H jing dit weiter, wa? (4b/ 192) [A second later WHAM, WHAM it kept up, ya know?] 36. Ja, und olle Klaus hat also ein paar Dinga von Achim jekricht. (4b/ 197)

90

Norbert Dittmar/Peter Schlobinski/Inge Wachs

[Yeah, so Achim let ol' Klaus have a couple.] 37. Kommt wieda rin, sacht wat zu Achim, BAB AB A jings weita, die Sache, wa? (4b/198) [Comes in again, says somethin' to Achim, WHAM, WHAM, WHAM, it kept up, ya know?] 38. Und der hat wie ein wie wie'n Jewehrfeuer uff ihn einjedroschen, wa? (4b/198) [And he whaled away at 'im like rifle fire, eh?] 39. Der sich mit Angelika kloppt. (4b/249) [He fights with Angelika.] 40. Neulich hat se der Vera eene uebajezogen. (22a/407) [He clouted Vera one the other day.] 41. Du krichst gleich eine gegen die Buerne, dass de nach hintn fliegst. ( ) [You're about to get one on the noodle tha'll send you flying.] 42. Hat man einen einzigen Jugenlischen taetlisch anjegriffen? (KH12) [Was a single youngster physically assaulted?] 43. Dann hat man Angst, dass man 'n paar uff de Fresse kricht. (KH12) [Then you start worryin' you'll get a few in the mouth.] 44. Habn se dich schon mal aufs Maul? (KH12) [Have they already hit you in the mouth at some point?] 45. Kannst ma glauben, Junge haut nie 'n Maedchen. (KH13) [Believe you me, a boy doesn't hit a girl.] 46. Zwei Mann komm wegen euch nich mehr in'ne Bruecke, ... die zusammjeschlagen worden. (KH15) [Because of you, two guys who got beaten up don't come around this place anymore.] 47. Bis er sich entgleisen lassen hat und den beiden Beratern wat uff de Schnauze jehauen hat. (KH15) [Until he let himself get riled up and let the two advisors have it in the mouth.] 48. Man stachelte ihn aber solange hinaus dadrauf auf, dass er eben zuj eschlagen hat. (KH16) [But they needled him so long about it that he just started swinging.] 49. Man schlaegt aber auch, wenn man nuechtern ist. (KH17) [A person punches when he's sober, too, though.] 50. Er hat jahnich zujeschlagen. (KH18) [He didn't throw any punches at all.] 51. Ick versteh' dit nisch, warum die'n paar ufPs Maul kriegen! [I don't understand why they're gettin' a few in the mouth!]

Berlin Style and Register

91

52. ... dass die ohne wat, ohne Grund 'n paar uffs Maul kriegen! (KH20) [... that they're gettin' a few in the mouth without havin' done anything, for no reason!] 53. Meinste ick schlage mich aus Langeweile? (KH24) [Ya think I fight 'cause I'm bored?] 54. Wenn hier eener ma rauslooft und der mich anjewixt hat, hat 'n Nasenbeinbruch oder sonst wat, ha' ick nur meine Haut verteidicht. (KH24) [If someone runs outa here and he's screwed me over, has a broken nose or somethin', I've only defended my own hide.] 55. (a) Dir wuerd' ick heute wieder eene uff de Nase hauen, weil du so'ne Scheisse erzaehlst. [I'd belt you in the nose again today because you're talkin' such shit.] (b) Aber ick kann doch ooch eene uff de Nase hauen, wenn ick nich jetrunken habe, wenn de so'ne Scheisse erzaehlst.(KH27) [But I can belt (you) in the nose, too, when I haven't drunk anything if you talk such shit.] 56. Is ooch schon mal vorjekomm, dass [...]wenn ick besoffen bin, einfach ein, ohne ohne Grund eine vors Maul jehau'n hab. (KH28) [It's also happened when I've been drunk that I've just hauled off and belted (somebody) in the mouth for no reason.] 57. Der erzaehln mir gleich eine Schtory, ja dat ick se 'ne Ohrfeige hau'n musste. (KH30) [They were just tellin' me a buncha bull, so I had to slap 'em.] 58. Dann haut Krawallo den sofort eins vor die Schnauze. (KH31) [Then Krawallo belted 'im right away in the mouth.] 59. Die Fuenfe sind nur Schlejer, haun alle raus. (KH32) [Those five are just brawlers, throw everybody out.] 60. Die schlagen wa doch alle raus hier. (KH32) [We'll throw all of 'em outa here.] 61. Habm wir schon ma grundlos zugeschlagen? (KH33) [Now did we start swingin' for no reason?] 62. Und hat nu vom Rudi grade wat gekricht. (KH34) [And now (he) got one from Rudi.] 63. (a) Weeste, du muesst eine uff de Nase kriegen, [Ya know, you oughta get one in the nose.] (b) und sachste: "Hau doch zu, [An' ya say: "Go ahead, hit me,] (c) schlach d o c h ' t Nasenbein ein."

92

Norbert Dittmar/Peter Schlobinski/Inge Wachs

[just break my nose.] (d) ick habe frueher mal die Schnauze hingehalten, verstehste? Kiek dir ma mein Gesicht an, wie't aussieht, nur weil ick gesacht hab: "Schlach doch ein!" (KH36) [I stuck my nose out once, ya know? Look at my face, what it looks like just because I said: "Go ahead and hit me!"] 64. Hau ick die eine auf de Nase. (KH40) [Yeah, I'm goin' to belt you in the mouth today, too!] 66. Ick hau' dir uff de Schnauze. (KH42) [I'll belt you in the mouth.] 67. (a) Hucker haut dir auf de Schnauze, wenn er, ... wennste ihm sachst: "Hucker, hoer uff zu trinken", [Hucker's goin' to belt you in the mouth if he, ... if ya say to 'im: "Hucker, stop drinkin'."] (b) hau ick dir uff de Schnauze. (KH42) [I'll belt you in the mouth.] 68. Und da hab ick die Frau eine jescheuert, ja? Dit dit jibs ja nich! (24a) [And then I, I clouted the woman, yeah? That's, that's incredible!] 69. "Hausbesetzerin" trau ick mir nich zu sagen, dann kriejn Se een eens uff de Gusche. (81/965) [I don't dare say "squatter" or they'll get aced in the mouth.] 70. Serge, du krist eene in de Fresse! (26a/520) [Serge, you're gonna get one in the trap!] 71. Wenn dit Ding anfaengt zu bruellen, dann komm ick hoch mit'n Knueppel: "NAAH! - BOM BOM!" (28b) [If that thing starts screamin', I'll go up there with a club: "NAH! - - BAM BAM!"] 72. Mein Bruder is schon een Jahr im Knast, wenn DER rauskommt! BAH, da is mein Vater ferti [...], denn ( ) mein Bruder eh den so zusamm, det jibs ja nit! Kann er gleich [hin] uff de Intensivstazjon jehn. (28a/130) [My brother has already BEEN in jail a year; when HE gets out! MAN, my father's through (...), 'cause my brother's goin' to beat him so badly, it'll be incredible! He can go straight into the intensive care unit.] 73. Haj'k immer eene jedroehnt jekrigt. (28a/536) [I always got clobbered.] 74. Pass auf, wenn du bei mir'n Funken rauf, krist paar jeknallt. (28a/565) [Watch it, if you ..., or you'll get pasted a couple.] 75. Ja, da kriste aber eene uff de Fresse. (29a/373)

Berlin Style and Register

93

[Yeah, you're really gonna get one in the trap.] 76. Muesstest de gleich um de Ohren kriej'n! (29a/177) [You oughta get slapped right off.] 77. Hau ihm ma u ooch von mir eens uff de Schnauze! (30a/58) [Let'im have one in the mouth for me, too!] 78. Schreib richtisch, du Pflaume, sons jibs heute noch Senge! (29b/55) [Write it right, ya nitwit, or you'll get a lickin' today yet.] 79. Lass Lynn in Ruhe, sonst schlag' ick dir'n Zahn ein! (29b) [Leave Lynn alone, or I'll knock your teeth out!] 3.2 The Types of References: "The Target of the Attack" and "The Blow" Our examples show that physical violence is directed primarily to a particular part of the body, the head. This involves either the head as a whole - Kopp (7, 8, 16, and 28), Buerne (41), and Schaedel (33) - or specific parts of the head. The main target is the mouth, expressed variously as Schnauze (11, 18, 20, 24, 47, 58, 63d, 65, 66, 67a, 67b, and 77), Maul (44, 51, 52, and 56), Fresse (5, 43, 70, and 75), Gusche (69), and, indirectly, Zahn (79). The nose {die Nase) is mentioned several times (54, 55a, 55b, 63a, 63c, and 64).9 Next in line are the ears {die Ohren-. 1, 57, 76), the eyes {die Oogn\ 10), the neck {der Hals-. 15), and the brain {das Gehuern: 3).10 The preference for aiming at the mouth can be explained by its function as the organ of articulation and by the significance it has for the Berliner as the instrument of linguistic skill. The verbal blow to the mouth becomes a symbol gesture the speaker uses to help disarm the adversary, to "silence" the person. The blow(s) are referred to primarily with semantically vague labels. "One" {einej eenej eens-. 14, 21, 30, 40, 41, 55a, 55b, 56, 58, 63a, 64, 68, 69, 70, 73, 75, and 77) ranks above "something" {wat-. 24, 28, 29, 47, and 62), "a few (things)" ('«paar (Dinger): 36, 43, 51, 52, and 62), or "what you're askin' for" (rvat de brauchst-. 13). The stylistically conspicuous metaphor Oelfmeter (literally): "penalty shot" in line 10 and the specific label Ohrfeige ("slap") in line 57 are exceptions. 9

10

Since all the examples are taken from the "bridge discussion", the assumption is that this is due to an "echo effect". It would be interesting to examine a broader data base and check whether this is a dialectal feature of the Berlin dialect, whether the Berliner also makes the so-called four letters the target of the verbal tour de force, or whether the head is similarly preferred in other regional varieties.

94

Norbert Dittmar/Peter Schlobinski/Inge Wachs

3.3 The Verbal Predicate The most frequently used verb variants are ein-/ zusammen) schlagen / sich schlagen (3, 9, 17, 25c, 46, 48, 49, 50, 53, 60, 61, 63c, 63d, and 79), (sich) hauen (5, 6, 7, 12a, 12b, 12c, 16, 18, 20, 21, 22, 25b, 45, 47, 55a, 55b, 56, 57, 58, 59, 63b, 64, 65, 66, 67b, and 77), and (eine/ was) kriegen/ gekricht (1, 2, 10, 13, 14, 24, 28, 29, 30, 33, 36, 41, 43, 51, 62, 63a, 69, 70, 73, 74, 75, and 76). Less frequent are variants like sich kloppen (39), auf jemanden eindreschen (38), sich keilen (25a), jemanden taetlich angreifen (42), jemandem eine scheuern (68), (sich) pruegeln (26, 31, 32, 34), jemandem eine knallen (74), jemandem eine droehnen (30, 73), jemandem eene rinschmoken (2), jemandem eine pellen (11), andjemandem eene ueber^iehen (40). In some examples the act of hitting is expressed through onomatopoetic utterances like KNICK KNACK (4), KLATSCH KLATSCH (35), BAB AB A (37), or BOM BOM (71). 3.4 Predications Having the Noun Schlag As an Object of Reference Predications having the noun Schlag (blow) as the object of reference relate primarily to intensity. In many cases the qualitative specification of the blow is expressed through the description of its effect. In formal terms, this is achieved in several examples by means of consecutive clauses. The consecutive clause relates to the expression standing for the blow in the main clause. It not only expresses a matter following from the main clause; the specification of the sequence primarily serves the step-by-step classification of the blow (see GRUNDZUEGE: 803-4): ... dit's ja nich anders jeht (2), ... dass die ja nich aus de Oogen kieken kann (10), ... dass de nach hinten fliegst (41). In our data, this type of predication is found in conjunction with formulation pattern I (see section 3.5). A step-by-step classification of the blow through the description of its consequences for the victim can also be expressed in terms of a temporal relationship, however. In such a case, the subject matters (blow/effect) are jointed coordinately by an adverbial pronoun (sometimes an elliptical one) communicating posteriority (dann or, in the Berlin dialect, denn). Some examples are ... und denn is allet spaet fuer den (4), ... und denn is Ende mit ihm (15), and ... kann a glei /hin/ u f f de Intensivstation jehn (72). The matter expressed in the second clause implies a "final state" for the victim, a state tantamount to a total knockout. In example (1) and (2), the blow's effect is manifested as an increase in heat. To achieve this, the speakers make use of rhetorical figures of speech. The metaphorical verb reinschmoken (to clobber) used in the second sentence

Berlin Style and Register

95

of our examples triggers a chain of associations that includes the English "to smoke" and "that it smokes". In example (1), the increase in heat is expressed through a figure similiar to metalepsis. This replaces the cause (the slap) with its effects (heisse Obren)-, "hot / that is, stinging / ears"). In example (10), too, predication about the blow is expressed with a rhetorical figure of speech — the metaphor Oelfmeter (penalty shot), which among goalkeepers is the most feared free kick that can be awarded to the opposing team in the game of soccer. The ball, placed just eleven meters away and directly in front of the goal, is kicked with ferocious power and keen precision at the nets — and almost invariably scores. In this example, an additional feature is that the intensity of the act is stressed by a specification of the blow according to the pattern "x is so y that ..." This paints a picture that highlights the impact of the blow: ... dass die ja nich aus de Oogen kieken kann (that she won't see straight). In the hyperbolic threat expressed in example (7), the power behind the act of violence is escalated to the extreme. The speaker does not content himself with wanting to slam his opponent against the wall; he claims that he can slam him right through it. Because of his unbelievable strength, the speaker even feels himself capable of slamming his antagonist through a cement wall. But even that is not enough. The element of strength surfaces again in an intensifying phrase. Though one normally uses an especially powerful drill to bore through a cement wall, the speaker will manage the job with his bare hands thanks to his awesome might: IVir brauchen kein Bohrer! (We won't need no drill!). The same motif is used again by the same speaker in (8) and (27): ... denn jeh ik mit dem durch de Wand, bloss dit den sein Kopp natuerlich vor mein is, wa? (... then I'll take 'im through the wall, 'cept, o'course, that his head goes first, eh?) and Ick geh mit dir durch de Wand. (I'll take ya through the wall.) Predications involving the blow can also have to do with acoustic features, as illustrated with the onomatopoetic expressions BOM BOM (71), KLATSCH KLATSCH (35), and BABABA (37) or the metaphoric verbs knallen and droehnen (30, 73, and 74). A qualitative specification in the form of a comparison was found only in example (38) ... ivie'n Jewehrfeuer (like rifle fire). 3.5 Patterns of Formulation Our data shows that two particular patterns of formulation are especially frequent: ( 1 ) sentence constructions in the Adressatenpassiv (see G R U N D ZUEGE, 1981: 556-557) with kriegen (colloquial for "to get") as a linking

96

Norbert Dittmar/Peter Schlobinski/Inge Wachs

verb or a verb with full meaning and (2) the active-voice construction with the verb hauen (colloquial for "to slam", "to hit").

3.5.1 Pattern of Formulation I {passive recipient with kriegen) This pattern is illustrated in examples 2, 10, 13, 14, 24, 28, 29, 30, 36, 41, 51, 52, 62, 63a, 69, 70, 73, 74, and 75. Just as with pattern of formulation II, one can refer to the: victim (dative) blow (factitive) target (locative) actor (agentive) 11 Beyond these cases, some of the examples have explicit 12 predications (P) concerning the quality or quantity of the blow(s). To join (D) and (F) syntactically, the speaker uses either the verb with full meaning (VI) or the linking verb in the phrase kriegen + verb with full meaning (VI + V2). The pattern of formulation can be described as shown in Figure 3.1. 1 3 "der die du/de olle Klaus se a/er man der Ti mm y -

"u

"

i

E G K íf & U >

§ g > S- §

"g

so eene so 'ne Ölfmeta rinje- \ wat de brauchst schmoktl wat jedröhnt 1 eine jeknallt 1 'n paar Dinga \ : / 'n paar

uff de Schnauze jegn Kopp u f f n Schädel jegn de Bürne uff de Fresse u f f s Maul uff de Nase

/von mir \ von Achim] vom Rudi I

C 0

:

D Dative

V,

v2

F Factitive

L Locative

A Agentive

P Predication

Figure 3.1: Patterns of formulation I {passive recipient with kriegen) 11

12 13

By this term, we mean the case as defined by C. FILLMORE (1968: 24-25): Agentive (A), "the case of the typically animate perceived instigator of the action identified by the verb." Dative (D), "the case of the animate being affected by the state or action identified by the verbs." Factitive (F), "The case of the object or being resulting from the action or state identified by the verb, or understood as a part of the meaning of the verb." Locative (L), "the case which identifies the location or spatial orientation of the state or action identified by the verb." Unlike the implicit predications expressed as rhetorical figures of speech. The description ties in with SANDIG (1978). Parentheses stand for facultative elements, brackets for obligatory.

97

Berlin Style and Register

By definition agentive and dative are characterized by the feature [ + human]. Most of the variants of the factitive illustrate semantically empty types — the exception being the metaphor Oelfmeter. Aside from the word Hackbrett, the elements in the locative all refer to the head or parts of the head of the dative object. Three of the examples of predication concerning (F) are consecutive clauses; one case is an indefinite statement of quantity (reichlich [plenty]). Table 3.1 shows which cases are illustrated in each instance. It presents information on the examples in which predictions (P) concerning (F) are made and shows whether the speaker is identical with the potential actor (sp. = act.) 14 We consider this aspect to be interesting because it directly coincides with an action context with emotional involvement and because one can assume that this is manifested on the verbal level.

Example

2 10 13 14 24 28 29 30 36 41 43 51 52 62 63 a 69 70 73 74 75

(D)

(V,)

(V»)

(F)

+ +

+

+

+ + + +

+

+ +

+ + + + +

+ +

+ + +

+ +

+ +

+

+ + + +

+ +

+ + + + + +

+ + + +

+ +

+

-

+ -

+ +

—

+ + +

+ + + + + + + + + + + + +

(L)

(A)

—

—

-

-

-

+

-

+ +

+ -

+ + + + -

+

(P)

(Sp.=Act.)

+

(+)

+

+

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

+ -

+

-

-

-

-

-

-

-

-

-

-

-

+ -

-

-

-

-

-

-

-

-

-

-

-

—

—

+ +

+

-

+

+

-

-

+ +

Table 3.1: Overview o f the realization o f various cases, predication concerning (F), and the identity o f speaker and actor in formulations with the meaning "A hits B " constructed in the Adressatenpassw 14

with krtegen.

This does not always follow from the examples isolated here, but can be inferred from the context in some cases. In (63a) and (70), the question cannot be answered clearly. The use o f direct speech to reproduce utterances o f third parties is marked with a ( + ) .

98

Norbert Dittmar/Peter Schlobinski/Inge Wachs

As can be gathered from Table 3.1, this pattern of formulation is chosen primarily when the speaker is talking about dealing with a third party in a physically violent way. The dative and factitive cases are obligatory. The semantic vagueness of the verb and the factitive elements suggest an explanation for the fact that both cases are supplemented by at least one other case and/or a predication. It also suggests an explanation for the fact that kriegen is replaced by a linking verb phrase containing a metaphorical verb with full meaning (droehnen, rinschmoken, knalleri). This makes it possible to relieve the polysemy of the phrase der dricht eine through the use of the locative (der kricbt eine jedroehnt). The facultative elements are thus key categories for the specification of the meaning. The deep structure of the formulation pattern can be represented as a case-frame. 15 We will introduce (P) as an additional element: (Part II) + kriegen /

D + F + (A) (L) I (P)

It seemed interesting to us that none of our examples had all the elements (all cases + second full verb + predication). As shown in the summary below, there is instead a tendency toward four semantic "stresses" that can be expanded by predication. We have been able to discern the following constellations: D + VI + V2 D + VI D + VI

+ (P)

One could hypothesize from this that the "normal type" of the formulation pattern has four "stresses", that the pattern is characterized by a syntactic and semantic "rhythm", and that a pragmatic component is inherent in deviations from the "normal type of rhythm". This thesis can be supported by the fact that speaker and actor or potential actor are identical in, for example, formulations with five "stresses" (2, 13, 41). The fifth element is a gradual specification of the blow and involves good chances for image maintenance. 3.5.2 Formulation Pattern II (Active-Voice Construction with hauen) Unlike formulation pattern I, pattern II is formed primarily when the speaker and the actor or potential actor are the same person (see Table 3.2). This is surprising in that the Adressatenpassiv — although it is not oriented to the agent — generally offers more chances for self-portrayal 15

Alternatives are noted in parantheses.

99

Berlin Style and Register

than formulation pattern II. The formulations constructed in the Adressatenpassiv have a number of expressive figures of speech (see droehnen, knallen, rinschmoken, Oelfmeter, and Hackbrett) as well as step-by-step predications about the act of violence. By contrast, there is nothing special about formulation pattern II. It lacks predications concerning (F). Example (7), with its colossal hyperbole, is quite different from the other active-voice constructions (see Table 3.2). If speaker and actor are not identical, the tendency is to reduce the pattern of formulation to a "minimal form", whereas more complex formations are preferred if speaker and actor are identical: Speaker # Actor

hauen /

D + A/

hauen /

D + A + (L) + (F)

Speaker = Actor

Ex. 5 6 7 16 18 20 21 22 25 b 45 47 55 a 55 b 56 57 58 59 64 65 66 67 a 67 b 77

(D)

(V)

+ + + + + + + + + + + + + + + + + + + + + + +

+ + + + + + + + + + + + + + + + + + + + + + +

(L)

(A)

—

+

-

-

+ + + + + + + + + + + + + + + + + + + + + + +

(F)

?

-

+

+ + + -

-

-

-

-

-

+ + + + + +

•

+ + + •

-

+ -

+

-

+ + + + +

(P) — -

(Sp. = Act.)

(+)

(+) + + + + +

-

-

-

-

-

-

+ + + + + -

+ + + + +

—

Table 3.2: Summary of the realization of various cases, predication concerning (F), and the identity of speakers and actor in active-voice constructions with hauen.

100

Norbert Dittmar/Peter Schlobinski/Inge Wachs

3.6 Summary To summarize, it can be said that statements about the use of physical force make an interesting field for contrastive studies in dialectology. FILLMORE'S case grammar and SANDIG'S method of describing style could be helpful analytic instruments. Although the illocutionary aspect of the formulations we have presented could not be included within the scope of the present work, we did manage to elaborate the propositional aspect by examining reference and predication. It was shown that the discreet kinds of reference have a certain typology (vagueness of the nomenclature for the reference type called Schlag [blow] and parts of the head for the reference type called "target of the attack"). Predications relating to the blow primarily communicate its effect on the victim. They are manifested as rhetorical figures of speech, statements of amount, consecutive clauses, temporal relationships, or onomatopoeia. The verbal predicate has a broad range of variants, the most dominant of which are /ein-¡^usammen-)schlagenjsich scblagen, (sich) hauen, and (eine/wai) kriegenjgekricht. Aided by case grammar, we were able to move beyond the differentiation between reference and predication and delineate two patterns of formulation in universal and abstractive categories that may be useful in contrastive analysis. To analyze the illocution of the formulations presented in this article, one must include the situational context. Beyond that, such analysis would have to consider both the type of sentence involved and the speaker's perspective on what is happening (reflected in questions like "who is hitting whom?" and "what is the relationship between the speaker and the actor or victim?").

4. Semantic Deviations and Rhetorical Figures of Speech 4.1 The Verbal Creativity of the City Dweller As often pointed out, particularly in popular publications, the linguistic creativity of the Berliner is almost inexhaustible. Oft cited examples of the Berliner's linguistic creations are the original nicknames given to public buildings and squares — "the pregnant oyster" (die schwangere Auster) for the Congress Hall, "Mt. Rubble" {Mont Klamotte) for the Islander, and the "hay rake" {die Hungerharke) for the memorial to the Berlin Air Lift to name just a few — and folk etymologies like "moth news" {Mottenpost for the daily paper known as the Berliner Morgenpost, "singing epidemic" {Singepidemie) for the Singakademie, and Null u f f s Pferd for the expression

101

Berlin Style and Register

Null ouvert in the game of skat. These expressions have meanwhile become firmly anchored in the regional lexicon. Linguistic creativity, the unconventional use of a language system's building blocks, is a special feature of the city dweller, who is even more burdened by norms and regimentation of various sorts than the inhabitant of rural regions. The masterful use of language as a system of symbols, the playful breaking of conventional and thus seemingly inviolate rules is interpreted by SORNIG (1981: 73) as an expression of protest against linguistic and social norming. In urban communities, in contrast to rural areas, a whole range of differing rules and standards of behavior, linguistic and otherwise, are in existence simultaneously, and if there is one thing that characterizes the urbanite, it is his skeptical attitude towards traditions and normative rules and, above all, the delight he takes in innovation and rulebreaking where — for various sociological reasons — confidence and belief in norms becomes doubtful, and it is especially the poet and the proletarian who may boast of their dynamic handling of rules and a clownesque courage and liberty not to abide by them. (SORNIG, 1981: 71-72)

The semantic and idiomatic deviations and rhetorical figures of speech we have compiled are excerpted from our own data base and from SCHLOBINSKI (1982). They document the pleasure that the Berliner takes in linguistic alienation as we routinely encounter it in everyday situations. 4.2 Semantic Deviations This group includes (a) neologisms, (b) terms that take on a different meaning (extension of the semantic field) when used unconventionally, and (c) variations in idiomatic phrases and routine formulations. Table 3.3 presents individual lexical elements, their meaning(s), and each utterance's corresponding context. 4.2.1 Neologisms Lexical Element

Meaning

Context

ausbuchteln

to write an essay (with much effort)

Ick wtrd den Nerv wahrschtinltch nie habm, mtch bins^uset^en, da wat Vanuenftjet aus^ubuchteln. (19a/63) [I'll probably never have the nerve to sit down and sweat it out writing something reasonable.]

Norbert Dittmar/Peter Schlobinski/Inge Wachs

102

Lexical Element

Meaning

Context

austutschn

to relax together over a nice bottle of wine

Da unten steht fuer Sie 'ne K eh Kiste Wein. Koenn Se jam^jemuetlich austutschnn mit Ihrem Mann! (23b/504) [There's a case of wine down there for you. You and your husband can relax together over a nice bottle of wine.]

Dreckbuddl

cleaning woman

Wir sind fuer andre keen Dreckbuddl! (22a/582) [We're not a cleaning woman for other people!]

Glueschn

(tired eyes)

Dann hah ick nur \rvee Stundn jeschlafen, steh'k mit so'ne Glueschn erst ma u f f . (7a/313) [Then I slept for only two hours, have such tired eyes when I first get up.]

hoehleti

to sleep in a (cramped?) sleeping compartment on a train

Jetn^ habm wir da noch drinne jehoehlt, in unsern Schlafwagen in Braunschweich. (4a/159) [Now we were still sleeping in there in our sleeping compartment in Brunswick.]

Jedaechtniswaermer

beret (literally: memory warmer)

Na jan% schlimm is ja so'n Jedaechtniswaermer! [Well, a beret like that is really bad!]

Kalmppe

run-down building, a real dump

Dit is ja OOCH da druebm so'ne Kalueppe. 921) [That's a real dump over there, TOO.]

kramich

messy, untidy, sloppy

Dit is kramich! (—) [That's too messy!]

mohndoof

extremely dimwitted

Noch dit Stueck [Mohnkuchen], denn is a mohndoof! (14b/247) [One more piece of [poppy-seed] cake and he'll be as dumb as a poppy seed.]

Muerbchen

especially tender meat skewer

" Fleischspiesse" duerft ick ga nich da%u sagen, ick muesste MUERBCHEN dazu sagen! (3a/107) [I wouldn't just call 'em meat skewers, I'd have to call 'em HEAVENLY!]

Platschpampe

slush

Wat hier [in Berlin] runtakommt, is ja nur Platschpampe. (16a/106) [The stuff that falls here (in Berlin) is only slush.]

raffitueckisch

clever, slick, extremely shrewd

Dit is ja raffitueckisch! (PS/113) [Boy, that's really slick!]

rinschmoken

to clobber someone

Die kriejn so eene rinjeschmokt, dat's ja nich andas jeht. (22a/322) [They'll get clobbered so hard that they won't know what hit 'em.]

(8a/

103

Berlin Style and Register

Lexical Element

Meaning

Context

stremplich

(in some way or other) not in order

Wenn se [die Rose] naemlich stremplich waer, haett ick se rausjeschmiss', verstehste, ick lass doch nuescht Stremplichet stehen. (PS/202) [If somethin' weren't right with it (the rose) somehow, I'd throw it out, ya know? I don't let anything stand around if there's somethin' wrong with it.]

vollklarren

to smear (smudge) the walls

Is schon allet wieder volljeklarrt und allet. (5b/ 582) [It's all completely smeared and everything AGAIN already.]

Table 3.3 Neologisms in the Berlin dialect

4.2.2 Extension of the Semantic Field The terms in Table 3.4 do not represent newly coined expressions; they are rather expressions whose "normal" meaning has been altered as a result of an unusual context. The extension of meaning almost always coincides with the use of a rhetorical figure of speech.

Lexical Element

Meaning

Context

'ne Mark abdruecken

to pay a pretty penny (literally: to make an imprint of a mark)

Da musste aber 'ne Mark abdruecken! (19a/123) [You've got to pay a pretty penny for THAT!]

'n Abjang machen

to fall down (literally: to make an exit)

Und mein Mann hielt mich dann, beziehungsweise die Leiter fest, damit ick nich 'n Abjang mache. (14a/510) [And my husband held me then, or rather, the ladder, so that I wouldn't fall.]

ausmaerts reden

to speak another language; not to speak German (literally: to talk abroad)

Als wenn ick auswaerts rede. (5a/165) [As if I were talkin' Greek.]

Bereifung

nerves (literally: set of tires)

Denn sind Se meistns so fertich uff de Bereifung. (161/636) [Then their nerves are usually pretty shot.]

104

Norbert Dittmar/Peter Schlobinski/Inge Wachs

Lexical Element

Meaning

Context

Biefies

inspectors of the Berlin City Transit (BVG)

Wo die jan^en Biefies rumbaengen. (15a/18) [Where all the "beefies" stand around.]

Bimbos

foreigners, nonwhites

Da kraueben mehr Bimbos rum wie Franzosen. (4a/159) [There're more coloreds crawlin' around there than French.]

droehnen

to lambast someone (literally: to boom, to resonate)

Haj'k imma eene jedroebnt jekricbt. (28/536) [I always got lambasted.]

Elefant

a big fuss (literally: elephant)

Das is also hier auch schon passiert, bloss da wuerd nich so 'n Elefant draus gemacht, (llb/633) [So that's also happened here; it's that no one made such a big fuss about it.]

erschrecken

to intimidate, to deter (literally: to frighten, to startle)

Eine erschreckte Wiener! (3a/107) [A frightened Viennesel]

'n Ganta

a tough goose (literally: a gander)

Wenn Se so'n Ganta erwischen, der da knallhart is. (16a/400) [If you get a goose that's tough.]

Jahreswagn

a jalopy (literally: car o' the year)

Koofen wa wieder 'n Jabreswagen! (14b/262) [Let's buy an old jalopy!]

jdm. de Karten lejn

to threaten someone (literally: to show someone one's cards)

Den bob ick jet% de Kartn jeleecht. (22a/262) [I've now laid it on the line to him.]

Klamotten

large boulders, stones (literally: junk, rags)

Chinesische Maua bochjeklettat, weesste, wo die Klamotten da rausbaengn. (14b/628) [Climbed the Great Wall of China, ya know, where the huge boulders jut out.]

Kompott

work that remains to be done after hours (literally: stewed fruit)

Dit haengt als Kompott hinten noch mit dranne. (16a/519) [That's still left to be done as overtime work.]

Kruemel

tiny pieces of meat (literally: crumbs)

Je mehr Fleischstueckn der [Schinken] hat, desto besser is a. Wenn a 'tuerlicb nur 'n paar Kruemel hat, denn koenn Se 'nglei wegschmeissen! (16a/314)

105

Berlin Style and Register

Lexical Element

Meaning

Context [The more pieces of meat it (the ham) has, the better it is. If it only as a couple of tiny pieces, ya c'n throw it right out.!]

Lauser

turnip

Eine Mark, so'n Lauser! (PS/136) [A mark; what a turnip!]

Mollekuele

cool (glass of) beer

Noch eine Mollekuele! (3a/107) [Another beer!]

pellen

to let (someone) have it, to slug (literally: to peel)

Halt die Schnauze, sonst pell ick Dir eene! (22a/ 394) [Shut your trap, or I'll let you have one!]

'n Pieps

a very small amount (literally: a chirp)

Wenn da so 'n Pieps [ungeloescbter Zement] in de Oogn kommt, na da is die Netzhaut vabrannt. (22a/94) [If even a peep (of unslaked cement) gets into your eyes, why, the retina is burned.]

Sit^kissen

Turkish flatbread (literally: seat cushion)

Hier haste 'n Sit^kissen! (—) [Here's a seat cushion for ya!]

Stoss

success (?) (literally: thrust)

Der bat Stoss jehabt bei die Weiber. (5b/500) [He had success with women.]

'n Strumpf

a story, article (literally: a stocking, a sock)

Machen Se ma irjnin netten Strumpf da draus. (16a/600) [Make some nice story out of it.]

trommeln

to brag, to boast (literally: to drum)

Der kann trommeln! (3a/107) [He sure brags a lot!]

ueberyehn

to clout (someone)

Neulich hat se der Vera eene ueberje^ogen. (22a/ 416) [He clouted Vera the other day.]

Ufftritt

to step on someone's foot or toes (literally: to make an appearance)

Dit war 'n Ufftritt! ( - ) [He really stomped on me!]

Table 3.4. Extension of the semantic field as illustrated in the Berlin Dialect.

4.2.3 Deviations in Idiomatic Phrases Our body of data agrees with that of GROBER-GLUCK (1975) in containing a number of supraregional idiomatic phrases. Originally hyperbolic phrases

106

Norbert Dittmar/Peter Schlobinski/Inge Wachs

that have lost their hyperbolic character (such as ick fress 'n Besen auf [I'll eat my hat; I'll be hanged], tausend Wuensche haben [to have a thousand desires]) are exaggerated and brought into use again through new variations (ick fress 'n Nuckel u f f [I'll eat a pacifier], achtun^wach^ichtausend Wuensche habm [to have twenty-eight thousand desires]). Since idiomatic phrases are highly habitualized in a speech community, just beginning such a phrase leads the communication parrtner to expect the utterance to continue in a certain way, an expectation that is not met, however. The result is surprise. (a) (b) (c) (d) (e) (f) (g) (h) (i) (j) (k) (1) (m) (n)

ik jeh hier vor de Mautn. ( ) / (Hunde) Der sah aus wie der Graf von Monte Christo. (3b) ( G r a f Koks) Wie du Schmalz jeraspelt hast! (4a/159) / (Suesshol Wie die alle hierher jefundn habm, dit weiss der Schinda! (5b/242) (Teufel) Rejelrecht war ick u f f de Buerke! (5a/165) / (Palme) Denn fress ick 'n Nuckel u f f ! (7b/52) / (Besen) Du jehst ma u f f n Senkel! ( ) / (Wecker, Geist) Geld looft nich. (PS/116) / (stinkt) Gleiche Laenge, gleiche Welle. (PS/200) / (Stelle) Pass uff, sonst brichste Dir de Ohren! ( ) / (den Hals) Achtundzwanzichtausend Wuensche! (PS/207) / (tausend) Wir habm jedacht, wir schiein! (23b/l 16) / (sehn nicht recht) Da iss die hin und weg. (22a/392) / (hin- und hergerissen; gan^ weg) Der guckt mich ganz schraeg an. ( ) / (schief)

In the following examples there is a reliteralization of the idiomatic phrase; that is, the phrase regains the denotative meaning it lost as its meaning eroded (COULMAS, 1 9 8 1 ) . (o) Die is schneller weg vom Fenster, als se ueberhaupt durchgucken kann. (p) A: "Gut'n Rutsch!" B: "Rutsch nich aus!" (16a/94) (q) Said as a man hands a woman the bread basket at breakfast: Darf ick dir 'n Korb geben? When greeting ceremonies were involved, we observed that the routine formulation, guten Tag/guten Morgen was varied, too — in a kind of "game" in which the person being greeted not only repeated but outdid the communication partner's salutatory formulation (SF) by adding at least one modifier to it ( C L A R K & C L A R K , 1977). In the process, the following pattern emerges:

107

Berlin Style and Register

1. Speaker:

2. Speaker:

( = modifier + SF 2 ) Sp3

SF]

SF2 ( = modifier + SF^

SF 4 ( = modifier + SF 3 )

Here are a few more examples: (r) A: "Gut'n Tach!" B: "Ein recht schoen gut'n Tach!" (3a/107) (s) A: "Schoen gut'n Tach!" B: "'n wunderschoen gut'n Tach!" (3b) Going your partner one better is only part of the game, however, as example (t) shows: (t)

A: B: A: B: A: B: A: B:

"Mojn!" "Gut'« Morgn!" "Einen vfutiAetschoenen gut'n Morgn!" "Einen recht schoenen wunderschoenen, na Mensch, sach bloss, du hast jut jeschlafen heut!" "Mornin'!" "Good morning!" "A wonderful good morning!" "Very good, wonderful, hey man, say, you slept well last night!"

4.3 Rhetorical Figures of Speech 4.3.1 Hyperboles The Berliner's "loud-mouth bluster" is illustrated by a good number of hyperbolic phrases in which certain characteristics like the appearance (see also 4.3.2), quality, or size of persons or objects are exaggerated to the point of being preposterous if the utterances are taken literally: (a) Quantity A salesman's praise of his lemons: Die sind jan% saftich, kommt een Eimer raus. (PS/141) [They're really juicy; you'll get a bucket from 'em.] Judgment about a type of apple: Da habm se festjestellt, dass die nicht ein Vitamin enthalten, nur Wasser. (PS/203) [They've found out they don't have a single vitamin only water.]

108

Norbert Dittmar/Peter Schlobinski/Inge Wachs

Salesman to a child: Und du kommst wieder mit achtundzwanzigtausend Wuenschen an. (PS/207) [And you're here again with twenty-eight thousand wishes.] Said about a pastor's drinking habits: Der bat gleich kranenweise Schnaps jekooft. (23b/116) [Right away he bought more booze than a crane could move.] Said of the vacation weather: Dit hat so jeregnet, wir dachten, wir kriejn Schwimmhaeute. ( ) [It rained so much, we thought we'd grow webbed feet.] (b) Size A salesman about the size of the moss roses he is selling: Dit sind Baumstaemme. (PS/178) [Those are tree trunks.] Response to a question about the cost of ornamental branches: Wenn Se nun een so'n Baum kriejn, denn muss ick Ihnen drei Mark f u f f c i c h abnehmen. (PS/111) [If you get a tree like that, I'll have to charge you three Marks fifty.] Conversation about rapidly growing geraniums: Dit sind richtje Buesche. ( ) [Those are outright bushes.] (c) Length of Time Said about tough Christmas geese: Wenn Se da so'n Ganta erwischen, der da knallhart is, den koenn Se vier Wochn braten, der wird nie weicher! (16a/400) [If you get a goose that's tough, you c'n cook it for four weeks and it won't get any more tender!] Dissatisfaction with a particular state of affairs can also be manifested as hyperbole, with the lexeme schmeissen (to throw, to chuck) being a characteristic element: (d) Expression of Dissatisfaction Said about the mandatory exchange of money required of foreigners upon entering the German Democratic Republic: Dit Bloede is natuerlich da dranne, wa? Ditte praktisch, wenn de jet^t da rueber faehrst, wa, fuemun%wan%ich Mark wegschmeissn kannst. (14b/486) [That's the dumb thing about it, o'course, huh? Every time ya go over there now, eh? (They think) ya c'n just throw away twenty-five marks.]

Berlin Style and Register

109

Said when the oven does not work at Christmas of all times: Schmeiss ick se aus'm Fenster und jeh essen. (15b/320) [I'll chuck it (the Christmas goose) out the window and go out to eat.] Should the neighbor again dare to park his car in the wrong spot: Denn ruf ick'n Abschleppdienst und lass'n in die Spree schmeissen. (15b/471) [Then I'll call to have it towed away and thrown into the river. Barkeeper telling what a butcher can expect if he delivers inferiorquality products: Dem wuerde ick diese Po^jon knallhart zurückschmeissen, uebern Tresen, in's Jesicht! (16a/314) [I'd chuck this cut over the counter right smack back in his face!] Said of inferior-quality ham: Wenn er nur'n Kruemel hat, koenn Se'n gleich wegschmeissen! (16a/314) [If it has only a few small pieces, ya c'n chuck it out right away!] 4.3.2 Figurative Comparisons One of the things to which similes and metaphors relate is the speed of occurrences. Speaker reporting how he hurried to prevent an accident: Ick bin jesprungn wie 'n Hase. (22a/102) [I jumped like a rabbit.] Speaker describing that he would leave Berlin if he were young again: Wat mein Sie, wie ick abhaun wuerde! Die die Socken qualm! (5a/562) [You've no idea how I'd beat it! (So fast) my socks would be smoking.] If the metaphors and similes relate to the external appearance of persons, they can be interpreted as the expression of a humorous attitude toward human "fate" (corpulence, disease, poverty). An obese woman describing the figure she cuts when washing windows: wie Ballarina auf der Leiter: (14a/510) [like ballerina on a ladder.] Said about a sick child: Der sieht aus wie der Tod uff Latschen.

(

)

[He looks like death warmed over.]

110

Norbert Dittmar/Peter Schlobinski/Inge Wachs

Said about an underweight infant: Der sah aus wie 'ne abjezogne Katze. (3b/440) [He looks like a skinned cat.] Said about a mortally ill daughter: Die sah aus wie Gratula (Dracula?) (3b/440) She looked like Gratula (Dracula?).] Said about a woman ill with mumps: Die sah aus wie 'n Ostaei -so mit Schleife so. ( ) [She looked hlike an easter egg — with a bow and all.] Woman speaking to a buxom girl: Du Futter, wa? ( ) [You're also well fattened, aren't you?]

stehst ooch janz jut im

Conversation about the people in Bolivia: A: "Die tragn Hemden, die sind voellig abgetragen und zerrissen." [They wear shirts that are totally worn out and torn.] B: "Die tragen Bruesseler Spitzn!" ( [They wear Brussels lace!]

