![10 Nikolaus P. Himmelmann
other things, native speakers may be more comfortable to discuss metalin-
guistic knowledge without being constantly recorded). But, to repeat, regard-
less of the recording method, records of observable linguistic behavior and
metalinguistic knowledge both contain primary data documenting linguistic
interactions in which native speakers participate.
In the following, we will use the label corpus of primary data as a short-
hand for corpus of recordings of observable linguistic behavior and meta-
linguistic knowledge for this component of a language documentation.
Throughout this book it is assumed that this corpus is stored and made
available in digital form.
To date, there is very little practical experience with regard to structuring
and maintaining such digital corpora. Consequently, no widely-used and
well-tested structure exists for them. Within the DoBeS program, it is a
widespread practice to operate with two basic components in structuring
primary data: records of individual communicative events and a lexical
database (this obviously follows a widespread practice in linguistic field-
work where apart from transcripts of recordings and fieldnotes the compila-
tion of a lexical database is a standard procedure).
Records of individual communicative events are called sessions (alter-
native terms would be “document”, “text”, or “resource bundle”). In the
manual for the IMDI Browser,4
a session is defined as “a meaningful unit of
analysis, usually […] a piece of data having the same overall content, the
same set of participants, and the same location and time, e.g., one elicita-
tion session on topic X, or one folktale, or one ‘matching game’, or one
conversation between several speakers.” It could also be the recording of a
two-day ceremony. Sessions are typically allocated to different sets defined
according to parameters such as medium (written vs. spoken), genre (mono-
logue, dialogue, historical, chatting, etc.), naturalness (spontaneous, staged,
elicited, etc.), and so on. It is too early to tell whether some of the various
corpus structures currently being used are preferable to others.
There are two reasons why a lexical database appears to be a useful
format for organizing primary data. On the one hand, there is a need to
bring together all the information available for a given item so that one can
make sure that the meaning and formal properties of the item are well un-
derstood.5
On the other hand, and perhaps more importantly, a list of lexical
items is a very useful resource when working on the transcription and trans-
lation of recordings. One of the most widely used computational tools in
descriptive linguistics, the program Toolbox (formerly Shoebox),6
allows for
the semi-automatic compilation of a lexical database when working through](https://image.slidesharecdn.com/2384689-250605164618-400bf7d4/85/Essentials-Of-Language-Documentation-Trends-In-Linguistics-Studies-And-Monographs-Tilsm-Nikolaus-P-Himmelmann-24-320.jpg)
![12 Nikolaus P. Himmelmann
– a brief characterization of the content of the session (what topic is being
talked about? what kind of communicative event [narrative, conversa-
tion, song, etc.] is being documented?);
– links between different files which together constitute the session, e.g. a
media file (audio or video) and a file containing a transcription, trans-
lation, and various types of commentary relevant for interpreting the re-
cording contained in the media file (on which see further below).
The metadata on both levels have two interrelated functions. On the one
hand, they facilitate access to a documentation or a specific record within a
documentation by providing key access information in a standardized for-
mat (what, where, when, etc.). In this function, they are similar to a cata-
logue in a library and we can thus speak of a cataloguing function.9
On the
other hand, they have an organizational function in that they define the
structure of the corpus which, in particular in the case of documentations in
digital format, in turn provides the basis for various procedures such as
searching, copying, or filtering within a single documentation or across a
set of documentations. Obviously, a metadata standard which targets the
organizational function has to be richer and more elaborate than one which
targets the cataloguing function. The former is actually a corpus manage-
ment tool, which defines digital structures and supports various computa-
tional procedures, rather than just a standard for organizing a catalogue.
Currently there exist two metadata standards which in fact complement
each other in that they target these different functions. The OLAC standard
targets exclusively the cataloguing function and provides an easy and fast
access to a large number of diverse repositories of primary data on a
worldwide scale (in both digital and non-digital formats). The IMDI stan-
dard, which incorporates all the information included in the OLAC standard
and hence is compatible with it, is actually a corpus management tool
which primarily targets digitally archived language documentations. Further
discussion of metadata concepts and standards is found in Chapters 4 and
13.
Apart from metadata, there is in most instances also a need for further
information accompanying each recording as well as the documentation as
a whole in order to make the corpus of primary data useful to users who do
not know the language being documented. On the level of individual ses-
sions, such additional information is called here an annotation.10
Thus, in
the case of audio or video recordings of communicative events, it is obvi-
ously useful to provide at least a transcription and a translation so that users](https://image.slidesharecdn.com/2384689-250605164618-400bf7d4/85/Essentials-Of-Language-Documentation-Trends-In-Linguistics-Studies-And-Monographs-Tilsm-Nikolaus-P-Himmelmann-26-320.jpg)
![Chapter 1 – Language documentation: What is it and what is it good for? 17
On a more profound level, in non-literate societies the documentation of
historical, cultural, and religious knowledge generally introduces a new way
for accessing such knowledge and thereby may change the whole psycho-
social fabric of the society (Ong 1982). This is particularly true of societies
where much of the social fabric depends on highly selective access to cultural
and historical knowledge, transmission of such knowledge thus involving
different levels of secrecy (see Brandt [1980, 1981] for a pertinent example).
That is, in some instances a documentation project may contribute to the
demise of the very linguistic and cultural practices it proposes to document.
In these instances, it would appear to be preferable not to document, but
rather to support language maintenance in other ways, if necessary and pos-
sible.
Note that in general, language documentation and language maintenance
efforts are not opposed to each other but go hand in hand. That is, it is an
integral part of the documentation framework elaborated in this book that it
considers it an essential task of language documentation projects to support
language maintenance efforts wherever such support is needed and wel-
comed by the community being documented. More specifically, the docu-
mentation should contain primary data which can be used in the creation of
linguistic resources to support language maintenance, and the documenta-
tion team should plan to dedicate a part of its resources to “mobilizing” the
data compiled in the project for maintenance purposes. Chapter 15 elabo-
rates some of the issues involved here.
4. Alternative formats for language documentations
The format for language documentations sketched in the preceding section
is certainly not the only possible format. In fact, within structural linguistics
there is a well-established format for language documentations consisting
primarily of a grammar and a dictionary. In this section, I will first briefly
present some arguments as to why this well-established format is strictly
speaking a format for language description and not for language documen-
tation proper, and thus is not a viable alternative to the basic documentation
format of Table 1. In Section 4.2, we will then turn to the question of
whether it makes sense to integrate the grammar-dictionary format with the
basic documentation format of Table 1 and thus make fully worked-out
grammars and dictionaries essential components of language documenta-
tions.](https://image.slidesharecdn.com/2384689-250605164618-400bf7d4/85/Essentials-Of-Language-Documentation-Trends-In-Linguistics-Studies-And-Monographs-Tilsm-Nikolaus-P-Himmelmann-31-320.jpg)
![Chapter 1 – Language documentation: What is it and what is it good for? 29
Acknowledgements
I wish to thank my co-editors and Eva Schultze-Berndt for critical discus-
sion of many of the issues touched upon here, as well as helpful comments
on the draft version of this chapter.
Notes
1. With regard to the latter point, compare the following quote from Luraghi
(1990: 128 FN1) which nicely illustrates the problems arising when data types
are missing in a given corpus: “As to the position of the verb, the most impor-
tant difference [between main and subordinate clauses, NPH] lies in the ab-
sence of VSO sentences in subordinate clauses. It can of course be objected that
this may be due simply to the shortage of sources, since VSO sentences are on
the whole very infrequent. However, in the light of comparative data from
other Indo-European languages, this objection could perhaps be rejected …”
2. The major limitation here are restrictions on access to recordings imposed by
speakers or communities which, of course, should be observed.
3. “Experiment” here is to be taken in a broad sense, including, for example, the
testing of the acceptability of invented examples.
4. IMDI = ISLE Metadata Initiative. The manual can be downloaded at http://
www.mpi.nl/IMDI/tools.
5. Note that this does not necessarily imply that all the information for a lexical
item has to be gathered in a single location (i.e. an entry in the database), as it
is currently done by most researchers. Alternatively, the lexical database could
consist simply of links to all the sessions where the item in question occurs.
This could include a session where the item is elicited as part of the elicitation
of a word list or semantic field, a session where the item has been recorded in a
list of items or a carrier phrase in order to document characteristic sound pat-
terns, and a session where it occurs as part of a procedural text.
6. Please refer to the appendix for further information on this program.
7. Note that linguistic interaction here includes interactions with native speakers
of other varieties inasmuch as they are a common occurrence in the speech
community which is being documented.
8. The following list takes an audio or video recording as its main example. Of
course, the same type of metadata is needed for primary data gathered in a dif-
ferent way such as written fieldnotes or photos.
9. Note that the term cataloguing is used here in a somewhat broader sense than
in Chapter 4 where it is used to refer to one particular subtype of metadata.
10. Strictly speaking, “annotations” could also be called metadata since the term
“metadata” in general refers to all kinds of data about data. However, within the](https://image.slidesharecdn.com/2384689-250605164618-400bf7d4/85/Essentials-Of-Language-Documentation-Trends-In-Linguistics-Studies-And-Monographs-Tilsm-Nikolaus-P-Himmelmann-43-320.jpg)
![32 Arienne M. Dwyer
1. Ethics
1.1. Research as mediation
Ethical behavior is often assumed to flow intuitively from the noble goals
of scientific research. Most fieldworkers consider themselves well-inten-
tioned, rational people. But have all participating individuals and groups
been considered in these research goals? Have their ethical standards been
considered?
Fieldwork methodology has in the last decades progressed from a typi-
cally non-cooperative model (research on a community) to a cooperative
model which in its strongest form explicitly empowers speech communities
(research on, for, and with a community) (Cameron et al. 1992: 22–24).
Assumptions about what is ethical for a particular field situation are best
avoided, especially assumptions on the part of the researcher about what
participants want.2
The researcher should also have a grasp of the legal
implications (local, national, and international) of data ownership.3
An un-
derstanding of ethical and legal responsibilities also facilitates the building
of trust – and thus a successful relationship – with a community research
team. Finally, making ethical and legal premises explicit, helps to anticipate
and avoid problems. A field researcher mediates between speakers, their
communities and the fieldworker’s own community, which includes an
institution, a funding body, and possibly an archive. Inevitably, all partici-
pants in a language documentation will face ethical dilemmas, in which no
course of action seems quite satisfactory. There may be “no right decision,
only… [one] ‘more right’ than the alternatives” (Hill, Glaser and Harden
1995: 19).
Distilled to its essence, the ethics of field research entails indigenous
people and field researchers mediating each other's cultural imperatives.
This contextualization of ethical principles can only occur through produc-
tive mutual negotiation at the local level. The ethical principles presented
here may seem as both imperious and overly generic, given that in this
chapter broad-brush principles are often preceded by the cajoling impera-
tive should or the bossy must. But these are suggestions awaiting contextu-
alization in a particular research situation. And this mediation of ethical
principles by all participants forms the nucleus of any research project.](https://image.slidesharecdn.com/2384689-250605164618-400bf7d4/85/Essentials-Of-Language-Documentation-Trends-In-Linguistics-Studies-And-Monographs-Tilsm-Nikolaus-P-Himmelmann-46-320.jpg)
![34 Arienne M. Dwyer
1.2.2. Balancing priorities
Since field linguistic situations are so diverse, one-size-fits-all codes of
conduct are impractical. Codes of conduct are voluntary and often largely
unenforceable, but good guidelines help ensure good working relationships
and a positive research outcome. For the sake of methodological transpar-
ency, and for smooth communications between all parties, some norms are
always part of the field experience.
Most research teams choose a pragmatic approach, making use of both
explicit ethical guidelines as well as drawing observations from specific
field experiences.5
No matter what form is chosen, research teams would do
well to make explicit the ethical norms of their particular project.
1.2.3. Normative ethics in language documentation
Individual teams should establish a code of ethical norms specific to their
particular area for a given research project. This code would encompass
detailed guidelines on consultation and negotiation between indigenous
people and researchers for all phases of the research, including planning
and dissemination.
Since such voluntary normative approaches have proven useful, the sci-
entific community can aim at establishing a two-tiered, flexible ethical code
for linguistic field research: a generic code of putatively universal ethical
norms, and as above a specific individual code for a research on an ethnic
group in a particular area, created by individual researchers.
At present, linguists lack a generic code of conduct. Ideally, field lin-
guists will work with the country’s linguists and social scientists to devise
this generic code. This code would be specific for field linguistics but could
be modelled on existing well-articulated guidelines (such as the Australian
Institute of Aboriginal and Torres Strait Islander Studies’ Guidelines for
ethical research in indigenous studies [AIATSIS 2000], the African Studies
Association’s Guidelines for ethical conduct in research and projects in
Africa [African Studies Association n.d.], and the American Anthropological
Association’s Code of ethics [AAA 1998]). Though the above are designed
as regional codes, they are actually generic enough as to be potentially ap-
plicable to any world region.
A generic statement on ethical principles should address all phases of re-
search: planning, fieldwork, analysis, archiving, and end products. Planning
ethically for each phase entails assessing the roles played by participants](https://image.slidesharecdn.com/2384689-250605164618-400bf7d4/85/Essentials-Of-Language-Documentation-Trends-In-Linguistics-Studies-And-Monographs-Tilsm-Nikolaus-P-Himmelmann-48-320.jpg)
![40 Arienne M. Dwyer
Principle 5: Archive and disseminate your data and results
Researchers must avoid being buried with their unpublished field notes and
recordings. Within bounds of informed consent, those working with endan-
gered-language communities have an obligation to appropriately store and
publish data and analyses. Even in imperfect form, ordered, shared data are
more useful than no data; disseminating or at least properly archiving col-
lected data is far more respectful to a speaker community than piling it in
the back of a closet. Hence, many field researchers now believe that best-
practice archiving (cf. EMELD 2000–2005) and dissemination (in any for-
mat) should be a requirement of fieldwork.
Such principles sketch out the bare minimum in ethical linguistic fieldwork
practice. For more elaborated documents, see AIATSIS (2000) and the Af-
rican Studies Association (n.d.).
1.5. Potential problems: some examples
1.5.1. The observer’s paradox and covert research
The requirement of obtaining informed consent rules out covert research,
i.e. recording without speaker’s knowledge. The deception inherent in covert
research renders it taboo for many who do fieldwork. Yet many social scien-
tists routinely pretend to be ordinary citizens in order to obtain a naturalistic
view of their research subjects: they, for example, join a group that believes
in UFOs, work desk jobs for the sensationalist newspaper Bild Zeitung, or
staff a Wal-Mart store to reveal the group or corporate practices (Wallraff
1977, Ehrenreich 2002). Such fieldworkers and journalists will vocifer-
ously defend their enterprise.
In anthropology and linguistics fieldwork, a researcher’s presence changes
the phenomena under observation, often making conversation less sponta-
neous. Most field workers simply attempt to minimize the intrusiveness of
their presence (the so-called observer’s paradox [Labov 1971: 171]) by, for
example, using a small recording device, or by having native-speaker insid-
ers conduct the field research. These methods have provided adequate data
and have been seen as ethically sound by the majority of field linguists and
community researchers.
However, since the observer is always intrusive to some extent, some
language researchers have decided to make surreptitious recordings. This](https://image.slidesharecdn.com/2384689-250605164618-400bf7d4/85/Essentials-Of-Language-Documentation-Trends-In-Linguistics-Studies-And-Monographs-Tilsm-Nikolaus-P-Himmelmann-54-320.jpg)
Essentials Of Language Documentation Trends In Linguistics Studies And Monographs Tilsm Nikolaus P Himmelmann
- 1. Essentials Of Language Documentation Trends In Linguistics Studies And Monographs Tilsm Nikolaus P Himmelmann download https://ebookbell.com/product/essentials-of-language- documentation-trends-in-linguistics-studies-and-monographs-tilsm- nikolaus-p-himmelmann-4769378 Explore and download more ebooks at ebookbell.com
- 2. Here are some recommended products that we believe you will be interested in. You can click the link to download. Essentials Of Language Documentation Jost Gippert Editor Nikolaus P Himmelmann Editor Ulrike Mosel Editor https://ebookbell.com/product/essentials-of-language-documentation- jost-gippert-editor-nikolaus-p-himmelmann-editor-ulrike-mosel- editor-49413086 Essentials Of The Language Of Conscience Illustrated Tieman Dipple https://ebookbell.com/product/essentials-of-the-language-of- conscience-illustrated-tieman-dipple-44468028 Essentials Of Programming Languages Daniel P Friedman Mitchell Wand https://ebookbell.com/product/essentials-of-programming-languages- daniel-p-friedman-mitchell-wand-56635878 Essentials Of Programming Languages 2nd Edition 2nd Edition Daniel P Friedman https://ebookbell.com/product/essentials-of-programming-languages-2nd- edition-2nd-edition-daniel-p-friedman-2151200
- 3. Essentials Of Programming Languages Third Edition Daniel P Friedman Mitchell Wand https://ebookbell.com/product/essentials-of-programming-languages- third-edition-daniel-p-friedman-mitchell-wand-32790316 The Essentials Of World Languages K12 Effective Curriculum Instruction And Assessment Janis Jensen https://ebookbell.com/product/the-essentials-of-world- languages-k12-effective-curriculum-instruction-and-assessment-janis- jensen-5046920 The Essential Role Of Language In Survey Research 1st Edition Mandy Sha Tim Gabel https://ebookbell.com/product/the-essential-role-of-language-in- survey-research-1st-edition-mandy-sha-tim-gabel-51706466 Greek An Essential Grammar Of The Modern Language David Holton https://ebookbell.com/product/greek-an-essential-grammar-of-the- modern-language-david-holton-4672704 Greek For The Rest Of Us The Essentials Of Biblical Greek Using Greek Tools Without Mastering Biblical Languages William D Mounce Mounce https://ebookbell.com/product/greek-for-the-rest-of-us-the-essentials- of-biblical-greek-using-greek-tools-without-mastering-biblical- languages-william-d-mounce-mounce-33497838
- 5. Essentials of Language Documentation ≥
- 6. Trends in Linguistics Studies and Monographs 178 Editors Walter Bisang Hans Henrich Hock Werner Winter Mouton de Gruyter Berlin · New York
- 7. Essentials of Language Documentation edited by Jost Gippert Nikolaus P. Himmelmann Ulrike Mosel Mouton de Gruyter Berlin · New York
- 8. Mouton de Gruyter (formerly Mouton, The Hague) is a Division of Walter de Gruyter GmbH & Co. KG, Berlin. 앪 앝 Printed on acid-free paper which falls within the guidelines of the ANSI to ensure permanence and durability. Library of Congress Cataloging-in-Publication Data Essentials of language documentation / edited by Jost Gippert, Niko- laus P. Himmelmann, Ulrike Mosel. p. cm. ⫺ (Trends in linguistics. Studies and monographs ; 178) Includes bibliographical references and index. ISBN-13: 978-3-11-018864-6 (cloth : alk. paper) ISBN-10: 3-11-018864-3 (cloth : alk. paper) 1. Linguistics ⫺ Documentation. 2. Language and languages ⫺ Documentation. I. Gippert, Jost. II. Himmelmann, Nikolaus P., 1959⫺ III. Mosel, Ulrike IV. Series P128.D63E85 2006 025.06141⫺dc22 2006001315 ISBN-13: 978-3-11-018864-6 ISBN-10: 3-11-018864-3 ISSN 1861-4302 Bibliographic information published by Die Deutsche Bibliothek Die Deutsche Bibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available in the Internet at ⬍http://dnb.ddb.de⬎. ” Copyright 2006 by Walter de Gruyter GmbH & Co. KG, D-10785 Berlin All rights reserved, including those of translation into foreign languages. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechan- ical, including photocopy, recording or any information storage and retrieval system, with- out permission in writing from the publisher. Cover design: Christopher Schneider, Berlin. Printed in Germany.
