Chapter 5 – The ethnography of language and language documentation
119
deployed, so that the face and reputation of all interlocutors are properly
attended to. Indeed, any community may have certain kinds of speech
events in which outsiders simply cannot successfully participate. For this
reason, and also because it is both ethical and sensible to build local capac-
ity, it is generally preferable to train local native speakers in recording
techniques and have them do most of the basic recording themselves.
The kinds of diversity in patterns of usage studied by ethnographers of lan-
guage have often been treated as relatively stable in communities. But
documentary linguists must also attend to contexts in which new conven-
tions and forms of diversity can emerge very fast: the contexts of elicitation
and adult second-language acquisition that are at the center of their work.
Linguists who do field work have understood for many years that elicita-
tion is a collaborative process that requires mutual adaptation on the part of
researcher and consultant. Early attention to the problem of what happens
in elicitation and in the kind of adult second-language learning that docu-
mentary linguists undertake focused mainly on problems that would
emerge from different patterns about matters like asking questions. Charles
Briggs’ Learning how to ask (1986), where he argues that the acquisition of
new information must be embedded in local social understandings of who
is permitted to ask what kinds of questions to whom, is a classic discussion
of this issue. Some anthropologists, including Briggs himself, have found
that the best way to work is to undertake what is locally understood as an
apprentice role. I don’t think this approach is a solution to the problems
faced by documentary linguists. Communities may have well-established
institutions for apprenticeship in wood-carving or divination. They will
certainly have very well-established patterns for first-language socializa-
tion. But it is highly unlikely that they will have well-established patterns
for adult second-language learning or elicitation. And certain local patterns
for adult learning may be quite inappropriate to the documentary linguist’s
task. A very good example is the routine of adult acquisition of ceremonial
orations and creation accounts among the Tohono O’odham of Arizona
described by Ruth Underhill (1946). A man (it was always a man) who
wished to learn a particular oration would approach someone who knew it
and present a very important gift, consistent with the significance of the
target text – blankets, a rifle, a horse. If the source accepted the gift, he
120
Jane H. Hill
would then recite the oration: once. The job of the apprentice was to listen
with the most intense focus, to try to master as much of the oration as pos-
sible from this single recitation. Because if he needed to hear it again, an-
other expensive gift would be required. This particular method really would
not work for most documentary linguistics – in fact, it has been tried. The
linguist Bill Graves described in his dissertation (Graves 1988) encountering
a Pima speaker, an immensely knowledgeable elder who had been very
highly recommended by everyone, who chose to organize his role as lin-
guistic consultant along the lines of the traditional model for learning that
Underhill had described. Graves had to arrive early, because if he was even
five minutes late for an appointment Mr. Brown would refuse to talk to
him. Graves had to listen with the most extreme care, because Mr. Brown
spoke very quietly, did not like repeating things, and refused to explain
things. Mr. Brown would occasionally rise abruptly and terminate a meeting
if he was annoyed. Finally, Mr. Brown required cash up front at every
meeting. After a summer of this sort of thing, Graves reluctantly concluded
that Mr. Brown was a bit too traditional and sought a consultant who was
willing to compromise.
The absence of established routines for adult second-language learning
and linguistic elicitation in most minority-language communities makes it
obvious that elicitation will produce some kind of new system that emerges
in collaboration. New theory in “learning how to learn” shows that such
emergent systems are always produced in learning communities, even in
ones that seem well-established and stable. Learning communities belong
to the category of social organizations that have come to be called “com-
munities of practice.” Eckert and McConnell-Ginet (1992: 464) provided a
founding definition of this entity: “A community of practice is an aggregate
of people who come together around mutual engagement in an endeavor …
practices emerge in the course of this mutual endeavor.” Meyerhoff (2002)
has usefully summarized the theory of communities of practice, which have
become an important unit of analysis in recent variationist sociolinguistics.
The key elements of Eckert and McConnell-Ginet’s definition are mu-
tual engagement – which may be “harmonious or conflictual,” and the en-
deavor, which Meyerhoff defines as a jointly negotiated enterprise, which
must be reasonably specific. Finally, a community of practice will develop
a shared repertoire of normative practices and interactional resources that
are “the cumulative result of internal negotiations” (Meyerhoff 2002: 528).
These subcomponents are in dialectical relationship: mutual engagement
both makes possible, and is made possible by, the negotiation of a joint