)

Some of the metaphors used in conjunction with the similes are characterized by grotesque combinations of fantasy like Ballarinaj Leiter and Todj Latschen. This is also exemplified by the description of the smell caused by kitchen waste that has been lying around for several days: Denn stinkt dit wahrscheinlich 333)

wie Jaensebraten mit Pflaummus. (15b/

[Then it probably reeks like roast goose with plum jam.] 4.3.3 Diaeresis Last but not least, the Berliner's rhetorical skills are expressed in diaeresis, piling up words in a certain way. In rhetoric, diaeresis is understood to mean the division of a main concept into numerous syndetic or asyndetic subsumptions, which can be recapitulated at the end by the generic term. With this rhetorical figure of speech, the meaning of terms like alles (all, everything), nichts (nothing), and viel (much, a lot, a great deal of) is specified. (a) jab's keen Zucker, keen Milch nuescht i>u essen. (3b/449)

Berlin Style and Register

[There was no sugar, no milk nothin'to eat.] (b) wenn bei mir 'ne Birne kaputt jeht, dann muss ick een komm lassen, wenn an mein Auto wat kaputt jebt, dann muss ick een komm lassen. Ick muss imma een komm lassn. (PS/121) When a light bulb burns out in my place, I have to have somebody come; when something goes wrong with my car, I have to have somebody come; I always have to have somebody come.] (c) Wir komm vom Schlafzimmer in Gartn, wir komm vom Kinderzimmer in Gartn, wir komm vom Bad in Gartn. (14a/488) [We come from the bedroom into the yard; we come from the children's room into the yard; we come from the bathroom into the yard.] (d) Sie sind doch Dame, Sie habm doch Jedankn, Sie sind doch denkend. (PS/147) [After all, you're a lady; you do have thoughts; you DO think.] (e) Fuer Sie is et nich zu schade, fuer Ihrn Sohn is et nich zu schade und fuer Ihrn Mann is et nich zu schade. (PS/122) [It doesn't matter for you; It doesn't matter for your son; and It doesn't matter for your husband.] (f) Pappa, kuck mal die Knospen. Pappa, kuck mal da'n Kaefer, Pappa, kuck ma! (PS/143) [Papa, look at the buds; Papa, look there, a beetle; Papa, look!]

112

Norbert Dittmar/Peter Schlobinski/Inge Wachs

(g) Denn weess se erst in wat ick schwach bin, wat ick jut bin, wat Sache is. (7a/271) [Then she'lll really know what I'm poor at, what I'm good at, what's what.] (h) Dit se dit abms und nachts, und mitternachts, dit stoert denn imma! (22a/269) [The fact they (do) it in the evening and at night, and at midnight, that's a constant bother!] (i) Es gibt nichts, die habm nuescht, die kriejn keene Seile, die kriejn keene Haken, die kriejn keene Karabiner, die kriejn nuescht. Dit habm se nichl (14b/586) As figure 3.2 shows, example (i) is an especially interesting example for its style. The structure is characterized by a combination of several rhetorical figures. This example was formulated during an argument about wether a person could obtain mountain climbing equipment in the German Democratic Republic. Our compilation of semantic deviations and rhetorical figures of speech gives an idea of the vitality of Berlin wit as it is experienced in everyday situations. With the playful manipulation of a seemingly fixed system of rules called language, the Berliner can still verbosely refute the current thesis that his idiom is dying out.

Berlin Style and Register

Figure 3.2: Illustration and examples of a collective concept («»«¿•¿//nichts) [nothing] for rhetorical support of an argument

Chapter 4 NORBERT DITTMAR/PETER SCHLOBINSKI/INGE W A C H S

Components for an overarching theoretical perspective in sociolinguistics 1. Introduction In a working paper on the theory of linguistic variation ("Linguistic Analysis of the Three Kinds"), R A L P H FASOLD ( 1 9 8 4 : 4 ) points out what many sociolinguists consider to be a crucial problem of this young discipine — the lack of theory. He asks about the foundations for a theory of "communicative competence" (HYMES) and quickly comes to a conclusion applicable to other approaches: ...So far as I am aware, no one has proposed a theory of this kind, although there are sets of intelligently organized observations of various communicative situations. Studies of language variation share, in broad outline , the same outlook. In my judgement, work on language variation is almost completely atheoretical. As far as I know, CHARLESJAMES BAILEY (1973) is the only linguist who has attempt to outline a serious theory of linguistic variation. Bailey's theory was never widely accepted, and Bailey himself seems to have lost interest in it ... Recently, variation studies have in practice given up theory development in favor of the intelligently organized display of observed data.

Discomforted by this state of affairs, some researchers react by critically examining practicable explanatory approaches ( D I T T M A R , 1 9 8 3 ; S C H L O B I N SKI, 1 9 8 7 ) . Others propose approaches like "interpretive sociolinguistics" ( A U E R / D I LUZIO, 1 9 8 4 ) , "connotation grammar" ( B I E R W I S C H , 1 9 7 6 ) , "accommodation theory" ( G I L E S / S M I T H , 1 9 7 9 ) , and "ethnomethodological conversational analysis" ( S T R E E C K , 1 9 8 3 ) . As in the research on second language acquisition, in which contrastive hypothesis, the theory of learner varieties, the identity hypothesis, monitor theory, and pidginization theory are more like labels than tried and tested theories, there seems to be a variety of unconnected sociolinguistic approaches amounting to what, euphemistically, can only be called the discipline's patchwork quilt. The fundamental problem of developing theory in sociolinguistics (formulating an integrated theory about the relationship between a theory of language

Components for an overarching theoretical perspective in sociolinguistics

115

structure and a theory of social structure) were outlined in B E C K E R , DITTMAR, and K L E I N ( 1 9 7 8 : 1 6 4 - 1 6 5 ) , without any attempt to solve them except for the modest appeal to conduct "specific individual studies" (p. 166). The thesis underlying the following comments is that sociolinguistic theory can be developed only to the extent that one manages to build a constructive, interdisciplinary bridge between language systems and language use, between the life world in a specific culture and the institutionalized social system. With HABERMAS ( 1 9 8 1 ) in mind, one may pose a basic question: What meaningful relationship exists between the everyday management and successful handling of tasks or actions and the institutional social forces "interpenetrating" them, creating the vortex of social integration, and intervening a tergo as "second-order" factors in the precursory happenings of "first-order" forces when everyday ways of doing things become inadequate, when alternatives demand decisions, when routines must be questioned and conflicts settled? At first glance it would appear that we are dealing with actions aimed at achieving mutual understanding, actions that are culturally transmitted and renewed. At second glance it would seem we are dealing with forces of social legitimacy that characterize a system of social integration on the basis of normative arrangements that, depending on the relevant sociological category, embrace a "center" and a "periphery" (see K R E C K E L , 1 9 8 3 ) . The complexes labeled "everyday world" and "system of social integration" in sociology evidently involve problems similar to those of langue and parole in linguistics. In sociolinguistics, the obvious counterpart of this contrast is methodological dichotomy: (1) Static categories like "strata", "age", "status", and "network", versus dynamic categories like "discursive negotiation", "organization of taking turns at talking", and "recipient design"; (2) Analysis on the macrolevel versus analysis on the microlevel (on this point see AUWAERTER, 1 9 8 2 ) ; (3) Quantitative, correlative description versus qualitative description oriented to interaction; (4) Observation of the "external perspective" versus observation of the "internal perspective"; (5) "System-related" versus "everyday" (discourse-related) approaches. AUWAERTER reduces these five dichotomies to the difference between microsociolinguistics (first-order factors in the sense described above) and macrosociolinguistics (second-order factors): The two "pillars" are dependent on each other in that the correlative, probabilistic relationships that microlinguistics finds — say, the relationships between social characteris-

116

Norbert Dittmar/Peter Schlobinski/Inge Wachs

tics of the speakers and language use — must have their origins on the microsociolinguistic level and must result from rule systems that govern the use of language as a factor of interpreted and defined speech situations. (Auwaerter, 1982: 7)

Because there is as yet neither a generally valid linguistic theory nor a generally recognized theory of society, we consider the following steps to be useful for bringing together components to construct a sociolinguistic theory: 1. Reconstruct sociolinguistic approaches and methodologies within the theoretical framework of sociology and linguistics; 2. Evaluate available concepts for their explanatory and descriptive value; 3. Critically screen present empirical knowledge with a view to formulating statements capable of developing theory; 4. Outline the foundations of the social use of language and the basic factors governing it. This step should be based on: (a) an interdisciplinary orientation; (b) the integration of matters related to systems and interaction; and (c) the constructive linking of macrolevels and microlevels of description and explanation. Regardless of the integrated theoretical perspective that may emerge, it should account for the difference between the "external" and "internal" perspective of observation and description, between statements about the particular and statements about the general, and it should balance theory and empirical results so that theory guides what might otherwise be "blind" empiricism and so that empirical work compels theory to remain firmly rooted in reality. As a contribution to step 4 (outline of foundations), the attempt is made in the following pages to isolate basic factors governing the social use of language and the features they condition. The following sketch was prompted by the discussion in the second chapter of KLEIN (1984). For lack of space, only the complex called structures of social meaning can be considered thoroughly here, with the other five complexes of factors being referred to only briefly without further comment. (A detailed illustration and discussion of all factors and their subcategories is available as an unpublished working paper.) 2. Factors Governing Socially Determined Use of Language At this point we shall delineate the essential parameters, whereby (1) through (3) are basic factors and (4) through (6) their corresponding

Components for an overarching theoretical perspective in sociolinguistics

117

structural properties. The following outline of the issue is presented as a series of hypotheses; it is neither explicit nor complete or sufficient. In short, it is a prolegomena to expounding a theory.

Outline 2.1 Social Integration 1. 2. 3. 4.

Culture-specific life world Socialization Group affiliation Legitimacy

2.2 Communicative Competence of Social Action 1. Linguistic competence 2. Knowledge of the world 3. Orientations to rational action 2.3 Identity 1. 2. 3. 4.

Motivation Attitude Accommodation Expressiveness

2.4 Structures of Linguistic Variation 1. 2. 3. 4. 5.

Linguistic Choices (Variants) Compatibility (restrictions on co-occurrences) Appropriateness Conflict Distribution of variants in speech

2.5 Regional Distribution 1. Cause and premise 2. Kinds and types of characteristics 3. Dynamics and satiation 2.6 Change 1. Linguistic and extralinguistic conditions 2. Actuation

118

Norbert Dittmar/Peter Schlobinski/Inge Wachs

3. Kinds and characteristics of transitions in time and space The first three parameters will receive cursory treatment. Point 4 is examined somewhat more thoroughly thereafter. Points 5 and 6 are dealt with elsewhere. 2.1 Social Integration If we take "society" to mean "the legitimate social order according to which the participants in communication regulate their affiliation to social groups and thereby assure solidarity" (HABERMAS, 1981, vol. 2: 209), then social integration means the sum of all factors that regulate group affiliation and the claims to solidarity and legitimacy connected with it. We must distinguish social integration that takes place directly in the context of the life world (social action in a real situation or interaction) from an institutional integration that transpires in the framework of society at large. The distinction between a societal integration of society beginning with the orientation to action and the systematic integration of society pervading the orientation to action necessitates a corresponding distinction in the concept of society itself... I would therefore like to propose thinking of societies as being both system and life-world at the same time. (HABERMAS, 1981, vol. 2: 179-180)

2.1.1 Culture-specific Life World Social integration first takes place in actual life worlds. Everyday reality in them is shaped by interactions with family members, relatives, friends, neighbors, business people, colleagues, etc. Integration occurs in discrete, concrete interactions with specific subject matters. Two aspects are decisive: (a) The social ties of a communicative context exist in and are constituted by the diversity of everyday interests and experiences that human beings share with each other —the love of dogs, the fun of playing at the pin-ball machine, the desire for political discussion, and the professional interest in one's work, for example, (b) There are no general regulations or norms that govern integration predictably. Certain varieties are typical for integration specific to a given culture. The vitality of a specific milieu is important to the preservation of local and regional varieties. It makes a key contribution to the cultural reproduction of society. "The cultural reproduction of the life world ensures that newly arising situations are coupled in the semantic dimension to the existing conditions of the world. It assures the continuity of transmittal and a coherence of knowledge sufficient for everyday praxis in each specific instance" (HABERMAS, 1981, vol. 2: 212).

Components for an overarching theoretical perspective in sociolinguistics

119

2.1.2 Socialization Socialization is an integrative and functional part of reproduction. For the members of a life world, it ensures that newly arising situations are coupled in the dimension of historical time to the existing states of the world. It assures for future generations the acquisition of generalised competences for action and provides for the harmonization of individual life histories and collective ways of life. Interactive abilities for and styles of leading one's life are gauged according to one's soundness of mind. (HABERMAS, 1981, vol. 2: 213)

Three features characterize socialization as a decisive process of the individual's integration into the life world or social group: (a) the individual's dependency and need for instruction, (b) adjustment to the expectations and attitudes of the primary reference persons, and (c) the learning and internalization of social competence and skills through language. The acquisition of the habitus (the linguistic habitus as well) becomes irreversible in the process of socialization. This includes such aspects as pronunciation, strategies of speech perception and processing, the interplay of linguistic and nonlinguistic behavior, and the emotive meaning of words and speech acts. 2.1.3 Group Affiliation There are many different reasons for the need to be integrated into a group: safety; solidarity; cooperation; representatation of one's interests; and protection against attack, disorder, and insecurity. It appears to be a prototype of social behavior. Group affiliation is determined a priori by birth, family, and relatives and a posteriori by the individual's specific choice of identity. As a rule, individuals have multiple group affiliations that can be described with the attributes of center versus periphery, vital interest versus instrumental interest, and active versus passive. Primary group affiliation is based on personal presence in group interactions, active participation, and vital interest. It exerts upon the group member's linguistic variety an abiding influence that can be equated with an emotionally laden social vortex. Secondary or tertiary group affiliation normally has much less direct impact on behavior in general and language behavior specifically. Choice can be more conscious, planned, and rational because the direct pressure of the interaction situation does not exist. The choices are thus more instrumental in nature. That is, they are shaped more by profession, status, and other institutional concepts of norms, values, and utility operating in society at large and are therefore less geared to integration in the life world than to integration in the existing, legitimate order.

120

Norbert Dittmar/Peter Schlobinski/Inge Wachs

2.1.4 Legitimacy Social integration without the claim of validity is unthinkable. What is accepted or stigmatized depends on what is permissible in a group or a certain social system. In other words, it depends on the political, legal, institutional, interactive order laid down in written or oral form for a specified period. In the sphere of the life world, this includes such things as table manners, etiquette, mating behavior, and rules of conversation (who is allowed to talk how, when, where, how long, and with whom). In short, it includes the conventions that groups set for interaction taking place during their activities. The establishment of legitimate order, however, only appears to be independent in the life world in that situations and interactions are ephemeral. They are pervaded and partially determined by values and norms laid down by a political, legal, and institutional system of society. To put it in rough terms, actions can be accorded legitimacy in two ways. First, in all cases codified in legal, institutional, and, in general, written form — as a rule these involve public affairs — a formal procedure with relatively set rules decides what is allowed and what is not. This goes for all institutions. In terms of the sociology of language, this means, for example, that an oath in a guest worker's German is invalid, a lecture delivered in a dialect is not accepted as a scientific contribution, school education conducted in Catalan is forbidden in Perpignan, and that an everyday story is admissible as evidence in court only if it provides information on which a verdict can be based. Second, there is a domain that is not codified so strictly in the same way, but principles of material exchange make it just as binding. That domain is the market and its system of material and symbolic values. Money, at least in the thesis of some sociologists like BOURDIEU and HABERMAS, has replaced numerous fields of interaction with symbolic ersatz actions. Many social sectors of material and cultural reproduction have meanwhile taken on prevailing market principles having a latent pull of social integration all their own. The linguistic market is an indirect controlling factor of social integration. It regulates the symbolic value of the varieties and indicates the cost of investment in a particular type of social integration. For one person it involves nine years of schooling and a tolerance for frustration; for the other, the prostitution of an acquired communicative identity in department stores and on the intermediate level of the sales representative. Legitimacy is an invisible guardian of upward mobility, social recognition, and prestige. The conditions of access to social positions are deter-

Components for an overarching theoretical perspective in sociolinguistics

121

mined in relation to the language legitimized by the market. Professions, paid work, and social positions can be grouped according to what they demand in the way of communicative skills and characteristics of linguistic varieties. Educational institutions are especially effective controllers of principles on the language market in that they filter access to social positions. Social categories like "prestige", "strata", "status", "sex", and "age" are closely related to "legitimacy" because their various natures define the center and the periphery on an axis labeled legitimate-illegitimate. Finally, everyday interactions specific to a group involve the private validity of values. This exists in long-term processes of group dynamics, often also being negotiated first in the communicational situation itself. We shall use the term "norms" to refer to systematic values (those values operating in public) and the term "maxims" to refer to private claims of validity (those associated with the life world). The former are legitimate or illegitimate; the latter, acceptable or unacceptable. 2.2 Communicative Competence of Social Action Objectives and purposes of verbal communication are embedded in social action and its interactive coordination. In order to understand and apply the functions of language's symbolic meaning and the social roles they convey and in order to anticipate them in mutual expectations of action, one must basically presuppose the existence of linguistic competence, social and pragmatic knowledge of the world, and a rational orientation to action. Only when combined do these three components constitute competence for social action. 2.2.1 Linguistic Competence Competence for social action does not just presuppose the acquisition of language; language acquisition is one of the key conditions for such socialization. Language acquisition occurs in complex operations of language processing that makes the production and understanding (reception) of utterances possible — relative to biological dispositions and the access of linguistic data. The acquisition of language by the child (or by the learner of a second language) requires the acquisition of (a) a number of discrete phonological, morphological, syntactic, and semantic rules, (b) the coordination of these rules on the various linguistic levels, and (c) the application of these rules in appropriate contexts. For the sociolinguist, the most important fact is that the acquired linguistic knowledge, both

122

Norbert Dittmar/Peter Schlobinski/Inge Wachs

active and passive, is accessible to different degrees and comes to be used in different ways accordingly. In various theoretical approaches sociolinguists talk a great deal about the "social meaning" of linguistc structures and their significance for communicative action. Unfortunately, no work as far as we can see addresses the question of where the "knowledge of variation" (see section 2.4) is seated in the psycholinguistic sense. Most naive statements hark back to the notion established by W H I T A K E R and others that linguistic skills (including those for the variation of linguistic structures) are united in the left half of the brain. We wish to contrast this view with L A M E N D E L L A ' S hypothesis, which we cannot go into further here, however. L A M E N D E L L A (1977) notes that neurolinguistics, following in the wake of Chomskian theory, is too fixated on studying the psycholinguistic moorings of grammar. The new approaches in pragmatics, functional and relational grammar, and in discourse linguistics call for a reorientation of research focused exclusively on the brain's neocortical system. L A M E N D E L L A ' S assumptions seem to be exceptionally important for linguistics in that they deal with the relationship between propositional content on one hand and connotations and pragmatic behavior competence on the other: The obvious candidate for the level of brain activity likely to be responsible for the bulk of nonpropositional human communication is the limbic system. This forebrain network of cortical and subcortical structures has often been thought of only in relation to its regulation of emotion and motivation, but in fact its range of functional responsibilities is quite large and includes major segments of our social and communicative behaviour. ... The majority of the limbic functions acquired by the child remain part of the adult communication repertoire as a neurobehavioral framework into which linguistic comunication is embedded. There can be no doubt that the limbic system plays an important role in human communication of all types, (p. 159) Speech is normally embedded within a matrix of behaviour patterns regulated by the limbic system. In decoding a speaker's message, the addressee consciously and unconsciously attends to these nonlinguistic behaviours in order to extract the entire message. (p.206) In considering limbic and language processing from a neurofunctional point of view we see that the two domains tend to be complementary. Higher neocortical systems develop new functions that augment the older limbic functional repertoire, but there is also overlap, so that the same function may be carried out by more than one system, albeit in different fashions. Much need not be encoded propositionally because it is already being expressed through the limbic system, and some functions are implemented preferentially at one level rather than another, (p. 209)

The biological foundation suggested here seems to be relevant especially for the interplay between communicative and sociopsychological factors.

Components for an overarching theoretical perspective in sociolinguistics

123

2.2.2 Knowledge of the World By "knowledge of the world" (see CARNAP, CRESSWELL, or STEGMUELLER, among others), we mean the social, linguistic, and cultural knowledge that enables an individual to act competently and to coordinate his own action with those of others rationally and purposefully in concrete situations in a specific local radius. We can distinguish linguistic, pragmatic knowledge, cultural knowledge, and social knowledge, with the lines between these stores of knowledge being fluid. To a certain extent, communicating for the purpose of coordinating action requires mutually shared knowledge. To establish a division of labor and to plan work, for example, one must lay down common local and temporal origins, for instance. Types of social deixis are grammaticali2ed, too (BROWN & LEVINSON, 1 9 7 8 ) . Of course, the competence for communicative action is not restricted to the knowledge of deictic relationships. A central question is which interactive and social meaning utterances have. That which is meant but often not said can be inferred from implicatures. There are many kinds of inferences with which one bridges gaps between what is said and what is meant, and they largely depend on cultural and social values and norms. Part of the communicative competence for action is thus the competence to make a number of linguistic choices that have social meaning (see section 2.4). The hypothesis seems plausible (see section 2.2.1) that the "knowledge of variation" linked with these choices is to be attributed to the limbic system (lateralized system of the right hemisphere) and not to the lateralized system of the left hemisphere (LAMENDELLA, 1 9 7 8 ) . 2.2.3 Orientations to rational action Another part of the capacity for communicative action is the ability to distinguish true from false and valid from invalid. These sorts of aspects play a key role in the linguistic activity of argumentation as social action. Granted, it is difficult to separate logical (rational) from strategic (pragmatic) content in argumentation (see K L E I N , 1 9 8 0 ) , but we consider the critical, rational components of action as relevant for sociolinguistics in that communicative actions are often oriented to social meanings disassociated from propositional, cognitive, and logical content. We wish to forward the thesis that the immediate interactions in the life world, as long as they cannot be isolated from each other rationally and cannot be examined for their content, remain "blind" to the communicative decoupling of fundamental needs and socially imposed demands.

124

Norbert Dittmar/Peter Schlobinski/Inge Wachs

If in the sociolinguistic context we bring in a competence called "critical rationality", we mean, in HABERMAS' terms (1981, vol. 1), the basic capacity: (1) to distinguish a message's content from social connotations of the utterance itself; (2) to differentiate the communicative functions o f a variety from its market value (prestige versus stigmatization); (3) to examine, take, and correct conversational turns (turntaking, type and length of the contribution) for their mutual claims to validity and legal right, be they connected with status, sex, age, or unequal competence to make use of knowledge; (4) to examine norms and standardization of linguistic varieties for their underlying bases and rational contribution (effect, utility) and, if necessary, to change them; (5) to select any kind of mode, style, or variety for use with a given topic, interaction partner, situation, overriding sociocultural context, etc., for the focus of a metacommunicative discourse. Naturally, sociolinguistics is primarily descriptive rather than prescriptive. In an expose about ecological sociolinguistics (DITTMAR, 1987 a), we consider the critical, rational components of the competence for action to be essential for imposing rational control on decoupling the linguistic market system from the use of varieties in the life world — a disengagement that the media, bureaucratization, and formal legal claims have already pushed into an advanced stage. 2.3 Identity The individual's counterweight to the social pull of group affiliation and social integration is his ego-stabilizing identity, which subjects his action to the polarity between developing the full potential of his personality and adjusting to society and which characterizes the entire range of emotional and rational states with which the individual identifies in his striving for continual self-experience. Linguistic identify is a symbolic expression of personality identity or group identity. It depends greatly on socialization, during which the individual's habitus takes shape. In encounters with the environment, the personality forms with a motivation commensurate to the individual's needs, with specific attitudes, strategies of adjustment to the social setting, and with an unmistakable expressive mode of behavior. Identity thus has four central functions for the personality or group: (1) drive for action,

Components for an overarching theoretical perspective in sociolinguistics

125

(2) preference for certain affective and cognitive dispositions and immunization against others, (3) balance between the interest of the individual and those of society, and (4) satisfaction of the need for individual-specific or group-specific expression. 2.3.1 Motivation "Motivation" is understood in general to mean that set of psychic forces that act upon the human organism to trigger and perpetuate behavior and orient it to a specific objective. Motivation thus has a drive function (psychic energy), a control function, and a directional function for action and behavior. There are many types of play involving intrinsic and extrinsic instrumental and integrational and integrative motivation that have the effect of, for example, making the individual want to take the initiative in discussion; behave in a cooperative manner; choose styles establishing distances or intimacy; pick up on, develop, or reject particular topics; learn a second language; or stop speaking one's mother tongue or dialect. 2.3.2 Attitude "Attitude" refers to a person's learned, relatively stable disposition or willingness to react to an object (a thing, person, or idea, for instance) with certain feelings, be they positive or negative, perceptions, notions, and behavioral patterns. One thereby refers to the feelings toward an object as the affective component of attitude; the perception, notions, ideas, etc., of the object as the cognitive component; and the behavioral tendencies toward the object as the conative or action component of attitude. Most of the time, the affective component is considered to be central. (FUCHS et al., 1978: 179)

If A is confronted with sentences of the variety X and variety Y and if A utters the propositions "X is good" and "Y is bad", then we are dealing with the "affective component" of attitude. It is a matter of taste in that all languages serve their functional purpose equally well (the principle of intrinsic equivalence; see DITTMAR, 1 9 8 7 a). If we then ask why X is good and Y bad, A could explain that "X sounds nice, soft, melodic; X reminds me of a brook at which I used to play as a child" and "Y is hard, ugly; I always think of people fighting". Such notions make up the "cognitive" component, whereas the resulting tendency to learn X as a foreign language and not Y represents the "conative component". Individuals as well as groups identify themselves with certain linguistic expressions, varieties, and patterns of language behavior. When they seek or reject certain choices, they are reacting in a manner complementary to

126

Norbert Dittmar/Peter Schlobinski/Inge Wachs

rationality with subjective or collective feelings of identity. It is deeply rooted historical, individual, and collective identification that conditions the emergence, disappearance, and maintenance of varieties. The attitudes tied to identity are sometimes at odds with social integration. They can be either conducive or detrimental to it. 2.3.3 Accommodation The constant conflict between preserving the boundaries of identity and needing to belong to a social setting with the objective of social integration is resolved by certain adjustments of the part of the individual. "Adjustment" is understood in general to mean the process by which the individual's abilities, needs, expectations, and goals are balanced with the demands that the social environment makes upon the individual and with the possibilities that the environment offers him to satisfy his needs. (FUCHS et al., 1978: 44)

Key driving forces of adjustment are psychosocial factors of exchange, that, is certain constellations of costs and benefits. Depending on the topic, the participants, the situation, and the circumstances, it is more or less useful to change style, to continue speaking a particular variety or to switch, or to adjust vocabulary, syntax, or pronunciation. Adjustments of speech behavior are reflected in such things as code switching, stylistic variation, negotiation of identities and subject matters, great disparities in verbal competence ("motherese" or "foreigner talk"), and attempts to cooperate effectively under adverse conditions. 2.3.4 Expressiveness The propositional content of messages in communication is accompanied by speakers' particular pattern of expression that connotes features of their personalities as well as their group affiliation. This pattern of behavior becomes obvious in characteristics of intonation and body language. The expressive behavior accompanying verbal interactions is often referred to as being "effect oriented" in that it helps to impress, convince, mitigate, intensify, and to signal consensus or dissent with certain parts of the spoken message (see Buehler's concept of "symptom"). Among the numerous dimensions that symptoms of speech behavior can be for "identity" and "group affiliation" are: (1) intimacy and esteem-. Certain lexical expressions, features of intonation and body language are more familiar and are valued more highly than others. The signal solidarity versus distance. (2) immediacy and intensity: They reflect the immediate feelings of the interactors about the subject of their discourse.

Components for an overarching theoretical perspective in sociolinguistics

127

(3) control of the conversation-. Eye contact, paralinguistic phenomena, body movements, and other things can allow one to control the conversation (see BERGER & B R A D A C , 1982). (4) speaking versus remaining silent: "Speaking" and "remaining silent" are extraordinarily meaningful activities. Higher-order states have "lowerorder" states. For example, A can talk while B is not listening; B can interrupt with the intention of getting the right to talk. The distribution of "talking" and "remaining silent" can thus have varied social significance. (5) register. It is a versatile source of information about the intent behind the speaker's message, about other things. The personality, sex, and ethnic background, among other things. The register is normally associated with stereotyped characteristics of personality and group affiliation. (6) phonetic variation-. Variation of sounds is an especially characteristic indicator of affiliation with a social group. Sound variants are versatile channels of social connotations (see section 2.4). (7) volume and tempo of speech-. These two paralinguistic factors refer to the construct of "intensity". Combinations of high/low volume and rapid/ slow speech lead to the classification of speakers as "introverted" or "extroverted". Volume also contributes to the power of persuasion speakers have. The complexes of factors cited in 2.1, 2.2, and 2.3 can be broken down into structural features, as will now be illustrated with "structures of linguistic variation" (complex 2.4).

2.4. Structures of Linguistic Variation Linguistic structures of social significance are manifested at the interface of three dimensions: "social integration", "communicative competence of social action", and "identity". We are dealing essentially with two basic issues: (a) How are the linguistic strata of linguistic significance organized in such a way that social interactions in the context of, for example, interpersonal responsiveness, of the preservation of individual and group limits, of consensus or dissent, and of cooperation or conflict transpire in a manner commensurate with the communicational purpose of the participants trying to make necessary adjustments while acting in their own interests? In a specific context, then, every individual has the task of orienting the linguistic organization of the utterances to prescribed rules, purposes, and interests of communication. This organization results essentially from a coordination of linguistic expressions and levels.

128

Norbert Dittmar/Peter Schlobinski/Inge Wachs

(b) The synchronization of linguistic expressions and levels is only one process among others, however, a process that helps us understand the social constitution of linguistic significance. The latter is tied at least to the characteristics of other dimensions: (1) the salience of features (quantity and quality), (2) the degree to which awareness is controlled (friction between the norms one is trying to ascribe to and one's behavioral habits), and (3) the intentions of action (habitus versus options available in a specific context). 2.4.1 Linguistic Choices (Variants) According to WURZEL (GRUNDZUEGE, 1981), the German language has over thirty-five phonemes. In addition to a knowledge of these phonemes, however, speakers of the language also possess knowledge of phonological variation. For example, /r/ can be pronounced as a uvular [r] (in northern Germany, say, Westphalia) or as an apical [R] (in Bavaria, such as in the Nuremberg area). The pronunciation of [ J ] instead of /g/ is typical for urban vernacular in Berlin (see chapter 1). Depending on their social and regional backgrounds, speakers produce variants that receivers hear as connoting "Bavarian", "Berliner", "ethnic German", "member of the lower class", "actor", etc. As on other levels of linguistic structure, certain variants consisting usually of a small number phonemes make up a configuration presumbly perceived as "prototypes" of social significance. Other variants of form (choices in the strict sense) are morphological: (2-1) (2-1') (2-2) (2-2') (2-3) (2-3')

und dann ging er mit die Kinder in die Schule und dann ging er mit den Kindern in die Schule [and then he went to school with the children] da war er schon wieder a m Routieren da rotierte er schon wieder [then he got nervous again] dat is ihm sein Bruder das ist sein Bruder [that is his brother]

The article in the prepositional phrase of example (2-1) should be "dative" (see example 2 - 1 ' ) . Example ( 2 - 2 ) is morphosyntactically different from example ( 2 - 2 ' ) . Example ( 2 - 2 ) is typical for Westphalian and Eastphalian. It is questionable whether example (2-2) implies a meaning different from example ( 2 - 2 ' ) . Example ( 2 - 3 ) is pleonastic (see RILEY, 1 9 6 7 , on pleonasms in Black English).

Components for an overarching theoretical perspective in sociolinguistics

129

As the following examples illustrate, syntax varies as well, especially when one considers it from the functional viewpoint. (2-4) Aufm Bau, da hamse nur Aerger gehabt. (2-4') Sie haben auf dem Bau nur Aerger gehabt [They had nothing but trouble on the construction site.] Das Auto habe ich vor der Tür stehen Das Auto steht vor der Tür [The car is parked in front of the door.] (2-6) Die Sowjets verlassen Kabul, weil sie fühlen sich schon halbwegs wie Amerikaner (2-6') ..., weil sie sich ...fühlen [The Soviets are leaving Kabul because they feel halfway like Americans.] (2-7) da faellt man sich ja ueber (2-7') darüber faellt man/kann man fallen [One can trip over that.] (2-8) 'nen Brief hab' ich nicht gekriegt, nae? [I didn't get a letter, did I?] Example (2—4) has an expressive function. The prepositional phrase as local adjunct is left-dislocated. Example (2-5) is a variant of the Berlin or central Brandenburg area. It is not clear whether example (2-5') is synonymous with example (2-5). In example (2-6) the position of the subject and verb in the dependent clause introduced by weil (because) is the same as that in the main clause. In this example, linguistic change in colloquial German speech appears to be introducing a shift in parts of speech, with the subordinating conjunction weil being used as if it were the coordinating conjunction denn (for). Example (2-7) is from Eastphalia (the Bielefeld area), where the verb fallen (to fall) can be used reflexively. In standard German, compound pronominal adverbs are decomposed into adverb + elliptical preposition (that is, the preposition takes no complementary noun phrase). Incidentally, the semantics of adv. + prep is usually identical with the meaning of the pronominal adverb. Example (2-9) is interesting because of the negating particle nae ending the utterance. In the dialect heard in Solingen (Bergisches Land), the use of the negating particle conveys a meaning similar to that expressed by the tag question in English. The particle nae expresses the request for someone to confirm the supposition "X did not receive a letter" by replying with something like Nein, du hast keinen Brief bekommen (No, you didn't get a letter), which is the normal reaction of a person speaking the dialect of Solingen. Much has been written about the function of such tag (2-5) (2-5')

130

Norbert Dittmar/Peter Schlobinski/Inge Wachs

questions. In example (2-8) the speaker presents his utterance as a hypothesis addressed to someone. This certainly has psychosocial reasons. In any case, the syntactic variation in the cases cited above have something to do with local identity, mode of interaction, or expressive function. The lexical domain has by far the greatest variation, of course. The words scher^en (to joke), huepfen (to hop, to skip), and Gegenstand %um Stochern (poker) are doelmern, huepkern, and Pinoekel in the Eastphalian spoken in Bielefeld. The Swiss German oeppes is etwas (something, somewhat) in High German. An Aspirantur in the German Democratic Republic is a position given a research fellow, with the understanding that that person is to complete his/her doctorate. As is well known, words are the most enduring reflection of social history, a fact that can easily be proven by citing the names of objects (LABOV, 1 9 7 3 ) . The varied expression for certain meanings are polyfunctional; they reflect history and traditions, taboos, expertise, differences between political and social groups, old and young, regional loyalties and much more. Last but not least, semantics and pragmatics merge in forms of address and courtesy. During and immediately after the student movement (from 1968 to 1975), the use of du (you, familiar address) or Sie (you, formal address) between students and instructors in the Federal Republic of Germany signaled "solidarity" or "distance", respectively, as far as radical innovations in academic style were concerned. Whereas at the beginning of the 1970s most students and instructors at the Free University of Berlin categorically expected the use of du, most students and instructors there now clearly expect the use of Sie. Moreover, variation in forms of courtesy is a broad field. SCHLIEBEN-LANGE and WEYDT (1978) report that a compliment paid when a person enters someone's home is received by northern Germans quite differently than by southern Germans. In reacting to the utterance Das ist aber scboen bei Ihnen (Your home is just lovely), the northern Germans simply thank the guest for the compliment, whereas the modesty of the southern Germans (in this case, Swabians) leads them to point out what is incomplete about the furnishings or what still needs to be done. Our examples show that a knowledge of the invariant basic structures of German is in fact a necessary, but not sufficient, condition for the success of communication — much like the fact that having absolute pitch indicates little about one's capacity for producing or appreciating music. The reduction of uncertainty in communication is largely part of the function that variation has in facilitating orientation. It is impossible to process information without at least some background knowledge and knowledge of the context. For reducing or eliminating cognitive uncertainties, speakers or

Components for an overarching theoretical perspective in sociolinguistics

131

listeners have knowledge of variation relative to their experiences, to their immediate and less-immediate social environment. That knowledge concerns all levels of linguistic communication: phonology, morphology, syntax, semantics, and pragmatics. It is associated with parts of invariant linguistic knowledge. It enables us to place ourselves as well as others into categories of status and changes of a complex social fabric. Part of this knowledge has to do with the coordinating linguistic domains of communication (the linguistic levels). This also includes "knowledge about compatibility". It is expanded by two other stores of knowledge that we wish to call "knowledge of appropriateness" and "knowledge of conflict". 2.4.2 Compatibility (Restrictions on Co-occurrence) Compatibility regulates the assignment of linguistic expressions and forms to communicative functions and contexts (see SCHLOBINSKI, 1 9 8 6 ) . (2-9) (2-10)

(2-11)

(2-12)

(2-13)

1

Ick jeb dia koa geld nich.1 [I'm not giving you any money.] Ich ban entlang de Neckarr gesung un do bot ea mi ae fisch gebracht. [I was strolling along the Neckar and singing, and he brought me a fish.] Der Generalsekretaer des ZK der sozialistischen Einheitspartei Deutschlands sprach sieb fuer eine Intensivierung der freien Marktwirtschaft und eine Reduktion der Lehrerschwemme und des Butterbergs aus. [The General Secretary of the Central Committee of the Socialist Unity Party of Germany expressed his support for an intensification of the free-market system and a reduction of the butter mountain.] Herr Praesident, werte Vorstandsmitglieder, verehrte Anwesende! Das war vielleicht 'n beschissenes Haushaltsjahr. Die Ganoven von der Finansyerwaltung sehen gan% schoen abgefuckt aus. Immer rin mit die Knete ins Scheisshaus. [Mr. President, respected members of the board, honored ladies and gentlemen, it was really a shitty fiscal year. The cheats from the finance office look pretty fucked up. Always throwing dough down the can.] Am sept hemes und am huit heures moins le quart mueemer ide schual si. [We have to be at school at seven and a quarter to eight.]

Although examples 2-9 through 2-13 would never be heard in a real situation, the English equivalents are given for the reader's convenience.

132

Norbert Dittmar/Peter Schlobinski/Inge Wachs

The combination of the Berlin and Bavarian dialects in example (2-9) is impossible except, as is true for all these examples, for ironic purposes; the variants are incompatible. This is even more subtle in example (2-10), in which "restrictions of co-occurrence" are ignored. Instead of entlang de and gebracht, anne (parallel to hinne or untie) and gebrung (parallel to gesung) are used. Restrictions on the co-occurrence of different content are violated in example (2-11). The social and institutional function of the agent is no longer compatible with the content of the predication. The example is "lexically" faulty as well. The expression Generalsekretaer des ZK der SED (General Secretary of the Central Committee of the Socialist Unity Party of Germany) and Lehrerschwemme (surplus of teachers)\Butterberg (butter mountain) come from two mutually exclusive social systems. This is clearly reflected in the social use of language. The passage in example (2-12) illustrates what one traditionally terms a "stylistic inconsistency". The violation of compatibility stems from the contrast between "formal" and "informal" language situations and between "public speech in institutions" versus "nonpublic group slang". The form of address conforms to the characteristics recognized as "formal", "public", and "institutional". The subsequent three sentences — in terms of the lexical choices as well as the morphological features they contain — are the epitome of group slang intended for informal usage. Functional incompatibility in this example does not coincide with formal incompatibility. Lastly, example (2-11) illustrates a sociolinguistic borderline case. The utterance is taken from recordings made in Alsace. The co-occurrence of French and Alemannic appears to be acceptable at the local level, but is rather incompatible outside the region and in the schools. As long as the mixing is regular and reflects the local identity of the speakers, the sociolinguist advocates their interpretation as "coherent, compatible patterns". In the demarcation of such entities as nation, ethnic background, and group, they are often considered as an "interference facto" or a "symbol of conflict". Language switching (and "code switching" or "variety switching") within and between utterances clearly shows that "compatibility" is tied to the norms of social groups. They follow certain shared "patterns of taste", usually ones derived from the individual's own linguistic habitus or the value attached to the habitus of another group. On the other hand, one can say that the conditions for compatibility of expressions, forms, and sentences are governed by their "acceptability", which is contingent on the awareness of norms. There are ... utterances that cannot be accepted everywhere without reservation despite [the fact that they are] grammatically correct and semantically interpretable. They do not occur everywhere, and if they were to occur, they would elicit rejection. They are

Components for an overarching theoretical perspective in sociolinguistics

133

utterances that, despite observance of all grammatical and semantic norms, are felt in some situations to be somehow deviant, misplaced, inappropriate, impolite, offensive, etc. Consequently, there must be demands on the product of the activity that go beyond grammatical correctness and semantic interpretibility and constitute special types of linguistic norms. A whole group of judgments on utterances relates to the question of whether and how much (positively or negatively) utterances or characteristics of utterances correlate with a certain situation. (HARTUNG, 1977: 34)

We conclude with an example taken from research (AUER, 1 9 8 7 ) on Italian children in Constance who are being raised bilingually (German-Italian). The example under ( 2 - 1 4 ) illustrates co-occurrence restrictions in Italian and include Italian that approximates the standard; italiano stentato (Italian that shifts irregularly between standard Italian and dialect because of linguistic uncertainty); and dialectal Italian. (2-14)

u u il il *u

patre padré padré padre padre

AUER (1987: 13-14)

a fénEsté a fénEstré la finEstré la finEstra *la finEstra

a skolé a skwOlé la skwolé la skwola *la skole

ag'g'a dévéndà ag'g'a dévendaré devo diventa devo diventare *devo dvéndà *ag'g'a diventare

ag g e purtate ag'g'é portate o portate o portato *ag'g'é portato *o purtaté

comments on the patterns in example ( 2 - 1 4 ) as follows:

Restrictions on co-occurrence specify which characteristics of certain parameters should not occur together with certain characteristics of other parameters. The restrictions can apply within a segment or can affect neighboring segments. They can be undirectional (implicative) or interactive. (In former case one characteristic of a variable restricts the other, but not vice versa.) Restrictions on co-occurrence are more rigid for some speakers than for others. The analysis can begin with the premise of code fluctuation. For a given speaker from the group of Italian children of Calabrian background in Constance, the following restrictions may apply [on this point see example (2-14)]. The speaker allows ... variation between voiced and unvoiced realization of the alveolar plosive in padre if the noun occurs with a dialect definite article (») and if the final vowel is softened to Schwa, but he allows only the standard form — if the standard from of the article is used {it). The dialectal u thus limits the softening of the final vowel in a unilateral restriction of co-occurrence affecting neighboring segments. Respecting this and other restrictions on co-occurrence, the speaker can make a second complex type of variation — a fluid transition from a quasi-standard way of speaking to a dialectal one ( il padre —• ¡1 padr9, il padrS -* u padre) or vice versa. The "NORMS" holding for a different level of dialectal speech in one and the same utterance, elements that can be combined in code-shifting. By contrast, restrictions on co-occurrence are not observed in italiano stentato. One of the ways this is shown is that forms too divergent to be used within the accepted scope of variation occur together.