- 9. Editors’ preface Language documentation is concerned with the methods, tools, and theoreti- cal underpinnings for compiling a representative and lasting multipurpose record of a natural language or one of its varieties. It is a rapidly emerging new field in linguistics and related disciplines working with little-known speech communities. While in terms of its most recent history, language documentation has co-evolved with the increasing concern for language endangerment, it is not only of interest for work on endangered languages but for all areas of linguistics and neighboring disciplines concerned with setting new standards regarding the empirical foundations of their research. Among other things, this means that the quality of primary data is carefully and constantly monitored and documented, that the interfaces between pri- mary data and various types of analysis are made explicit and critically reviewed, and that provisions are taken to ensure the long-term preservation of primary data so that it can be used in new theoretical ventures as well as in (re-)evaluating and testing well-established theories. This volume presents in-depth introductions into major aspects of lan- guage documentation, including a definition of what it means to “document a language,” overviews on fieldwork ethics and practicalities and data processing, discussions on how to provide a basic annotation of digitally- stored multimedia corpora of primary data, as well as long-term perspectives on the preservation and use of such corpora. It combines theoretical and practical considerations and makes specific suggestions for the most com- mon problems encountered in language documentation. The volume should prove to be most useful to students and researchers concerned with documenting little-known languages and language varie- ties. In addition to linguists and anthropologists, this includes students and researchers in various regional studies and philologies such as African Studies, Indology, Turkology, Semitic Studies, or South American Studies. The book presupposes familiarity with the basic concepts and terminology of descriptive linguistics (for example, basic units such as phoneme or lex- eme), but most chapters will also be accessible and useful to non- specialists, including educators, language planners, politicians, and govern- ment officials concerned with linguistic minorities.
- 10. vi Preface Nearly all chapters of this volume are based on a series of lectures and seminars presented during the First International Summer School on Lan- guage Documentation: Methods and Technology held in Frankfurt/Main (Sept. 1–11, 2004). While not a textbook in the strict sense (which would in- clude exercises, etc.), the volume is designed to serve as the main source of readings for a university class on language documentation (for third-year students and above). Parts of it can also be used as readings in fieldmethod classes and classes in linguistic anthropology. However, it is not a guide to linguistic fieldwork. Instead, it focuses on issues which are typically not men- tioned at all, or all too briefly, in fieldwork manuals such as, for example, the cooperative interaction between researcher(s) and speech community, orthography development, the function of metadata, archiving recordings and transcripts. When used as a textbook in a language documentation class, it should be complemented with readings on linguistic fieldwork and lin- guistic anthropology from other sources (see further Section 5 of Chapter 1). Of major import to documentary linguistics is the technology used in recording and preserving linguistic primary data, most of which is IT- related today. Since this is a rapidly changing field, we have kept the dis- cussion of specific technological aspects and procedures to an absolute minimum, focusing on conceptual issues and practicalities which we be- lieve will stay with us for some time to come. Nevertheless, a considerable number of technical standards, software programs, and institutions con- cerned with corpus building and preservation are mentioned in this book in order to provide examples for a given conceptual issue or a recommended general procedure. The appendix provides an alphabetical list of all the abbreviations used in this regard, as well as internet links providing more up-to-date information on them. This information is continuously updated on the book’s website at: http://titus.uni-frankfurt.de/ld On this website, the reader will also find video and audio files for some of the examples given in this book as well as links and suggestions for topics which could not be adequately dealt with here. Finally, it bears emphasizing once again that language documentation in many ways is still a rather new discipline where many basic concepts and procedures are in the process of being tested and fully elaborated (see also Section 3.2 in Chapter 1). In particular, while considerable progress has been made in recent years with regard to the compilation and archiving aspects of language documentation, to date there is very little experience
- 11. Preface vii indeed with regard to actually working with digitally-stored multimedia corpora of lesser-known languages. In the coming years, we expect to see major developments here with regard to the etiquette of working with such corpora (How are they evaluated? How are they referred to in publications? How can work by different investigators on the same variety be combined into a single coherent corpus?) as well as with regard to the technology used in exploring them and extracting relevant information for a specific project. We also expect an impact on the methodological and theoretical debate in the subject areas working most intensively with data from such corpora, including linguistic typology, linguistic anthropology, and oral literature. As a part of these developments, it may well turn out that some of the suggestions made in this book, e.g. with regard to the structuring of the corpora or the format for annotations, will need to be revised or perhaps even be discarded. Still, we trust that the discussion of the basic conceptual issues as laid out here will be of continued interest and relevance for many years to come and thus truly merit to be considered “essentials of language documentation.” Nikolaus P. Himmelmann, Bochum Jost Gippert, Frankfurt Ulrike Mosel, Kiel
- 12. viii Preface Acknowledgements We gratefully acknowledge the very generous support of the Volkswagen- Stiftung (http://www.volkswagenstiftung.de) which has been instrumental in producing this book. The foundation not only funded the summer school for which most chapters were drafted, but also provided the means to dis- tribute a substantial number of copies of this book free of charge outside of Western Europe, North America, and Japan. By granting a research fellow- ship for Himmelmann in 2004–2005, it has allowed him to focus his re- search on the issues dealt with in Chapters 7 and 10 and to engage in the editing of the book in a way which otherwise would not have been possible. Through its DoBeS Programm (Documentation of Endangered Languages program), which started in the year 2000, it has made a major contribution to the development of documentary linguistics as an innovative field of study and practice within the humanities. Our sincerest thanks are due to the contributors of the volume who spent a lot of time on conceiving their chapters and have always been ready to cooperate with us in the difficult task of preparing a consistent book. We also gratefully acknowledge much practical help we have received in putting the volume together. Marcia Schwartz checked English and style conventions; Judith Köhne compiled the combined list of bibliographical references at the end. At Mouton, Ursula Kleinhenz did a great job of seeing the book through to press. Many thanks to all of you.
- 13. Contents Editor’s preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v Acknowledgements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii Chapter 1 Language documentation: What is it and what is it good for? . . . . . . . . . . . . . . . . . . . . . 1 Nikolaus P. Himmelmann Chapter 2 Ethics and practicalities of cooperative fieldwork and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Arienne M. Dwyer Chapter 3 Fieldwork and community language work. . . . . . . . . . . . . . 67 Ulrike Mosel Chapter 4 Data and language documentation . . . . . . . . . . . . . . . . . . . . . 87 Peter K. Austin Chapter 5 The ethnography of language and language documentation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 Jane H. Hill Chapter 6 Documenting lexical knowledge . . . . . . . . . . . . . . . . . . . . . . 129 John B. Haviland Chapter 7 Prosody in language documentation . . . . . . . . . . . . . . . . . . . 163 Nikolaus P. Himmelmann Chapter 8 Ethnography in language documentation . . . . . . . . . . . . . . . 183 Bruna Franchetto Chapter 9 Linguistic annotation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 Eva Schultze-Berndt Chapter 10 The challenges of segmenting spoken language . . . . . . . . 253 Nikolaus P. Himmelmann
- 14. x Contents Chapter 11 Orthography development . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275 Frank Seifart Chapter 12 Sketch grammar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301 Ulrike Mosel Chapter 13 Archiving challenges. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 Paul Trilsbeek and Peter Wittenburg Chapter 14 Linguistic documentation and the encoding of textual materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337 Jost Gippert Chapter 15 Thick interfaces: mobilizing language documentation with multimedia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363 David Nathan Abbreviations and resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 391 Language index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 417 Subject index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
- 15. Chapter 1 Language documentation: What is it and what is it good for? Nikolaus P. Himmelmann Introduction This chapter defines language documentation as a field of linguistic inquiry and practice in its own right which is primarily concerned with the compi- lation and preservation of linguistic primary data and interfaces between primary data and various types of analyses based on these data. Further- more, it argues (in Section 2) that while language endangerment is a major reason for getting involved in language documentation, it is not the only one. Language documentations strengthen the empirical foundations of those branches of linguistics and related disciplines which heavily draw on data of little-known speech communities (e.g. linguistic typology, cognitive anthropology, etc.) in that they significantly improve accountability (verifi- ability) and economizing research resources. The primary data which constitute the core of a language documentation include audio or video recordings of a communicative event (a narrative, a conversation, etc.), but also the notes taken in an elicitation session, or a genealogy written down by a literate native speaker. These primary data are compiled in a structured corpus and have to be made accessible by various types of annotations and commentary, here summarily referred to as the “apparatus”. Sections 3 and 4 provide further discussion of the components and structure of language documentations. Section 5 concludes with a pre- view of the remaining chapters of this book. 1. What is a language documentation? An initial, preliminary answer to this question is: a language documenta- tion is a lasting, multipurpose record of a language. This answer, of course, is not quite satisfactory since it immediately raises the question of
- 16. 2 Nikolaus P. Himmelmann what we mean by “lasting”, “multipurpose” and “record of a language”. In the following, these constituents of the definition are taken up in reverse order, beginning with “record of a language”. At first sight, a further definition of “record of a language” may look like a bigger a problem than it actually is since it involves the highly com- plex and controversial issue of defining “a language”. The main problem with defining “a language” consists in the fact that the word language refers to a number of different, though interrelated phenomena. The problems in defining it vary considerably, depending on which phenomenon is focused upon. That is, different problems surface when the task is to define lan- guage as opposed to dialect, or language as a field of scientific enquiry, or language as a cognitive faculty of humans, and so on. Unless we want to postpone working on language documentations until the probably never arriving day when all the conceptual problems of defining language in all of its different senses are resolved and a theoretically well-balanced delimi- tation of “a language” for the purposes of language documentations is pos- sible, we need a pragmatic approach in dealing with this problem. The basic tenet of such a pragmatic approach is implied by the qualifiers multipurpose and lasting in the definition above: The net should be cast as widely as possible. That is, a language documentation should strive to in- clude as many and as varied records as practically feasible, covering all aspects of the set of interrelated phenomena commonly called a language. Ideally, then, a language documentation would cover all registers and varie- ties, social or local; it would contain evidence for language as a social prac- tice as well as a cognitive faculty; it would include specimens of spoken and written language; and so on. A language documentation broadly conceived along these lines could serve a large variety of different uses in, for example, language planning decisions, preparing educational materials, or analyzing a set of problems in syntactic theory. Users of such a multipurpose documentation would include the speech community itself, national and international agencies concerned with education and language planning, as well as researchers in various disciplines (linguistics, anthropology, oral history, etc.). In fact, the qualifier lasting adds a long-term perspective which goes beyond current issues and concerns. The goal is not a short-term record for a specific pur- pose or interest group, but a record for generations and user groups whose identity is still unknown and who may want to explore questions not yet raised at the time when the language documentation was compiled. Obviously, this pragmatic explication of “lasting, multipurpose record of a language” rests on the assumption that it is possible and useful to com-
- 17. Chapter 1 – Language documentation: What is it and what is it good for? 3 pile a database for a very broadly defined subject matter (“a language”) without being guided by a specific theoretical or practical problem in mind which could be resolved on the basis of this database. With regard to its use in scientific inquiries, the validity of this assumption is shown by the suc- cess of all those social and historical disciplines working with data not spe- cifically produced for research purposes. Thus, for example, cave dwellers in the Stone Age did not discard shellfish, animal bones, fragments of tools, and the like within the cave with the purpose in mind of documenting their presence and aspects of their diet and culture. But archeologists today use this haphazardly discarded waste as the primary data for determining the length and type of human occupation found in a given location. Similarly, inscriptions on stones, bones, or clay tablets were not produced in order to provide a record of linguistic structures and practices, but they have suc- cessfully been used to explore the structural properties of languages such as Hittite or Sumerian, which had already been extinct for millennia before their modern linguistic analysis began. However, it is also well known that historical remains and records tend to be deficient in some ways with regard to modern purposes. Stone in- scriptions and other historic documents with linguistic content, for exam- ple, never provide a comprehensive record of the linguistic structures and practices in use in the community at the time when these documents were written. Thus, given that the Hittite records discovered to date mostly per- tain to matters of government, law, trade, and religion, it remains unknown how Hittite adolescents chatted with each other or whether it was possible to have the verb in first position in subordinate clauses.1 The experience with historical remains and records thus is ambivalent: On the one hand, it clearly shows that they may serve as the database for exploring issues they were not intended for. On the other hand, they show that haphazardly compiled databases hardly ever contain all the information one needs to answer all the questions of current interest. Based on this ob- servation, the basic idea of a language documentation as developed here can be stated as follows: The goal is to create a record of a language in the sense of a comprehensive corpus of primary data which leaves nothing to be desired by later generations wanting to explore whatever aspect of the language they are interested in (what exactly is meant by “primary data” here is further discussed in Section 3.1.1 below). Put in this way, the task of compiling a language documentation is enormous, and there is no principled upper limit for it. Obviously, every specific documentation project will have to limit its scope and set specific
- 18. 4 Nikolaus P. Himmelmann targets. Guidelines and suggestions as to how to go about setting such limits and targets are further discussed below and in the remaining chapters of this book. But to begin with, the fundamental importance of taking a prag- matic stance in all matters of language documentation needs to be empha- sized once again. There are major practical constraints on the usefulness of targets and delimitations for language documentations which are exclu- sively based on theoretical considerations regarding the nature of language and speech communities. In most if not all documentation settings, the range of items that can be documented will be determined to a significant degree by factors that are specific to the given setting, most importantly, the availability of speakers who are willing and able to participate in the documentation effort. In fact, recent experiences make it clear that encour- aging native speakers to take an active part in determining the contents of a documentation significantly increases the productivity of a documentation project. Consequently, a theoretical framework for language documentation should provide room for the active participation of native speakers. While the input of native speakers and other factors specific to a given setting is not completely unpredictable, it clearly limits the level of detail of a general framework for language documentation which can be usefully explored in purely theoretical terms. This assessment, however, should not be construed as denying the rele- vance of theorizing language documentations. Not everything in a docu- mentation is fully determined by the specifics of a given documentation situation. Speakers and speech communities usually do not have a fully worked-out plan for what to document. Rather, the specifics of a documen- tation are usually established interactively by communities and research teams. On the part of the research team, this presupposes a theoretically grounded set of basic goals and targets one wants to achieve. Furthermore, without theoretical grounding language documentation is in the danger of producing “data graveyards”, i.e. large heaps of data with little or no use to anyone. While language documentation is based on the idea that it is possible and useful to dissociate the compilation of linguistic primary data from any particular theoretical or practical project based on this data, language documentation is not a theory-free or anti-theoretical enterprise. Its theoretical concerns pertain to the methods used in recording, processing, and preserving linguistic primary data, as well as to the question how it can be ensured that primary data collections are indeed of use for a broad range of theoretical and applied purposes. Among other things, documentation theory has to provide guidelines for determining targets in specific documentation projects. It also has to develop
- 19. Chapter 1 – Language documentation: What is it and what is it good for? 5 principled and intersubjective means for evaluating the quality of a given documentation regardless of the specific circumstances of its compilation. A further major concern pertains to the interface between primary data and analysis in a broad range of disciplines. Based on a detailed investigation and evaluation of basic analytical procedures in these disciplines, it has to be determined which type and format of primary data is required for a par- ticular analytical procedure so that it can be ensured that the appropriate type of data is included in a comprehensive documentation. The present book provides an introduction to basic practical and theo- retical issues in language documentation. It presents specific suggestions for the structure and contents of language documentations as well as the methodologies to be used in compiling them. To begin with, it will be useful briefly to address the question of what language documentations are good for. That is, why is it a useful enterprise to create lasting, multipurpose re- cords of a language? 2. What is a language documentation good for? From a linguistic point of view, there are essentially three reasons for engag- ing in language documentation, all of them having to do with consolidating and enlarging the empirical basis of a number of disciplines, in particular those branches of linguistics and related disciplines which heavily draw on data of little-known speech communities (e.g. descriptive linguistics, lin- guistic typology, cognitive anthropology, etc.). These are language endan- germent, the economy of research resources, and accountability. Certainly the major reason why linguists have recently started to engage with the idea of multipurpose documentations is the fact that a substantial number of the languages still spoken today are threatened by extinction (see Grenoble and Whaley 1998; Hagège 2000; Crystal 2000; or Bradley and Bradley 2002 for further discussion and references regarding language en- dangerment). In the case of an extinct language, it is obviously impossible to check data with native speakers or to collect additional data sets. Creating lasting multipurpose documentations is thus seen as one major linguistic response to the challenge of the dramatically increased level of language endangerment observable in our times. In this regard, language documenta- tions are not only seen as data repositories for scientific inquiries, but also as important resources for supporting language maintenance. Creating language documentations which are properly archived and made easily accessible to interested researchers is also in the interest of
- 20. 6 Nikolaus P. Himmelmann research economy. If someone worked on a minority language in the Philippines 50 years ago and someone else wanted to continue this work now, it would obviously be most useful if this new project could build on the complete set of primary data collected at the time and not just on a grammar sketch and perhaps a few texts published by the earlier project. Similarly, even if a given project on a little-known language is geared to- wards a very specific purpose – say, the conceptualization of space – it is in the interest of research economy (and accountability) if this project were to feed all the primary data collected in the project work into an open archive and not to limit itself to publishing the analytical results plus possibly a small sample of primary data illustrating their basic materials. While the set of primary data fed into an archive in these examples would surely fail to constitute a comprehensive record of a language, it could very well be of use for purposes other than the one motivating the original project (data from matching tasks developed to investigate the lin- guistic encoding of space, for example, are also quite useful for the analysis of intonation, for conversation analytic purposes, for grammatical analysis, and so on). More importantly, if it were common practice to feed complete sets of primary data into open archives (which do not necessarily have to form a physical unit), comprehensive documentations for quite a number of little-known languages could grow over time, which in turn would strength- en the empirical basis of all disciplines working on and with such lan- guages and cultures. That is, while much of the discussion in this chapter and book is concerned with projects specifically targeted at creating sub- stantial language documentations, the basic idea of creating lasting, multi- purpose documentations which are openly archived is not necessarily tied to such projects. It is very well possible and desirable to create such docu- mentations in a step-by-step fashion by compiling and integrating the pri- mary data sets collected in a number of different projects over an extended period of time. In fact, it is highly likely that in most instances, really com- prehensive documentations can only be created in this additive way. Finally, establishing open archives for primary data is also in the interest of making analyses accountable. Many claims and analyses related to lan- guages and speech communities for which no documentation is available remain unverifiable as long as substantial parts of the primary data on which the analyses are based remain inaccessible to further scrutiny. Accountability here is intended to include all kinds of practical checks and methodological tests with regard to the empirical basis of an analysis or theory, including replicability and falsifiability. The documentation format developed here
- 21. Chapter 1 – Language documentation: What is it and what is it good for? 7 encourages, and also provides practical guidelines for, the open and widely accessible archiving of all primary data collected for little-known lan- guages, regardless of their vitality.2 3. A basic format for language documentations This section presents a basic format for language documentations and then highlights some features which distinguish this format from related enter- prises. 3.1. The basic format 3.1.1. Primary data Continuing the argument developed in the preceding sections, it should be clear that a language documentation, conceived of as a lasting, multipur- pose record of a language, should contain a large set of primary data which provide evidence for the language(s) used at a given time in a given com- munity (in all of the different senses of “language”). Of major importance in this regard are specimens of observable linguistic behavior, i.e. examples of how the people actually communicate with each other. This includes all kinds of communicative activities in a speech community, from everyday small talk to elaborate rituals, from parents baby-talking to their newborn infants to political disputes between village elders. It is impossible to record all communicative events in a given speech community, not only for obvious practical, but also for theoretical and ethical reasons. Most importantly, such a record would imply a totalitarian set-up with video cameras and microphones everywhere and the speakers unable to control what of their behavior is recorded and what not. A major theoretical problem pertains to the fact that there is no principled way for determining a temporal boundary for such a recording (all communicative events in one day? two weeks? one year? a century?). Consequently, there is a need to sample the kinds of communicative events to be documented. Once again, we can distinguish between a prag- matic guideline and theoretically grounded targets. The pragmatic guideline simply says that one should record as many and as broad a range as possible of communicative events which commonly occur in the speech community.
