MICHAEL SILVERSTEIN

The Fieldwork Encounter and

the Colonized Voice of Indigeneity

E T H N O G R A P H I C A N D L I N G U I S T I C F I E L D W O R K generates
inscriptions of various sorts and, in our contemporary multimedia world,
in various modalities as well. A mode of Amerindianist fieldwork rendered
canonical by Franz Boas and his students centers on native language texts
taken down from dictation-speed informant speech and later translated and
published in bilingual editions. In this philological enterprise on behalf of
the otherwise unlettered, the goal was to establish through publication a reli-
able corpus bespeaking a culture’s—not merely an individual’s—cosmogony
and reflexive historical consciousness, its members’ view of their sociocul-
tural universe, no less than to provide sufficient primary verbal material for
an inductive grammatical analysis of the indigenous language of the corpus
of texts.1
But of course even such a situation, bringing together a dictating speaker
and a transcribing anthropological amanuensis, is a two-party discursive
interaction. It is a social event in which individuals inhabit role relationships
based on parameters of identity from which they are, as we say, relationally
“recruited” to their roles in institutional circumstances that depend on wider
background forces of sociohistorical reality. So the dictated material must
perforce be read as a text precipitated in and pointing to (“indexing”) a com-
plex and multilayered interactional context, to be treated no differently in
this respect from the transcripts we make these days from videotaped inter-
actions for purposes of sociological and anthropological analysis of their
dynamics.
In such analysis, we understand the self-contextualizing power of dis-
course to be semiotically parallel to that of pantomime. In both, much of
what is interpretable in the interval of multiparty engagement is built up

a b s t r a c t This essay follows the indexical (context-indicating) clues of linguistic form in spoken
Kiksht (Wasco-Wishram Chinookan) and reconstructs the emerging poetic or metrical structures of
a long-ago Kiksht-mediated encounter during anthropological linguistic fieldwork, memorialized in a pub-
lished text. In this way we can hear something of the voice of a Native American speaker coming to grips
with the impact of social and cultural change in the American settler state of the turn of the twentieth
century. Rep re s en ta t io ns 137. Winter 2017 © The Regents of the University of California. ISSN
0734-6018, electronic ISSN 1533-855X, pages 23--43. All rights reserved. Direct requests for permission to
photocopy or reproduce article content to the University of California Press at http://www.ucpress.edu/
journals.php?p¼reprints. DOI: https://doi.org/10.1525/rep.2017.137.2.23. 23
 from individual gestural acts and from the sequencing and chunking, the
metricalization, of whole segments of behavior, whether verbal or kinesic,
from which an addressee must reconstruct a cultural context in which the
textual form—gradually coming, over space-time, to be “entextualized,” that
is, rendered coherent as text—comes to make cultural sense (and by making
cultural sense, affords one or more interpretations of what is going on). The
relationship of any feature of text to its cultural context is, semiotically speak-
ing, dynamically indexical; at every instant, such features of talk or movement
point to an already in-play sociocultural frame and to one about-to-come-
into-being, the first licensing the “appropriateness” of the occurrence of
some textual feature, the second entailed in-and-by its very occurrence. The
second is the so-called performative meaningfulness of what speakers do with
words (as with kinesic motions), the social acts we understand their perfor-
mance will have effected as social actors of particular characteristics in par-
ticular circumstances.2 Such indexical reading is central to discerning
a generationally new kind of historical consciousness and hence indigenous
voice in the long-ago event of fieldwork encounter on which I concentrate.3
Peter McGuff, aged about thirty in the summer of 1905 and a speaker of
Kiksht, the easternmost Chinookan language along the Columbia River—as
well as of Klickitat Sahaptin and English—dictated a short text to the anthro-
pologist Edward Sapir that the latter published in 1909 in Wishram Texts.
A doctoral student working under Boas at Columbia University, Sapir pub-
lished the text along with much other material spoken by far older speakers,
principally Louis Simpson, then, in 1905, aged about seventy-five. As some-
one who has also done fieldwork on the language, in the 1960s and 1970s
with a number of Kiksht speakers roughly of Mr. McGuff’s generation and life
experiences, I have returned to this text several times in relation to the state
of the language as I observed it now forty and more years ago, closer indeed
to Mr. McGuff’s usage than to Mr. Simpson’s. I would like here to focus
attention upon a grammatical hapax legomenon in Mr. McGuff’s dictation,
a unique textual occurrence in the whole Sapir collection in fact, and to
contextualize its occurrence in respect of the McGuff-Sapir interaction and
what it seems to reveal about Mr. McGuff’s generational experience in the
rapidly encroaching colonial context of turn-of-the-twentieth-century Native
American reservation life.
To do so, we will have to consider Wasco-Wishram social life diachroni-
cally, allowing us to contextualize this particular linguistic form in relation to
the accumulated corpus of primary Chinookan language materials that we
have from the turn of the nineteenth century onward. The earliest are word
lists and expressions written down in the languages of the environs of the
Columbia River mouth. The late nineteenth- and twentieth-century material
spans text collections in two Chinookan languages by Franz Boas; Sapir’s

24 Representations
1905 Kiksht corpus from Mr. Simpson and other speakers, material of the
1930s (compiled by Melville Jacobs for Clackamas and by Walter Dyk for
Wasco-Wishram) and from the 1950s (compiled by David and Kathrine
French; Dell Hymes) as well as my own material developed some seventy-
five years after Boas’s. Mr. McGuff’s textual features allow us to catch lan-
guage and cultural change in progress in what had long been a plurilingual
speech community, by 1905 transforming in new ways, and to elaborate the
several orders of contextualization that are indexically implicated by the
linguistic forms in question.
.
Here is the unique grammatical form in question: ugwíłxix “it [a fire, wá-tuł
(fem.sg.)] must have been kindled/lit,” occurring as the last word in the first
line—Sapir’s line number 3b—of the second page of the original printed
Kiksht version, visually reassembled in figure 1.4 Sapir translates it as “[a fire]
was already burning”—which is true of the remembered and represented
situation, but not a grammatically accurate rendering. The Kiksht form is
a tenseless passive inflected verb with feminine singular grammatical subject
(it would appear overtly as initial a- were the next sound not -u-), agreeing with
the subject of the clause beginning the next printed line, watúł. The stem is
. .
-u-gwiłx- based on the root -√kiłx- “to kindle, light something on fire,” a transi-
tive verb that takes something combustible as its “Patient” or object. The final
suffix -ix, a general-purpose deictic and locational element we might gloss as
“in this/that place,” here points to something in the default context of com-
munication. In this here-and-now of communication, the utterer of the form
indexes the fact that he/she infers that some immediately present sensory
evidence indicates—points to the fact—that prior to the moment of commu-
nication someone has done to the “Patient” or object whatever the verb-stem
denotes.5 Here, to the inferring consciousness of young Pete, the roaring
flames in the fire pit of the traditional Wasco-Wishram semisubterranean
winter dwelling evoked by Mr. McGuff’s account indexes that someone had
kindled the fire prior to his arrival back home. Why is this form, in type what I
have termed an “evidential passive,” so revelatory in the textual and sociocul-
tural contexts in which it occurs in Mr. McGuff’s dictation to Sapir in 1905?
Briefly, I see it as indexing a new kind of consciousness on Mr. McGuff’s part,
one that brings the consciousness of a remembered past together with that of
a remembering present in a new way, very much akin to the indirect free style
of narrations of psychological interiority. Before that moment of Mr. McGuff’s
use it is uncountenanced in the Chinookan language corpus, including even in
the other text-artifacts in Sapir’s collection, taken from older speakers who did
not share the generational experiences of Mr. McGuff in the settler state.
To understand all this, we must center discussion on the Kiksht language
community, including those speakers of Upper Chinookan dialect localisms,
like Messrs. Simpson and McGuff, earlier distributed in a geographical range