2.4.3 Appropriateness "Appropriateness" refers to the relation between the language, the language user, and the situation. In more specific terms, it refers at least to the

134

Norbert Dittmar/Peter Schlobinski/Inge Wachs

participants in a conversation, the level of the situation's formality, and the subject matter. The participants in the conversation must bear various dimensions of appropriateness in mind, by all means the proper forms of address and greeting. In Korean, for example, forms of address must be appropriately chosen according to age, sex, social status, and consanguinity (KOHZ, 1984). Forms for increasing or reducing distance (courtesy) have an absolutely central function in the managment of social relations. Adressing the German employer with the familiar du form stigmatizes a foreign worker as a social "outsider". Calling an adult female Fraeulein in the academic milieu gives the speaker a "macho", male chauvinist character in the eyes of the woman so addressed. The matter of "appropriateness to the situation" has been discussed many times. Much attention has been drawn to Black civil rights advocates in the United States, who conduct their public political discourse in standard English, not in Black English, as would be possible. No one has described the criteria for the success and failure of speech acts better than AUSTIN ( 1 9 6 2 ) . The ceremony of baptism is null and void without the "ritual situation"; the lecture delivered in a dialect fails to meet the expectations that one will hear superregional, commonly understood speech norms; and swearing or joking in the standard language is robbed of its effect because the standard cannot imbue the necessary expressions with local flavor. Situational inappropriateness is especially evident in the second language behavior of foreigners. The subject is another factor determining what is considered "approporiate" or "inappropriate" in a conversation. When fate strikes, it is appropriate to offer the victim words of consolation and condolence. A scientific subject usually demands great syntactic and semantic complexity. A conversation about personal feelings requires informal registers that appropriately filter the articulation of emotions through particles and graduated expressions. 2.4.4 Conflict Linguistic behavior in conflict situations naturally depends in general on conflict behavior. Linguistically, conflict behavior is directly relevant when different languages and varieties are involved. It is certain that speakers of a stigmatized or low-prestige language or variety switch as much as possible to the prestige language or variety in conflict situations. What does the "knowledge of conflict" cover? It defintely covers a complex area, only a small part of which we can focus on.

Components for an overarching theoretical perspective in sociolinguistics

135

One can infer from the "characteristics of identity" that "limits of the ego" are not infinitely flexible. Identity must be preserved to a certain degree. On the other hand, permanent conflicts are linked with great tension, and because lack of cooperation or extremely social, cultural, and linguistic distance result in overloading the processing system they are ecologically arising from interaction and reduce high cognitive uncertainty, instability, and the amount of energy needed to process information. One component of the knowledge of conflict is thus the habitual recourse to "acts of compensation" in difficult situations caused by conflict. Such acts of compensation can be manifested in (a) adjustments, (b) complementary behavior, and (c) strategies of negotiation. (a) Adjustment Provided that there is a willingness for cooperation to begin with, communication conducted under "adverse" circumstances is compensated by rapid speech and omission, codified redundancies (often exemplified by the language heard during a football game or sports commentary), slow speech and simplification of complex formulations or grammatical structures (the language used with or by foreigners or that associated with an all-toomotherly nurse), or the production of novel structures and turns of phrase (such as italiano stentato, which develops through contact with other languages). Certain difficulties with communication are mitigated by the speaker's or receiver's adjustment to the conditions under which communication is taking place. The issues merely alluded to here gave rise to the so-called accommodation theory ( G I L E S & S M I T H , 1 9 7 9 ) . (b) Complementary Behavior With complementary behavior, little or no adjustment takes place because asymmetries in power, influence, and technical expertise are involved. Conflicts go hand in hand with the knowledge that these asymmetries cannot be alleviated through "adjustment". Most workers are not in the position to achieve the status of the employer through finite steps of adjustment. Even a copy of the employer's behavior (produced by making an extreme number of adjustments) does not constitute the employer. There are good reasons to retain the complementary roles — the "costs" of doing otherwise are too high in relation to the "benefits" sought. And when technical expertise is involved, asymmetry is not based on behavior that is determined by power but otherwise equally matched in principle; it rests on what usually are generally recognized in competence and knowledge. In this context, "complementary" means that no adjustment whatsoever is considered suitable for redressing the existing asymmetry. Rather, symbolic capital acquired over many years does not allow "disparities" to be offset spontaneously and thus sets up conflict.

136

Norbert Dittmar/Peter Schlobinski/Inge Wachs

(c) Strategies of Negotiation Strategies of negotiation enter weakly structured conflict situations as "knowledge". They occur when neither symbolic capital nor a difficult communicational situation demanding cooperation "rigidly structures" the conflict, meaning that taking direct recourse to the habitus does not resolve the conflict. It is clear that "courtesy", the "styles of packaging" used in propositions, and intensification and mitigation as resources of expression are some of the negotiation strategies classified as "knowledge of conflict". We submit that varying and variable communicative resources play a decisive part in conflicts and the possibility of solving them. 2.4.5 The Distribution of Variants in Speech We have already mentioned that linguistic choices made in connection with knowledge of the interacting participants coincide with the meaning and function of variation. The parameters developed in the course of that discussion might have given the impression that social meanings have just as many different nuances as there are contexts in which utterances are made. In other words, the number of specific differences in the contexts of utterances, more or less explicitly formulated by exponents of "reconstructive sociolinguistics" (AUER, 1 9 8 7 ) , can be considered as a strong hypothesis in the research on variation because of its profound, pragmatic assumptions; it is an important theoretical aspect in sociolinguistics if it is viewed as one aspect among others. Sociolinguistics fixated solely on such reconstructions of meaning would seem to have its problems, however, as the following arguments will show. Let us present the premises upon which we will proceed. For cognitive reasons, it is impossible for speakers or listeners to use language in every situation in such a flexible way that new, creative, nonroutine social meanings appropriate to the given contexts are created through linguistic expressions. There is much to be said for the fact that there is individual arid group variation, but it cannot be doubted that speakers acquire a language system of whose rules they must, for cognitive reasons, have a certain degree of automatic mastery if they are to cope with the manifold demands of communication. Our knowledge of first and second language acquisition shows that "variation" lies within certain limits of the individual's ability to process data. The assumption — a justified one in our opinion — that the speaker or receiver commands a system of rules suggests the view that a part of communication must be organized recur-

Components for an overarching theoretical perspective in sociolinguistics

137

rently, with more or less close attention and with more or less explicit goals and alternatives of action. Variation is subject to intrinsic and extrinsic conditions. The former constitute the habitus; the latter, the individual's alternatives for action relative to his or her own intentions, the goals of the interaction partner, and various contexts. In the first case variation is a program constituted by recurrence. In the second case, however, one must distinguish the context-intensive and functional senses in which resources are to be used. Context-free communicational resources (those that are recurrent) and context-sensitive communicational resources (those that are to be used according to whether they are appropriate) must be reconstrued both as a system and as a local way of managing the organization of social meanings in a given environment. For the moment, these considerations are nothing more than assumptions needing further elaboration. The following fragmentary observations are offered as a modest beginning for such explication. If there is context-free and context-sensitive social meaning in linguistic variation, then it can be documented both qualitatively and quantitatively. We do not assume that an average native of Nuremberg can modulate his habit-formed apical /r/ according to context to any great degree. Slightly shaded variants of the apical /r/ may sometimes indicate the situation (formal versus informal), the "subject", and the "interaction partner". The main segments of the apical /r/ will repeatedly connote "a native of Nuremberg", further specified if necessary by "woman", "city dweller", and "upper middle class". There are simple, clear, qualitative limits to how far reconstructive sociolinguistics can go in this regard. To present such limits as "rashly prescribed" would be something like saying it is unjustified to speak of "children" and "adults" in communication because one cannot categorize people as such until contexts of their behavior have been analyzed by abstracting from "children" and "adults". By contrast, we maintain that contexts without the presence of at least content-related qualities or predicates cannot be interpreted or reconstructed. The input of interpretative and reconstructive processes for constructing meaning is thus limited right from the outset by basic qualities, for they establish the framework for the scope of relevance that social meanings have. One could object now that the apical /r/ plays no role at all in certain interaction contexts as far as meaning or connotations of "(like) Nuremberg" are concerned. It would be irrelevant in these contexts. But this view presupposes evidence that the participants subjectively perceive meaning — and it implies the theses that the "objective" (meaning visible,

138

Norbert Dittmar/Peter Schlobinski/Inge Wachs

audible, perceptible, for example) behavioral characteristics are relevant only insofar as they become "salient" in the interaction. There is much to suggest that the "qualities" of the interaction partners (being "black" or "white", "man" or "woman") as well as their actual behavior constantly contribute to interpretations. That is why it seems fitting to link the qualities of both "objective" and "subjective" behavior with each other. Intrinsic and extrinsic conditions of the social meaning that linguistic variation has are guided by (1) quality and quantity, (2) salience of features that vary, and (3) ways of affecting the outcomes of action. (1) Quality and Quantity The fact that basic characteristics like "black" versus "white" and "male" versus "female" are perceived in interactions and used as a framework for interpretation is just as undeniable as the fact that incremental, quantitative behavioral features are relevant for interpretation. Recurrent patterns are shaped by habitus. They are indispensable prerequisites for competent interaction. At the same time, they provide information to which the receiver can orient himself. The characteristics of the habitus are an expression of the individual's identity and are thus a kind of "accessory", a token, and an aid to interpretation. The example of the "swans", which are usually white but which can also be "black", has shown that quantitive statements are necessary if one wishes to avoid unjustifiably excluding cases that could be expected in the future. We need such quantitative statements because language families and variety families are characterized by numerous gradual transitions and because language change, a key problem of linguistics, can be based scientifically only on the view that language, as an organism, is fundamentally an open-ended system that is only incompletely embodied in or still in the process of being incorporated by a corpus C at a given point in time ti. If language is considered as an object of hermeneutics, quantitative arguments can be ignored as basically inappropriate. Each character or text is then to be understood as being "unique in this context and in this meaning". To us, the legitimacy of such a radical version of restructive sociolinguistics does not seem compelling. We can proceed with the reasonable conception of languages and varieties as system as long as the continual contribution of pragmatics at present makes it possible to overcome the taboos against investigating the integrated context of linguistic action in communication and of the knowledgeable use of a system of linguistic rules (in terms of "modules", for example, see KATZ, 1983). "Quantities" play a role in at least two respects: (1) to record recurrent regularities of

Components for an overarching theoretical perspective in sociolinguistics

139

behavioral elements in the form of "multiple-possibility statements" and (2) to make transitions and changes evident. Aside from the fact that increased recurrence is given and that it "expresses something" if behavior is thought of as "meaningful" in principle, the "semantic configurations" emerging from the number of words and their presence in grammatical meanings are similar to types of cloud formations. As researchers we thus have ... the possibility of formulating statistical sentences through inductive methods. We called these statistical sentences "multiple-possibility sentences" because they provide for several possibilities from the outset and thus do not begin to falter the moment one or the other of these possibilities occurs. (SEIFFERT, 1969: 171)

We maintain that habitus-conditioned behavioral characteristics recurring at certain intervals restrict open expectations of behavior and leave their mark on "horizons of action". Like skin color, body height, and other features, they constitute an empirically supported background to stereotyped reactions that establish a certain framework for the interpretation of action. By contrast, there are difficulties in explaining them and placing them in a single theoretical framework (see FASOLD, 1984: 4). Despite the differences between quantitive and interpretative procedures, the identical natures of methodological approaches in sociolinguistics is, paradoxically, to be seen in the fact that recurrent features or the social organization of sequences of utterances meaningful to speakers can be observed ad hoc without there being any explanatory framework in sight. This does not speak against these studies at all, but it does suggest that one can intensify the efforts to formulate theories capable of providing explanations. Several observations necessitate the consideration of qualitative and quantitative parameters in an integrated type of sociolinguistics. (B-l) Linguistic rules are not always used categorically; social change is reflected in variable structures. (B-2) The quantitative distributions of patterns of variations initially follow laws inherent in the structure of a language. Their relationship with external parameters can be understood as habitus-conditioned, stereotyped background to behavior that is to be interpreted, context-free, in terms of regulatory rules ( D I T T M A R , 1983). Beyond this, however, such behavior can take on no — or no other — social meaning in specific contexts than that of the recurrent meanings that have been noted. In other words, distributions of varying patterns must constantly be examined for the degree to which they are determined by pragmatic parameters. (B-3) Utterances and sequences thereof do not have a creative, new social meaning in every specific context. Certain patterns occur more or less

140

Norbert Dittmar/Peter Schlobinski/Inge Wachs

frequently in certain people. Recurrences necessarily exist, for otherwise great cognitive uncertainty would doom every stable social organization to failure. With statement (B-2), we call for retraction of the hypothesis that the social meaning of variable structures is quantitatively overdetermined; with statement (B-3), for the retraction of the idealized hypothesis that utterances are overdetermined by contextual conditions. (B-4) The strict positivist observation of such things as utterances, occurrences of linguistic units, and sequences of utterances and of social meaning attached only to their "aspects of form" is an "idealization" as far as reality is concerned. Observation from the "external perspective" (and not, as called for in ethnography, from the internal perspective) prevents subjects participating in actions from contributing to the understanding of their action. Because of the respective assumptions made in the paradigm of linguistic variation and the paradigm of discourse analysis, a particular occurrence must thus, for example, be provided with a particular "constant" social meaning when linguistic variables are involved. Or, for instance, a feature like aeh must be interpreted as a particle that is meaningful for and integral to the social organization of the verbalization (ex post facto interpretations), although the function of this particle may be merely to indicate linguistic planning. We do not wish to assert that we must know the much maligned intention of the interactors. But as long as we do not know their "internal perspective", ascribing of meaning "from the outside" are speculations. (B-5) There are good reasons for the fact that we need qualitative and quantitative analysis. Let us take example (2-15) (2-15)

A: Chef, wat soil ick nu machen? [Boss, what should I do now?] B: Haettest du besser aujjepasst, waer' das ganv^e nicht passiert. [If you had watched better what you were doing, none o' this would have happened.]

A is speaking the Berlin dialect, whereas B actualizes only one feature of it — the /g/ —> [J] in aufjepasst instead of aufgepasst. The [J] variant is not actualized in the word gan%e. In terms of Giles' sociopsychological approach, we could explain this by saying that B, being the supervisor and speaker of the standard language, adjusts slightly to A's Berlin dialect, overtly signaling in an informal way a willingness to cooperate. A precise linguistic analysis of the variation of /g/ reveals that the sociopsychological explanation is insufficient, however. We know from a detailed quantitative

Components for an overarching theoretical perspective in sociolinguistics

141

analysis that among the phonological segments following the velar stop the Schwa favors the voiced palatal spirant [J] the most, the [r] the least (see chapter 1). The example clearly shows that understanding the utterance in the actualized linguistic structure depends on our qualitative and quantitative knowledge. The"downward adjustment" is evidenced by a variant that is more apt to be favored when it precedes Schwa than when it precedes /a/. This is a case in which quantitative arguments dovetail with qualitative ones. As in many areas of linguistics, form and function must be carefully distinguished from one another. When two variants are involved, their dissimilar functions are relevant. But, and this seems to be a principle documented to an increasing extent, we cannot determine the extralinguistic function more precisely until the inner linguistic form is described. Whatever kind of description is to be provided in sociolinguistics, we must (1) determine the form of that which is to be described and (2) reconstruct the function of that form. Form and function can be determined on the intralinguistic level by distribution analysis and quantitative weighting. On the extralinguistic level there is the much discussed question of social meaning, whose treatment (see AUWAERTER, 1 9 8 2 , for example) varies according to the descriptive approaches chosen. It is customary to distinguish between correlative and interactionist sociolinguistics. The former focuses on systems and habitus; the latter, on pragmatics and changing contextual frameworks. The interactionist approach considers language to be both a product of an interaction and a factor that construes social reality (see also BERGER & LUCKMANN, 1 9 6 9 ) . Both approaches are necessarily reconstructive, however, in that they determine the meanings of linguistic expression and utterances ex post facto, proceeding in more or less context-sensitive and context-free fashion, respectively. This has been elaborated upon much more clearly in research on second-language acquisition than in sociolinguistics: "Most, if not all, second language acquisition studies done to date ... have been ex postfacto, in that investigators have decided what to look for after the data have been collected" (TARONE, FRAUENFELDER, & SELINKER, 1 9 7 6 : 115). In this sense, all available sociolinguistic studies can probably be called "reconstructive". A few descriptions in the paradigm of the "ethnography of communication" are exceptions (see HEATH, 1 9 8 3 , for example). In these studies, which feature a holistic approach, documentation of linguistic utterances merely illustrates some of the many other examples possible. Participant observation is used in an attempt to mitigate the disadvantages

142

Norbert Dittmar/Peter Schlobinski/Inge Wachs

of ex-post-facto observations drastically. Prevailing social and normative meanings are to be revealed and documented as directly as possible in the context in which they occur. (2) Salience of Linguistic Characteristics The word "salient" can relate to linguistic structure or to its social reception. It is certainly important whether variants turn up at the beginning or at the end of an utterance. Their interplay with the curve of intonation is also important. After all, the salience of variants depends on stress of the syllable or morpheme in which they occur. We must assume there is a typology and hierarchy of more and less salient linguistic structures. This is where natural laws of psycholinguistics intervene, laws that have apparently received little attention from sociolinguists. They are, however, essentially different from the restrictions on co-occurrence mentioned earlier. Another important aspect is the control of the style of speaking. The question of whether verbal interactions take place as routine and informal exchanges or with a certain degree of tension in a formal vein has been a recognized subject of research in sociolinguistics since LABOV (1966). The principle of attentiveness holds that styles can be ordered along a single dimension, depending on the measure of attentiveness being accorded the act of speaking. ... Unforced and familiar styles can be placed at one end of this continuum; aloof, ritual s t y l e s a t t h e o t h e r . (LABOV, 1 9 8 0 : 1 6 - 1 7 )

The conveyence of social meaning varies in intensity depending on the degree of formality in the style. Our hypothesis is that social meanings are conveyed "in greater r e l i e f in informal conversations than in formal ones, in which social meanings take a back seat to diverse lexical expressions and idiomatic phrases. "Speech control" influences more than just the organization of social meaning on various linguistic levels, however; it also influences reception. It depends entirely on the "focus" of conversations and verbal interactions and on their being embedded in a setting. Formal linguistic features of identity in the habitus may not be perceived by a participant in the interaction if that person is concerned primarily with sentimental conversations, hobby gardening, or having fun for fun's sake. In the initial conversation or job interview, in which there is a certain amount of pressure to become acquainted and, possibly, to establish a relationship with one another, characteristics of the habitus receive much greater attention — if they provide important clues that reduce uncertainties about the other person. From this it follows that the control of awareness helps determine the intensity of the production and reception of social meaning.

Components for an overarching theoretical perspective in sociolinguistics

143

The production of social meaning is subject to the fluctuations of the focusing that takes place in the interaction. Much the same can be said of its reception, which is also greatly dependent on the context of the action and the situation in which the interactors find themselves. From this it follows that social meanings are produced and received with differing degrees of relevance and intensity. It does not follow, however, that the necessary reconstruction of these fluctuations is solely of a qualitative nature. To the extent that they are qualitative in nature, one must figure out what their pragmatic determinants are. In the process, their mutual characteristics, weightings, and relations will regain quantitative transitional qualities. (3) Ways of Affecting the Outcomes of Action This point can be grasped in only a vague way, but has to do with the speaker's intentions. Although little is known about this area, the problem can still be examined. One of the things with which the linguistic calibrations of "register" have to do is the openness of interaction situations. GOFFMANN ( 1 9 7 4 ) , for example, has pointed out that the mode of address varies with the "symmetrical" versus "asymmetrical" interaction relationship. In a hierarchical relationship it is likely that the subordinate partner will relate to the higher-ranking partner by means of "neutral" turns of phrase not relating directly to him. Vice versa, it is easier for the higherranking person to relate directly to the subordinate person. There seem to be different degrees of "empathy" and "desirability" in communication. It is plausible that the use of hypercorrect features, the prestige of which is presumed, is more frequent in a job interview than in informal conversation with people the same age. To make a correspondingly good impression, the appropriate stylistic resources are brought into play. Depending on the state of the interaction and one's own interest, there are dynamic fluctuations in how the fields of variation are filled in. A possibility "of deriving a particular advantage from a situation" might well lead the speaker to a give preference to the prestige features. The certainty of having an unchallenged position in the interaction might well coincide with the relatively stable production of semiformal "basic language". It may be similar if the interaction partner is categorized as an opponent who is more powerful and who has views that one does not share. Taking recourse to basic language in this case is often a clear act of "drawing the line". Symbolically, solidarity with the social group to which one feels affiliated is stressed by one's linguistic behavior. Such a demarcation can produce very sharp linguistic edges along with the countervalues corresponding to them. Lastly, it is also possible to approach

144

Norbert Dittmar/Peter Schlobinski/Inge Wachs

the other person on the basis of mutual affiliation with a specific group. The approach might stem from a difference between individual and group interests but ensues because of affinity and a desire for cooperation. The linguistic choices serving the interest of such options spur dynamics of variation whose quantitative transitional character must be recorded. Underlying this quite general interpretation of the influence that the goals of action have on linguistic choices and stylistic calibrations is the initial premise that there are dynamics of variation in the interaction process. These dynamics can be treated adequately only in terms of the ways in which they intermesh quantitatively and qualitatively.

Part 2 The Analysis of Urban Varieties: Some Methodological Considerations

Chapter 5 WOLFGANG KLEIN

The Unity of a Vernacular. Some Remarks on "Berliner Stadtsprache" Und wenn da eena.vastehste, mit'm Auto ruffjeht, dit hat u f f m Damm zu stehn, den schlag ick u f f t Jehiirn.

1. Measuring a cloud When I was ten years old, I entered - for the first time in my life - a cloud. It was a small bright white oblong cloud, and one could clearly see it from the valley when we started to climb up that mountain in the Austrian Alps. But as we came closer, the shape of the cloud became less distinct, and eventually, there was only fog around us. It was very disappointing. Like most speakers of German, I have a quite distinct feeling of what Berlin dialect is, and as a rule, I can easily recognize a speaker of that vernacular after a few words, such as those quoted above. I think this feeling is rather solid, and it would surely survive an empirical test. I have no doubt that there is something like "Berliner Schnauze". But as soon as one begins to have a closer look at it, it seems to fade away. Linguistic investigation shows that Berlin vernacular is apparently a quite heterogenous phenomenon, composed of many variants, that many of the linguistic features which are generally thought to be constitutive of it appear in other dialects and vernaculars as well, and that its specific flavour may arise from features hardly ever mentioned and surely never carefully studied in the literature, such as speech rate, pause structure or pitch range. But this latter possibility is pure speculation, and as for the facts and findings, it seems that, as often happens in research, increasing detailed knowledge based on careful empirical studies annihilates the object under investigation.

148

Wolfgang Klein

Now, the fact that the cloud dissolved before our eyes as we approached it does not mean that there was no cloud, nor that it had no clear shape. We can take pictures of it, from all sides, measure its diameter, compare it to other clouds. It is an entity. And similarly, we have a clear feeling that a vernacular, such as the "Berliner Schnauze", is an entity. But what constitutes the unity of such a vernacular? I am not in a position to answer this question. The project discussed here has brought together, in the first years of its existence, a great deal of the necessary bricks, and this breathing-spell before putting them together may be the right time to raise some general issues in connection with the question above. This is what I intend to do in the sequel.

2. Congruency of features There are two ways of circumscribing a vernacular, just as any language or dialect: first, by referring to the people who speak it, and second by giving all the linguistic properties which characterize it, or at least those in which it differs from other vernaculars. Let us call these "external" and "internal" definition, and for a vernacular to be a unity, there must be a straightforward correspondence between external and internal characteristics. Now, the problem is that both external and internal definition are compound, that is, each of them is based on a complex cluster of features. In general, it makes no sense to say, for example: "Berlin vernacular is that language spoken by the people living now in the (geographically or politically defined) area of Berlin". Apparently, there are other external characteristics that must be included in the definition, such as social class, the "nativeness" of the speakers, the geographical stratification within the city, etc. So, on the external side, we have a whole set of features Si XS2X... xs n , and the question arises of how we can cluster these features in such a way that they indeed constitute a unity rather than an arbitrary collocation. The unifying trait is apparently that the people in question speak Berlin vernacular - not that they regularly speak to each other and, in this sense, form a speech community. A speech community is a unity, and an important one, and in many cases, there is a one-to-one mapping between a speech community and a vernacular. But this is clearly not the case for the Berlin vernacular, where regular communication between vernacular and non-vernacular speakers is far from being exceptional. So,

The Unity of a Vernacular

149

if there is a unity somewhere, we cannot base it on the external features. It must be based on the internal, the linguistic characterisation. A vernacular may be described on various levels, such as phonology, morphology, syntax, lexicon, specific types of discourse behavior, etc. Traditional research is mostly concerned with one or at most two of these domains, and most often with some isolated variables within them. This is perfectly understandable, given the mundane restrictions of all empirical research projects, such as time and money; but it obviously limits the scope of possible results. So, it is clearly one of the main merits of the present project "Berliner Stadtsprache" that it approaches its objects on various levels at the same time: phonology, structure of the lexicon, interactive behavior, attitudes of speakers. And with the rich data that are available now, it would not be difficult to include other areas, such as syntax or morphology. Now, the aim of the project is not to study a series of independent properties of Berlin vernacular, but that vernacular, and the investigation of all of these properties is a crucial prerequisite to that end. What this investigation buys us - and this is an important achievement in itself - is a series of linguistic properties f 1 x f 2 x . . . x f n on various levels. But this is not all we want. First, we must relate these properties to the external features mentioned above, and this leads to the problem of congruency, and second, we must relate them to each other, and this leads

to the problem of

interplay.

The problem of congruency is simply that internal features do not coincide in their relation to external features. It may be that the use of some specific lexical items is typical for an externally defined population a, the occurrence of a phonological feature for a different and only partly overlapping population b, a special type of discourse behaviour for a third population c, and so on. In a sense, this corresponds to the traditional problem of isoglosses in dialect geography, where - to take a classical example - the boundaries of p, t, k versus pf, ts, 9 do not or not fully coincide. But the traditional case, although not different in principle, is relatively straightforward when compared to the congruency problem in the present case - first, because in the isoglosses example, only one external correlate is considered, namely geographical space 1 , and second, because the linguistic variables are closely related, part of the same process and

1

Geographical space, of course, reflects other features, such as communication channels and barriers. But these are only indirectly related to the linguistic characteristics. Isoglosses in the traditional sense are not between social classes or religions, but between countries and villages.

150

Wolfgang Klein

thus show a related distribution: p is compared to t with respect to the same phonological process and its results, but not p to same lexical or syntactical feature, or even to discourse behaviour. This is quite different in the case of Berlin vernacular and the various linguistic variables which characterize it. Thus, the use of a word such as "Keule", the change from "g" to "j" in many positions, and a specific kind of interactive behavior (which I won't try to specify here) are three variables which are generally felt to be, and have actually been shown to be, constitutive for this vernacular. But apparently, their distribution with respect to external factors is totally different. It is different already with respect to geographical space, and it is the more so, if one includes other external variables. In other words: Relating all the linguistic features which characterize Berlin vernacular 2 individually to external features does not buy us anything with respect to the question what constitutes the unity of this vernacular. Hence, we first must somehow bring these features together before we can take the step into the external world. Note that this is not to say that there is no point in studying the distribution of individual variables. If it can be shown that the specific type of interactive behavior displayed by speakers of "Berliner Schnauze" shows up everywhere, in Mannheim, Hamburg or Vienna, where specific conditions of early socialisation or family structure obtain, then this is a highly interesting finding in itself. But it does not answer the unity question.

3. Interplay of features A language, and so a vernacular, is not an agglomeration of independent linguistic features but a system - at least to a large extent. If a language, for example, drops final plosives, such as "d" and "t", then this is a phonological feature of that language. But this phonological feature may have dramatic consequences for inflectional morphology: tense marking or case marking may be made impossible, unless other expressive devices are used of a particular intonation, and vice versa, or of a specific lexical particle etc. etc. I think no more examples are needed: the case is obvious, or at least should be obvious, and if it is not always taken into account in empirical research, then this is mainly due to real world restrictions of research projects such as those mentioned above. 2

We could have, for example, a "full account" of the linguistic characteristics of say 10 speakers which everyone feels to be typical of "Berlin vernacular".

The Unity of a Vernacular

151

This obvious fact seems to give us a handle on the unity question: what we have to look for is a specific clustering of linguistic properties. I indeed think that this is the most promising step to take. I give this advice for free. But as with so much free advice, this suggestion solves one problem and creates another - namely the problem of identifying principles on which this clustering can be based. How can we proceed from a mere collocation of linguistic properties to a meaningful interrelationship? There is no straightforward answer. In the examples of interplay considered above, there was an implicit assumption that there are certain expressive needs which, when not expressible in one way, must be expressed in some other way. Thus, if reference to the past cannot be expressed by some other construction, and similiary, if sentence mood cannot be expressed by syntactic means, such as inversion, then it must be expressed by intonation, or by a tag, etc. All of these cases may be arguable, because we cannot be sure which expressive needs are really indispensable and therefore necessitate compensatory devices, if not expressible in a specific way. But in principle, it seems plausible that there is such a balance between various possible expressive devices, and the specific clustering of linguistic features in the Berlin vernacular may be based on this sort of communicative compensation. There is another possibility. Consider the consequences which the replacement of /g/ by /j/ in at least many positions may have: it is not just the local replacement of one sound by another but the entire phonemic system may be affected, since the phonemic distinction between /g/ and HI is lost. It has been argued, especially in the structuralist tradition (MARTINET), that such "gaps" in the system lead to changes in other parts of the system. Now, it is surely disputable whether this is always true; but it is at least possible that specific features in one subsystem (such as phonology) may lead to adaptations in other parts of that subsystem. This would be another type of non-arbitrary clustering of linguistic features. There are surely other kinds of necessary - or at least non-contingent linkage between various variables, and the general strategy must be to look for these. But still, it seems highly unlikely that all features of Berlin vernacular could be mutually related in a meaningful way. Thus, the occurrence of some specific lexical items, which are felt to be so typical for Berlin vernacular (cf. section 2 of the project report) can hardly be related to phonological or syntactic features. But they belong crucially to what constitutes for us "Berliner Schnauze". Hence, the specific clustering of linguistic properties which makes up the unity of that vernacular cannot be exclusively based on non-contingent relations in the sense outlined above. There must be other ingredients.

152

Wolfgang Klein

4. Flavour So far, we have assumed that the unity of a vernacular, if at all, can be based on a specific clustering of linguistic properties. And I think, this is true. But it may well be that our feeling of a vernacular being a unity and this feeling was our starting point - is much less based on this specific cluster of linguistic properties, which the vernacular objectively has, but on some of them whose taste is particularly salient. They are the curry in the dish, which gives it its particular flavour, although there are other and even more substantial ingredients. So, it may be that we identify Berlin vernacular just by means of a few cues, which may be a minor component amongst the objective properties that characterize the whole vernacular. At this point, the "flavour hypothesis" is obviously a mere speculation. The only way to find out whether there is some substance to it is to isolate some features which look like good candidates for the "curry", and to test whether Berlin vernacular sounds like Berlin vernacular without them. In any event, it seems clear that a speaker is usually more sensitive to some properties of a dialect he does not speak than to others. I think most Germans would consider the substitution of /j/ for /g/ to be a particularly salient characteristic of Berlin vernacular. But it can at best be but one of the spices, since it also appears in other German dialects. Another good candidate are lexical items which are known to be specific to Berlin vernacular. But again, this can't be the whole story, since some of the most typical words are rare in everyday communication, and usually identification of a speaker as a representative of "Berliner Schnauze" does not hinge on the appearance of these specific lexical items - although they are a good cue if they show up. If I may trust my own weighting of features, prosodic properties are most important. This is evidenced by the following somewhat anecdotal observation which is shared by several speakers. When reading through the various texts transcribed and analysed within the Berlin project, one does not have the impression that these texts are typical for Berlin speakers. Often, one could easily imagine them to be collected in Mannheim or Köln (barring some occasional idiosyncrasies, as those just mentioned). This totally changes if one then listens to the original recordings: they are clearly identifiable as Berlin vernacular. This cannot be due to the phonetic properties alone, because at least some of the phonetic properties - actually quite many - are maintained in the transcription. What is totally lost, however, are prosodic properties. No one seems ever to have tested the relative weight of

The Unity of a Vernacular

153

prosodic features 3 for the identification of a dialect. But, just as a guess, I am quite convinced that the small text quoted as a motto at the beginning of this chapter, when read with Köln intonation, would not be identified as Berlin vernacular but as a sort of Köln vernacular which - depending on the real knowledge about that vernacular - is felt to be somewhat odd. This last remark has carried us into realm of pure speculation, and so, it seems time to leave the arena. Before doing so, however, we will have a brief look on what the "flavour hypothesis" implies for the question of unity. Our central argument for the assumption that something like "Berlin vernacular" indeed exists as a unity was the fact that every German speaker could identify it. I still think that this observation is correct, albeit with some qualifications. But the flavour hypothesis could easily acount for this fact without the assumption that there is indeed a unity such as "Berlin vernacular". There might be an agglomeration of language varieties which are quite different in a lot of essential linguistic properties but which still share the same "curry". But this is an open question, and only further empirical investigation can help us here. 5. The more we know the less we know About many things in life, we have a quite stable everyday knowledge which is not based on methodological scientific investigation but on experience, tradition and the remarkable intellectual capacities generally attributed to us. It gives us a clear and consistent picture of the world without which we could not exist. This knowledge also includes knowledge about linguistic facts, sometimes quite detailed, sometimes quite global and gross. And so, some of us feel tempted to check, to explore, to extend and to refine this knowledge. And indeed, linguistic research has taught us a lot - a gigantic amount of facts. On the other hand, the clear shape of the phenomena from which we started dissolves, and we are left in the fog of truth. But it may be that the mountain is higher than our present position, and when we continue climbing, we may see the cloud from the other side. But from here, we don't know, of course, how high the mountain really is 4 . 3

4

I am not sure what the relevant prosodic features in the present case are. It seems, however, that it is not the basic course of the pitch contour, i.e. of intonation in the narrow sense of the word. What seems more important are speech rate, "timber" and the specific way in which pitch changes are realized within a segment. I am greatful to Jane Edwards for stylistic comments.

Chapter 6 PLERETTE THIBAULT

Discourse Analysis in Sociolinguistics Since the pioneer work of LABOV on the Black English vernacular (1972b), meaningful steps have been made in the sociolinguistic characterization of working-class vernaculars in urban speech communities. Sociolinguists have turned themselves into good ethnographers, in order to get access to spontaneous verbal interaction. Dealing with rather homogeneous subcommunities, they had to look for internal social factors to be correlated with linguistic variants. The degree of integration in a social network, well investigated by MILROY ( 1 9 8 0 ) proved to be an important factor to consider in sub-community sociolinguistic research. Meanwhile, studies on urban speech based on socially stratified samples have also progressed. After twenty years of empirical work, it is now possible to obtain 'real time' data 1 so as to document sociolinguistic changes. HENRIETTA CEDERGREN has led the way with her comparison of Montreal French "r" at two time periods and by replicating her 1967 field work in Panama. Our Montreal research group ( D I A N E VINCENT, DAVID SANKOFF, WILLIAM KEMP and myself) entered this era in the summer of 1984 when we re-interviewed half of the 120 informants whose speech production formed the well-known 1 9 7 1 SANKOFF-CEDEGREN corpus (cf. SANKOFF AND SANKOFF 1 9 7 3 ) .

The Berlin research group has certainly set itself ambitious goals, since it has undertaken the description of the Berlin speech variety, as well as the analysis of sub-community spontaneous speech interactions. In this paper, I will concentrate on one particular aspect of sociolinguistic research, that is: discourse analysis. First, I will comment on the Berlin group studies on style, focusing on two of the questions they raise in their work. Then I will introduce the approach our Montreal group is taking to discourse analysis, in relation to our global research goals. ' As opposed to 'apparent time' data, obtained through the synchronic comparison of generational differences (cf. Labov 1966; 1972b).

Discourse Analysis in Sociolinguistics

155

1. Is There a Regional Variation in Style? A quick answer to this question would be a definite "yes", since there is a widely known name for the Berlin speech style: the "Berliner Schnauze" or "mouth" (as in "loud-mouth"). Dialectal phonetic and lexical features are certainly part of it, but rhetorical aspects, such as witty remarks of a special type, are also involved. In the sixties and seventies, a similiar label raised passionate debates about the language spoken in Quebec. It was calles "joual" (from a local pronunciation of the word "cheval" meaning "horse"). This very pejorative label was meant to qualify the poor variety of French spoken in Quebec as compared to standard French 2 . Within the community, some people thought everybody spoke "joual" in Quebec; others excluded themselves by saying it was a working-class speech variety, to be found only in Montreal. Evaluations of the "joual" by middle-class urban speakers, as manifested in our 1971 corpus, were similiar to those of middle-class Berliners with respect to their local "Schnauze", reported in SCHLOBINSKI'S dissertation (1987). In Quebec, the "joual" label still exists and still carries its connotations but it has become a lower-key issue with the emergence of a standard local variety. Analogies between the Berlin and the Montreal situations lead me to question the status of the "Berliner Schnauze". Is it part of a regional speech variety or is it an urban working-class vernacular style? In order to analyze characteristic discursive patterns of the Berlin speech community, the Berlin research group has focused on five narratives taken from lengthy interviews of two working-class speakers relating conflict situations. From this they formulated a set of programmatic rules corresponding to individual speech acts (as defined by SEARLE and AUSTIN), each speech act being viewed as a step in the development of a conflict situation. In the studied accounts of conflicts, direct and indirect requests, rejection of requests, insults, threats, and status claims are performed in a structured way in reaction to the opponent's moves. These pragmatic rules could very well be universal with specific realization patterns according to the type of interaction and the social group involved. That the specific array of rules in the Berlin narratives corpus is part of working-class vernacular norms is a workable hypothesis. Indeed, sociologists studying stratified urban societies have characterized the two 2

The word "joual" was first used in reference to the 'barely-human' utterances of illeducated teen-agers.

156

Pierette Thibault

ends of the social scale as displaying well-integrated and relatively homogeneous networks with specific norms of behavior. It is doubtful, however, that the social groups in the middle of the social scale would show any particular array of rules. What could be expected from them is occasional style switches within the discursive repertoire of their speech community, be it urban or regional. The Berlin group also looked at the problem of circumscribing a regional speech style from another angle.LASCH (1928) and others had reported a collective image of superiority associated with the Berlin community. Considering the five narratives in the light of the speaker's self-image, it appears that whatever moves are made by the opponents during the course of the conflict, the final conversation step always leaves the speaker with the victorious 'last word'. But there are also specific values attached to this positive image of the speaker. In all the narratives, through variations with regard to verbal tenses, use of pronouns, intensifiers and figures of style, the speaker expresses personal authority, rhetorical abilities and wit. Less frequently, he will stress his physical superiority and in only one instance, the image of the speaker emerges as being that of a tolerant man. Are these values specific to the Berlin community? I cannot help thinking of personal experiences in Montreal with taxi drivers and garage mechanics who, time and again, have bombarded me with witty remarks that left me speechless. The language was different but the discursive strategy the same. "Sounding" in the U.S. Black community also comes to mind. Such cross-cultural similarities lead me to believe that rather than being regional, discursive strategies tend to be characteristic of specific social strata (social class, sex, age groups). Of course, a regional dimension is attached to sub-group discursive style, in the sense that it integrates dialectal linguistic features. It appears then that in order to account for the Berlin way of telling stories, 1) research should concentrate on lexicon and phonology, as traditional dialectology has always done, but also on specific rhythmic patterns found in discourse 3 , and 3

Interestingly enough, the German colleagues participating in the conference all noted that while reading the Berlin narratives examples, they thought similiar accounts could be found in other regions of Germany but as soon as they listened to the tapes, they changed their mind. Rhythmic segmentation and intonation are probably largely responsible for that.

Discourse Analysis in Sociolinguistics

157

2) comparative research should be carried out in different areas of Germany. The lack of comparative controls weakens the claims about speech style made by the Berlin goup. Let us now turn to another question. .

2. Can the Benefits of Quantitative and Qualitative Studies be Linked at a Theoretical Level 4 Prior to any quantitative treatment of discursive features, comes the unveiling of the organization of discourse at all pertinent levels. In that perspective, looking at the structure of a narrative, identifying speech acts, and pinning down individual strategies in specific contexts are fundamental steps for sociolinguistic research on discourse. Thus, such exploratory studies as those reported by the Berlin group cannot be called qualitative analyses. In many instances, as D A V I D SANKOFF says, 'qualitative' is only a euphemism for 'not enough data'. One study presented by the Berlin group will help me illustrate this point. It concerns the semantic configurations related to the verb "schlagen" ('to beat'). They identified two basic patterns: a passive-recipient model centered on the verb "kriegen" ('to get') and an active model built around the verb "hauen" ('to hit') and synonyms. Using FILLMORE'S case grammar, they found that in both patterns, reference can be made to the victim (dative case), the blow (factitive case), the target (locative case) and the actor (agentive case). Each of these elements as well as the verb can be stressed through the use of intensifiers, metaphors, onomatopoeia, subordinate clauses and so on. In the Berlin corpus of 79 utterances, it appears that the passive-recipient model shows a more complex stress structure than the other model. However, for both patterns, when the speaker and actor are identical, more cases are realized and more elements are stressed. This study of the semantic configurations of the beating action qualifies as a typical empirical quantitative study. The small number of occurrences prevents any statistical weighting of individual constraints but patterns of verbal behavior emerge as structured and variable, two basic components of sociolinguistic analysis. 4

"...wie man eine theoretische Perspektive der Soziolinguistik skizzieren kann, die die Vorteile der quantitativen mit der qualitativen Analyse verbindet." (Dittmar, Schlobinski and Wachs 1985).