- 22. 8 Nikolaus P. Himmelmann The theoretically grounded sampling procedure will be determined to a significant degree by the purposes and goals of the particular project. The rather broad and unspecific goal of a lasting, multipurpose record of a lan- guage envisioned here implies that, as much as possible, a sufficiently large number of examples for every type of communicative event found in a given speech community is collected. This in turn raises the highly complex issue of how the typology of communicative events in a given speech com- munity can be uncovered. Within sociolinguistics, the framework known as the ethnography of communication provides a starting point for dealing with this issue. Chapter 5 provides a brief introduction to major concepts relevant here. Chapter 8 lists a range of important topics and parameters. Besides observable linguistic behavior, is there anything else that needs to be documented in order to provide for a lasting, multipurpose record of a language? Or can all relevant information be extracted from a comprehen- sive corpus of recordings of communicative events? One aspect of “a lan- guage” that is not, or at least not easily, accessible by analyzing observable linguistic behavior is the tacit knowledge speakers have about their lan- guage. This is also known as metalinguistic knowledge and refers to the ability of native speakers to provide interpretations and systematizations for linguistic units and events. For example, speakers know that a given word is a taboo word, that speech event X usually has to be followed by speech event Y, or that putting a given sequence of elements in a different order is awkward or simply impossible. Similarly, metalinguistic knowledge as understood here also includes all kinds of linguistically based taxonomies, such as kinship systems, folk taxonomies for plants, animals, musical in- struments and styles, and other artifacts, expressions for numbers and measures, but also morphological paradigms. The documentation of metalinguistic knowledge, while not involving principled theoretical or ethical problems, is also not a straightforward task because much of it is not directly accessible. To be sure, in some instances there are conventional speech events involving the display of metalinguistic knowledge, such as reciting a genealogy or lengthy mythological narratives which sketch a cognitive map of the landscape. In many societies, there are also a number of well established and much discussed topics where speakers engage in metalinguistic discussions regarding the differences between dif- ferent varieties (in village X they say “da” but we say “de”; young people cannot pronounce our peculiar /k/-sound correctly anymore, etc.). Further- more, transcripts prepared by native speakers without direct interference by a linguist often provide interesting evidence regarding morpheme, word,
- 23. Chapter 1 – Language documentation: What is it and what is it good for? 9 and sentence boundaries (see Chapters 3 and 10 for further discussion). But very often documenting metalinguistic knowledge will involve the use of a broad array of elicitation strategies, guided by current theories about different kinds of metalinguistic knowledge and their structure. One very important type of elicited evidence are monolingual definitions of word meanings provided by native speakers. See Chapters 3 and 6 for further discussion and exemplification. The documentation of metalinguistic knowledge as understood here in- cludes much of the basic information that is needed for writing descriptive grammars and dictionaries. In particular, it includes all kinds of elicited data regarding the grammaticality or acceptability of phonological or mor- phosyntactic structures and the meaning, use, and relatedness of lexical items. However, it should be clearly understood that documentation here means that the elicitation process itself is documented in its entirety, in- cluding the questions asked or the stimuli presented by the researcher as well as the reaction by the native speaker(s). That is, documentation per- tains to the level of primary data which provide evidence for metalinguistic knowledge, i.e. what native speakers can actually articulate regarding their linguistic practices or their recordable reactions in experiments designed to probe metalinguistic knowledge.3 A grammatical rule as stated in a grammar or an entry in a published dictionary are not primary data in this sense, even though some linguists may believe that they are part of a native speakers’ (unconscious) metalinguistic knowledge. In this view, grammatical rules and dictionary entries are analytical formats for metalinguistic knowledge. Whether and to what extent these have a place in a language documentation is an issue we will take up in Section 4.2. It is also worth noting that the documentation of observable linguistic behavior and metalinguistic knowledge are similar in that they basically consist of records of communicative events. In the case of observable lin- guistic behavior, the communicative event involves the interaction of native speakers among themselves, while in the case of metalinguistic knowledge it involves the interaction between native speakers and documenters. There is a superficial difference with regard to the preferred documentation for- mat in that it is now standard practice to make (video) recordings of ob- servable linguistic behavior, while for the elicitation of metalinguistic knowl- edge it is still more common simply to take written notes. In principle, (video-)recording would also be the better (i.e. more reliable and compre- hensive) documentation format for elicited metalinguistic knowledge, but there may often be practical reasons to stay with paper and pencil (among
- 24. 10 Nikolaus P. Himmelmann other things, native speakers may be more comfortable to discuss metalin- guistic knowledge without being constantly recorded). But, to repeat, regard- less of the recording method, records of observable linguistic behavior and metalinguistic knowledge both contain primary data documenting linguistic interactions in which native speakers participate. In the following, we will use the label corpus of primary data as a short- hand for corpus of recordings of observable linguistic behavior and meta- linguistic knowledge for this component of a language documentation. Throughout this book it is assumed that this corpus is stored and made available in digital form. To date, there is very little practical experience with regard to structuring and maintaining such digital corpora. Consequently, no widely-used and well-tested structure exists for them. Within the DoBeS program, it is a widespread practice to operate with two basic components in structuring primary data: records of individual communicative events and a lexical database (this obviously follows a widespread practice in linguistic field- work where apart from transcripts of recordings and fieldnotes the compila- tion of a lexical database is a standard procedure). Records of individual communicative events are called sessions (alter- native terms would be “document”, “text”, or “resource bundle”). In the manual for the IMDI Browser,4 a session is defined as “a meaningful unit of analysis, usually […] a piece of data having the same overall content, the same set of participants, and the same location and time, e.g., one elicita- tion session on topic X, or one folktale, or one ‘matching game’, or one conversation between several speakers.” It could also be the recording of a two-day ceremony. Sessions are typically allocated to different sets defined according to parameters such as medium (written vs. spoken), genre (mono- logue, dialogue, historical, chatting, etc.), naturalness (spontaneous, staged, elicited, etc.), and so on. It is too early to tell whether some of the various corpus structures currently being used are preferable to others. There are two reasons why a lexical database appears to be a useful format for organizing primary data. On the one hand, there is a need to bring together all the information available for a given item so that one can make sure that the meaning and formal properties of the item are well un- derstood.5 On the other hand, and perhaps more importantly, a list of lexical items is a very useful resource when working on the transcription and trans- lation of recordings. One of the most widely used computational tools in descriptive linguistics, the program Toolbox (formerly Shoebox),6 allows for the semi-automatic compilation of a lexical database when working through
- 25. Chapter 1 – Language documentation: What is it and what is it good for? 11 a transcript, and the existence of this program is certainly one reason why the compilation of a lexical database currently is almost an automatic pro- cedure when working with recordings. However, as with all other aspects of organizing a digital corpus of primary data, it remains to be seen and tested further whether this is indeed a necessary and useful procedure. 3.1.2. Apparatus Inasmuch as linguistic and metalinguistic interactions cover the range of basic interactional possibilities,7 a documentation which contains a com- prehensive set of primary data for both types of interactions is logically complete with regard to the level of primary data. However, it is well known that a large corpus of primary data is of little use unless it is pre- sented in a format which ensures accessibility for parties other than the ones participating in its compilation. To be accessible to a broad range of users, including the speech community, the primary data need to be accom- panied by information of various kinds, which – following philological tradition – could be called the apparatus. The precise extent and format of the apparatus is a matter of debate, with one exception: the uncontroversial need for metadata. Metadata are required on two levels. First, the documentation as a whole needs metadata regarding the project(s) during which the data were com- piled, including information on the project team(s), and the object of docu- mentation (which variety? spoken where? number and type of records; etc.). Second, each session (= segment of primary data) has to be accompanied by information of the following kind:8 – a name of the session which uniquely identifies it within the overall corpus; – when and where was the data recorded?; – who is recorded and who else was present at the time?; – who made the recording and what kind of recording equipment was used?; – an indication of the quality of the data according to various parameters (recording environment and equipment, speaker competence, level of detail of further annotation); – who is allowed to access the data contained in this session?;
- 26. 12 Nikolaus P. Himmelmann – a brief characterization of the content of the session (what topic is being talked about? what kind of communicative event [narrative, conversa- tion, song, etc.] is being documented?); – links between different files which together constitute the session, e.g. a media file (audio or video) and a file containing a transcription, trans- lation, and various types of commentary relevant for interpreting the re- cording contained in the media file (on which see further below). The metadata on both levels have two interrelated functions. On the one hand, they facilitate access to a documentation or a specific record within a documentation by providing key access information in a standardized for- mat (what, where, when, etc.). In this function, they are similar to a cata- logue in a library and we can thus speak of a cataloguing function.9 On the other hand, they have an organizational function in that they define the structure of the corpus which, in particular in the case of documentations in digital format, in turn provides the basis for various procedures such as searching, copying, or filtering within a single documentation or across a set of documentations. Obviously, a metadata standard which targets the organizational function has to be richer and more elaborate than one which targets the cataloguing function. The former is actually a corpus manage- ment tool, which defines digital structures and supports various computa- tional procedures, rather than just a standard for organizing a catalogue. Currently there exist two metadata standards which in fact complement each other in that they target these different functions. The OLAC standard targets exclusively the cataloguing function and provides an easy and fast access to a large number of diverse repositories of primary data on a worldwide scale (in both digital and non-digital formats). The IMDI stan- dard, which incorporates all the information included in the OLAC standard and hence is compatible with it, is actually a corpus management tool which primarily targets digitally archived language documentations. Further discussion of metadata concepts and standards is found in Chapters 4 and 13. Apart from metadata, there is in most instances also a need for further information accompanying each recording as well as the documentation as a whole in order to make the corpus of primary data useful to users who do not know the language being documented. On the level of individual ses- sions, such additional information is called here an annotation.10 Thus, in the case of audio or video recordings of communicative events, it is obvi- ously useful to provide at least a transcription and a translation so that users
- 27. Chapter 1 – Language documentation: What is it and what is it good for? 13 not familiar with the language are able to understand what is going on in the recording. However, the exact extent and format of the annotations that should be included in each session is a matter of debate. It is common to distinguish between minimal and more elaborate annotation schemes. A widely as- sumed minimal annotation scheme consists of just a transcription and a free translation which should accompany all, or at least a substantial number of, primary data segments. More elaborate annotation schemes include various levels of interlinear glossing, grammatical as well ethnographical commen- tary, and extensive cross-referencing between the various sessions and re- sources compiled in a given documentation. See further Chapters 8 and 9. On the level of the overall documentation, information accompanying the primary data set other than metadata is, for lack of a well-established term, subsumed here under the heading general access resources (alterna- tively, it could also simply be called “annotation”). Such general (in the sense of: relevant for the documentation as a whole) access resources would include: – a general introduction which provides background information on the speech community and language (language name(s), affiliation, major varieties, etc.), the fieldwork setting(s), the methods used in recording primary data, an overview of the contents, structure, and scope of the primary data corpus and its quality; – brief sketches of major ethnographic and grammatical features being documented; – an explication of the various conventions that are being used (orthogra- phy, glossing abbreviations, other abbreviations); – indices for languages/varieties, key analytic concepts, etc.; – links and references to other resources (books and articles previously published on the variety or community being documented; other pro- jects relating to the community or its neighbors, etc.). For further discussion of some aspects of relevance here, see Chapters 8 and 12. Table 1 provides a schematic overview of the components of the lan- guage documentation format sketched in this section.
- 28. 14 Nikolaus P. Himmelmann Table 1. Basic format of a language documentation Primary data Apparatus Per session For documentation as a whole recordings/records of observable linguistic behavior and metalinguistic knowledge (possible basic formats: session and lexical database) Metadata – time and location of recording – participants – recording team – recording equipment – content descriptors … Annotations – transcription – translation – further linguistic and ethnographic glossing and commentary Metadata – location of documented community – project team(s) contributing to documentation – participants in documentation – acknowledgements … General access resources – introduction – orthographical conventions – ethnographic sketch – sketch grammar – glossing conventions – indices – links to other resources … 3.2. What’s new? Language documentation in the way depicted in Table 1 is not a totally new enterprise. The compilation of annotated collections of written historical documents and culturally important speech events (legends, epic poems, and the like) was the major concern of philologists in the nineteenth cen- tury. Linguistic and anthropological fieldwork in the Boasian tradition has also always put major emphasis on the recording of speech events. Within linguistic anthropology, recording and interpreting oral literature is a major task. All of these traditions have had a major influence on documentary linguistics as developed in this book.