The Fieldwork Encounter and the Colonized Voice of Indigeneity 25

figure 1a. Textual features of McGuff’s reminiscence in Kiksht.
figure 1b. Sapir’s original English version.
 northward from Clackamas in villages on the Willamette River south of
Portland, Oregon, and ranging eastward along both sides of the Columbia
River—now the boundary between the states of Washington and Oregon—to
.
the easternmost pair of villages, Nixluidix (in the Sahaptin language, Wišxami)
on the present Washington side of the Columbia, and Ałasq’u (Sahaptin,
Wasq’upam) on the present Oregon side, just below the former Celilo Falls,
at present The Dalles, Oregon.6 By the time of Sapir’s graduate school field-
work in 1905, a number of families from both villages and from other riverine
locations had accepted allotments of farm- and ranchland on one of the two
sister reservations, one in Washington state, one in Oregon, that came into
being consequent upon the signing of treaties, the cession of riverine lands,
and the administration of the federal Dawes Act of 1887. Mr. Simpson and his
brother, Tom Simpson, both became landholders outside the town of White
Swan on the Yakima [now Yakama] Reservation in south-central Washington,
where no fewer than twenty-one different named ethno-tribal units of Native
Americans were consolidated, from three different historical language fami-
lies, Salishan, Sahapti(a)n, and Chinookan. As well, a long period of contact
from the earliest days of the Oregon fur trade in the late eighteenth century
had already resulted in multiple non-Indian partial ancestries for many Colum-
bia River people, such as Mr. McGuff himself, whose father was black, though
his maternal affiliations by residence, mode of life, and language seem to have
been definitive in his own self-identification as indigenous.
There is a long history of plurilingualism in the historically dynamic speech
community that encompasses these Kiksht-speaking peoples and various sur-
rounding groups and newcomers to the area. There was an indigenous system
of rank-sensitive intermarriage-trade-warfare-slavery that promoted linguistic
exogamy as well as linguistic exolocality. Further, the Chinookan speakers, in
particular, were renowned for their status as middlemen in trade in the lower
Columbia River and tributary drainages, both among indigenous groups sea-
sonally pouring into the WišXami-Wasq’upam area for the products of the
extraordinary local salmon fishery, for basketry and carved items, for denta-
lium (“wampum”), for slaves, and so on, and later, among the early
nineteenth-century Euro-Americans and -Canadians coming as traders, as mis-
sionaries, and then settlers. It was, of course, the Chinookan language, Lower
Chinook, spoken at and around the mouth of the Columbia River and Willapa
Bay, that was the principal component, along with words from Nootka, French,
and a little English, of Chinook Jargon (“Tsinuk Wawa”). This trade language
emergent from the Europeans’ maritime exploration of the Northwest Coast
was eventually used—in the late nineteenth–early twentieth century—from the
Alaska panhandle almost to northern California, and from the Pacific coast
inland to Saskatchewan and Idaho, as the intergroup contact vernacular used
beyond the bounds of normal indigenous polyglot sociality.7

28 Representations
 By Mr. McGuff’s generation, English, learned in a relatively standardized
form through Bureau of Indian Affairs boarding-school education, of which
he had some, was clearly being added to the repertoire of languages among
Yakima Reservation and Columbia River people; thus could Mr. McGuff serve
as interpreter, translator, and later collaborator by postal correspondence for
Sapir’s research both during fieldwork and afterward as he prepared the text
collection for publication (Sapir taught McGuff to write the heretofore
unwritten Kiksht language using a roughly phonemic orthography, as he,
Sapir, had done during the 1905 field season). Mr. Simpson, by contrast, born
circa 1830–1835, while able to speak Klickitat Sahaptin and Chinook Wawa,
barely knew English beyond some words and phrases: “His English is extremely
broken, hardly intelligible at times,” noted Sapir in his introduction to the
texts.8 (My own experience, too, in 1970 with Mrs. Martha Sconawah, who still
lived at Nixluidix and was born circa 1865, was that she knew essentially no
English, while my consultants born during the period of 1878 through 1893
were quite fluent at least in local vernacular forms of the rural Northwest.)
Though the situation on the Northwest Coast has long attracted students,
such as Franz Boas himself, of linguistic and cultural diffusion across social
groups—resulting in linguistic and cultural areal phenomena—this tale of
plurilingualism should not, in fact, be seen as particularly unusual. We can be
certain that Boas himself, from central Europe, understood something of the
polyglot abilities of most educated Europeans and thus of the plurilingualism
of Europe as a linguistic as well as cultural area, with certain intense mani-
festations such as the Balkan southeast, then under Habsburg and Ottoman
rule. Boas used diffusional arguments against the social evolutionary, racial-
izing discourses of unique, pristine, autonomously evolving language-culture-
race (including “mentality”) linkages that were the stock-in-trade of late
nineteenth-century technical as well as popular theorizing, coming ulti-
mately, if erroneously, to question the validity of Stammbaum (branching-
tree) linguistic “family” relationships for any North American language.
Here, in fact, in Kiksht, we have lots of evidence of diffused or borrowed
linguistic material that came from the speakers’—like Messrs. Simpson and
McGuff and my consultants— contact with Sahaptin and their at least partial
bilingualism in this numerically more dominant language spoken on both
reservations where Kiksht speakers were allotted land. Such contact must
have been over the course of several hundred years, in fact, since certain
borrowed forms already occur in the late eighteenth- and early nineteenth-
century attestations of Chinookan languages even further downriver toward
the Pacific Ocean, where no Sahaptin speakers were living. What I am par-
ticularly concerned with here, however, is not a singular linguistic form as
such, so much as a “voicing,” a way of orienting one’s consciousness in
narration toward narrated—and narratable—worlds, that is innovative and