158

Pierette Thibault

For the purpose of the present discussion, let us consider 'stress' as a discursive variable. The variants would be: presence and absence of stress elements on each of the actualized cases. In the given example, the constraints groups would be: a) The passive-recipient and the active models; b) the case per se; c) the speaker-actor relationship (identical or not). Another constraint, not mentioned here, might be whether the beating action is part of a structured narrative or not. A quick look at our 1971 Montreal corpus shows that such a constraint seems to apply 5 . Here are three examples: 1) "Je te l'accroche par les cheveux, elle était à genoux devant moi". ('I grab her by the hair, she was kneeling in front of me') 2) "Il m'achalait, il était dans les rangs. On était en rangs puis on montait l'escalier. Je me suis retourné puis pow! Il se retrouve au bas de l'escalier. Le principal m'a appelé. Il avait l'oeil fendu ici." (He was bugging me, he was in line. We were in line and we were going up the stairs. I turned around and pow! He found himself at the bottom of the stairs. The principal called me. He had a cut on his eye here.') 3) "Non je me suis jamais cassé la mâchoire, j'en ai cassé. D'ailleurs j'en ai cassé une il y a deux ans. Un Français." (No. I've never got a jaw broken, I broke some. Matter of fact I broke one two years ago. A Frenchman.') The second example, the only true narrative, shows stress on the victim, on the blow, on the target and on the actor. In the first example, the stress is on the victim and in the last, of course, the accent is on the target. Could a semantic configuration be treated as a variable? Arguing about the functional equivalence of different stress patterns would not be an easy task because of their diversity in forms and syntactic structure. Weak complementarity of the patterns in the social distribution (cf. SANKOFF and THIBAULT 1 9 8 1 ) would be another argument in favour of such an analysis. Again, corpus limitations prevent this type of demonstration. So, the highly debatable question as to whether or not discursive patterns can be treated within the sociolinguistic variable framework cannot be solved here. What is needed is comparative data, and large amounts of it. 5

We were fortunate enough to obtain numerous accounts of fights, since one of our interview questions was: "Were there neighborhood fights where you grew up?"

Discourse Analysis in Sociolinguistics

159

Going back to the association of quantitative and qualitative results at a theoretical level, I cannot see why a theory could not integrate different types of analyses. Life histories and content analysis centering on attitudes definitely contribute to a better understanding of linguistic behavior. SCHLOBINSKI'S dissertation is a good illustration of that. As to the possibility of accounting for a speech style from a qualitative angle, it is no less ambitious (not to say questionable) than the qualitative account of other aspects of linguistic behavior.

3. The Montreal Research Group Approach to Discourse When listening to tapes and reading transcriptions of natural speech samples, one encounters grammatical connectors that hardly connect anything, lexical items with no apparent referential meaning and with prosodie features, such as intonation patterns and pauses that no linguistic description can account for. At the present stage of our work, these are the discursive elements our research group is concerned with. We proceed with discourse as we do with other sociolinguistic business, that is: we look at the distribution of discursive markers, identifying possible structural and contextual constraints. Our general approach to discourse could be labeled as a text grammar approach. In other words, rather than focusing on individual discursive strategies or interactional patterns, we concentrate on the internal structure of discourse, not leaving aside, of course, contextual conditioning and social distribution. With regard to the study of discursive markers, we have two specific research objectives: 1) to make a typology of discursive markers according to their function; 2) to describe structural features of different discursive units such as the information unit (question-answer), the argumentation unit (comments, judgements), the story unit (narrations, descriptions). D I A N E VINCENT has already achieved part of our objectives in her 1 9 8 3 dissertation. She identified three types of discursive markers: structure signals such as connectors, interaction markers, and punctors or prosodie markers. Since she concentrated on punctors, work needs to be done on other types of markers. She also analyzed the distribution of punctors according to the above-mentioned discursive units, although she had few formal criteria to justify her divisions. In order to facilitate the identification of discourse structural features, special attention is being given to the transcription of our 1984 corpus.

160

Pierette Thibault

Under the leadership of DIANE VINCENT, transcribers use a punctuation code to signal pauses and phonetic lengthening. They also carefully transcribe all the back-channel due to the interlocutor. Contextual elements such as baby cries, dish-washing machine noises, the entrance of an additional person are indicated in a parenthetical comment. In our case, getting comparable data is not a problem, for both our 1971 and 1984 corpora are based on semi-directed interviews. Moreover, in both sets of interviews, people talked about residence, work, education and language. With the help of a formal discourse model and formal discourse features, we hope to be able to account for some aspects of discourse variability. An important question remains to be discussed. Given our approach to discourse, what is to be expected from the comparison of our two corpora? Again VINCENT'S work has given rise to interesting questions. Her study was based on twelve interviews, each showing an average of 255 occurrences of punctors. One general tendency is that long 'turns' contain proportionally more punctors than the short ones. One uppermiddle class man and two young men did not seem to follow the pattern. The upper-middle class man had long turns with very few punctors and the two young men, although coming from quite different social backgrounds, showed short turns filled with punctors. Do young people use more punctors than older ones? Do upper-middle class people tend to use fewer punctors than the other social groups? And what about other discursive markers? If there are speech styles characteristic of age and social classes, differences should show up in individual behavior between 1971 and 1984, especially for those who were teen-agers in the seventies and those who have had upward social mobility. The use of discursive markers is the first aspect of discourse we are dealing with but perhaps someone from our group might be interested in looking at variable stress patterns in conflict narratives, following the initiative of the Berlin group. In any case, we are quite optimistic about future developments because other researchers, s u c h as OLYNYK et al (1983), ERMAN (1986) a n d SCHIFFRIN ( 1 9 8 2 ) , h a v e

taken an approach similiar to ours and so far, their results integrate well with VINCENT'S.

Chapter 7 STEPHEN C . LEVINSON

Conceptual Problems in the Study of Regional and Cultural Style

0. Style and linguistic variation These comments are specifically addressed to the study of Berlin style and register by W A C H S , S C H L O B I N S K I & D I T T M A R , but since that study raises quite general issues about how the study of distinctive verbal styles should be conducted, I shall try to draw out the general methodological problems that beset any study of this kind, and for which, at present, we seem hardly to have adequate solutions. I shall concentrate on the attempt made in that study to characterize 'Berlin wit', or the mixture of verbal aggression and inventiveness that seems characteristic (and indeed is stereotypical) of Berliners. 1 The authors are to be congratulated for insisting that we pay attention to phenomena that we are inclined to brush under the rug because we don't really know how to study them. They remind us forcibly that there is more to language variation thanjust the superficial phonological and morphosjntactic variables that have been the focus of sociolinguistic studies of urban dialects. The additional elements are, to put it loosely, patterns of language usage, that is, patterns of message-construction, of preferred syntactic constructions, of verbal routines; in short, in a very broad sense, matters of style and register. However, good hunches can have disasterous theoretical consequences, as witness B E R N S T E I N ' S ( 1 9 7 1 ) attempts to conceptualize class-stratified patterns of language use (ably criticized by C O U L T H A R D ( 1 9 6 9 ) , D I T T M A R ( 1 9 7 6 ) and others). It is therefore important to try and get conceptual 1

This paper is an expanded version of the oral presentation, taking into account comments of the participants, which were most helpful.

162

Stephen C. Levinson

clarification in this area. However, this is not easy, as is made clear by the uncertain and various uses of the terms 'style' and 'register', to which we should turn.

1. Muddles in the models: what should we mean by 'style'? Sociolinguists are of course not the only scholars interested in the characterization of style, and thought on the subject in literary analysis and rhetoric goes back at least as far as Aristotle's Rhetoric. But perhaps I may be forgiven for synthesizing (and radically oversimplifying) what one might term a traditional approach to style (the little book by N . E . E N K V I S T , 1973, is very helpful here). On this view, we can perhaps characterize the nature of style along the following lines: 1. Style is about form and not content, a distinction that presupposes that we can compare different ways of saying the 'same thing'. 'Form' is to be characterized across all the linguistic levels, from paralanguage, to prosodies, to phonetics, to phonology, to morphology, to syntax and lexicon (as for example, in CRYSTAL & D A V Y , 1 9 6 9 ) . 2. Style is largely a matter of tendencies, rather than strict co-occurrence constraints. It is thus to be characterized as a distinctive set of preferences in 'form', or a distinctive cross-level assemblage of prosodic, phonetic, lexical and syntactic features. 3. Style is explicitly, or more often, implicitly, an inherently comparative concept. Any isolation of a specific style presuppose a benchmark norm against which that style is observably distinctive. 4. Styles correlate with contexts, whether those are viewed primarily in terms of the producer (or author), the audience (or interlocutor, or reader), or the social situation (or event). All of this is eminently sensible, and close to the ordinary ('lay') use of the term 'style', which is, after all, part of our everyday metalanguage. But from a sociolinguistic perspective, this traditional concept of style will hardly serve to distinguish style from other linguistic and sociolinguistic concepts like dialect, diglossic level, genre, or code as in code-switching. We therefore turn to the sociolinguistic literature in search of a technical concept of style. But there seems to be some very fundamental confusion, or at least disagreement, in that literature about what we should mean by 'style' and 'register'. First, there is a relatively well-defined notion of style in correlational sociolinguistics, namely the Labovian concept of style as the linguistic reflexes of degree of attention to speech ( L A B O V , 1972a: Ch. 3 ) .

Conceptual Problems in the Study of Regional and Cultural Style

163

This has been operationalized, for the purpose of sociolinguistic interview techniques, as the distinctions between reading styles, word-citation styles, and more casual styles of varying degrees (as elicited by asking for certain kinds of absorbing narrative). There are reasons to doubt that the linguistic reflexes thereby induced are really unitary phenomena; for example, DRESSLF.R & W O D A K ( 1 9 8 2 ) , in a study of Viennese German, were able to tease apart the 'natural' phonological processes of fortition from the switch to prestige phonological variants, both to be found in 'formal style' (high degree of attention to speech). In any case, this notion of style, although relatively clear, is also relatively irrelevant to our problem, the characterization of distinctive patterns of language-use, message construction and verbal routines. For one thing, this notion of style is unidimensional (from 'casual' to 'formal') while the phenomena we wish to charaterize are multidimensional (e.g. 'aggressive' vs. 'friendly' style, as well as 'preaching style' vs. 'disc-jockey style', etc.). Secondly, since all speakers move along this one dimension (so that one can talk of shared 'channel cues' for 'casual' speech, for example), it is not clear how this concept of style could be used to pick out, e.g., the distinctiveness of 'Berlin wit'. In short, the Labovian notion of style is tangential to our problem. It may be related, in that certain styles in our broad sense may only or mostly occur in certain styles in the narrow Labovian sense, but we can expect no greater relevance than that. 2 We turn therefore to another 'technical' sociolinguistic concept, register, which is often equated with style. Halliday, who seems to have originated the sociolinguistic use of the term, characterizes register in opposition to dialect thus: while dialect is variation correlated with users, register is variation correlated with uses ( H A L L I D A Y , MCINTOSH & STREVENS, 1964). However, we see immediately that this definition effectively makes 'Berlin wit' a matter of dialect, and in no way helps us to distinguish a special kind of linguistic variation or patterning, let alone characterize it. One of the problems here, of course, is that all sociolinguistic concepts are unholy mixtures of formal (linguistic) and functional (social/contextual) criteria — this is no less true of concepts like 'language' ("a dialect with 2

There is a different concept of 'style' introduced in LABOV & FANSHEL, 1977, which may be of more relevance. In an analysis of a therapeutic interview, while noting that style in the narrow sense is held pretty much constant throughout (p. 115n), they point out that one can detect switches from 'narrative style' to 'interview style' to 'family style' which correspond with 'fields of discourse'. But although we are offered phonological and lexical features of these 'styles' (p. 129f.), no clear characterization of this additional concept of style is given.

164

Stephen C. Levinson

an army, a navy and a flag" as the saying goes), than of the less clear notions with which we are grappling. Thus a textbook treatment of the distinctions between dialect, diglossic level, lexical register and style might be based on a number of binary formal and functional criteria, as in Table 7.1. Here Halliday's functional distinction between variation with respect to 'use' vs. 'users' provides one dimension of contrast, while other dimensions are given by (a) the formal distinction between rule-governed (systemic) pattern vs. mere preferential usage, and (b) the formal distinction between patterns that have to be stated across linguistic levels and those that can be stated at just one level (as perhaps with 'argot', 'jargon' and lexical register). On such a treatment, style will be defined as a distinctive preferred pattern of use within a set of rules (at all linguistic levels) that specify a dialect, which correlates more with uses than users. Dialect Functional: use (vs. users) Systemic: rules usage Levels: cross (vs. uni-level)

Accent

_ +

•

Diglossia

Register

Style

+

+

+

+

+ +?

-

-

-

+

—

+

—

•

+

Table 7.1: Features of the variation of style

This is all laughably inadequate in relation to the complexities of the phenomena, and in any case the notion of a variable rule will render the distinction between rule-governed pattern and usage-preferences quite obscure. One response to these difficulties is to abandon the formal criteria altogether, and define style and register purely on functional grounds, as any kind of distinctive linguistic variant appropriate to, or correlated with, a specific kind of social situation (viewed, for example, in terms of HYMES (1972) characterization of speech events). We would then talk about individuals' linguistic repertoires (GUMPERZ, 1 9 6 8 ) as consisting of stylistic variants used in specific speech events. Such a notion of style would not contrast directly with dialect, digossic level, or even language, etc., since any of those could effectively function as stylistic variants. Although such a purely functional view is certainly a useful perspective, it does not solve our problem of course — how to define a notion of style or register (or call it what you will) that will capture patterns of language

Conceptual Problems in the Study of Regional and Cultural Style

165

use that are not part of a dialect but are typical and distinctive of speakers of a dialect. There is, finally, a sociolinguistic notion of style based directly on the traditional notion, namely the view of style characterized in terms of choice between linguistic alternates (on the paradigmatic dimension) and in terms of limitations of choice specified as co-occurrence constraints (on the syntagmatic dimension), a view crystallized nicely by ERVIN-TRIPP ( 1 9 7 2 ) . This view shares with the traditional view an inability to make the distinctions between dialect, diglossic level, style, etc., which led us to turn to the sociolinguistic literature in the first place. The point of rehearsing these difficulties is just to make the point that sociolinguistics does not provide a clear notion of style and register (at least, not of the desired kind). The absence could, and should, be seen as a severe failing. I wish now to pursue just one of the underlying sources of conceptual confusion in this area, a conceptual problem which I feel is especially pertinent and troublesome to the Berlin study.

2. A conceptual dilemma: two views about the nature of sociolinguistics Conceptual confusion about the nature of style can in part, I shall argue, be attributed to a more general sociolinguistic schizophrenia, a deep ambivalence between two rather different ways of thinking about linguistic variation. The dilemma is which of these ways of thinking is right. One horn of the dilemma is a view of sociolinguistics that makes absolutely central the concept of the sociolinguistic alternate (as e.g. in ERVINTRIPP, 1 9 7 2 ) . A sociolinguistic alternate is, of course, one of two more linguistic forms that are in paradigmatic opposition and express the same meaning or at least perform the same function (there is an important slippage between moving from 'meaning' to 'function', to which I shall return). Classic examples are Labov's sociolinguistic variables (class-stratified realizations of phonological and morpho-syntactic variants), Brown and Gilman's polite vs. familiar pronouns of address (and indeed all address forms), Fergusons's diglossic levels, even Gumperz's codes in code-switching situations. Why should the notion of sociolinguistic alternate be so central? The rationale, presumably, goes like this: if two variants VI and V2 express the same meaning M l , then any difference in value or use must be purely social — that is, be attributable to contextual factors (like speaker's social identity, or the relationship between speaker and addressee) rather than to

166

Stephen C. Levinson

message content. If this is indeed the rationale, it is not actually all that clear, because it will only be by the definition of what counts as 'meaning' that Ml will exclude an indication of, say, speaker identity. However, an example will immediately make the idea plausible again: suppose VI is ain't and V2 aren't, and that in the appropriate linguistic environments they both mean 'are not' (there'd be a parallel but distinct pair ain't and hasn't, etc.; see CHESHIRE, 1 9 8 2 ) . But suppose further that ain't expresses emphatic negation, so that it is, for example, the preferred way to deny propositions put forward by an interlocutor (this may be counterfactual for working class (WC) British speakers, but is possibly correct for those middle class (MC) speakers who use it occasionally). Now suppose we find (as Cheshire has) that VI is favoured by WC speakers. Since VI is (on our hypothesis) emphatic, we do not know for certain whether WC speakers are simply more emphatic more of the time (perhaps accusations are more common and so the corresponding denials are too), or whether VI is a true sociolinguistic alternate, directly indicating WC speaker identity. For this reason, two variants VI and V2 only count as sociolinguistic alternates if they express exactly the same meaning ML. Now, as pointed out by LAVANDERA ( 1 9 7 8 ) and others, this criterion of synonymity of variants makes difficult the extension of the notion of sociolinguistic alternate from phonological variables to morphosyntactic variables, where subtle shades of meaning difference may distinguish between the variants. 3 Lexical alternates will be even harder to match exactly in meaning, let alone matters of style and register operating simultaneously at many different levels of linguistic realization. Further, as pointed out in painstaking detail by LYONS ( 1 9 7 7 ) , the notion of synonymity is itself not a straightforward concept: we must distinguish synonymy in all contexts (perhaps jumper¡pullover) from synonymy in some contexts (big house ¡large house) but not others (big sister ¡large sister), cognitive synonymy from affective or connotational synonymy (horse¡steed), and so on. And however we define the notion of synonymity it must exclude 'social meanings' while not referring to them (that would make the definition of 'sociolinguistic alternate' circular). A further difficulty is the extension from sameness of meaning to sameness of function. It seems plausible enough: just as the French may indicate politeness by a choice of a polite pronoun Vous, so the English 3

Unless we exclude non-segmental phonology of some kinds, the notion of a phonological variable won't be that clear either; contrastive stress, for example, will make a meaning difference between two variants of the same vowel.

Conceptual Problems in the Study of Regional and Cultural Style

167

may choose a complex request form like Would you by any chance be able to change this pound note (as opposed to, say, Please change this pound note). What stays constant now is not meaning in the narrow sense but illocutionary (or at least perlocutionary) force. Similarly we might compare hey vs. excuse me as forms performing the same function, a summons, although it is far from clear what meaning, if any, hey (for example) has. Instead then of requiring sameness of meaning, we will demand identity of function, and using this criterion we can ask how various identical functions (they might be requesting, summonsing, greeting, etc.) are differentially realized by persons of different social identity, or persons of the same social identity in different social situations, etc. Plausible though the extension seems, it is fraught with difficulty, for if functions are allowed to become very abstract (like 'having a chat') almost anything can be subsumed under them as alternates. I raise these difficulties for what I shall call the alternates view because they do not seem to have been properly thought through. Nevertheless, I assume there is something basically right about this whole approach to the subject, even if it is not entirely clear why! On this view, sociolinguistics is the study of different realisations of the same meaning or function, and the study of style and register would be the study of different ways of saying the same thing within a dialect, a repertoire, or other restriction. Let us turn now to the other horn of the dilemma, a view of sociolinguistics that would make absolutely central the cultural relativity of linguistic functions. This view is familiar from the work called 'ethnography of speaking', instigated especially by HYMES ( 1 9 7 4 ) , where the emphasis is on the cultural uniqueness and distinctiveness of speech acts, speech events and verbal routines. Again, to see the plausibility of this view, consider an example: the Tamils of South India have a special kind of spirit possession, where roving forest demons inhabit the body of some luckless woman, who is then brought to the temple of a particular deity to be exorcised. The resultant speech event is very interesting for a theory of communication because the woman's speech is not attributed to her, but to the demon, and words addressed to her carry the honorifics appropriate to the possessor and not to the possessed — in short the notions of speaker and addressee are shown to be more complex than one might have thought (LEVINSON (in press)). The associated speech acts are specialized speech functions; invocations, spells, exorcisms, etc. Clearly this is exotic enough, and distinctive by virtue of the absence of any thing like it in our own culture. We can easily enough find the reverse case, as for example the notable absence in some cultures (as among Guugu Yimidhirr speakers of

168

Stephen C. Levinson

Cape York; BROWN & LEVINSON, in prep.) of the kinds of greetings and partings that punctuate interactions in our own culture. In short we find that tribe A has speech function F1 and tribe B has no corresponding function, but a unique function F2. F1 and F2 are quite incommensurable, but are each interesting in their own right. The study of distinctive verbal routines (as in LABOV'S (1972b) Black American 'rules for ritual insults', or IRVINE'S (1974) elaborate Wolof greeting formulae) would fall within this way of thinking. So here is another paradigm: sociolinguistics is the study of the cultural distinctiveness of speech functions. There is no necessity for the comparision of two or more different realizations of the same function or meaning — we are interested precisely in the unique and incomparable. I now suggest that the Berlin study of style and register falls directly in the middle of the two horns. On the one hand, Berliners can be said to have an aggressive style by virtue of, for example, the unmitigated realization of requests (i.e. for the same speech function, requesting, there are a number of alternate possible realizations, and Berliners chose a distinctive variant). On the other hand, Berliners seem to have some distinctive speech functions, for example, they appear to pick quarrels and bluff their way through them, and then boast about them, in a way that is perhaps as culturally specific as Labov's Black American verbal routines of ritual insult. The question is which horn to choose? It is an important choice because the two views about sociolinguistics have quite different methodological consequences. Under the first view, which focuses on the notion of the sociolinguistic alternate, the following methodological procedures are applicable: (a) If one is to compare VI with V2 as alternative realizations, it is essential to check for sameness of meaning or function (even in the most empirical work, this is done intuitively). It is essential to check that they are not in 'complementary distribution', i.e. it is essential that they both could have occured in the same syntagmatic slot, i.e. are not determined by collocation or syntagmatic rule. When we move from sameness of meaning to sameness of function, intuitions of sameness are more shaky: we must then attempt to standardise the speech functions that we are interested in, by looking at highly constrained situations. BROUWER et al. ( 1 9 7 9 ) , for example, compared men's and women's relationship of the same speech functions in a railway ticket office. (b) Ultimately, though, we have no great interest in the speech functions differentially realized, except in so far as we need to understand them in

Conceptual Problems in the Study of Regional and Cultural Style

169

order to be sure that they are held constant while we vary the realizations. So, for example, we would have no great interest in the underlying sequence of speech actions in the Berliner's reported quarrels: we are interested only in how the threats, insults, etc., are realised (in the Berlin case, realized in what we may call 'Berlin brutal style'). (c) If VI and V2 realize the same speech function, we must now look for the social value of each realization (or connotation to use Bierwisch's terminology) by examining the contexts of use. We are therefore obliged to use techniques of comparision and social distribution (does speaker A of social category 1 use the same variant as speaker B of social category 2, etc.). Of course, it is no good comparing speakers A and B where they do not belong in some important way to the same speech system, or speech community viewed exactly and only in the way Labov insists, as a community united by shared speech values or connotations (cf. Schlobinski's (or Gumperz's) definition in terms of shared behaviour — useless in this context). For the method is strictly Saussurean: VI and V2 may not contrast at the level of meaning, in some strict sense, but to be sociolinguistic alternates they must indeed contrast for the body of speakers who use either V1 or V2 or both at the level of value or connotation or social significance. In the case of Berlin style, then, we must compare our 'brutal' realizations of requests, threats and insults, in two kinds of social populations; first, we need to know the distribution of these realizations throughout the Berlin community, East, West, male, female, young and old, working class and middle class, etc. Perhaps 'Berlin brutal style' is internally distinctive, associated with middle aged lower middle class male proprietors, and is in opposition to, say, delicate female speech, or upper middle class punctilious courtesy. Secondly, though, we need to bear in mind that the style might be contrastive in another larger speech community of values; for example, it might be in symbolic opposition to peasant taciturn politeness in the immediate rural environs, or to rival realizations in the nearest great cities. In this case; the opposition would be 'insiders' vs. 'outsiders', as in the Martha's Vineyard study ( L A B O V , 1972a: Ch. 1 ) . But there is no particular point in comparing Berlin style to, say, Mannheim style (KALLMEYER & K E I M , this volume) without a theory about how these two communities might be in symbolic opposition within a speech community of shared evaluations of stylistic variants. (d) Finally, there is no special interest, within the alternates perspective, in matters of motivation-, it is silly to ask why Group A uses variant VI and Group B uses variant V2, and not vice-versa, for in the best Saussurean sense VI and V2 are arbitrary carriers of social value. This does not, of

170

Stephen C. Levinson

course, mean that an interesting historical account cannot be given associating groups with variants, as has indeed been done, e.g., for the New York Ir/; but this account will only make clear, by the typical oscillations that variants have as positively valued in one era and negatively in the next, how essentially arbitrary the values of each variant are (LABOV, 1972a; DOWNES, 1984: 112ff.). Turning now to the other view, the approach from the ethnography of speaking, we can see that the methodological consequences are quite different: (a) Instead of looking for sociolinguistic alternates, we should look for verbal routines and stylistic elements that are distinctive of, and possibly unique to, Berliners. We do not need to control for sameness of meaning or function: we are interested precisely in distinctive functions. (b) By the same token, we are interested not just in surface realizations of underlying meanings or functions, but in the underlying meanings or functions themselves. Thus, studying the reported quarrels which, it is claimed, make up part of Berlin style, we would be interested not just in how the reported threats, etc., are realized, but also in the way the narrative is constructed at the level of speech acts and their chaining — perhaps there are special sequences of speech actions distinctive of the style. (c) On this approach, there is no essential methodological reason (as there is on the alternates approach) to be especially interested in matters of comparison and distribution. For example, the description of Tamil spiritpossession does not essentially rely on a comparison of spirit-possessions in different milieu or by different practitioners; we only need to know whether some performance is more or less representative of a class of events. Nor do we absolutely need to know exactly who gets possessed or why; the social distribution of possession is a secondary question, best left to a mode of analysis that mixes psychoanalysis and the study of stress in social relationships. In fact, Tamil spirit possessions are both highly visible and distinctive affairs and highly restricted by time, place, personnel and event. Their distributional rarity is quite beside the point. In an exactly similar way, some of the extreme exemplars of 'Berlin wit' might be highly visible, stereotypical and distinctive, yet only exercised by forty year old male proprietors (as in the recorded examples)! 4 (d) Unlike on the alternates approach, on the ethnographic approach we have every reason to be interested in the social motivation of distinctive 4

Some aspects of this way in which 'high profile' speech habits may lead to the attribution of cultural 'ethos' by outside observers are discussed in BROWN & LEVINSON, 1978: 248ff.

Conceptual Problems in the Study of Regional and Cultural Style

171

realizations and functions. It makes perfect sense to ask why the natives have some particular verbal routine, to ask what social purpose it fulfills and how it fits into the details of the local social organization. Indeed, on this view, this is an essential series of questions. There is no assumption of arbitrariness, crucial on the alternates view. So, which is the right approach? We had better make up our minds, as we shall go about our investigations in quite different ways depending on the line we take. I suppose most sociolinguists, like myself, will find our Labovian instincts (which after all are essentially based on the Saussurean or structuralist axiom of contrast within a paradigmatic set) in favour of the alternates approach. But it is interesting to consider what happens if we suppress them. In this regard, an abortive study of my own may be instructive. Emulating LABOV'S (1972a) ingenious study of variables in New York department stores, we set out to study whether there are any class-stratified aspects of pragmatics. In a small English city, a wired-up investigator went into twenty seven shops, all specializing in male attire, carefully class stratified on the basis of objective criteria, including price of goods, location, advertising and style of window display. It was possible also to take into account not only the sex, age and dress of the shops assistants, but also their accents, from marked hyper-RP, through RP, to local regional. The investigator attempted to use the same initial greeting and request forms in all the shops, viz, "Hello, I was looking for a pullover". We were hoping, as on the alternates approach, to hold function constant in this highly restricted setting, so that different class-stratified realizations of greetings, requests, offers of service, etc., could be studied. This method worked to some limited extent in some of the shops, and we found reliable class-stratified differences in some of the opening moves. Thus, the entrance of the investigator (potential customer) operated as a summons, to which the shop assistant responded with an answer which doubled as a greeting, and an offer to serve; in high class stores, the token was 'Good morning' (or 'good afternoon' as appropriate), in middle ranking stores it was more often 'Hello, can I help you', in lower ranking stores the greetings were more often absent as in 'Can I help you?' (even more often, there was no verbal response at all, as I shall describe). The address forms that accompanied these predictably varied from "sir" at one end of the spectrum to "love" or "dear" at the other. At this point, our investigator said his prepared line, "Hello, I was looking for a pullover". And at this point too our study foundered, for there were two quite different interpretations of this utterance: in the higher ranking stores it was taken

172

Stephen C. Levinson

to request the initiation of an exclusive dyadic service interchange (one might gloss the apparent implicature as "Please help me select a pullover"), while in the lower ranking stores it was taken to be simply a request for information, receiving responses of the kind "Over there behind the jackets". And from there on, the verbal interactions in the two kinds of store diverged in incommensurable ways. What had gone wrong of course was that, despite the restricted setting, the kinds of speech functions relevant to shops of different kinds had diverged and become irreconcilably different. For in the higher and middle ranking shops there was a quite different definition of the activity or speech event: customer and assistant expect to be paired in an intense and prolonged verbal interaction which will only cease when the customer leaves, satisfied or dissatisfied. In the lower ranking stores, there was no such expectation, for they are organized on a self-service basis; customers can enter, look through the merchandise and leave without a word having been spoken. Even when interaction is initiated by the customer, there are no greetings, and minimal help is given. From this basic distinction between service-stores and self-service stores (and there are gradations in between), all sorts of structural concomitants follow: the goods are displayed differently, access to goods is restricted or open, but above all the roles of customer and assistant are differently defined, so that in the higher ranking store the customer expects help, advice and even aesthetic judgements, while in the lower ranking store he can expect only to have goods pointed out, and to be suspiciously surveilled. From these structural facts, the quite different and incommensurable speech functions attended to in verbal interaction naturally follow. With such divergence in function, there was no hope of studying variant realizations of the same function. And so, in the grip of the alternates approach, I abandoned the study. But I now wonder if this was not a serious error. For if we suppress the differences of speech function, and restrict ourselves to studying circumstances which are strictly commensurable, what we are doing is divorcing sociolinguistics from the study of social practice. For what could be a more direct and systematic relation between social class and language than the chain of connections through class, the economics of different kinds of shops, the consequent differences in storeorganization, to the kinds of verbal routine appropriate in the different kinds of shop? Perhaps we could formulate a counter-manifesto to the alternates approach along these lines: the alternates view, by insisting on studying only variants of identical meanings and functions (in which it has no intrinsic interest), threatens to sanitize and sterilize sociolinguistics,

Conceptual Problems in the Study of Regional and Cultural Style

173

divorcing it from the study of social life and the different social worlds that construct different speech functions. 5 I have tried to argue that there are pro's and con's to both the alternates approach and the ethnography of speaking approach. In the study of style, I think we must be open to both approaches, but given the divergent methodological consequences of each approach, we cannot afford to be carelessly eclectic. It is essential to be clear about whether, in a particular case, one is handling phenomena that are best treated as different realizations of the same function, or are best treated as distinct and incommensurable functions.

3. Application of these remarks to the Berlin study 3.1 Vacillation between the two views of sociolinguistics? In the report on Berlin style and register, W A C H S , SCHLOBINSKI & DITTMAR argue that Berlin 'loud-mouthed' style was especially evident in their tapes in a recurrent genre, where the speaker reports a quarrel or argument in which he defeats an opponent. (It is important here of course to distinguish reported arguments or quarrels from real ones - a matter to which I'll return.) In this study the authors employ the following methods: (1) Just half a dozen reported quarrels from just two speakers constitute the data. Such a qualitative rather than quantitative selection will be fine on the 'ethnography of speaking' view outlined above, but it is likely to be judged inadequate evidence on the 'alternates' view. For, on the latter perspective, we require comparative evidence for the distinctiveness of this particular realization or variant of a speech function: we want to know about what this variant contrasts with, and the distribution of all these different realizations throughout the social population for whom the various variants are in paradigmatic contrast (i.e. within the speech community in the strictly Labovian sense). (2) The characterization of the reported quarrels employs the speech act categories of LABOV & FANSHEL ( 1 9 7 7 ) . If we ask what the purpose of these categorizations are, we would obtain different responses according 5

An example of the kind of work that would be legitimated by our counter-manifesto would be the well-known analysis by EHLICH & REHBEIN 1972 of the way in which verbal interaction in a restaurant is structured by the economic, social and ideological factors that lie behind the institution of the restaurant itself.

174

Stephen C. Levinson

to which view of sociolinguistics was adopted. On the alternates view, we have no intrinsic interest in the speech acts or functions at all, for we are interested only in distinctive realizations of the same functions; the only point in assigning speech acts to utterances would be in order to hold functions constant while comparing realizations. On the ethnography of speaking viewpoint, though, we should be interested in the functions themselves, either on the grounds that they might individually be distinctive and unique to Berliners, or because they are 'chained' in distinctive and uniquely Berliner sequences of speech acts. The authors though are mute on the rational behind their employment of the LABOV & FANSHEL categories; they are caught, I suspect, on the horns of our dilemma. (3) The mode of expression of certain speech acts is carefully noted. For example, one source of the impression of aggressiveness is the 'aggrevated' realization of request and threats, in contrast to the normal 'mitigation' of such speech acts (for three different kinds of account see LABOV & FANSHEL, 1977: 84ff.; BROWN & LEVINSON, 1978; LEVINSON, 1983: 332ff£). In employing this kind of methodology, comparing contrastive realizations of the same speech acts, we seem to be firmly within the alternates perspective. (4) Finally, they relate the 'Berlin brutal style', as I have called it, to the competitiveness of the city-dweller, and the associated importance of negotiated social roles. In other words, the authors assume that the style is not arbitrary, but is rather motivated by the social conditions of inner city Berlin life. Here, by contrast, we seem to be equally firmly within the ethnography of speaking perspective, since it makes no sense to seek synchronic motivations for arbitrary sociolinguistic alternates. If this critique is correct, then the authors have failed (just like the rest of us) to maintain a clear conception of the beast they are busy chasing, namely speech style. One must at least attempt to decide which aspects of Berlin style are best approached from the alternates perspective and which from the ethnography of speaking perspective, and then apply the methods which are uniquely relevant to that perspective. Meanwhile we do not know whether to castigate them for, on the one hand (assuming the alternates view), insufficient data, lack of comparative variants, lack of information about the social distribution of such styles of speech, or, on the other hand (assuming the ethnographic view), for the absence of clear claims about uniquely Berliner speech acts or sequences of speech acts, and an overly sketchy attempt to describe social motivations. Until we make up our minds, we shall not be able to recommend the direction for future research.

Conceptual Problems in the Study of Regional and Cultural Style 3.2

Use of the

LABOV &

FANSHEL

175

speech act categories

I would like now to attend to a rather specific issue, namely the pertinence of the Labov & Fanshel speech act categories to the sociolinguistic analysis of style and discourse. I have elsewhere spelt out theoretical reservations about the whole approach (LEVINSON, 1983: 286ff.), and I want here merely to point out some practical problems in application. Recollect that the theory of speech acts defines particular illocutionary forces in terms of certain distinctive felicity conditions (henceforth, FCs); where a linguistic expression is used to perform an illocution other than that indicated by its syntactic form, we talk of 'indirect speech acts' (SEARLE, 1969, 1975; reviewed in LEVINSON, 1983: Ch. 5). The distinct contribution of the Labov & Fanshel framework is to recast this account in terms of inferential rules; if (i) speaker S addresses addressee A with linguistic form F, and F is sometimes used to perform illocution I, and (ii) I has the FCs 1 to n, and (iii) the FCs 1 to n obtain (i.e. the context satisfies the conditions), then A may infer that in saying F, S meant I. It is this third, italicised, condition that Labov & Fanshel emphasize: to find out what speech act an utterance performs, check to see whether certain conditions are met. Now, if such inferential rules can be made to work, they would, needless to say, give us a powerful account of how we interpret utterances as doing things, like requesting, threatening and so forth. I believe, however, that it is almost impossible to formulate such inference rules in a way which will yield any generality of application, without begging all sorts of questions. The Berlin study indicates quite neatly the kinds of problems involved, so let me criticize some of the formulations offered therein. But let me emphasize that my point in singling these out is not to suggest that the authors were in any way remiss or careless in their analysis, but rather to suggest that there are intrinsic difficulties with this kind of analysis. First, the inference rules are phrased in ways that raise many puzzles. Consider Rule (4), which is intended I think to be equivalent to the more perspicuous reformulation: 6

6

The original version, following L A B O V & F A N S H E L , has the form 'B does not believe that A believes that (a) A has the information I, (and) (b) B does not have the information I'. Whether not believing that not p is equivalent to believing that p depends on whether we want to allow for epistemic uncertainty: John may not believe that not p, because he has never entertained beliefs about p, or has doubts whether p.

176

Stephen C. Levinson

If A addresses B with an utterance U requesting information I, then U is to be interpreted as a valid request for information if B believes that A believes that: (a) A does not know I; & (b) B knows I. What this, for the most part, amounts to is a definition of the validity of an illocution, not a characterization of the illocution itself. (The L A B O V & FANSHEL ( 1 9 7 7 : 8 9 ) version, more useful on the face of it, substitutes 'an imperative requesting information I' for 'an utterance requesting information I', thus linking linguistic form to iillocutionary inference.) What is validity? L A B O V & FANSHEL ( 1 9 7 7 : 8 1 ) state they have substituted 'validity' for 'sincerity', to emphasize the social objectivity of the interpretation; thus, they say, a military request for information might be known by the enlisted addressee to be fatuous, but would nevertheless have to be treated as 'valid'. Similarly, in the Berlin study, the attraction of the Labov & Fanshel rules is said to be that they make it "possible to objectify classifications concerning the pragmatic and interactive components of speech utterances" (p. 55). In any case, the purpose, of course, of distinguishing 'validity' from 'invalidity' is to make the distinction between direct (valid) and indirect speech acts (or at least something similar to it): when must the addressee do some additional reasoning to a further illocution? 7 But how is an interpretation to have this 'social objectivity' when the conditions in the rule above are epistemic, a matter of private belief? And how, to be practical, are we to do the analysis of texts, where we have no direct access to the epistemic states of the participants? If the distinction between 'valid' and 'invalid' requests is to be empirical for both participants and analysts, it had better be directly inferrable from the texts. Or, to put the matter in another way, how does B know what A's epistemic state is? The answer is, in at least some cases, reasonably clear: if A asks B "What time is it?", B infers from the utterance that A probably doesn't know what time it is. This is by far the most likely way in which B will infer A's epistemic state. But if this is so, then the key idea of Labov & Fanshel's analysis, that we can use validity or sincerity conditions on speech acts to infer what act is being done, seems to need inversion: assuming we know what speech act is being done, then we can infer the probable subjective states of participants (which, I take it, was pretty much Searle's original position). The advantage of the latter analysis is that it

7

This formed the basis of the GORDON & LAKOFF (1972) analysis o f the 'blocking' of literal forces in the derivation of indirect speech acts.

Conceptual Problems in the Study of Regional and Cultural Style

177

gives us an account of the otherwise ineffable knowledge we apparently have about others' mental states. Of course, it is quite possible that for some objectively ascertainable sincerity conditions (like the preparatory condition on a request to open the door, that there is a door and that it is closed) we partially infer from the conditions to the illocution, while for subjective sincerity (or validity) conditions we infer from the illocution to the conditions! There are many detailed ways in which one could quarrel with the formulations of the inference rules. Thus, for example, we are given lists of conditions, but these are sometimes conjunctive (e.g. in Rule 1), sometimes disjunctive (e.g. Rule 3), and sometimes indeterminate. Secondly, the conditions are often much too strong: for example, a precondition for an insult is said to be that the participants are already in a conflict situation (p. 56), which would make impossible an acount of the role of insults in instigating conflicts; or, a precondition on threats is said to be that the speaker has the ability to do the threatened action (p. 57), whereas the data is full of threats like "I'll slam your head through that cement wall" (p. 79) where it is doubtful that the physical properties of heads and cement make such an action possible. But there are not only difficulties of formulation, there are also difficulties of application. Take the following lines (in English glosses) from example 1 (p. 60), with their assigned first-order speech act categories: 1. A. "What's going on here? (Request for Information) 2. Hold it! Right now!" (Request) 3. B. "There he goes bitchin' again" (Request Refusal) Does utterance 1 actually meet the conditions in Rule 5 (or my version above)? Certainly A knows the information requested (the prior part of the story goes "All of a sudden I see what's going on over there ..."); but does B know that A knows, for if so it is not in fact a valid request? How could we decide? On the ground that no response is waited for (or at least reported), intuitively B does not take it as a valid request for information, in which case the analysts have made the wrong assigment. However, if that is granted, we have brought to bear a kind of information, viz. conversational sequencing, notably absent from the speech act characterizations, and we are perhaps now involved in a different kind of analysis. Similar uncertainties arise about utterance 2 and its categorization: does it really meet the conditions in the author's Rule 1 (p. 55)? For example, does the recipient B have the obligation to do the request thing (condition 4), in this case to cease a specified action? (Conversely, does A have the

178

Stephen C. Levinson

right, as required in condition 5, to so request?) One reading of utterance 3 (and now we are appealing to sequential information again) is that it specifically repudiates these rights and obligations. If so, the conditions are not met and no request has taken place. Perhaps this should be the hub of the analysis of such conflicts: we find 'hung' speech acts where a proposed action can not even get off the ground because the interactional preconditions are contested (see 3.3 below). Yet again, it is not at all clear how utterance 3 meets the conditions for 'Request Refusals' specified in Rule 3 (p. 56). Partly this is because the conditions themselves are unclear (and in part surely wrong - not every request for information following a request can be interpreted as a refusal; see LABOV & FANSHEL, 1 9 7 7 : 9 1 ) . But anyway there does not seem to be a clearly specificable delimited set of ways of refusing requests; B could for example have said just "Hell", or turned his back. Here in fact he made a metalinguistic comment to another interlocutor, and has thereby implicated derision of A; current theories of speech acts cannot handle how remarks addressed to one party can be intended as different illocutions to third parties (as pointed out by C L A R K & CARLSON, 1 9 8 2 ) nor is it clear that they should attempt to do so (LEVINSON, in press). Interestingly, it would seem on intuitive grounds that an utterance of this sort could either be a refusal to co-operate, or a mere 'bitch' accompanying acquiescence how is the analyst to decide? In any case, the point is that the utterance is altogether more interesting than this unprincipled application of the authors' Rule 3 would allow. Finally, we are told of this exchange that two meta-speech acts (my term) can be assigned on the following basis: utterances 1 and 2 count as 'arrogating status' because they are requests in a conflict situation where B believes A has the right to so request (Rule 9); while utterance 3 counts as a 'repudiation' of that status because it is a request-refusal following an 'arrogation of status' (Rule 10; p. 57). But if B believes A has the right to so act, then how can he be challenging that right? And surely the 'arrogation of status' can be the very thing that instigates conflict, as in this example, rather than conflict being an antecedent condition on 'arrogation', as stated in Rule 9. But the main difficulty with these metaspeech-acts is that they seem to be quite different kinds of things from, say, requests, threats and the like. From an Austinian point of view, this whole mode of analysis makes a gross conflation between illocutionary force and perlocutionary effect: illocutions are conventional and limited in number (and thus appropriately designated by rule); perlocutions are

Conceptual Problems in the Study of Regional and Cultural Style

179

non-conventional and unlimited in number and kind, and thus cannot be encapsulated as rules (see L E V I N S O N , 1 9 8 3 : Chapter 5 ) . 8 My point in going through this example is merely to draw attention to the intrinsic difficulties of applying this style of analysis to empirical materials. If the speech act categories are clear and well-defined, then they are too rigid to apply to the flexibility of actual use; and if they are not properly defined, then they merely trade on our intuitions without explicating them. A final issue is whether this whole mode of analysis is not much too particularistic - do we need such ad hoc inference rules, or are there much more general processes at work, which will pari passu account for these specific inferences? Consider, for example, Rule 5 (p. 56), the rule for 'reinstating requests', which states that when A answers a question, which itself follows a request from A, then A should be heard as repeating the request. This rule is really quite unnecessary; the reading follows from the much more general properties of insertion sequences and conditional relevance, namely the principles that establish, for all kinds of speech acts, that when an expected response is not forthcoming what occurs in its place is heard as preliminary to providing that response. There is thus a structural symmetry between: (i) Request (Question - Answer) Compliance (ii) Question 1 (Question 2 - Answer 2) Answer 1 (iii) Invitation (Question - Answer) Acceptance, and so on. In an exactly similar way, if we bring Gricean principles of inference to bear, we may not need many of the Labov & Fanshel rules. For example, Labov & Fanshel have a 'rule of narrative sequencing' (1977: 107), which already has a rival account in the Maxim of Manner (see L E V I N S O N , 1983: 108). Or, consider even the central rule for requests: as we noted above, instead of trying to infer the force from the conditions, we could argue that it is the force that allows the inference of the conditions. Thus, a request from A to B to do X would not even be rational, let alone Griceanly co-operative, unless (i) A wanted X done, (ii) A believed B could do it, (iii) there was a reasonable expectation that, if 8

Another kind of difficulty with Rules 9 and 10 is that they confuse social status with interactional dominance-, social status is about the right and duties that pertain to social identities (e.g., in example I, being a custodian); interactional dominance is a potentially independent dimension. The lay phrase 'social relationship' confuses these of course, but that provides no excuse for the analyst (see e.g. GOODENOUGH, 1967).