- 29. Chapter 1 – Language documentation: What is it and what is it good for? 15 Nevertheless, the idea of a language documentation as sketched above is new for mainstream linguistics, and even compared to these earlier ap- proaches, it is new with regard to the following important features: – Focus on primary data: The main goal of a language documentation is to make primary data available for a broad group of users. Unlike in the philological tradition, there is no restriction to culturally or historically “important” documents, however such importance may be defined. – Explicit concern for accountability: The focus on primary data implies that considerable care is given to the issue of making it possible to evaluate the quality of the data. This in turn implies that the field situa- tion is made transparent and that all documents are accompanied by metadata which detail the recording circumstances as well as the further steps undertaken in processing a particular document. – Concern for long-term storage and preservation of primary data: This involves two aspects. On the one hand, metadata are crucial for users of a documentation to locate and evaluate a given document, as just men- tioned. On the other hand, long-term storage is essentially a matter of technology, and while compilers of language documentations do not have to be able to handle all the technology themselves, they need to have a basic understanding of the core issues involved so that they avoid basic mistakes in recording and processing primary data. Among other things, the quality of the recording is of utmost importance for long- term storage and hence needs explicit attention. See further Chapters 4, 13, and 14. – Work in interdisciplinary teams: Work on a truly comprehensive lan- guage documentation needs expertise in a multitude of disciplines in ad- dition to the basic linguistic expertise required in transcription and trans- lation. Such disciplines include anthropology, ethnomusicology, oral history and literature, as well as all the major subdisciplines of linguis- tics (socio- and psycholinguistics, phonetics, discourse analysis, corpus linguistics, etc.). There are probably no individuals who are experts in all of these fields, and few who have acquired significant expertise in a substantial number of them. Hence, good documentation work usually requires a team of researchers with different backgrounds and areas of expertise. – Close cooperation with and direct involvement of speech community: The documentation format sketched above strongly encourages the active involvement of (members of) the speech community in two ways. On
- 30. 16 Nikolaus P. Himmelmann the one hand, as mentioned above, native speakers are among the main players in determining the overall targets and outcomes of a documenta- tion project. On the other hand, a documentation project involves a sig- nificant number of activities which can be carried out with little or no academic training. For example, the recording of communicative events can be done by native speakers who know how to handle the recording equipment (which can be learned in very short time), and it is often preferable that they do such recordings on their own because they know where and when particular events happen, and their presence is fre- quently felt to be less obtrusive. Similarly, given some training and regular supervision, the recording of metalinguistic knowledge and also the transcription and translation of recordings can be carried out by na- tive speakers all by themselves. See further Chapter 3. 3.3. Limitations As with most other scientific enterprises, the language documentation for- mat developed here is not without problems and limitations. Some of the theoretical and practical problems have already been mentioned in the pre- ceding discussion, and it will suffice here to emphasize the fact that the documentation format in Table 1 is based on a number of hypotheses which may well be proven wrong or unworkable in practical terms (see further Section 4 below). In addition to theoretical and practical problems, there are also ethical problems and limitations which are related to the fact that even the most circumspectly planned documentation project has the poten- tial to profoundly change the social structure of the society being docu- mented. This may pertain to a number of different levels, only two of which are mentioned here (see Wilkins 1992, 2000; Himmelmann 1998; and Grinevald 2003: 60–62 for further discussion). On a somewhat superficial level, there are usually a few, often not more than one or two native speakers who are very actively involved in the pro- ject work. Through their work in the project, their social and economic status may change in a way that otherwise may have been impossible. This in turn may lead to (usually minor) disturbances in the wider community, such as inciting the envy or anger of relatives and neighbors. It is also not unknown that affiliation with an externally funded and administered project is used as an instrument in political controversies and competitions within the speech community.
- 31. Chapter 1 – Language documentation: What is it and what is it good for? 17 On a more profound level, in non-literate societies the documentation of historical, cultural, and religious knowledge generally introduces a new way for accessing such knowledge and thereby may change the whole psycho- social fabric of the society (Ong 1982). This is particularly true of societies where much of the social fabric depends on highly selective access to cultural and historical knowledge, transmission of such knowledge thus involving different levels of secrecy (see Brandt [1980, 1981] for a pertinent example). That is, in some instances a documentation project may contribute to the demise of the very linguistic and cultural practices it proposes to document. In these instances, it would appear to be preferable not to document, but rather to support language maintenance in other ways, if necessary and pos- sible. Note that in general, language documentation and language maintenance efforts are not opposed to each other but go hand in hand. That is, it is an integral part of the documentation framework elaborated in this book that it considers it an essential task of language documentation projects to support language maintenance efforts wherever such support is needed and wel- comed by the community being documented. More specifically, the docu- mentation should contain primary data which can be used in the creation of linguistic resources to support language maintenance, and the documenta- tion team should plan to dedicate a part of its resources to “mobilizing” the data compiled in the project for maintenance purposes. Chapter 15 elabo- rates some of the issues involved here. 4. Alternative formats for language documentations The format for language documentations sketched in the preceding section is certainly not the only possible format. In fact, within structural linguistics there is a well-established format for language documentations consisting primarily of a grammar and a dictionary. In this section, I will first briefly present some arguments as to why this well-established format is strictly speaking a format for language description and not for language documen- tation proper, and thus is not a viable alternative to the basic documentation format of Table 1. In Section 4.2, we will then turn to the question of whether it makes sense to integrate the grammar-dictionary format with the basic documentation format of Table 1 and thus make fully worked-out grammars and dictionaries essential components of language documenta- tions.
- 32. 18 Nikolaus P. Himmelmann It should be clearly understood that this section is merely intended to draw attention to this important topic at the core of documentation theory. It barely scratches the surface of the many complex issues involved here. For more discussion, see Labov (1975, 1996), Greenbaum (1984), Pawley (1985, 1986, 1993), Lehmann (1989, 2001, 2004b), Mosel (1987, 2006), Himmelmann (1996, 1998), Schütze (1996), Keller (2000), Ameka et al. (2006), among others. 4.1. The grammar-dictionary format The grammar-dictionary format of language description targets the language system.11 That is, it is based on the notion of a language as an abstract sys- tem of rules and oppositions which underlies the observable linguistic be- havior. In this view, documenting a language essentially involves compiling a grammar (= set of rules for producing utterances) and a dictionary (= a list of conventional form-meaning pairings used in producing these utter- ances). To this core of the documentation, a number of texts are often added, either in the form of a text collection or in the appendix to the gram- mar, which have the function of extended examples for how the system works in context. These texts are usually taken from the corpus of primary data on which the system description is based, but they do not actually pro- vide access to these primary data because they are edited in various ways. Providing direct access to the complete corpus of primary data is typically not part of this format. The compilation of grammars and (to a lesser extent) dictionaries is a well-established practice in structural linguistics, with many fine specimens having been produced in the last century. But even the best structuralist grammars and dictionaries have been lacking with regard to the goal of presenting a lasting, multipurpose record of a language. Major problems with regard to this goal include the following points: a. Many communicative practices found in a given speech community remain undocumented and unreconstructable. That is, provided with a grammar and a dictionary it is still impossible to know how the lan- guage is (or was) actually spoken. For example, it is impossible to derive from a grammar and a dictionary on how everyday conversational rou- tines look like (how does one say “hello, good morning”?) or how one linguistically interacts when building a house or negotiating a marriage.
- 33. Chapter 1 – Language documentation: What is it and what is it good for? 19 b. In line with the structuralist conception of the language system, gram- mars and dictionaries contain abstractions based on a variety of analyti- cal procedures. With the data contained in grammars and dictionaries, most aspects of the analyses underlying the abstractions are not verifi- able or replicable. There is no way of knowing whether fundamental mistakes have been made unless the primary data on which the analyses build are made available in toto as well. c. Grammars usually only contain statements on grammatical topics which are known and reasonably well understood at the time of writing the grammar. Thus, for example, grammars written before the advent of modern syntactic theories generally do not contain any statements re- garding control phenomena in complex sentences. Many topics of cur- rent concern such as information structure (topic, focus) or the syntax and semantics of adverbials have often been omitted from descriptive grammars due to the lack of an adequate descriptive framework. As pointed out in particular by Andrew Pawley (1985, 1993, and elsewhere), there is a large variety of linguistic structures often subsumed under the heading of speech formulas which do not really fit the structuralist idea of a clean divide between grammar and dictionary and thus more often than not are not adequately documented in these formats. d. Grammars and (to a lesser extent) dictionaries provide little that is of direct use to non-linguists, including the speech community, educators, and researchers in other disciplines (history, anthropology, etc.). These points of critique mostly pertain to the fact that structuralist language descriptions are reductionist with regard to the primary data on which they are based and do not provide access to them. Or, to put it in a slightly dif- ferent and more general perspective, they document a language only in one of the many senses of “language”, i.e. language as an abstract system of rules and oppositions. Inasmuch as structuralist language descriptions are intended to achieve just that, the above “critique” is, with the possible ex- ception of point (b), not fair in that it targets goals for which these descrip- tions were not intended.12 In this regard, it should be emphasized that the above points in no way question the usefulness and relevance of descriptive grammars and diction- aries with regard to their main purpose, i.e. to provide a description and documentation of a language system. While there is always room for im- provement (compare points (b) and (c) above), there is no doubt about the fact that grammars and dictionaries are essentially successful in delivering
- 34. 20 Nikolaus P. Himmelmann system descriptions. What is more, the above points also do not imply that grammars and dictionaries do not have a role to play in language documen- tations, as further discussed in the next section. The major thrust of the critical observations above is that a description of the language system as found in grammars and dictionaries by itself is not good enough as a lasting record of a language, even if accompanied by a text collection. And it is probably fair to say that the way primary data have been handled in the grammar-dictionary format is now widely seen as not adequate and thus in need of improvement. From this assessment, however, it does not necessarily follow that the basic format of Table 1 is the only imaginable format for lasting, multipur- pose records of a language. Instead, it may reasonably be asked, why not combine the strong sides of the two formats discussed so far and propose that language documentations consist of the combination of a large corpus of annotated primary data as well as a full descriptive grammar and a com- prehensive dictionary? This is the question to be addressed in the next sec- tion. 4.2. An extended format for language documentations Assuming that the structuralist notion of a language as a system of rules and oppositions is a viable and useful notion of “a language”, though not necessarily the only useful and viable one for documentary purposes, and assuming further that a descriptive grammar and a dictionary provide ade- quate representations of this system, it would seem to follow that a truly comprehensive language documentation does not simply consist of a large corpus of annotated primary data – as sketched in Section 3 – but instead should also include a comprehensive grammar and dictionary. Along the same lines, one may ask why the apparatus in Table 1 should only contain a sketch grammar and not a fully worked-out comprehensive grammar, thus replacing the format in Table 1 with the one in Table 2.13
- 35. Chapter 1 – Language documentation: What is it and what is it good for? 21 Table 2. Extended format for a language documentation Primary data Apparatus Per session For documentation as a whole Metadata General access resources – introduction – orthographical conventions – glossing conventions – indices – links to other resources … recordings/records of observable linguistic behavior and metalinguistic knowledge Metadata Annotations – transcription – translation – further linguistic and ethnographic glossing and commentary Descriptive analysis – ethnography – descriptive grammar – dictionary The difference between the basic format for language documentations in Table 1 and the extended format depicted in Table 2 pertains to the addition of fully worked out descriptive analyses on various levels (as indicated by the shaded area in Table 2), replacing the corresponding sketch formats (sketch grammar, ethnographic sketch) under general access resources in the basic format. Whether this is in fact a fundamental difference or rather a gradual difference in emphasis, is a matter for further debate. In actual practice, the difference may not be as relevant as it may appear at first sight, as we will see at the end of this section. Still, in the interest of making clear what is involved here, it will be useful to highlight the differences between the two formats and to indicate some of the problems that are created by incorporating comprehensive descriptive formats in the extended documen- tary format. There are at least two types of such problems, one relating to theoretical issues, the other to research economy. The theoretical problem pertains to the fact that it is not at all clear how exactly the descriptive grammar (or the ethnography or the dictionary)14 should look that is to be regarded as an essential part of a language docu- mentation. As is well known, for much studied languages such as English,
- 36. 22 Nikolaus P. Himmelmann Latin, Chinese, Arabic, Tagalog, Quechua, or Fijian, there exist not only different types of grammars (pedagogical, historical, descriptive) but also different descriptive grammars, each having its particular emphasis and way of presenting the structure of the language system. This simply reflects the fact that at least according to the current state of knowledge, there is not just exactly one descriptive grammar which correctly and comprehensively captures the system of a language. Instead, any given descriptive grammar is a more or less successful attempt to capture the system of a language (variety), rarely if ever comprehensive, and usually also including at least some contested, if not clearly wrong, analyses. As a consequence of this state of affairs, the following problem arises with regard to the extended format for language documentations in Table 2. Either one has to specify a particular type of descriptive grammar as the one which is the most suitable one for the purposes of language documenta- tions and thus is able to provide a reasonably precise definition of this part of a documentation. Alternatively, one allows for a multitude of descriptive grammars to be included in a documentation, thus declaring it a desirable goal to include a number of different analyses of the language system as part of the overall documentation of a language. The latter option clearly raises the issue of practical feasibility, which leads us to the second problem men- tioned above, i.e. the essentially pragmatic problem of research economy. Practical feasibility also is an issue if just one analysis of the grammati- cal system is assumed to be an essential part of a language documentation, for the following reason. It is a well-known fact that it is possible to base elaborate descriptive analyses exclusively on a corpus of texts (either texts written by native speakers or transcripts of communicative events) – and most good descriptive grammars are based to a large degree on a corpus of (mostly narrative) texts. A large corpus of texts in fact provides for the pos- sibility of writing a number of interestingly different descriptive grammars, targeting different components of the language system and their interrela- tion. Consequently, one could argue that even if one accepts the claim that a comprehensive documentation should also document the language system, there is no need to include a fully worked-out descriptive grammar in a language documentation. The information needed to write such a grammar is already contained in the corpus and the resources needed to extract this information and to write it up in the conventional format of a descriptive grammar are not properly part of the documentation efforts. In this view, resources allocated to documentation should not be “wasted” on writing a grammar but are better spent on enlarging the corpus of primary data, the
- 37. Chapter 1 – Language documentation: What is it and what is it good for? 23 quantity or quality of annotations, or on the “mobilization” of the data (mo- bilization is further discussed in Chapter 15). The major counterargument against this position would be the claim that actually producing a descriptive grammar is a necessary part of a language documentation because otherwise, essential aspects of the language system would be left undocumented. The evaluation of this claim rests on the ques- tion of whether there is some kind of important evidence for grammatical structure which, as a matter of principle, cannot be extracted from a suffi- ciently large and varied corpus of primary data as sketched in Section 3 above. As far as I am aware, there is especially one type of evidence of this kind, i.e. negative evidence. Obviously, illicit structures cannot be attested even in the largest and most comprehensive corpora.15 However, the lack of explicit negative evidence in a corpus of texts does not per se necessitate the inclusion of a descriptive grammar in a language documentation. On the one hand, with regard to the usual way of obtaining negative evidence (i.e. asking one or two speakers whether examples x, y, z are “okay”), it is doubtful whether this really makes a difference in quality compared to evidence provided by the fact that the structure in question is not attested in a large corpus. Elicited evidence is only superior here if it is very carefully elicited, paying adequate attention to the sample of speakers interviewed, potential biases in presenting the material, and the like. On the other hand, and more importantly, the basic documentation format of Table 1 does not only consist of a corpus of more or less natural communicative events but also of documents recording metalinguistic knowledge. Metalin- guistic knowledge includes negative evidence for grammatical structuring, as already mentioned above. Obviously, gathering negative evidence on grammatical matters presup- poses that the researcher asks the right questions, which in turn presupposes grammatical analysis. In this regard, it bears emphasizing that documenta- tion does not exclude analysis. Quite the opposite: analysis is essential. What the documentary approach implies, however, is that the analyses which are carried out while compiling a documentation do not necessarily have to be presented in the format of a descriptive grammar. Instead, analyses can (or should) be included in a documentation through (scattered) annotations on negative evidence, the inclusion of experiments generating important evi- dence for problems of grammatical or semantic analysis, and so on (see further Chapters 8 and 9). The major reason for choosing a distributed grammatical annotation format instead of the established descriptive grammar format is one of time
- 38. 24 Nikolaus P. Himmelmann economy. The writing of a descriptive grammar involves to a substantial degree matters of formulation (among other things, the search for the most suitable terminology) and organization (for example, chapter structure or the choice of the best examples for a given regularity; see Mosel 2006 for further discussion and exemplification). These are very time consuming activities which in some instances may enhance the analysis of the lan- guage system, but in general do not contribute essential new information on it. Thus, with regard to the economy of research resources, it may be more productive to spend more time on expanding the corpus of primary data rather than to use it for writing a descriptive grammar. In short, then, the difference between the basic and the extended formats as conceived of here is one between different formats or “styles” for the inclusion of analytical insights in a documentation. In the basic format, analyses are included in the form of scattered annotations and cross- references between sessions (and, of course, indirectly also by the fact that for topics for which little or no data can be found in the recordings of communicative events, elicited primary data are included). In the extended format, analyses are presented as such in full, i.e. as descriptive statements about the language system, usually accompanied by (links to) relevant ex- amples. In actual practice, there will be many instances where this apparently clear difference will become blurred. For example, when the number and types of communicative events that can be recorded in a given community is severely limited, it may be more useful to work on full, and fully explicit, descriptions of aspects of the grammatical system not represented in the texts, rather than recording more texts of the same kind with the same speaker. Furthermore, on a much more mundane level, there are (individu- ally widely diverging) limits as to the time and energy that can be produc- tively spent on the not always thrilling routine work involved in documen- tation (filling in metadata, checking translations and glossing, etc.), and it would be a counterproductive and rather ill-conceived idea generally to restrict work with a speech community to “pure” documentation to the ex- clusion of all fully explicit (= publishable) analytic work. It is thus unlikely that linguists undertaking language documentations will stick to the basic format in its purest form and refrain from working on aspects of a fully explicit descriptive analyses while compiling the annotated corpus of pri- mary data. It should, then, also not come as surprise that many researchers – including some of the contributors to this volume – tend to ignore the difference between the two formats and to remain implicit as to what ex-
- 39. Chapter 1 – Language documentation: What is it and what is it good for? 25 actly they have in mind when referring to grammatical analyses and dic- tionaries. Most language documentations that have been compiled in recent years are actually hybrids with regard to the two formats. They tend to include many scattered analytical observations as well as substantial fully worked- out descriptive statements of some aspects of the language system (rarely comprehensive grammars). It remains to be seen whether this practice is actually viable in the long-term or whether there are clear advantages at- tached to adhering to either the basic or the extended format as discussed in this section. 5. The structure of this book The following chapters provide in-depth discussions and suggestions for various issues arising when working on and with language documentations. While the authors have slightly different views of what a language docu- mentation is (or should be) and clearly differ with regard to their major topics of interest and theoretical preferences, they share a major concern for the maintenance of linguistic diversity, including the quality, processing, and accessible preservation of linguistic primary data, which in some way or other all these chapters are about. The focus of each chapter is on a topic which is rarely dealt with within descriptive linguistics (and mainstream linguistics in general), reflecting the fact that issues relating to the collection and processing of primary data have been widely neglected within the discipline until very recently. For each topic, both theoretical and practical issues are discussed, although the chapters differ quite significantly as to how much space they allot to either, in accordance with the topic being dealt with. Apart from the present introduction, there are roughly four parts to this book which, however, are closely linked to, and overlap with, each other. Chapters 2 to 4 deal with general (i.e. not specifically linguistic) ethical and practical issues which have to be considered and reconsidered from the earliest planning stage of a documentation project through to its completion. The guiding questions here are: How to interact with speech communities and individual speakers; and how to capture, store, and process relevant data. These issues are interrelated, in that data capture and processing is not just a technological issue, but also has to pay attention to sensitivities and interests of the speech community and the individual speakers contributing
- 40. 26 Nikolaus P. Himmelmann data. Chapter 3 includes suggestions for getting started with the actual lin- guistic documentation work in the field. The next eight chapters (Chapters 5 to 12) pertain to the recording and processing of primary linguistic data from an anthropological and linguistic point of view. The first three of these chapters (Chapters 5 to 7, but also a considerable part of Chapter 8) are primarily concerned with the issue of how and what to document, given the goal of creating a lasting and multi- functional record of a language. Chapter 5 provides an introduction to a cultural and ethnographic understanding of language. This is essential for the success of a documentation project, not only with regard to the neces- sity of being able to identify the types of communicative events that should be recorded, but also for being able to successfully interact within a speech community which has a different set of norms of interaction. In the latter regard, Chapter 5 complements and expands Chapters 2 and 3. Chapter 6 addresses the issue of how to access and represent meta- linguistic knowledge, focusing primarily on lexical knowledge. Chapter 7 briefly discusses the kinds of data needed for prosodic analysis, while Chapter 8 reports on the demands of anthropologists for language docu- mentations, which complements the discussion of this topic in Chapter 5. Chapter 8 also addresses the issue of ethnographically relevant annota- tion and commentary and thus forms a group with the next four chapters (Chapters 9 to 12) all of which are concerned with the part of a documenta- tion called “apparatus” in Table 1. That is, they deal with the processing of primary data necessary for them to become useful and accessible to a broad range of users. While Chapters 8 and 9 provide an overview of the basic structure and various practical aspects of ethnographic and linguistic anno- tation and commentary, respectively, the following two chapters address some more specific issues with regard to the written representation of re- corded communicative events. Chapter 10 is concerned with one major aspect of transcription, namely, the need to segment the continuous flow of spoken language into smaller units, in particular words and intonations units. Issues relating to the development of a practical orthography which can be used for the written representation of the recordings, for educational materials, etc., and which is acceptable and accessible to the speech com- munity are discussed in Chapter 11. The final chapter in this part of the book, Chapter 12, discusses the structure and format of the sketch grammar which is part of the overall apparatus of the documentation, intended to facilitate access to the primary data themselves as well as the grammatical information to be found in sessions and lexical database.