The Fieldwork Encounter and the Colonized Voice of Indigeneity 29

 unique in all the Chinookan corpus and bespeaks a new kind of contact-era
consciousness on the part of Mr. McGuff and those of his generation and
later whose language I was also able to study.9
Now Mr. McGuff was usually the interpreter mediating between Sapir
and Mr. Simpson and other oldsters, but in the stretch of talk I am concerned
with here, he himself served as “informant,” talking slowly about “winter
bathing” as Sapir wrote it all down in the summer of 1905 for later translation
and publication as an “ethnologic text,” one revealing or explicating cultural
customs, in the Boasian way of distinguishing types of texts. In such an
enterprise, an “ethnologic text” is differentiated from a “myth text,” in which
is narrated the cultural cosmogony, how things became as they now are based
on plots of transformation of a previous moral order of the universe. In
previous work I have analyzed several features of this particular dictation,
from the particular constructional details of grammar to its genre character-
istics.10 Here, I want to recapitulate some of these matters with an eye spe-
cifically to linguistic and cultural change, embedding that moment in the
White Swan, Washington, apple orchard between the thirtyish-year-old Peter
McGuff and the twenty-one-year-old Edward Sapir in its proper envelope of
cultural and linguistic histories. I start with the larger genre issues and pro-
ceed to the linguistic forms of particularly revelatory value that illuminate
.
what the evidential passive form [a]ugwíłxix tells us.
As noted, Sapir thought that this dictation fit into the rubric of explicating
“customs,” in the instance that of “punishing” pubescent children who fell
asleep during long nights of wintertime myth-telling. The “punishment” was
a solitary, if highly choreographed ritual dip in the icy waters of the Columbia
River. What could be the cultural logic here? It involves the connection
between the characters in cosmogonic myths via the tendrils of immanence
that epistemically remote world still had with the current-era Wascos and Wish-
rams through the “spirit power complex.”11 As I have explicated this dictation,
in fact, it is not merely Mr. McGuff’s recounting of custom in this respect, saying
that when one fell asleep as a child one had to jump into the water through
a hole cut in the ice of the frozen Columbia River; it is a genre of interaction, I
maintain, that my own consultants of that very generation performed with me
as well in the 1960s and 1970s, one I caption “Why I Don’t Have a ‘Spirit
.
Power.’” (The “spirit power” [Kiksht -iúłmax] quest of late pubescence/early
adolescence, if successful, endowed an individual in the regional societies with
characterological, personality, and status attributes necessary to successfully
navigate the adult life course. Spirit powers came from direct, secret, generally
nocturnal encounters with faunal avatars of the very species that, as narrative
characters, populate the cosmogonic world of creation myths.)
Such a text of “Why I Don’t Have a Spirit Power” has a specific Bakhtinian
“voicing” structure. It is not the voice merely detailing, however elaborately

30 Representations
or telescopically, distantly remembered events—Mr. Simpson’s long story in
the same volume of his participation in the Paiute (1860) or perhaps Ban-
nock War (1878) is of this type. In such a narrating stance, the historicity of
such past events is uniformly projected from the here-and-now of an event in
which “what happened”—and even why and wherefore—is in a coherently
constructed prior time-and-space. In Mr. McGuff’s presentation, rather,
Sapir encountered a confrontation in the narrating present of a recuperated
consciousness situated within earlier childhood events in relation to the nar-
rating consciousness of the present 1905 moment. I know of no other text in the
whole Chinookan corpus—four different dialect collections—that features
this kind of split or fully doubled consciousness. That recuperated conscious-
ness in this narrative event in 1905 is Mr. McGuff’s young self: hearing, seeing,
experiencing things with a Wishram youngster’s naive belief, hope, and ingen-
uous expectation of encountering a spirit power. Mr. McGuff’s 1905 conscious-
ness comments in conclusion—perhaps wryly, perhaps wistfully, perhaps
ruefully—on the fact that, even though earlier as a child he had done every-
thing correctly, or nearly correctly, in submitting to the inevitable “punish-
ment” for falling asleep during the nightlong telling of cosmogonic myths, he
did not after all acquire a spirit power (though perhaps, in keeping with the
goal of the regime, he owes his always excellent health to the presumptively
fortifying and ultimately also protective effect of the bracing “winter bathing”).
How is that recuperated consciousness made manifest on the plane of
morphology and syntax and word choice, making up what we term the deno-
tational text of the dictation event? It is clearly through verbally constructing
a secondary (transposed) deictic center from the perspective of which events are
projected as experienced in the distinct chronotope (“spacetime”) of the
young Pete, who is a character in the narrative world, not a participant in
the event of narration. Let us see very briefly how that works. As shown in
figure 1, Mr. McGuff’s text is built around four direct quotes (which I have
boxed in and labeled with roman numerals), each bit of quoted utterance
occurring in a situation set up by putting himself—and us—retrospectively
into the situation described. Think of the parallel “narrative present” of
vernacular American English: when, recounting a past event, we might say,
“So he says ‘Here’s your lost book!’ And I say ‘Great! Now I won’t have any
overdue fine!’” Note, in contrast to the expected past inflection (said), the
present tense inflection of the framing predicate lexeme say-, in which the
sayings and doings of the characters in the past event being narrated seem to
be treated as, in effect, within an experiential envelope knowable to the
participants of the current ongoing event of communication, for which
a so-called “present,” that is, non-past tense would be used (though note that
the framing verb say- here remains in the same simple un-auxiliated form in
this narrative usage as it would have with the past inflection; this is clearly not