180

Stephen C. Levinson

A asks, B will do X. Hence all B has to do in order to infer (i)-(iii) is to assume A is rational, and assume that the utterance in question is a request on the basis of some features of its form and content. Even the Labov & Fanshel social conditions (to which they attach great importance), specifying A's right to ask and B's obligation to comply, would in their turn be derivative from the inferred condition (iii), and a general assumption of rationality, since such social conditions would confer such reasonable expectation. I am not proposing these alternatives with any commitment, but merely pointing out that, given the inherent leakiness of speech act categorizations of the sort illustrated in our discussion of example 1 in the Berlin study, we would do well to search for much more general inferential principles of a more flexible sort. 3.3 A Genre: some properties of quarrels and reported quarrels However, let us return to empirical issues and in particular to the nature of the speech genre that is captured in the examples. Clearly, they fall into some general category of reported quarrels or arguments. Is there something special about that category? One important issue is what the relationship is between real quarrels and reported ones. One can detect at once in reported quarrels elements of self justification and self-presentation, as the authors of the Berlin study point out (as, e.g., in example 3, where the protagonist holds that he instructed his adversary "in a nice and friendly way", p. 71). But are there, for example, features of quarrels that are not reportable? The only way to find out would be to look in detail at the relation between actual quarrels and the narratives describing them, but there have been very few empirical studies of actual quarrels as far as I can ascertain (a notable exception is the work of M . GOODWIN on Black children in Philadelphia, which I shall refer to shortly). 9 So I proffer an example in Appendix 1 (a rare example, I may add, in many hundreds of hours of taped natural talk) 10 , and venture a few unsubstantiated remarks on a subject I don't pretend to understand. 11 In this particular example, two features of vituperative argumentation stand out which also have intuitive generality. The first we may call the ® There are related dimensions of confrontation discussed, inter alia, in TANNEN (1984), LABOV & FANSHEL ( 1 9 7 7 ) a n d ATKINSON & D R E W ( 1 9 7 9 ) . 10 II

At the workshop, LABOV confirmed the rarity for his own corpus. The subject is of interest, of course, to anyone interested in politeness; and it is a matter of regret that existing theories of politeness (BROWN & LEVINSON, 1978) do not automatically predict the properties of rude quarrels! See however 3.4 below for some predictions.

Conceptual Problems in the Study of Regional and Cultural Style

181

'pushdown-stack' nature of quarrels, where A does speech action X, B does not respond directly but replies with the first part of an insertion sequence, A in turn refuses to respond to the insertion sequence, etc. In effect, we have a structure of the following sort, where XI, Y l , etc., are the first parts of adjacency pairs expecting X2, Y2, etc., as immediate reponses: [XI [Yl [W1 [Z1 ... with the notable absence of the responses in the reverse sequence: ...Z2] W2] Y2] X2] A push-down stack structure, then, where the 'plates' (turns) continue to be added to the stack, but are never lifted off again in the reverse order. Such a structure occurs at the beginning of the transcript in Appendix 1, where the request for payment for the developed photographic prints (1. 1-2) is met with an insertion-sequence request ("Can I just see them" (prior to paying), which is met with a refusal, which is in turn challenged with a question (1. 5-8), which is not directly answered, etc. In effect, then, an admittedly imperfect sequence of the sort I have characterized: [Request, [Inserted Prior Request — Refusal], [Questioning of Consequences of Refusal], Justification of basis of Refusal... If this characterization is roughly correct, then it goes some way towards explaining one salient subjective property of quarrels, namely that it is easier to get into them than to get out of them - extrication is difficult in part because one has become embedded so deeply in insertion sequences that one can't even remember how one got into them! It might also explain another subjective property of quarrels, namely their 'jumping connectedness': in performing speech actions XI, Y l , W l , etc., without performing their immediately relevant responses, the parties to a quarrel have left open a large range of pertinent next actions, which may finally get attended to in an apparently random order. Another property of quarrels has been better studied, namely that they are often constructed out of 'rounds' ( M . GOODWIN, in press). A 'round' consists of a sequence of speech actions of like type — e. g. threat — counter-threat, insult — insult, assertion — counter-assertion (or denial), often carefully matched in surface form. For example, in the Appended quarrel (1. 76 ff.), L:

... I am the governor (two thirty please)

[

M: No no Cabinet is the governor L: Two- No Cabinet is NOT the governor

182 If we see that reports, through

Stephen C. Levinson

now compare the reported quarrels of the Berlin study, we can some aspects of actual quarrels are accurately reflected in the and others perhaps are not. 1 2 Thus, the use of 'rounds' shows clearly, as e.g. in Example 4:

A: I'll belt ya right in your trap B: ...I'll wipe my mouth with you, you ape! But the 'push-down stack' character of quarrels is clearly much less reportable, due to the sheer complexity. Nevertheless, there are perhaps traces of the associated 'jumping connectedness', as for example in Example 1, where the irate custodian's instruction to the builders to "hold it!" is not directly dealt with, being met with insults, demands for reason, etc., so that when the custodian issues his Gandhian threat ("I'll sit down in a chair in front of it") it has to be understood as supporting his earlier instruction (the gloss might go 'Either you hold it, or I'll sit down in front of it'). The point in comparing reports of quarrels to the actual things is partly to assess just how the reports are being formulated to show the protagonistreporter in the best possible light, as assessed within Berlin cannons of style. But there are other reasons to be interested in quarrels in their own right. First, they suggest one empirical way of pursuing issues raised in speech act theory: for if, as I have suggested, quarrels often have a 'pushdown stack' structure generated by questioning or challenging preconditions to a prior action, then we should find 'felicity conditions' empirically exhibited in quarrels in a way that we are unlikely to find elsewhere (or at least only in agonistic settings). This method is touched upon by LABOV & F A N S H E L (1977: 93f£), A T K I N S O N & D R E W (1979), W O O T T O N (1981), but deserves full-scale study. 3.4 The 'strategy' analysis of style: 'face' and conversational preference A second reason to be interested in quarrels is that, being rents in the social fabric, they indicate just what kind of cloth it is made of. In particular, to pursue one of Goffman's themes, they reveal the hidden mechanisms of face-preserving techniques precisely by violating them. This 12

KLEIN & LABOV pointed out at the workshop that an important feature that distinguishes actual quarrels from reported ones is the extreme affectual loading in the former. A special feature of the Berlin reported quarrels, only apparent from recordings, is the prosodic 'flatness' associated with low speech energy, with implications of 'damped' affectual. content. Whether this prosodic 'flatness' is characteristic of the quarrels themselves, as opposed to reports of them, we are not told.

Conceptual Problems in the Study of Regional and Cultural Style

183

orientation suggests a rather different approach to the problem of 'Berlin wit' which I shall now pursue. Consider the fact, which any visitor to foreign lands will attest, that interaction styles can vary culturally, from (for example) the combative to the polite, and within the latter spectrum from the saccharine sweet to the austere formal. B R O W N & LEVINSON ( 1 9 7 8 ) tried to relate this systematically to the notion of 'face', and its two essential components, exploiting ideas of Goffman. It is possible to give an account of distinct cultural 'ethos' (1978: 248ff.) in terms of different interactional (and specifically verbal) strategies for dealing with participants' vulnerability to loss of 'face'. We were keen to point out that 'ethos' refers to the dominant style of interaction as observed in public places, and may convey a quite inaccurate picture of whar occurs in more private settings (which raises again the issue of the representativeness of 'Berlin wit'). The Brown & Levinson study is just one example of a set of studies that emphasize the strategic nature of style (see e.g. L A K O F F ( 1 9 7 5 ) , GUMPERZ ( 1 9 8 2 ) , TANNEN ( 1 9 8 4 ) ) . According to such analyses, style is NOT to be viewed as an assemblage of forms with arbitrary social or connotative valuations; for such an approach would be to analyze style within the alternates paradigm for sociolinguistic research, and would have as consequence a description of a style as a mere inventory of arbitrary formal features. Rather, on the strategic view, styles should be seen as systematically motivated, as essentially rational adaptations to certain contextual circumstances. For example, a strategic analysis of the style of legal documents would not consist just of a list of register features (cf. C R Y S TAL & D A V Y , 1 9 6 9 ) , but would also try to show how the forms used were designed to fall within the terms of particular statutes and legal instruments. In a similar way, the polite style of an English request form like: "Excuse me, sorry to bother you, would you mind awfully if by any chance I was to ask you to exchange seats, so that I can sit next to my companion?" is not to be thought of as a basic request form with a heap of conventional 'mitigators' attached. Rather, we should note the careful ways in which a request that is basically presumptuous and imposing is so phrased as to preserve the addressee's 'face': it is preceded by apologies and announcements of reluctance to impinge, and is phrased as a question about whether the addressee would mind in the highly unlikely circumstance that the speaker had the impertinence to make an invasive request despite having good reasons for doing so! Many little details, like the polite past tense,

184

Stephen C. Levinson

the recurrence of phrases like 'by any chance', the polite conditional, etc., of English requests can be seen as perfectly rational features on this account. To round it out, we need to specify the exact principles whereby these inferences about the speaker's attention to the preservation of the addressee's face are calculated or implicated (see e.g. LEECH, 1 9 8 3 ) . This approach contrasts of course with the traditional view that style is a matter of a distinctive or different way of saying exactly 'the same thing'. Face preservation is an important element, we claim, in everyday styles of speaking. 13 If we now combine this view with some recent developments in the analysis of conversation, we shall begin to see another possible approach to the problem of 'Berlin wit'. One notion of importance to recent conversation analysis is the concept of preference organisation (see review in LEVINSON, 1 9 8 3 : 332ff. and papers in ATKINSON & HERITAGE, 1 9 8 4 ) . A 'preferred' response or speech action or sequence of such actions, on this view, is one that, whatever the actual psychological preferences of participants, has a certain characterizable simple format which is statistically predominant over more complex forms. The clearest cases, perhaps, are responses to first parts of 'adjacency pairs' like questions, requests, invitations, etc., which seek immediate responses of particular kinds; here answers can be shown to be preferred (in the special sense) to non-answers, acceptances or compliances to rejections. Thus a typical acceptance of an invitation is prompt and direct, while a refusal is typically delayed by pauses, particles (like 'well'), appreciations, and accompanied by reasons, etc., as in the following (from ATKINSON & D R E W , 1 9 7 9 : 5 8 ) : A: Uh, if you'd care to come and visit a little while this morning I'll give you a cup of coffee B: hehh well that's awfully sweet of you, I don't think I can make it this morning, hh uhm I'm running an ad in the paper and-and uh I have to stay near the phone. Thus preference organization imposes an ordering (as preferred vs. dispreferred) on potential responses to first parts of adjacency pairs, and dictates the likely structure of preferred vs. dispreferred responses. In a similar way, preference organization ranks not only responses but entire sequences of certain kinds. Thus, for example, it has been shown that when a speaker makes an error of some kind, other participants tend

13

Thus, TANNEN (1984) attempts to show that the intense enthusiastic style associated with New York Jews is derived from attention to 'positive face', the aspect of self-esteem based on positive appreciation by others.

Conceptual Problems in the Study of Regional and Cultural Style

185

to wait to see if he'll correct himself (unprompted self-correction being preferred), failing which they may prompt him (other-prompted selfcorrection being next preferred). Least preferred, statistically at least, is for participants to correct the speaker without giving him the chance, unprompted or prompted, to correct himself (SCHEGLOFF, JEFFERSON & S A C K S , 1977). Again, by slight extension, preference organizations can be seen to hold across different possible sequences, as between sequence I and II below: I. A: B: A: B: II. A: B:

pre-request ("Would you be in your office on Monday?") go-ahead ("Sure") request ("Could I come and see you?") compliance ("Yes why not") pre-request ("Would you be in your office on Monday?") offer ("Sure, come around and we'll have a cup of coffee")

Sequence II is structurally reduced, and can be shown to be specifically "fished for" on many occasions (see LEVINSON, 1983: 345ff.). These examples must suffice to illustrate what is in fact a rich and detailed area of conversational dynamics. The relevance of these findings, which are based it must be noted on British and American English and are of uncertain cross-cultural generality, is this: these aspects of preference organization (and there are many more) can be seen in a quite direct way to be motivated by considerations of face-saving (a point made by BROWN & LEVINSON, in press, M . GOODWIN, in press, HERITAGE, 1 9 8 5 and no doubt others). For, to refuse a request or invitation, to avoid answering a question, to disagree with a judgement, etc., could potentially implicate non-cooperation, or a lack of caring for the other or his opinions. And to correct another, might be to suggest incompetence, while to wait for a request when it was already patently in the offing might be to suggest reluctance and lack of interest in compliance. On the basis of this, one might hazard the following prediction: if Berliners are really interactionally 'combative', then in general they ought to indulge in less 'face'-saving behaviour. So, for example, we might expect them: (a) to routinely use dispreferred responses to adjacency pair first parts without the normal dispreferred format (e.g. to refuse requests with a simple 'No!'); (b) to correct other speakers instead of prompting them to correct themselves; (c) to not indulge in pre-emptive offers, or other helpful truncations of conversational sequences. That is to say that a whole slew of interactional features should follow as motivated consequences of a certain lack of interest in face-preservation.

186

Stephen C. Levinson

Although there is only evidence for the first of these patterns in the Berlin data, there is evidence from elsewhere that these sorts of features do tend to occur together as a package, as shown by M. Goodwin's careful work on the verbal style of pre-teenage Black children in Philadelphia ( G O O D W I N , 1 9 8 2 ; in press). Of course, the prediction may be quite wrong. Perhaps Berlin aggressiveness is mere fantasy surfacing only in reports of one's own interactional exploits. But if this account is on the right lines, then it would attest to the utility of (a) the strategic analysis of style, (b) the parameters of conversational organization revealed by conversation analysts (as displayed, e.g., in ATKINSON & H E R I T A G E , 1 9 8 4 ) .

4. Conclusions The intuition that there are distinctive styles of speaking, as well as distinctive accents and dialects, associated with regional and social categories of speakers, is too strong to be entirely wrong. The difficulty, though, is to find a proper way of studying these phenomena. That difficulty in turn, perhaps, is partly due to our vacillation between two different ways of doing sociolinguistics, what I have called the alternates view vs. the ethnography of speaking view. For, as I've indicated, the alternates view would direct one to look for contrastive realizations of the same functions, with the associated methods of controlled comparison and community-wide survey. In contrast, the ethnography of speaking view would lead one to search for distinctive functions, distinctive verbal routines and sequences, and would seek an account of why particular social groups favour such distinctive styles. In the case of 'Berlin wit', perhaps we need both approaches, where following the alternates view we look, for example, for different kinds of realization of requests, and try to delimit the community for which each realization has its distinct value. On the other hand, we need also to consider the possibility of distinctive verbal routines, for example, boasts of argumentative exploits, where the quarrels reported might themselves have a distinctive Berlin pattern. Also, perhaps, there may be a distinctive approach to matters of face-preservation, which may be shown to directly motivate many features of Berlin style. Following this line, we ought then to attempt to find the social structural motivations for this distinctive approach to matters of face, for having done that we would have linked

Conceptual Problems in the Study of Regional and Cultural Style

187

style to social structure in the motivated way that the ethnography of speaking view requires. Appendix I

Quarrel in a Film Processing Store

In this extract, M is a male customer who has come to pick up his developed prints, but wants to check their quality before paying for them. L is the proprietor who refuses to hand them over before receiving the money. 1 2 3 4 5 6 7 8 9 10 11

1; m; 1; m;

1;

... Brookland's Avenue right? 2.30 = = please Can I just see them? Uh well I'm not allowed to, to do that Well um if there are any in there = = which I don't think you ought to have printed, what'll I do = - then? Well you get a free film with your = = developing and printing uh so to

[

12 13 14 15 16 17 18 19 20 21 22 23 24 25

m;

26 27 28

1;

1; m; 1; m;

1; m;

No, what happens if you've = = printed any that I don't think was right in focus for example Well if they're out of focus they're = = not our fault are they Ah now come off that No! It's all right, it's all right, it is = = your fault for printing them Oh no, we give a free film for = = developing and printing No, that's your trading methods, I = = mean, no no

[

No the thing is if you want to = = have just the- only the good ones you don't get the free =

188

Stephen C. Levinson

29 30

= film you see you see, you have your choice [

31 32 33 34 35 36 37 38 39 40 41

m;

42 43 44

m;

Ah yes perhaps it does, but you're = = not allowing me to see the qual- the quality of the prints

45 46 47 48 49

1;

we don't do that, you never = = do, um Well, that's not business practice, is = = it, I mean I ought to be able at least to ( ) [

50 51 52 53 54

1;

55 56 57 58

1;

1; m; 1; m; 1;

m;

m;

m;

59 60

1;

61

m;

Ah yes but you see that that oh yes = = indeed! It's a choice I can believe that! Yes this is well, you know But if they're all coveredif uh, wait a minute that's 2.30 that's = = uh look's as if they should all have come = - out anyway

[

[

Well if you send them in the (post) you don't do that, do you I can't send them away, I still have, I = = still have the right to send them back, they ( )

[

You = = have here, you have here I have here well it's = all right then. Well that's alright = [ You have a ( ) yes, you have a, = if they're our fault = [ It's like saying to somebody,

Conceptual Problems in the Study of Regional and Cultural Style

62 63 64 65 66 67

1;

If it's our fault. But I mean, when you = = bring the film in here you bring it in for developing and = = printing. And our machine doesn't know if you've made a mistake = = in your ( )

68 69

m;

I'm sorry that is your method = = of dealing ( )

[

[ 70 71 72

1;

Well look, you, you = = take it somewhere else in the future, all right?

73 74

m;

I wi- well can = = I can give your governor them in that

75 76 77

1;

Yes, yes, yes, you — = can get - I am the governor 2.30 please [

78 79 80

m; 1;

No, no Cabinet is the governor Two- No Cabinet is N O T the governor = 2.30, 2.40, 2.50 = [

81

m;

82 83

1; m;

84 85

1; m;

Three and proprietor

86 87 88

1; m; 1;

four here Five. I am managing this shop (

89 90 91 92

m; 1; m;

No, no And I have every right to say that if I Yes, are you the proprietor and = = director of the firm

[

[

You're the governor eh?

[

This is my shop Are you the, are you the director

[

[

[

)

189

190

Stephen C. Levinson

93 94 95 96 97

1; m;

98 99 100 101 102 103 104 105 106 107 108

m;

1;

1; p; m; 1;

As far as you're concerned, yes No I'm not- no. You are the person who = = pays the wages and everything? I have the right to say that ( )

[

Are you = = the proprietor and owner of the shop? ((turns to next customer)) Can I serve = = you dear? Yes please, heh Alright I can see Alright, you can find out, I have every = = right, if you arrange that you either accept our conditions = = or you don't You can please yourself!

Chapter 8 NORMAND LABRIE

Comments on Berlin Urban Vernacular Studies The Berlin Urban Vernacular studies discussed during the workshop on "Methods of Sociolinguistic Description and Explanation" represent important progress on two counts. First, few studies have ever integrated as many valuable types of sociolinguistic analysis as these do. The integration of phonetic variable rules is completed by investigations on attitudes as well as investigations of integration into local social networks. This integration is also considered in relation to analyses of the stylistic characteristics of the Berlin variety, known as the "Berliner Schnauze". Second, this rare integration of sociolinguistic means of investigation applied to the complex Berlin situation allowed, on the one hand, to probably achieve the most objective description of the Berlin variety ever accomplished. On the other hand, it enabled the researchers to confirm the relevance of many sociolinguistic tools in a milieu as complex historically, socially or politically as Berlin. My intention is not to comment on the BUV studies in their totality but simply to concentrate on a point which is of great interest to me personally for it relates to my current research. This is the sociolinguistic significance of integration into local social networks. Social networks are taken into consideration first in D I T T M A R and S C H L O B I N S K I ' S (1985) Die Bedeutung von socalen Net^werken für die Erforschung von Ortssprachen and second in S C H L O B I N S K I ' S (1987) dissertation: Stadtsprache Berlin. Eine so^iolinguistische Untersuchung. The first paper represents a compendium of the major works done in the past on social networks in relationship to linguistic behavior. It can, therefore, be used as a valuable theoretical tool for sociolinguistic scholars interested in social networks. S C H L O B I N S K I ' S dissertation elaborates on a particular conceptualization of social networks and describes its application to the Berlin context. The results thus obtained, although not significant statistically for the most part, are of great interest because they shed light on

192

Normand Labrie

the difficulties presented by a sociolinguistic conceptualization of social networks. What is more, they underline the relevance of pursuing the study of social networks. The hypothesis explored by SCHLOBINSKI is based on MILROY'S ( 1 9 8 0 ) own hypothesis: the more a speaker is influenced by the standard norm, the less dense and multiplex his social network should be. The reason is that when a speaker loosens ties within his social network, he loses by the same token the possibility to counteract the standard (institutional and oppressive) norms. MILROY ( 1 9 8 0 ) demonstrated that the linguistic loyalty (i.e. the affinity the speaker has towards the vernacular norms) was positively correlated with the degree of integration into social networks. This hypothesis was demonstrated by MILROY ( 1 9 8 0 ) as part of a study carried out in three different working-class communities of Belfast, three communities which were homogeneous on the social level compared to the three Berlin communities. In order to carry out the BUV study, a 10 point scale was used for the answers to 3 questions about integration into the local milieu: 1. What kind of relationship do you maintain with the people living in your neighborhood? 2. Do you have any friends or intimate acquaintances who live near your home? 3. Do you have contacts with relatives other than those living in your home, who live close to you? A high score corresponds to strong integration into a local social network while a low score corresponds to an isolated position into the network. This social network scale corresponds, in other words, to a measure of integration into the local milieu. The hypothesis tested by SCHLOBINSKI is that the production of dialect variables, characteristic of the BUV, is positively correlated to the degree of integration into a local social network. The higher the score of integration into the local social network, the higher the percentage of production of the stigmatized variables. Except for the variable [au 2 ] —> [u] found in the preposition ( a u f ) and its constructed forms ( d a r a u f a u f m a c h e n , etc.) which presented a relatively weak correlation (cf. p. 147), no other correlation could be observed. This absence of significant results was interpreted in the following way: 1. the subsample (25 informants) was too small; or 2. the scale, as modified for the Berlin context, (cf. p. 148) led to different (non-significant) results than those obtained in Belfast; or

Comments on Berlin Urban Vernacular Studies

193

3. the integration into a local social network would not have the same meaning in Berlin as in Belfast in regard to the variation and the linguistic change. My own contention, as mentioned above, is that there was an important difference between the degree of homogeneity of the Berlin communities and those in Belfast. I would tend to attribute the absence of significant results to this difference. The three Belfast communities were relatively homogeneous: all three were working-class communities with a more or less important degree of mobility with respect to job location. The three Berlin communities are quite different on the social level: Prenzlauer Berg is a community of workers (which used to be a working-class community) whereas Wedding is a working-class community. These two communities differ in respect to the geographic mobility of their residents. The third community, Zehlendorf, is a typical upper middle class community and as such differs from the two others. In addition to the differences in the localities, it could have been meaningful to investigate various individual characteristics of the integration into local social networks. But I won't stress the individual aspects in these comments, since they are more fully described in my article in this volume. While measuring the degree of integration into a local social network, my own hypothesis is that one cannot really predict the use of the stigmatized variety norms in the Berlin communities (whereas this could have been done in the three Belfast communities because of the relative homogeneity). What can, however, be predicted is the degree of respect of the local norms. In accordance with P I E R R E B O U R D I E U ' S (1982, 1984) conceptualization, one can imagine that for each community there is a local lingustic market. A stronger integration into a local social network would lead to the adoption of the linguistic norms inherent to that local linguistic market. In order to test this hypothesis, one possibility would be to use the Student " t " test (for the comparison of the means) for example. This test would enable one to determine if the individuals who are strongly integrated into a local social network make use of the stigmatized variables corresponding to the mean use of the variables in each of the communities. Another type of analysis would be to suppose an inverted U-curve for each community (quadratic and negative curves) along a continuum corresponding to the percentage of use of the dialect variable. The second dimension would be the level of integration into a local social network. The summit of these 3 curves should be localized at the mean level of utilisation of the stigmatized variables (see Fig. 8.1). In

194 Level of integration into a local social n e t w o r k

Normand Labrie ( 30"

1

(70%)

(90%)

P r e n z l a u e r Iierg \ Wedding Zehlendorf

0 30

70

90

100

Figure 8.1: % of utilisation of the dialect variable

each community, individuals who show the strongest local integration would behave linguistically according to the local norms, individuals who are less integrated would make less or greater use of the dialect variable than the mean use depending on other factors, such as geographic or social origins. This hypothesis and the proposed interpretation do not, however, allow an explanation for the following phenomenon: there is a correlation, although weak, between the integration into a local social network and the production of the variable [au 2 ] —> [u]. This phenomenon would have to be observed more closely. The hypothesis suggested above coincides with the orientation of my own research on social networks. This research is described more fully in my article in this volume. In the article I propose that an absolute value should not be attributed to the notions of density and multiplexity — the two main characteristics of integration into social networks. I relate the meaning of density and multiplexity to different empirically observed situations.

Part 3 Investigations of Urban Varieties Research in Progress

Chapter 9 KLAUS J . MATTHEIER

Attitudes in Communication Profiles. Some Remarks on the 'Erp-Project' The Erp-project, 1 which I shall briefly outline for you, is based on two strings of tradition. On the first side it is embedded in the German dialectological tradition. Although this research tradition originally was centered in dialect geography, during the 50s and 60s the attempt was made more and more to integrate sociological categories. 2 The second tradition — and that is not surprising for the early 70s in Germany as all over Europe when this project began — is that of American sociolinguistics. Owing to the rural character of our research place, a village of 1700 habitants between Cologne and Bonn in the Rhine district, we oriented our work not to the main urban language studies of New York, Detroit and Buffalo 3 but more to the small but very interesting study on the language situation in Martha's Vineyard LABOV offered in 1963/64.4 The central aim of our research-project was the description and, if possible, explanation of the communication profiles 5 of the rural community Erp. The notion of a 'communication profile of a local speech community' as it is used here is connected with the notions of 'local speech community' and 'social group'. The local speech community 'Erp' is constituted by all the inhabitants of the village, and it is not necessary that these persons have direct communication contacts with each other. But it is important that the village has existed for a long time so that we have specific language varieties typical of this place which differentiate it from all the surrounding villages. The local speech community must have a regional-historic and therefore a linguistic identity. 1 2 3 4 5

Cf. Mattheier 1981 and Mattheier 1983 Hard 1966, Mattheier 1980 Cf. Dittmar, Schlieben-Lange 1982 Labov 1963, now in Labov 1972b, 1-42 For the discussion of this notion cf. Bell 1976, 181-186

198

Klaus J . Mattheier

In the community we have a number of social groups, which all together form the social structure of the community. The notion of 'social group' used here is different from most of the classical definitions of 'social group'. 6 This term comprises all persons combined in one social group, who live under comparable objective social conditions and potentially are in communicative contact with each other, in a local community with a regional identity, and who interpret these living conditions in a comparable manner on the basis of a shared complex of meaning structures and attitudes. As we suppose the members of this kind of a social group have been socialised in comparable processes and are exposed to comparable fields of experience. Therefore they develop and follow more or less the same strategies of social and linguistic acting. The language of such a social group is an ensemble of varieties. The difference between the language of a social group like this and that of the whole local community consists in the fact that in a social group all persons have internalized the same spectrum of varieties and use these varieties according to comparable rules of language use. On the other hand in the local speech community totally different speech varieties and rules of language use may be found. To this extent the spectrum of varieties used in a social group can be one of the basic elements for the description of the communication profile of a local speech community. The description of the communication profiles of local speech communities therefore has to describe both the stratification of social groups in local communities as well as the ensembles of speech varieties that are used in these groups. Together these complexes form the objective side of the data. But for the description of such a social group it is not enough to determine objective social categories as income, and the objectively given and linguistically describable speech varieties. The definition of this phenomenon in addition to the objective categories consists of subjective components, pointing out the way persons that belong to a social group interpret the objective facts of their social existence. This subjective aspect is very difficult to handle, in so far as it is not an unbroken reflex of objective conditions of life. Here the attitudes and evaluations, the meanings and the social norm structure, that had been adapted during the socialisation and the former social experiences take effect. A socially declassed, former independent farmer, now a commuting semi-skilled worker in a factory, develops his social and also his speech behavior not in accordance with the objective conditions of his new life. His social and language behavior furthermore 6

Cf. Hufschmidt/Mattheier 1981, 76f.

Attitudes in Communication Profiles

199

is superposed on the social evaluation systems of his former existence. He uses these systems and the behaviour derived from them as an immunisation strategy against the objective social descent that has taken place. To describe and explain social groups precisely this field of social evaluation therefore is of special importance. Unfortunately up to now there have been a lot of problems in finding adequate methods of research and description. But this is not the right place to discuss these problems. Let me just make a few remarks. GUMPERZ for example uses the notion of social network to isolate such structures in his Hemnetsberget-Study 7 . In my opinion one of the central factors in describing this structure of everyday experience and attitude is what can be called 'local loyality' 8 , that is an emotional relationship to the place of living. In regions without important ethnic differences it stands for the notion 'ethnic identity' FISHMAN uses in his research. 'Local loyality' expresses a central part of the linguistically effective subjective components of the notion of 'social group' we are using in our research project. For this phenomenon includes all social and linguistic behavior that is historically connected with this location. If a social group is characterized by a high degree of local loyality, then one can also expect a high degree of behavior typical of the village in local customs as well as in speech behavior. After these more general remarks to explain the implications of the terms "communication profile", "social group" and "local identity" that are central for the theoretical concept used in our project we can describe three fields of research that are essential for the description and explanation of a communications profile of a village like our 'Erp'. 1. The local community has to be described as a historical entity in relation to other nearby local communities and to the overlying social and administrative stratifications. 2. The local community has to be described as a system of social groups. a) Therefore the objective conditions of life of the different social groups within the community must be registered and described in comparison to each other. b) Then the central aspects of the attitude structure and the social evaluation system in everyday life have to be described comparatively. In this context one can expect that, for example, the local loyality mentioned above is one of the most important factors especially for the control of language use. 7 8

Gumperz 1971, for the concept of'social network' cf. also Dittmar, Schlobinski 1985 Cf. Mattheier 1985, further references there.

200

Klaus J . Mattheier

3. The local community must be described as a system of speech varieties and rules of language use. a) Therefore one has to describe first the linguistically distinguishable speech varieties in comparison with one another. b) Then one must register the structures of language attitude and the mental concepts in reference to availability and efficiency of the different varieties that are normally used in the everyday life of the community. These fields of description make it possible to analyse a communication profile of a local community. The form in which such a communication profile can be analysed and presented will certainly never have such a strict formality as the national sociolinguistic profiles of S T E W A R D , F E R G U S O N , F I S H M A N and K L O S S . But it is quite conceivable that after having solved some of the problems mentioned above one can come to a standardized description of communication profiles of local communities of different size. This standardized description would make possible the comparability of such communication profiles at a national and international level and the elaboration of a typological structure for local speech communities. But for now a standardized description is only a distant aim for sociolinguistic and dialectological research. The Erp-project, about which I will give a short outline now, can't be more than an explorative study to work out a number of methods for the analysis of communication profiles. That especially concerns the field of attitude and evaluation structure in which we are interested here. In a number of detailed studies the members of the research project have dealt with four central problems of the attitude and evaluation structure in the local speech-community Erp. Let me show you, by taking some selected examples, some of our working methods, and the subjective attitude and evaluation structures we have pointed out thus far. My colleague J O C H E N H U F S C H M I D T has presented an analysis of the observations and attitudes the speakers themselves have concerning their use of dialect in everyday life9. H E I N R I C H M I C K A R T Z has analyzed the complex of positions the inhabitants of the village have concerning the problem 'dialect and speech training in family and school' 10 . E V A K L E I N studied the conception of the speakers referring to the availability of speech varieties in different social situations 11 . And I myself have tried to describe the ability of the inhabitant 9 10

Hufschmidt 1983 Mickartz 1983.

» Klein 1983.

Attitudes in Communication Profiles

201

to distinguish the different varieties within the dialect-standard-continuum 12 . In his contribution about 'experiences, observations and evaluations of dialect use' JOCHEN HUFSCHMIDT tries to answer the question of how the relation between dialect, regional vernacular and standard is mirrored in the knowledge of the speakers themselves. In this connection he is interested in two facts - first the continual conflict of dialect speakers with the surrounding standard language - secondly the problem standard language speakers have with a dialect variety, which is, in their mind, strange and socially marked. HUFSCHMIDT distinguishes in this connection between observations and experiences of the speakers on the one hand and attitudes on the other. Whereas the experiences and observations are made by the speakers themselves during their lifetime and therefore form the foundation for a certain behaviour in relation to the speech varieties, the evaluations are almost based indirectly on such observations. But evaluations on the same extent are founded on general attitude structures used in the village, into which the village people are placed by birth and which influence the experience and observation possibilities indirectly in a high measure. The language material HUFSCHMIDT used in his analysis as well as the material for the other analyses come from part III of the research observation interview that consists of an informal but interviewer-guided interview on the language use and the language attitudes in the village 13 . In most of the more than 140 interviews with two informants at each time this part has been an informal and relatively open discussion. This method has the disadvantage that it does not lead to clearly distinguishable and comparable data as come from test methods for the evaluation of dialects by the semantic differential, the Lickert-scale or the matched guise technique. For analyzing communication profiles certainly one will have to work with those methods. In the present research situation at the beginning of such analyses it is primarily necessary to accumulate data for variety evaluation under more informal conditions. For his analysis of the discussions on the use of speech varieties in the village and their evaluation HUFSCHMIDT chose a content analysis and he tried to form types of statements on this question. We have been conscious 12 13

Mattheier 1983a. Part I consisted of a participant observation of an informal chat between two villagers who know each other; part II consisted of an interviewer-guided interview, concentrating on employment affairs. Cf. Mattheier 1981.

202

Klaus J. Mattheier

of the problems of this method. The analyst who uses an interpretative method to describe the behavior and the attitude of a group of informants that is fundamentally strange to himself, finds himself in a well-knitted unresolvable hermeneutic vicious circle for he uses his own interpretive frame, bound to his own sociocultural network, to analyze the statements of the informant, living in another everyday-world 14 . It is only possible to moderate these problems a little by having a lot of such statements that point in a similar direction. HUFSCHIDT distinguishes in his analysis four types of statements: 1. Dialect is the main instrument of communication. 2. There are only positive experiences combined with dialect. Dialect is evaluated positively. 3. The influence of the standard language grows, that of the dialect diminishes. 4. Dialect is evaluated negatively. In the first part of his analysis he discusses these statements and tries to show their pattern of argumentation and their differentations. In the second part he correlates the statement with a number of objective data of social and speech behaviour. This correlation should provide initial evidence of one central hypothesis of the Erp-project, which states that inhabitants of a village who live under comparable objective circumstances of life also share some typical experiences and attitudes structures and therefore form social groups as pointed out above. The first type of statement 'dialect is the main instrument of communication' is the one expressed most often by the whole group of informants. In addition to this, dialect as a means of normal communication in the village is important not only for the inhabitant born there but also for the group of inhabitants that comes from outside the village and lives here for only a short time. For those people dialect is not important as a communicative barrier but as a problem of social integration into the village which is only possible by speaking dialect. On the other hand in this group of speakers all members are aware of the situational restrictions of dialect and of the difficulties dialect speakers have with their language outside their local everyday world, especially in school. Whereas in the first type of statements it is mainly observations and experiences which point to dialect as the main communication medium, the second type combines all those statements that can be interpreted as a positive evaluation of dialect. Informants, who are of this opinion, 14

Cf. Mattheier 1982, 633f.

Attitudes in Communication Profiles

203

emphazise especially the high grade of confidence which is created in the village by dialect. Another fact of positive evaluation is the deep connection with local folklore tradition that is constituted by dialect. But there is no villager who speaks of "love of the dialect" as the official "watchers of dialect correctness" so often do. From these two statements the types III and IV are fundamentally distinct. In these two types the experience of a continuing increase of standard language is emphasized and positively evaluated. These two theses of a decay of dialect are seldom expressed explicitly by our informants. But they notice two types of decay of dialect: first a quantitative decline of dialects as a result of the increase of the members of new inhabitants coming from surrounding towns and on the other side a deformation of the old dialect by an increasing number of interferences from standard language. The language mixture resulting from this process is called 'Hochdeutsche mit Knubbel' that is 'standard with stripes' and is negatively evaluated throughout the village. I have tried in another study to describe the variable language structure of this phenomenon using a type of variable rules that is based on the Labovian concept of variable rules 15 . The great number of statements of type IV makes it evident that negative experience with one's own dialect and also negative evaluation of dialect is widespread. The informants fear they will ridicule themselves by using their dialect outside local and private situations. That is mainly due to negative experiences they have had in school with dialect as their first language. The negative evaluation of dialect in school is in our opinion the basis of negative evaluation of dialect in general. Efforts to revitalize dialect would have to begin in the primary school and not with an increased usage of dialectal speaking in radio and tv, as can be observed in German today, a development that is called 'dialect wave' or 'dialect renaissance' 16 . The above mentioned statements on dialect come from 79 informants, that is 55.6 percent of our informants. The distribution of the statement types over the whole population of informant makes it possible to isolate certain groups of speakers by their social position in the community. It can be pointed out that the typical dialect speaker is in a somewhat schizophrenic position. On the one side their experiences of dialect are very negative and this generally should lead to a negative evaluation, but on the other side they evaluate their dialect positively. It is especially the 15 16

Mattheier 1979. Cf. Mattheier 1980, 171-174.

204

Klaus J. Matthcier

social group of farmers, local craftsmen and small tradesmen which form the so-called 'old middle class' but also parts of the blue collar workers and the farm labourers who evaluate their dialect positively although their experiences normally would give occasion to a more negative attitude. These informal results indicate how important the description of the attitude towards different speech varieties is and that it is to a certain degree independent of people's observations and experience of their language. It may be supposed that the further development of the dialects in this community will be more influenced by the positive attitude to dialect than by the negative experience people have of it. I for myself have tried to analyse another central problem of the everyday knowledge people in our village have of their own spectrum of speech varieties. This knowledge consists not only of a system of attitudes and evaluations concerning the dialect and the other speech varieties. It also consists of the ability of the speakers to differentiate the language continuum they are living in. In the analyses of the subjective concepts of situational language use that I have not discussed here, there have been hints that the connotations of 'dialect', 'regional vernacular' and 'standard' in this small speech community differ in accordance with the social environment people live in. To analyze the ability of the informants to differentiate the speech continuum I used what we call a 'tape-test' 17 . In this test all informants were confronted with a tape recording that consisted of six speech-items of 15 seconds each. These items differed in their depth of dialectality. There were two recordings of deep local dialect, one of which we called 'half dialect', one of 'regional vernacular', one of spoken standard with a non-regional accent and one of high standard. The informants were confronted two times with these tape recordings and then were asked to note on a scale that symbolizes the dialect-standard-continuum the degree in which they thought the recordings differ from standard and from each other. Then I used the frequency of the incorrect recognitions of two speech items as a measure for defective consciousness of the differences between the varieties represented. With this analysis I tried to answer two questions: 1. Which social groups in our population succeed best in identification and differentiation of the speech items heard during the test and which group has the greatest difficulties in doing so?

17

Cf. Mattheier 1983.

205

Attitudes in Communication Profiles

total group

new middle-class

old middle-class

high standard non-regional marked vernacular regional marked vernacular half-dialect

full dialect

Figure 9.1: Distribution of language-variety-differentiation in different social groups

2. How do different social groups differentiate the continuum between dialect and standard? The analysis of the first question does not show — as one could expect — the group of regional and social mobile commuters with a lot of experience with all possible varieties and regional Ribuarian 18 vernacular as their main variety as the group having the best ability to evaluate the varieties of the dialect-standard-continuum. The informants who did best in the test were those who because of their emotional connection to the village and their social position in the old middle class use dialect as well as the standard each according to the appropriate social situation and therefore have a balanced and stabile language-attitude-structure. For the second question it can be shown that in this small village of 1700 inhabitants there are obviously two different models to graduate the dialectstandard-continuum in everyday-knowledge. On the one hand there is the group of Younger Rhinelanders out of the new middle class who commute daily into the surrounding towns. These informants divide the speech continuum into four varieties. They separate deep dialect and so-called 'half dialect' on the one side and standard on the other, so that a cluster of different vernaculars remains in which they don't differentiate between vernacular with and without regional marks.

18

The dialect, spoken in the village Erp belongs to the Ribuarian dialect, which together with the Moselle-Franconian dialect of the south form the Middle-Franconian dialect.

