Chapter 5 – The ethnography of language and language documentation
125
3. Language ideology and documentary linguistics
The last set of ideas to be presented here involve how we can attend to the
very fast-moving dynamics of language ideology in endangered language
communities. Something of the significance of language ideologies has
been recognized for a very long time. For instance, the early ethnographers
and linguists working in indigenous North America discovered that accounts
of the creation were fully performed only in the winter, and so they could
not be elicited in the summertime; indeed, people thought it was dangerous
to do so.
But the early ethnographers thought of this kind of ideologically-driven
pattern as simply one more stable difference between them and their con-
sultants. Today we are finding, though, that these ideological systems can
evolve and spread in communities with astonishing rapidity. I will discuss
an example that unfortunately I had to observe at immediate second hand –
the contretemps around the publication of the Hopi Dictionary, for which
my husband Kenneth C. Hill was project director. The Hopi, who live in
northeastern Arizona, are the western-most of the Puebloan societies. Paul
Kroskrity (1998) has shown how in the Puebloan communities of the U.S.
Southwest, all indigenous language tends to be ideologically assimilated to
the prototype of ritual language, the language of the kivas. Kiva knowledge
is not shared with people who have not been initiated into the relevant ritual
societies, and many of the pueblos have decided that their language is
strictly for insiders. Indeed, one Hopi linguist, briefly employed at the Uni-
versity of Arizona about 30 years ago, refused to teach the language to non-
Hopi students. A second point is important in understanding the dictionary
controversy: During the period when public ceremonies are underway, the
Hopi villages construct a sort of “anti-market” economy that extends the
practice of the kiva to the entire village: nothing is sold, everything that one
might need is given as a gift.
This was the background ideological context in which my husband
worked for more than a decade with colleagues Emory Sekaquaptewa,
Ekkehart Malotki, Mary Black, and others to compile the great dictionary
of Third Mesa Hopi (Hopi Dictionary Project 1998). During the period of
the research only the most minor difficulties appeared; all tribal officials
were involved and participating. They all knew that the project was the
brainchild and dream of a senior Hopi, Emory Sekaquaptewa. The diction-
ary research group was extremely careful of Hopi ritual sensitivities, and a
committee of Hopi elders made sure that the dictionary would not contain
126
Jane H. Hill
anything that would be in violation of ritual prohibitions. Arrangements
were made to distribute dictionaries free to schools and at a greatly reduced
price to Hopis, and all royalties were to be paid to the Hopi Foundation, a
non-profit foundation dedicated to Hopi education. However, when the
publication date of the dictionary neared, the University of Arizona Press
proudly published a handsome full-color brochure as an announcement of
this major work, in which a price of $ 80.00 for the volume was mentioned.
This announcement finally made public and unavoidable what everyone
had managed to keep in the background – that the dictionary, which had
been largely funded by money from the U.S. government’s National Endow-
ment for the Humanities, would be available to non-Hopis, and that it would
be sold. This precipitated a difficult year during which the Hopi Director of
Cultural Affairs, Leigh Kuwanwisiwma, supported by many other Hopis,
argued that the dictionary should not be published at all because the Hopi
language should not be bought and sold, and certainly not for the benefit of
non-Hopis. Eventually the political faction that supported the dictionary
prevailed and it was published, but this result was by no means guaranteed
(Hill 2002 discusses this episode).
Recent theoretical work on linguistic ideologies can help us to under-
stand this sort of episode, and perhaps to work better and more compre-
hendingly with community members who support documentation of their
heritage language in dictionaries and development projects like language
classes. Susan Gal and Judith Irvine (1995) showed that language ideolo-
gies nearly always invoke three major semiotic principles. These are “ico-
nization,” “recursiveness,” and “erasure.” In “iconization,” elements of
language are shaped to match elements “in the world” – and by erasure, any
dimension of language that does not conform is ignored. By “recursive-
ness,” “iconization” operates throughout the system, bringing elements at
every level into line. Michael Silverstein (1996, 2003) has pointed out the
operation of what he calls the “dialectic of indexicality,” by which indexi-
cality is reshaped as reference. Miyako Inoue (2004) has shown how cer-
tain kinds of social circumstances – episodes of rapid political economic
change, in which identities are being rapidly restructured – heighten the
rapidity and strength of these processes.
Using these theoretical tools, we can say something about the Hopi case,
in which a language and an associated way of life that had always been
taken for granted becomes the object of the most acute attention and reflec-
tion. Such attention and reflection, and the iconization principle, yields an
exaggerated purism. In the Hopi case, by iconization the Hopi community