The Fieldwork Encounter and the Colonized Voice of Indigeneity 31

 a mere substitution of the descriptive or true present-of-ongoing-activity,
which in English requires a progressive aspect construction, [I] am/[he] is
saying). Note also the spatial and temporal organization in what is directly
quoted from the characters in such vividly recounted situations—“Here is
your lost book!” (proximal-presentational here; present tense is; earlier inter-
locutory participant your) and “Now I will not have any overdue fine!” (adverb
of concurrence now; earlier interlocutory participant I; earlier futurity of will
have). Such narratively quoted representations of what was uttered by inter-
locutory parties in some prior communicative event deictically project from
the configuration of roles of the interlocutors in that very situation of interac-
tion; note how the narrating frame itself is, as well, figuratively transposed, via
tense, at least in part to that spacetime. Contrasting with the situation in which
the story is being recounted, it becomes a secondary deictic center, a virtual
“here-and-now-in-the-past” in which the uttered deictic (indexical-denota-
tional) “here” and “now,” “I” and “you” that our characters speak as well as the
tense deixis of the frame anchor whatever and whoever is spoken about in the
there-and-then context of the grammatical subjects’ consciousness.
To be sure, Mr. McGuff uses the tenseless “simultaneous/ongoing” verb
construction several times with a force equivalent to such an English narra-
.
tive present. In the second printed line, line 9a, we have Ag_ a kwá=ba n3-x-u4-
gwi-|wícˇa-√tk-t “Now there I3 . . . [am] listening to them4” ( . . . am in a state of
“giving ear,” to gloss the literal Kiksht idiom). There is no actual tense prefix
in verb-initial position, but rather only the final suffix -t indicating a frozen
interval of time for young Peter in a durative or stretched-out event of a recur-
ring type. Similarly, in describing young Peter’s very wet, shivering body just
returned from the frigid dip in the Columbia in line 4b, Mr. McGuff vividly
reports icicles sticking out in every direction from his hair “they3-[are]-sever-
ally8.1-flying7+8.2+11-out9-on5 it4” i3-ł4-x5-√k’wá7-iu8.1-l8.2-k9-ł11, again a tenseless
form representing a continuous, spatially distributed state of individuable
things (the icicles) stretched out and around in a space the center point of
.
which is the youngster’s sensation of his head and hair (ił-náłxat). Let us pay
closer attention to the distribution of such forms in the overall textual struc-
ture (entextualization) that emerges in the course of Mr. McGuff’s dictation,
highlighted in figure 2 by extracting the poetic parallelism of the initial, scene-
setting introduction and the culminative focal episode upon young Peter’s
return from the Columbia River.
In the earliest narrative section, Mr. McGuff starts off by framing the
temporal relation to the current moment of interaction with Sapir: “Long
ago when-I-was-a-child.” What took place then? He narrates in the habitual
aspect about events that were a regular occurrence: “The old people would
.
(a- . . . -a) tell myths ([i]qánucˇk -√x7- ‘do, make myth’) in wintertime”; “[A-
certain-] someone would (a- . . . -a) give to (-l5-u6-√t7-) me an axe”; and so on.

32 Representations
 figure 2. Framework of parallelisms in narrative deixis.

In Mr. McGuff’s youth, like all children, he experienced multiple, recurring

episodes of being “punished” for falling asleep during those long winter
nights of cosmogonic mythopoesis.
Note by contrast how he frames the specific events of interest to his young
self: “Now there [= at these occasions where the myths were being told, in the
semisubterranean winter house] I’m-in-the-state-of-listening-to-them.” This last
verb form, the translation of which I have made a bit complex, is a formally
tenseless “simultaneous/ongoing” form, in the default deictic case of here-
and-now conversation, the equivalent of the English true “present” in be- . . .
V+ing, for example, [he] is walking (“now”). Mr. McGuff is using the closest
thing in Kiksht to what we term in our European literary tradition the “vivid
narrative present style” to capture what it felt like to be at the (here, generic
or recurrent) event from the deictic perspective of his now narratively
conjured-up young self. With him, as readers peering over Sapir’s, the orig-
inal addressee’s shoulder to his notebook page now printed, we, too, are
transposed into this event from that perspective.
To be sure, there are some passages in the myth and historical narrative
texts we have from older narrators like Mr. Simpson that also use some of the
deictic categories, like spatializing directional in predicates—“proximad” -t6-
“toward ‘here,’” “distad” -u6- “toward ‘there’”—as a narrative resource so as to
capture the onstage orientation of complex events that are experienced
through a myth-age character’s consciousness. Its use here in the first person,
the voicing of autobiographical memory is, I would submit, an indicator of

The Fieldwork Encounter and the Colonized Voice of Indigeneity 33

 a cultural transformation taking place in Mr. McGuff’s generation of Anglo-
schooled Native Americans, in which their prereservation youths take on
something of the quality of a “myth” era, that is, an era before extraordinary
and violent change from the older, riverine fishing, hunting, and gathering
life to the life of small-scale ranchers and farmers making a living in the
middle ranges of the Yakima Valley agricultural lands.
But I want to focus on the unmistakable parallel to Euro-American “indi-
rect free style” recuperation of consciousness that occurs in how Mr. McGuff
frames the fourth remembered quotation, labeled with Roman numeral IV,
on the return home of his youthful self, dripping and uncomfortable, from
the “punishment.” Again, a scene-setting habitual form: “When (=b t) I would
e
(a- . . . -a) come back (-t6-√i7-mam9 ‘hither-go-to.goal’),” spoken from the nar-
rator’s 1905 sensibility. And then a switch to Peter’s sensory consciousness-in-
the-typical-moment: the icicles felt sticking to his hair; a warm roaring fire
suitably prepared. What precedes this quotation, the culminating directive of
the elders waiting for him to return, features a historically new grammatical
verb form, the evidential passive, a form unattested not only in the Boas texts
in Lower Chinook and Kathlamet from downriver, or the Clackamas texts
from the Willamette, but also unattested in any older speaker’s Kiksht of
which we have records, though it is attested in Walter Dyk’s (1933) and in
.
my own later field materials. The form in question, as earlier noted ugwíłxix
“it [a fire, wá-tuł (fem.sg.)] must have been kindled/lit,” signals unequivo-
cally the transposition of deictic center to the consciousness of the moment.
It is indirect free style, narrating the event as it would be thought/narrated
in-and-at the moment it describes by a participant—here, the young, wet,
cold, dripping Peter McGuff—in that narrated event.
Indirect free style, long associated with literary “stream of consciousness”
authors, is, to be sure, a particular Bakhtinian “voicing” effect, in which, as
also in so-called indirect discourse, the permeable boundary is crossed
between the narrator’s perspectival origin-point in the world of the actual-
ized event of communication and a character’s perspectival origin-point in
the event of the narrated world. In direct discourse, narrating frame and
narrated utterances and thoughts are formally separated; in indirect dis-
course, the narrator’s framing perspective indexically colonizes that of the
character, interpreting it by in effect describing it; in free indirect style, the
directionality is reversed, and the ascriptive conceptual-affective and discur-
sive formulations of a character within the narrated world seem themselves to
give communicative shape to the narration, in effect transposing the char-
acter’s perspectival origin-point out into the world of narration through the
character’s now quasi-re-presented thoughts and words.
Virginia Woolf starts out Mrs. Dalloway with canonical indirect dis-
course, a metapragmatic framing construction, Mrs. Dalloway said, in the