206

Klaus J . Mattheier

On the other hand there is a group of elderly, village-born dialectspeakers belonging to the old middle class who divide the continuum only into three strata, the deep dialect, the high standard, combined with nonregional effected vernacular and between both a regional vernacular that consists of the 'half-dialect' and the regionally marked standard. We have no time to discuss these results in detail. So let me point out only the possible consequences, the existence of the two different models of structuring the dialect-standard-continuum will probably have for language change in coming years. One can expect that interference in the dialectstructure will first affect that group which in its perception of the varieties does not make a difference between non-regional-marked, standard-bound vernacular and regional vernacular. At the same time this group — the commuting new middle class — is much more exposed to standardlanguage everyday life, because of the social environment they live in. So the subjective perception-structure of the dialect-standard-continuum and the objective social structure of life — conditions couple with each other and form a social group with comparable social and — what is important for us — also comparable speech-behaviour. The research of the Erp-project as it has been published so far, of which I gave you two small examples here, shows the importance of evaluations and attitudes for the analysis of the communication profile of a community. That does not mean that there is no necessity for the analysis of objective speech behavior and of the variable structure of the language our villagers use in everyday life. But especially if you wish to reach the explanatory level of studies of local communication profiles you are forced to study both sides, the objective and the subjective structure of social life.

Chapter 10 A L B E R T O SOBRERO

Villages and Towns in Salento: the Way Code Switching Switches The study of the Berlin vernacular (see this volume) draws the sociolinguist's attention to general problems of urban dialectology, which I shall apply to Italian society in this paper. Regarding the relationship between urban and rural dialectal varieties (§1), I shall report the preliminary results of sociolinguistic research carried out in the Salento (southeastern Italy) as to the various functions of code switching in two towns in Puglia, Brindisi and Lecce, and in some neighboring villages (§2).

1. What are the sociolinguistic characteristics of urban varieties compared to rural ones? In other words, is there an urban specificum with identifiable characteristics? Despite inadequate literature about the Italian language situation, we can take the following considerations into account. An urban linguistic space does not, by itself, have any linguistic characteristics which clearly distinguish it from a non-urban linguistic space. In addition, the greater tendency toward standardization which is usually considered a distinctive feature of towns, cannot be assumed to be a general characteristic because 1) it is variously distributed according to extra-linguistic variables, such as residential quarters, social class, age, education, immigration waves, etc. (SOBRERO, 1 9 7 8 ; G R A S S I , 1 9 8 2 ) ; 2 ) the greater tendency toward standardization is subject to phenomena of hyperconservative reaction, to which phenomena of hyper-standardization are opposed (SOBRERO-ROMANELLO, 1 9 8 1 ) in a non-urban area, so that the town/rural area relationship may appear to be reversed. On the other hand, the sociolinguistic approach seems to be useful one. From the available

208

Alberto Sobrero

literature we can trace at least five directions of research regarding the different sociolinguistic structures of urban and rural speech communities. 1) The balancing point between tradition and innovation, existence and strength of a "linguistic loyalty". The united feeling which binds the speakers of a speech community, the "esprit de clocher" of a rural variety, is based on cultural homogeneity which is, by nature, weaker or virtually absent in town. If it is true, as it appears to be, that "Uunità di un punto linguistico minimo non è semplicemente uniformità d'uso, ma deve piuttosto considerarsi come il risultato di un perpetuo equilibrio che interviene fra gli atteggiamenti particolari di chi sta alla testa o alla retroguardia del movimento linguistico di quel punto e l'anione di una massa livellatrice'''' (TERRACINI, 1981: 333), the effect of the levelling mass is much weaker in town, and so the disgregation of the unitary tendency is quicker. The relation between avant-garde and reargarde becomes crucial compared to the persistence of a (weak and heterogeneous) levelling action. The preferred research field is the situational textual and inter-textual variation: the rules of choices and code switching, the continuum within which the change takes place (and the direction it takes), the innovative and conservative factor of the different kinds of speech in the various social classes, quarters, situations, etc. 2) Social networks and the complex articulation and social stratification. The social structure of an industrialized town can be better analyzed with a Marxist approach, and it is possible to apply such concepts as linguistic market and habitus to an urban sociolinguistic situation (BOURDIEU, 1984). But the extra-urban linguistic point — or area — is based on social contracts of quite a different historical origin (see 3), and this can produce remarkable effects on the system of interpersonal relations (see footnote 1)3) Historical depth. This point is closely related to the previous one. In order to understand the sociolinguistic structure of a town, it is necessary to give it historical depth, but not only in the sense of reconstructing the course of events. Especially in Italy, the time has come to make use of the criteria of synthétisante historiography for linguistic analysis. We expect important results, for instance, from the application of the concept of longue durée to the domain mentality (SCHLIEBEN-LANGE, 1 9 8 4 ) and to the habitualized semantic network (SOBRERO, 1979: 85-92). The assumption to be proved, in this case, is the effect of historical patterns of social time on the social and linguistic facts, which we understand through synchronic analysis. In other words, it is a matter of understanding the mechanisms connecting the speaker's choices on an unconscious level (for instance,

Villages and Towns in Salento: the Way Code Switching Switches

209

conversational strategies) with the patterns of civil, economical and linguistic history of his community (for a detailed exposition and some examples of analysis, see SOBRERO 1 9 7 9 ) 1 . 4) Prestige. Being intensely exposed to innovating tendencies, the urban linguistic community is much more sensitive than the rural one to the connotation of prestige or stigmati^ation which any of the linguistic varieties of the repertoire can little by little assume or change. In light of this consideration, we shall have to reanalyze the concept of linguistic loyalty ('why not disloyalty?) in the urban context, and study deeply, in town and in the country, attitudes (cultural stereotypes, comparative evaluations, etc.) on the one hand and behaviors on the other: first of all, the linguistic change in formal situations. 5) Double convergence. As a rule, a double convergence (vertical and horizontal) occurs among the varieties in the repertoire (AUER-MOESSLE, in press). Regarding Italian, the former concerns dialectal varieties and (Regional) Italian, the latter concerns hegemonic dialectal varieties (almost always urban) and subordinate dialectal varieties (almost always rural). Of course, the rules of such a convergence in town differ from those in the village, and must be carefully examined both on the conversational level (code switching, code shifting, code fluctuation, language mixing, etc.) and on the structural level (forms of coine in dialectal syntax or lexicon, Italianization of dialectal forms or of forms of coine, etc.) All five directions of investigation are influenced by phenomena which, as we know, play a particularly important role in the processes of linguistic transformation of bilingual societies: code choice and code switching 2 . On the basis of these considerations, in the first months of 1985, during a seminar on Italian dialectology at the University of Lecce, I collected sociolinguistic material on code switching in rural and urban communities. 1

In conclusion, we have to deal with the problem of the relationship between micro- and macro-analysis, -clearly stated by BREITBORDE in 1983- from an historical point of view. Point 3 is closely related to point 2. For instance, on the basis of the Italian historical peculiarity, with the combined use of such tools as social network and historical pattern, we might prove a stimulating hypothesis for our urban sociolinguistics: if the density and multiplexity (MILROY-MARGRAIN, 1980) of social networks are higher in artisan quarters than in residential and working quarters, the cause is to be found in the different historical typology of the social aggregation, which in societies with a prevalent handicrafts economy is more deeply rooted in the medieval corporation than in the capitalistic manufacture (PROCACCI,

2

1955).

For a history of the approaches to code switching, a discussion on the main problems and the bibliography on this topic, see GARDNER-CHLOROS, 1983 and Volume 39 (1983) of the International Journal of the Sociology of Language.

210

Alberto Sobrero

I am presenting here my initial considerations on that material. I hope to point out both the richness of possible results and the importance of the topic to the dialectologists and sociolinguists working in areas which can be compared to the Italian ones.

2.

We worked in the towns of Lecce and Brindisi and in seventeen villages of the two provinces. In each place we gathered data on the linguistic behavior of one or more local informants, matching them for social class, type of employment, and education, by using the following elicitation and analytic techniques. a) Initially we carried out a short participant observation: each researcher observed the linguistic behavior of a relative or a close friend for an uninterrupted period of at least five days. Each researcher took field notes on the distinctive characteristics of the various communicative situations and on the interactant's individual linguistic characteristics paying attention to the in-network interaction and the informant's linguistic options in each situation. Whenever possible, the researcher surreptitiously recorded spontaneous speech in natural settings. b) Following the participant observation, three types of elicitation were carried out: 1) a translation of a classic Italian dialect questionnaire; 2) a guided interview with a visible tape recorder; 3) a discussion or spontaneous informal talk with the same interviewer using a hidden tape recorder. We focused on the code selection among the varieties of the repertoire (Regional Italian, dialectal Italian, local dialect) and the changes of choice occurring in the verbal interaction: code adoption, code switching, code shifting, code fluctuation (inside the Regional Italian) 3 . In this case, however, we shall use the expression code switching in the most general sense of changementjalternance de langue ou de varieté linguistique dans un discours ou une conversation (GARDNER-CHLOROS, 1 9 8 3 : 2 1 ) . The results are complex, but regarding the dialect/language code switching they are quite clear. Behaviors can be grouped into well-defined types, the first one occurring in rural communities, the second one in urban communities. In sections 2.1 and 2.2 I present cases typical of both types of behavior. 3

Code switching, code shifting and code fluctuation are based on the alternating use of more than one realization for linguistic categories or features on the phonetic-phonological, morphological, syntactical, lexical or textual level (AUER-DI LUZIO, 1983: 1).

Villages and Towns in Salento: the Way Code Switching Switches

211

2.1

In Acquarica del Capo 4 , informant D., a female dressmaker with a primary school education, who is well integrated into the village social network 5 , shifts between two extreme varieties: a dialect with some code switching towards Italian, and Italian with some code switching towards dialect. The basis is usually dialectal and the choice of dialect, in the five days of observation, is categorical and identical with those who are immersed in the same network: eighteen persons, including friends and acquaintances, all with a network score of at least 70/100. The topics of the conversations can be compared with one another, because they concern events of everyday life. When D. interacts — always about daily topics — with educated relatives (a high school graduate and a university student), or with educated friends of her husband's (a teacher and a geometer), who are at the periphery of the network, her speech includes a good deal of code switching towards Italian. When she talks to two friends of her daughter's — young university students — Italian is prevalent with comparatively frequent code switching towards dialect. She speaks a more elegant Italian with reduced code switching during the guided interview (topics discussed were young people's neglect of dialect; dialect as school subject matter; dialect as social marker; religious and civil traditional customs; the feast of the patron saint. Type of discourse: narrative-argumentative). Thus, the network tie, the degree of formality of the situation and the topic not only determine the choice of the code-of-basis, but they control the mechanism of the code switching as well. The setting seems to be particularly strong. For example, when an analyst, who is a graduate, but one well integrated in the network (score: 91/100), went to D.'s house to take her blood to be analyzed, she spoke only dialect to him. Therefore, in settings connoted by informality and by uniformity of network, the very transactional situations also become personal, and for this reason they orient towards the choice of dialect. 4

65 kms. from Lecce, on the farthest southern limit of the Salento: on March 31, 1985 it had 4,697 inhabitants, with problems of unemployment and flowing emigration (in 1985 the emigrants numbered 596, equal to 12.6% of the resident population). Agriculture occupies 60.5% of the active population; commerce and small factories occupy the remaining active population. Main meeting places: parishes, party rooms. Surrounding towns and villages where young people go to study or work: Presicce, Casarano, Tricase, Alessano. Illiteracy is widespread among elderly people.

5

For each informant we calculated a network score (index of cultural integration) in hundredths taking into account the characteristics of each place. Here, the network is close, dense and multiplex, and its dimensions practically coincide with the speech community.

212

Alberto Sobrero

These findings are confirmed by surveys in Sava, Cellino S. Marco, Mesagne, Guagnano, S. Pietro, Vernotico, Novoli, Trepuzzi, S. Pietro in Lama, Casarano, Allìste, Sannicola, and can be summarized in the following diagram: network familiarity education young formal setting

V — — 44+

network familiarity education young formal setting

D
I

Diagram 10.1: The Relationship between Code-switching and Social Network D = Dialect with some code switching towards Italian I = Italian with some code switching towards Dialect

We notice that a diglossic evaluation of the language/dialect relationship orients this informant's behavior, which is, moreover, characterized by variable use of code switching, which assumes two different functions: — an unconscious, conversational function of convergence in spontaneous speech (GILES-SMITH, 1979) 6 ; 6

At first glance,"the case would seem similar to the Puerto Rican case in New York 1980). In fact, in spite of a constant alternation between Italian and dialect — also on an infratextual level — and of the cultural and social predominance of Italian, the mixed code produces a third available variety. This, however, does not allow us to (POPLACK,

Villages and Towns in Salento: the Way Code Switching Switches

213

— in formal speech a strategical function (DITTMAR, 1987b) aware of adaption to strong factors in the situation: topic (in the guided interview) and purpose (let strangers know her competence in Italian). It is only in this sense that we can still speak (with GUMPERZ, 1982) of a contraposition between we code and they code. If we analyze the semantic content of the code switching we find very frequent changes because of a lexical gap or a quotation, as, for instance: D: dice ka 1-annu truvatu nu poku de bronkopolmonite (lexical gap) D: nu pikka ingeneria nu pikka dice medicina (lexical gap) A: e labbra viola tieni D: ci gge labbra viola? [...] A le labbra viola si m-aggu (quotation) D: Sono andata 1-altro gorno e diku me kummare (quotation) Which type of Italian does the informant use? She uses words learned not at school or from newspapers, but through television, and mainly telenovelas, telefilms, quiz-programs: psychiatry and psychoanalysis, as she herself declared, are words she heard in a telefilm, and enthusiastically adopted, after asking a niece to be certain of the meaning. Thus, the innovation — produced by the most popular television programs — enters D.'s linguistic repertoire through the code switching mechanism applied to in-network interactions. The code switching becomes one of the rules of the speaker's linguistic use; it is not a valve which allows alternate entry to two different paths, but is an internal mechanism of multiple competence which contains both forms of dialect and forms of Italian (POPLACK, 1980; SANKOFF-POPLACK, 1981). 2.2

In Brindisi 7 , informant G., a 70 year-old housewife with a primary school education who has an in-network integration score of 75/100, exhibits an italophone and italophile attitude, which appears not only in the narration classify ours as code switching without convergence. The convergence, in fact, is systematically registered on different levels: a) in the conversational interaction (SOBRERO, in press); b) on a phonetic (SOBRERO-ROMANELLO, 1980), a lexical (ROMANELLO-SOBREROZAMPOLLI, in press), and even a morpho-syntactical level, in Regional Italian. 7

Provincial capital, with 91,857 inhabitants on March 31, 1985. Important commercial (Turkey, Israel, Egypt, Lebanon, Southern Africa, the Far East) and touristic port (ferry-boats to Corfu, Paxoi, Ithaca, Cefalonia, Patras, Igoumenitza: harbor passengers movements, about 80,000 a year). There is a national airport with a stop for international network. Its commercial harbor is the most important in Southern Italy after Naples. 3,759 people work in commerce, 7,085 in industry, 6,996 in other activities. The town is endowed with sports facilities, museums, art galleries, libraries, big stores, 5 cinemas, 3

214

Alberto Sobrero

of her autobiography - obstinately rich in readings and efforts to learn and speak Italian in spite of her limited education - but also in the repeated declaration of her incompetence to translate the dialectal questionnaire because of her poor knowledge of dialect. Her actual performance contains both the standard and dialect, in different doses and with different rules of use compared to the previous case. In the five days during which she was observed, G. met eight relatives (brother, sister, children, son-in-law, daughter-in-law, niece: network score between the son-in-law's score of 5 0 / 1 0 0 and the daughter-in-law's score of 9 0 / 1 0 0 , the average score of the relatives being 7 4 / 1 0 0 ) , two friends in the neighborhood ( 8 6 and 9 0 / 1 0 0 ) , her shoemaker and her hairdresser ( 8 3 and 9 0 / 1 0 0 ) . With her relatives, her shoemaker and her hairdresser, she exhibits a basis of Italian with dialectal insertions. We notice, however,that such insertions are not as frequent while speaking with her daughter-inlaw and son-in-law; for instance, in an interaction with her son-in-law, in 15 cues of G.'s, we find only one item we can phonetically classify as a dialectal one: vintikuattru (twenty-four). Instead, in seven cues, some deictics and pragmatic locutions (nu, mo!) appear with a tag switching function (POPLACK, 1 9 8 0 ) . On the other hand, in surreptitiously-taped conversations between the informant and her friends, code switching toward dialect is very frequent. The tape recordings are interesting: during the first one, G. receives a telephone call from her daughter, who was asked by the field worker to call her mother. G. quickly realizes that the call is rather strange, and becomes suspicious. At this moment, G. performs a mixed discourse, with considerable prevalence of Italian. In the second case, when G. discusses in a relaxed manner with the same daughter and her niece L. (who is the field worker) her plan to knit a sweater, the alternation is generalized. G. performs a mixed discourse, in which the shift Italian/dialect is continuous, bi-directional and quite natural (see Appendix). We also notice the typological and functional difference of the code switching between the in-network informal speech and the formal speech with peripheral8 speakers of the network. In the guided interview (formal speech) and in the conversations with her son-in-law we have a conversational code switching towards dialect only for pragmatic locutions or deictics (na [here], mo [now], none [no], nu [a, article]), and for quotations

8

theaters, quite a number of cultural, sporting and recreational associations, 6 consulates, 2 hospitals, 9 junior high schools, 18 secondary schools. For a distinction among in-network, peripheral and extra-network speakers, see SOBRERO, in press.

Villages and Towns in Salento: the Way Code Switching Switches

215

(GUMPERZ, 1 9 8 2 ) up to the limit of tag switching (whole sentences interpolated by dialectal n a f . On the other hand, in the mixed discourse performed in-network, the shift involves more complex structures of the dialect (see Appendix). Furthermore, the semantic equivalence forms of the two codes can be perfectly replaced, even with a syntagmatic inversion of code: tutto un pesgp (20, 19) I tuttu nu pevgo (20, 13); n-altro (5, 20) / l-atru (6, 8). Today the orientation that is more frequent among the elderly in the urban area of Brindisi and Lecce can be represented by Figure 1, but the extremities of the abscissa are respectively D/I (mixed discourse) and Regional Italian. The diglossic evaluation of the community, if there is one, is much less obvious than in the language used by the informant of Acquarica, and in the code switching the function of conversational convergence10 is prevalent. Since, compared to the rural pattern, it acts on a much lower — or even absent — level of awareness, the increase of code switching from the Italian basis towards dialect with in-network interlocutors or, more often, interlocutors characterized by features [—young, (—son), —education, -(-closeness], gives origin to a mixed discourse full of replaceable lexical items. But deictics — especially anaphorical — and pragmatic dialectal locutions enrich the most formal speech as well: they make up the base of the structure of the code switching and spread from the origin of the abscissa (mixed discourse) almost to the limit of its highest value — that is, close to Italian —. In this way those particles become "stateless" in the speaker's awareness and, on the level of the community's repertoire, penetrate not only into colloquial Italian but also into the most formal Regional Italian 11 . We can say that, in an urban community characterized by frequency, intensity, and variety of contacts, one of the dominant characteristics of

9

10

11

In contrast to S A N K O F F and P O P L A C K (1981), this is not a matter of incompetence in one of the two varieties, but of transcode marks. The general results here set out through the emblematic presentation of the two main investigations in Acquarica and Brindisi are confirmed by the conclusions to which we have come after examining the investigations of Sava, Cellino S. Marco, Mesagne, S. Pietro Vernotico, Novoli, Sannicola, Trepuzzi (2 surveys), Lecce (4), Brindisi (4), Casarano, Alliste. I have registered very frequent uses of the article nu, na (in place of »»/»»»/»«a), (stu, sta (in place of questo (mas.), questa (fem.) = this), mo (now) even in the most formal situations: university lecture, discussion between a publisher and a writer, meeting of the Board of Governors of the University, etc.

216

Alberto Sobrero

the linguistic change is this: code switching with the characteristics of an extended tag switching leads dialectal micro-structures (deictics, pragmatic locutions, interjections, phonological interferences) from conversational convergence to vertical convergence ( A U E R - M O E S S L E in press) between dialect and language. This occurs within a generalized process of shifting of the extremes in the rural continuum Dialect-Regional Italian (the continuum in an urban setting tends to become: mixed discourse-Regional Italian) 12 . The fact that code switching serves a different function in town than it does in the village seems to be evident from our investigations. But, more on a general level, is it a matter of characteristics of the Salento area, or of constants within the town-village relationship?

Appendix L. G. L. G. G. L. G. L. G. L. G. L. G. L. G. 12

mbé — sta lana é kaldissima e ka grad^je é lana pura lui — e no ti kosta na lira veramente solo la patsjentsa tua la patjentsa mia veramente ((pause)) gwarda e — e kwesto per kwedda fessa di tsia lutjia — lutjia vedjamo ((laugh)) ( ) t f e si le f f e vedea tutte ste kose tutte sfruttate ( ) wéj nna bortsa de — de kwidde gwarda k-adjdju dato sotto e — kwesta no — a:: lana bjanka ( ) la lana bjanka é kwella ke skutjimmo none kwedda sta ankora kwesta — saj t f e d éì kwiddu maAAone bjanko kwale bjanko? kwiddu [grande ke kompraj io | il m a g o n e o la d3akka? il m a g o n e . We can add that this occurs also within a change in the hierarchic order of the extralinguistic variables acting on the speaker's choices. In fact, for our elderly informant the social network — in the quarter dimension — is still the determining factor for the structuring of her linguistic competence; but the other urban interviews that we recorded in Brindisi and Lecce, have produced different resultes: they pointed out that social network in the towns has lesser weight for younger generations — under 30 years — with a higher degree of education, and that the father/son relationship has greater weight.

Chapter 11 NORMAND L A B R I E

Social Networks and Code-Switching: a Sociolinguistic Investigation of Italians in Montreal* Many recent sociolinguistic studies have tried to integrate investigations of social networks as a means for describing individual social environments. If the social network concept is useful for the study of behavior in urban settings ( M I L R O Y , 1 9 8 0 ) , this concept should be even more useful for the study of bi- or multilingual settings. The main purpose of the study described in this paper is to establish the existence of relationships between the social network of an individual and the aspect of his linguistic behavior known as code-switching, typical of situations where languages are in contact (BREITBORDE, 1 9 8 3 ) . This project involves an empirical knowledge of linguistic facts linked to the conjunctural aspects of specific verbal interactions as well as to the structural aspects of the environment of the individual ( B A K H T I N E , 1 9 7 7 ) . Indeed, one of the goals of this study is to demonstrate how particular junction points between the social universe and individual linguistic behavior could be considered as sociolinguistic objects. Linking social networks and code-switching is not a new concept (GUMPERZ, 1 9 6 4 ; G A L , 1 9 7 8 ; P O P L A C K , 1 9 8 0 ) . This idea became all the more relevant when I tried to understand how the linguistic behavior of individuals in a bilingual context was dependent on the social environment, that is on the organization of linguistic communities ( L A B R I E , 1 9 8 4 ; L A B R I E & CLEMENT, in preparation). While undertaking a sociolinguistic study, it seems obvious that a flexible conceptualization of the linguistic community is needed (ROMAINE, 1 9 8 2 ) as well as a real sociolinguistic definition of the social network, rather than an ethnographic one. A * This research is conducted with the assistance of a doctoral fellowship f r o m the Social Sciences and Humanities Research Council of Canada. The author wishes to thank Denise Deshaies, Louise Rousseau and Shelley Faintuch for their comments on this article.

218

Normand Labrie

sociolinguistic concept of the social network has to be consistent with concepts of linguistic community, linguistic market, domains and situation (ROMAINE, 1982; International Journal of Sociology of Language, 39, 1 9 8 3 ) . This paper will make reference to a study in progress; definitive results are therefore not available at present. Attention will be focused on the theoretical framework establishing the relationship between social networks and code-switching.

1. Situation of the Italians in Montreal The 2 0 0 , 0 0 0 Italians living in Montreal represent the third largest group in Quebec after the Francophones and the Anglo-Saxons. An interesting aspect of this situation is that Montreal is already a bilingual city, the most bilingual city in Canada (Statistics Canada, 1 9 8 5 ) 1 . The same study shows that the allophones in Quebec (those Québécois whose mother tongue is neither French nor English) form a category of citizens among the most bilingual in the country. Almost half of them claim to speak both French and English (compared to 15% of the total population of Canada) in addition to their mother tongue in most cases. Italians also have the reputation of being more bilingual than any other ethnic group. Besides the statistical importance of this ethnic group, it is interesting to note that (for almost 20 years) the whole question of language status in Quebec has been influenced by pressure from the Italian community. Italians who came to Montreal after the Second World War settled in the French parts of the city and assumed the same types of occupations as the Francophones, whose social status was comparably low. This situation gave rise to special and social competition between those two linguistic groups ( S M I T H , 1 9 7 4 ) . As a matter of fact, the generation that immigrated bears many resemblances to the Francophones. The main difference between the groups, which arose in part from the competition between them, is that the generation that immigrated decided to give its children an English education in order to help them become more competitive on the labor market (BOISSEVAIN, 1 9 7 0 ; TADDEO & TARAS, in press). Over the last 20 years this decision has proven to be totally antithetical to the nationalistic aspiration of the Francophones. According to PAINCHAUD & POULIN ( 1 9 8 3 ) , 9 0 % of Italian children attend English schools 1

To be bilingual as defined by Statistics Canada means having a knowledge of French and English.

Social Networks and Code-Switching

219

in 1978 (compared to 70% today after the introduction of Bill 101). Three languages are used by these children: Italian with their parents (83%); English at school with their siblings (72.3%) and friends (79.8%); and French for playing in the neighborhood. As for the older generation, the languages used are: a regional or local dialect, or standard Italian (with family, neighbors, friends and merchants); French and to a lesser extent, English at work and with public services. As a result — compared to Toronto where only 26.3% of the children speak Italian with their parents (PAINCHAUD & POULIN, unpublished manuscript) — Montreal is characterized by a high level of maintenance of Italian. Because of the possibility to integrate into two different groups and because that possibility has different implications for each generation, Italian is used as a lingua franca by the different generations. It is highly relevant, therefore, to study the linguistic behavior of Italians in Montreal in relationship to their social integration. Differences between the generations shall be taken into account as well as differences in behavior in different domains such as the family and the work place. In reference to BOURDIEU'S ( 1 9 8 2 ; 1 9 8 4 ) conceptualization, it will be possible to consider code-switching as a constitutive and a cohesive element of a complex linguistic capital linked to a multilingual linguistic market. It will then be possible to verify differences and similarities in the linguistic production for individuals of different generations, on different markets.

2.Theoretical framework The concept of code-switching refers, in this paper, to the most general definition given by POPLACK (1980:583): "Code-switching is the alternation of two languages within a single discourse, sentence or constituent". It also takes into account the alternation which can occur between units larger than these, that is the use of different languages within different discourses or the use of appropriate languages in different situations or on different linguistic markets. This conceptualization of code-switching makes use of a linguistic analysis on which a conversational analysis is superposed. While carrying out a conversational analysis in determining code-switching types in addition to a linguistic analysis, it is possible, as a result, to take into account the identity of the interlocutors, their individual characteristics and the social conditions prevailing for a particular type of interaction.

220

Normand Labrie

P O P L A C K ' S three code-switching types (1980) were identified by means of a linguistic analysis based on conversations held in peer groups. P O P L A C K considered that the conversational situation was constant. The three types are (see Table 1): 1. Emblematic code-switching 2. Intra-sentential code-switching 3. Extra-sentential code-switching

Categories

Sub-categories

Relevant aspects

Extra-discursive

Code-adoption

A tensed market forces the use of one particular language

Code-choice

A change in the situation, a new interlocutor, for example, possibly implies the choice of a different language

Intrasentential

Without any change in the situation, a change in language is produced:

Intra-discursive

— inside a sentence Extrasentential

— between two sentences

Emblematic

— between elements of a same sentence, if these are not syntactically linked. These elements can be:

— syntagmatic

— pragmatic locutions

— intra-syntagmatic

— loans — lexical interferences — morphological interferences — phonological interferences — individual loans (recourse)

Table 11.1: Categories of code-switching

In analysing spontaneous speech, where changes inevitably occur in the conversational situation, it is sometimes difficult to distinguish extrasentential code-switching from an alternation of language corresponding to a change in the conversational situation. In order to allow for such a distinction, I shall call the three linguistic types of code-switching described by P O P L A C K (1980) intra-discursive, since they are all realized within constant discursive conditions. When changes occur in the discursive

Social Networks and Code-Switching

221

conditions (a change of interlocutor; a change in the social determinants between the locutor (speaker) and his interlocutors; or a change in the tension of the linguistic market) the resulting code-switching can be considered to be of the extra-discursive type (situational code-switching in GUMPERZ'S terms [ 1 9 7 6 ] ; diglossic situation in FERGUSON'S [ 1 9 5 9 ] although this way of categorizing seems less flexible). Extra-discursive code-switching consists of two different types of alternation: Code-switching which corresponds to a change in the situation caused, for example, by the presence of a new interlocutor. AUER (1981) showed that one very important point in the conversation (MCSP: mögliche code-switching Punkte) where code-switching occurs, is actually when interlocutors change. D I SCIULLO et al. ( 1 9 7 6 ) also showed that selfselection and/or selection of a new interlocutor represent structural factors promoting language alternation. Code-adoption, the exclusive adoption of a particular language in a situation strongly determined by social constraints (language classroom, TV program, etc.), depends on the degree of tension in the linguistic market as conceived of by BOURDIEU ( 1 9 8 2 ; 1 9 8 4 ) . As BAKER ( 1 9 8 0 ) pointed out, different definitions of the base language can lead to different results with regard to code-switching. The. base language concerns a particular verbal interaction, or a particular part of it. If it is determined by the presence of unilingual interlocutors or by the tension of the linguistic market, it is said to be determined by extradiscursive factors. If a situation allows for only one extra-discursive base language, code-switching will be of the extra-discursive type. If a situation allows for two or more extra-discursive base languages, code-switching will be of intra-discursive type. In so far as intra-discursive code-switching is concerned, each speaker has his/her own base language, that is one or more languages preferred at some moments during the conversation. The intra-discursive base language corresponds to the one the speaker deliberately activates. The intra-discursive base language of a speaker can also alternate. The based language could also be determined by a theme (a thematic base language). Theme should not be understood as a factor determining the possibility of producing code-switching. That possibility is merely determined by social factors concerning the conversational situation. But once that possibility exists, theme might favor a particular base language. Thus, once a thematic base language is imposed or suggested, any type of code-swithing is possible depending on the social factors concerning the conversational situation.

222

Normand Labrie

Before examining the role of social networks, a final factor in codeswitching should be taken into consideration: the competence of the interlocutors ( H A M E R S & B L A N C , 1 9 8 4 ) . D I S C I U L L O et al. ( 1 9 7 6 ) have pointed out that this is a necessary but not sufficient factor for the prediction of code-switching, that is the relative frequency of the use of language in a conversation between multilinguals. All of the studies relating code-switching to social networks (GUMPERZ, 1 9 6 4 ; G A L , 1 9 7 8 ; P O P L A C K , 1 9 8 0 ) refer to an anthropological concept of social networks. This anthropological conception goes hand in hand with longer periods of observation of the social context as exemplified in W H Y T E ' S study ( 1 9 5 5 ) . By following this inductive procedure it was discovered whether a locutor (speaker) belonged to a particularly significant peer group, socio-professional group or not. The relationship between the use of code-switching strategies and involvement in such social network could be established in most studies, although this kind o f conceptualization almost always implied a simple dichotomic distinction of the networks, that is belonging or not belonging to a particular group. Interesting conclusions can be drawn, particularly from POPLACK'S (1980) studies. POPLACK (1980) identified two ethnographic networks, similar to these described by L A B O V (1972b). Both included bilingual speakers and Spanish dominant speakers. Statistical analyses revealed no relationship between the involvement in a network and prediction of codeswitching types. This absence of significant results might be attributed to the fact that POPLACK (1980) did not take into account the individual characteristics of each member of the network, be it their position, their role, etc. But, on the other hand, some findings in P O P L A C K ' S results are of interest. Types of code-switching are influenced mostly by the dominant language (bilinguality and Spanish dominance) and the work place (either on their own block or not). Balanced bilinguals do more intrasentential code-switching than those who are Spanish-dominant. Those who work off the block do more extrasentential code-switching than those who stay within the block. And finally, those who are Spanish-dominant tend to do more emblematic code-switching. The originality of P O P L A C K ' S research is that through the study of actual code-switching occurrences, indicators of preference for certain linguistic forms have been determined (i.e. potential for linguistic behavior). The principal contribution of this study is that it established that this phenomenon occurs within its own inherent grammatical framework. It implies that the speaker has a special ability. Thus, the three types of code-switching can serve as a means for measuring the degree of bilingual ability.

Social Networks and Code-Switching

223

In order to be able to relate this individual potential for linguistic behavior to the concept of social network, it becomes necessary to have at command a concept which allows explanation of individual differences. As a result, a priori definition of the concept of social network is an absolute necessity. In this vein, the most concrete concept available in sociolinguistics is without any doubt the one developed by MILROY ( 1 9 8 0 ) for her study of vernacular use in Belfast. Once rendered operational through notions of density and multiplexity 2 , this concept allowed MILROY ( 1 9 8 0 ) to demonstrate that the people who most used the vernacular had a closed network, that is a social network characterized by density and multiplexity links. DITTMAR & SCHLOBINSKI ( 1 9 8 5 ) used the theory of probabilities in order to illustrate some weaknesses in MILROY'S ( 1 9 8 0 ) operationalization of the notions of density and multiplexity. In fact, the scale used by MILROY (1980) seems arbitrary in respect to the weight given to its components. DITTMAR & SCHLOBINSKI ( 1 9 8 5 : 1 8 0 ) propose to take into account the differences (1) between persons who are integrated into distinct clusters or into several clusters; (2) between the different items of the network strength scale (it might not be appropriate to suppose that kin relationships represent the same weight in respect to the integration into social networks as friendship relationships do); and (3) between frequency of contact. In fact, individual differences should be considered more precisely. This implies a new dimension to the notions of density and multiplexity. In another perspective, it is also possible for linguistic research to consider that an exchange network forms part of a social network (DITTMAR & SCHLOBINSKI, 1 9 8 5 ) . RUSSEL ( 1 9 8 2 ) mentions that the social network ties could correspond to potential communication channels, and could thus help explain code-switching behavior. ROMAINE ( 1 9 8 4 : 3 1 ) also insists on the fact that the social network concept is limited as such: "...networks don't explain anything over and beyond what can be explained by reference to the agents who form the networks". The social position of agent and his relations to others may constrain his behavior on a particular occasion in specific ways. It seems then evident that the social network concept should still be refined in order to be more fully integrated into sociolinguistics.

2

Density refers to members of he social network knowing each other. Multiplexity refers to members of the social network playing more than one role (i.e. belonging to more than one domain).

224

Normand Labrie

In this regard, a research group in Quebec has planned to investigate situations characterized by languages in contact, taking into account some aspects of social networks, including measures of density and multiplexity. For PRUJINER, DESHAIES, HAMERS, B L A N C , CLEMENT, L A N D R Y et.al. ( 1 9 8 4 ) individuals living in language contact situations (French and English) have a network including linguistic sub-networks. The linguistic composition of their network and the characteristics of their sub-networks could influence their linguistic behavior ( D E S H A I E S , HAMERS & L A B R I E [ 1 9 8 4 ] ; HAMERS, DESHAIES & L A B R I E [ 1 9 8 4 ] . The underlying conceptual definition of the social network implies that each individual is the focal point of a network formed by a definite number of persons in different domains. These domains, according to FISHMAN'S ( 1 9 6 8 ) notion of domains, are linked to the different aspects of the family, professional and social life. Instead of using an approach in which all members of a network are being observed, in the methodology that has been developed, the individual (as an informant) makes himself the description of his network. This conception focuses on the perception the individual has of the members belonging to his network.

3.Theoretical model and hypotheses In the study of code-switching among the Italians in Montreal, we might expect to find social networks including three kinds of linguistic subnetworks: Italian, French and English. Each sub-network can be related to the others by a definite number of density and multiplexity links. Each of these sub-networks can also be related internally by a definite number of such links. The image the network projects to the researcher includes the three following aspects: 1. the linguistic composition of the social network; 2. the dynamics of the social network (density and multiplexity between the sub-networks); 3. the dynamics of the sub-networks (density and multiplexity inside each sub-network). Considering these three aspects, it is possible to determine three different kinds of social networks: 1. A network with a multiethnic type of dynamics, that is a network formed of different ethnic sub-networks, including density and multiplexity links between the sub-networks (some Francophones know some Italophones, etc.)

Social Networks and Code-Switching

225

2. A network with an ethnocentric type of dynamics, that is a network formed of different ethnic sub-networks which include density and multiplexity links. There would be, however, no such links between the sub-networks (Italophones know each other, Francophones know each other, etc.) 3. A network with a monoethnic type of dynamics, that is a network formed of almost only one ethnic sub-network (which doesn't necessarily correspond to ethnic origins). However, even the marginal existence of relationships with people of different ethnic origins seems unavoidable in a multiethnic context. Considering such a conceptualization of the social networks, new observations about the meaning of density and multiplexity can be recorded: 1. An individual who has a network with a multiethnic type of dynamics (i.e. Francophones know Italophones, Anglophones know Francophones, etc.) often meets with people of different ethnic origins. According to his competence in the different languages (which can be improved over time) he will be brought almost inevitably to play the role of interpreter, that is to make use of code-switching strategies in order to assure intercomprehension. In the long run, he will have the opportunity to develop a greater mastery of code-switching strategies. 2. In a different but not contradictory fashion, an individual who has a network with an ethnocentric type of dynamics (i.e. Italophones know each other, Francophones know each other, etc. but Italophones don't know Francophones) often meets people of the very same ethnic origins. While investigating homogeneous groups, this is the level at which M I L R O Y (1980) developed her conceptualization of the social network. In a multiethnic context, the more a linguistic sub-network includes density and multiplexity links, the more powerful it should be in respect to imposing the use of a particular language. This would mean a higher level of respect of the norms prevailing in a sub-network 3 . 3. Finally, when an individual meets an interlocutor having density and multiplexity links, this, paradoxically, does not imply that he will have to observe the norms of the sub-network the interlocutor belongs to. On the contrary, density and multiplexity seem here to correspond to an expression of the familiarity of the relationship between the two. The locutor will have the possibility of playing a more active role. This might be how the individual contributes to define the norms inside a sub-network of even the social network he belongs to. 3

The norms of a sub-network could also eventually dictate the use of code-switching.

1. Composition of the social network

2. Dynamics of the social network and relative dynamics of the subnetworks

Organic

Conjunctural

Explanatory factors (situational)

+ 1 1

Ethnocentric

M C

IS

O CX M) C

-a o

-a

c

!•

trt C o

XI

fi

2.3

— X ^Nl-H + 1-1- 1 + 1 -1-1 + ISN (FSN) (ESN)

ISN FSN ESN

3 ® üTiTw g

Use of extra-discursive types of code-switching

2.2

k f c f c S g ^ 1+ —1 +^ +^ S[jj— XH

+ / -/ +

üj PT 1+ —1+

- /+ ISN ESN

ISN ESN - /+ + / ISN FSN ESN - 1 + 1 -1-1 + + 1-1ISN FSN + / -

- /+ + /-

ISN FSN

Types of Codeswitching and preferred base language % k, txT £C H + S+

Multiethnic

2.1

Relative dynamics of the sub-N.

Dominanttypesof intra-discursive code-switching

Catalytic Dynamics of the SN

Presence or absence of code-switching (in relation to the presence or absence of Italian, French or English sub-networks)

Conditioning

Dynamics of the SN and sub-N.

Types of linguistic behavior

Effect

•a c g

2.4 Existence of a subnetwork without any dynamics

Predictive factors (structural)

Status of the relationship

226 Normand Labrie

g w ÜJ | g ? g k , fc 3 fT - X NH XHH^ß r^^

3

227

Social Networks and Code-Switching

! W

> M

:

1u 1a

SV S , S 6 u »

su a 3 ST E w a

=3 1 »3 X V u

§ C

-D o u u -i VI 3

•ä 2 S X W

•I I

y

c 1 ä HS

9 " v T) o U

W x — ^ &

O W J3 u ^ il [Ii

.2

I 1

"

I

•c -S u **

6U •s

a •s o c o 2

Q «

3

£ w

3 w y O bO 9 & I 2 1 2 o

J 'S

£ 2 •a a

««A1 O

2« '3 -S a

a 1 äö a« »C aw en ' .S fi « fi I I I

Jä.g

o i»

•S

- a s ü c . S u "

c 8

i R . 5 « e « P "O _ e • i g ì •a v .a « l o I Jg 3 CO O I

I S S u § E S i 3 S. M 1 & .s -s gro:, grey) and by diphthongation (Hose > House, trouser). On the basis of such phonological differences, people of the Odenwald are said by urban people to be barkening. And different sound boundaries cross the territory of the actual Mannheim (BRÄUTIGAM 1934, 145). In Mannheim (in the central parts of the city) exists a traditional urban vernacular, which is put into opposition to the rural dialects of the surrounding region by BRÄUTIGAM in his study ( 1 9 3 4 ) . Remnants of the differences between the old rural dialects still remain on the urban territory. The most frequent exemples used by Mannheimers to indicate this internal differentiation are e.g. Monnem (center) vs. Mannem (south of M.), Neggarau vs. Neggara:, or Roi vs. Reí (Rhein). The linguistic presence of the historical map seems to be quite typical for a city like Mannheim with a rather modest and historically rather new centrality (for Duisburg see MIHM 1985b). Inhabitants with a clear consciousness of the local culture know such features. But in the process of the urbanization of the whole region,

238

Werner Kallmeyer / I n k e n K e i m

the opposition 'rural' — 'urban' is changing its character and is partly losing its social significance. There is probably — as in other urbanized regions too — a process of regionalization, where different variants are still identified with certain areas, but where social and situational factors play a more important part in the definition of the social meaning of language, than geographical ones. Great parts of the population in Mannheim are using a dialectal variety as normal colloquial language (like generally in the southern part of German; SCHUPPENHAUER / W E R L E N 1983; CLYNE 1984). Naturally, there is a high presence of other German varieties as well as many foreign languages in a highly mixed population, which is due to the great population movements after the second world war, to the immigration of people of German origins from the eastern parts of Europe since the 1970s, the inner-German labour migration and, finally, the international labour migration; even in local groups, it is quite common to find people from other parts of Germany, who have to develop specific speech characteristics for their social integration. There is a specific cultural identity in Mannheim, at least the image of such an urban culture indicated by elements of an urban mythology such as originals, literary figures etc. (e.g. the "Blumepeter", a flower seller in the 20s, well known for his merriness and mother wit), and by stereotypes about the Mannheimer and his speech. The idea of a specific speech style which is significant for the local identity is expressed by names as "Mannemer Gosch" ("Gosch" is a dialect word for "mouth"), or more specifically "Filsbach Gosch" (with reference to one part of the inner city, where historically the "broadest" urban dialect was spoken by the lower class population) or a name like "Bloomaul" ("blue muzzle"; a "Bloomaul" is somebody with a humorous, but direct speech style, speaking his mind and never being at a loss for a witty commentary). Such stereotypes are traces of already historical processes in the formation of local culture, and their mere existence doesn't tell us very much about the vitality of this culture, but only the actual use made of such elements and the formation of new ones. The official cultural politics are consciously using the historical resources in the last years in the course of efforts for a renaissance of urban culture ("Bloomaul-orders" are conferred, "Blumepeter-festivities" are celebrated etc.). These practices are changing the character of the symbols, and in this sense, they are not continuing the tradition. But the official practices are part of a probably more general search for local identity, which is carried out by the daily activities of certain groups of inhabitants too.