- 41. Chapter 1 – Language documentation: What is it and what is it good for? 27 The last part of the book, consisting of the final three chapters, relates to the long-term perspectives of a documentation, in particular, archiving issues and its use in language maintenance. Apart from an obvious focus on tech- nological issues, the main concern of Chapter 13 on “Archiving challenges” is a critical review of the different interests and goals of the three major groups involved in the archiving process: the donators (the people handing material to the archive), the archivists (the people running and maintaining the archive), and the users of archival sources. Chapter 14 takes up one particularly critical issue in long-term preservation, i.e. the changing stan- dards in character and text structure encoding which very easily render digitally-stored information uninterpretable. Finally, Chapter 15 focuses on speech communities as potential users and argues that there is a need for elaborate and creative concepts for mobilizing primary data, i.e. creating language resources from archival data which are of interest and use to a given community. There are a number of important topics which actually should also be dealt with in a book such as the present one but which unfortunately and for reasons beyond the control of the editors could not be included at this point. In particular, the following three topics are also of critical importance to language documentation (see the book’s website for additional and up-to- date information on these and other topics). – One major aspect of linguistic interactions which has to be attended to in documentations are so-called paralinguistic features, in particular ges- ture. The recent textbook on gesture by Kendon (2004) provides a thor- ough general introduction to this topic. See also Section 2.5 in Chapter 9 for a brief note on paralinguistics. – There is no chapter on the basics of producing high-quality audio and video recordings. While this topic in part involves a lot of technological aspects which change rather rapidly and thus would in any event not have been included in this book, there is a need to be aware of what de- fines good recordings. In addition to the book’s website, see the Lan- guage Archiving Newsletter and the DoBeS and ELDP websites for relevant pointers and links. – Apart from the kind of mobilization of primary data for language main- tenance purposes discussed in Chapter 15, there are also more traditional, but equally important contributions that a language documentation can make to language maintenance efforts. These include, in particular, the development of teaching materials in the documented variety. See von Gleich (2005) for a brief discussion and references.
- 42. 28 Nikolaus P. Himmelmann The book is also heavily biased towards the more narrowly linguistic ap- proaches to language. Documentary work that aims at a truly comprehensive record of a language also has to engage with ethnobotany, musicology, human geography, oral history, and so on. We hope that it will be possible before too long to compile a further introductory volume where the core issues and methodologies of these and related disciplines are presented from the point of view of enhancing language (and culture) documentations. Even though the focus is on linguistic approaches to language, it should be clearly understood that even for this domain the ability to engage in lan- guage documentation projects cannot be gained by mastering only the topics and techniques presented here. Ideally, training in language documentation includes a training in the basics of a broad range of linguistic subdisciplines and neighboring disciplines. Training in descriptive and anthropological linguistics is indispensable. The latter two topics are not dealt with here because good textbooks for them are readily available. As for descriptive linguistics, the classic text- books by Hockett (1958) and Gleason (1961) still provide an excellent in- troduction which, however, should be complemented by typologically grounded surveys of major categories and structures as, for example, in the second edition of Shopen’s Language Typology and Syntactic Description or in Kroeger (2005). As for anthropological linguistics, Duranti (1997) introduces the most important concepts and issues, which could be com- plemented with the more in-depth discussion of the ethnography of com- munication by Saville-Troike (2003). Finally, the contributions in Newman and Ratliff (2001) combine descriptive and ethnolinguistic topics and in- sights and complement the discussion of linguistic fieldwork in Chapters 2 and 3 of this volume. In conclusion, it may be worthwhile to emphasize the fact that docu- mentary linguistics is an emerging field where many things are still in flux. Most importantly perhaps, large multimedia corpora on lesser-known lan- guages are very new and largely unexplored entities. It is very well possible that new techniques for working with such corpora will emerge before too long, requiring major adjustments to the format for language documenta- tions discussed in this chapter and book. But rather than a shortcoming, this should be seen as one of the exciting aspects of language documentation. Apart from being a useful introduction to language documentation, provid- ing theoretical grounding as well practical advice, this book should make it clear that language documentation is an important, engaging and rewarding enterprise with many repercussions for linguistics and other language- related disciplines and projects.
- 43. Chapter 1 – Language documentation: What is it and what is it good for? 29 Acknowledgements I wish to thank my co-editors and Eva Schultze-Berndt for critical discus- sion of many of the issues touched upon here, as well as helpful comments on the draft version of this chapter. Notes 1. With regard to the latter point, compare the following quote from Luraghi (1990: 128 FN1) which nicely illustrates the problems arising when data types are missing in a given corpus: “As to the position of the verb, the most impor- tant difference [between main and subordinate clauses, NPH] lies in the ab- sence of VSO sentences in subordinate clauses. It can of course be objected that this may be due simply to the shortage of sources, since VSO sentences are on the whole very infrequent. However, in the light of comparative data from other Indo-European languages, this objection could perhaps be rejected …” 2. The major limitation here are restrictions on access to recordings imposed by speakers or communities which, of course, should be observed. 3. “Experiment” here is to be taken in a broad sense, including, for example, the testing of the acceptability of invented examples. 4. IMDI = ISLE Metadata Initiative. The manual can be downloaded at http:// www.mpi.nl/IMDI/tools. 5. Note that this does not necessarily imply that all the information for a lexical item has to be gathered in a single location (i.e. an entry in the database), as it is currently done by most researchers. Alternatively, the lexical database could consist simply of links to all the sessions where the item in question occurs. This could include a session where the item is elicited as part of the elicitation of a word list or semantic field, a session where the item has been recorded in a list of items or a carrier phrase in order to document characteristic sound pat- terns, and a session where it occurs as part of a procedural text. 6. Please refer to the appendix for further information on this program. 7. Note that linguistic interaction here includes interactions with native speakers of other varieties inasmuch as they are a common occurrence in the speech community which is being documented. 8. The following list takes an audio or video recording as its main example. Of course, the same type of metadata is needed for primary data gathered in a dif- ferent way such as written fieldnotes or photos. 9. Note that the term cataloguing is used here in a somewhat broader sense than in Chapter 4 where it is used to refer to one particular subtype of metadata. 10. Strictly speaking, “annotations” could also be called metadata since the term “metadata” in general refers to all kinds of data about data. However, within the
- 44. 30 Nikolaus P. Himmelmann context of language documentations it is useful to distinguish between different types of metadata (in this broad sense), and it is now a widely-used practice to use the term “metadata” in the context of language documentations exclusively for data types which have a cataloguing or organizational function and to use “annotation” (or “commentary”) for other types of information accompanying segments of primary data. 11. The structuralist idea of language as an abstract system has been articulated in a variety of oppositions including the well-known Sassurean distinction of langue vs. langage vs. parole and the Chomskyan distinction of competence vs. performance. For the present argument, the details of how the abstract lan- guage system is conceived of do not matter and thus are ignored. 12. With regard to falsifiability (point (b)), not providing access to the primary data is indeed a major problem for the scientific status of these descriptions. However, the basic assumption here appears to have been that whoever wanted to replicate and possibly falsify a descriptive analysis on the basis of material other than the one made available in examples and texts could compile their own set of primary data. This assumption is no longer viable in the case of en- dangered languages and, as already pointed out in Section 2, it is hence not by chance that a close connection exists between language endangerment and the recent increased concern for the preservation of primary data in linguistics and related disciplines. 13. The part called “descriptive analysis” in the rightmost column could also be added in other ways to the overall format, for example as an additional column of its own, on a par with “primary data” and “apparatus”. While there are theo- retical issues associated with these alternative overall organizations, these do not play a role for the argument in this section and hence can be safely ignored. 14. Essentially the same points made here and in the following with regard to de- scriptive grammars could also be made with regard to conventional dictionaries and ethnographic monographs (see Chapter 6 for a brief discussion of different types of dictionaries, which is also relevant here). Including these two other main analytical formats in the discussion would, however, unnecessarily com- plicate the exposition. Hence, dictionaries and ethnographies are not further discussed in this section. The choice of descriptive grammars as the main ex- ample is simply due to the fact that it is the format the author is most familiar with. 15. Very occasionally, though, especially in the interaction between parents and children, unacceptable or highly marked structures might be attested in admon- ishments of the form: Don’t say X, say Y.
- 45. Chapter 2 Ethics and practicalities of cooperative fieldwork and analysis Arienne M. Dwyer Introduction This chapter examines central ethical, legal, and practical responsibilities of linguists and ethnographers in fieldwork-based projects. These issues span all research phases, from planning to fieldwork to dissemination. We focus on the process of language documentation, beginning with a discussion of common ethical questions associated with fieldwork: When is documenta- tion appropriate in a particular community, and who benefits from it? Which power structures are involved, both in and out of the field? Section 1 ex- plores key concepts of participant relations, rights, and responsibilities in fieldwork in the context of ethical decision-making. It introduces a set of guiding principles and examines some potential pitfalls. Section 2 discusses the legal rights issues of data ownership (intellectual property rights and copyright) and data access. Such information aids planning before field- work and especially the archiving phase. Sections 3 and 4 cover the more concrete practical aspects of the field- work situation: developing a relationship with a speech community and organizing and running a project. We survey what may be termed “the five Cs” critical to planning and executing a project: criteria (for choosing a field site), contacts, cold calls, community, and compensation. Finally, since even the best-planned projects encounter logistical and interpersonal chal- lenges, we present several generic case studies and some possible methods of resolving such disputes. Such ethical and logistical planning is essential to successful commu- nity-centered knowledge mobilization, from which documentation products useful for both academics and community members are produced in an environment of reciprocity. It is the linguist’s responsibility to focus on process (Rice 2005: 9)1 as much as the end goals.
- 46. 32 Arienne M. Dwyer 1. Ethics 1.1. Research as mediation Ethical behavior is often assumed to flow intuitively from the noble goals of scientific research. Most fieldworkers consider themselves well-inten- tioned, rational people. But have all participating individuals and groups been considered in these research goals? Have their ethical standards been considered? Fieldwork methodology has in the last decades progressed from a typi- cally non-cooperative model (research on a community) to a cooperative model which in its strongest form explicitly empowers speech communities (research on, for, and with a community) (Cameron et al. 1992: 22–24). Assumptions about what is ethical for a particular field situation are best avoided, especially assumptions on the part of the researcher about what participants want.2 The researcher should also have a grasp of the legal implications (local, national, and international) of data ownership.3 An un- derstanding of ethical and legal responsibilities also facilitates the building of trust – and thus a successful relationship – with a community research team. Finally, making ethical and legal premises explicit, helps to anticipate and avoid problems. A field researcher mediates between speakers, their communities and the fieldworker’s own community, which includes an institution, a funding body, and possibly an archive. Inevitably, all partici- pants in a language documentation will face ethical dilemmas, in which no course of action seems quite satisfactory. There may be “no right decision, only… [one] ‘more right’ than the alternatives” (Hill, Glaser and Harden 1995: 19). Distilled to its essence, the ethics of field research entails indigenous people and field researchers mediating each other's cultural imperatives. This contextualization of ethical principles can only occur through produc- tive mutual negotiation at the local level. The ethical principles presented here may seem as both imperious and overly generic, given that in this chapter broad-brush principles are often preceded by the cajoling impera- tive should or the bossy must. But these are suggestions awaiting contextu- alization in a particular research situation. And this mediation of ethical principles by all participants forms the nucleus of any research project.
- 47. Chapter 2 – Ethics and practicalities of cooperative fieldwork and analysis 33 1.2. Normative ethics The ethical decisions made during fieldwork belong to the domain of pro- fessional ethics. Since many field research networks also create codes of conduct, we are also concerned here with normative ethics. Normative practices attempt to prescribe best-practice standards for field situations. A research team might make the normative decision to adhere to a de- tailed set of ethical principles determined in advance, asking “is our aim just to evaluate the resolution of past ethical dilemmas in the field by con- sensus?” Normative guidelines generally follow a deductive or an inductive approach. Some researchers review such a list of field experiences and at- tempt to achieve consensus on future ethical research behavior. Another less normative approach might simply be to observe and note the ethical dilemmas that appear. This descriptive list of relevant field di- lemmas and how they were resolved could serve as a reference for future field researchers. An example of a less normative approach is the “do no harm” credo discussed below. The dangers of excessive normativity are well-known; colonial subjuga- tion, religious or cultural conversion-induced linguicide, and business profit are all examples of normative frameworks which are tendentially destruc- tive. Such frameworks are assumed by their proponents to be universally held, and universally beneficial. 1.2.1. Documenting endangered languages as a normative framework Claiming that languages should be documented before they disappear is also a normative act, and it is a framework in which not everyone believes.4 But most researchers strongly support the documentation of endangered languages, arguing that a decline in linguistic diversity constitutes a decline in specific forms of knowledge and expression. Speakers of endangered languages also often support such a normative framework, since language is a central part of culture and of ethnic identity. Should a language be documented when its speakers would prefer it to disappear? How should community priorities and external western-scientific priorities be weighed? Many would argue that documentation should make the language available to future generations; most would also argue that both sets of priorities should be accommodated, to the extent possible.
