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-
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