The Symbolization of Social Identity

239

The evaluation of the Mannheim urban vernacular (m.u.v.) is controversial. On one side, it is the positively evaluated language of the urban identity (since the efforts of a renaissance of urban culture, this is the case even in the official perspective), on the other side, the m.u.v. is defined as sociolect in opposition to near standard language, at least certain levels of the m.u.v. (especially the "broad" variety of the lower class milieus in the inner city). In comparison to other regional or local dialects, the m.u.v. seems not to be connoted as an original, salient and expressive language. Finally, the m.u.v. is no language of prestige for the local elite (like the "Honoratiorensschwabisch" in Stuttgart, ENGEL 1 9 6 2 ) . The Mannheimer has his language pride, but it has a defensive character, which is often expressed by a formula like "our language is a language too".

3. The ethnographic and linguistic approach The project is designed as an ethnographically based qualitative study, combining global orientation with intensive observation of small populations and detailed analysis of their speech behaviour. Our plan of ethnographic observation comprises the following steps: — general view of the city, its historical development, its regional integration and internal segmentation, the main characteristics of the linguistic situation and the symbols of urban culture. — ethnographic panaroma of different, mainly or partially residential districts, contrasting in regard to criteria such as 'center/periphery', 'old/ new part of the city', 'incorporated older community etc. / new construction', 'relation to the organizations of the community' etc.; central points of the description are the socio-spatial structures, social categories and segmentations, forms of organization, central events and problems of the social life. — portraits of local groups of different life ages (youth, 'middle age' with family duties and integration into professional worlds, and post-retirementage). These groups were observed in the surroundings of their living places in situations that are linked with the local organization of life, and that form part of their local world (insider communication); furthermore they are observed in contact with relevant others in situations at the borders of that local world (very often situations of conflict with other groups or with institutions), and thirdly, in contact with mostly unknown others in situations outside the reach of local organization, situations that

240

Werner Kallmeyer / Inken Keim

are for the inhabitants foreign or open territory (mainly downtown, but also public recreation places). The speech behaviour of these groups is analyzed with respect to their activities of social organization and their forms of symbolizing identity. In the first perspective we concentrate on: — the organization of contacts — the constitution of discourse universes by talking about the world (see the second part of this paper and in K A L L M E Y E R / K E I M 1986a, K A L L M E Y E R I K E I M 1986b) — the cooperation for practical purposes. The analysis of symbolization has three main objects: — code switching (see the second part of this paper) — stereotypes and formulaic speech ( K A L L M E Y E R / K E I M 1986a) — lexicon and social categorization ( K E I M 1 9 8 4 ) . The observations at these levels will be used to characterize significant forms of speech behaviour in Mannheim. The result of the comparison of the different portraits will be a description of the speech style of different urban groups.

4. Analysis of the symbolization of social identity In the second part of this paper, we will briefly demonstrate our sociolinguistic analysis of one instance of symbolizing behaviour. Our example is a story about a contact situation at the border of the locals' own world, treating one specific problem of local life. The story is told by a member of an inner-city local group in a typical in-group communication at the usual gathering place. The following remarks may indicate the significance of the example. The district of Mannheim we are talking about, the "Westliche Unterstadt", is a part of the inner city, located at the lower end of the social scale. It is the only part of a specific urban culture of the lower class which existed here at least before the second world war, and in the first years afterwards; remains of this culture still exist. It is the only inner-city district with a popular name ("Filsbach"). The population of this place is said to speak the broadest Mannheim variety. Since the 1960s, activities of urban renewal have been going on, destroying large areas of the historical Filsbach. Great population movements took place, and today the population is highly mixed; there are Filsbach natives, Germans from other parts of Mannheim or from other parts of Germany, and many

The Symbolization of Social Identity

241

foreigners, above all Turks. The area is socially rather unstable, and one of the main problems of the native population is the destruction of their local networks. There are different reactions within this population to the structural change: resignation, gathering movements etc. In comparison to other urban districts, this one has a clear want of formal organization with regard to community life (e.g. the district does not correspond to a political unity). Structural changes and political or social movements are experienced by the locals as essentially alien to them and initiated from outside. The participants in our example are women of post-retirement-age mainly coming from the district or having lived there for a long time, and belonging to the Filsbach world. They are representative of a greater part of the German population due to their age and sex, their attitudes towards the Filsbach world and their reactions to the structural changes. They form a group only with respect to the group meetings. Outside, they are scattered over the district and neighbouring parts of the city and have little contact with one another. But the group meetings are highly relevant to them, because they play an important role in their social life. The group meetings are a point of crystallization for their existence within the framework of the Filsbach world. Here, they can produce and actualize their knowledge about the spatial organization of Filsbach, its protagonists, its central values, the significant types of social deviances, and the problems of the area. The history of the group reflects the processes of social destruction and reconstruction within the area. The women, now participating in the group life, needed a place where they could meet, and the stimulus of an institutional organization to come together and to develop a group structure. The place is provided by a socio-pedagogical organization that organizes meetings with a culture programme, mainly based on middle class concepts of leisure time activities. But once put together, the women developed a group life of their own on the basis of certain cultural procedures particular to the Filsbach world as, e.g., the manifestation of a marked attitude of hedonism by telling dirty jokes or by being very direct about physical and sexual matters in gossip and in biographic narrations. So, during the group meetings, the frame of reference is mainly the Filsbach world. The setting itself is in some sense extraterritorial for that world (an institutional place run by people belonging to other social categories and with quite different ideas), but to a certain degree the place has been taken over and transformed into a part of the Filsbach territory. Here, the participants can practice their cultural patterns, whereas in other places, in which they come together (provided e.g. by Christian

242

Werner Kallmeyer / Inken Keim

organizations or by the community), this is not guaranteed; there, their territorial rights are even disputed. This is what the story of our example is about. The content of the narration, which has a clear episodical structure, is the following: (a) W1 (the narrator) together with her husband and W24, another member of the group, had fixed a meeting at the usual Monday meeting place which is a municipal institution organizing a social programme for elder people in the district. Our women are using this place as a meeting place, without participating in the programme's activities. (1,1-1,7) (b) W1 and her husband find the door closed, and the manageress of the institution refuses to allow them to enter because of a closed party, and she proposes they wait for W24 on the street (it was a cold day in the winter). (1,7-2,9) (c) Using the opportunity when other guests go in, W1 enters, meets the manageress and insists on explaining her concern, which is refused once more by the manageress. She points out that she doesn't even know W24. (2,10-2,25) (d) On the street, W1 and her husband find a solution for the practical problem of how to meet W24, who is already on her way. (2,25-3,15) (e) W24 manages to get into the place, meets the manageress and confronts her with her claim to be a known and admitted member of her clientele, but the manageress still treats her as an unknown person. (3,16-4,3) (f) W1 phones the superior institution and complains about the manageress who, in the meantime, evidently had already spoken with the employee W1 is talking to. The employee declares that the behaviour of the manageress was wrong with regard to the institutional rules. He advises W1 not to give up the place, and he offers to speak once more to the manageress and to reprimand her. (4,3-5,23) The topic is quite interesting for the group, and the manner of telling the story is evidently successful, as can be seen by the reactions of the audience. At least one part of the success is due to the use of symbolization as a means of performance, especially in the reported speech. Reported speech is a very common place for symbolization, and upward or downward code-switch is a common device for it. Symbolization as procedure of expression is part of the normal speech activities in conversation. It is incorporated in the constitution of conversation, and it has to prove successful as a conversational technique. So it is

The Symbolization of Social Identity

243

useful to analyze the structure of the symbolizing procedures starting from the conditions of their success. And these are: (a) their functionality for the constitution of conversation, (b) their perceptibility, (c) their interpretability, (d) their usefulness as social analysis, (e) their quality as reflection of the linguistic conditions of the participants. These different aspects may serve as a guideline for the following analysis of our example. a) The functionality of symbolization for the constitution of conversation: Symbolizing procedures are, in different ways, functional for the constitution of conversation; they may be applied to mark the reported speech of others and to mark central parts in narrations; or they may be used for indicating conversational modes, e.g. the change from earnest talk to playful attacks etc. Two of these ways of functionality can be demonstrated in our example. Here, symbolizing procedures are used to mark the speech of others, and they are applied to emphasize important points of the narration. As the reactions of the listeners indicate, this is narratively very successful: — The first appearance of the narrator's opponent (episode b) is an important point in the narrative structure; here, the presentation of the conflict takes its start, and the social constellation the conflict is based upon is indicated. At her opponent's first appearance, the narrator quotes her by changing the local speech variety (to near standard) as well as her manner of speaking (higher voice, sharp rhythm, overarticulation). This first quotation of the narrator's opponent raises the listener's spontaneous indignation (1, 13 u. 14; for the analysis see below p. 245 f.) — In the judgements of her opponent's actions, the norm system of the narrator's social world and that of her listeners is involved and confirmed in the evaluation of her opponent's actings as deviant. In the first judgement (end of episode b) the narrator depicts an alternative way of acting in accordance with her own social norms: (2,1-2,6): an anschdannischi mensch hatt gsacht (a decent person would have said) do=s e vorraum (there is an anteroom) g6h se mol doroT (come in) setze sisch do hie (sit down)

244

Werner Kallmeyer / Inken K e i m

wié sisch dés géhert 1 (as it should be done) This judgement is given in the narrator's dialectal variety (in contrast to her opponent's near standard) combined with a specific rhythmical figure. Her listeners agree spontaneously with her (2,4 u. 5). Her second judgement (end of episode d), in which she gives an explicit evaluative term for her opponent's actions: (3.10-13): des iss doch eine Unverschämtheit 2 des is dox asna unf3Kjse:mdh$d h (that is an impertinence)

is marked by near standard phonetics (contrasting with her preceding narrative style in dialect) combined with a specific staccato-rhythm. Transposed in dialectal phonetics, this sequence would be: des iss e üvaschämtheit dss is 9 u:fa3fa£:mdhaed At the climax of the narration (end of episode f), when the superior of the manageress shows his agreement with the narrator's idea of adequate social behaviour, the narrator quotes this superior, changing in the last quotation line from the preceding institutional language into a very low dialectal variety (for analysis see p. 246 f.). This last quotation evokes the listeners' spontaneous expressions of admiration and astonishment. (5,2023). These are three examples where symbolizing procedures are functional for the constitution of the narration. b) The perceptibility of symbolization There are different ways of marking symbolizing procedures relating to their conversational context. For example, specific means of expressions that signal specific meanings by routinized use in specific groups can be applied casually and unobstrusively, nevertheless evoking the whole meaning system of this group, but being incomprehensible to outsiders. In contrast to that, in our example, symbolizing procedures are marked by deviation from the "normal" speech of the narrator, or by the deviation

1

The rhythmic signs: " = unstressed; = secondary accent; = main accent.

2

The phonetic transcription follows mainly the API-system.

245

The Symbolization of Social Identity

of "others" from their normal speech towards the dialectal variety of the narrator. Here, the perceptibility of symbolization is constituted by locally produced contrasts between the narrator's speech and the speech of "others" on different levels: the phonetic level (switch from a dialectal variety towards a near standard and vice versa), the lexical level (switch from colloquial style towards official/institutional style and vice versa), and in the manner of speaking (change in pitch, rhythm, tempo and articulation). This locally produced contrast can be demonstrated at two important parts of the narration: At the first appearance of her opponent, (episode b) the narrator changes on the phonetic level from her preceding dialectal variety into a near standard, on the prosodic level from her normal pitch and her quietly and equally flowing speech-rhythm into a higher pitch and a sharp rhythm. Even the lexical items belong to an official style. (1,10-12) preceding sequence:

appearance of the opponent:

un no kummt die fra: H. raus un w no kumt di fRaa H. Ra°s (and there comes Mrs. H.) sie konnen heut nicht rein zi: kceni" hoyt nift Rawen (you can't come in today) heut ist geschlossene gesellschaft hoyt ist g a j b s ' n a g3zeljaft h (today, there is a closed party)

The change on the phonetic level is marked by the following characteristics 3 : Near standard vs. Dialect

Main changing processes (from standard towards dialect would be)

können kceni"

unrounding of the vowel, deletion of the final nasal in unstressed /-en/endings

3

kenne ksna

Here the standard characteristics are compared with the narrator's "normal" dialectal variety in the text. The term "changing processes" does not refer to the historical development of the dialect, but to processes taking place in the narrator's code-switches.

246

Werner Kallmeyer / Inken Keim

heut hoyt

heit haet

unrounding of the diphthong;

nicht ni$t

net net

lexical variant in the dialect

rein Raen

roi Roy

rounding and nasalization of the diphthong, deletion of the final /n/

geschlossene gafbs'na gesellschaft g 3 ? elfaft h

gschlossini gflos'ni gsellschaft gselfaft

deletion of the vowel in the /ge/-prefix and /¿/-ending in the fem. adjectives in nom. and acc. sg.

The change on the prosodie level is marked by: — higher pitch — faster tempo — change of rhythm: in the preceding parts, every syntactic unit has one or two main accents and a few secondary accents. In the quotation, in contrast, a syntactic unit has one main and only one secondary accent; the other syllables are unstressed. The change on the lexical level is marked by [geschlossene gesellschaft] and especially by the following [interner geburtstag] (a friend's birthday party), expressions, that do not belong to the "normal" language use of the narrator, as all the recordings do prove. Quite another kind of linguistic changing takes place in the last quotation of the manageress' superior, when he offers to speak with the manageress and to reprimand her (episode f; 5,17-19) preceding sequence:

hoffentlisch hat sie e paar gute ausre- reden hofontli5 hat a pha:K gv:ta aosRe:- Re:dn (I hope she has some good excuses zur

hand hanj at hand)

JSV:K

last quotation:

schunsch scheiß isch sie zomme fünf faps if si: tsoma (otherwise I will dress her down)

247

The Symbolization of Social Identity

Here, change takes place especially on the phonetic and the lexical level. The preceding sequence shows signs of a kind of institutional language in the verbal construction "ausreden zur hand haben" (to have excuses at hand), and it is given in near standard phonetics: Near standard vs. Dialect

Main changing processes (from standard towards dialect would be)

hat hat

hot hot

darkening of the vowel

hand hant

hond hond

gute

gude gv:da ausredde aosRE^a

gv:£3 ausreden ausRe:dn

softening of the fortis; shortening of a long vowel, hardening of the lenis, and deletion of the final nasal.

The following sequence contrasts very clearly with the preceding: the pitch becomes lower and, especially, the lexical items and the phonetics change towards a low dialectal variety: Standard

Dialect

Low dialect

sonst sonsth

sunsch

schunsch JunJ

Main changing processes (from standard towards low dialect) receding of the vowel and deletion of the final /t/ from standard to dialect; regressive assimilation of [2] in front of a vowel and following [J] to [J] from dialect to low dialect.

Standard

Dialect

zusammen tsu-sam'n

zusammen tsoms

darkening of the vowel; deletion of the vowel in the /zu/-prefix and coincidation of [ts] + [z] to [ts]; deletion of the final nasal in unstressed /en/-endings.

The rough scolding expression [zommescheiBe] contrasts very clearly with the preceding institutional language. c)The interpretability of symbolization Symbolizing procedures are brought about by applying a specific kind of marking and evaluating that establishes relations between social constella-

248

Werner Kallmeyer / Inken Keim

tions (social situations, units, events), specific features and specific values. Mostly, single components and relations are not locally but only contextually conceivable. The interpretability is guaranteed by the following textual characteristics: There is a textual coherence in referring to social constellations and in expressing evaluations. In our example, the social constellation the symbolizing procedure is referring to is a central part of the topical structure of the story. The story presents an argument between a woman of the local "Filsbachworld" and the manageress of a social institution. This manageress belongs to one of the new groups, invading the district after the urban renewal in order to establish new forms for social gatherings and new programmes for free-time-activities that are for the most part not ¡accepted by the Filsbach population. But the Filsbach women try to use the new spatial and material offerings in accordance with their own concepts for free-time-activities, which is rejected by the manageress, who wants her programme to be accepted. The core of the argument between the narrator and the manageress is a cultural and territorial conflict between a part of the local (working class) population and the newcomers with social conceptions contrasting sharply with the established local conceptions. But this social constellation is not only topicalized in the story, it is also present in the actual conversational situation itself. In the actual meeting one of the members of the socio-pedagogical organization is participating (see above p. 241), who is organizing the free-time-activities for this group of elderly women. It is to this woman especially that the story is directed. In the actual situation, she is treated by the local women as belonging to the same social category as the manageress in the story, being an institutional agent and, therefore, belonging to the "others". But in contrast to the manageress, she is a decent kind of institutional agent and a person, who tolerates the local women's different living style. Nevertheless, the story's subject, the presentation of the complete misbehaviour of an institutional agent and the narrator's contra-actions, is a kind of warning to her, that she should not try to change her present tolerant attitude towards the women and to take over a hard institutional line. The next point is the expression of characteristics and evaluations with various means. To begin with, there is the linguistic change on several language-levels in relation to the context, as described above. Secondly, there have to be explicit verbal expressions for the relevant social categories which the symbolizing procedures are referring to. In our example, after the presentation of the manageress by means of switching on several language-levels, we find an explicit judgement on her verbal behaviour

The Symbolization o f Social Identity

249

"in dem ton" (1, 17) (she used the wrong tone), and on her acting. Her behaviour is opposed to that of an "anschdännischa mensch" (2, 1) (a decent person), and it is explicitly evaluated as "eine Unverschämtheit" (3,10-13) (an impertinence). A third feature constitutive for perceptibility is the consistent use of the symbolizing means. In our example, there are switch-phenomena hardly comprehensible in their function, when taken alone and isolated. But seen in the context of the clearly marked parts, these phenomena (consisting of the same types of symbolizing means, perhaps not all of them, and less marked) become interpretable as symbolizing forms. In symbolizing procedures, certain sequences of expressions regulate the sequential structure of meaning. For instance, an alluding announcement is followed by a manifest symbolization, or the symbolizing procedure starts with clearly marked forms and weakly marked forms follow. In our example, the sequential structure of the symbolizing procedure is: 1. clearly marked linguistic forms on the phonetic, prosodic and lexical level for the different participants of the story; 2. explicit verbal expressions for the social categories which the different linguistic forms are indicating; 3. weakly marked linguistic forms, either on the phonetic, the lexical or the prosodic level. The way of marking is consistent with the marking under 1. The fourth constituent for interpretability is the knowledge of the social constellation the story is based upon for the participants in the conversational situation. Using presuppositions and implications, the participants indicate to each other their familiarity with the story's subject. In our example, the social constellation is never explicitly depicted, neither by the narrator nor by her listeners. A lot of information is left out that would be necessary in order for an outsider to grasp "the point" of the story. Nowhere in the story is the narrator's opponent presented in her institutional function, nor is the relation between the manageress and the institutional agent described; nowhere is the institutional situation of the district and the relation between the local population and these new offerings explicitly depicted. And above all, the overall situation of the district, its socio-ecological change, is left out of the story. Names, indexical items and, especially, the symbolizing procedures in the story suffice to evoke to the participants the whole social constellation. d) The usefulness of symbolization processes as social analysis. Symbolization procedures refer to living conditions and to experiences especially relevant for the participants, and they make these conditions

250

Werner Kallmeyer / Inken Keim

present to them. The participants show their understanding of the indicated relations in their comments and their appreciation of the narration. In our example, the narrator points out some of the relevant experiences of the group: their dependency on institutions, and their living conditions being determined by others. Institutions and their offers of free-time-possibilities as well as of financial help in many respects constitute an essential part in the life of many of the local people. Therefore, the narrated event, that an institution they had access to, suddenly closed on them, is a threatening experience, and the narrator's successful solution of the problem has a reassuring and comforting effect. The narrator presents her story as a kind of recipe for success in dealing with institutions. But the story leaves the question open as to whether the participants really caught the institutional mechanisms and fitted the superior's actions into institutional requirements. Here, the institutional problem is only dealt with on a personal level. The participants feel reassured, because the narrator found a decent person in the hierarchically higher institution who showed interest in her problem. The institutional and the political aspects of this person's actions are not focussed on in the story. In handling the story's subject the participants do not analyse the institutional conditions and do not show a critical and political attitude towards the institution; there is only personal relief in finding understanding in an alien and distant world. These findings correspond very well with other observations about these women, concerning their dealing with institutions. Everywhere there can be found a great distance between local people and the institutions, and little knowledge on the side of the local people about the institutional organization, institutional constraints and mechanisms. The actors in the narrations are clever in using institutional offerings and in coping advantageously with the institutional agents, but they accept the world of the institution without questioning its ends and mechanisms. The second important point of the story is the implicit reference to one aspect of the local people's present living conditions in the district: there is a clear deficit in formal organization for the local population and, therefore, they have no chance of participating in decisions concerning the rapid change in the district's socio-ecological structure. All the decisions concerning the district are made elsewhere, and out of the reach of the local people; they are alien to them. Even the socio-pedagogical organization (which the local women came to an agreement with), coming from outside the district, but with the proclaimed aim to offer possibilities of participation on the level of formal organization, and even of taking over the whole organization, has not yet succeeded in cooperating with

The Symbolization of Social Identity

251

the local population on the institutional level. Local people do not really participate in the organization, they use it in accordance with their own conceptions. e) Symbolizing procedures as a reflection of the linguistic conditions of the participants Symbolizing procedures, as procedures for the local and textual constitution of meaning, do not imply a fixed relation between certain means of expression and certain social meanings. In our example, the use of near standard forms has different meanings; in the definition of a specific meaning prosodic features play a central part. The near standard forms in the presentation of the manageress, given in the specific articulation described above, signal a kind of social distance that is clearly negatively valued. On the other hand, the standard forms of the manageress' superior, combined with specific lexical forms and different prosodic features imply social distance too, but one that is clearly positively valued. The near standard forms in the narrator's explicit evaluations give the utterance force and emphasis. In other contexts, this narrator uses standard forms simply to quote a standard speaking friend. Nevertheless, in symbolizing procedures, the use of special linguistic forms reflects specific linguistic conditions, as there are: the knowledge of the language use of "others" as well as within the own social world, and the knowledge of the usual symbolizing practices. In the Filsbach world, the use of the standard language is certainly a traditional means of expressing social distance (but the use of standard does not always have that meaning!); and the overarticulated, sharp and pointed way of speaking is a routinized procedure for the exaggerated presentation of a specific social type. In our example, the narrator uses dialectal forms to refer to people and situations of her own social world; these people really are dialect speakers. In contrast to that, the world of "others" is indicated by standard forms. The manageress of the institution, coming from northern Germany, really speaks a near standard variety, while the manageress' superior is a Mannheimer, knowing official near standard varieties as well as dialectal varieties. So, symbolization reflects in some aspects the linguistic reality; but, at the same time, the (real) linguistic forms of others are used to express the users' evaluation of the others' actions. Symbolizing procedures not only reflect a part of the linguistic reality, but also the linguistic repertoire of the participants. The dialectal features of the participants are roughly those that were described by BRAUTIGAM (1934) in his study of the urban dialect in Mannheim. But the participants' dialectal variety is not homogeneous; there is variation throughout their

252

Werner Kallmeyer / Inken Keim

language use. One type of the non-symbolizing variation is the switch from dialectal to standard forms and vice versa in order to perform conversational tasks, e.g. emphasizing, focussing, framing, etc. This switching has the same phonetic range as the symbolizing switching. That means that variation between dialectal and standard forms is part of the "normal" language use of the participants. But, in symbolizing procedures, the linguistic continuum between dialect and standard is segmented; different languages are marked by consistent choice of specific characteristics on different language levels, and different social meanings are attributed to these different languages. One of the results of this paper, as has been indicated above, is that in symbolizing procedures the linguistic reality is reflected differently. Our findings in this matter — in this example and in the whole corpus (as analyzed till now) — are, in short, the following: In symbolizing procedures there is — the mirroring of the linguistic reality: As in our example, standard speakers are presented in standard forms. Here, according to the linguistic ability of the speakers, standard may only be indicated by some features. Dialect speakers are presented in dialectal forms. — the presentation of a social topology of nearness and distance: Here, as in our example, dialectal forms are used to refer to persons and situations of one's own social world, and to indicate social nearness. In contrast to that, standard forms are used to refer to persons and situations of an alien (not always standard speaking) social world, and to indicate social distance. Here, the speaker's evaluations of her own world as well as the "other's" are mostly indicated. — the caricaturing exaggeration of special features of the linguistic behaviour of "others": These are based on perceived linguistic features. The procedures are e.g., as in our example, over-articulation and sharp pronunciation, or underarticulation and mumbling pronunciation (as in other texts). Here, the speaker's evaluations are always indicated, but there is no strict ascription of positive or negative evaluations to specific linguistic features. — the caricaturing use of linguistic stereotypes to characterize a specific, not only linguistic, but overall social behaviour: Here, the linguistic stereotypes used do not necessarily have a relation to the presented person's "normal" linguistic behaviour. In our corpus, although not in this text, the participants of the group use a special kind of exaggerated standard features together with a specifically sweetened prosody to refer to a person's faked nobleness. The linguistic forms used may, to

253

The Symbolization of Social Identity

some degree, reflect the linguistic behaviour of this person; but the depicted person could also be a dialect speaker throughout with other (non-linguistic) attributes of faked nobleness. Here, the linguistic forms used function as a metaphor for fake behaviour.

APPENDIX: Text Example transcript conventions: =

0,00 (...) gern )

a

"

5s5 5f5 3s 31 21 2s ä 1,1 W1 2 W1

contraction pause not understandable simultaneous speaking strong accent rising intonation falling intonation gem s3 louder oft 13 softer verstehst du 12 slower tempo verstehst du s2 faster tempo nasalization of a vowel un am letschte möndach sin mer a hie un die frau W24

hot noch gsacht 0 ah na 0 isch kumm nunner un do haww-isch

3 W1 gsad 0 ah nit so frie so um drei rum nit 0 des reischt 4 W24 ia jetz 5 W1 6 W24 un isch hab noch en (...) getroffe nit mer wollte donn die (...) 7 W1

O loß misch mol des weitererzähle jetz geh mer hie 0 un jetz war

8 W1 die düer vaschlosse net 0 zugeschlosse 0 no haww=isch no haww« 9 W24

u

10 W1

isch gschellt 0 geläudet un no kummt die frau hansen raus un mach

11 W1

die tür auf also nach de stra"ß 2s sie können heut ni"cht rein

12 W1 heut ist geschlossene eesellschaft s2 un na haww«isch (rocht 0 ia ach" 13 WE 14 W24 ach"des gibts doch net

Werner Kallmeyer / Inken Keim

254

15 W1

was" für ä gschlossini gsellschaft 0 aja" also das äh 0 da iss

16 W1

interne geburtstach und äh 0 also sie können nich"t rein 0 2s sie

17 W1

können am mi"ttwoch kommen aber heu"te nicht s2 0 net in dem ton 0

18 W1

nit 2s sie können äh am mittwoch kommen aber heute nicht s2 0 äh un ...

19 W1

dann 0 no haww=isch gsacht ja wer wer leid wer is=n do 00 nit

20 W l 21 W8 22 WE

wer machten do de (...) ob=s stimmt odder nit 0 wer hot-n dezu oigelade odder sowas

23 W1

amerikaner 00 un do haww=isch gsad so" 00 un 0 un wie isch

24 W1

dann weiter no macht se die düer zu un sescht 0 a warde se mal 0

2,1 W1 2 3 4 5

uff de 5s5 str"aß 5s5 2s än anschdännischa mensch hätt gsacht

d o s e Vorraum gehe se mol do roi setzt sisch do hie Wl WE W24 G^Lia W8 gehe se roi

u

K 6 7 8 9

ah die deut"sch-

Zustimmendes Gemurmel

W1 wie sisch des gehert (...) er nit 0 odda de Vorraum s2 00 un WE es hat ja en schönen Vorraum W24 U ) W8 ja genau K

Beifälliges Gemurmel

10 W1

isch hab gewart 0 ä weill 0 un-d-äh uff ämol kummt die 0 mit

11 W1

der 0 mit der drehorgel" dear net 0 die sin dann rein un 3s donn bi

12 W1

isch auch reingegangen s3 un hab zu moin mönn gsacht wadde mol do

13 W1

21 isch will gugge wer do drin is do wäre die leut drin die so

14 W1 15 WI

immer halt 12

16 W1 17 WI

wäre net drin wu 0 nur möndachs kumme sondern wu dienschtags un ah so ja

ah no 0 ah s=war ziemlisch mehr drin ne die" die mondags immer kumme jaja

18 W l mittwochs die warn drin 19 W24 die wäre die hawwe all 20 W l 21 W4

un 00 no haww=isch donn do geguckt nit 0 wer do drin is bescheid gewißt

255

The Symbolization of Social Identity

22 W1

no haww=isch ja 0 no haww=isch zu=re gsacht 0 es iss so" isch treff

23 W1 24 WI

misch mit de frau W24 hier deswege 0 nit 05s5 wer iss=n die mhm

25 W1

frau W24 5s5 die weeß des ganz genau 0 nit hab gefrocht un do haw=isch

3,1 W l

gsacht willli was mache ma donn jetz net isch weeß jetz net isch kann

2 3 4 5

Wl nimmer telefoniere s=iss kurz vor drei 0 die frau W24 is nimmer W24 äh W8 U WI mhm

6 Wl

dahäm die is schun unnerwegs net komm mer laafe mol vor= n= de

7 Wl

marktplatz 0 äh un gugge no un=an hamma se aa 0 getroffe also mer

8 Wl

hawwe und donn getroffe 0 no haww=isch gsad luis denke mol was.

9 Wl

mer ewe passiert is die hot uns garnit noi uff de + Straß + 0

K 10 W l 11 WE 12 W8 K

+ Betont + uff de + Straß + abgferdischt nit (...) 21 des iss doch eine unverun verschämt + Betont +

un"Verschämtheit fresch iss des 12 un 13 W l un isch bin dann noi 14 W24 15 W8 e freschheit des is e freschheit ja K

Sehr erregt

Zustimmendes Gemurmel

00 ja un no is sie noi 0 ne 0 hawwisch gsad geh mol 16 W l 17 W24 allä 18 W l

noi 00 un no hot sie erzählt un nach=ere weil sach isch 0 herrgott

19 W l

will die drin bleiwe wenn isch des wisst ginge mir häm odder

ginge wu annerschd hie 0 un do sescht de willi ei isch gugg 20 W l 21 W24 ah sie hot (...) 22 W l emol nit un isch hab 0 wesch jo wenn isch sag ( „ J die hot misch 23 W24 24 W24 vielleisch fünf mol gfrocht ob isch die frrau W24 war haw=isch (Lacht) also sie 4,1 W l 2 W24 gsad sie wisse ganz genau 5f5 wer isch bin 5f5

256

Werner Kallmeyer / Inken Keim

3 W1 is dann noi net 0 j a = s kummt noch weider geht noch weider de zirkus 4 W24 GJ K Lachen 5 W1 6 WI

bisch du ferdisch 5s5 ja 0 un dann haw isch gedacht so" isch erzähle se weiter

7 W1

loß ma des jo net gfalle i"sch 0 un dann haw isch zu=er gsacht

8 Wl 9 WI

morje frie ruf isch=s Sozialamt ä net 0 haww«isch gedenkt 0 daß die 0 ah aha

10 W l 11 WE 12 W8 K 13 W l 14 M

isch will des wisse

un no haww=isch

un 0 no

U)

u

Begrüssung eines Hinzukommenden (M)

hots gheeße erseht ja der herr sowieso iss net

der iss 0 guten tag

15 W l noch unnerwegs isch soll dann 16 WI grad so ä interessante gschicht 17 W24

u

K 18 W l 19 WI K

Geschirrklappern, andere sprechen

ja äh un dann 0 äh 00 ja no haww»isch ägerufe mache se mol weiter Neuankömmling an

20 W l

un do het er gsacht das hätte ja 0 3s un no hot des lu"der hot donn

21 W l

ägerufe wie se net getraut hat 21+wie sie sich verhalten soll+12 nit

K

+ Standard +

22 W l

2s als leiderin weeß die net emol wie se sisch verhalde soll s2 s3 0 der

23 W l

hot mer des gsacht nit (zu M) here se mol zu isch sag ihne ja 5s5

24 W l

un no sacht=a 21 das hätt sie nischt machen dürfen 0 sie hätt sie

5,1 W l

reinbitten können un dann" hot er gsacht 3s ist das n"ischt gestattet

2 Wl

s3 0 isch hab gsacht frau hansen machen sie 3s sofort die tagesstätte

3 Wl

auf un lassen die leute rein s3 0 nit 0 das geht ja nicht 12

4 Wl

un wenn die äh deutschamerikaner was machen wollen dann haben

5 Wl

sie des bei mir" vorher anzumelden und dann wird des geregelt

6 W1

und dann wird die 0 tages- äh tagesstätte geschlossen un do könne

257

The Symbolization of Social Identity

7 W1 se ( ^ J ja sie 0 un no haww»isch 8 W24 u 9 W8 grad wollt isch saache 10 W1

gsad hern se mol herr baumann heeßt der glaub isch in dem

11 W1

grüne heftel steht ... baumann baunmann nit 0 21 un dann sacht=a 12

12 W1

no haw isch gsad ja wisse sie was isch geh do nimmer hin 0

13 W1 14 W8

un no seschtsa ach" des würd isch net mache net

15 W1

haww=isch gsacht 0 des würd isch net mache gehe se na hie un

16 W1

sachere bscheid 0 no haw isch gsacht 0 äh mhm 0 no seschta 0

17 W1

noja 0 isch werd se nochemol zur red stelle 0 hoffentlisch hat

18 W1

sie e paar gute ausre- reden zur hand 0 schunsch scheiß isch

19 20 21 22 23

jetz erseht rescht

W1 sie zomme hot der mir gsacht un do haww=isch gsad 2s au"srede ach WI W24 oh WE oh W8 och

24 W l

um ausrede iss die nicht verlege haww=isch gsacht

25 W l K

liewer gott Gelächter

K s2 soviel ausredde hot kän mensch uff de weit wie die

Lachen

Bibliography Ideologie und ideologische Staatsapparate. In: L . Althusserl: Marxismus und Ideologie. Berlin, 1 3 7 - 1 5 7 . A L Z H A U S , H . P . / H E N N E , H . / W I E G A N D , H . E . (1973). Lexikon der Germanistischen Linguistik, Vol. II. Tübingen, 333. A M M O N , U . ( 1 9 8 3 ) . "Explicit-Definition von 'Dialekt' und benachbarten Begriffen mit Mitteln der formalen Logik." In: K. MATTHEIER (ed.): Aspekte der Dialekttheorie. Tuebingen, 2 7 - 6 8 . APELTAUER, E. (1977). Elemente und Verlaufsformen von Streitgespraechen. Eine Analyse von Texten und Tonbandprotokollen unter sprechhandlungstheoretischen Gesichtspunkten. Ph.D. Muenster. ATKINSON, J.M./J. HERITAGE (1984). Structures of Social Action: Studies in Conversation Analysis. Cambridge. ATKINSON, J.M./P. D R E W (1979). Order in Court. London. A U B E R T , V. (1972). "Interessenkonflikt und Wertekonflikt: Zwei Typen des Konflikts und der Konfliktloesung." In: W.L. Buehl (ed.): Konflikt und Konfliktstrategie. Ansaetze zu einer soziologischen Konflikttheorie. Muenchen, 178-205 A U E R , J. C. P. (1981). Einige konversationsanalytische Aspekte der Organisation von „Code-switching" unter italienischen Immigrantenkindern. In Revue de Phonétique Appliquée, 58: 126—148. A U E R , J.C.P./A. DI LUZIO (1983). "Structure and Meaning of Linguistic Variation in Italian Migrant Children in Germany." In: R. Baeuerle/R. Schwarze/C. von Stechow (eds.): Meaning, Use and Interpretation of Language. Berlin, 1-21. A U E R , PETER ( 1 9 8 7 ) . Phonologie des Konstanzer Stadtdialekts. Universität Konstanz. Unveröfftl. Ms., Konstanz. A U E R , P . / A . DI LUZIO (eds.) (1984). Interpretative Sociolinguistics. Migrants—Children—Migrant Children. Tuebingen. A U E R , P . / A . MOESSLE (in press). " A Case of Convergence and its Interpretation: MHG Î and û in the City Dialect of Konstanz. In: Auer/di Luzio (eds.) Convergence and Variation. Berlin/New York. A U S T I N , J. ( 1 9 6 2 ) . How to Do Things with Words. The William James Lectures, 1955, Oxford. ALTHUSSERL, L . ( 1 9 7 3 ) .

260

Bibliography

M. (1982). Sprachgebrauch in Abhaengigkeit von Merkmalen der Sprecher und der Sprechsituation. Eine soziolinguistische Untersuchung. Berlin. B A C H , A D O L F (1970). Geschichte der deutschen Sprache. Heidelberg. B A K E R , O . R . ( 1 9 8 0 ) . "Categories of Code Switching in Hispanic Communities: Untangling the Terminology." In: Working Papers in Sociolinguistics Series 76, Austin (Texas)/Washington. AUWAERTER,

M. (1977). Le marxisme et la philosophie du langage. Essai d'application de la méthode sociolinguistique en linguistique. Paris. B A U S C H , K A R L - H E I N Z (Hrsg.) ( 1 9 8 2 ) . Mehrsprachigkeit in der Stadtregion. Jahrbuch 1981 des Instituts fuer deutsche Sprache. Duesseldorf 1982 (Sprache der Gegenwart 56). B A U S I N G E R , H. (1972). Dialekte, Sprachbarrieren, Sondersprachen. Frankfurt (M.). B E C K E R , S./N. D I T T M A R / W . K L E I N ( 1 9 7 8 ) . "Sprachliche und soziale Determinanten im kommunikativen Verhalten auslaendischer Arbeiter." In: U. Quasthoff (ed.): Sprachstruktur - Sozialstruktur. Zur linguistischen Theoriebildung. Koenigstein/Ts., 1 5 8 - 1 9 2 . BAKHTINE,

A. ( 1 9 8 4 ) . "Language Style as Audience Design." In: Language in Society 1 3 , 1 4 5 - 2 0 4 . B E L L , R . T . ( 1 9 7 6 ) . Sociolinguistics. Goals, Approaches, and Problems. London. BELL,

C H . R . / J . J . B R A D A C ( 1 9 8 2 ) . Language and Social Knowledge. Uncertainty in Interpersonal Relations. London. BERGER, PETER L./THOMAS LUCKMANN ( 1 9 6 9 ) . Die gesellschaftliche Konstruktion der Wirklichkeit. Eine Theorie der Wissenssoziologie. Frankfurt (M). BERGER,

(1981). "Modalities, and the Process of Cultural Reproduction: A Model." In: Language in Society 10, 327-363. B E R R U T O , G. (1984). "Italian Contributions to the Sociolinguistic Theory." In: Linguistische Berichte 90, 58-70. BERNSTEIN, B .

BESCH, W . / J . HUFSCHMIDT/A. KALL-HOLLAND/E. K L E I N / K . J .

MATTHEIER

(1981). Sprachverhalten in laendlichen Gemeinden. Ansaetze zur Theorie und Methode. Forschungsbericht Erp-Projekt I. Berlin. BESCH, WERNER/

KNOOP, ULRICH/ PUSCHKE, W./WIEGAND,

HERBERT

E.

(Hrsg.) (1983). Dialektologie. Ein Handbuch zur deutschen und allgemeinen Dialektforschung, Berlin. W./K. M A T T H E I E R (ed.) (1985). Ortssprachenforschung. Beitraege zu einem Bonner Kolloquium. Berlin.

BESCH,

261

Bibliography

BIERWISCH, M. (1975). "Variation im Sprachsystem." In: Linguistische Studien. Akademie der Wissenschaften der D D R . Zentralinstitut fuer Sprachwissenschaften. Reihe: Arbeitsberichte Berlin 19, 65-137. BIERWISCH, M. (1976). "Social Differentiation of Language Structure." In: A. Kasher (ed.): Language in Focus: Foundations, Methods and Systems. Dodrecht, 407-456. B L O C H , E. ( 1 9 6 2 ) . Erbschaft dieser Zeit. Frankfurt ( M . ) . B L O M , J A N P E T E R / G U M P E R Z , J O H N J . ( 1 9 7 2 ) . "Social meaning in linguistic structure: Code switching in Norway", in: Gumperz/Hymes 1972, S. 407-434.

Wo sind Grossstaedter daheim? Studien ueber Bindungen an das Wohnviertel. Wien. BOISSEVAIN, J . (1970). The Italians in Montreal. Social Adjustment to a Plural Society. Ottawa. BODZENTA, E R I C H / S P E I S E R ,

I./THUM, K .

(1981).

"The Economics of Linguistic Exchanges." In: Social Science Information 1 6 ( 6 ) , 6 4 5 - 6 6 8 . BOURDIEU, P. ( 1 9 7 9 ) . La distinction. Critique social du judgement. Paris. B O U R D I E U , P. ( 1 9 8 2 ) . Ce que parler veut dire. L'économie des échanges linguistiques. Paris. BOURDIEU, P. ( 1 9 7 7 ) .

Die feinen Unterschiede. Frankfurt (M.). "Oekonomisches Kapitel, kulturelles Kapitel, soziales Kapital." In: Kreckel, B. (ed.): Soziale Ungleichheiten. Goettingen, 183-

B O U R D I E U , P. ( 1 9 8 2 ) . B O U R D I E U , P. ( 1 9 8 3 ) . 198.

BOURDIEU, P. (1984). "Capital et marche linguistiques." In: Linguistische Berichte 90, 3-24. R. Y . & G E N E S E E , F . ( 1 9 8 0 ) . Evaluative Reaction to Code Switching Strategies in Montreal. In Giles, Robinson, Smythe (Eds.) Language. Social Psychological Perspectives, Oxford, Pergamon Press, 335-343.

BOURHIS,

Die Mannheimer Mundart. Heidelberg. L.B. (1983). "Levels of Analysis in Sociolinguistic Explanation: Bilingual Code Switching, Social Relations, and Domain Theory." In: International Journal of the Sociology of Language 39, 5-43.