- 48. 34 Arienne M. Dwyer 1.2.2. Balancing priorities Since field linguistic situations are so diverse, one-size-fits-all codes of conduct are impractical. Codes of conduct are voluntary and often largely unenforceable, but good guidelines help ensure good working relationships and a positive research outcome. For the sake of methodological transpar- ency, and for smooth communications between all parties, some norms are always part of the field experience. Most research teams choose a pragmatic approach, making use of both explicit ethical guidelines as well as drawing observations from specific field experiences.5 No matter what form is chosen, research teams would do well to make explicit the ethical norms of their particular project. 1.2.3. Normative ethics in language documentation Individual teams should establish a code of ethical norms specific to their particular area for a given research project. This code would encompass detailed guidelines on consultation and negotiation between indigenous people and researchers for all phases of the research, including planning and dissemination. Since such voluntary normative approaches have proven useful, the sci- entific community can aim at establishing a two-tiered, flexible ethical code for linguistic field research: a generic code of putatively universal ethical norms, and as above a specific individual code for a research on an ethnic group in a particular area, created by individual researchers. At present, linguists lack a generic code of conduct. Ideally, field lin- guists will work with the country’s linguists and social scientists to devise this generic code. This code would be specific for field linguistics but could be modelled on existing well-articulated guidelines (such as the Australian Institute of Aboriginal and Torres Strait Islander Studies’ Guidelines for ethical research in indigenous studies [AIATSIS 2000], the African Studies Association’s Guidelines for ethical conduct in research and projects in Africa [African Studies Association n.d.], and the American Anthropological Association’s Code of ethics [AAA 1998]). Though the above are designed as regional codes, they are actually generic enough as to be potentially ap- plicable to any world region. A generic statement on ethical principles should address all phases of re- search: planning, fieldwork, analysis, archiving, and end products. Planning ethically for each phase entails assessing the roles played by participants
- 49. Chapter 2 – Ethics and practicalities of cooperative fieldwork and analysis 35 and the potential benefits and detriments of research; it also ideally includes local participants’ participation at every phase. In the planning phase, re- searchers should identify all the potential participants (see Section 1.3 be- low), including sponsoring institutions, and estimate remuneration for local participants. During fieldwork, the researchers establish and maintain rela- tionships, and negotiate contracts or protocols for obtaining data. It is at this crucial phase that the researchers must obtain informed consent (see Sections 1.5 and 2.2.1 below). The analysis phase includes such normative ethical decisions as the number of minimally adequate levels of annotation. Annotation decisions are questions of ethics, as what annotation is included will determine the accessibility of the materials to particular audiences.6 During the archiving phase, the researcher must carry through the wishes of consultants in terms of anonymity and recognition by making speakers anonymous; decisions must be taken on user access to the materials (com- munity, scientific researchers, general public) and which materials are to be accessed. In the longer term, such codes of conduct could be developed for spe- cific regions (countries or ethnolinguistic areas), based on a comparison of individual codes of conduct from the same area. This would result in a third tier of guidelines, a regional code. Though regional codes are the least critical of the three types of guidelines, such a code would outline certain region- or country-specific practices spanning a number of ethnic groups for a given area, e.g. archival practices for material from a consultant who passed away since the data collection. 1.3. Players The practical application of ethical principles entails the specification of ethical and legal relationships between all participants in the documentation process. These relationships should be made explicit and clearly differenti- ated. First, consultants (speakers/singers) are part of a certain sociocultural context in a certain country (see Figure 1). The sociocultural context con- sists not only of the speaker community itself but its relationship nested within local society. Then, the interaction between researcher(s) and con- sultant(s) occurs within a regional and national context, which includes governments, officials, subject experts, and eventually users of the analyzed data. Speaker-consultants are part of both linguistic and administrative com- munities; language communities are usually part of larger ethnolinguistic or
- 50. 36 Arienne M. Dwyer ethnoreligious regions. These regions, in turn, may be contiguous with or reach across provincial or national boundaries. The roles and perspectives of participants are gradient and dynamically created. We can use “insider/outsider” as shorthand to describe two ex- tremes of how a researcher situates himself or herself with regards to the research situation, as well as how other participants view that researcher. The researcher might be an insider (i.e. accepted as a member of that com- munity) or an outsider (from a distant community, whether in that country or in another). These roles are gradient rather than absolute, since a foreign researcher and a native speaker from a distant community may both be con- sidered “outsiders” from the community under investigation. A local re- searcher often assumes multiple insider/outsider roles: it is often the case that a researcher is part of the ethnolinguistic group, but not or no longer from the particular community. In this situation, that researcher is both an insider and an outsider. The distinction may be relevant for research plan- ning, as it often facilitates research to work with a person from the actual community under investigation. Furthermore, researchers’ institutional connections play an important role in determining both the direction and scope of the research. Every in- stitution has its own agenda. If a researcher is funded by a university in that nation’s capital, for example, in some cases he/she might be expected to produce a study that enhanced that country’s ethnic policy. A researcher from overseas might, in contrast, be subtly pressured by the home univer- sity or the funding agency to quickly obtain a lot of data and produce publi- cations, while overlooking the need for reciprocity with the speech com- munity. Creating research products useful to communities is an issue which will become more and more central to the ethical practice of the research enterprise, though currently grant funding is mostly limited to products for a scientific audience. Institutional affiliations almost invariably insinuate themselves into the power relationships between players. Though outsiders may be regarded with more suspicion than insiders, the affiliations of outsiders generally are seen as prestigious. Usually enhancing this prestige is the economic means of the researcher as a result of the funding. Then in this web of relations there is the archive, in which the researcher deposits his or her materials. Though requirements of the granting agency vary, each has specific guidelines for data depositing and use. Finally, the archive disseminates data to users.
- 51. Chapter 2 – Ethics and practicalities of cooperative fieldwork and analysis 37 That these players – individual fieldworkers, communities, research consor- tia, funding agencies, archives and users – may all be located in different countries has legal implications for the storage, ownership, transfer, and publication of the data (see Section 2 below). But more important to the success or failure of a given research collaboration are the shifting and highly contextual nets of power and belonging (insider/outsider) between these players. A research project on any scale would do well to evaluate both these legal and social relationships in the planning stage. insider/outsider insider/outsider Figure 1. Participants in linguistic fieldwork (adapted from Hiß 2001, Wittenburg 2001–2004) 1.4. Ethical principles Heritage can never be alienated, surrendered or sold, except for conditional use. Sharing therefore creates a relationship between the givers and receiv- ers of knowledge. The givers retain the authority to ensure that knowledge is used properly and the receivers continue to recognize and repay the gift. (Daes 1993: 9) We can outline the following five fundamental ethical principles for lan- guage documentation: country government: national, local experts: national, local users country (country) community consultant funding organization (country) institution researcher institution archivist country user
- 52. 38 Arienne M. Dwyer Principle 1: Do no harm (including unintentional harm) Though inarguable, this maxim requires individuals to specify what “harm” means in the specific local context. Since research is a kind of prying, pro- tecting privacy largely concerns deciding which information to protect from public view. Harm to privacy may come from revealing information that discredits a person (Thomas and Marquart 1987: 90). There are, of course, many kinds of inadvertent harm. For example, publicizing one person’s name might result in embarrassment, whereas not publicizing another’s name may be viewed as a slight. Moreover, the people with whom an outsider-researcher associates could be stigmatized by the community for giving away cultural or even national security secrets, for example, which might lead to trouble with community leaders or police. Also, since many researcher-consultant exchanges involve compensation, unintentional harm can be caused by arousing financial or material envy in the indigenous community. Part of fairness is being attentive to relative compensation: what one person acquires in material or political gains as a result of participation may cause envy or ill will in others in the community. Such attentiveness re- quires researching not only what is the appropriate form of compensation (e.g. money, goods, recognition) and the appropriate amount, but also re- quires knowledge of project participants’ status in and relationship with the community (see Section 3.5). Gifts or payments of goods or money, where culturally appropriate, compensate for both the expertise of another individual and the inconven- ience caused him or her. Even where no overt compensation changes hands, the core participants create a dynamic of reciprocity, whereby the gift of language knowledge is reciprocated by the researcher in some way, e.g. by compiling a community course book. After all, the term compensation lit- erally means ‘hanging together.’ Underlying this equilibrium is the second principle that we might simply articulate as: Principle 2: Reciprocity and equity The research relationship must be consultative, continuously negotiated, and respectful. Accommodate community input into your research goals, or, better yet, plan the research collaboratively with the indigenous com- munity. Re-negotiation of methodologies and goals is a normal part of this process. Part of the culture of respect is acknowledging that one’s view- points may not be universally held. The researcher should also respect both
- 53. Chapter 2 – Ethics and practicalities of cooperative fieldwork and analysis 39 the indigenous knowledge system under study and respect the confidence and trust of individual participants. One area of normative ethics that modern researchers generally think of right away is the idea of “giving something back” to the community. This notion is not altruistic, but rather reflects the consideration that when re- searchers enter a community, they disturb it at least temporarily, and also take data away. Even with compensation, research behavior is nearly al- ways a lopsided proposition, with clear benefits accorded more to the re- searcher than the community. Thus, many researchers in recent years have come to feel strongly that they should additionally compensate communi- ties with scientific products or even economic development aid. Therefore, our generic code also includes: Principle 3: Do some good (for the community as well as for science) What constitutes a generous act of “giving back” varies greatly depending on community needs. Such acts are more abstract than mere compensation for a consultant’s time; they are also never 1:1, in the sense that a re- searcher can never repay a community for the rich but nonetheless snap- shot-like view of the culture obtained during a particular field research ex- perience. The most common examples of “giving back” include preparing peda- gogical and cultural materials useful to the community, such as promulgat- ing an orthography, developing textbooks and primers, making audio CDs, VCDs and documentary film, and creating picture books on material culture, e.g. embroidery or architecture. Principle 4: Obtain informed consent before initiating research It is critical for the researcher to establish an agreement with data producers (speakers, singers and/or a community) to record, archive and disseminate these data. Researchers are ethically obligated to inform data producers of all possible uses of the data so as to implement the do no harm principle above. Permission should be recorded in a culturally appropriate form: written, video or audio-taped. A detailed discussion of the issues and pro- cedures in informed consent are found below in Section 2.2.1. Such mandatory contracts certainly encourage researchers to document permissions. However, in some local situations, unrecorded oral contracts may be most conducive to mutual trust, though they usually do not fulfil the legal requirements of IRBs (Institutional Review Boards).
- 54. 40 Arienne M. Dwyer Principle 5: Archive and disseminate your data and results Researchers must avoid being buried with their unpublished field notes and recordings. Within bounds of informed consent, those working with endan- gered-language communities have an obligation to appropriately store and publish data and analyses. Even in imperfect form, ordered, shared data are more useful than no data; disseminating or at least properly archiving col- lected data is far more respectful to a speaker community than piling it in the back of a closet. Hence, many field researchers now believe that best- practice archiving (cf. EMELD 2000–2005) and dissemination (in any for- mat) should be a requirement of fieldwork. Such principles sketch out the bare minimum in ethical linguistic fieldwork practice. For more elaborated documents, see AIATSIS (2000) and the Af- rican Studies Association (n.d.). 1.5. Potential problems: some examples 1.5.1. The observer’s paradox and covert research The requirement of obtaining informed consent rules out covert research, i.e. recording without speaker’s knowledge. The deception inherent in covert research renders it taboo for many who do fieldwork. Yet many social scien- tists routinely pretend to be ordinary citizens in order to obtain a naturalistic view of their research subjects: they, for example, join a group that believes in UFOs, work desk jobs for the sensationalist newspaper Bild Zeitung, or staff a Wal-Mart store to reveal the group or corporate practices (Wallraff 1977, Ehrenreich 2002). Such fieldworkers and journalists will vocifer- ously defend their enterprise. In anthropology and linguistics fieldwork, a researcher’s presence changes the phenomena under observation, often making conversation less sponta- neous. Most field workers simply attempt to minimize the intrusiveness of their presence (the so-called observer’s paradox [Labov 1971: 171]) by, for example, using a small recording device, or by having native-speaker insid- ers conduct the field research. These methods have provided adequate data and have been seen as ethically sound by the majority of field linguists and community researchers. However, since the observer is always intrusive to some extent, some language researchers have decided to make surreptitious recordings. This
- 55. Another Random Document on Scribd Without Any Related Topics
- 56. the sleek angora, and her desire but strengthened to possess her peer. But white angoras are a luxury, and an expensive one at that, and, however prosperous the Salon Malakoff might be, its proprietors were not as yet in a position to squander eighty francs upon a whim. So, until profits should mount higher, Madame Sergeot was forced to content herself with the voluntary visits of her neighbor's pet. Madame Caille did not yield her rights of sovereignty without a struggle. On the occasion of Zut's third visit, she descended upon the Salon Malakoff, robed in wrath, and found the adored one contentedly feeding on fish in the very bosom of the family Sergeot. An appalling scene ensued. "If," she stormed, crimson of countenance, and threatening Espérance with her fist, "if you must entice my cat from her home, at least I will thank you not to give her food. I provide all that is necessary; and, for the rest, how do I know what is in that saucer?" And she surveyed the duck-clad assistants and the astounded customers with tremendous scorn. "You others," she added, "I ask you, is it just? These people take my cat, and feed her—feed her—with I know not what! It is overwhelming, unheard of—and, above all, now!" But here the peaceful Hippolyte played trumps. "It is the privilege of the vulgar," he cried, advancing, razor in hand, "when they are at home, to insult their neighbors, but here—no! My wife has told me of you and of your sayings. Beware! or I shall arrange your affair for you! Go! you and your cat!" And, by way of emphasis, he fairly kicked Zut into her astonished owner's arms. He was magnificent, was Hippolyte! This anecdote, duly elaborated, was poured into the ears of Abel Flique an hour later, and that evening he paid his first visit in many months to Madame Caille. She greeted him effusively, being willing to pardon all the past for the sake of regaining this powerful friend.
- 57. But the glitter in the agent's eye would have cowed a fiercer spirit than hers. "You amuse yourself," he said sternly, looking straight at her over the handful of raisins which she tendered him, "by wearying my friends. I counsel you to take care. One does not sell inferior eggs in Paris without hearing of it sooner or later. I know more than I have told, but not more than I can tell, if I choose." "Our ancient friendship"—faltered Alexandrine, touched in a vulnerable spot. "—preserves you thus far," added Flique, no less unmoved. "Beware how you abuse it!" And so the calls of Zut were no longer disturbed. But the rover spirit is progressive, and thus short visits became long visits, and finally the angora spent whole nights in the Salon Malakoff, where a box and a bit of carpet were provided for her. And one fateful morning the meaning of Madame Caille's significant words "and above all, now!" was made clear. The prosperity of Hippolyte's establishment had grown apace, so that, on the morning in question, the three chairs were occupied, and yet other customers awaited their turn. The air was laden with violet and lilac. A stout chauffeur, in a leather suit, thickly coated with dust, was undergoing a shampoo at the hands of one of the duck-clad, and, under the skillfully plied razor of the other, the virgin down slid from the lips and chin of a slim and somewhat startled youth, while from a vaporizer Hippolyte played a fine spray of perfumed water upon the ruddy countenance of Abel Flique. It was an eloquent moment, eminently fitted for some dramatic incident, and that dramatic incident Zut supplied. She advanced slowly and with an air of conscious dignity from the corner where was her carpeted box, and in her mouth was a limp something, which, when deposited in the immediate centre of the Salon Malakoff, resolved itself into an angora kitten, as white as snow! "Epatant!" said Flique, mopping his perfumed chin. And so it was.
- 58. There was an immediate investigation of Zut's quarters, which revealed four other kittens, but each of these was marked with black or tan. It was the flower of the flock with which the proud mother had won her public. "And they are all yours!" cried Flique, when the question of ownership arose. "Mon Dieu, yes! There was such a case not a month ago, in the eighth arrondissement—a concierge of the avenue Hoche who made a contrary claim. But the courts decided against her. They are all yours, Madame Sergeot. My felicitations!" Now, as we have said, Madame Sergeot was of a placid temperament which sought not strife. But the unprovoked insults of Madame Caille had struck deep, and, after all, she was but human. So it was that, seated at her little desk, she composed the following masterpiece of satire: Chère Madame,—We send you back your cat, and the others —all but one. One kitten was of a pure white, more beautiful even than its mother. As we have long desired a white angora, we keep this one as a souvenir of you. We regret that we do not see the means of accepting the kind offer you were so amiable as to make us. We fear that we shall not find time to shampoo your cat, as we shall be so busy taking care of our own. Monsieur Flique will explain the rest. We pray you to accept, madame, the assurance of our distinguished consideration, Hippolyte and Espérance Sergeot. It was Abel Flique who conveyed the above epistle, and Zut, and four of Zut's kittens, to Alexandrine Caille, and, when that wrathful person would have rent him with tooth and nail, it was Abel Flique who laid his finger on his lip, and said,—
- 59. "Concern yourself with the superior kitten, madame, and I concern myself with the inferior eggs!" To which Alexandrine made no reply. After Flique had taken his departure, she remained speechless for five consecutive minutes for the first time in the whole of her waking existence, gazing at the spot at her feet where sprawled the white angora, surrounded by her mottled offspring. Even when the first shock of her defeat had passed, she simply heaved a deep sigh, and uttered two words,— "Oh, Zut!" The which, in Parisian argot, at once means everything and nothing.
- 61. T C a f f i a r d D E U S E X M A C H I N A HE studio was tucked away in the extreme upper northeast corner of 13 ter rue Visconti, higher even than that cinquième, dearly beloved of the impecunious, and of whoso, between stairs and street odors, chooses the lesser evil, and is more careful of lungs than legs. After the six long flights had been achieved, around a sharp corner and up a little winding stairway, was the door which bore the name of Pierre Vauquelin. Inside, after stumbling along a narrow hall, as black as Erebus, and floundering through a curtained doorway, one came abruptly into the studio, and, in all probability, fell headlong over a little rattan stool, or an easel, or a box of paints, and was picked up by the host, and dusted, and put to rights, and made much of, like a bumped child. Thus restored to equanimity one was better able to appreciate what Pierre called la Boîte. The Box was a room eight metres in width by ten in length, with a skylight above, and a great, square window in the north wall, which latter sloped inward from floor to ceiling, by reason of the mansarde roof. Of what might be called furniture there was but little, a Norman cupboard of black wood, heavily carved, a long divan, contrived from various packing boxes and well-worn rugs, a large, square table, a half dozen chairs, three easels, and a repulsive little stove with an interminable pipe, which, with its many twists and turns, gave one the impression of a thick, black snake, that had, a moment before, been swaying about in the room, and had suddenly found a hole in the roof through which to thrust its head. But of minor things the Box was full to overflowing. The Norman cupboard was crammed with an assortment of crockery, much of it
- 62. sadly nicked and cracked, the divan was strewn with boxes of broken pastels, paint-brushes, and palettes coated with dried colors, the table littered with papers, sketches, and books, and every chair had its own particular trap for the unwary, in the form of thumb-tacks or a glass half full of cloudy water: and in the midst of this chaos, late on a certain mid-May afternoon, stood the painter himself, with his hands thrust deep into the pockets of his corduroy trousers, and his back turned upon the portrait upon which he had been at work. It was evident that something untoward was in the air, because Pierre, who always smoked, was not smoking, and Pierre, who never scowled, was scowling. In the Quartier—that Quartier which alone, of them all, is spelt with a capital Q—there was, in ordinary, no gayer, more happy-go-lucky type than this same Pierre. He lived, as did a thousand of his kind, on eighty sous a day (there were those who lived on less, pardi!), and breakfasted, and dined, at that,—yes, and paid himself an absinthe at the Deux Magots at six o'clock, and a package of green cigarettes, into the bargain. For the rest of the time, he was understood to be working on a portrait in his studio, and, what is more surprising, often was. There was nothing remarkable about Pierre's portraits, except that occasionally he sold one, and for money—for actual money, the astonishing animal! But if any part of the modest proceeds of such a transaction remained, after the rent had been paid and a new canvas purchased, it was not the caisse d'épargne which saw it, be sure of that! For Pierre lived always for the next twenty-four hours, and let the rest of time and eternity look out for themselves. Yet he took his work seriously. That was the trouble. Even admitting that, thus far, his orders had come only from the more prosperous tradesmen of the Quartier, did that mean, par exemple, that they would not come in time from the millionaires of the sixteenth arrondissement? By no means, whatever, said Pierre. To be sure, he had never had the Salon in the palm of his hand, so to speak, but what of that? Jean-Paul himself would tell you that it was all favoritism! So Pierre toiled away at his portrait painting, and made a
- 63. little competency, but, if the truth were told, no appreciable progress from year's beginning to year's end. For once, however, his luck had played him false. The fat restaurateur, whose wife's portrait he had finished that afternoon and carried at top speed, with the paint not yet dry, to the rue du Bac, was out of town on business, and would not return until the following evening; and that, so far as Pierre was concerned, was quite as bad as if he were not expected until the following year. Pierre's total wealth amounted to one five-franc piece and three sous, and he had been relying upon the restaurateur's four louis, to enable him to fulfill his promise to Mimi. For the next day was her fête, and they were to have breakfasted in the country, and taken a boat upon the Seine, and returned to dine under the trees. Not at Suresnes or St. Cloud, ah, non! Something better than that—the true country, sapristi! at Poissy, twenty-eight kilometres from Paris. All of which meant at least a louis, and, no doubt, more! And where, demanded Pierre of the great north window, where was a louis to be found? For there was a tacit understanding among the comrades in the Quartier that there must be no borrowing and lending of money. It was a clause of their creed, which had been adopted in the early days of their companionship, for what was, clearly, the greatest general good, the chances being that no one of them would ever possess sufficient surplus capital either to accommodate another or to repay an accommodation. For a moment, to be sure, the thought had crossed Pierre's mind, but he had rejected it instantly as impracticable. Aside from the unwritten compact, there was no one of them all who could have been of service, had he so willed. Even Jacques Courbet, who possessed a disposition which would have impelled him to chop off his right hand with the utmost cheerfulness, if thereby he could have gratified a friend, was worse than useless in this emergency. Had it been a matter of forty sous—but a louis! As well have asked him for the Vénus de Milo, and had done with it.