34 Representations
past tense projected from a moment in which the character is walking in
London on her household errand.12 (To be a persnickety grammarian, had
Woolf been a nineteenth-century novelist writing in a different “voice” she
probably would have written “Mrs. Dalloway had said,” to differentiate her
earlier resolve and communication of that resolve before setting out into
the city on her errand. That Woolf does not, leaves the reader a bit floating
in time, if not in space, an effect perhaps specifically—and, in the passage,
ingeniously—planned by the author.) The frame is followed by a subordi-
nate clause that explains that errand—were it in full form it would be
introduced by the subordinator that—that gives a propositional interpreta-
tion of some utterance earlier in time, here to the effect that Mrs. Dalloway,
rather than her servant named Lucy, committed herself—a performative
act effected by her uttered words!—to buy some flowers for a party to take
place at her home later that day.
Mrs. Dalloway said she would buy the flowers herself.
At this point in the narration, Woolf runs through the mindset, the
justification to herself for Mrs. Dalloway to have committed herself to under-
taking the errand. Here is indirect free style. Note that it begins with the
logical connector for, synonymous with because, and is followed by a version
of what, presumably, Mrs. Dalloway had earlier said to Lucy, or at least had
thought at the moment when she formulated her resolve at that earlier point
in the day: “You have your work cut out for you,” having to undertake the
major preparations for the party, including opening up connecting rooms
with the assistance of a certain Mr. Rumpelmayer’s crew. All this is still in her
mind as she ambulates:
For Lucy had her work cut out for her. The doors would be taken off their hinges;
Rumpelmayer’s men were coming.
And her mind comes quickly to appreciate the particular delight of being
out-and-about on this particularly bright London morning, experienced as
her unanticipated reward, as it were, for her kindness to Lucy. Woolf uses as
parenthetical metapragmatic frame, thought Clarissa Dalloway (and note the
use of full name precisely at this moment at which the character’s affective
engagement with her world through her sensorium becomes the focus of
attention!) the frame interrupting an essentially direct-discourse citation of
affect, characterized by Woolf by including a striking simile: “And then, what
a fresh morning—!”
And then, thought Clarissa Dalloway, what a morning—fresh as if issued to
children on a beach.
Observe the details here for Kiksht, then. Mr. McGuff uses a temporal
adverbial clause to introduce quotation IV that is completely parallel to the
very opening phraseology in part I. He says, “Now at-this-(place/time)
I-would-arrive; now long-since it-must-have-been-kindled the fire; [from] the

The Fieldwork Encounter and the Colonized Voice of Indigeneity 35

 ends-of my-hair icicles (are-)sticking-out-from (in all directions).” Along with
the simultaneous and continuative form “they(-are)-hanging-from it” (i-ł-x-
√k’wá-iu-l-k-ł), he formulates what I have termed an “evidential passive,” [a-]
.
u-gwíłx-ix wa-túł “it [namely, the fire] must-have-been-kindled.” Such a form is
generally used when, in the presence of sensorily available evidence, the
speaker infers that someone must have done something to someone or some-
thing, such that the evidence is the index of this prior agentive action. It is the
passive verb form of that inferentially prior action, used together with a char-
acteristic deictic suffix -ix that independently occurs on verbs of positionality.
The form as such is strongly deictic, linking the inference of the utterer of the
form that “[someone] must-have-V-ed” to the perception of something in the
context of inference-stimulating consciousness. Here, note, the recuperated
consciousness of Mr. McGuff—young Pete—reacts to the proximity of a roar-
ing fire, from which it infers that “someone must-have-lit/kindled-it.”
This evidential passive form, providing precisely the parallelism in a verb
form and thus predication to the “simultaneous” forms of the earlier seg-
ments of the text, is, moreover, an innovation within the Wishram-Wasco
dialect of Kiksht of Mr. McGuff’s generation. As I remarked earlier, it does
not occur either in Clackamas, at the western extension of Kiksht, in which
we have extensive text materials, dictated to Melville Jacobs by Mrs. Victoria
Howard, born circa 1865, nor in the Kiksht of the other speakers from whom
Sapir gathered materials in 1905–09. It does, however, occur extensively in
the Kiksht of my own consultants born around the same time as Mr. McGuff.
(Walter Dyk’s “Wishram” fieldwork of 1930–33 recorded the form, but does
not recognize its proper position within Kiksht grammatical structure.)
As an index of historical process, note that the particular voicing of
narrative chronotopes by Mr. McGuff points to a shift of perspective from
one in which the myth era was the generative, cosmogonic space-time in
which was ordained the orderliness of the way humans related both to each
other and to the faunal universe. The shift is to what we would recognize as
a “before” and “after” defined by the definitive encompassment of Native
American life by the settler state in the grip of which Mr. Simpson and Mr.
McGuff both found themselves, Mr. McGuff all the more so as a product of
a colonial Anglophone educational system.
There is a sense in which, then, for Mr. McGuff English-derived modes of
forming narrative allow the interplay of the immediate and the recuperated
consciousness revealed in this plaint to Sapir (which apparently fell on deaf
ears). Discourse structures we can recognize as speakers of English seem to be
incipiently “enregistered”—rendered as coherently co-occurring text-forming
features—in Mr. McGuff’s generation’s Kiksht usage. Thus, note the textual
series of (a) tenseless punctual present-simultaneous forms, such as n-x-u-gwi-|
wícˇa-√tk-t “I . . . in a state of ‘giving ear’ [=listening] to them”; (b) habitual