BRAEUTIGAM, K U R T ( 1 9 3 4 ) . BREITBORDE,

(Hrsg.) ( 1 9 8 6 ) . Sprachwissenschaft und Volkskunde. Perspektiven einer kulturanalytischen Sprachbetrachtung. Opladen.

BREKLE, HERBERT E./MAAS, UTZ

D./M. G E R R I T S E N / D . DE HANN ( 1 9 7 9 ) . "Speech Differences Between Women and Men: On the Wrong Track?" In: Language in Society 8(1): 33-50.

BROUWER,

262

Bibliography

"Universals in Language Use: Politeness Phenomena." In: E. Goody (ed.): Questions and Politeness. Cambridge,

BROWN, P . / S . LEVINSON ( 1 9 7 8 ) .

56-290.

P./S. L E V I N S O N (1987). Introduction, in: Politeness. Cambridge, (a re-issue of Brown & Levinson 1978).

BROWN,

P./S. L E V I N S O N (in preparation). Cross-Cultural Communication in Cape York. Report to Australian Institute of Aboriginal Studies. B U E H L , W.L. (1972). "Entwicklungslinien der Konfliktsoziologie." In: W.L. Buehl (ed.): Konflikt und Konfliktstrategie. Ansaetze zu einer soziologischen Konfliktstrategie. Muenchen, 9-64. C E D E R G R E N , H . C . / D . S A N K O F F (1974). "Variable Rules: Performance as a Statistical Reflection of Competence." In: Language 50 (2), 333-355. BROWN,

U./S. R O M A I N E (1984). "Evaluative Reactions to Panjabi/English Code-Switching." In: Journal of Multilingual and Multicultural Development 5/6, 447-474.

CHANA,

J . ( 1 9 8 2 ) . Variation in an English Dialect. Study. Cambridge.

CHESHIRE,

A

Sociolinguistic

Psychology and Language. An Introduction to Psycholinguistics. New York/Chicago/San Francisco/Atlanta. C L A R K , H./T. C A R L S O N (1982). "Hearers and Speech Acts." In: Language 58 (2), 332-373. C L É M E N T , R./Y. B E A U R E G A R D (1986). Peur d'assimilation et confiance en soi: leur relation à l'alternance des codes et à la compétence communicative en langue seconde. In: Canadian Journal of Behavioral Sciences.

CLARK, H.H./E.V. CLARK (1977).

G. (1984). Language and society in the German-speaking countries. Cambridge.

CLYNE, MICHAEL

P. (Hrsg.) (1982). Belonging. Identity and Social Organization in British Rural Cultures. Manchester.

COHEN, ANTHONY

Theorie sozialer Konflikte. Neuwied/Berlin. F. (1981). Routine im Gespräch. Zur pragmatischen Fundierung der Idiomatik. Wiesbaden.

COSER, L . A . ( 1 9 7 2 ) . COULMAS,

M. (1969). "A Discussion of Elaborated and Restricted Codes." In: Educational Review 22 (1): 38-50.

COULTHARD,

Investigating English Style. London. "Implicational Scales and Sociolinguistic Linearity."

CRYSTAL, D . / D . D A V Y ( 1 9 6 9 ) . D E CAMP, D . ( 1 9 7 1 ) .

In: Linguistics

73,

30-43.

"Zwischen Mundart und Hochsprache. Ein Beitrag zur Stadtsprache- Stadtmundart und Umgangssprache", in: Zeitschrift fuer Mundartforschung 29, 1962, S. 1-43.

DEBUS,

FRIEDHELM

(1962).

263

Bibliography

D . / J . F . H A M M E R S / N . L A B R I E (1984). "La relation entre les réseaux et le comportement langagier d'adultes en milieu de contact des langues: une étude pilote." Paper presented at the 7 th Congress of the I.A.A.L., August 5 — 10, Brussels.

DESHAYES,

DI

SCIULLO,

A. I A.

VAN A M E R I N G E N / H .

CEDERGREN/P.

PUPIER

(1976).

"Etude de l'interaction verbale chez les Montréalais d'origine italienne." In: Cahiers de Linguistique 6: 127-153. D I T T M A R , N. (1976). Sociolinguistics: A Critical Survey of Theory and Application. London. D I T T M A R , N. (1980). "Ordering Adult Learners According to Language Abilities." In: S. Felix (ed.): Second Language Acquisition. Tuebingen, 205-231.

N. (1982). "Soziolinguistik: Theorie, Methodik und Empirie ihrer Forschungsrichtungen." In: Studium Linguistik 12, 20-52. D I T T M A R , N. (1983). "Descriptive and Explanatory Power of Rules in Sociolinguistics." In: B. Bain (ed.): The Sociogenesis of Language. New York/London, 225-255. D I T T M A R , N. (1987a). Soziolinguistik: Grundlagen, ausgewählte Ergebnisse, Perspektiven, unpublished ms., to appear: de Gruyter, Berlin. D I T T M A R , N. (1987b). Variatio delectat. Sociolinguistica e linguaggi in contatto. Lecce. D I T T M A R , N./B. S C H L I E B E N - L A N G E (1982). "Stadtsprache. Forschungsrichtungen und -perspektiven einer vernachlaessigten soziolinguistischen Disziplin." In: K.-H. Bausch (ed.): Mehrsprachigkeit in der Stadtregion. Mannheim, 9-86. D I T T M A R , N./B. S C H L I E B E N - L A N G E (eds.) (1982a). La Sociolinguistique dans les pays de langue romane. Tuebingen. D I T T M A R , N./P. SCHLOBINSKI (1985). "Die Bedeutung von sozialen Netzwerken für die Erforschung von Ortssprachen." In: W. Besch/K.J. Mattheier (eds.): Ortssprachenforschung. Beitraege zu einem Bonner Kolloquium. Berlin, 158-188. D I T T M A R , N./P. S C H L O B I N S K I (in press). "Convergence, Discourse and Variation." In: C.J.P. Auer/A. di Luzio (eds.): Convergence and Variation. Berlin/New York. D I T T M A R , N./P. S C H L O B I N S K I / I . W A C H S (1986). Berlinisch. Studien zum Lexikon, zur Spracheinstellung und zum Stilrepertoire. Berlin. D I T T M A R , N./E. T H I E L I C K E (1979). "Der Niederschlag von Erfahrungen auslaendischer Arbeiter mit dem institutionellen Kontext des Arbeitsplatzes in Erzaehlungen." In: G. Soeffner (ed.): Interpretative Verfahren in den Text- und Sozialwissenschaften. Stuttgart, 65-103. DITTMAR,

264

Bibliography

Language and Society. London. "Sociophonological Methods in the Study of Sociolinguistic Variation in Viennese German." In: Language in

DOWNES, W . ( 1 9 8 4 ) .

DRESSLER, W . / R . W O D A K ( 1 9 8 2 ) .

Society 11:3, 339-70.

et. al. ( 1 9 7 4 ) . Allgemeine Rhetorik. Muenchen. DUDEN. Das große Wörterbuch der deutschen Sprache. Mannheim 19761981 E H L I C H , K . / J . REHBEIN ( 1 9 7 2 ) . Zur Konstitution pragmatischer Einheiten in einer Institution: Das Speiserestaurant. In D. Wunderlich, (ed.) Linguistische Pragmatik. Frankfurt/M. ENCREVÉ, P. (1982). "A propos du 'marché' linguistique'. In: N. Dittmar/ B. Schlieben-Lange (eds.): Die Soziolinguistik in romanischsprachigen Laendern. La Sociolinguistique dans les pays de langue romane. Tuebingen, 97-103. ENGEL, U L R I C H ( 1 9 6 2 ) . "Schwaebische Mundart und Umgangssprache", DUBOIS, J .

in: M u t t e r s p r a c h e 72 (1962) S. 257-261.

(1973). Linguistic Stylistic. The Hague. The Counselor as Gatekeeper: Social Interaction in Interviews. New York. E R M A N , B. (1986). Some Pragmatic Expressions in English Conversation. In: Tottie, G./Bäcklund, J. (eds.) English in Speech and Writing. A Symposium. Stockholm, 131 — 148. E R V I N - T R I P P , S. ( 1 9 7 2 ) . "On Sociolinguistic Rules: Alternation and Cooccurrence in Sociolinguistics." In: Gumperz/Hymes (1972), 213 — 250. ESSER, P. (1983). Dialekt und Identitaet. Diglottale Sozialisation und Identitaetsbildung. Frankfurt (M.)/Bern. FASOLD, R. ( 1 9 8 4 ) . Linguistic Analysis of the Three Kinds. NWAVE 1 9 8 4 , Montréal, unpublished ms. FERGUSON, CH. (1959). "Diglossia." In: Word 15 (2): 325-340. FILLMORE, C H . ( 1 9 6 8 ) . "The Case for Case." In: E . Bach/R.T. Harms (eds.): Universals in Linguistic Theory. New York, 1-90. FISHMAN, J. (1968). "Sociolinguistic Perspectives on the Study of Bilingualism." In: Linguistics 39, 21-49. FISHMAN, J . ( 1 9 7 1 ) . Sociolinguistics. A Brief Introduction. Massachusetts. FISHMAN, J.A. (1975). Soziologie der Sprache. Eine interdisziplinaere sozialwissenschaftliche Betrachtung der Sprache in der Gesellschaft. Muenchen. ENKVIST, N . E .

ERICKSON, FREDERIC/SHULTZ, J . ( 1 9 8 2 ) .

FISHMAN, JOSHUA A./COOPER, ROBERT L . / M A . R O X A N A e t a l . ( 1 9 6 8 ) .

Bi-

lingualism in the barrio. The measurement and description of language dominance in bilinguals. Washington D.C.

Bibliography

265

D. (1979). Grammatik und Konversation. Stilistische Pragmatik des Dialogs und die Bedeutung deutscher Modalpartikeln. Tuebingen. F R A N K E N B E R G , H . ( 1 9 7 6 ) . Vorwerfen und Rechtfertigen als verbale Teilstrategien der innerfamiliaeren Interaktion. Ph.D. Duesseldorf. F U C H S , W . et. al. ( 1 9 7 8 ) . Lexikon zur Soziologie. Opladen. F U E H R E R , B. (1982). Das Berlinische im Tagesschrifttum von 1848/49: Studien zum Verhaeltnis von Idiolekt, Soziolekt und Dialekt. Frankfurt (M.)/Bern. G A N S , H E R B E R T , J . ( 1 9 6 2 ) . The Urban Villagers. Group and Class in the Life of Italian-Americans. Glencoe. G A R D N E R - C H L O R O S , P. (1983). "Code-switching: approches principales et perspectives." In: La linguistique 19, 21-53. G I L E S , H . ( 1 9 7 3 ) . "Accent Mobility: A Model and Some Data." In: Anthropological Linguistics 1 5 (2): 8 7 - 1 0 5 G I L E S , H./P.M. S M I T H (1979). "Accommodation Theory: Optimal Level of Convergence." In: Giles, H./R. St. Clair (eds.): Language and. Social Psychology. Oxford, 45-65 G I L E S , H./R. S T . C L A I R (eds.) (1979). Language and Social Psychology. Oxford. G O F F M A N , E . (1971). Interaktionsrituale. Ueber Verhalten in direkter Kommunikation. Frankfurt (M.). G O F F M A N , E . (1974a). Frame Analysis. New York. G O F F M A N , E . (1974b). Das Individuum im oeffentlichen Austausch. Mikrostudien zur oeffentlichen Ordnung. Frankfurt (M.). G O O D E N O U G H , W. (1967). Rethinking Status and Role. In: S. Tyler (ed.) Cognitive Anthropology, New York. G O O D W I N , M. ( 1 9 8 2 ) . "Processes of Dispute Management Among Urban Black Children." In: American Ethnologist 9 , 7 6 - 9 6 . G O O D W I N , M . (in press). "Children's Arguing". In: S. Philips et. al. (eds.): Sex Differences in Language. Cambridge. G R A S S I , C . ( 1 9 8 2 ) . "Ville et campagne dans la socio linguistique italienne." In: N . Dittmar/B. Schlieben-Lange (eds.) ( 1 9 8 2 ) , 1 4 3 - 1 5 2 . G R E E N , J U D I T H L . / W A L L A T , C Y N T H I A ( 1 9 8 1 ) . Ethnography and Language in Educational Settings. Norwood, N . J . (Advances in Discourse Processes, 5). G R U N D Z Ü G E E I N E R D E U T S C H E N G R A M M A T I K (1981). Ed. by K . H. Heidolph, W. Fläming & W. Mötsch, Berlin (East). G U E L I C H , E. (1976). "Ansaetze zu einer kommunikationsorientierten Erzaehltextanalyse." In: W. Haubrichs (ed.): Erzaehlforschung 1. Goettingen, 224-256. FRANCK,

266

Bibliography

J.J.(1964). "Linguistic and Social Interaction in Two Communities." In: American Anthropologist 66 (6): 137-154. G U M P E R Z , J . J . (1968). "The Speech Community." In: International Encyclopedia of the Social Sciences. Macmillan, 391-6, Reprinted in P. Giglioli (ed.) (1972): Language and Social Context. Harmondsworth, 219-231. G U M P E R Z , J . J . ( 1 9 7 1 ) . "On the Ethnology of Linguistic Change." In: W . Bright (ed.): Sociolinguistics. Proceedings of the UCLA Sociolinguistic Conference, 1 9 6 4 . The Hague, 2 7 - 3 8 . G U M P E R Z , J . J . (1976). "The Sociolinguistic Significance of Conversational Code-Switching." In: J . Cook-Gumperz/J.J. Gumperz (eds.): Working Paper 46, Language Research Laboratory, Berkeley. G U M P E R Z , J O H N J . (1982a). Discourse Strategies. Cambridge (Studies in interactional sociolinguistics 1). G U M P E R Z , J O H N J . (Hrsg.) (1982b). Language and Social Identity. Cambridge (Studies in interactional sociolinguistic 2). G U M P E R Z , J O H N J . ( 1 9 8 4 ) . "Ethnography in Urban Communication", in Auer/di Luzio 1984, S. 1-12. G U M P E R Z , J . J . / J . C O O K - G U M P E R Z (1982). "Introduction into Language and the Communication of Social Identity." In: J . Gumperz (ed.): Language and Social Identity. Cambridge, 1-21. G U M P E R Z , J . J . / D . H Y M E S (eds.) ( 1 9 7 2 ) . Directions in Sociolinguistics. New York. H A B E L , C. (1979). Aspekte bewertender Grammatiken. Berlin. H A B E R M A S , J . (1981a,b). Theorie des kommunikativen Handelns, Vol. I, II. Frankfurt (M.). H A L L I D A Y , M . A . K . ( 1 9 7 8 ) . Language as Social Semiotic. The Social Interpretation of Language and Meaning. London. H A L L I D A Y , M . / A . M C I N T O S H / P . S T R E V E N S ( 1 9 6 4 ) . The Linguistic Sciences and Language Teaching. London. GUMPERZ,

K . ( 1 9 7 9 ) . Alltagssprachliche Diskussion: Struktur - Verlauf Techniken. Ph.D. Berlin.

HALM,

HAMERS,

J.E/M.

HAMERS,

J.F./D.

(1984). Bilingualité et bilingualisme. Brussels. L A B R I E (1984). "Aspects socio-psychologiques du compartement langagier en fonction des réseaux sociaux chez des adolescents en milieu de contact de langages." Paper presented at the 7 th Congress of the I.A.A.L., August 5 — 10, Brussles. BLANC

DESHAIES/N.

G. (1966). Zur Mundartgeographie. Ergebnisse, Methoden, Perspektiven. Duesseldorf.

HARD,

267

Bibliography

W. (1981). "Differenziertheit der Sprache als Ausdruck ihrer Gesellschaftlichkeit." In: Autorenkollektiv: Kommunikation und Sprachvariation, Berlin (DDR), 26-72. H Ä R T U N G , W. (ed.) (1977). Normen in der sprachlichen Kommunikation. Sprache und Gesellschaft Vol. II. Berlin (GDR). H E A T H , S.B. (1983). Ways With Words. Language, Life, and Work in Communities and Classrooms. Cambridge. H E F N E R , W E R N E R / U R E L A N D , S T U R E ( 1 9 8 0 ) . "Areale und soziolinguistische Variation: Die p-jpf-Isoglosse im Raum Rhein-Neckar-Main", in: Ureland, Sture (Hrsg.), Sprachvariation und Sprachwandel. Tuebingen 1980, HÄRTUNG,

S.

51-94.

HENNE,

H./H.

REHBOCK

(1979).

Einführung in die Gespraechsanalyse.

Berlin/New York. J.C. ( 1 9 8 5 ) . "Recent Developments in Conversational Analysis." In: Sociolinguistics Newsletter X V (1), 1-19. H O F F M A N N , W A L T E R / M A T T H E I E R , K L A U S ( 1 9 8 6 ) . "Stadt und Sprache in der neueren deutschen Sprachgeschichte: eine Pilotstudie am Beispiel von Koeln". in: Sprachgeschichte. Ein Handbuch zur Geschichte der deutschen Sprache und ihrer Erforschung. Band 2.2 Berlin/New York, HERITAGE,

1837-65

W. (1976). "Selbst- und Partnereinschaetzung in Gespraechen." In: H. Weber/H. Weydt (eds.): Sprachtheorie und Pragmatik. Akten des 10. Linguistischen Kolloquiums Tuebingen 1975, Vol. I. Tuebingen, 175-

HOLLY,

186.

R. VAN (1981). "Word Final t, d Deletion in a Dutch Urban Dialect and Some Problems in Variable Rule Analysis." In: S. Daalder/M. Gerritsen: Linguistics in the Netherlands. Amsterdam, 23-35.

HOUT,

HUDSON, R . A . ( 1 9 8 0 ) .

Sociolinguistics. Cambridge.

HUFSCHMIDT,

KLEIN/K.J.

J./E.

MATTHEIER/H.

MICKARTZ

(1983).

Sprachverhalten in laendlichen Gemeinden. Dialekt und Standardsprache im Sprecherurteil. Forschungsbericht Erp-Projekt Berlin II. Berlin. HUFSCHMIDT,

J./K.J.

"Sprache und Gesellschaft. integrierenden Beschreibung." In: J.

MATTHEIER

(1981).

Ueberlegungen zu einer Hufschmidt et al. 1 9 8 3 , 1 1 7 - 2 0 0 .

D. (1972). "Models of the Interaction of Language and Social Life." In: J . J . Gumperz/D. Hymes (1972), 35-71. H Y M E S , D. (1974). Foundations in Sociolinguistics. An Ethnographic Approach. Philadelphia. International Journal of the Sociology of Language 39 (1983). HYMES,

268

Bibliography

J . (1974). "Strategies of Status Manipulation in the Wolof Greeting." In: B. Bauman/J. Sherzer (eds.): Explorations in the Ethnography of Speaking. Cambridge, 167-191.

IRVINE,

KALLMEYER, W E R N E R / K E I M , INKEN/NIKITOPOULOS, PANTELIS ( 1 9 8 2 ) .

Zum

Projekt ,Kommunikation in der Stadt', in: Bausch, Karl-Heinz (Hrsg.), Mehrsprachigkeit in der Stadtregion. Jahrbuch 1981 des Instituts fuer deutsche Sprache. Duesseldorf, S. 3 4 5 - 3 9 0 (Sprache der Gegenwart 5 6 ) . K A L L M E Y E R , W E R N E R / K E I M , I N K E N (1986a). "Formulierungsweise, Kontextualisierung und soziale Identitaet. Dargestellt am Beispiel des formelhaften Sprechens", in: LiLi (in press). K A L L M E Y E R , W E R N E R / K E I M , I N K E N (1986b). Talking About the World. Formulation, Universe of Discourse, Group and Social World. Paper, Mannheim. K A N N G I E S S E R , S. (1972). Aspekte der synchronen und diachronen Linguistik. Tuebingen. K A N N G I E S S E R , S. ( 1 9 7 8 ) . "Modalitaeten des Sprachprozesses I . " In: C . Habel/S. Kanngiesser (eds.): Sprachdynamik und Sprachstruktur. Tuebingen, 8 1 - 1 3 9 . K A T Z , J . ( 1 9 8 1 ) . The Modularity of Mind. Cambridge, Massachussets. K E I M , I N K E N ( 1 9 8 4 ) . "Talking about Foreigners - Some Ethnographie Remarks", in: Auer, Peter/di Luzio, Aldo (Hrsg.): Interpretative sociolinguistics. Migrants - Children - Migrant Children. Tuebingen, S. 259283.

W. (1974). Variation in der Sprache. Kronberg/Ts. W. (1980). "Argument und Argumentation." In: Zeitschrift fuer Literaturwissenschaft und Linguistik 38/39, 9-57. K L E I N , W. (1983). "Wie Kinder miteinander streiten. Zum sprachlichen Verhalten von Grundschulkindern in Konfliktsituationen." In: D. Boueke/W. Klein (eds.): Untersuchungen zur Dialogfaehigkeit von Kindern. Tuebingen. K L E I N , W. (1984). Zweitspracherwerb. Eine Einfuehrung. Koenigstein/ Ts. K L E I N , W./N. D I T T M A R (1979). Developing Grammars. The Acquisition of German by Foreign Workers. Heidelberg/New York. K L O S S , H. (1978). Die Entwicklung neuer germanischer Kultursprachen. Duesseldorf. K O H Z , A. (1984). "Das System der Anredeformen im Koreanischen." In: W. Winter (ed.): Anredeverhalten. Tuebingen, 293-325. K R A P P M A N N , L. (1971). Soziologische Dimensionen der Identitaet. Strukturelle Bedingung fuer die Teilnahme an Interaktionsprozessen. Stuttgart.

KLEIN,

KLEIN,

Bibliography

269

KRECKEL, REINHARD (ed.) (1983). Soziale Ungleichheiten. Göttingen. KRUSE, D. (1986). Glassbrenner und der Berliner Dialekt. Berlin. ( = Schriften zur Berliner Sprache und Literatur I). KUEHN, P. (1982). "Typen lexikographischer Ergebnisdarstellung." In: W. Besch et al. (eds.): Dialektologie. Ein Handbuch zur allgemeinen Dialektforschung, Vol.I. Berlin/New York, 702-723. LABOV, W. (1966). The Social Stratification of English in New York City. Arlington: Center for Applied Linguistics. LABOV, W. (1972a). Sociolinguistic Patterns. Pennsylvania. LABOV, W. (1972b). Language in the Inner City. Philadelphia. LABOV, W. (1973). " T h e Boundaries of Words and Their Meaning." In: C.J. Bailey/R.W. Shuy (eds.): New Ways of Analyzing Variation in English. Washington D.C., 340-373. LABOV, W. (1980). Sprache im sozialen Kontext. Ed. Norbert Dittmar/ Bert-Olaf Rieck, Königstein/Ts. LABOV, W./D. FANSHEL (1977). Therapeutic Discourse. Psychotherapy as Conversation. New York. LABRIE, N. (1984). La vitalité ethnolinguistique et les characteristiques socio-psycholinguistiques de l'individue vivant en milieu bilingue. Québec, International Center for Research on Bilingualism, Publication B133. LABRIE, N./R. CLÉMENT (1986). Ethnolinguistic Vitality, Self-Confidence and Second Language Proficiency: An Investigation. In: Journal o f Multilingual and Multicultural Development 7 (4), 269 — 282. LAKOFF, R. (1975). Language and Woman's Place. New York. LAKS, B. (1983). "Langage et pratiques sociales." In: Actes de la recherche en science sociales 46, 73-97. LAMANDELLA, J . T . (1977). " T h e Lymbic System in Human communication." In: H. Whitaker/H.A: Whitaker (eds.): Studies in Neurolinguistics, Vol 3, 157-222. LASCH, A. (1928). Berlinisch. Eine berlinische Sprachgeschichte. Berlin. LAUMANN, E.O. (1973). Bonds o f Pluralism: The Form and Substance o f Urban Social Networks. New York/London/Sydney/Toronto. LAVANDERA, B. (1978). "Where Does the Sociolinguistic Variable Stop?" In: Language in Society 7, 171-182. LEPAGE, R.B. (1968). "Problems o f Description in Multilingual Communities." In: Transactions of the Philological Society, 189-212. LEECH, G. (1983). Principles of Pragmatics. London. LEVINSON, S. (1982). Explicating Concepts o f Participant Role. Unpublished paper presented to the International Conference on the Sociology o f Language, Plymouth, July 1982.

270

Bibliography

S. (1983). Pragmatics. Cambridge. S. (in press). Putting Linguistics on a Proper Footing: Explorations in Goffman's Concepts of Participant Role. In P. Drew/T. Wooton (eds.) Goffman: An Interdisciplinary Appreciation. Oxford. L U C K M A N N , THOMAS ( 1 9 8 5 ) . Kommunikative Gattungen im 'Haushalt' einer Gesellschaft. Ms. Konstanz. LYONS, J. (1970). New Horizons in Linguistics. Harmondsworth. LYONS, J . (1977). Semantics (2. Vol.). Cambridge. M A A S , U T Z ( 1 9 8 6 ) . "Volkskundliches (Kultur) in der Sprachwissenschaft", in: Brekle/Maas 1 9 8 6 , S. 3 3 - 6 9 . M A L I N O W S K I , B. (1974). "Das Problem der Bedeutung in primitiven Sprachen." In: C.K. Odgen/A. Richards (eds.): Die Bedeutung der Bedeutung. Frankfurt (M.), 323-384. M A T T H E I E R , K . J . ( 1 9 7 9 ) . Sprachvariation und Sprachwandel. Untersuchungen zur Struktur und Entwicklung von Interferenzprozessen zwischen Dialekt und Hochsprache in einer laendlichen Sprachgemeinschaft des Rheinlandes. Habil., ms. MATTHEIER, K . J . (1980). Pragmatik und Soziologie der Dialekte. Einführung in die kommunikative Dialektologie des Deutschen. Heidelberg. M A T T H E I E R , K . J . (1981). "Chronologischer Ueberblick ueber Planung und Durchfuehrung der Datenerhebung." In: W. Besch et al. 1981, 16-42. M A T T H E I E R , K . J . ( 1 9 8 2 ) . "Datenerhebung und Forschungsziel." In: W . Besch et al. 1 9 8 2 , 6 2 2 - 6 3 9 . M A T T H E I E R , K.J. (1983a). "Sprachbewertungen im Kommunikationsprofil der Ortsgemeinschaft Erp." In: J. Hufschmidt et al. 1983, 265-280. M A T T H E I E R , K . J . (1983b). "Sprachlage und sprachliches Kontinuum. Auswertung des Tonbandtests zur Sprachlagendifferenzierung." In: J. Hufschmidt et al. 1983, 226-264. M A T T H E I E R , K.J. (1985). "Ortsloyalitaet als Steuerungsfaktor beim Sprachgebrauch in oertlichen Sprachgemeinschaften." In: W. Besch/K.J. Mattheier 1985, 149-157. M A T T H E I E R , K L A U S J . / B E S C H , W E R N E R (1985). "Ortssprachenforschung. Einleitende Ueberlegungen", in: Besch/Mattheier 1985, S. 9-23. M I C K A R T Z , H. (1983). "Einstellungsaeusserungen zur Verwendung von Hochsprache und Mundart in der Kindererziehung." In: J. Hufschmidt et al. 1983, 60-116. M I H M , A R E N D (Hrsg.) (1985a). Sprache an Rhein und Ruhr. Dialektologische und soziolinguistische Studien zur sprachlichen Situation im RheinLEVINSON, LEVINSON,

Bibliography

271

Ruhr-Gebiet und ihre Geschichte. Stuttgart 19 (Zeitschrift fuer Dialektologie und Linguistik, Beihefte, Heft 50). M I H M , A R E N D (1985b). "Sprachlandschaft Duisburg", in Mihm (1985a), S. 201-222. M I L R O Y , L. ( 1 9 8 0 ) . Language and Social Networks. Oxford. M I L R O Y , L./S. M A R G R A I N (1980). "Vernacular Language Loyalty and Social Network." In: Language in Society 9, 43-70. M U N S K E , H.H. ( 1 9 8 3 ) . "Umgangssprache als Sprachkontakterscheinung." In: W. Besch et al (eds.): Dialektologie. Ein Handbuch zur deutschen und allgemeinen Dialektforschung, 2. Halbband. Berlin/New York, 1002-1018. N E U L A N D , E. ( 1 9 8 1 ) . "Punkt zwoelf muss et Essn aufm Tisch stehn!' Analyse alltaeglicher Kommunikation in einer Arbeiterfamilie." In: Linguistische Berichte 7 6 , 6 5 - 8 9 . N E U M A N N , W . (ed.) ( 1 9 7 6 ) . Theoretische Probleme der Sprachwissenschaft. Berlin (GDR). O L Y N E K , M./D. S A N K O F F / A . D ' A N G L E J A N (1983). "Second Language Fluency and the Subjective Evaluation of Officer Cadets in a Military College." In: Studies in Second Language Acquisition 5 (2): 213-249. P A I N C H A U D , C./R. P O U L I N (1983). Italianité, conflict linguistique et structure du pouvoir dans la communité italo-québécoise." In: Sociologie et Sociétés X V (2): 89-104. C./R. P O U L I N (1983). Les Italiens du Québec. Histoire et sociologie d'une communauté, unpublished manuscript.. P O L E N Z , P. VON (1978). Geschichte der deutschen Sprache. Berlin. P O P L A C K , S. (1980). "Sometimes I'll Start a Sentence in English Y T E R MINO EN ESPANOL: Toward a Typology of Code-Switching." In: Linguistics 18, 581-618. P R O C A C C I , G. (1955). "Dal feudalesimo al capitalismo: una discussione storica." In: Societa', 123-138. P R U J I N E R , A. et al. (1984). Variation du comportement langagier lorsque deux langues en contact. Research Project. Québec, International Center for Research on Bilingualism. Publication G-5. PAINCHAUD,

U. ( 1 9 8 0 ) . Erzaehlen in Gespraechen. Linguistische Untersuchungen zu Strukturen und Funktionen am Beispiel einer Kommunikationsform des Alltags. Tuebingen. R E H B E I N , J . (1977). Komplexes Handeln. Elemente zur Handungstheorie der Sprache. Stuttgart. R E H B E I N , J . (1979). "Handlungstheorien." In: Studium Linguistik 7, 1-25. QUASTHOFF,

272

Bibliography

ROMAINE, S. (1980). "What is a Speech Community?". In: Belfast Working Papers in Language and Linguistics 4, 41-59: ROMAINE, S. (1984). " T h e Status of Sociological Models and Categories in Explaining Language Variation." In: Linguistische Berichte 90, 2538.

ROMAINE, S. (ed.) (1982). Sociolinguistic Variation in Speech Communities. London. M.T./A.A. SOBRERO/A. ZAMPOLLI (in press). "Per un atlante 'modulare': il N A D I R . " Proceedings of the Meeting on Atlanti regionali. Aspetti metodologici linguistici e etnografici, Palermo. RÜSSEL, J . (1982). "Networks and Sociolinguistic Variation in an African Urban Setting." In: S. Romaine (ed.): Sociolinguistic Variation in Speech Communities. London, 125.141.

ROMANELLO,

E. B . / H . G I L E S (eds.) ( 1 9 8 2 ) . Attitudes Towards Language Variation. London.

RYAN,

SCHEGLOFF ( 1 9 7 8 ) . " A Simplest Systematics for the Organization of Turn-Taking in Conversation." In: J . Schenkein (ed.): Studies in the Organization of Conversational Interaction. New York, 7-55.

SACKS, H . / J . J E F F E R S O N / E .

SANDIG, B. (1978). Stilistik. Sprachpragmatische Grundlegung der Stilbeschreibung. Berlin/New York. Formal Grammar for Code-Switching." In: Papers in Linguistics: International Journal o f Human Communication 14 (1), 3-45.

SANKOFF, D . / S . POPLACK ( 1 9 8 1 ) . " A

D . / G . SANKOFF (1973). "Sample Survey Methods and Computer Assisted Analysis in the Study of Grammatical Variation." In : R. Darnell (ed.): Canadian Languages in their Social Context. Edmonton, 7-64.

SANKOFF,

D./P. THIBAULT (1981). "Weak Complementarity: Tense and Aspect in Montreal French." In: B.B. Johns/D.R. Strong (eds.): Syntactic Change. Ann Arbor: University of Michigan Press, 205-216.

SANKOFF,

" T h e Preference for Selfcorrection in the Organization of Repair in Conversation." In: Language

SCHEGLOFF, E . / J . J E F F E R S O N / H . SACKS ( 1 9 7 7 ) . 53, 361-382.

C. (1982). „Versuch über das Aushandeln", in: W. L. Buhl (ed.): Konflikt und Konfliktstrategie. Ansätze zu einer soziologischen Konflikttheorie. München, p. 236 — 283.

SCHELLING, T H .

SCHENKEIN, J . (1978). Studies in the Organization of Conversational Interaction. New York.

Bibliography

273

D. ( 1 9 8 2 ) . Discourse Markers: Semantic Resource for the Construction of Conversation. Ph.D. University of Pennsylvania. SCHLIEBEN-LANGE, B. ( 1 9 7 5 ) . Linguistische Pragmatik. Stuttgart/Berlin/ Koeln/Mainz. SCHLIEBEN-LANGE, B. (1984). "La langue durée en sociolinguistique." In: Sociolinguistique des langues romanes. Actes du XVIIème Congrès International de Linguistique et Philologie romanes, Aix-en-Provence/ Marseille, 289-297. SCHLIEBEN-LANGE, B . / H . W E Y D T ( 1 9 7 8 ) . "Fuer eine Pragmatisierung der Dialektologie." In: Zeitschrift fuer Germanistische Linguistik 6, 257-

SCHIFFRIN,

282.

P. (1982). "Das Verkaufsgespraech. Eine empirische Untersuchung zu Handlungsschemata und kommunikativen Zielen." In: Linguistische Arbeiten, 18, Berlin 1-234. SCHLOBINSKI, P. (1986). "Code-switching im Berlinischen." To appear in N. Dittmar/P. Schlobinski (eds.): Wandlungen einer Stadtsprache — Berlinisch in Gegenwart und Vergangenheit. Berlin. SCHLOBINSKI, P. ( 1 9 8 7 ) . Stadtsprache Berlin. Eine soziolinguistische Untersuchung. Berlin/New York. SCHLOBINSKI, P./U. B L A N K (1985). Sprachbetrachtung: Berlinisch. Ein Schulbuch fuer den Deutschunterricht ab Klasse 10. Band I und II. Berlin. SCHUETZE, Y . ( 1 9 7 7 ) . Interfamiliaere Kommunikation und kindliche Psyche. Eine exemplarische Analyse der Kommunikation und Rollenstruktur zweier Familien. Berlin. SCHUPPENHAUER, CLAUS/WERLEN, I W A R ( 1 9 8 3 ) . "Stand und Tendenzen in der Domaenenverteilung zwischen Dialekt und deutscher StandardspraSCHLOBINSKI,

che", in: Besch/Knoop/Puschke/Wiegand 1983, S. 1411-1427.

J. (1969). Speech Acts. Cambridge. C.D.C. (1974). Invasion, Succession, and Conflict: The Case of StLéonard, Québec. Ph.D., Pennsylvania State University. SOBRERO, A . A . ( 1 9 7 8 ) . "Borgo, citta', territorio: alcuni problemi di metodo nella dialettologia urbana." In: Rivista Italiana di Dialettologia II (1), SEARLE, SMITH,

9-21.

"Modelli sociolinguistici e modelli storici." Quaderni storici 40, 80-104. SOBRERO, A.A. (in press). "Microconvergences in the Conversation Between Dialects and Language." In: C.J.P. Auer/A. di Luzio (eds.): Convergence and Variation. Berlin/New York. SOBRERO, A . A .

(1979).

274

Bibliography

SOBRERO, A.A./T. ROMANELLO (1981). L ' i t a l i a n o c o m e si p a r l a in S a l e n t o .

Lecce. SORNIG, K. (1981). "Lexical Innovation: A Study of Slang, Colloquialisms and Casual Speech." In: Pragmatics and Beyond II (5). Statistics Canada (1985). Language in Canada. Ottawa, Minister of Supply and Services. Catalogue 99-935. STOLT, B. (1979). "Ein Diskussionsbeitrag zu mal, eben, doch aus kontrastiver Sicht. (Deutsch-Schwedisch)." In: H. Weydt: Die Partikeln der deutschen Sprache. Berlin/New York, 479-487. SUTTLES, GERALD (1968). The social order of the slum. Ethnicity and territorry in the inner city. Chicago. SUTTLES, GERALD (1972). T h e social c o n s t r u c t i o n of c o m m u n i t i e s . C h i -

cago/London. TADDEO, D./R. TARAS (1982). The Language of Education Debate. A Study of the Political Dynamics between Quebec's Education Authorities and the Italian Community 1918-1982, unpublished manuscript. TANNEN, D. (1984). Conversational Style. Ablex. TARONE, E./U. FRAUENFELDER/L. SELINKER (1976). " S y s t e m a t i c i t y / V a r i a -

bility and Stability/Instability in Interlanguage Systems." In: H.D. Brown (ed.): Papers in Second Language Acquisition. Ann Arbor. TERRACINI, B.A. (1981). Linguistica al bivio. Napoli. THIBAULT, P. (1982). "Style, Sense, Function." In: N. Dittmar/B. SchliebenLange (eds.): Die Soziolinguistik in romanischsprachigen Laendern. La sociolinguistique dans les pays de langue romane. Tuebingen, 73-79. VEBLEN, T. (1981). Theorie der feinen Leute. Eine oekonomische Untersuchung der Institutionen. Muenchen. VINCENT, D. (1983). Les ponctuants de la langue. Ph.D. at University of Montreal. WAGNER, R.G. (1983). "Nicht jeder kann alles - zum Bildungssystem in der DDR", part III. Sender Freies Berlin I, January 21, 1983. WATZLAWICK, P./J.H. BEAVIN/D.D. JACKSON (1969). M e n s c h l i c h e K o m -

munikation: Formen, Stoerungen, Paradoxien. Bern/Stuttgart/Wien. WEINER, E.J./W. LABOV (1983). "Constraints on the Agentless Passive." In: Journal of Linguistics 19, 29-58. WHYTE, W.F. (1955). Street Corner Society. Chicago. WILLIS, PAUL (1978). Profane culture. London. Dt.: 'Profane Kultur'. Rocker, Hippies: Subversive Stile der Jugendkultur. Frankfurt 1981. WILPERT, G. VON (1969). Sachwoerterbuch der Literatur. Stuttgart. WIRTH, L. (1974). "Urbanität als Lebensform". In: U. Berlyn (ed.) Stadt und Sozialstruktur. Arbeiten zur sozialen Segregation, Ghettobildung und Stadtplanung. Dreizehn Aufsätze, München 42—66.

Bibliography

275

A. ( 1 9 7 5 ) . Dilemmas of Discourse. London. A. ( 1 9 8 1 ) . The Management of Grantings and Rejections by Parents in Request Sequences. In: Semiotica 37, 59 — 89. W U N D E R L I C H , D . ( 1 9 7 6 ) . Studien zur Sprechakttheorie. Frankfurt ( M . ) .

WOOTTON,

WOOTTON,

Contributors Norbert Dittmar Fachbereich Germanistik, Freie Universität Berlin Habelschwerdter Allee 45, 1000 Berlin 33, Germany. Wolfgang Klein Max-Planck-Institut für Psycholinguistik Berg en Dalse Weg 79, 6522 Nijmegen, Netherlands. Normand Labrie Centre International de Recherche sur le Bilinguisme, Université Laval Québec GIK 784, Canada. Stephen C. Levinson Linguistics Department, University of Cambridge Cambridge CB3 9DA, Great Britain. Klaus Mattheier Germanistisches Seminar, Universität Heidelberg Karlstraße 16, 6900 Heidelberg, Germany. Peter Schlobinski Fachbereich Germanistik, Freie Universität Berlin Habelschwerdter Allee 45, 1000 Berlin 33, Germany. Alberto Sobrero Dipartimento di Filologia Linguistica Letterature, Université 73100 Lecce, Italy. Pierette Thibault Département d'anthropologie, Université de Montréal C.P. 6128, Montréal H3C 3S7, Qué-Canada. Inge Wachs Fachbereich Germanistik, Freie Universität Berlin Habelschwerdter Allee 45, 1000 Berlin 33, Germany.

SOZIOLINGUISTIK UND SPRACHKONTAKT SOCIOLINGUISTICS AND LANGUAGE CONTACT CHRISTIANE VON STUTTERHEIM

Temporalität in der Zweitsprache Eine Untersuchung zum Erwerb des Deutschen durch türkische Gastarbeiter Groß-Oktav. XIV, 364 Seiten. 1986. Ganzleinen DM 1 5 6 , ISBN 3110106965 (Band 2) Sprachwissenschaftliche Untersuchung, die sich mit dem ungesteuerten Zweitsprachenerwerb beschäftigt. Analyse des Erwerbs von Ausdrucksmitteln zur Zeitreferenz und deren Gebrauch im Diskurs. Aus dem Inhalt: Temporale Konzepte, Temporalität im Diskurs. Temporale Referenz im Türkischen. Analysen der Lernsprachen, sprachliche Mittel zur temporalen Referenz, Erzählanalysen, Implizite Referenz, Diskursorganisationsprinzipien. PETER SCHLOBINSKI

Stadtsprache Berlin Eine soziolinguistische Untersuchung Groß-Oktav. XIV, 302 Seiten, diverse Abbildungen und Tabellen. 1987. Ganzleinen DM 1 3 6 , - ISBN 3 11 010914X (Band 3) Empirische Untersuchung zur inner- und außersprachlichen Variation des Berlinischen sowie zu Spracheinstellungen gegenüber dem Berlinischen. Aus dem Inhalt: Grundlegende Aspekte der Variationslinguistik Quantitative Analysen zur Sprachvariation Spracheinstellungsstudien Die soziale Bedeutung des Berlinischen Geteilte Stadt — geteilte Sprachgemeinschaft? Preisänderungen vorbehalten

Walter de Gruyter

w DE

G

Berlin • New York

SOZIOLINGUISTIK UND SPRACHKONTAKT SOCIOLINGUISTICS AND LANGUAGE CONTACT

Variation and Convergence Studies in Social Dialectology Edited by Peter Auer and Aldo di Luzio 1988. Large-octavo. VIII, 320 pages, various illustrations. Cloth DM 1 7 8 , - ISBN 3 11 011045 8 (Volume 4) A collection of papers at the interface between dialectology, sociolinguistics and conversational analysis in various speech communities. From the contents: The interpretive analysis of historical records MHG i and u in the city dialect of Constance Sociophonology The acquisition of Standard Swiss German by Swiss children Some properties of the dialect of Sicily Processes of Standardization in Africa An international account of word final /r/-variation in a Northern English accent conversational microconvergences between dialect and standard — etc.

Preisänderung vorbehalten

Walter de Gruyter

w DE

G

Berlin • New York