- 64. So it was that, with the premonition of Mimi's disappointed eyes cutting great gaps in his tender heart, Pierre had four times shrugged his shoulders, and quoted to himself this favorite scrap of his remarkable philosophy,—"Oh, lala! All this will arrange itself!" and four times had paused, in the act of lighting a cigarette, and plunged again into the depths of despondent reverie. As he was on the point of again repeating this entirely futile operation, a distant clock struck six, and Pierre, remembering that Mimi must even now be waiting for him at the west door of St. Germain-des-Prés, clapped on his cap, and sallied forth into the gathering twilight. It was apéritif hour at the Café des Deux Magots, and the long, leather-covered benches against the windows, and the double row of little marble-topped tables in front were rapidly filling, as Pierre and Mimi took their places, and ordered two Turins à l'eau. A group of American Beaux Arts men at their right were chattering in their uncouth tongue, with occasional scraps of Quartier slang, by way of local color, and now and again hailing a newcomer with exclamations, apparently of satisfaction, which began with "Hello!" The boulevard St. Germain was alive with people, walking past with the admirable lack of haste which distinguishes the Parisian, or waiting, in patient, voluble groups, for a chance to enter the constantly arriving and departing trams and omnibuses; and an unending succession of open cabs filed slowly along the curb, their drivers scanning the terrasse of the café for a possible fare. The air was full of that mingled odor of wet wood pavements and horse- chestnut blossoms, which is the outward, invisible sign of that most wonderful of inward and spiritual combinations—Paris and Spring! And, at the table directly behind Pierre and Mimi sat Caffiard. There was nothing about Caffiard to suggest a deus ex machina, or anything else, for that matter, except a preposterously corpulent old gentleman with an amiable smile. But in nothing were appearances ever more deceitful than in Caffiard. For it was he, with his enormous double chin, and his general air of harmless fatuity, who edited the little colored sheet entitled La Blague, which sent half Paris into convulsions of merriment every Thursday morning, and he
- 65. who knew every caricaturist in town, and was beloved of them all for the heartiness of his appreciation and the liberality of his payments. In the first regard he was but one of many Parisian editors: but in the second he stood without a peer. Caran d'Ache, Léandre, Willette, Forain, Hermann Paul, Abel Faivre—they rubbed their hands when they came out of Caffiard's private office, and if the day chanced to be Saturday, there was something in their hands worth rubbing. A fine example, Caffiard! Mimi's black eyes sparkled like a squirrel's as she watched Pierre over the rim of her tumbler of vermouth. She was far from being blind, Mimi, and already, though they had been together but six minutes, she had noted that unusual little pucker between his eyebrows, that sad little droop at the corners of his merry mouth. She told herself that Pierre had been overworking himself, that Pierre was tired, that Pierre needed cheering up. So Mimi, who was never tired, not even after ten hours in Madame Fraichel's millinery establishment, secretly declared war upon the unusual little pucker and the sad little droop. "Voyons donc, my Pierrot!" she said. "It is not a funeral to which we go to-morrow, at least! Thou must be gay, for we have much to talk of, thou knowest. One dines at La Boîte?" "The dinner is there, such as it is," replied Pierre gloomily. "What it is now, is not the question," said Mimi, with confidence, "but what I make of it—pas? And then there is to-morrow! Oh, lala, lalala! What a pleasure it will be, if only the good God gives us beautiful weather. Dis, donc, great thunder-cloud, dost thou know it, this Poissy?" Pierre had begun a caricature on the back of the wine-card, glancing now and again at his model, an old man selling newspapers on the curb. He shook his head without replying. "Eh, b'en, my little one, thou mayest believe me that it is of all places the most beautiful! One eats at the Esturgeon, on the Seine, —but on the Seine, with the water quite near, like that chair. He
- 66. names himself Jarry, the proprietor, and it is a good type—fat and handsome. I adore him! Art thou jealous, species of thinness of a hundred nails? B'en, afterwards, one takes a boat, and goes, softly, softly, down the little arm of the Seine, and creeps under the willows, and, perhaps, fishes. But no, for it is the closed season. But one sings, eh? What does one sing? Voyons!" She bent forward, and, in a little voice, like an elf's, very thin and sweet, hummed a snatch of a song they both knew. "C'est votre ami Pierrot qui vient vous voir: Bonsoir, madame la lune!" "And then," she went on, as Pierre continued his sketch in silence, "and then, one disembarks at Villennes and has a Turin under the arbors of Bodin. Another handsome type, Bodin! Flut! What a man!" Mimi paused suddenly, and searched his cloudy face with her earnest, tender little eyes. "Pierrot," she said, softly, "what hast thou? Thou art not angry with thy gosseline?" Pierre surveyed the outline of the newspaper vender thoughtfully, touched it, here and there, with his pencil-point, squinted, and then pushed the paper toward the girl. "Not bad," he said, replacing his pencil in his pocket. But Mimi had no eyes for the caricature, and merely flicked the wine-card to the ground. "Pierrot"—she repeated. Vauquelin plunged his hands in his pockets and looked at her. "Well, then," he announced, almost brutally, "we do not go to- morrow." "Pierre!"
- 67. It was going to be much worse than he had supposed, this little tragedy. Bon Dieu, how pretty she was, with her startled, hurt eyes, already filling with tears, and her parted lips, and her little white hand, that had flashed up to her cheek at his words! Oh, much worse than he had supposed! But she must be told: there was nothing but that. So Pierre put his elbows on the table, and his chin in his hands, and brought his face close to hers. "Voyons!" he explained, "thou dost not believe me angry! Mais non, mais non! But listen. It is I who am the next to the last of idiots, since I have never a sou in pocket, never! And the imbecile restaurateur, whose wife I have been painting, will not return until to-morrow, and so I am not paid. Voilà!" He placed his five-franc piece upon the table, and shrugged his shoulders. "One full moon!" he said, and piled the three sous upon it. "And three soldiers. As I sit here, that is all, until to-morrow night. We cannot go!" Brave little Mimi! Already she was winking back her tears, and smiling. "But that—that is nothing!" she answered. "I do not care to go. No— but truly! Look! We shall spend the day in the studio, and breakfast on the balcony, and pretend the rue Visconti is the Seine." "I am an empty siphon!" said Pierre, yielding to desperation. "Non!" said Mimi firmly. "I am a pierced basket, a box of matches!" "Non! Non!" said Mimi, with tremendous earnestness. "Thou art Pierrot, and I love thee! Let us say no more. I shall go back and prepare the dinner, and thou shalt remain and drink a Pernod. It will give thee heart. But follow quickly. Give me the key." She laid her wide-spread hand on his, palm upward, like a little pink starfish.
- 68. "We go together, and I adore thee!" said Pierre, and kissed her in the sight of all men, and was not ashamed. Caffiard leaned forward, picked up the fallen wine-card, pretended to consult it, and ponderously arose. As Pierre was turning the key in the door of the little apartment, they heard a sound of heavy breathing, and the deus ex machina came lumbering up the winding stair. "Monsieur is seeking some one?" asked the painter politely. There was no breath left in Caffiard. He was only able, by way of reply, to point at the top button of Pierre's coat, and nod helplessly: then, as Mimi ran ahead to light the gas, he labored along the corridor, staggered through the curtained doorway, stumbled over a rattan stool, was rescued by Pierre, and, finally, established upon the divan, very red and gasping. For a time there was silence, Pierre and Mimi busying themselves in putting the studio to rights, with an instinctive courtesy which took no notice of their visitor's snorts and wheezes; and Caffiard taking note of his surroundings with his round, blinking eyes. Opposite him, against the wall, reposed the portrait of the restaurateur's wife, as dry and pasty as a stale cream cheese upon the point of crumbling, and on an easel was another—that of Monsieur Pantin, the rich shirt- maker of the boulevard St. Germain—on which Pierre was at work. A veritable atrocity this, with a green background which trespassed upon Monsieur Pantin's hair, and a featureless face, gaunt and haggard with yellow and purple undertones. There was nothing in either picture to refute one's natural suspicion that soap had been the medium employed. Caffiard blinked harder still as his eyes rested upon the portraits, and he secretly consulted the crumpled wine-card in his hand. Then he seemed to recover his breath by means of a profound sigh. "Monsieur makes caricatures?" he inquired.
- 69. "Ah, monsieur," said Pierre, "at times, and for amusement only. I am a portraitist." And he pointed proudly to the picture against the wall. For they are all alike, these painters—proudest of what they do least well! "Ah! Then," said Caffiard, with an air of resignation, "I must ask monsieur's pardon, and descend. I am not interested in portraits. When it comes to caricatures"— "They are well enough in their way," put in Pierre, "but as a serious affair—to sell, for instance—well, monsieur comprehends that one does not debauch one's art!" Oh, yes, they are all alike, these painters! "What is serious, what is not serious?" answered Caffiard. "It is all a matter of opinion. One prefers to have his painting glued to the wall of the Salon, next the ceiling, another to have his drawing on the front page of La Blague." "Oh, naturally La Blague," protested Pierre. "I am its editor," said Caffiard superbly. "Eigh!" exclaimed Pierre. For Mimi had cruelly pinched his arm. Before the sting had passed, she was seated at Caffiard's side, tugging at the strings of a great portfolio. "Are they imbeciles, these painters, monsieur?" she was saying. "Now you shall see. This great baby is marvelous, but marvelous, with his caricatures. Not Léandre himself—it is I who assure you, monsieur!—and to hear him, one would think—but thou tirest me, Pierrot!—With his portraits! No, it is too much!" She spread the portfolio wide, and began to shuffle through the drawings it contained. Caffiard's eyes glistened as he saw them. Even in her enthusiasm, Mimi had not overshot the mark. They were marvelous indeed, these caricatures, mere outlines for the most part, with a dot, here and there, of red, or a little streak of green, which lent them a curious,
- 70. unusual charm. The subjects were legion. Here was Loubet, with a great band of crimson across his shirt bosom, here Waldeck- Rousseau, with eyes as round and prominent as agate marbles, or Yvette, with a nose on which one might have hung an overcoat, or Chamberlain, all monocle, or Wilhelmina, growing out of a tulip's heart, and as pretty as an old print, with her tight-fitting Dutch cap and broidered bodice. And then a host of types—cochers, grisettes, flower women, camelots, Heaven knows what not!—the products of half a hundred idle hours, wherein great-hearted, foolish Pierre had builded better than he knew! Caffiard selected five at random, and then, from a waistcoat pocket that clung as closely to his round figure as if it had been glued thereto, produced a hundred-franc note. "I must have these for La Blague, monsieur," he said. "Bring me two caricatures a week at my office in the rue St. Joseph, and you shall be paid at the same rate. It is not much, to be sure. But you will have ample time left for your—for your portrait-painting, monsieur!" For a moment the words of Caffiard affected Pierre and Mimi as the stairs had affected Caffiard. They stared at him, opening and shutting their mouths and gasping, like fish newly landed. Then, suddenly, animated by a common impulse, they rushed into each other's arms, and set out, around the studio, in a mad waltz, which presently resolved itself into an impromptu can-can, with Mimi skipping like a fairy, and Pierre singing: "Hi! Hi!! Hi!!!" and snapping at her flying feet with a red-bordered handkerchief. After this Mimi kissed Caffiard twice: once on the top of his bald head, and once on the end of his stubby nose. It was like being brushed by the floating down of a dandelion. And, finally, nothing would do but that he must accompany them upon the morrow; and she explained to him in detail the plan which had so nearly fallen through, and the deus ex machina did not betray by so much as a wink that he had heard the entire story only half an hour before. But, in the end, he protested. But she was insane, the little one, completely! Had he then the air of one who gave himself into those
- 71. boats there, name of a pipe? But let us be reasonable, voyons! He was not young like Pierre and Mimi—one comprehended that these holidays did not recommence when one was sixty. What should he do, he demanded of them, trailing along, as one might say, he and his odious fatness? Ah, non! For la belle jeunesse was la belle jeunesse, there was no means of denying it, and it was not for a species of dried sponge to be giving itself the airs of a fresh flower. "But no! But no!" said Caffiard, striving to rise from the divan. "In the morning I have my article to do for the Figaro, and I am going with Caran to Longchamp, en auto, for the races in the afternoon. But no! But no!" It was plain that Caffiard had known Mimi no more than half an hour. One never said, "But no! But no!" to Mimi, unless it was for the express purpose of having one's mouth covered by the softest little pink palm to be found between the Seine and the Observatoire,— which, to do him justice, Caffiard was quite capable of scheming to bring about, if only he had known! He had accepted the little dandelion-down kisses in a spirit of philosophy, knowing well that they were given not for his sake, but for Pierre's. But now his protests came to an abrupt termination, for Mimi suddenly seated herself on his lap, and put one arm around his neck. It was nothing short of an achievement, this. Even Caffiard himself had not imagined that such a thing as his lap was still extant. Yet here was Mimi, actually installed thereon, with her cheek pressed against his, and her breath, which was like clover, stirring the ends of his moustache. But she was smiling at Pierre, the witch! Caffiard could see it out of the corner of his eye. "Mais non!" he repeated, but more feebly. "Mais non! Mais non! Mais non!" mocked Mimi. "Great farceur! Will you listen, at least? Eh b'en, voilà! Here is my opinion. As to insanity, if for any one to propose a day in the country is insanity, well then, yes,—I am insane! Soit! And, again, if you wish to appear serious,— in Paris, that is to say—soit, également! But when you speak of odious fatness, you are a type of monsieur extremely low of ceiling,
- 72. do you know! Moreover, you are going. Voilà! It is finished. As for Caran, let him go his way and draw his caricatures—though they are not like Pierre's, all the world knows!—and, without doubt, his auto will refuse to move beyond the porte Dauphine, yes, and blow up, bon Dieu! when he is in the act of mending it. One knows these boxes of vapors, what they do. And as for the Figaro, b'en, flut! Evidently it will not cease to exist for lack of your article—eh, l'ami? And it is Mimi who asks you,—Mimi, do you understand, who invites you to her fête. And you would refuse her—toi!" "But no! But no!" said Caffiard hurriedly. And meant it. At this point Pierre wrapped five two-sou pieces in a bit of paper, and tossed them, out of a little window across the hallway, to a street-singer whimpering in the court below. Pierre said that they weighed down his pockets. They were in the way, the clumsy doublins, said wonderful, spendthrift Pierre! For the wide sky of the Quartier is forever dotted with little clouds, scudding, scudding, all day long. And when one of these passes across the sun, there is a sudden chill in the air, and one walks for a time in shadow, though the comrade over there, across the way, is still in the warm and golden glow. But when the sun has shouldered the little cloud aside again, ah, that is when life is good to live, and goes gayly, to the tinkle of glasses and the ripple of laughter, and the ring of silver bits. And when the street-singer in the court receives upon his head a little parcel of coppers that are too heavy for the pocket, and smiles to himself, who knows but what he understands? For what is also true of the Quartier is this—that, in sunshine or shadow, one finds a soft little hand clasping his, firm, warm, encouraging and kindly, and hears a gay little voice that, in foul weather, chatters of the bright hours which it is so sweet to remember, and, in fair, says never a word of the storms which it is so easy to forget!
- 73. The veriest bat might have foreseen the end, when once Mimi had put her arm around the neck of Caffiard. Before the deus ex machina knew what he was about, he found his army of objections routed, horse, foot, and dragoons, and had promised to be at the gare St. Lazare at eleven the following morning. And what a morning it was! Surely the bon Dieu must have loved Mimi an atom better than other mortals, for in the blue-black crucible of the night he fashioned a day as clear and glowing as a great jewel, and set it, blazing with warm light and vivid color, foremost in the diadem of the year. And it was something to see Mimi at the carriage window, with Pierre at her side and her left hand in his, and in her right a huge bouquet—Caffiard's contribution —while the deus ex machina himself, breathing like a happy hippopotamus, beamed upon the pair from the opposite corner. So the train slipped past the fortifications, swung through a trim suburb, slid smoothly out into the open country. It was a Wednesday, and there was no holiday crowd to incommode them. They had the compartment to themselves; and the half hour flew like six minutes, said Mimi, when at last they came to a shuddering standstill, and two guards hastened along the platform in opposite directions, one droning "Poiss-y-y-y-y!" and the other shouting "Poiss'! Poiss'! Poiss'!" as if he had been sneezing. It was an undertaking to get Caffiard out of the carriage, just as it had been to get him in. But finally it was accomplished, a whistle trilled from somewhere as if it had been a bird, another wailed like a stepped-on kitten, the locomotive squealed triumphantly, and the next minute the trio were alone in their glory. It was a day that Caffiard never forgot. They breakfasted at once, so as to have a longer afternoon. Mimi was guide and commander-in- chief, as having been to the Esturgeon before, so the table was set upon the terrasse overlooking the Seine, and there were radishes, and little individual omelettes, and a famous matelote, which Monsieur Jarry himself served with the air of a Lucullus, and, finally, a great dish of quatre saisons, and, for each of the party, a squat brown pot of fresh cream. And, moreover, no ordinaire, but St.