36 Representations
 .
conditionals (formally “futures”), such as a-łg-n-u-√lx+ám-a “a certain one
would (on occasion) say to me”; (c) tenseless evidential passives, such as the
.
form in question, [a-]u-√gwíłx-ix “it [fire] must have been lit”; and (d) tenseless
duratives-of-state, such as i-ł-x-√k’wá-iu-l-k-ł “they are hanging down from it”
(icicles from the ends of his hair). This series of verb forms congruently con-
stitutes predications deictically centered in the recuperated consciousness of
the then young Pete McGuff, with the exception of “habitual” type (b) all of
them tenseless—hence predicable only of some spacetime projected from
a remote here-and-now deictic center, one that clearly is described in the
narrative as (e) n-k’áškaš=b t “when I was a child” during which period all the
e
myth-telling and “winter bathing” took place. That narrative consciousness
shifts suddenly back to the deixis of the ongoing interactional moment with
Sapir as Mr. McGuff sums up right after the last quotation of the elders; here
.
he switches to the “remote/mythic past” form (f ) ga-q- ′n-t-√x “they [imper-
e
sonal] did to me” (lines 7–8b) with distinctive ga1- . . . -t6- . . . inflectional form
for temporally remote events experienced in a long-ago earlier phase of life.
This cleavage of predicational deixis into two intersecting planes—one
projected from the here-and-now of consciousness in the moment of discur-
sive interaction with Sapir, the other projected from the recuperated con-
sciousness of the young Peter—creates the effect of two “first persons.” But
what I wish to emphasize here is the fact of two predicational registers, in effect,
“[objective] past” and “free indirect [reinhabited] past” as the student of
Kiksht stylistics notices in the chain of coherent, cotextual continuities and
compatibilities of verb forms (a) through (d), intersecting with another
chain, that of (b)–(e)–(f ). (Note the parallelism here to how Émile Benve-
niste discovered two registers of predication in French, one, objectively
“voiced,” as Bakhtin would say, admitting the “passé simple” [for example,
je fus] and several other tense, aspect, and perfect forms, the other, subjec-
tively “voiced,” excluding the “passé simple” though admitting several of the
other forms—thus two intersecting but distinct planes defined by enregis-
tered chains of predicational compatibilities within discourse segments.13)
And further, the historical point to hand: This kind of textual assemblage,
bespeaking a contrast of “voicings” and thus indexing a historical change in
historical consciousness, seems to be an innovation of Mr. McGuff’s genera-
tion, the first to be schooled at places like his boarding school, Chemawa, in
the colonial language, English. It is, I would say, the adaptive assimilation of the
deictic narrative pragmatics of English that we see in his transcribed text, one
in which the specific historical emergence of the evidential passive form-type
makes perfect sense as part of a predicational register that contributes to
establishing two serially organized modes of consciousness, a present-and-pro-
spective-in-the-past (the young and obedient Peter paying anxious attention
to his old people in the hope of getting a spirit power) and a present-and-

The Fieldwork Encounter and the Colonized Voice of Indigeneity 37

 retrospective-in-the-present (the sadly spirit-power-less Mr. McGuff, older and
wiser, interacting with Sapir). Mr. McGuff’s narrative deployment of the variously
compatible and coherent predicational forms in entextualizing/contextualizing his his-
torical consciousness is the very locus of emergence and change of the grammar of Kiksht.
That is to say, this newly emergent discursive use of the Kiksht grammatical
category of “evidential passive” for the “indirect free style” of personal memory
points to a change in temporal or historical consciousness across the cataclysm
of settler-state reorganization of social life—even of where Wishram and Wasco
culture could exist in the peripheries of the settler-state political economy.
To be sure, not only plurilingual speech communities, but every unilin-
gual language community is internally diverse from a sociolinguistic point of
view—as Bakhtin long ago emphasized with his concept of разноречие or
“heteroglossia.” A “language,” then, actually consists of the logical union of all
the repertoires of distinctive registers controlled by people who are, to different
degrees, encompassed in, and oriented to (or by) its linguistic norms for
using and evaluating the well-formedness of utterances.14 (As we use the
term “language,” it denotes norms for denotational—referring and modally
predicating—verbal communication, cross-cut and intersected, then, by what
we term “indexical” values associable with specific signaling forms that link
language to socially definable contexts of interaction.) Thus, as socially self-
contextualizing actors, all communicationally competent speakers of a lan-
guage actually orient their usage to many such registers implicit in the total
community repertoire, among a subset of which they move unremarkably as
a function of interactional situation (or with noticeable performative entail-
ments when communicating against expectations).15 Various studies have elab-
orated such institutionalized registers as “sports announcer talk,” bespeaking
a professional voice in broadcast or print, or “motherese,” a register in Euro-
American society for caregivers who are interacting with infants and young
children, the latter itself sometimes incorporating features of the “baby talk”
register adults project onto children (who actually speak quite otherwise).16
Of course, these registers are characterizable as contextually, and thus
ethno-demographically and -situationally specific clusters of criterial indexi-
cal features, together with all the machinery necessary to render denota-
tional discourse entextualized form-in-context. To the native speakers, as it
were, registers are “alternate ways of ‘saying the same thing,’” that is, in
a sense situationally appropriate or expectable equivalents or near-enough
equivalents of communicated purport [“meaning”] formulated in a particu-
lar way. (Of course, no such thing can in fact be true when we analyze the
language, making instability of denotational code an essential outcome of the
cultural fact of enregisterment.) Every register has features that are highly
salient indexicals of communicating “in register,” the occurrence of which
keys communication as to relevant participant identities being invoked, to

38 Representations
ongoing social action, and so on; these stand out in relation to all the rest of
linguistic form necessary to making denotational sense in an utterance, but
which clearly are features of the language. Registers differ, moreover, in
which kinds of structural features of linguistic signs are paramount in their
indexical salience: note how some registers are predominantly phonological
(see Peter Trudgill’s study of pop music register in the 1960s and 1970s);
others, such as our registers of variously tribal “academese,” being principally
lexical—through terminologization [Ecce!]—and morphosyntactic—featur-
ing sesquipedalian Graeco-latinate derivational morphology, involved syntax
of subordination and modification, and so on.17
Being drawn into pragmatic paradigms of indexicality across register
divides—enregisterment of linguistic forms, as we term it—and its opposite, the
collapse of such distinctions, constitutes a dynamic force acting upon the
organization of any language so as to change it. Salience of formal features
that have an isolable, performative indexical “punch” (reliable indexical
semiosis) thus fluctuates in time according to cultural values, coming into
being and disappearing through the many links of interdiscursivity and of
ideologically (culturally) informed metapragmatic (un)consciousness, all the
stuff of culture in its usual operation.18 When we study the contours of usage
and the degree of indexical salience of so-called sociolinguistic variables in
the real time of a synchronicized slice—the ethnographic horizon—of diach-
rony, we get a glimpse of the register structure that is the real functional locus
of any language form. Such register structures frequently operate at the
historical intersection of language communities that have come into perdur-
ing, as opposed to fleeting contact for one or another reason. In such situa-
tions, a plurilingual speech community is in evidence, such as frames the
interaction of Peter McGuff and Edward Sapir, where alternate “languages”
as such come to be contextually mobilized by thus polyglot speakers, as
a function of the sociocultural contexts in which communication goes on.
With the concept of register in mind, careful philological study of such
texts-in-context focusing on the internal poetics of grammatical form thus
opens a larger vista of sociohistorical phenomena. The linguistic forms are
organized into registers that are associable with denotational and interac-
tional genres of particular sorts, first-person memory narrative here explain-
ing a personal deficit—no spirit power—that coincides with massive social
upheaval and cultural change. To be sure, the manifold of registers of which
any “language” is composed is in constant flux, but we see how the emergent
“evidential passive” linguistic form reveals the modernizing force of a new
predicational style assimilated from Anglophone others in the plurilingual
speech community in which intersect English, Sahaptian, and Kiksht.
Of course, it is precisely in the kind of interaction where an indigenous
Mr. McGuff would be telling a Euro-American Mr. Sapir about “Why I Don’t

The Fieldwork Encounter and the Colonized Voice of Indigeneity 39

 Have a Spirit Power” that the ironic or wistful or angry split consciousness
would become institutionalized, to the performance of which the innovative
enregisterment of predication-types, including the “evidential passive” spe-
cific to Wishram-Wasco in this generation, would become useful—in a sense,
“selected for” as a grammatical affordance of “voice.” That is to say, there is
a substantive link between what we might term linguistic change of this sort
and cultural change insofar as erstwhile efficacious ritual practices are put
into question by the kind of asymmetric contact with the Anglo society that
Mr. Simpson’s and Mr. McGuff’s groups have suffered. But there is also an
intersection of linguistic and cultural change insofar as the very interactional
genre we have encountered via Sapir’s inscription is the vehicle of shifting
cultural normativities. Here, we catch a shift in the very culturally specific
anxieties about identities and their bases in moving from those defined by
the spirit quest “complex”—to use the Boasian diffusionist term—to those of
importance in the situation of contact and settler colonization. To Sapir,
Mr. McGuff had obviously moved a good way into the encompassing larger
society:
Pete McGuff . . . may serve as a type of the younger generation of Indian. . . . He has
not of course that feeling for the old Indian life, and faith in the truth of the myths,
that a man like Louis Simpson has; nevertheless, in spite of his white man’s ratio-
nalism, he is not at all disposed to dismiss as idle the ideas of the Indians in regard
to medicine-men and guardian spirits. He has been trained in the Agency school,
reads and writes English well, and in general displayed throughout remarkable
intelligence.19

How much of the nuance of Sapir’s characterization of the divided alle-

giances of his field assistant and interlocutor now seems to us to be contained
in the very text we are examining! Built as it is on the duplex—even dialec-
tical—voicing structure of the colonized organic intellectual’s split conscious-
ness, to the attuned and sensitive reader the “Winter Bathing” chapter in
Wishram Texts indexically contextualizes itself through an entextualized struc-
ture of intersecting predicational registers, in one of which the “evidential
passive” is a useful affordance of a culturally emergent voice.

Notes

Earlier versions of this paper were presented as follows: “The Multiple Linguistic
and Cultural Contextualizations of Language Change,” circulated and discussed at
the Wenner-Gren Foundation-National Science Foundation Workshop “Analyzing
Change: Linguistic and Cultural Models,” New York City, 9–12 April 2008 (Joel
Robbins and Bambi Schieffelin, organizers). “Culture and History in a Kiksht
hapax legomenon: The Ethnohistorical Contextualization of Linguistic Form,” given

40 Representations
 as the keynote address of the Wyoming Humanities Council conference “Lan-
guage, Culture and History” held at University of Wyoming, Laramie, 1 July
2010 (Leila Monaghan, organizer). An expanded version was circulated and dis-
cussed at the Semiotics: Culture in Context Workshop, at the University of Chi-
cago, 14 October 2010. Thanks to the organizers of the first two occasions for the
invitations to present this work, and to all three interlocutory groups for useful
comments for revision. I am grateful as well to the three editors of this issue of
Representations, originally organizers of a symposium at the 2015 annual meeting of
the American Comparative Literature Association in Seattle, for comments on
a prefinal draft, now extensively revised in the light thereof.
1. See Regna Darnell, “Franz Boas, Edward Sapir, and the Americanist Text Tradi-
tion,” Historiographia Linguistica 17, nos. 1–2 (1990): 129–44; and Michael Silver-
stein, “From Baffin Island to Boasian Induction: How Anthropology and
Linguistics Got into Their Interlinear Groove,” in Franz Boas as Public Intellectual:
Theory, Ethnography, Activism, ed. Regna Darnell et al. (Lincoln, NE, 2015), 83–127.
2. For a reasonably nontechnical conspectus on such analysis of dyadic interaction,
worked in detail on an extended example, see Michael Silverstein, “The Voice of
Jacob: Entextualization, Contextualization, and Identity,” ELH 81, no. 2 (2014):
483–520. For an elaborate guide through the analytic procedures necessary to
contemporary linguistic anthropological work, see Stanton Wortham and
Angela Reyes, Discourse Analysis Beyond the Speech Event (New York, 2015).
3. I have treated this text in respect of linguistic and cultural details in two prior
publications, emphasizing in the first its genre as a fieldwork encounter, and in
the second innovations in the syntax of its language consistent with influence
from English. See Michael Silverstein, “The Secret Life of Texts,” in Natural
Histories of Discourse, ed. Michael Silverstein and Greg Urban (Chicago, 1996),
81–105; and Michael Silverstein, “Kiksht ‘Impersonals’ as Anaphors and the
Predictiveness of Grammatical-Categorial Universals,” Proceedings of the Berkeley
Linguistics Society 21 (1995): 262–86.
4. Our modern Americanist orthographic practices differ somewhat from those of
1905, when Sapir transcribed into his notebooks (now in the possession of the
American Philosophical Society in Philadelphia), but in the main are consistent.
My representations of Kiksht forms are in a phonological orthography, more-
over, while Sapir’s are closer to phonetic-as-heard. For those readers wishing to
get a sense of the phonetic realization of Kiksht, note that orthographic <ƚ>, so-
called “voiceless ell,” is pronounced like English orthographic <l> without any
.
voicing; orthographic <x>—here used for a more canonical letter with a dot
underneath—represents a uvular (far-back-of-tongue) pronunciation as at the
end of German Bach, while orthographic <x> represents a similar sound made
slightly in the direction of the last sound of German ich (Sapir uses, respectively,
_
<x> and <xˑ>); and <g>—more canonically with dot underneath—represents
a uvular correspondent to velar <g>, voiced sounds paralleling the uvular <q>
and velar <k> made without voice. Stress is indicated on the relevant vowel, all
three of which, written <a, i, u>, are pronounced as in Continental orthogra-
phies. In Sapir’s orthography, the unstressed voicing, the so-called schwa of
syllables like the second one in English <kettle> or <button> is indicated with
a small cap <E>; our modern orthographies use a turned letter, thus: < >. e
5. I have treated this grammatical form in some descriptive and historical detail,
including its pragmatic, or context-sensitive aspects in “Deixis and Deducibility
in a Wasco-Wishram Passive of Evidence,” Proceedings of the Berkeley Linguistics
Society 4 (1978): 238–53.