- 74. Emilion, if you please, with a tin-foil cap which had to be removed before one could draw the cork, and a bottle of Source Badoit as well. And Caffiard, who had dined with the Russian Ambassador on Monday and breakfasted with the Nuncio on Tuesday, and been egregiously displeased with the fare in both instances, consumed an unprecedented quantity of matelote, and went back to radishes after he had eaten his strawberries and cream: while, to cap the climax, Pierre paid the addition with a louis,—and gave all the change as a tip! But it was unheard-of! Afterwards they engaged a boat, and, with much alarm on the part of Mimi, and satirical comment from Caffiard, and severe admonitions to prudence by Pierre, pushed out into the stream and headed for Villennes, to the enormous edification of three small boys, who hung precariously over the railing of the terrace above them, and called Caffiard a captive balloon. They made the three kilometres at a snail's pace, allowing the boat to drift with the current for an hour at a time, and, now and again creeping in under the willows at the water's edge until they were wholly hidden from view, and the voice of Mimi singing was as that of some river nixie invisible to mortal eyes. She sang "Bonsoir, Madame la Lune," so sweetly and so sadly that Caffiard was moved to tears. It was her favorite song, because—oh, because it was about Pierrot! And her own Pierrot responded with a gay soldier ballad, a chanson de route which he had picked up at the Noctambules; and even Caffiard sang—a ridiculous ditty it was, which scored the English and went to a rollicking air. They all shouted the refrain, convulsed with merriment at the drollery of the sound:—
- 75. "Qu'est ce qui quitte ses père et mère Afin de s'en aller S'faire taper dans le nez? C'est le soldat d'Angleterre! Dou-gle-di-gle-dum! Avec les ba-a-a-alles dum-dum!" Caffiard was to leave them at Villennes after they should have taken their apéritifs. They protested, stormed at him, scolded and cajoled by turns, and called him a score of fantastic names—for by this time they knew him intimately—as they sat in Monsieur Bodin's arbor and sipped amer-menthe, but all in vain. Pierre had Mimi's hand, as always, and he had kissed her a half-hundred times in the course of the afternoon. Mimi had a way of shaking her hair out of her eyes with a curious little backward jerk of her head when Pierre kissed her, and then looking at him seriously, seriously, but smiling when he caught her at it. Caffiard liked that. And Pierre had a trick of turning, as if to ask Mimi's opinion, or divine even her unspoken wishes whenever a question came up for decision—a choice of food or drink, or direction, or what-not. And Caffiard liked that. He looked across the table at them now, dreamily, through his cigarette smoke. "Pierrot," he said, after he had persuaded them to let him depart in peace when the train should be due,—"Pierrot. Yes, that is it. You, with your garret, and your painting, and your songs, and your black, black sadness at one moment, and your laughter the next, and, above all, your Pierrette, your bon-bon of a Pierrette:—you are Pierrot, the spirit of Paris in powder and white muslin! Eigho! my children, what a thing it is, la belle jeunesse! Tiens! you have given me a taste of it to-day, and I thank you. I thought I had forgotten. But no, one never forgets. It all comes back,—youth, and strength, and beauty, love, and music, and laughter,—but only like a breath upon a mirror, my children, only like a wind-ripple on a pool; for I am an old man."
- 76. He paused, looking up at the vine-leaves on the trellis-roof, and murmured a few words of Mimi's song:— "Pierrette en songe va venir me voir: Bonsoir, madame la lune!" Then his eyes came back to her face. "I must be off," he said. "Why, what hast thou, little one? There are tears in those two stars!" "C'est vrai?" asked Mimi, smiling at him and then at Pierre, and brushing her hand across her eyes, "c'est vrai? Well then, they are gone as quickly as they came. Voilà! Without his tears Pierrot is not Pierrot, and without Pierrot"— She turned to Pierre suddenly, and buried her face on his shoulder. "Je t'aime!" she whispered. "Je t'aime!"
- 78. A T h e N e x t C o r n e r NTHONY CAZEBY was a man whom the felicitous combination of an adventurous disposition, sufficient ready money, and a magnificent constitution had introduced to many and various sensations, but he was conscious that, so far as intensity went, no one of them all had approached for a moment that with which he emerged from the doorway of the Automobile Club, and, winking at the sting of the keen winter air, looked out across the place de la Concorde, with its globes of light, swung, like huge pearls on invisible strings, across the haze of the January midnight. He paused for a moment, as if he would allow his faculties to obtain a full and final grasp of his situation, and motioned aside the trim little club chasseur who stood before him, with one cotton-gloved hand stretched out expectantly for a supposititious carriage-check. "Va, mon petit, je vais à pied!" Afoot! Cazeby smiled to himself at the tone of sudden caprice which rang in his voice, and, turning his fur collar high up about his ears, swung off rapidly toward the Cours la Reine. After all, the avenue d'Eylau was only an agreeable stroll's length distant. Why not go home afoot? But then, on the other hand, why go home at all? As this thought leaped suddenly at Cazeby's throat out of the void of the great unpremeditated, he caught his breath, stopped suddenly in the middle of the driveway, and then went on more slowly, thinking hard. It had been that rarissima avis of social life, even in Paris, a perfect dinner. Cazeby had found himself wondering, at more than one stage of its smooth and imposing progress, how the Flints could afford to do it. But on each recurrence of the thought he dismissed it with a little frown of vexation. If there was one thing more than
- 79. another upon which Cazeby prided himself, it was originality of thought, word, and deed, and he was annoyed to find himself, even momentarily, on a mental level with the gossips of the American and English colonies, whose time is equally divided between wondering how the Choses can afford to do what they do, and why the Machins cannot afford to do what they leave undone. People had said many things of Hartley Flint, and still more of his wife, but no one had ever had the ignorance or the perversity to accuse them of inefficiency in the matter of a dinner. Moreover, on this particular occasion, they were returning the hospitality of the Baroness Klemftt, who had, at the close of the Exposition, impressed into her service the chef of the Roumanian restaurant, and whose dinners were, in consequence, the wonder and despair of four foreign colonies. After her latest exploit Hartley Flint had remarked to his wife that it was "up to them to make good," which, being interpreted, was to say that it was at once his duty and his intention to repay the Baroness in her own sterling coin. The fact that the men of the party afterwards commended Hartley's choice of wines, and that the women expressed the opinion that "Kate Flint looked really pretty!" would seem to be proof positive that the operation of "making good" had been an unqualified success. Now, Cazeby was wondering whether he had actually enjoyed it all. Under the circumstances it seemed to him incredible, and yet he could not recall a qualm of uneasiness from the moment when the maître d'hôtel had thrown open the doors of the private dining room, until the Baroness had smiled at her hostess out of a cloud of old Valenciennes, and said, "Now there are two of us who give impeccable dinners, Madame Flint." Even now, even facing his last ditch, Cazeby was conscious of a little thrill of self-satisfaction. He had said the score of clever things which each of his many hostesses expected of him, and had told with great effect his story of the little German florist, which had grown, that season, under the persuasive encouragement of society's applause, from a brief anecdote into a veritable achievement of Teutonic dialect. Also, he had worn a forty franc orchid, and had left it in his coffee-cup because it had begun
- 80. to wilt. In brief, he had been Anthony Cazeby at his extraordinary best, a mixture of brilliancy and eccentricity, without which, as Mrs. Flint was wont to say, no dinner was complete. But the sublime and the ridiculous are not the only contrasting conditions that lie no further than a step apart, and Cazeby was painfully conscious of having, in the past five minutes, crossed the short interval which divides gay from grave. Reduced to its lowest terms, his situation lay in his words to the little chasseur. With the odor of the rarest orchid to be found in Vaillant-Rozeau's whole establishment yet clinging to his lapel, Anthony Cazeby was going home on foot because the fare from the Concorde to the avenue d'Eylau was one franc fifty, and one franc fifty precisely ninety centimes more than he possessed in the world. For a moment he straightened himself, threw back his head, and looked up at the dull saffron of the low-hanging sky, in an attempt to realize this astounding fact, and then went back to his thinking. Well, it was not surprising. The life of a popular young diplomat with extravagant tastes is not conducive to economy, and the forty thousand dollars which had come to Cazeby at the beginning of his twenty-eighth year had proved but a bad second best in the struggle with Parisian gayety. His bibelots, his servants, Auteuil, Longchamp, his baccarat at the Prince de Tréville's, a dancer at the Folies- Marigny, Monte Carlo, Aix, Trouville,—they had all had their share, and now the piper was waiting to be paid and the exchequer was empty. It was an old story. Other men of his acquaintance had done the same, but they had had some final resource. The trouble was, as Cazeby had already noted, that, in his case, the final resource was not, as in theirs, pecuniary. Quite on the contrary, it was a tidy little weapon, of Smith and Wesson make, which lay in the upper right hand drawer of his marqueterie desk. He had looked long at it that same afternoon, with all his worldly wealth, in the shape of forty-two francs sixty, spread out beside it. That was before he had taken a fiacre to Vaillant-Rozeau's.
- 81. At the very moment when Cazeby was contemplating these doubtful assets, a grim old gentleman was seated at another desk, three thousand miles away, engaged upon a calculation of the monthly profits derived from a wholesale leather business. But Cazeby père was one of the hopeless persons who believe in economy. He was of the perverted opinion that money hardly come by should be thoughtfully spent, or, preferably, invested in government bonds, and he had violent prejudices against "industrials," games of chance, and young men who preferred the gayety of a foreign capital to the atmosphere of "the Swamp." Also he was very rich. But Anthony had long since ceased to regard his father as anything more than a chance relation. He could have told what would be the result of a frank confession of his extremity as accurately as if the avowal had been already made. There would have been some brief reference to the sowing of oats and their reaping, to the making of a metaphorical bed and the inevitable occupancy thereof, and to other proverbial illustrations which, in a financial sense, are more ornamental than useful,—and nothing more. The essential spark of sympathy had been lacking between these two since the moment when the most eminent physician in New York had said, "It is a boy, sir,—but—we cannot hope to save the mother." The fault may have lain on the one side, or the other, or on both, or on neither; but certain it is that to Anthony's imagination Cazeby senior had never appealed in the light of a final resource. Somehow, in none of his calculations had the idea of invoking assistance ever played a part. Naturally, as a reasoning being, he had foreseen the present crisis for some months, but at the time when the inevitable catastrophe first became clear to him it was already too late to regain his balance, since the remainder of his inheritance was so pitifully small that any idea of retrieving his fortunes through its instrumentality was simply farcical. The swirl of the rapids, as he had then told himself, had already caught his boat. All that was left to do was to go straight on to the sheer of the fall, with his pennant flying and himself singing at the helm. Then, on the brink, a well-placed bullet—no bungling for Anthony Cazeby!—and
- 82. the next day people would be talking of the shocking accident which had killed him in the act of cleaning his revolver, and saying the usual things about a young man with a brilliant future before him and everything in life for which to live. And this plan he had carried out in every detail—save the last, to which he was now come; and his was the satisfying conviction that not one of the brilliant, careless men and women, among whom he lived, and moved, and had his being, suspected for a moment that the actual circumstances differed in the least from the outward appearances. He thought it all over carefully now, and there was no play in the entire game that he felt he would have liked to have changed. Sentiment had no part in the makeup of Anthony Cazeby. Lacking from early childhood the common ties of home affection, and by training and profession a diplomat, he added to a naturally undemonstrative nature the non-committal suavity of official poise. But that was not all. He had never been known to be ill at ease. This was something which gained him a reputation for studious self- control. As a matter of fact it was due to nothing of the sort. No one had ever come fairly at the root of his character except Cazeby père, who once said, in a fit of passion, "You don't care a brass cent, sir, whether you live and are made President of the United States, or die and are eternally damned!" And that was exactly the point. Something of all this had passed through Cazeby's mind, when he was suddenly aroused to an appreciation of his whereabouts by the sound of a voice, to find that the curious instinct of direction which underlies advanced inebriety and profound preoccupation alike, had led him up the avenue du Trocadéro, and across the place, and that he had already advanced some little way along the avenue d'Eylau in the direction of his apartment. The street was dimly lighted, but, just behind him, the windows of a tiny wine-shop gave out a subdued glow, and from within came the sound of a violin. Then Cazeby's attention came around to the owner of the voice. This was a youngish man of medium stature, in the familiar street dress of a
- 83. French laborer, jacket and waistcoat of dull blue velveteen, peg-top trousers of heavy corduroy, a crimson knot at his throat, and a dark tam o'shanter pulled low over one ear. As their eyes met, he apparently saw that Cazeby had not heard his first remark, and so repeated it. "I have need of a drink!" There was nothing of the beggar in his tone or manner. Both were threatening, rather; and, as soon as he had spoken, he thrust his lower jaw forward, in the fashion common to the thug of any and every nationality when the next move is like to be a blow. But, for once, these manifestations of hostility failed signally of effect. Cazeby was the last person in the world to select as the object of sudden attack, with the idea that panic would make him easy prey. In his present state of mind he went further than preserving his equanimity: he was even faintly amused. It was not that he did not comprehend the other's purpose, but, to his way of thinking, there was something distinctly humorous in the idea of holding up a man with only sixty centimes to his name, and menacing him with injury, when he himself was on his way to the upper right hand drawer of the marqueterie desk. "I have need of a drink," repeated the other, coming a step nearer. "Thou art not deaf, at least?" "No," said Cazeby, pleasantly, "no, I am not deaf, and I, too, have need of a drink. Shall we take it together?" And, without waiting for a reply, he turned and stepped through the doorway of the little wineshop. The Frenchman hesitated, shrugged his shoulders with an air of complete bewilderment, and, after an instant also entered the shop and placed himself at the small table where Cazeby was already seated. "A vitriol for me," he said. Cazeby had not passed three years in Paris for nothing. He received this remarkable request with the unconcern of one to whom the slang of the exterior boulevards is sufficiently familiar, and, as the
- 84. proprietor leaned across the nickled slab of his narrow counter with an air of interrogation, duplicated his companion's order. "Deux vitriols!" The proprietor, vouchsafing the phrase a grin of appreciation, lumbered heavily around to the table, filled two small glasses from a bottle of cheap cognac, and stood awaiting payment, hands on hips. "Di-ze sous," he said. There was no need to search for the exact amount. Cazeby spun his fifty-centime piece upon the marble, added his remaining two sous by way of pourboire, and disposed of the brandy at a gulp. "Have you also need of a cigarette?" he inquired, politely, tendering the other his case. For some minutes, as they smoked, the diplomat and the vagabond took stock of each other in silence. In many ways they were singularly alike. There was in both the same irony of lip line, the same fair chiseling of chin and nostril and brow, the same weariness of eye. The difference was one of dress and bearing alone, and, in those first moments of mutual analysis, Cazeby realized that there was about this street-lounger a vague air of the gentleman, a subtle suggestion of good birth and breeding, which even his slouching manner and coarse speech were not wholly able to conceal: and his guest was conscious that in Cazeby he had to deal with no mere society puppet, but with one in whom the limitations of position had never wholly subdued the devil-may-care instincts of the vagabond. The one was a finished model of a man of the world, the other a caricature, but the clay was the same. "I am also hungry," said the latter suddenly. "In that respect," responded Cazeby, in the same tone of even politeness, "I am, unfortunately, unable to assist you, unless you will accept the hospitality of my apartment. It is but a step, and I am rather an expert on bacon and eggs. Also," he added, falling into the
- 85. idiom of the faubourgs, "there is a means there of remedying the dryness of the sponge in one's throat. My name is Antoine." "I am Bibi-la-Raie," said the other shortly. Then he continued, with instinctive suspicion, "It is a strange fashion thou hast of introducing a type to these gentlemen." "As a matter of fact," said Cazeby, "I do not live over a poste. But whether or not you will come is something for you to decide. It is less trouble to cook eggs for one than for two." Bibi-la-Raie reflected briefly. Finally he had recourse to his characteristic shrug. "After all, what difference?" he said. "As well now as another time. I follow thee!" The strangely assorted companions entered Cazeby's apartment as the clock was striking one, and pressure of an electric button, flooding the salon with light, revealed a little tea-table furnished with cigarettes and cigars, decanters of Scotch whiskey and liqueurs, and Venetian goblets of oddly tinted glass. Cazeby shot a swift glance at his guest as this array sprang into view, and was curiously content to observe that he manifested no surprise. Bibi-la-Raie had flung himself into a great leather chair with an air of being entirely at ease. "Not bad, thy little box," he observed. "Is it permitted?" He indicated the table with a nod. "Assuredly," said Cazeby. "Do as if you were at home. I shall be but a moment with the supper." When he returned from the kitchen, bearing a smoking dish of bacon and eggs, butter, rye bread, and Swiss cheese, Bibi-la-Raie was standing in rapt contemplation before an etching of the "Last Judgment." "What a genius, this animal of a Michel Ange!" he said.
- 86. Welcome to our website – the perfect destination for book lovers and knowledge seekers. We believe that every book holds a new world, offering opportunities for learning, discovery, and personal growth. That’s why we are dedicated to bringing you a diverse collection of books, ranging from classic literature and specialized publications to self-development guides and children's books. More than just a book-buying platform, we strive to be a bridge connecting you with timeless cultural and intellectual values. With an elegant, user-friendly interface and a smart search system, you can quickly find the books that best suit your interests. Additionally, our special promotions and home delivery services help you save time and fully enjoy the joy of reading. Join us on a journey of knowledge exploration, passion nurturing, and personal growth every day! ebookbell.com