The Fieldwork Encounter and the Colonized Voice of Indigeneity 41

 6. The definitive anthropological account of these people is David H. and Kathrine
S. French, “Wasco, Wishram, Cascades,” in Handbook of North American Indians,
vol. 12, Plateau, ed. Deward E. Walker, Jr. (Washington, DC, 1998), 360–77.
An earlier treatment emphasizing the history of cultural transformation in the
encroaching settler state is David H. French, “Wasco-Wishram,” in Perspectives in
American Indian Culture Change, ed. Edward H. Spicer (Chicago, 1961), 337–430.
7. On Chinook Wawa or Jargon, see Michael Silverstein, “Dynamics of Recent
Linguistic Contact,” in Handbook of North American Indians, vol. 17, Languages,
ed. Ives Goddard (Washington, DC, 1997), 117–36. On the Chinookan-speaking
peoples of this region, see Michael Silverstein, “Chinookans of the Lower Colum-
bia,” in Handbook of North American Indians, vol. 7, Northwest Coast, ed. Wayne
Suttles (Washington, DC, 1990), 533–46; and, in a more popular vein, Robert
H. Ruby and John H. Brown, The Chinook Indians: Traders of the Lower Columbia
(Norman, OK, 1976).
8. Edward Sapir, Wishram Texts (Leiden, 1909), xi.
9. The concept of “voicing,” that is, coming to inhabit a particular “voice” in the
role of narrator, is a central focus of Bakhtinian criticism, developed in Mikhail
Mikhailovich Bakhtin, The Dialogic Imagination, ed. Michael Holquist, trans.
Michael Holquist and Caryl Emerson (Austin, TX, 1982), in which are elabo-
rated not only this concept, derived from musicological discourse, but as well the
related notions of “polyphony” and “double voicedness” in the way a narrator
(dis)aligns with the characters and social interests that populate a narrated
universe.
10. See note 3.
11. This geographically widespread system of customs and beliefs, whereby the eth-
ical, moral, and psychodynamic forces embodied in the cosmogonic “myth” era
license and guard over the being-in-the-social-world of an individual who has
successfully encountered an avatar of that earlier order of things. Ruth Bene-
dict’s Columbia University doctoral dissertation was a comparative study, pub-
lished as The Concept of the Guardian Spirit in North America (Menasha, WI, 1923).
For the specifically Wasco-Wishram manifestations, see Leslie Spier and Edward
Sapir, Wishram Ethnography (Seattle, 1930), and, based on more recent fieldwork,
Michael Silverstein, “Private Ritual Encounters, Public Ritual Indexes,” in Ritual
Communication, ed. Günter Senft and Ellen Basso (Oxford, 2009), 271–92.
12. Virginia Woolf, Mrs Dalloway (London, 1925).
13. Émile Benveniste, “Les relations de temps dans le verbe français,” in Problèmes de
linguistique générale (Paris, 1966), 237–50. See also Monika Fludernik, “Narratol-
ogy and Literary Linguistics,” in Oxford Handbook of Tense and Aspect, ed. Robert I.
Binnick (Oxford, 2012), 75–101, for a critique of how Benveniste’s analysis has
been made use of by subsequent narratologists, such as Ann Banfield, Unspeak-
able Sentences: Narration and Representation in the Language of Fiction (London,
1982), which especially treats indirect free discourse.
14. Mikhail Bakhtin, “Discourse in the Novel,” in The Dialogic Imagination, 259–422.
For the derivation of the entailments of heteroglossia/enregisterment to con-
ceptualizing a language sociolinguistically, see Michael Silverstein, “Denotation
and the Pragmatics of Language,” in Cambridge Handbook of Linguistic Anthropol-
ogy, ed. Nick Enfield, Paul Kockelman, and Jack Sidnell (Cambridge, 2014),
128–57.
15. See especially Asif Agha, “Registers of Language,” in The Blackwell Companion to
Linguistic Anthropology, ed. Alessandro Duranti (Malden, MA, 2004), 23–45; and
Asif Agha, Language and Social Relations (Cambridge, 2007).

42 Representations
16. Charles Ferguson, “Sports Announcer Talk: Syntactic Aspects of Register Varia-
tion,” Language in Society 12, no. 2 (1983): 153–72; Catherine Snow, “Mothers’
Speech to Children Learning Language,” Child Development 43, no. 2 (1972):
549–65; Charles Ferguson, “Baby Talk in Six Languages,” American Anthropologist
66, no. 6, part 2 (1964): 103–14.
17. Peter Trudgill, “Acts of Conflicting Identity: The Sociolinguistics of British Pop-
Song Pronunciation,” in On Dialect: Social and Geographical Perspectives (Oxford,
1983), 141–60.
18. See references in note 14, and on the dynamics of the semiotic processes of
enregisterment, see Michael Silverstein, “Language and the Culture of Gender:
At the Intersection of Structure, Usage, and Ideology,” in Semiotic Mediation: Socio-
cultural and Psychological Perspectives, ed. Elizabeth Mertz and Richard J. Parmentier
(Orlando, FL, 1985), 219–59; and Michael Silverstein, “Indexical Order and the
Dialectics of Sociolinguistic Life,” Language and Communication 23, no. 3–4 (2003):
193–229.
19. Sapir, Wishram Texts, xii.

The Fieldwork Encounter and the Colonized Voice of Indigeneity 43